Applied Energy: Luigi Rubino, Clemente Capasso, Ottorino Veneri

Applied Energy 207 (2017) 438–464
Contents lists available at ScienceDirect
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
Review on plug-in electric vehicle charging architectures integrated with

distributed energy sources for sustainable mobility
Luigi Rubino a, Clemente Capasso b, Ottorino Veneri b,⇑
a
DIII – Università degli Studi della Campania Luigi Vanvitelli, Via Roma 29, Aversa (CE) 81031, Italy
b
Istituto Motori – National Research Council of Italy, Via Marconi 4, Naples 80125, Italy
h i g h l i g h t s
Outlook of PEV charging infrastructures.

Integration of electric and hybrid mobility with distributed energy sources plants.
Traditional and innovative power electronics architectures.
Inductive and conductive PEV charging systems.
PEV battery management.
a r t i c l e i n f o a b s t r a c t
Article history: In this review, the aim is to present a complete outlook for innovative charging infrastructures. In a real
Received 31 January 2017 smart grid scenario, these infrastructures are candidates to support the integration of electric and hybrid
Received in revised form 22 June 2017 mobility with distributed energy sources. In this paper, at the outset, an analysis of the scientific and
Accepted 28 June 2017
technical literature about main international standards and classifications has been provided. Also taken
Available online 2 August 2017
into consideration in this analysis are the expected challenges related to charging technologies for elec-
tric and plug-in hybrid vehicles, giving specific details on current and possible future trends for both sta-
Keywords:
tionary and dynamic inductive charging systems. In particular, for each charging level, traditional and
Plug-in electric vehicles
PEV charging infrastructure
more innovative power electronic architectures—equipped with the new technologies that support both
Energy storage systems slow and fast conductive charging operations for the new-generation road vehicle–have been reported,
Power converters described and analysed in detail. The analysis has been conducted through a comparison of power archi-
Smart grid tectures, in terms of efficiency, scalability and charging power/time of the vehicle battery packs. Specific
attention has also been devoted to off-board DC fast-charging architectures, which play a fundamental
role in the integration of stationary energy storage systems and renewable energy sources with the main
grid. Finally, in this review, a wide range of the most interesting applications, technical experiences and
international pilot projects have been summarized and discussed, with specific references to the new
technologies mentioned above. The overview reported in this paper highlights the importance of a proper
charging infrastructure, in combination with next generation energy storage technologies, to support the
large-scale diffusion of electric and plug-in hybrid vehicles.
Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction growth in the number of road vehicles and electric utilities—is

expected to increase continually in the next years [1,2]. Even
The huge consumption of liquid fossil fuels in transportation though recent advancements in internal combustion engine tech-
and energy generation sectors is considered to be one of the main nologies—encouraged by new international legislations that are
causes for the recent increase in atmospheric concentrations of enforceable—have motivated improvements in vehicle conversion
greenhouse gases; in particular, that of carbon dioxide. Moreover, efficiencies and tailpipe emissions, environmental issues still per-
the large demand for fossil fuels—mainly due to the worldwide sist, with drastic effects on earth’s atmosphere and human health
[3]. Consequently, the long awaited transition towards sustainable
transportation systems–supported by the efficient integration of
⇑ Corresponding author. renewable energy sources with the main grid—is becoming an
E-mail address: o.veneri@im.cnr.it (O. Veneri).
http://dx.doi.org/10.1016/j.apenergy.2017.06.097
0306-2619/Ó 2017 Elsevier Ltd. All rights reserved.
L. Rubino et al. / Applied Energy 207 (2017) 438–464 439
important challenge to be addressed from the technical and scien- would support the main grid in its interaction with the plug-in sus-
tific point of view. tainable mobility. In this case, the efficient integration of PEV is
The use of electric drives for vehicle propulsion systems repre- obtained through the direct or indirect interaction between the
sents an attractive solution, which involves various advantages main grid and on-board or off-board charging devices.
such as the high values of well-to-wheel conversion efficiency A large number of review papers have addressed the above
and the possibility to operate in urban areas in full-electric topics. In the first group of papers [7–11], an overview of the
mode—without local exhaust emissions. Encouraged by the above main technologies and issues related to PEV propulsion systems
advantages, the automotive industry commercially introduced have been presented. The aim of these papers is the evaluation
the first generation of Hybrid Electric Vehicles in 1997 (HEV) [4]. of the most promising on-board energy management/control
These vehicles were characterized by a propulsion system based strategies, associated with different power-train configurations,
on an Internal Combustion Engine (ICE), which works in combina- adopted by the main car manufacturers. Smart charging
tion with an electric drive with positive consequences related to approaches and vehicle-to-grid operations are analysed in detail
vehicle performance, fuel consumption and environmental impact. in [16–18], with particular reference to the improvement of grid
Nowadays, HEV represent mature technology and have been safety and reliability through the smart integration of PEV with
adopted by a conspicuous number of users in both the private the main grid. Other papers focus on traditional and innovative
and the corporate sector. Many HEV with various configurations power conversion architectures for PEV battery-charging devices.
are now available on the market and can be classified on the basis In particular, paper [19] presents an overview of the main solu-
of either hybridization degree, which is related to the ratio tions adopted for on-board battery charging devices, which are
between the ICE and electric drive power, or propulsion system generally suitable for low and medium power charging opera-
architecture (parallel, serial, parallel-serial, power-split, etc.) tions in a PEV. In [20], Yilmaz et al. report a review of the most
[5,6]. Recent developments in battery technologies have supported used topologies and standards for PEV battery chargers character-
the spread of a new generation of hybrid electric vehicles on the ized by different peak power, with details on both uni-directional
market, characterized by the capability to cover long distances in and bidirectional solutions.
full-electric mode. In this case, the vehicle battery pack can be In this context, this manuscript is aimed at providing a com-
recharged by using the energy coming from the main grid, by con- plete and updated analysis on charging architectures for PEV, with
necting the vehicle to an external charging plug using a specific reference to conductive and inductive charging technologies. This
connector. For this reason, these vehicles are also called Plug-in analysis is particularly focused on innovative power conversion
Hybrid Electric Vehicles (PHEV) [7]. A good alternative, especially architectures, based on the use of modular multi-level converters,
for urban use, is represented by Battery Electric Vehicles (BEV or which allow the optimal integration of PEV with the main grid
EV), where power to the electric drive is supplied from the on- through the efficient use of distributed energy storage systems
board battery pack. In this case, the ICE is completely excluded, and renewable energy sources. The presented topologies have been
as the large battery pack in the vehicle can ensure a daily travel analysed and compared in terms of scalability, efficiency and peak
range of about 150 km. Such vehicles can be connected to specific charging power, taking into account their impact on the main grid
charging points and recharged from the main grid using connec- in terms of power quality. The review ends with a description of
tors. Therefore, PHEV and BEV are both generally referred to as the main results related to worldwide applications of the above
Plug-in Electric Vehicles (PEV) [6]. Further details on hybrid and charging technologies in the context of pilot research projects
electric vehicles classification are widely available in the scientific funded by public or private institutions.
literature [8–11].
In fact, the large-scale spread of PEV in the automotive market
is still affected by battery technologies and the charging infrastruc- 2. Main standards for charging operations of PEV
ture. A slight increase in the demand for PEV in the private and cor-
porate sectors has been supported by new lithium based energy The main elements of PEV charging architectures are repre-
storage systems [12,13]. In fact, lithium compounds have allowed sented by the power grid, the communication network, and PEV
storage system technology to reach high values of energy and charging systems. The power grid supplies electric power to charge
power densities, with positive effects on PEV performance in terms PEV. For this reason, a large deployment of PEV involves prelimi-
of acceleration and daily travel range in full electric mode. The nary evaluations of the impacts that vehicle charging operations
above battery technologies can be charged during the night may have on the power grid and the identification of the best
through private low-power charging equipment, with charging charging strategies to be adopted. In this regard, a communication
times of up to 8 h. This type of operation presents a low power network is required to support the efficient integration of PEV with
requirement for the main grid and therefore can be considered the main grid on the basis of agent and multi-agent aggregation
generally acceptable for urban mobility. On the other hand, the schemes [21]. PEV Charging systems are devoted to the control
autonomy of PEV when compared with traditional vehicles based and management of vehicle charging operations, with smart inter-
on internal combustion engines, is still considered, by a large sec- actions with both vehicle Battery Management Systems (BMS) and
tion of the users, insufficient to cope with the well-known ‘range other centralized control systems at a higher hierarchical level. The
anxiety’ [14,15]. For this reason, the capillary diffusion of a proper development of power electronic technologies related to PEV
charging infrastructure would be required in order to encourage charging systems can play a key role in reducing both vehicle
the use of PEV for travel paths as well, which are longer than the charging times and the impact of charging operations on the main
typical daily urban paths. This infrastructure should support vehi- grid, which are still considered the main technical bottlenecks for a
cle owners with commonly available fast or semi-fast charging sta- large-scale proliferation of sustainable mobility [22].
tions or with mobile wireless power transfer charging systems A high-level classification of PEV charging systems is based on
during their journey. This means that high-power requirements, the presence or absence of wires for the electric power transfer
involving a large number of simultaneous PEV charging operations, between vehicles and charging units. On the basis of this classifica-
should be addressed through the existing electric network. In a tion, PEV charging systems can be classified as either conductive or
smart grid scenario, a PEV would be charged in an efficient and wireless PEV chargers. The second group is also referred to as
clean way through the intelligent integration of renewable energy inductive chargers, because in this case the electric power is trans-
sources and stationary energy storage systems. This scenario ferred through magnetic fields by using inductive coupling
440 L. Rubino et al. / Applied Energy 207 (2017) 438–464
between two or more coils, which work as a transformer with a big reported in [19] and described in detail in the next sections. This
air gap [23]. charging mode, highlighted in green in Fig. 2, is identified as Level
IEC, SAE and ISO standards are now available in Europe and the 2 and is used for power values up to 19.2 kW, with a three- phase
United States of America to completely regulate PEV conductive voltage AC source. The Level 3 charging devices, highlighted in
charging operations with a complete set of classifications and orange in Fig. 2, are related to charging operations with peak
rules, which range from the charging connectors to the topology power values greater than 19.2 kW and PEV charging times lower
of charging devices. A schematic overview of the above standards than 30 min. For this reason, Level 3 power converters are also
with their reference field is shown in Fig. 1 [19–29]. referred to as Fast- Charging Devices. Fast-charging operations
The first classification of conductive PEV chargers, which is can be performed through either on-board vehicle supply equip-
reported in the standard SAEJ1772, has been introduced by the ment, with the electric drive power converter and a three-phase
Society of Automotive Engineers (SAE) and is based on PEV peak AC voltage source, or with an external off-board DC power source.
charging power [25]. As a matter of fact, this classification has been In this last solution, the AC/DC power conversion is performed off-
in continuous evolution because new PEV manufacturers are intro- board, which is advantageous with regard to vehicle weight, space
ducing new standards on the basis of new charging power levels and available charging power. In addition, the use of DC voltage
and country of sale [25–28]. As reported by the International charging devices allows battery chargers to overcome the issues
Energy Agency (IEA) [28], in North America, all car manufacturers related to the difference in AC voltage and frequency values
have adopted the above SAE standard. Asia and Europe are also try- adopted by different countries. For the DC fast-charging mode,
ing to harmonize their standards with North America, but the glo- the SAEJ1772 defines two further levels, which are referred to as
bal standardization challenge is still open due to different energy DC Level 1 and DC Level 2. As reported in Fig. 2, each DC Level is
distribution infrastructures and different choices of various car characterized by a different value of peak charging power. In addi-
manufacturers. A detailed description of all available standards is tion, a new level, called DC Level 3, is in definition with a charging
outside the scope of this manuscript; however, it is interesting to peak power of up to 240 kW [30].
take into consideration some of them in order to understand the A similar classification of charging systems for PEV has been
trend in terms of charging methods, voltage, and power for the also introduced by the International Electrotechnical Commission
next years. (IEC), with IEC 62196 standard, which defines four PEV conductive
As mentioned above, the SAEJ1772 represents one of the most charging types and four connection types on the base of voltage
adopted standards for power converters devoted to PEV charging levels, charging power and charging devices [31,32].
operations. This standard defines the charging power levels, the The rules and modes defined by both SAE and IEC standards for
output voltage, the location (on/off board) and the connectors— AC charging operations can be considered stable and those stan-
with the related communication protocol—between the charging dards have now adopted by many vehicle manufacturers. On the
device and the vehicle [28]. Fig. 2 shows a schematic classification other hand, new standards have recently been developed for DC
of the charging converters in accordance with the standard fast-charging operations. Among those, one of the most adopted
SAEJ1772 and with its possible extension. In particular, Level 1, solution for the DC fast-charging operations is represented by the
highlighted in blue in Fig. 2, refers to slow charging operations, CHAdeMO (CHArge de MOve) Standard. This standard was
which can be carried out at home or in corporate parking-areas designed and proposed in 2010 by the Tokyo Electric Power Com-
with power levels lower than 1.9 kW and a single phase voltage pany (TEPCO), Nissan, Mitsubishi, Fuji Heavy Industries and Toy-
AC source. In this case, an on board power converter, specifically ota. The first CHAdeMO protocol was introduced by TEPCO’s
devoted to low power charging operations of the battery pack, patents, allowing voltage and current values of up to 500 V –
can be used. For higher charging power values, on board weight 125 A. In the year 2016, the CHAdeMO association defined a new
and space constraints do not allow the use of an additional power protocol enabling charging operations with 100 kW of continuous
converter for the charging operations. A possible solution to over- power and 150–250 kW peak power (350 A/500 V). The new charg-
come this issue is the use of the electric drive inverter and AC ing stations will be released in the year 2017. The association has
motor windings for charging operations of the battery pack, as also announced that a new protocol for 350–400 kW peak power
Fig. 1. Schematic overview of available standards for PEV charging operations [29].
Fig. 2. Classification for PEV conductive charging architectures based on the most used standards [19–28,30–38,41].
(350–400 A), working with the output voltage of up to 1 kV, is connectors, used for AC and DC charging operations, are described
under study. The new protocol is announced to be released in in Table 1.
2020 [33]. The CHAdeMO standard is now included in the IEC In particular, the Type 1 and Type 2 Mennekes connectors are
62196 standard as type 4 and, since 2015, into the IEEE Standard now widely used for Level 1 and Level 2 charging operations. The
2030.1.1TM-2015 [34,35]. first connector, specifically designed for DC charging operations,
In 2012, Tesla introduced the first fast DC charger working at has been introduced with the CHAdeMO standard. This connector,
480 V DC with a rated current value of 250 A, resulting in also referred as CHAdeMO connector, is based on two central pins,
250 kW of peak charging power. By using the entire available elec- which allow the electric power exchange between the vehicle and
tric power at the charging station, the PEV battery pack can be the charging unit. The other pins are devoted to communication,
charged in less than 15 min. In this case, the power modulation using the CAN bus protocol between the vehicle and charging sys-
allows the battery to be charged to 100% of its SoC. When multiple tem in order to control safety and main electrical parameters (such
PEV are connected to the same charging station, its control system as battery SoC and PEV charging current) of PEV charging opera-
reduces the peak power of each charging operation. Tesla has also tions [34,35]. The most relevant drawback of the CHAdeMO con-
announced that a new charging station of 350 kW, integrated with nector is related to the fact that this kind of connector is suitable
renewable energy sources, is under study [36,37]. A similar solu- only for DC charging Operations. For this reason, an additional
tion has also been proposed by Audi with a fast DC charger of vehicle inlet is generally required to additionally support Level 1
150 kW peak power, which will be extended to the maximum or Level 2 AC charging operations. As a consequence, the CHAdeMO
power value, allowed by the SAE Combo connector of 350 kW rat- connector has been widely adopted only by Japanese car manufac-
ing power [38]. turers, supported by the large development in Japan of DC fast-
The above-mentioned standards also define the main character- charging infrastructure. In order to solve the main issues related
istics of charging connectors, which represent the power and com- to the CHAdeMO standard, another connector has been introduced
munication interface between the grid and the vehicle on charge. on the basis of collaboration among European and American
These connectors differ from each other with regard to the charg- automotive manufacturers and the Society of Automotive Engi-
ing power level, the voltage source (i.e., AC or DC), communication neers (SAE). In this case, AC and DC pins are integrated into the
protocol and mechanical characteristics. The most common same connector. Thus, AC single phase (AC level 1), AC tri-phase
Table 1
Main AC and DC charging connectors.
Standard IEC 62196 - Type 1 (J1772)

