Bozhko 2018
Bozhko 2018
Bozhko 2018
fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
The state of the art 24 krpm 115V or 230V variable- been reviewed and the main results achieved have also been
frequency alternating-current (VFAC) S/G has a presented in this paper.
relatively high power density, but can only deliver a
small fraction of the generation power in starting II. BASIC CONCEPT
mode (most are found on APUs and the biggest ones This section gives a brief introduction of the concept of a
are on B787 main engines) starter/generator system. The starter/generator system is
Advances in power electronics result in the possibility to essentially a system which allows bi-direction power flow
consider other machine types for the S/G operations which between two voltage sources.
have the potential of a substantial improvement in power
density (kW/kg), overall system performance, functionality,
reliability, diagnostics, and availability for the overall system
compared to the previous or existing solutions.
The AEGART project [3] (stands for Aircraft Electrical
Starter-Generation System with Active Rectification
Technology) targeted the development of such a novel system.
Figure 1 Example circuit
The AEGART project aim was to develop an efficient and
lower-mass S/G system for next-generation business Jet Consider a simple RL-electric circuit connected to two voltage
applications. The AEGART project aimed to deliver 20% sources VA and VB on each side, as shown in Figure 1. If we
lower mass and 10% better efficiency than comparable assume that voltage source VA is fixed. Varying the magnitude
conventional starter/generators with better or improved safety. and the phase of the source VB, the current in the circuit can be
This project was part of “Eco-Design for Systems” ITD within controlled in both the magnitude and the phase with respect to
CleanSky JTI and has become a cornerstone of the MEA. VA. It means that the power flow through the circuit (active P
Some of the key features of the AEGART system have been and reactive Q) can be controlled in either direction. If the role
described in TABLE I. of VA is given to the electric machine, and the role of VB – to
TABLE I Key features of AEGART system the power electronic converter that acts as a voltage source,
then one can control the machine as a motor (directing the
Key features Description active power flow to machine) or as a generator (power is
extracting from the machine) as illustrated by Figure 2.
Operation speed range 0 - 32,000rpm
Continuous electrical 45kW with 270Vdc voltage
power delivered source: R L
VM VC C
Maximum power can be can go up to 150kW with EM
VA VB Vdc
delivered by the permanent increased DC-link voltage
I
magnet machine
Cooling of the machine Stator cooling
Cooling of the power Power converter is located at controls
converter the power electronics bay and
thus liquid cooling is not Figure 2 Electric Machine – Power Converter system
available. Forced air cooling The developed AEGART S/G system implements the
is used in the AEGART described idea. The speed-torque characteristic requirement of
the S/G is shown in Figure 3. The machine runs as a motor
Consideration of many design aspects was required for this during the engine starting period. During this period, the
complex multi-disciplinary task including definition of the power is flowing from the converter to the electric machine.
best system topology, electrical machine and power electronic When the engine is self-sustained, it starts to drive the electric
converter design, thermal management and cooling machine as a generator and the power starts to flow from the
arrangements, development of control techniques, health electric machine to the converter. The dc-link voltage of the
monitoring and fault management systems. This paper reports power converter is set at 270VDC. An approximate torque
the final achievements of the AEGART project. One of the speed characteristics requirement of the starter generator is
challenges encountered in the design of a starter/generator for shown in Figure 3. As can be seen, the machine runs as a
aero-engines is the need to satisfy the two fundamental motor during engine start to ωstart. The engine ignites at the
functions, namely to provide mechanical power to the engines speed ωstart and starts to drive the machine to the speed ω min,
during start-up (as a starter) and to generate electrical power where the machine will start to run as a generator. The speed
during normal engine operation (as a generator). The range for the electric machine running as a generator is from
developed solution utilizes recent advances in modern power ωmin (20,000rpm) to ωmax (32,000rpm). When running as a
electronics allowing the use of novel machine types together generator, the electric machine can supply a maximum
with the introduction of controlled power electronic converters constant power of 45 kW.
into the main path of energy flow. This paper summarized the
AEGART project including machine design, power converter
design and S/G testing. Different aspects of the project have
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
T(Nm)
III. ELECTRIC MACHINE DESIGN
ωbase In the design process, fault tolerance is considered by adopting
Tstart
a redundancy solution in the event of a failure. This allows a
distributed winding to be adopted, consequently minimizing
rotor losses in both permanent magnet and the rotor back iron.