Name Type 1 Connector SAE Combo Connector
Connector picture and scheme
Communication protocol Power-Line Communication (PLC)

Standard IEC 62196 -
Name Type 2 Mennekes Connector

Standard IEC 62196 - (CHAdeMO Japan - IEEE Standard 2030.1.1TM-2015)
Name Type 4 Chademo Connector
Communication protocol CAN bus

Standard Tesla Custom Standard
Name TESLA combo connector
(AC level 2) and DC charging operations can be performed through DC Levels allow higher power charging operations, a lot of studies
the same connector. For this reason, the above connector, which is on AC charging technologies are still in progress. These studies are
reported in Table 1, is also referred to as a SAE combo connector mainly related to innovative power architecture to improve safety
[39]. A custom proprietary connector, referred as TESLA combo for the end users, bidirectional power fluxes management, power
connector and reported in Table 1, has been introduced by TESLA quality and overall system efficiency [39–42].
in the US market for the Model S. This connector can handle both
AC and DC charging operations [36,37]. 3.1. Level 1- on board slow charging devices
As clear from the above considerations, a large variety of con-
nectors adopted by different car manufacturers are now available The Level 1 battery chargers for PEV are designed for low power
especially for DC charging operations. On the other hand, this vari- (lower than 3 kW) charging operations and are generally located
ety limits the opportunity for the end users to recharge their PEV at on-board [19,27]. Their power conversion architecture can be con-
the closest charging station, which should be equipped with the sidered as a double stage converter, which can be also provided
specific connector required for PEV charging operations. A possible with galvanic insulation. The first stage is devoted to the AC/DC
solution would be the use of connector-adapters, integrated with conversion and is based on an active or a passive rectifier with
digital gateways in order to deal with communication and safety Power Factor Correction (PFC) hardware. In both cases, the har-
issues. Anyway, this solution is generally considered too expensive monic impact on the main grid must be contained within the limits
for the vehicle owners. For this reason, various power electronics imposed by the standards IEC 61000-3-2, IEC-1000-3-2, and IEEE-
manufacturers (such as: ABB, IES, Fuji, BTC Power, and Eaton) have 519. The second stage is related to the DC/DC power conversion,
introduced new multi-standard charging stations in the market. In which controls the charging operations of the battery pack. This
this case, a large number of connectors are available on the same last conversion stage, for low power battery chargers, is generally
charging station and the vehicle owner can choose the proper con- provided with galvanic insulation and, depending on the first stage
nector, avoiding the use of adapters [40]. topology, can support bi-directional power fluxes. The most used
configuration presents a first stage, generally based on a classical
3. AC conductive charging technologies diode rectifier, followed by a DC boost converter also used for
PFC. The boost converter can be parallelized and driven in an inter-
As mentioned in the above section, AC conductive charging leaved mode in order to increase the efficiency, reduce the boost
operations involve the use of on board AC/DC converters to inter- inductance and the size of input filters required to control
face the vehicle battery pack with the AC voltage supplied by the Electro-Magnetic Interference (EMI). The power quality and the
main-grid. AC charging modes are used for Level 1 and Level 2 efficiency of the battery charger can be further increased by adopt-
charging operations with a rated power up to 50 kW. Although ing power switches based on silicon carbide (SiC); for example, SiC
Fig. 3. On-board battery chargers: (A) rectifier + pfc + dc/dc converter (B) afe + dc/dc converter (C) rectifier + pfc + resonant converter (D) rectifier + resonant converter + pfc
(E) afe + resonant converter.
MetalOxide-Semiconductor Field-Effect Transistors (MOSFETs) size and weight. The developed prototype of 3.3 kW rated power
[43]. is able to reach an efficiency value of 93%, with a volume of
The paper [44] presents a Level 1 battery charger, which is 5.84 dm3, and a volumetric power density of 0.56 kW/dm3. A sim-
shown in Fig. 3A, whose architecture is based on the cascade con- ilar power conversion architecture, which is shown in Fig. 3D, is
nection of a full wave rectifier, a PFC circuit and a ZVS isolated DC/ described in the paper [47]. The first conversion stage comprises
DC converter. The passive rectifier is used to perform the AC/DC a diode passive rectifier and a series compensated resonant con-
power conversion from the main grid, obtaining the DC voltage verter working at a fixed frequency of 90 kHz. The main advantage
for the second stage. The presence of the PFC circuit guarantees of such converters is that the voltage ratio between the input and
high levels of power quality for the AC side, as required by the the output at the resonant frequency does not depend on the load.
standards. An interleaved PWM is used to increase the overall effi- In fact, in this case, the converter works as a DC transformer. The
ciency, with the advantages of reducing filter dimensions and second stage is a classical PFC boost converter. Experimental
capacitor current values in comparison with traditional architec- results have been reported in the above paper with reference to
ture working at the same switching frequency. The above paper a full-size prototype, with the resonant converter and PFC respec-
also reports experimental results obtained with a high integration tively working at 90 kHz and 45 kHz switching frequency. The
level full-size prototype of 3.3 kW. In this case, the converter is liq- maximum evaluated efficiency is 92.5% with a volumetric and
uid cooled and the maximum evaluated efficiency is 94%. Here, the gravimetric power density of Pv = 0.46 kW/dm3 Pd = 0.45 kW/kg.
volume and the weight of the converter have been respectively Both the above two solutions can be also based on a single phase
reduced by 5.46 dm3 and 6.2 kg, when compared with a traditional AC voltage power supply.
solution. The power densities are Pv = 0.6 kW/dm3 Pd = 0.53 kW/kg. The last architecture analysed in this paper for the Level 1 is
This solution can be also based on a single phase AC voltage power reported in Fig. 3E [48]. The proposed architecture is based on a
supply. single-phase, two-level active front end (AFE) converter for the
A similar solution, which is shown in Fig. 3B, is described by first power conversion stage, and a double compensated DC/DC
Jaganathan et al. in [45]. The first power conversion stage is converter for the second stage. Both converters allow bi-
obtained through a bi-directional NPC active front end, whereas directional power fluxes, useful to provide ancillary services to
the second stage is based on a bi-directional single channel buck the main grid, on the basis of the Vehicle-to-Grid (V2G) paradigm
converter, which is devoted to the control of battery charging oper- [18]. Examples of ancillary services are the active filtering and the
ations. The proposed solution is based on the use of a three-level static compensation (STATCOM). The double compensated config-
converter, which allows reducing the dimension of AC filters with uration of the DC/DC converter allows maximizing the electric
an overall reduction of converter size. The topology analysed in power transfer, as the reactive power of the circuit is largely
[45] requires an external transformer for the insulation between reduced when working at the resonant frequency. Although some
the grid and the EV. Anyway, this issue can be overcame by using experimental validations have been performed, the power densi-
an isolated DC/DC power converter. The above paper is based on ties have not been reported in the above-mentioned paper because
evaluations carried out in a simulated environment without the design procedure is validated only as a prototype and not with
reporting experimental results. its integration. The 3.3 kW prototype can attain a 94.5% efficiency.
The SungKyunKwan University and Samsung R&D team propose Two or more AFE converters can be connected in parallel and
a different topology, which is shown in Fig. 3C [46]. Also, in this driven synchronously. The obtained group of converters can be
case, the first conversion stage is obtained through a diode passive referred to as Synchronous AFE (SAFE) converter. The advantages
rectifier with PFC and an EMI filter. The second stage is a series of the SAFE converter over the AFE consist in the increased effi-
compensated resonant converter, which allows increasing the efficiency, the reduction of filter weight and volume, and the reduc-
ciency and the switching frequency with benefits in terms of the tion of EMI. The papers [49–51] describe the mathematical
modelling procedures and the design criteria to be taken into the shape of the current and increase the transformer utilization.
account to ensure that SAFE converters satisfy the requirements The above paper reports different experimental evaluations, but
and limits imposed by the standards. In addition, the above papers no information about the power density is reported.
evaluate the effects of the number of parallelized AFE converters A comparison of the above-described converter topologies, on
and line inductance on the efficiency, the switching frequency, the basis of their main characteristics and performance, is reported
the nominal power and the power quality. On the basis of these in Table 2, as described in the analysed references. The most inter-
modelling procedures, design criteria and evaluations, the selec- esting topologies are based on isolated solutions, which allow
tion of the best solution for a specific application can be arrived avoiding the use of heavy external transformers for galvanic insu-
at; therefore, the use of slow, iterative procedures can be avoided. lation. The bi-directionality is also an important characteristic,
A different method to reduce the EMI and the power losses is the which is required for the battery charger to support the main grid
use of AFE based on multilevel topologies. with ancillary services. Nowadays, ancillary services are not
The most used topology for the Level 1 isolated DC/DC conver- mandatory and have not yet been regulated by the law of each
sion stage is the Zero Voltage Switching (ZVS) converter. In fact, country; however, when the number of PEV is expected to increase
this solution is characterized by high-efficiency values (up to over the coming years, the use of big storage capacities—repre-
96%) at rated power (typically 3.3 kW) and good controllability. sented by PEV on charge—will be mandatory for grid stability.
The papers [44,52,53] describe the design procedures of the con- For this reason, further developments of the AFE solutions are
verters and the possibility of obtaining soft switching under any expected over the coming years. Further studies are required
load condition. The input and output filters can be further reduced for the solution of row E in Table 2, in order to compare the
by interleaving two ZVS converters as suggested in [54]. Other ben- performance of this battery charger with traditional on-board
efits of the interleaved architecture are the reduction of the reverse topologies.
recovery losses for diodes with an increased efficiency. Another
common type of DC/DC converter for this application is the reso- 3.2. Level 2 – On board slow semi fast charging devices
nant power converter, which works at a high switching frequency
with consequent reductions in weight, volume and switching The Level 2 battery chargers are designed to work with a peak
losses. The resonant converter, at the resonant frequency, works charging power of 19.2 kW, allowing PEV charging operations in
with current in phase with the full bridge voltage. This operating few hours, without the need for external off-board battery charg-
mode presents a great advantage in terms of efficiency, as the ers. This sub-section highlights some of the most interesting
switching losses depend on the product of the voltage with the topologies adopted for semi fast battery chargers.
current; this product is zero or very small at the resonant fre-
quency. One of the most adopted resonant converters for the bat-
tery charger is the LLC topology, also called the series compensated
resonant converter. The voltage ratio between the output and the
input is independent of the load for the LLC converters when work-
ing at the resonant frequency. Thanks to this characteristic, they
can be used at a fixed switching frequency, working as a DC/DC
transformer. The switching frequency is chosen in order to maxi-
mize the efficiency, as reported in [55]. The efficiency can be fur-
ther improved by adopting an asynchronous asymmetrical
driving signal, as studied in [56].
The schemes in Fig. 3 show only the most commonly used
scheme for on-board battery chargers. In recent years, new topolo-
gies have been proposed to increase the efficiency of the DC/DC
converter with a reduced number of active/passive components.
An example, not reported in Fig. 3, is detailed in [57] and reaches
98% peak efficiency. In this case, the proposed topology can be clas-
sified as a hybrid pulse-width-modulation resonant converter. In
fact, the power conversion architecture is based on a resonant con-
verter equipped with auxiliary circuits, which are used to modify Fig. 4. Three-Phase battery charger scheme.
Table 2
Short comparison among Level 1 converter topologies.
Isolated Bi-directional Power supply Dc/Dc converter Performance Reference