Different slot-pole combinations were investigated in view of
0 the overall losses and performance and the compromise in the
ωstart ωmin ωmax ω (rad/s)
design of the machine for operating as an engine starter and as
generator. Implications of different magnet’s retention
Figure 3 Torque-Speed characteristics of the aircraft starter- material and magnetic materials were studied.
generator
Different electrical machine topologies, namely induction A. Slot/pole combinations
machines, switched reluctance machine [4, 5] and topological The slot (Q) /pole (p) combinations considered during design
variants of permanent magnet based machines [6, 7] were are shown in Table I. Both single (Nlay = 1) and double layer
considered together with their respective driving converters (Nlay = 2) winding topologies are considered in the selection
during the trade-off study period with the process shown in procedure. The number of rotor poles is limited to 4, 6 and 8
Figure 4. as any higher number of poles will lead to high electrical
Initially, the identified machine types have been subject to a frequencies and higher power converter losses. In order to
basic design analysis to establish the optimal aspect ratio, limit the temperature increase in the rotor as a consequence of
cross section, to guarantee a mechanical integrity, to define a the eddy currents induced by magnetomotive force (MMF)
number of turns to suit the required DC-side voltage, etc. harmonics at high speed, the winding topology for each
Based on these criteria, the possibility of achieving a peak slot/pole combination was considered such that the winding
torque has been assessed. If failed, then the next harmonics are minimized for an acceptable trade-off with the
approximation cycle was initiated with modified initial fundamental winding factor.
selections. If the peak torque can be achieved, then the next
stage of design was undertaken: this includes calculation of TABLE II Considered slot/pole combinations
machine losses and calculation of machine lumped Q Nlay 2p Q 2p Q 2p Q
parameters. In parallel, each machine type requires an 0.5 2 4 6 6 9 8 12
appropriate converter topology and thermal arrangement. 1 1 4 12 6 18 8 24
These were included into the trade-off process as well, as 1 2 4 12 6 18 8 24
Figure 4 shows. A machine lumped parameter model, together 1.5 2 4 18 6 27 8 36
with the converter topology candidate, allows for calculation 2 1 4 24 6 36 8 48
of converter losses and design of basic heatsink. Machine 2 2 4 24 6 36 8 48
losses, together with possible machine cooling arrangements,
allows for assessment of machine temperature – if this appears Normally, the highest efficiency can be achieved by machines
to be higher than allowed, then another cycle of basic design is with higher pole numbers. However, a high pole number will
required. If passed, then the total system mass and system result in a higher fundamental frequency of the machine and
efficiency can be evaluated, as well as system reliability and thus require a higher switching frequency of the power
safety. converter. With these considerations, the 6-pole machine
designs were considered for the final machine design. Out of
the 6-pole machine variants, the best machine candidate was
selected as 6-poles 36-slots as it has the lowest eddy-current
losses.
Table III represents the winding factors associated with
different slot-per-pole-per-phase (q) and winding layer
number combinations considered in the comparison designs,
where the parameter ᴦc represents the coil pitch.
TABLE III Winding factors associated with different slot-per-pole-
per-phase and winding layer number combinations
Winding factors of prominent harmonics
q Nlay ᴦc 1 2 4 5 7 11
0.5 2 1 0.866 0.866 0.866 0.866 0.866 *
1 1 3 1 0 0 1 1 1
1 2 2 0.866 0 0 0.866 0.866 0.866
Figure 4 Illustration of AEGART trade-off process
1.5 2 4 0.945 0.061 0.140 0.140 0.061 *
As the trade-off study result, for the detailed development of 2 1 5 0.966 0 0 0.259 0.259 0.966
the AEGART system, the topology “permanent-magnet 2 2 5 0.933 0 0 0.067 0.067 0.933
machine with surface-mounted magnets (SPM) with three- * represent cases where non-zero intermediate values exist between the 7th
level IGBT converter” was selected. and the 11th harmonic orders.
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
When comparing the different slot/pole combinations, in each the electrical machine is estimated using Finite Element
design, the stator tooth maximum flux density at no-load was Software MAGNET. The maximum Eddy losses were
considered to have 75% of the saturation flux density of the achieved at 32,000rpm at rated power 45kW. The stress plot
core material. The total weight of each machine was of the rotor with the CF sleeve at 32,000rpm with the worst
maintained the same for the comparative study. This included temperature is shown in Figure 6. It can be seen that the
the stator, rotor and the copper associated with the winding, maximum stress under the worst case scenario (highest
including the end windings. Each machine was thus designed operational speed 32,000rpm), is around 715 MPa which is
for a constant weight and fixed outer diameter so as to fulfil within safe margin of the given CF sleeve’s allowable stress.
the torque-speed criteria presented in Figure 3. The final
TABLE IV. Mechanical parameters of Carbon Fiber and Inconel
selection of the winding configurations is made to be slot-per- sleeves at the maximum speed 32krpm
pole-per-phase q=2, winding layer number to be 1 and coli
Parameter Carbon Fiber Inconel Alloy
pitch ᴦc to be 5 due to its lower harmonics.
Sleeve thickness (mm) 4 4
B. Implication of Magnet retention Sleeve undersize (mm) 0.134 0.308
To retain the surface magnet, two different sleeve materials Sleeve prestress (MPa) 510 676
were investigated during the design period: Carbon Fiber (CF) Eddy losses at 32,000rpm (W) 12.1 383.7
and Inconel 718. The sleeve thickness calculation was
performed based on a 110% over-speed safety margin (i.e.
designed at 32,000rpm *110% =35,200rpm) . The relation
between the sleeve thickness and the rotor radius with
different magnet thickness is shown in Figure 5.
Figure 6 Stress plot of machine rotor with the Carbon Fiber sleeve at
32 krpm and under a worst case temperature gradient of 200 0C
sleeve, 1700C magnets and 1200C rotor core
C. Magnetic material selection
From Figure 3, it is possible to identify that the designed
machine will be working under flux weakening conditions
with a wide speed range (speed from ω base = 12,000rpm up to
32,000rpm). Under the flux weakening conditions, the
machine magnet will be handling demagnetizing current
injected from the machine terminals. In order to prevent a
design of a machine with a high demagnetization risk, the
maximum demagnetization field condition is incorporated in
the design process The magnets are arranged in a Halbach
structure using Samarium cobalt magnet (Sm2Co17) material.
Magnet demagnetization at different operating points is
Figure 5 Variation of sleeve thickness requirement with rotor outer considered and the designs are modified when it was subject
diameter: above, Carbon fibre sleeves; below, Inconel sleeves
to strong demagnetization field.
It can be seen that the required sleeve thickness is highly
influenced by the rotor radius. This is due to the fact that the
centrifugal force proportionally increases with the rotor radius.
The magnet thickness, however, has a less significant
influence over the sleeve thickness requirement. It was
identified that the carbon fibre sleeve was required to be
thicker than that of the Inconel sleeve, especially when the
rotor radius is beyond 60mm.
The pre-stress of the 4 mm sleeve of CF and Inconel at (a) (b)
maximum speed (32krpm) is shown in Table III. Both sleeves
provide reasonable pre-stress with corresponding sleeves. The Figure 7. AEGART machine: manufactured stator and rotor. (a)
CF has been the final choice since it has negligible eddy cross-section view of the manufactured 36 slot stator; (b)
manufactured rotor with CF sleeves
current losses compared to Inconel sleeve. The eddy losses of
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
D. Magnet Demagnetization Direct oil cooling of the machine stator was identified as the
optimal solution. The core principle utilised for machine
Operating conditions in both motoring and generating modes
cooling was to use ducts through the stator core along the
demand high currents which may result in high
stator outer diameter and the slots existing between stator
demagnetization fields. In order to prevent a design of a
teeth. This arrangement guaranteed an even distribution of the
machine with a high demagnetization risk, a maximum
coolant. An electrically non-conductive sleeve separated the
demagnetization field condition is incorporated in the design
airgap and rotor from the stator. This solution keeps windage
process. Based on the cooling available, worst-case rotor
losses manageable (windage loss is the viscous friction loss
temperature for the aero-engine starter-generator is considered
generated between a fluid and the rotating elements). The
at 300oC. The magnets are arranged in Halbach structure using
electromagnetic losses (including copper loss, eddy current
Samarium cobalt magnet (Sm2Co17) material. Magnet
losses in magnets and iron losses) were derived using co-
demagnetization at different operating points is considered and
simulation of FE machine model and power converter in
the designs are modified when it was subject to strong
Matlab/Simulink. The windage loss was estimated both
demagnetization field.
numerically (CFD) and analytically whilst bearing losses
where only analytically predicted as detailed in [10]. Further
enhancing cooling techniques, such as radial jets at the drive
end and a flow deflector at the non-drive end were also
implemented to improve system efficiency and hence reduce
machine weight.