A RECTIFIER + PFC + DC/DC CONV YES NO Single phase or Three phase ZVS DAB P = 3.3 kW, g% = 93.6%, [44]
AC Voltage Pv = 0.6 kW/dm3
Pd = 0.53 kW/kg
B AFE + DC/DC CONV NO YES Three phase AC Voltage Buck-boost P = 3.3 kW, g% = 93.6%, [45]
Pv = N.A.
Pd = N.A.
C RECTIFIER + PFC + RESONANT CONV YES NO Single phase or Three phase LLC resonant P = 3.3 kW, g% = 93%, [46]
Pd = N.A.
D RECTIFIER + RESONANT CONV + PFC YES NO Single phase or Three phase LLC resonant P = 3.3 kW, g% = 92.5%, [47,58]
Pd = 0.45 kW/kg
E AFE + RESONANT CONV YES YES Three phase AC Voltage Resonant – SS P = 3.3 kW, g% = 94.5%, [48]
compensation Pv = N.A.
Pd = N.A.
converters working at high frequency that permits achieving better

power quality and reduction of harmonic filters, which in turn lead
to the reduction of heavy magnetic elements. In fact, the use of the
High Frequency (HF) transformer inside the DC/DC converter pre-
cludes the use of a heavy low-frequency transformer. In [58], Chae
et al. propose a 3 kW converter with a resonant converter followed
by a PFC converter for battery charging control. The experimental
results confirm that this solution reduces losses while offering a
high level of power density.
By increasing the charging power, further reduction in weight
and volume can be achieved through the integration of the DC/
DC converter—used for battery management and the regulation
of the DC bus for the electrical drive with an AC/DC converter.
The DC/DC converter can reach an efficiency greater than 94%.
The off-board battery charger can integrate further with the local
battery. The paper [59] proposed a new topology that has three dif-
ferent working phases. The first is the slow AC/DC charging of the
local storage at 1.44 kW. The nominal voltage of the storage is
180 V and is less than the high voltage DC bus inside the PEV,
Fig. 5. Single phase integrated battery charger. assumed for the prototype to as 330 V. The second phase is used
to boost the local storage voltage to feed the HV PEV DC bus with
5 kW peak power. The third phase is introduced for regenerative
One of the most critical targets to be achieved for on-board
charging, when the energy flux is inverted and can reach 5 kW.
Level 2 battery chargers is the reduction of on-board weight and
The provisioning of local storage permits installation of the off-
cost, as these characteristics directly affect the overall vehicle per-
board charger, where only Level 1 is admitted, thereby reducing
formance and cost. This target can be achieved by using the same
the charging time.
hardware for controlling different PEV operations (i.e. driving or
A different approach, which at the moment seems to be one of
charging operations), with different integration levels. In particu-
the most promising solution, is based on the utilization of the elec-
lar, PEV configurations are generally based on a single high voltage
tric machine windings as boost inductances for the electric drive
DC bus, which is used to supply the electric drive and to interface
inverter, which operates as AFE converter. This solution, whose
the battery to this bus with a bi-directional DC/DC converter. For
scheme is reported in Fig. 5, is also referred to as ‘‘combination
low power levels (3.3 kW), PEV battery chargers are based on an
of topology”, because the same hardware is used for different pur-
AC/DC followed by a DC/DC converter for the battery charging con-
poses [60]. The above topology can be adopted for single phase
trol, as shown in Fig. 4. As described in [57], some authors have
integrated battery chargers and the battery can be charged even
studied high efficiency with a reduced number of components in
when the vehicle is stationary. Similar solutions have been pro-
Fig. 6. Single or three phase integrated battery charger [61].

posed by various authors [60,61], but such battery chargers have of the overall system. Although the combination of topology repre-
some drawbacks. In fact, this configuration can operate only with sents a simple architecture, its control differs from the classical AFE
a single-phase AC voltage source, with corresponding limits in control scheme, because this topology must ensure the absence of
PEV charging power. In addition, high power and expensive AC mechanical torque for the motors, especially in the case of Perma-
contactors are required for switching from charging mode to driv- nent Magnets (PM) motors. In this regard, the papers [63,64] pro-
ing mode. Further improvements have been realized with AFE/- pose a detailed study of the mathematical model of the topology of
electric drive supplied by multilevel converters, which allow a Fig. 6, with a control method aimed to avoid generating torque
reduction of the grid interfacing filters, owing to their better per- during charging. The described control method is related to the
formance in terms of power quality [62]. case of a Permanent Magnet Synchronous Motor (PMSM).
The topology described in [60] can be adapted for a three-phase The reduction of the harmonic impact on the main grid, with a
AC voltage source by using two three-phase inverters, as shown in consequent reduction in the AC filter size, can be pursued by
Fig. 6 and proposed by [61]. This solution requires a connection increasing the number of phases of the electric machine. In
with the main grid in the middle of each phase. This way, the AC Fig. 7, the topology related to a nine-phase electric machine is
currents are split into two opposite currents of equal values, ensur- shown as an example of this concept. In this case, it is possible
ing a complete magnetic decoupling between the rotor and the sta- to increase the equivalent switching frequency taking advantage
tor and avoiding undesired torque generation. In addition, the of the interleaving modulation scheme, which simplifies the har-
introduction of a mid-point connection with the inverter reduces monics filtering with power electronics components driven at a
the rated power of AC contactors, increasing the power density low switching frequency. This solution can be used with both the
Fig. 7. Nine phase machine used for Level 2 semi-fast charging and electric drives with reduced EMI.
Fig. 8. Operating modes of a PMSM with split windings: (A) charging, (B) traction.
single and three phase AC voltage source. A detailed study on the The analysed charging devices have reached the maximum effi-
effect of the switching frequency, inductance filter, and number ciency value of 80%, evaluated for a real vehicle equipped with
of parallelized converters on the power quality is reported in 1:1 scale prototypes.
[50]. The papers [65–67] analyse the decoupling transformation Another class of motors, which can be used for PEV, is the
for multiphase machines with the related control. For each system, Switched Reluctance Machine (SRM). This class is a promising
a detailed study is provided and an experimental prototype has alternative to induction and PM machines for vehicle applications.
been used to validate the models and the control logic. This is mainly due to SRM robustness and simplicity of control.
The above described topologies are not provided with galvanic Also, for this kind of machine, the stator winding can be used in
insulation. A method used to introduce the insulation into the combination with the traction inverter to control the charging
combination of topology involves the use of brushless motors with operations of the battery pack [70]. The behavior of the SRM and
split windings, as shown in Fig. 8. Owing to the magnetic coupling of its converter during driving (A) and charging (B) operations is
between inductances, it is possible to insulate the PEV from the reported in Fig. 9.
grid during charging operations. In [68,69], experimental evalua- When the system is in the phase (A), the boost converter is used
tions on this method have been carried out for 20 kW charging sys- to keep the voltage value of the DC bus constant, while the battery
tems. These papers start with an analysis of a detailed pack voltage is changing. During the demagnetization of the coil,
mathematical model, evaluated for the electric machine during the winding energy is recovered and stored directly into the bat-
charging operations. Then, specific attention is devoted to the con- tery pack. This topology also allows regenerative braking opera-
trol system for the grid synchronization and charging operations. tions, which can be obtained by changing the reference values
This control system uses the same current control loop for both and the pulse modulation angle of the control system. The (B)
charging and driving operations. A different control scheme is used phase is used for the charging operations of the battery pack. The
only for the synchronization, voltage balancing and speed control. power source is connected to diodes D1D4, which together work
Fig. 9. Behavior of the SRM + converter during (A) driving operations and (B) charging operations [70].
as a passive rectifier, with the inductances L1 and L2. The switches

Q5 and Q6 are supported by the diodes D6 and De, which—with the
inductance L3—work as a PFC to improve the AC power quality.
Paper [70] reports a detailed description of the control system with
theoretical and experimental evaluations. These evaluations are
related to a prototype which is based on a small 2.3 kW SRM, with
four phases, 8–6, 48 V, and 6000 rpm. In this case, the dependency
of efficiency and the PFC on switching frequency is also evaluated.
The best results for both parameters are obtained with the proto-
type working at a 12.5 kHz switching frequency.
4. DC conductive charging technologies
As mentioned in the previous sections, the above-described

charging systems, classified as Level 1 and Level 2, are suitable
for slow charging operations and limit the use of PEV for applica-
tions where long parking times are expected (i.e. at home and at
work) [71]. Level 3 fast-charging operations represent one of the
most attractive solutions to enable the use of PEV for long distance
travel as well, with the possibility—in combination with the last
Fig. 10. Level 3 fast charging – (A) LF transformer and (B) HF transformer.
battery technologies—of reducing PEV charging times to a few
minutes [72]. The charging power related to Level 3 battery Low Frequency (LF) or High Frequency (HF) transformer, can be
chargers ranges from 20 kW to 250 kW and more (the CHAdeMO considered; see Fig. 10 [72].
Association has announced that a 350 kW fast-charging system is The first scheme in Fig. 10A is based on the use of a low fre-
under study [73]). quency transformer between the grid and the converter, resulting
The selection of the proper power converter topology for fast- in a big, bulky and expensive converter. The switching frequency
charging systems should take into account various requirements, must be maintained low to reduce the high switching and reverse
which have been synthetically reported in Table 3. The first recovery losses. The second scheme, in Fig. 10B, is based on a HF
requirement is related to the galvanic insulation, which guarantees solution, which is realized through an isolated DC/DC converter
safety for vehicle owners during fast-charging operations. A wide working at 8 kHz. In this case, the high switching frequency—in
range of charging voltage should be available at the output of the comparison with the first solution—involves advantages in terms
charging station to interface different kinds of PEV supplied by dif- of size, weight and cost. However, the DC/DC design is strictly
ferent battery technologies. The charging station should also guar- related to the parasitic elements of the transformer and the switch-
antee a limited current and voltage ripple for the PEV battery pack ing losses may limit the overall efficiency. The proposed scheme
to avoid undesired operating conditions, which might negatively allows the parallelization of converters to increase the charging
affect the battery lifespan. Peak Power and charging times are power [72].
strictly related to recent and emerging battery technologies, which The fast-charging systems can be connected to a low- or
can be charged at high charging rates in a relatively short time. As a medium-voltage AC grid. Focusing the attention on the low-
consequence, these two parameters are expected to significantly voltage AC source, the first AC/DC conversion stage can be realized
change in the coming years. Other constraints that must be taken through various topologies of an active front end [75]. The first con-
into account are related to the effects on the distribution system sidered topology is the two-level bidirectional voltage-source-
and transformers, and the dimensioning of cables, circuit breakers, inverter, reported in Fig. 11A. The connection in parallel with such
and fuses [74,72]. a converter—controlled through phase shift PWM and commonly
The drawbacks of Level 3 battery chargers are related to their called Synchronous Active Front End (SAFE)—increases the nominal
size, weight and cost; for these reasons, they are suitable for off- power and reduces the filter weight/size and the switching losses
board installation in a dedicated area, which is generally located with good THD, as described in [50]. The number of switches for
in conventional petrol stations. the SAFE is a function of the number, N, of AFE in parallel, and
can be calculated as 6 N, whereas the number of inductors is 3 N.
Such converters are conveniently used at low voltage, with the
4.1. DC charger topologies for power up to 100 kW
DC bus working under the switch breakdown voltage. This solution
can be also supplied by a medium-voltage AC source through the
As mentioned above, safety requirements impose the use of gal-
use of a bulky LF transformer. Another AC/DC topology is proposed
vanic insulation between PEV and the electric grid. In this regard,
in [76] and is generally referred to as the VIENNA rectifier. The main
two different topologies, which are based on the use of either
advantages of this solution, which is shown in Fig. 11B, consist in
the reduced number of switches. On the other hand, such a con-
Table 3 verter allows only unidirectional power fluxes and for this reason
DC fast charging specs. cannot be involved in V2G operations A different way to have the
Safety Insulation
same benefits of the SAFE converter working at medium or low
AC voltage, with reduced THD, is the use of a Neutral-Point Clamped
Charging voltage Variable between 100–600 V depending
(NPC) converter to feed a bipolar DC bus, with a rated power of up to
on the battery technology
Maximum battery current and <1% current a few MW [77]. The main scheme of an NPC converter is shown in
voltage ripple allowed <5% voltage Fig. 11C, with reference to a three-level architecture. For this topol-
Minimum peak power 125 kW ogy, the use of diodes is required to clamp the switching voltage to
Charging time <10 min to reach the 80% of the battery
half of the DC bus, where two capacitors are connected in an elec-
SoC
Fig. 11. Common AC/DC converters adopted for PEV (A) bi-directional two-level inverter, (B) uni-directional VIENNA rectifier, and (C) bi-directional NPC three-level inverter
[77].
trical series. In this case, the middle point is also referred to as neu- the converter efficiency can be optimized by selecting the right
tral point. With the increasing number of levels, each diode should number of active legs as a function of the load demand, as shown
support higher voltage values. When the voltage required for each in Fig. 12 [78].
diode is greater than its reverse breakdown voltage limit, other
diodes must be added in series. The number of required diodes 4.2. DC charger topologies for power greater than 100 kW
can be calculated as a function of the number of levels N, and is
equal to 2(N-1). For this reason, the NPC topology is not generally High power fast-charging operations are characterized by a rec-
used for high values of DC voltage. ognized impact on the main grid, which may be affected by voltage
For Level 3 battery chargers, after the above described AC/DC instability, increased peak demand, power quality issues, increased
conversion stages, a second stage is obtained through DC/DC con- power losses, transformer heating and overloading conditions
verters devoted to the control of PEV charging operations. [71,79,80]. Some of the above issues could be limited through
Although different DC/DC converters can share the same DC bus, the proper design of the converter topology, filter and control.
each converter can be interfaced with only one PEV on charge, as On the other hand, future issues related to overload conditions—
the output charging current/voltage depend on the status of the owing to simultaneous charging operations of a large number of
PEV battery pack [76]. Adopting the LF transformer solution of PEV—need to be taken into account [71].
Fig. 10A, a non-isolated DC/DC converter can be used, with advan- In fact, the overload conditions related to fast-charging opera-
tages in terms of efficiency, ripple and controllability. In this case tions would require high-power peak and overload capacity for
stations with an experimental validation of the proposed model,