(a) (b)
Figure 8 Demagnetization prediction plot at maximum speed
Figure 8a shows the demagnetization prediction plot
obtained in FE at 32 krpm under worst case temperature. The
constant value of 1 shown in the plot represents the magnet
under risk of demagnetization whilst 0 is no demagnetization
risk. From the results, it can be clearly seen that there are
slight demagnetization effects during generating mode. This is
due to the magnet arrangement of the Halbach structure. To
avoid the demagnetization on the edges, the magnets are
modified as shown in Figure 8b. From Figure 8b, it can be
clearly seen that the magnets are not subject to
demagnetization.
E. Thermal management
The thermal design of the developed starter/generator has been
detailed in [8] and will be only briefly discussed in this paper.
During the design process, the following aspects were
accounted for in design of an optimised cooling system: Figure 9 Main components of the designed starter/generator
Effectiveness: Intensive cooling can improve the overall
machine efficiency by reducing the temperature F. Mechanical design
dependent losses such as copper losses. The detailed mechanical design of the machine is presented in
Reliability: High temperatures accelerate the aging of our previous publications [8]. The final rotor, stator and
some components, reducing their lifetime; hence machine housing are shown in Figure 9. The rotor shaft is
operating at low temperatures can increase the overall made of high strength stainless still with a diameter of 40mm
reliability of the machine. As an example, a temperature and a length of 250mm. The shaft is connected to a gear box
rise of 10°C can reduce the winding insulation life by through a spline at the drive end. At the non-drive end, a
around 50% [9]. resolver is connected for speed measurement and control
Reduction of size and weight: In aerospace applications, applications. The shaft is supported by bearings at each end
size and weight are particularly significant factors to be mounted in the two flanges of the machine. The stator, made
taken into account, having a direct impact on the fuel of iron laminates and copper windings, was directly inserted
consumption and hence CO2 emissions. into the machine casing. The rotor consisted of a shaft and
magnets, and was inserted into the stator central chamber,
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
Idc(A)
SM PMM
ωm
Idc_max idc R= 1/kdroop
kdroop
vd2 vq2
Coordinate θ
Iabc
transform
270 Vdc(V)
abc→dq0 +
VC VC
VC max ref
V=270 Vdc
Flux
weakening
Id D-axis
current -
controller controller Vdref Vabcref
ωm Coordinate
ωmref
Iq max transform
Speed dq0→abc
controller
S Iqref Q-axis Vqref -Idc_max
current
Im a x
270V
Idcref DC current G controller (a) (b)
controller TSR
-Im a x
limiter Idc Cdc
Vdc Figure 12 Representation of droop control for S/G systems during
Vdc
{ the generation mode. a) in diagram, b)An equivalent circuit
270V
DC Bus
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
PMs (Sm2Co17) are arranged in Halbach array structure which used at these temperatures are provided by our industrial
is retained by directly wound CF sleeves. The prototyped partner. Under these two conditions, the engine friction T and
stator, rotor and complete assembly of the starter-generator speed ω profile is shown in Figure 18.
machine are shown in Figure 16. It is worth highlighting here Shaft, T is idc
controlled
T
that the mass density of the machine is 33kW/L & 16kW/kg. T*
Dynamometer
Dynamometer
Electrical
machine GSS
controller of S/G Vdc
ω
ω T
The final system scheme for S/G functional test is also shown ω, iabc
Sabc
Vdc
in Figure 16. During our test campaign, the S/G was coupled 'S/G controller in generation mode
control (switch to ‘S’ state in Figure 9)
mechanically with a dynamometer machine which can be
operated in four quadrants in the torque-speed plane.