through a 1:1 scale energy storage system.
Paper [90] reports a detailed study of a 220 kW ultra-fast battery
charger, supplied by low AC Voltage at 400 V – 50 Hz, and reported
in Fig. 14. The new scheme differs from the above cited solutions of
Fig. 10 for the introduction of the split local storage, which allows
fast charging operations at high efficiency and with reduced impact
on the main grid. The first conversion stage of this device, reported
in Fig. 14A, is based on a T-type bi-directional AC/DC AFE with a
peak power of 20 kW, which can feed the local battery charger.
The second stage is a 22 kW isolated quasi-resonant double output
DC/DC converter, used for the slow charging operations of the local
storage, as shown in Fig. 14B. The last stage is devoted to PEV fast-
charging operations and is based on a not isolated DC/DC converter
working up to 220 kW with the topology shown in Fig. 14C. The
Fig. 12. Non- isolated multiphase DC/DC converter.
authors have realized a full-size prototype with efficiency values
of 98% and 99.57%, which have been respectively evaluated during
local storage and PEV charging operations. Such fast battery charg-
the main grid, with consequent over-dimensioning of cables, ers are suitable when the number of charging operations during the
power transformers and other components. To mitigate the above day is small. In this case, the ES buffer can be charged in a few hours
issues, some authors have proposed the integration of renewable at Level 1 or 2, whereas the PEV can be charged in few minutes by
sources and energy storage units (batteries, flywheel, etc.) into using the locally stored electric energy.
the charging station architecture, as shown in Fig. 13. This way,
the load represented by PEV on charge can be decoupled from
the main grid, taking advantage of the locally stored/generated 4.3. High power modular multilevel converters for distributed sources
energy [81–84]. This scheme also allows the charging station to integrations
provide ancillary services to the main grid, through the use of bidi-
rectional power converters, with an improvement in the overall The idea of slow buffering using local storage units, and fast PEV
power quality and reliability. Although the most used storage sys- charging operations using the energy stored in the ES buffers can
tems are based on batteries, flywheels and supercapacitors, the be also extended to the case of big charging stations able to feed
Level 3 battery chargers can be also integrated with other storage more PEV at once. With this aim, starting from the generic topol-
units based on different storage principle. Some examples are ogy reported in Fig. 10B, it is possible to develop a specific architec-
given by the pumped hydroelectric, compressed air and hydrogen ture which is based on the direct connection to the AC medium-
storage systems. These alternative storage systems are extensively voltage bus and ES buffer integration. In this architecture, AC/DC
studied and compared in [85]. multilevel converters are connected to medium-voltage AC and
The fast-charging stations play a key role in the smart micro cascaded with isolated DC/DC converters. The outputs of DC/DC
grid, and various studies have been performed to understand the converters are connected in parallel to each other and connected
integration issues, limitations and solutions related to the differ- to the high-power DC bus. The ES buffer and PEV are connected
ences in the communication protocols, baud rate and interfaces to the DC bus through not isolated DC/DC converters. This solution
of each PEV supply equipment [86]. The control of each charging presents a large number of advantages which mainly consist in the
system and its integration can be centralized or distributed. In both reduction of volume, weight and cost. In addition, this topology
cases, different optimization logics can be introduced taking into allows removing the LF transformer and increasing efficiency and
account the SoC of batteries, the power flow exchange between power quality level through the connection in parallel with high-
the grid and the fast-charging station, and the availability of the efficiency DC/DC modular converters with galvanic insulation.
renewable source [87,88]. In this regard, Ding et al. [89] propose The resulting topology is shown in Fig. 15 and is one of the most
a linearized model for optimal sizing and control of fast-charging promising topologies in terms of modularity and scalability
Fig. 13. Classical Level 3 battery charger for EV with single DC bus, local storage, and renewable sources.
Fig. 14. T-type AC/DC converter with split storage and fast DC/DC converter.
because it allows to easily increase the rated power of the charging values required by the vehicles on charge. As an alternative, the
station [91–93]. excess charging power could be supplied by the main grid with dis-
The main limitation of the above power architecture is the use continuous power absorption, which can be partially mitigated
of a single DC bus. In fact, with this architecture, simultaneous PEV through charging/discharging operations of the ES buffer with ben-
fast-charging operations would require the support of a huge efits related to the power quality. A different approach proposed in
amount of power supplied by the ES buffer through its DC/DC con- the scientific literature is based on split battery energy storage sys-
verter. Consequently, the ES buffer power converter should be tems (sBESS). This solution involves the integration of sBESS into a
sized for a discharging power equal to the sum of the peak power multilevel AFE; the electric power is supplied to the PEV on charge
Fig. 15. Insulated multilevel fast charging station with ES buffer and renewable sources.
through small-size isolated DC/DC converters connected in paral- by adopting the DSBC topology, as shown in Fig. 18 and proposed
lel. Each storage unit is directly connected to the output of one in [100], according to the terminology introduced by Akagi [101].
AC/DC cell and provides electrical power through the isolated This topology allows for doubling the AC current with semiconduc-
DC/DC converter. In papers [94,95], the authors analyse the control tors working at half the peak current. The presence of the batteries
of modular multilevel cascade PWM converters with star configu- for each cell is permitted when the number is higher than the num-
ration, using sBESS connected at the output of each AC/DC cell. ber of charging points, to quickly discharge only a fraction of the
These papers describe the model and the control scheme of such energy stored. In this case, the supervisor and the control system
a topology and report experimental evaluations for a 200 V– are required to dynamically change the sBESS in charge/discharge,
10 kW power converter with 3.6 kW h of sBESS. The drawback of thereby reducing the electric power coming from the grid and at
this configuration is mainly related to the control of charging dis- the same time supplying the charging points with the energy
charging operations of the storage units on the base of their main stored in sBESS.
working parameters, such as SoC, lifetime, technology, and tem- The main advantage of using DSBC multilevel converter, in com-
perature . The above drawback can be solved by adding a bi- parison with the DSCC configuration based on chopper cells, is the
directional DC/DC power converter for each storage unit, as shown capability of avoiding DC short circuits, by turning off the cell’s
in Fig. 16. In this regard, paper [96] evaluates star and delta config- switches due to the presence of diodes inside the cells. The paper
urations, focusing attention on the control developed as a function [102] highlights this issue with a formal study and a model of
of the limitations of the converter. This study evaluates the main the DC fault using the half bridge cell. The main drawback is a
grid issues due to voltage asymmetry and verifies the behavior of reduction in the converter efficiency due to the higher number of
the proposed topologies during the grid faults. Paper [97] analyses switches [103]. To overtake the efficiency reduction and maintain
the control of the same power conversion architecture, focusing on the capability to block the DC fault, some authors have proposed
reduction of the effects of the second order voltage harmonics on to add one or more inverter cells into the chain to avoid reverse
the battery pack. The grid power quality can be further improved DC current [104–111].
through the dead time and non-linearities compensation described The connection of the above described cells with the common
in [98]. DC bus is obtained through isolated DC/DC converters, whose out-
The modular converter cell, considered for the above architec- put terminals are connected in parallel. The elementary cell can be
ture, with the isolated DC/DC converter, the storage unit and similar to the cell reported in Fig. 17 or a modified topology adopt-
related bi-directional DC/DC converter, is shown in Fig. 17 [99]. ing, for example, different isolated DC/DC converters. This conver-
The introduction of insulation at the cell level allows for serializ- sion stage is commonly realized through soft switching, owing to
ing/parallelizing cells, thereby satisfying the voltage and power good controllability, low losses, and high switching frequency that
constraints. In fact, the use of distributed storage can support the reduces the volume/weight of the magnetic elements. The paper
grid in reducing the peak absorption of power. [74] shows a multilevel fast battery charger with isolated ZVS con-
The resulting converter can be considered the most general case verters connected at 11 kV on the AC side and can reach a peak
of the family of multilevel sBESS fast-charging stations, because it power of 250 kW. In this case, AC/DC cells are driven at a 500 Hz
allows the optimization of the charging/discharging operations of switching frequency, whereas the DC/DC converters work with a
sBESS and interaction with the main grid. The converter power switching frequency of 10 kHz, with 13 cells for each phase. The
can be further increased without changing the switch technology above paper analyses the mathematical models of the proposed
Fig. 16. Conventional fast battery charging for PEV based on Cascaded H-bridge with integrated energy storage and isolated DC/DC converters.
Fig. 17. State-of-the-art elementary cell with integrated battery charger and ZVS isolated DC/DC converter.
architecture with the related control schemes, describing the the basis of a multi-objective optimization. As an example, the con-
advantage of this topology—in terms of impact on the main grid, trol system can charge the batteries by using the power coming
sBESS management and performance—in comparison with tradi- from renewable sources, when available, and by using the power
tional solutions. In [99], Vasiladiotis et al. report the design proce- coming from the main grid in other cases, balancing the contribu-
dure, the main control schemes and experimental evaluations/ tion to battery charging operation as a function of energy cost and
results for a modular multilevel converter based on the cell of available power. The number of the DC bus can be also extended in
Fig. 17 in its main operative conditions. case of a high number of cells, thereby reducing the power on each
By considering bi-directional isolated DC/DC converters for the single DC bus. In addition, the scheme reported in Fig. 16 and its
cell in Fig. 17, the power architecture reported in Fig. 16 also allows extension on the AC side of Fig. 18 present high levels of modular-
the integration of renewable sources and provides ancillary ser- ity and scalability. The modularity allows the design of a single ele-
vices to the main grid. In this case, the control system can manage mentary cell and the addition of several such cells to accept
the energy exchange between DC bus, sBESS and the main grid on different input voltages. The scalability permits an increase in the
Fig. 18. Basic structure of DSBC/DSCC with common DC bus (A) Chopper cell (DSCC) and (B) Bridge cell (DSBC).
power of the charging station, adding one or more converters and When a cell is in fault state, the AC input terminals are short-
connecting their output in parallel. The parallelization can be done circuited and the voltage level of the faulted cell is shared between
at the component, cell, and converter level. The last solution is the other cells. The availability can be further extended by adding
more reliable because it introduces the redundancy of the plant, many idle cells that are normally switched off with short-circuited
reducing the stop time after a fault occurs, and allows the opti- inputs. In case of faults, those cells can be woken up and used to
mization of the number of cells as a function of the power in order substituted the faulted cells.
to maximize the efficiency. The design can allow that one or more The PEV Level 3 fast charging configuration can be compared
cells are in the fault state before shutting down the converter. each other in a compact tabular form as reported in Table 4.
Table 4
DC fast charging topology comparison.
Converter Grid voltage level Figure Advantages Drawbacks

LF transformer + AFE + not isolated Low voltage Fig. 10A – Simple construction – High dimensions. The insulation
DC/DC converter – High Efficiency is obtained by using a LF
transformer
– Slow charging. The peak power is
limited by the grid
AFE + isolated DC/DC converter Low Voltage Fig. 10B – High Efficiency – Slow charging. The peak power is
– Reduced dimensions limited by the grid
T-type AC/DC converter + isolated Low Voltage Fig. 14 – High Efficiency – The local BS are charged slowly at
DC/DC converter with split – Reduced dimensions level I or II, resulting in a limited
storage + High power DC/DC – Fast Charging – the peak power is not number of fast charging per day
converter limited by the grid
Multilevel converter + isolated DC/DC Medium Voltage Fig. 15 – Direct connection to the MV without LF – The number of simultaneous PEV
converter connected in parallel transformer charging is limited to the ES DC/
+ single DC bus storage + high – Good power quality DC converter
power DC/DC converters for fast – Cannot be optimized the ES charging/ – Oversizing of the
charging discharging
Multilevel converter + isolated DC/DC Medium Voltage Fig. 16 – Direct connection to the MV without LF – Complex hardware with high
converter connected in parallel transformer number of switches and magnetic
+ split storage + high power DC/DC – Good power quality components
converters for fast charging – Allow the optimization of the charging/
discharging of each ES
– Allow simultaneous fast charging. Each
cell works at fraction of the whole power
Double Multilevel converter Medium Voltage Fig. 18 – Direct connection to the MV without LF – Complex hardware with high
+ isolated DC/DC converter transformer number of switches and magnetic
connected in parallel + split components
storage + high power DC/DC – Good power quality – The chopper cell solution (DSCC),
converters for fast charging – Allow the optimization of the charging/ adopted to increase the overall
discharging of each ES efficiency cannot limit the DC
– Allow simultaneous fast charging. Each bus short circuit. Protection cir-
cell works at fraction of the whole power cuits must be added for safety
– Double the nominal power with High
efficiency
From the description of the power architectures reported above, Table 5

it is clear that a strong contribution of energy storage systems is Double sided compensation methods.
required both on converter and vehicle side. For this reason, the PRIMARY
development of new battery technologies represents one of the Series - Parallel -
key elements, which could boost the wide diffusion of Level 3 compensated compensated
fast-charging systems. In this regard, recent scientific literature SECONDARY Series -
presents new alternatives to the currently used lithium-ion batter- compensated
ies. One of the most attractive technology in terms of energy den-
sity comprises lithium-air batteries that can reach the theoretical
Parallel -
limit of an energy density that is 10-times higher in comparison
compensated
with modern lithium-ion batteries. This means that a PEV with a
lithium-air battery pack could reach an autonomy of up to
800 km in full electric mode [112]. This new technology would
make PEV electric driving range comparable with that of tradi-
tional vehicles supplied by gasoline engines. However, commer-
cialization of these new batteries has a long way to go because compensation can be obtained on each side, connecting the capac-
different challenges have to be addressed in terms of reduced life- itors in series or in parallel to the transformer, resulting in four
time and peak power. Both limitations make this solution unusable possible compensation circuits. The transformer, on the primary
for the present EV market [113]. Solid state lithium-ion batteries, side, is typically driven by a full-bridge or a half-bridge converter.
by eliminating the electrolytes, represent an attempt to overcome This choice can be done on the basis of operative power/voltage
the limitations of the lithium-ion batteries in terms of safety. The and technology. The secondary side generally comprises a cascade
new battery can operate like a super capacitor with a full charge/ of a rectifier and a DC/DC converter devoted to battery charging
discharge in few minutes with high current density (18 C), stable operations. The two sides of the RIPT battery charger communicate
cycling, and a wide range of working temperatures between with each other for control and status updates with a wireless
30° and 100 °C. For this reason, Toyota is conducting research communication protocol [117–120].
in this field, but the results are still far short of the target for com- In reference [121], the authors describe one of the first induc-
mercialisation [114]. The Grabat battery is based on a different tive charging systems for EV. This system is based on a split mag-
material, graphene, and support a charging/discharging rate that netic power transformer. The primary side, with the magnetic core,
is 33-times higher in comparison with lithium-ion without mem- is integrated into the charging station; whereas the secondary side
ory effect. The energy density of the Grabat battery is 1 kW h/kg is installed in the connector of the vehicle charging cable. In this
(lithium-ion has a 0.18 kW h/kg energy density). For this reason, way, when the vehicle is on charge, the two sides of the power
it is considered a real alternative to past technology. Some com- transformer are coupled [122]. This solution presents various
mercial graphene batteries are already in the market, and have advantages, such as the reduction of power losses due to the mag-
been available since 2014. There are also tens of studies for the netic flux leakage, the reduction of weight for the on-board mag-
development of next generation batteries, each with advantages/ netic parts (the magnetic core is inside the charging station),
disadvantages and maturity levels, but all research teams have possibility to reach high-power values (up to 100 kW), safety for
the same target: reduce the costs, increase the lifetime, power den- the end users due to the absence of electrical contacts, etc. As a
sity, and the charging/discharging current. matter of fact, this system is not adopted anymore because it
requires a cable connection with no advantage in comparison with
conductive charging systems; however, this system still offers a
5. Wireless charging technologies cheap solution and can guarantee a high level of safety for users.
Nowadays, new WPT solutions are based on charging opera-
Wireless charging technologies allow removing cables between tions, which are carried out through the use of pads, without user
the power source and the vehicle on charge. In fact, in this case, the intervention and with no contact between the charging station and
electric power is transferred via electromagnetic induction. This the vehicle. For this kind of application, the pad design plays a key
charging method presents various advantages, in comparison with role because it widely affects the maximum transferred power and
conductive charging operations. In particular, wireless charging its dependency on the displacement between primary and sec-
technologies are characterized by intrinsic galvanic insulation, ondary pads. Various kinds of pads—which differ from each other
simple and rough structure, charging power from Level 1 to Level on their physical characteristics, geometrical shapes and number
3. The main drawbacks consist in the costs and the need of custom of coils—are presented in the scientific literature [123].
on-board hardware for the PEV. A first classification of pads is reported in [124] and is based on
A large number of wireless charging systems are proposed in the presence or absence of a magnetic core; the related applica-
many scientific papers, on the base of different working principles. tions are respectively referred to as ferrite and air cored pads. From
Anyway, Resonant Inductive Power Transfer (RIPT) represents the an experimental comparison between the two RIPT pads, it has
most widely used technology for wireless charging operations been observed that the air cored devices can work with an air
[115]. Starting from the coupled coils, RIPTs can be modelled as a gap of 200 mm obtaining a coupling coefficient of 0.16 [125],
transformer, but the air gap between the coils reduces the mutual whereas the ferrite cored devices can work with an air gap of
inductance, which results up to 10–20% lower than the self- 70 mm resulting in a coupling coefficient of 0.35 [126]. For the last
inductance. The coupling factor of RIPT chargers is generally about case, it is possible to further increase the coupling coefficient up to
10–20%. To maximize the transferred power, compensation circuits 0.72, by reducing the distance to 6 mm [127]. It is clear that the air
can be introduced on both sides to guarantee that the circuit impe- cored systems are suitable for wide air gaps with low power trans-
dance—at the switching frequency—is represented by a real num- fer. For PEV charging operations, the presence of a core is preferred
ber. In [116], Wang et al. present the compensation circuits due to the possibility of optimising the energy transfer between
shown in Table 5 with the related analytical models. These models the vehicle and the charger; however, in this case, an air-gap
can be used to size the electrical components to obtain maximum reduction of up to a few hundreds of millimetres is required. The
power transfer for different switching frequency and load. The ferrite core pads can be classified on the basis of their shapes. The
first studied and used cores for the IPT transformers were based on the above-described AWC system. Z-Mover reaches 92% efficiency,
E, U and Pot Cores geometries; but these solutions cannot be is light (<5kg), and small (25 25 cm); this permits for installation
adopted for PEV due to their high thicknesses [123]. For this rea- on a wide typology of vehicles. The system also integrates many
son, an alternative solution based on a circular structure, which protections to increase safety, for example, the anti-crushing sys-
is shown in Fig. 19A, has been proposed. In this case, the coil is tem, rain and snow protection, metal detection on the pad, etc. This
mounted on a ferrite plate, whereas an aluminium shield can be new concept in charging systems presents a better experience
placed on the opposite side of the ferrite plate. The main issue with using the EV, reducing the need for human work to charge the sys-
such cores is related to the large reduction of overall efficiency and tem. Another opportunity given by the WPT is the ability to charge
output power when the misalignment increases. In particular, the while the vehicle is moving, also referred to as dynamic electric
output power may fall to zero for misalignment values greater than vehicle charging (DEVC) [136] or charging while driving (CWD)
40% of the pad size. [128]. The misalignment capability can be [137], as opposed to the Static WPT (SWPT) described above. A
extended with the Double D Quadrature shape (DDQ), reported power transmitter embedded in the roads would feed one or more
in Fig. 19B, which allows a maximum power 2 time higher than pick-ups. This solution allows for a reduction of the energy stored
the circular structure, whereas the transferred power is reduced on board or for an increase in the vehicle all electric range. The
up to 1/6 of its maximum value, for misalignments of 100% along drawback of this solution is in the use of on-board dedicated con-
one direction [129]. verters, which is based on a different hardware of the Level 2
The misalignment performance of charging devices can be fur- charging converters. The scheme of this solution is shown in
ther extended by adopting the so called two-coil or four-coil sys- Fig. 20 [116,123].
tems [112]. According to theoretical and experimental results, Another issue of the distributed systems of Fig. 20 consists in
the two-coil system gives a better performance with a small air- the effects of the Electro Magnetic Field (EMF) on humans and ani-
gap that is typically under 300 mm [130]; the four-coil system mals. Paper [138] proposed a solution to this problem, introducing
gives better results for mid-range applications [131–133]. For power line segmentation with EMF shields. In particular, the
PEV applications, the two-coil systems are generally preferred described solution is based on a dynamic control system, which
because the distance between the vehicle and the ground is low. activates only the pad in correspondence to the vehicle and turns
In [134], Choi et al. have shown—through analytical and experi- off the remaining pads on the power line. This way, the EMF is at
mental evaluations—that the use of custom designed pads allows its maximum value only in the air-gap between the vehicle and
accepting large misalignments (100% of lateral and 50% of longitu- the road, whereas its influence can be negligible in the narrow
dinal misalignment), without a decrease in the maximum trans- areas. A prototype of the above proposed system has been set-up
ferred power. The prototype was designed for 15 kW peak power, and tested for the case study of an electric bus. In this case, a
15 kHz switching frequency, and a 15 cm air gap. The experimental 100 kW peak power charging system has been considered with a
results gave, at 5 kW, an EMF equal to 6.1 µT. resonance frequency of 20 kHz and an air-gap, between the vehicle
Audi [38] adopted a different Wireless Power Transfer (WPT) and the road, of 26 cm. The maximum efficiency value of 80% has
solution called Audi Wireless Charging (AWC). The primary coil been experimentally assessed for this prototype.
with the inverter is installed on the floor, and the secondary coil The DEVC system is not ready for the market due to technical
is connected to the onboard battery charger under the chassis plat- and standardization limits, but gives some opportunities to the
form of the car. When the car is parked, the onboard computer users and to the grid; for this reason, it is predictable that in the
helps the driver (in the future this function will be done automat- future the SWPT will use the same equipment as the DEVC, giving
ically) to align the two coils. The system moves the primary coil up, users many charging opportunities.
close to the secondary and starts the charging process. The first Another issue to be taken into account for DEVC charging sys-
generation of AWC has a nominal charging power of 3.6 kW, but tems is their integration with the existing roads [139]. In fact, pre-
the carmaker is planning to increase the power up to 11 kW in liminary experimental tests on road prototypes result in
the next release. The idea is to adopt this charging station at home construction difficulties and cracks. In addition, the literature is
or work as an alternative to the classical Level 1 and Level 2 con- still lacking of theoretical and thermal analysis related to the
ductive charging stations. A similar solution has been proposed behavior of these systems, working operative conditions represen-
by the PRIMOVEÓ department of the Bombardier group, with its tative of their behavior under the roads. Recent experimental eval-
Z-Mover [135]. The power and the working idea is the same as uations have been performed for SWPT systems. In this case, the
Fig. 19. Pad Circular (A) and DDQ (B) structure.

Fig. 20. Dynamic electric vehicle charging (DEVC) used to feed the moving vehicle from under asphalt, through water, ice, and snow [112,116,123].
installation of these devices requires excavation of the road line,

the installation of ferrites and cables and the sealing of coils with
cement before covering the road with asphalt [140–142].
By adding bi-directionality to the DEVC system, it is possible to
use the on-board battery pack to support the main grid during
simultaneous PEV charging operations. In paper [58], authors have
experimentally validated bi-directional systems with a 100 W low
power prototype working at 40 kHz and 15 cm of air gap between
the coils. The work also gave an accurate model of the system com-
prehensive of the battery model that permits analytical calcula-
tions of the power flow of the WPT and simulates the interaction
Fig. 21. Main topics covered by international pilot projects.
with the grid. The bi-directionality added to the DEVC crates a
new scenario for energy management. Assuming that all the PEV
in a place are connected with a bi-directional DEVC, the local grid integration in a smart grid concept, through the use of ICT devices.
supervisor can decide to use the power stored in some vehicles to The study and experimentation of smart charging strategies are
help others that are too discharged to arrive at a feeder. Another referred to the smart integration of a large number of PEV with
application is to give the surplus of power during braking or a long the main grid. Pilot projects in this field can take advantage of col-
slope back to the grid. This new concept implies that the power lection and analysis of data related to driving and charging habits
absorbed by the grid after at the start-up of the system is equal of a large number of vehicle owners. In this case, vehicle owners
to the losses or, in any case, less than the total power. At the could be encouraged in using PEV through economic and social
moment, the literature is still lacking of specific studies with benefits. Finally, the development of both on-board and off-board
numerical examples to understand these improvements; however, new charging devices could help reducing PEV charging times
according to the recycling principle adopted from railways, where and costs, with particular reference to wireless charging
this idea has been working for years, this opportunity can be con- operations.
sidered sustainable also for PEV.
6.1. Pilot projects on charging infrastructure
6. Projects and applications of PEV charging technologies
The first set of research projects is aimed to the development of
Recent developments in energy storage and power electronics charging infrastructures supporting the large scale adoption of
technologies have supported the increasing diffusion of PEV in PEV. In this regard, relevant results have been obtained with EV
the automotive sector. Unfortunately, the market share of these and ChargePoint America Projects, which have been funded by
vehicles has been limited through the years by various barriers, the U.S. Department of Energy [143]. The combination of the above
which require economic, scientific, and technical efforts to be com- two projects has allowed for the development of a large PEV charg-
pletely overcome. In fact, the reduced electric range of PEV still ing infrastructure, involving 22 regions of the U.S. During the
represents a limitation for a bulk of the users, who are encouraged research activities, the charging infrastructure has been extended,
to use the electric propulsion only for short distance travel that is reaching 17,000 AC Level 2 charging stations and about 100 DC
generally limited to urban areas. In addition, the costs of PEV are Level 3 fast charging stations. The above projects have been sup-
still very high when compared to the cost of conventional vehicles ported with data coming from 300 Smart For-Two vehicles
based on internal combustion engines. As a consequence, govern- involved in the Car2Go car-sharing service and from more than
ment grants, which could incentivize users and manufacturers 8000 private owners of the Nissan Leaf and Chevrolet Volts
with socio-economic advantages, play an important role in sup- [144]. The collection of data related to vehicles and charging infras-
porting the spread of sustainable mobility. tructure under study have allowed for obtaining a comprehensive
Various pilot projects, which have been funded by either gov- analysis over a period of three years related to PEV driving pat-
ernment or industrial companies, have been carried out all over terns, frequency of Level 2 and Level 3 charging operations, and
the word. As synthetically reported in Fig. 21, these projects aim power requirements for the main grid during simultaneous charg-
to support the diffusion of PEV at different hierarchical levels. ing operations. Similar objectives are expected over the next com-
In particular, the first step is represented by the development of ing years for the DRIVETHEARC project, which has been funded by
charging infrastructure, which can be obtained by increasing the the Japan New Energy and Industrial Technology Development
number of Level 2 and Level 3 charging points and supporting their Organization (NEDO), as part of the agreement between NEDO
and the State of California Governor’s Office of Business and Eco- related to driving and charging behavior of each vehicle owner
nomic Development [145]. The main activities of this project are have been acquired and analysed to preliminarily evaluate the
aimed at supporting and promoting the use of electric vehicles effect of different PEV charging/discharging control methods. In
for long distance travel, along the southwest side of California this context, the Fraunhofer Institute for System and Innovation
[146]. For this reason, this area will be provided, with a charging Research ISI, in collaboration with other partners of iZEUS project,
infrastructure based on 50 DC Level 3 fast charging stations has developed a prototype of a 20 kW bi-directional charging
[147]. During the above project, PEV owners will be provided with device that can monitor the grid status and—on the basis of the
a navigation app, to collect important data related to their charging information received (i.e., frequency and voltage at the PEV’s grid
habits and driving patterns [145]. connection point)—manage charging and vehicle-to-grid opera-
REPLICATE (Renaissance of Places with Innovative Citizenship tions. In addition, the on-board integration of the bi-directional
and Technology) project, which has been co-funded by the Horizon charging equipment has been performed for the specific case of
2020 Framework Programme of the European Union, is also the Opel Meriva in its electric version [6,150].
devoted to the development of charging infrastructures in a smart
city context [148]. In particular, the main objective of this project is
the development of integrated energy, mobility, and ICT solutions 6.3. Pilot projects on charging devices
in urban areas. This project involves various international indus-
trial partners and the cities of Florence (Italy), Bristol (UK), and The final set of research projects is devoted to the development
San Sebastian (Spain) with the installation of about 250 Level 2 of charging system. In this context, the great part of pilot projects is
and Level 3 charging stations in the areas of interest. In addition, devoted to the feasibility analysis and development of wireless PEV
navigation apps, supported by public high-speed wireless mobile charging systems. Among those, relevant results have been
networks, will be developed by ICT research groups. This activity obtained with the UNPLUGGED (Wireless Charging for Electric
will make it possible to acquire and monitor driving and charging Vehicles) project, which has been funded by the European Union’s
data related to a vehicle fleet composed of 27 electric cars, Seventh Framework Programme for research, technological devel-
4 e-buses, 112 e-taxis, and 32 e-bikes. opment, and demonstration. The main objective of this project is
the evaluation of technical feasibility, issues, interoperability, and
6.2. Pilot projects on smart charging socio-economic impacts of inductive charging systems for PEV,
with particular focus on dynamic wireless charging en-route.
Another set of projects is focused on the evaluation of PEV Two wireless charging system prototypes, with a rated power of
charging/discharging strategies, which aim to reduce the impact 3.7 kW and 50 kW, have been developed during the research activ-
on the main grid related to a large diffusion of PEV and to satisfy ities in this project. These levels of electric power have been chosen
vehicle owners’ needs in terms of charging time, cost, and battery as respectively representative of small passenger cars and urban
durability. Particular relevance on the above aspects is given in the commercial vehicle applications. The prototypes are equipped with
framework of the iZEUS (Intelligent Zero Emission Urban System) interoperable communication modules to support the smart inte-
Project [149]. This project has been funded by the German Federal gration of vehicles with the main grid. In particular, the 50 kW
Minister for Economic Affairs and Energy in the context of the charger is based on an innovative multilevel architecture, which
funding programme ‘‘Information and Communication Technolo- allows the PEV to be recharged with a voltage ranging from
gies for Electric Mobility II—Smart Car—Smart Grid—Smart Traffic.” 200 V to 750 V. The block diagram of the above 50 kW battery
The main objective of the iZEUS project is related to the develop- charger is shown in Fig. 22 [151].
ment of exemplary applications supporting the adoption of electric On the primary side, the battery charger is connected to the AC
sustainable mobility for private and urban commercial use. These three-phase main grid through the AC/DC grid side converter. This
objectives are reached through the investigation and analysis of converter is realized as a controlled multilevel rectifier, which is
control strategies, which optimize the integration of a large num- based on the NPC three-level architecture shown above in
ber of PEV and renewable energy sources in a smart grid scenario. Fig. 11C. The realized DC stage has, with adjustable DC voltage val-
The evaluations carried out in this project are based on data com- ues, two 25 kW resonant power converters working in the Zero
ing from a test fleet comprising more than 60 electric and plug-in Current Switching (ZCS) mode. Each resonant converter feeds the
vehicles, which have been provided by Daimler AG, Opel AG, and primary coils with a square voltage wave. On the secondary side,
Toyota Motor Corporation. In addition, the test fleet has been sup- the induced AC voltage is rectified and filtered through passive
ported with a public charging infrastructure, developed in previous components to perform the charging operations of the PEV battery
and parallel projects, of more than 600 Level 2 charging stations, pack. It is clear that this configuration involves only passive com-
located in the area of Baden-Wuerttemberg. Many users have been ponents on the secondary side of the inductive charger. Conse-
incentivized to participate in the research project, with economic quently, the charging operations are completely managed
advantages on battery leasing rates and with tablet computers through the control of the power converter located on the primary
having a navigation applet installed on-board. In this way, real data side [151,152].
Fig. 22. Block diagram of the 50 kW inductive battery charger realized for the UNPLUGGED project [151].
Similar topics have been addressed in the FABRIC (feasibility into account that they can be cooled during the night, when the
analysis and development of on road charging solutions for future power demand for the main grid is low [155]. On the other hand,
electric vehicles) project. This project has also been funded by the statistical analyses, conducted by Progress Energy INC, have shown
European Union’s Seventh Framework Programme for research, that the great part of PEV owners have the habit to put their vehi-
technological development, and demonstration. The main objec- cle on charge at the end of their working day. Therefore, the night-
tive of the FABRIC project is related to the reduction of ‘‘range anx- time charging operations would require an expensive upgrade of
iety” caused by the limited autonomy of electric vehicles. In order the pre-existing distribution systems. The use of active Level 1 bat-
to pursue this objective, fast charge, battery swap, and advanced tery chargers would enable smart charging operations supervised
on-road charging solutions are being analysed to extend the by external control units, which operate at different hierarchical
expected autonomy of future electric vehicles. In particular, during levels and manage electric power flows in order to limit the peak
the FABRIC project, three dynamic inductive charging pad proto- power requirements, reduce hot spots and avoid instability condi-
types were developed. These prototypes are characterized by a tions for the main grid. This scenario, which is characterized by the
rated power of about 20 kW. Both on-board and off-board proto- management of one-way power flows, is referred to as Grid to
types have been realized and are currently being replicated. The Vehicle (G2V).
three prototypes present different physical dimensions, electrical In this regard a lot of control schemes based on multi-agent
and electronic design, and operating air gaps between the road power and communication architecture are reported in the scien-
and the vehicles. Supporting navigation apps have X also interested tific literature [155,156]. In particular, the evaluations reported
developed to help drivers in following optimal coil alignment dur- in these papers result in a reduction of about 30% in peak power
ing dynamic charging operations [153,154]. requirement from the main grid, which can be obtained with a
smart charging G2V scenario. The main barrier for the wide spread
of G2V logic is related to the absence of communication standards
7. Current and future trends of PEV charging architectures for Level 1. For this reason, a unified approach based on mobile
protocols (e.g. 3G, 4G, etc.) could be proposed for the communica-
As clear from the analyses reported in this paper, the develop- tion among the supervising control systems and vehicles in charge.
ment of a proper charging infrastructure, in combination with next With this approach, each vehicle, identified according to its geo-
generation energy storage technologies, can support the large pro- graphical location and local communication network, can commu-
liferation of PEV, which is expected for the next decades. As matter nicate with the supervisor and receive information about the
of fact, Level 1 and Level 2 charging devices represent the most maximum power level accepted, on the base of vehicle state of
convenient solutions for an urban use of PEV, as these solutions charge and main grid status. Geographical location of the vehicle
are characterized by low power requirements from the main grid, can be carried out with GPS or similar systems (such as Galileo,
reduced costs and easy availability. In particular, Level 1 charging Glonass, BeiDou, etc. . .), which also provide reduced synchroniza-
operations can be easily obtained at home and the technology tion times of about 10 ls. In this way, mobile communication pro-
behind such on-board battery chargers is characterized by high tocols will allow all PEV to be integrated with G2V without the
reliability and safety for the final user. Future trends for Level 1 definition of an integrated shared protocol for the battery chargers.
chargers are mainly related to wireless charging operations, which Level 2 battery chargers involve higher charging power values,
have been already described in the previous sections. In fact, the in comparison with Level 1, but also higher initial costs are
absence of electrical contacts, which allows the automatic charging required. In this regard, government funding and pilot projects,
operations to be carried out without human intervention, would supporting corporate electric fleets or car-sharing services could
make the home charging systems more user friendly by reducing allow a wide diffusion of Level 2 charging systems, especially for
safety risks for PEV owners during charging operations. In this private or public parking areas. In this context, the example of
regard, Audi and Bombardier have proposed suitable solutions rep- the iZEUS project gives useful results on driving and charging
resented by the AWC and Z-Mover systems, which have been behavior of a great number of PEV owners, which interact with
already described in the specific section related to wireless charg- the main grid through smart charging/discharging strategies, sup-
ing technologies. Although wireless charging represents a very ported by an existing Level 2 charging infrastructure.
attractive solution, its technology development is still at an early The high power values exchanged between the PEV and the
stage due to various technical bottlenecks. The first one is the low- main grid justify the use of charging devices able to manage also
est efficiency in comparison with the conductive charging systems. bi-directional power flows. In this case, the electric energy, which
For low power charging operations this drawback can be neglected, is stored in the vehicle battery pack, can be fed back towards the
but a large use of wireless devices involves extra load for the net- main grid through the Vehicle to Grid (V2G) operative scheme. In
work, which is required to compensate the efficiency losses. A sec- this regard, the new updated version of standard J1772 has been
ond disadvantage is the absence of international standards in released by SAE in October 2012, with new guidelines related to
terms of operating frequency, communication protocols between PEV integration and V2G in the smart grid scenario [18]. V2G oper-
charging station and PEV, pad geometry, etc. Further enhance- ations allow PEV to support the main grid with ancillary services
ments for Level 1 conductive charging systems can be achieved able to perform power factor correction, active filtering, voltage
by increasing the smart integration of PEV with the main grid. In drops compensation and network frequency correction. In addi-
fact, the use of Level 1 charging systems is expected to continue tion, the high amount of energy stored in PEV battery packs can
also for future applications as it gives the possibility of recharging be used to support the main grid during peak power demand
the vehicle even when no public charging stations are available phases. The proper management of V2G operations represent an
and without installing expensive home power converters. Nowa- open challenge, as the requirements of both Grid System Operator
days, the great part of Level 1 PEV battery chargers operate as a (GSO) and vehicle owners need to be taken into account. In partic-
passive load, without on board intelligence for their smart integra- ular, from the operator point of view, the availability and reliability
tion with the main network. In this case, although the maximum of high power V2G services are considered as fundamental features
charging power for each vehicle is low (limited to 3.3 kW for Level to be satisfied. On the other hand, the engagement of PEV in V2G
1), simultaneous charging operations of a large number of PEV may operations involves an increase in vehicle initial cost, related to
involve peak power demand, which would be too high for the main the use of bidirectional battery chargers, and a high number of
grid. In addition, the existing power transformers are sized taking PEV battery pack charging/discharging operations, which strongly
coming from each vehicle in charge, with positive effect on battery

durability and PEV owners’ acceptance of V2G operations. Vehicle
owners could be also encouraged in participating V2G operations,
through economic benefits related to their participation in energy
and ancillary services market charging operations. It is clear that
with the proposed architecture, the control of V2G operations
should be carried out by taking into account the different commu-
nication protocols between vehicles and battery charging devices
[157,158].
The use of electric vehicles also for long travel distances could
be enabled by a capillary diffusion of a new generation Level 3
fast-charging architecture, which would allow fuel-stop equivalent
charging times. The development of Level 3 charging infrastructure
has been widely supported by EV, ChargePoint America and DRI-
VETHEARC projects with the installation of more than 150 new fast
charging stations. Future challenges are related to the reduction of
PEV charging times, by increasing the peak power of the DC fast-
charging devices. In this regard, as clearly shown in Fig. 24, a con-
tinuous growth in peak power for DC Chargers has been observed
in the last years, with an increasing rate of more than 30 kW/year
for the last decade [19–28,30–39]. On the basis of the observed
rate, and according to the expected development of future battery
Fig. 23. V2G aggregative architecture. technologies, a peak power of 1 MW could be reachable in 2040.
The above scenario needs to be taken into account in the choice
of proper topologies for DC chargers. In particular, scalable archi-
affect the storage system durability. In addition, the electric power tectures should be preferred, as they allow to increase the of a
values supplied by single vehicles during V2G operations are often charging station by adding new modules, without any change on
negligible, in comparison with the requirements of the main elec- the charging devices already in use. Another issue to be seriously
tric grid. taken into account for future development of DC chargers is the
The above issues can be faced, in the near future, through the impact of simultaneous PEV fast-charging operations on the main
use of an aggregative indirect V2G architecture, which is proposed electric grid. In fact, various PEV market penetration scenarios
and described in [18]. With this power and communication archi- [159–161] suggest that the development of charging infrastruc-
tecture, a large number of vehicles in charge are grouped and man- tures and battery technology could enable a wide diffusion of
aged as Virtual Power Plants (VPP), through the supervision of PEV in the next years, with a market share up to 40% before the
external control entities, which are called aggregators. The main year 2030. It is clear that the main electric grid, in its present con-
scheme of a V2G aggregative architecture is reported in Fig. 23. figuration, would be directly affected by this increase in electric
As shown in the above figure, each aggregator receives from the power requirements involving high-power peak and overload
GSO specific requests for supplying ancillary services to the main capacity, with the consequent over-dimensioning of cables, power
grid. Those requests can be satisfied, through the management of transformers and other components.
power fluxes coming from a large number of PEV in charge, dis- The main research challenge is the resolution of the above
tributed energy generation systems and stationary storage units, issues in a smart grid scenario—where the existing electric power
which are grouped under the same aggregator. In this way, the system is supported—through the supervision of ICT devices by
aggregator can support the main grid with high power ancillary distributed generating units which are based on renewable energy
services at any time, by using a small amount of electric power sources, stationary energy storage systems and electric/hybrid
Fig. 24. Increase of the peak power per year for fast DC battery chargers [19–28,30–38,41].
mobility [18]. In this regard, power distribution companies are

beginning the integration process of storage units into the main
grid. An example is represented by the development plan launched
by the Italian company TERNA in 2011, for the installation of three
large storage units in the South of Italy. Those storage units should
be able to optimize the integration of renewable sources with the
main grid, with a rated storage capacity of about 38 MW h [162].
Further developments in power architecture for level 3 DC battery
chargers are expected to perform the integration of local storage
units through the topologies based on modular multilevel convert-
ers, which are shown in Section 4. In addition, new energy storage
technologies characterized by high charging/discharging rate capa-
bility and increased durability are expected for the next years.
Those technologies will allow vehicle charging time up to 10 min
without relevant reductions in battery lifecycle. In this regard,
lithium-titanate batteries represent a promising technologies, even
though, at the best of our knowledge, they are still affected by
energy density issues [163].
Wireless charging systems will be used in the future as a valu-
able support for the conductive charging infrastructure. In particu-
lar, the DEVC charging systems represent an attractive solution to
increase the full electric range of PEV along specific roads. DEVC
technology could also allow the energy sharing among PEV, on
the base of the Vehicle to Vehicle (V2V) paradigm. V2V operations Fig. 25. Functional scheme of V2V operations.
are based on bi-directional exchange of electric power among vehi-
cles, with positive effects on the electric power supplied by the
main grid. A simplified scheme representing the V2V operations Table 6
is shown in Fig. 25. open issues regarding PEV charging architecture.
The vehicles represented in the above pictures are characterized Technical issues Possible solutions
by different state of charge and operative conditions. In particular, High number of connectors Multi-standard charging stations
the vehicles reported in green are fully charged and do not require and plugs PEV charging adapters
external power contribution from the main grid. During their brak- Low vehicle autonomy New battery technologies with
ing operations the kinetic energy of the vehicle can be used to high energy density
recharge the red vehicles, whose battery pack presents low values Wireless dynamic electric vehicle
of state of charge. This scheme allows to perform roadside assis- charging (DEVC)
tance services where a fully charged vehicle can help stalled vehi- Range Anxiety Public funding and pilot projects
cles along the road, with the energy coming from its battery pack, for the installation of a large
number of DC - fast charging
to reach the next charging station. The above operations can be
stations
managed through an external V2V power exchange control system, New battery technologies with
which is based on mobile communication protocols [164]. high charging rate
A lot of challenges on wireless charging systems are still opened Peak power demand due to a high Dynamic charging power limita-
and require efforts from both corporate and scientific sectors. One number of simultaneous PEV tion through G2V services
of those is represented by the integration of DEVC with the existing charging operations may cause Ancillary services for the main
grid instability and power line grid provided by PEV in charge
road systems. In particular, the electromagnetic losses caused by
overloads through V2G operations
the road surface materials have shown a relevant increase in wet Energy sharing between PEV to
weather conditions. In many countries, these conditions are gener- reduce peak power demand for
ally considered frequent for various months of the year and, for the main grid (V2V)
this reason, the related losses, during charging operations, cannot Integration of stationary energy
storage systems and distributed
be neglected [165]. The indirect costs of an e-road, which are due
generation units with the main
to surface quality and underground hydraulic/electrical system grid
maintenance also represent an important aspect to be properly
evaluated during the designing phase. The above issues can make
DEVC technology unattractive in terms of costs, since several inter- additional technology improvements in the energy storage or
ventions could be required to meet all the needs of an urban area. power electronics sector; other issues, such as the development
Further studies are more focused on the performance improve- of a proper charging infrastructure, are related to political inter-
ment of charging devices in terms of efficiency and allowed vention through public funding and pilot projects.
misalignments. In this regard, the prototypes realized in the con-
text of UNPLUGGED and FABRIC projects represent an important
step towards the development of on-road charging solutions. 8. Conclusions
Future pilot projects are expected to support technology advance-
ments for these devices in terms of charging power, efficiency, A comprehensive overview about innovative PEV charging
safety and integration with the road infrastructures. architectures, devoted to the integration of sustainable mobility
In order to draw the conclusions of this review, Table 6 reports and distributed energy sources in the smart grid scenario, has been
the main issues regarding PEV charging architecture which still presented in this paper.
have to find a definitive implementation. In particular, some of This overview has been preliminary focused on the classifica-
these issues, reported in that table, can be addressed through tions and standards related to conductive PEV charging devices,
with specific details on charging connectors and power levels. [10] Sabri MFM, Danapalasingam KA, Rahmat MF. A review on hybrid electric
vehicles architecture and energy management strategies. Renew Sust Energy
Then, traditional and more innovative charging architectures have
Rev 2016;53:1433–42.
been described, with reference to both Level 1 and Level 2 on- [11] Chau KT, Wong YS. Overview of power management in hybrid electric
board conductive charging systems. In addition, high-power con- vehicles. Energy Convers Manage 2002;43(15):1953–68.
ductive charging systems, based on AC and DC micro-grid power [12] Fergus JW. Recent developments in cathode materials for lithium ion
batteries. J Power Sources 2010;195(4):939–54.
architectures have been analysed and compared with specific ref- [13] Capasso C, Veneri O. Experimental analysis on the performance of lithium
erence to Level 3. In this case the paper gives wide evidence on based batteries for road full electric and hybrid vehicles. Appl Energy
how Level 3 represents a valid solution, in terms of efficiency, mod- 2014;136(31):921–30.
[14] Pearre NS, Kempton W, Guensler RL, Elango VV. Electric vehicles: How much
ularity and charging times, in order to really support the smart range is required for a day’s driving? Transport Res C-Emer 2011;19
integration of PEV with distributed energy sources and stationary (6):1171–84.
energy storage systems. [15] Wang TG et al. Path-constrained traffic assignment: A trip chain analysis
under range anxiety. Transport Res C-Emer 2016;68:447–61.
A dedicated section of this paper has also analysed the charging [16] García-Villalobosa J et al. Plug-in electric vehicles in electric distribution
architectures that are based on a wireless connection between the networks: A review of smart charging approaches. Renew Sust Energy Rev
vehicles under charge and the main electric grid. For this technol- 2014;38:717–31.
[17] Zhang L, Jabbari F, Brown T, Samuelsen S. Coordinating plug-in electric
ogy a detailed description of the power architectures, able to sup- vehicle charging with electric grid: Valley filling and target load following. J
port both stationary and mobile inductive charging systems, is Power Sources 2014;267:584–97.
reported in this paper. Anyway, this charging solution still presents [18] Mwasilu F et al. Electric vehicles and smart grid interaction: A review on
vehicle to grid and renewable energy sources integration. Renew Sust Energy
few issues, and for this reason the inductive charging has to be
Rev 2014;34:501–16.
considered as a valid solution only available in the near future. [19] Haghbin S. Grid-connected integrated battery chargers in vehicle
The last section of the paper reports information about the most applications: Review and new solution. IEEE Trans Ind Electron 2013;60
relevant pilot projects, which have recently experimented and (2):459–73.
[20] Yilmaz M, Kreint P. Review of battery charger topologies, charging power
tested some of the above innovative PEV charging technologies levels, and infrastructure for plug-in electric and hybrid vehicles. IEEE Trans
supported by the development of the related infrastructure and Power Electr 2013;28(5):2151–69.
management strategies. These projects have been described in [21] Quinn C, Zimmerle D, Bradley TH. The effect of communication architecture
on the availability, reliability, and economics of plug-in hybrid electric
terms of main results, limitations and possibilities for their large vehicle-to-grid ancillary services. J Power Sources 2010;195(5):1500–9.
scale applications. [22] Pollet BG, Staffell I, Shang JL. Current status of hybrid, battery and fuel cell
Finally, this overview brings to the main conclusions that electric vehicles: From electrochemistry to market prospects. Electrochim
Acta 2012;84:235–49.
national or regional charging infrastructures have to be conceived [23] Bi Z, Song L, De Kleine R, Mi CC, Keoleian GA. Plug-in vs. wireless charging:
and designed, in the prospective of the long awaited sustainable Life cycle energy and greenhouse gas emissions for an electric bus system.
mobility, taking into account the technical characteristics and Appl Energy 2015;146:11–9.
[24] Rawson M, Kateley S. Electric vehicle charging equipment design and health
advantages of both innovative power conversion architectures and safety codes. SAE Technical Paper 1999; No. 1999-01-2941.
and stationary energy storage systems. In fact, only in this way [25] SAE Electric Vehicle and Plug-in Hybrid Electric Vehicle Conductive Charge
the smart grid based sustainable mobility can take real advantage Coupler, SAE Standard J1772; 2010.
[26] Bohn T. Plug-in Electric (PEV) Standards, Upcoming PEV/Features, charging
of the charging infrastructure and relative integration with clean
system overview, U.S Dept. of Energy, Clean Cities Coalition, Golden, CO.
energy generation systems. [Online]. Available: http://www1.eere.energy.gov/cleancities/.
[27] Khaligh A, Dusmez S. Comprehensive topological analysis of conductive and
inductive charging solutions for plug-in electric vehicles. IEEE Trans Veh
Acknowledgements Technol 2012;61(8):3475–89.
[28] http://www.ieahev.org/about-the-technologies/charging-equipment/.
[29] Soylu S, editor. Electric Vehicles: The Benefits and Barriers. InTech; 2011.
This work has been supported by the Horizon 2020 Project [30] Borlase S. Smart grids: infrastructure, technology, and solutions. CRC Press;
titled REPLICATE (REnaissance of PLaces with Innovative Citizen- 2012.
ship And TEchnology), number: 691735. The authors gratefully [31] IEC. 61851-1: Electric vehicle conductive charging system-part 1: General
requirements. IEC (International Electrotechnical Commission), Geneva,
acknowledge Mr. Antonio Rossi and Mr. Salvatore Gabriele, who
Switzerland; 2010.
work as technicians of Istituto Motori, for their cooperation in [32] IEC 62196-3: Plugs, socket-outlets, vehicle couplers and vehicle inlets –
drawing the figure of this paper. The authors also gratefully Conductive charging of electric vehicles-part 3. IEC (International
Electrotechnical Commission), Geneva, Switzerland; 2014.
acknowledge Pompeo Marino, full professor at the Università degli
[33] https://www.chademo.com/technology/high-power/.
Studi della Campania Luigi Vanvitelli, for his help in discussions [34] http://www.chademo.com/chademo-is-published-as-ieee-standard/.
and comments about the contents of this paper. [35] 0.1.1-2015 - IEEE Standard Technical Specifications of a DC Quick Charger for
Use with Electric Vehicles.
[36] https://www.tesla.com/en_EU/supercharger.
References [37] https://twitter.com/elonmusk/status/812708946225963008?ref_src=twsrc%
5Etfw.
[38] https://audi-illustrated.com/en/future-performance-2015/Ladetechnologien.
[1] Villalobos JG et al. Plug-in electric vehicles in electric distribution networks: A
[39] Bakker S, Leguijt P, Van Lente H. Niche accumulation and standardization -
review of smart charging approaches. Renew Sust Energ Rev 2014;38:717–31.
the case of electric vehicle recharging plugs. J Clean Prod 2015;94:
[2] Annual Energy Outlook. Energy information administration. Department of
155–64.
Energy; 2010.
[40] http://tec.ieee.org/newsletter/may-june-2015/ieee-pev-dc-fast-charging-part-2.
[3] Kalghatgi G. Developments in internal combustion engines and implications
[41] http://www.ieahev.org/about-the-technologies/charging-equipment/.
for combustion science and future transport fuels. Proc Combust Instit
[42] Veneri O, Capasso C, Ferraro L, Del Pizzo A. Performance analysis on a power
2015;35(1):101–15.
architecture for EV ultra-fast charging stations. In: International Conference
[4] Alawi A, Baha M, Bradley TH. Review of hybrid, plug-in hybrid, and electric
on Clean Electrical Power (ICCEP); 2013. p. 183–88.
vehicle market modeling studies. Renew Sust Energy Rev 2013;21:190–203.
[43] Xiaofeng D et al. Effects of silicon carbide MOSFETs on the efficiency and
[5] Scrosati B, et al. Advances in battery technologies for electric vehicles.
power quality of a microgrid-connected inverter. Appl Energy
Woodhead Publishing; 2015.
2017;201:270–83.
[6] Veneri O. Technologies and applications for smart charging of electric and
[44] Deepak SG, et al. An automotive onboard 3.3-kW battery charger for PHEV
plug-in hybrid vehicles. Springer; 2016.
application. IEEE Trans Veh Technol 2012; 61(8): p. 3466–74.
[7] Mi C, Abul Masrur M, Gao DW. Hybrid electric vehicles: principles and
[45] Jaganathan S, Gao W. Battery charging power electronics converter and
applications with practical perspective. Chichester: Wiley; 2011.
control for plug-in hybrid electric vehicle. In: 2009 IEEE Vehicle Power and
[8] Hannan MA, Azidin FA, Azah M. Hybrid electric vehicles and their challenges:
Propulsion Conference.
A review. Renew Sust Energy Rev 2014;29:135–50.
[46] Jong-Soo K, et al. Design and implementation of a high-efficiency on-board
[9] Bayindir KC, Mehmet AG, Ahmet T. A comprehensive overview of hybrid
battery charger for electric vehicles with frequency control strategy. In: IEEE
electric vehicle: Powertrain configurations, powertrain control techniques
Vehicle Power and Propulsion Conference; 2010.
and electronic control units. Energy Convers Manage 2011;52(2):1305–13.
[47] Chae HJ, et al. 3.3 kW on board charger for electric vehicle. In: IEEE Power [79] Shareef H, Islam MM, Azah M. A review of the stage-of-the-art charging
Electronics and ECCE Asia (ICPE & ECCE); 2011. technologies, placement methodologies, and impacts of electric vehicles.
[48] Jih-Sheng L, et al. A high-efficiency 3.3-kW bidirectional on-board charger. In: Renew Sust Energy Rev 2016;64:403–20.
IEEE Future Energy Electronics Conference (IFEEC) 2015. [80] Flores RJ, Brendan PS, Brouwer J. Electricity costs for a Level 3 electric vehicle
[49] Fioretto M, et al. Power losses analysis in AC/DC conversion based on Active fueling station integrated with a building. Appl Energy 2017; 191: p. 367–84.
Front End systems. In: IEEE Symposium on Power Electronics, Electrical [81] Buchanan, William D., Joseph F. Mohos, and Wally E. Rippel. Battery charging
Drives, Automation and Motion (SPEEDAM); 2010. system and method. U.S. Patent No. 7,256,516. 14 Aug. 2007.
[50] Fioretto M, et al. Efficiency advantages by increasing converter’s number in [82] Planas E et al. AC and DC technology in microgrids: A review. Renew Sust
synchronous Active Front End systems for the AC/DC conversion. In: Energy Rev 2015;43:726–49.
International Conference on Power Electronics, Intelligent Motion and [83] Capasso C, Veneri O. Experimental study of a DC charging station for full
Power Quality; 2011. electric and plug in hybrid vehicles. Appl Energy 2015;152:131–42.
[51] Fioretto M, et al. Considerations on boost inductor design in back-to-back [84] Veneri O, Capasso C, Iannuzzi D. Experimental evaluation of DC charging
converters for renewable energy. In: IEEE International Conference on Clean architecture for fully-electrified low-power two-wheeler. Appl Energy
Electrical Power (ICCEP); 2011. 2016;162:1428–38.
[52] Gautam DS et al. A zero-voltage switching full-bridge DC-DC converter with [85] Xing L et al. Overview of current development in electrical energy storage
capacitive output filter for plug-in hybrid electric vehicle battery charging. technologies and the application potential in power system operation. Appl
IEEE T Power Electr 2013;28(12):5728–35. Energy 2015;137:511–36.
[53] Gautam DS, et al. A zero voltage switching full-bridge DC-DC converter for an [86] Sbordone D et al. EV fast charging stations and energy storage technologies: A
on-board PHEV battery charger. In: IEEE Transportation Electrification real implementation in the smart micro grid paradigm. Electr Pow Syst Res
Conference and Expo (ITEC); 2012. 2015;120:96–108.
[54] Gautam DS, et al. An interleaved ZVS full-bridge DC-DC converter with [87] García-Triviño P et al. Control and operation of power sources in a medium-
capacitive output filter for a PHEV charger. In: IEEE Energy Conversion voltage direct-current microgrid for an electric vehicle fast charging station
Congress and Exposition (ECCE); 2012. with a photovoltaic and a battery energy storage system. Energy
[55] Chang-Yeol O, et al. Design of resonant network based on power losses 2016;115:38–48.
analysis of series resonant converter for on-board battery charger in EVs. In: [88] Hafez O, Bhattacharya K. Optimal design of electric vehicle charging stations
IEEE Vehicle Power and Propulsion Conference; 2012. considering various energy resources. Renew Energy 2017;107:576–89.
[56] Seung-Hee R, et al. Operating optimization at light load of series resonant DC- [89] Ding H, Zechun H, Yonghua S. Value of the energy storage system in an
DC converter with duty-adjusted frequency control in EVs on-board charger. electric bus fast charging station. Appl Energy 2015;157:630–9.
In: IEEE Vehicle Power and Propulsion Conference; 2012. [90] Christen D, Jauch F, Biel J. Ultra-fast charging station for electric vehicles with
[57] Il-Oun L. Hybrid PWM-Resonant Converter for Electric Vehicle On-Board integrated split grid storage. In: IEEE European conference on power
Battery Chargers. IEEE Trans Power Electron 2016;31(5):3639–49. electronics and applications (EPE’15 ECCE-Europe); 2015.
[58] Chae HJ, Moon HT, Lee JY. On-board battery charger for PHEV without high- [91] Chandra GR, Bauer P, Zeman M. System design for a solar powered electric
voltage electrolytic capacitor. Electron Lett 2010;46(25):1691–2. vehicle charging station for workplaces. Appl Energy 2016;168:434–43.
[59] Lee YJ, Alireza K, Emadi A. Advanced integrated bidirectional AC/DC and DC/ [92] Ratil HA, et al. Integrated photovoltaic-grid dc fast charging system for
DC converter for plug-in hybrid electric vehicles. IEEE Trans Veh Technol electric vehicle: A review of the architecture and control. Renew Sust Energy
2009;58(8):3970–80. Rev 2017; 69(c): p. 1243–57.
[60] Lisheng S, Meintz A, Ferdowsi M. Single-phase bidirectional AC-DC converters [93] Sanzhong B, Lukic SM. Unified active filter and energy storage system for an
for plug-in hybrid electric vehicle applications. In: IEEE Vehicle Power and MW electric vehicle charging station. IEEE Trans Power Electron 2013;28
Propulsion Conference; 2008. (12):5793–803.
[61] De Sousa L, Silvestre B, Bouchez B. A combined multiphase electric drive and [94] Laxman M, Yamagishi T, Akagi H. Active-power control of individual
fast battery charger for electric vehicles. In: IEEE Vehicle Power and converter cells for a battery energy storage system based on a multilevel
Propulsion Conference; 2010. cascade PWM converter. IEEE Trans Power Electron 2012;27(3):
[62] Martel T, Rufer A. Configurable Modular Multilevel Converter (CMMC) for 1099–107.
flexible EV. In: IEEE European Conference on Power Electronics and [95] Laxman M et al. State-of-charge (SOC)-balancing control of a battery energy
Applications (EPE ECCE-Europe); 2015. storage system based on a cascade PWM converter. IEEE Trans Power
[63] Lacroix S, Labouré E, Hilairet M. An integrated fast battery charger for electric Electron 2009;24(6):1628–36.
vehicle. In: IEEE Vehicle Power and Propulsion Conference; 2010. [96] Michail V, Rufer A. Balancing control actions for cascaded H-bridge converters
[64] Bruyere A. A multiphase traction/fast-battery-charger drive for electric or with integrated battery energy storage. In: IEEE European conference on
plug-in hybrid vehicles: Solutions for control in traction mode. In: IEEE power electronics and applications (EPE); 2013.
Vehicle Power and Propulsion Conference; 2010. [97] Trintis I, Munk-Nielsen S, and Teodorescu R. Cascaded H-bridge with
[65] Subotic I. Overview of fast on-board integrated battery chargers for electric bidirectional boost converters for energy storage. In: IEEE European
vehicles based on multiphase machines and power electronics. IET Electr conference on power electronics and applications (EPE); 2011.
Power Appl 2016;10(3):217–29. [98] Abronzini U, et al. Dead time and non-linearities compensation for CHB
[66] Bodo N, et al. Single-phase on-board integrated battery charger based on a multi-level converters with integrated ESS feeding EV’s ultra-fast charging
nine-phase machine. In: Annual conference of the IEEE industrial electronics stations. In: Conference ESARS ITEC; 2016.
society (IECON); 2014. [99] Vasiladiotis M, Rufer A. A modular multiport power electronic transformer
[67] Bodo N et al. Efficiency Evaluation of Fully Integrated On-Board EV Battery with integrated split battery energy storage for versatile ultrafast EV charging
Chargers with Nine-Phase Machines. IEEE Trans Energy Convers 2017;32 stations. IEEE Trans Ind Electron 2015;62(5):3213–22.
(1):257–66. [100] Hillers A, Biela J. Optimal design of the modular multilevel converter for an
[68] Haghbin S et al. An isolated high-power integrated charger in electrified- energy storage system based on split batteries. In: IEEE power electronics and
vehicle applications. IEEE Trans Veh Technol 2011;60(9):4115–26. applications (EPE); 2013.
[69] Haghbin S et al. An integrated 20-kW motor drive and isolated battery [101] Akagi H. Classification, terminology, and application of the modular
charger for plug-in vehicles. IEEE Trans Power Electr 2013;28(8):4013–29. multilevel cascade converter (MMCC). IEEE Trans Power Electron 2011;26
[70] Chang HC, Liaw CM. Development of a compact switched-reluctance motor (11):3119–30.
drive for EV propulsion with voltage-boosting and PFC charging capabilities. [102] Leterme W, Beerten J, Van Hertem D. Equivalent Circuit for Half-Bridge MMC
IEEE Trans Veh Technol 2009;58(7):3198–215. Dc Fault Current Contribution. In: IEEE ENERGYCON; 2016.
[71] Botsford C, Szczepanek A. Fast charging vs. slow charging: Pros and cons for [103] Ladoux P, et al. Comparative study of variant topologies for MMC. In:
the new age of electric vehicles. In: International battery hybrid fuel cell International symposium on power electronics, electrical drives, automation
electric vehicle symposium; 2009. and motion (SPEEDAM); 2014.
[72] Aggeler D, et al. Ultra-fast DC-charge infrastructures for EV-mobility and [104] Zedong Z et al. A hybrid cascaded multilevel converter for battery energy
future smart grids. In: IEEE PES innovative smart grid technologies management applied in electric vehicles. IEEE Trans Power Electron 2014;29
conference Europe (ISGT Europe); 2010. (7):3537–46.
[73] http://www.chademo.com/chademo-announces-150kw-protocol/. [105] Gui-Jia S. Multilevel DC-link inverter. IEEE Trans Ind Appl 2005;41
[74] Carlos JG, Morcos MM. Impact of EV battery chargers on the power quality of (3):848–54.
distribution systems. IEEE Trans Power Deliver 2003;18(3):975–81. [106] Wang X, et al. Equivalent electromagnetic model of self-blocking MMC with
[75] Kolar JW, Friedli T. The essence of three-phase PFC rectifier systems—Part I. DC fault isolation capability. In: IEEE power and energy society general
IEEE Trans Power Electron 2013;28(1):176–98. meeting (PESGM); 2016.
[76] Serkan D, Cook S, Khaligh A. Comprehensive analysis of high quality power [107] Ja MK, et al. Real-time control system and entire control structure for a test
converters for level 3 off-board chargers. In: IEEE vehicle power and bench for a MMC-based HVDC transmission system. In: IET conference
propulsion conference; 2011. proceedings. The Institution of Engineering & Technology; 2014.
[77] Rivera S et al. Electric vehicle charging station using a neutral point [108] Cuzner RM, et al. Converter topological and solid state protective device
clamped converter with bipolar DC bus. IEEE Trans Ind Electro 2015;62(4): trade-offs for future shipboard MVDC systems. In: IEEE electric ship
1999–2009. technologies symposium (ESTS); 2015.
[78] Christen D, Tschannen S, Biela J. Highly efficient and compact DC-DC [109] Cuzner RM. Power electronics packaging challenges for future warship
converter for ultra-fast charging of electric vehicles. In: Power electronics applications. In: IEEE international workshop on integrated power packaging
and motion control conference (EPE/PEMC); 2012. (IWIPP); 2015.
[110] Kohlmann M., et al. Test-bench layout emulating a mmc-based hvdc [136] Zhou S, Chris Mi C. Multi-paralleled LCC reactive power compensation
transmission system. In IET MedPower; 2014. networks and their tuning method for electric vehicle dynamic wireless
[111] Bouabana A., et al. GRID fault generator for the assessment of loads under charging. IEEE Trans Ind Electron 2016;63(10):6546–56.
grid fault conditions. In: IET MedPower. [137] Deflorio F, Castello L. Dynamic charging-while-driving systems for freight
[112] http://www.nature.com/news/breathing-battery-advance-holds-promise- delivery services with electric vehicles: Traffic and energy modelling. Trans
for-long-range-electric-cars-1.18683. Res Part C-Emerg Technol 2017.
[113] Tao L et al. Cycling Li-O2 batteries via LiOH formation and decomposition. [138] Jaegue S et al. Design and implementation of shaped magnetic-resonance-
Science 2015;350(6260):530–3. based wireless power transfer system for roadway-powered moving electric
[114] Kato Y et al. High-power all-solid-state batteries using sulfide superionic vehicles. IEEE Trans Ind Electron 2014;61(3):1179–92.
conductors. Nat Energy 2016;1:16030. [139] Feng C, Taylor N, Kringos N. Electrification of roads: Opportunities and
[115] Bi Z, Kan T, Mi CC, Zhang Y, Zhao Z, Keoleian GA. A review of wireless power challenges. Appl Energy 2015;150:109–19.
transfer for electric vehicles: Prospects to enhance sustainable mobility. Appl [140] Flanders Make. Technical report inductive charging; 2013.
Energy 2016;179:413–25. [141] Swedish ICT Viktoria. Slide-in Electric Road System-Inductive project report.
[116] Wang CS, Covic GA, Stielau OH. Power transfer capability and bifurcation Project report, Phase 1.
phenomena of loosely coupled inductive power transfer systems. IEEE Trans [142] Rim C T. The development and deployment of On-Line Electric Vehicles
Ind Electron 2004;51(1):148–57. (OLEV). In: Energy conversion congress and exposition; 2013.
[117] Fariborz M, Edington M, Eberle W. Wireless power transfer: A survey of EV [143] Idaho National Laboratory ‘‘Plugged In: How Americans Charge Their Electric
battery charging technologies. In: 2012 IEEE energy conversion congress and Vehicles Findings from the largest plug-in electric vehicle infrastructure
exposition (ECCE). demonstration in the world” Summary Report.
[118] Kurs A et al. Wireless power transfer via strongly coupled magnetic [144] Francfort J, Carlson B, Shrik M, Smart J. Light-Duty Plug-in Electric Vehicle &
resonances. Science 2007;317(5834):83–6. Charging Infrastructure Data Collection in the U.S. In: 9th US – China electric
[119] Hasanzadeh S, Vaez-Zadeh S. Resonance based contactless energy transfer. vehicle and battery technology workshop; 2014.
In: 2012 IEEE 3rd power electronics and drive systems technology (PEDSTC). [145] http://www.greencarcongress.com/2016/11/20161116-drivethearc.html.
[120] Kunwar A, Williamson SS. Design considerations for loosely coupled [146] http://www.prnewswire.com/news-releases/drivethearc-ev-fast-charging-
inductive power transfer (IPT) system for electric vehicle battery charging- corridor-breaks-ground-in-northern-california-300362625.html.
A comprehensive review. In: IEEE transportation electrification conference [147] https://drivethearc.com/getting-started/.
and expo (ITEC); 2014. [148] http://replicate-project.eu/.
[121] Klontz KW, et al. Converter selection for electric vehicle charger systems with [149] http://www.izeus.kit.edu/downloads/iZEUS_Flyer_EN.pdf.
a high-frequency high-power link. In: 1993 power electronics specialists [150] http://www.izeus.kit.edu/english/62.php.
conference. PESC’93. [151] http://cordis.europa.eu/result/rcn/174238_en.html.
[122] http://www.altfuels.org/events/testdriv/farewell.shtml. [152] Sanz, JF et al. UNPLUGGED project: Development of a 50 kW inductive
[123] Villa JL et al. Design of a high frequency inductively coupled power transfer electric vehicle battery charge system. In: World electric vehicle symposium
system for electric vehicle battery charge. Appl Energy 2009;86(3):355–63. and exhibition (EVS27); 2013.
[124] Ahmed KK, Aamir M, Mekhilef S. Inductively coupled power transfer (ICPT) [153] http://cordis.europa.eu/result/rcn/192481_en.html.
for electric vehicle charging–A review. Renew Sust Energy Rev [154] http://www.fabric-project.eu/.
2015;47:462–75. [155] Brooks A, Gage T, AC Propulsion. Integration of electric drive vehicles with
[125] Shinohara N. Beam efficiency of wireless power transmission via radio waves the electric power grid—a new value stream. In: 18th international electric
from short range to long range. J Korea Electromagn Eng Soc vehicle symposium and exhibition, Berlin, Germany; 2001.
2011;10:224–30. [156] García-Villalobos J et al. Plug-in electric vehicles in electric distribution
[126] Ning P, Miller JM, Onar OC, White CP, Marlino LD. A compact wireless networks: A review of smart charging approaches. Renew Sust Energy Rev
charging system development. In: IEEE applied power electronics conference 2014;38:717–31.
and exposition (APEC); 2013. [157] Tseng HR. A secure and privacy-preserving communication protocol for V2G
[127] Sibue J, Kwimang G, Ferrieux J, Meunier G, Roudet J, Périot R. A global study of networks. IEEE wireless communications and networking conference
a contactless energy transfer system: analytical design, virtual prototyping (WCNC); 2012. p. 2706–2711.
and experimental validation. IEEE Trans Power Electron 2012;28:4690–9. [158] Zhong F et al. Smart grid communications: Overview of research challenges,
[128] Budhia M, Covic GA, Boys JT. Design and optimisation of magnetic structures solutions, and standardization activities. IEEE Commun Surveys Tutorials
for lumped inductive power transfer systems. In: IEEE energy conversion 2013;15(1):21–38.
congress and exposition ECCE; 2009. [159] Hazeldine T, et al. Market outlook to 2022 for battery electric vehicles and
[129] Covic GA, Boys JT. Modern trends in inductive power transfer for plug-in hybrid electric vehicles. In: Report to the committee on climate
transportation applications. IEEE J Emerg Selected Topics Power Electron change; 2009.
2013;1(1):28–41. [160] Nemry F, Brons M. Plug-in Hybrid and Battery Electric Vehicles: market
[130] Hui SYR, Zhong W, Lee CK. A critical review of recent progress in mid-range penetration scenarios of electric drive vehicles. In: JRC technical notes; 2010.
wireless power transfer. IEEE Trans Power Electron 2014;29(9):4500–11. [161] Humayd AS, et al. A Novel Framework for Evaluating Maximum PEV
[131] Zhu Q, Wang L, Liao C. Compensate capacitor optimization for kilowatt-level Penetration into Distribution Systems. IEEE Trans Smart Grid; 2017.
magnetically resonant wireless charging system. IEEE Trans Ind Electron [162] http://www.terna.it/it-it/azienda/chisiamo/ternastorage.aspx.
2014;61(12):6758–68. [163] Shenouda AY, Murali KR. Electrochemical properties of doped lithium
[132] Zhu Q, Guo Y, Wang L, Liao C, Li F. Improving the misalignment tolerance of titanate compounds and their performance in lithium rechargeable
wireless charging system by optimizing the compensate capacitor. IEEE Trans batteries. J Power Sources 2008;176(1):332–9.
Ind Electron 2015;62(8):4832–6. [164] Gil A, Taiber J. A literature review in dynamic wireless power transfer for
[133] Bloom MA, Niu G, Krishnamurthy M. Design considerations for wireless electric vehicles: Technology and infrastructure integration challenges.
electric vehicle charging. In: 2013 IEEE transportation electrification Sustainable automotive technologies - Springer International Publishing;
conference and expo (ITEC); 2013. p. 1–6. 2013. p. 289–298.
[134] Choi, Su Y., et al. Asymmetric coil sets for wireless stationary EV chargers [165] Onar OC, et al. A novel wireless power transfer for in-motion EV/PHEV
with large lateral tolerance by dominant field analysis. IEEE Trans Power charging. In: IEEE applied power electronics conference and exposition
Electron 2014; 29.12: p. 6406–20. (APEC); 2013. p. 3073–3080.
[135] http://primove.bombardier.com/media/news/news-detail-page/
article////experts-convinced-by-primove-solution-for-cars.html.

Applied Energy: Luigi Rubino, Clemente Capasso, Ottorino Veneri

Uploaded by

Copyright:

Available Formats

Applied Energy: Luigi Rubino, Clemente Capasso, Ottorino Veneri

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applied Energy: Luigi Rubino, Clemente Capasso, Ottorino Veneri

Uploaded by

Copyright:

Available Formats

Applied Energy 207 (2017) 438–464

Contents lists available at ScienceDirect

Review on plug-in electric vehicle charging architectures integrated with

 Outlook of PEV charging infrastructures.

1. Introduction growth in the number of road vehicles and electric utilities—is

Standard IEC 62196 - Type 1 (J1772)

Communication protocol Power-Line Communication (PLC)

Connector picture and scheme

Communication protocol Power-Line Communication (PLC)

Connector picture and scheme

Communication protocol CAN bus

Communication protocol Power-Line Communication (PLC)

Isolated Bi-directional Power supply Dc/Dc converter Performance Reference

converters working at high frequency that permits achieving better

Fig. 6. Single or three phase integrated battery charger [61].

as a passive rectifier, with the inductances L1 and L2. The switches

4. DC conductive charging technologies

As mentioned in the previous sections, the above-described

stations with an experimental validation of the proposed model,

Converter Grid voltage level Figure Advantages Drawbacks

From the description of the power architectures reported above, Table 5

Fig. 19. Pad Circular (A) and DDQ (B) structure.

installation of these devices requires excavation of the road line,

coming from each vehicle in charge, with positive effect on battery

mobility [18]. In this regard, power distribution companies are

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Outlook of PEV charging infrastructures.