Figure 17. Dynamometer control when S/G is in the motoring mode
idc
VC Cdc
Dynamometer S/G GSS
DC source
Engine emulation
Iabc
{
Figure 16. Testing bench for the S/G system, left to right:
dynamometer, Starter/generator, power electronics and GSS DC
source
Figure 18 Engine friction torque corresponding to the engine speed
The dynamometer allows operation in two modes: the torque during the starting process. Solid line: cold-start condition,
control mode and the speed control mode. These two modes dashed line: hot-start condition.
are used for testing the motoring and generation functions During the motoring mode test, it is required that the S/G
respectively. All these tests were conducted using facilities should be able to accelerate aircraft engines up to 10,000rpm
within the University of Nottingham and implemented at smoothly. The motoring test result under a cold-start torque
ambient temperature. The machine is pre-heated up to 50oC profile is shown in Figure 19 (a1- d1). From the speed profile
before test (this is due to the fact that 50oC is the minimum in Figure 19(a1), it can be seen that the S/G is able to
temperature at which the temperature management system smoothly accelerate the aircraft engine during a cold
functions properly) environment (-40oC) to a requested speed 10,000rpm in 40
A. Motoring mode test of S/G seconds. Since the speed is lower than ωbase, there is no need
for flux weakening operation, i.e. the current id = 0A. The
During the S/G motoring mode test, the dynamometer behaves electromagnetic torque Te is proportionate to the q-axis current
as a mechanical load to the S/G and represents the engine iq and thus the current iq reflects the cold-start torque profile as
friction torque during engine start-up. When running as a shown in Figure 19(b1). The temperatures of machine
mechanical load, the dynamometer is working in a torque windings and the power converter are shown in Figure 19 (c1)
control mode and the torque is controlled by its own and Figure 19 (d1). It can be seen that the maximum
controller. Since the engine friction torque is dependent on the temperature is 60oC for the power converter devices and is
engine shaft speed and the ambient temperature, different T-ω 72oC for electric machine stator windings. Both temperatures
curves at different temperatures are expected. As can be seen are below the maximum allowed values. The temperature of
from Figure 17, the reference torque T * of dynamometer is power converter devices and electric machine windings are
dependent on the shaft rotational speed ω with the T-ω profiles closely related to the active power (providing torque during
pre-programmed in its controller before testing. motoring) current iq. It can be seen that both machine and
For aircraft applications, it is required that the S/G should be power converters reached their highest temperature
able to start up the engine successfully in any potential simultaneously after iq decreased from the maximum point
ambient temperature conditions. Two extreme cases were (Figure 19 b1-c1).
tested during the S/G system functional test campaign, i.e.
ambient temperature T= 55oC (engine hot start-up condition)
and T= -40oC (engine cold start-up condition). The T-ω curves
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
12000
(a1) (a2)
Speed (rpm)
8000
4000
0
0 5 10 15 20 25 30 35 40 45
400
(b1) (b2)
300
id,iq(A)
200
100
-100
0 5 10 15 20 25 30 35 40 45
50
(c2)
(c1)
PE Temp(C)
40
30
20
0 5 10 15 20 25 30 35 40 45
58
(d2)
(d1) 56
EM Temp(C)
54
52
50
0 5 10 15 20 25 30 35 40 45
Time(s)
Figure 19 Machine speed, id. and iq response starter/generator system for cold-start (a1- d1) and hot-start (a2- d2)
conditions. a) speed response; b) id & iq; (a) Temperature of power converter, (b) Temperature response of electric
machine stator windings
The hot-start testing results are shown in Figure 19 (a2 – d2). reference value by its own controller. As illustrated in Figure
It can be seen that the machine successfully drove the engine 20, the dynamometer essentially behaves as a mechanical
up to 10,000rpm in 40 seconds time very smoothly. The speed source during this mode and drives the S/G to any
current iq of machine stator reflected the hot-start engine requested speed up to 32,000rpm (the maximum generation
torque profile. Since the speed has not reached the flux- speed). The S/G transforms the mechanical power to electrical
weakening range, the flux-control current id also kept at zero. power and supplies electrical power to the loads (i.e. GSS as it
The maximum temperature of the power converter during hot- is able to operate in four quadrant and thus can be used as a
engine start mode is about 40oC and the maximum load).
temperature of the electric machine is 57oC. During the cold- Shaft, ω is idc
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
Cooling
100 Station
Current Idc (A)
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
REFERENCES
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2018.2863031, IEEE
Transactions on Transportation Electrification
IEEE Transactions on Transportation Electrification - 2018- 01- 0006
2332-7782 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.
Alternative Proxies: