art10_int

Download as pdf or txt
Download as pdf or txt
You are on page 1of 16

ARCHIVES OF ELECTRICAL ENGINEERING VOL. 72(1), pp.

173 –188 (2023)

DOI 10.24425/aee.2023.143696

Generalized state space model and small signal stability


analysis of Z-source converter

MASOUD JOKAR KOUHANJANI1 o B , SINA SOLTANI2 o , MOHAMMAD MARDANEH3 o

1 Department of Technical Study

Shiraz Electric Distribution Company


Shiraz, Iran
2 Departmentof Control and Management
Neyriz Ghadir Steel Complex (NGHSCO)
Fars, Iran
3 Department of Power and Control Engineering
Shiraz University of Technology
Shiraz, Iran
e-mail: B masoudjokar@hotmail.com, sinasoltani@shirazu.ac.ir, mardaneh@sutech.ac.ir, mail4

(Received: 23.08.2022, revised: 04.11.2022)

Abstract: This paper proposes two nonlinear exact and simple state space models of a Z-
source converter (ZSC) connected to an ac grid. A generic model of a ZSC accompanied
with proper controllers are proposed and a dynamic model of the whole system is derived;
as a result, based on a simple one, an equivalent block diagram of the current-controlled
ZSC system is proposed. The ac small signal stability method is applied and the impact of
controller parameters on network’s stability is discussed. Besides, overall system dynamic
performance has been assessed in the event of perturbations. Time-domain simulations
have been implemented in PSCAD/EMTDC to validate the accuracy of the models and
effectiveness of the proposed controllers. The results of the exact model are compared with
the response of the equations which are applied in MATLAB.
Key words: small signal stability, state space model, Z-source converter (ZSC)

1. Introduction
A Z-source converter, a novel replaceable energy conversion topology that is equipped by
passive components, has been proposed in [1]. Various topologies, control and modulation strate-
gies are considered in [2, 3], respectively. In contrast with conventional voltage-source converters
© 2023. The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution-
NonCommercial-NoDerivatives License (CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/), which per-
mits use, distribution, and reproduction in any medium, provided that the Article is properly cited, the use is non-commercial,
0 and no modifications or adaptations are made.
174 Masoud Jokar Kouhanjani et al. Arch. Elect. Eng.

(VSCs) and current-source converters (CSCs), the specific characteristic of ZSCs is both buck
and boost capabilities which is the outcome of:
i) shoot through switching state, at least two switches of one leg are on,
ii) X-shaped impedance network that includes two inductors (𝐿 1 and 𝐿 2 ) and two capacitors
(𝐶1 and 𝐶2 ), as well as the connection between a dc power source and a set of switches, as
shown in Fig. 1.

Fig. 1. Generic model of ZSC connected to ac grid

Nowadays, there is an extension to the power system to utilize power electronic-based equip-
ment such as: distributed generation (DG) units, HVDC lines and FACTS. Furthermore, due to the
remarkable capability of ZSCs that allows them to work as a buck or boost converter relevant to
desired DC-bus or ac terminal voltages, there will be a wide usage of ZSCs instead of VSCs in the
near future, particularly in the case of a power source with conversional output terminal voltage.
A large number of literatures have been allocated to various uses of ZSCs, which is clear evidence
of the steep increase in research on finding various suitable applications for this novel converter,
e.g., for adjustable-speed drives [4,5], electric vehicles [6–9], distributed generation [10–15] and
so on. In order to perform a feasibility study and detailed design, a precise comprehensive model
of the entire considered system should be provided.
Publications related to the model of Z-source converters mostly concentrate on the circuit of
the converters with a simple load regardless of the dynamic behavior of the ac grid and without
considering active and reactive power controllers [16–19]. In [20], a steady state analysis of
a Z-source dc-ac converter operating in continuous conduction mode (CCM) is presented. The
circuit-averaging technique is used to present ac small-signal modeling of the power stage of
a Z-source converter in [21].
Vol. 72 (2023) Generalized state space model and small signal stability analysis 175

In this paper, a Z-source dc-ac converter equipped with proper active and reactive power
controllers and connected to the infinitely stiff ac grid, which has constant frequency and balanced
sinusoidal voltage, is considered. The rest of this paper is organized as follows. In Section 2,
state space equations of electrical circuits, an impedance network and the ac grid, are derived.
Moreover, a coupling equation that makes relation between the impedance network and the ac
grid is derived. In order to track the desired active and reactive power a suitable control system
is proposed in Section 3, according to the conventional control methodology that is widely used
in VSCs. Section 4 is allocated to show the simplified model. Subsequently, the ac small-signal
stability method is applied in Section 5 in order to analyze the stability of the system and propose
an equivalent block diagram of the system. Section 6 is allocated to show the time-domain
simulation which is implemented in Matlab. The results are compared to the exact switching
circuit developed in PSCAD/EMTDC. Finally, Section 7 concludes the paper and comments on
possible future works.

2. Dynamic equations of electrical circuit


Figure 1 represents the included ZSC impedance network with a set of IGBTs, and its
corresponding ac-side equivalent circuit. There are four well-known methods to implement the
PWM switching strategy for Z-source converters: simple constant boost applied in this study [1],
maximum boost [22], maximum constant boost [23] and the triangle-shift control strategy [24].
The performance region of the converter is shown in Fig. 2. The basic equations of Z-source
converter terminal voltage are as follows [1], according to the stated switching function:
1
𝐵max = , (1)
1 − 2𝐷 max
1−𝐷
𝐺=𝑀·𝐵= , (2)
1 − 2𝐷
where: 𝑀 is the buck modulation index, 𝐷 is the shoot-through duty ratio, 𝐵 and 𝐺 are the boost
factor and the voltage gain of the converter in inversion mode, respectively. The system comprises
a power source that could be the representative of a DG unit, dc link or a part of a back-to-back
converter. The resistance 𝑅 𝑝 represents the total switching loss of the system [25]. The ac side of
the converter is connected to a utility grid through a series 𝑅𝐿 filter and three-phase transformer
that are modeled by 𝑅𝑠 and 𝐿 𝑠 .The synchronization signals for the control system are taken from
the low-voltage side of the corresponding transformer. The ZSC of Fig. 1 exchanges real and
reactive power components, 𝑃𝑠 (𝑡) and 𝑄 𝑠 (𝑡), with the ac grid at the point of common coupling
(PCC). For simplicity and without loss of accuracy, we assume the following:
– 𝑉dc is an ideal and independent voltage source. The forward voltage drop of 𝑆1 is considered
as a constant voltage drop 𝑉𝐷 . More explanation is delivered in the control system section.
– The carrier frequency is assumed much larger than that of the modulating signal (e.g. more
than 10 times).
– The dc capacitors and inductances are adequately large so that the changes of their volt-
ages and currents within one carrier period can be neglected. The impedance network is
considered lossless.
176 Masoud Jokar Kouhanjani et al. Arch. Elect. Eng.

– IGBT is presumed fast, therefore, we do not consider a time constant in a switch dynamic
model; moreover, in its conducting state, it is modeled by the series resistant (𝑟).
– A large number of dynamic models proposed for VSCs in literatures do not take into
account an ac-side capacitor filter [25–27].

Fig. 2. Capable performance region of ZSC according to simple constant boost switching strategy method

Two operation modes including two different circuit topologies can be introduced in ZSC
operation relevant to non-shoot through and shout through states [1, 16] as shown in Fig. 1 and
Fig. 3, respectively. Dynamics of the ac side of the ZSC system are described by the following
space-phasor equation:
d ®𝑖
𝐿𝑠 = −(𝑅𝑠 + 𝑟)®𝑖 + 𝑉®𝑡 − 𝑉®𝑠 , (3)
d𝑡
where 𝑉𝑠 and 𝑉𝑡 are the voltages of the power system and ac terminals of the converter, respectively.
For the traditional V-source PWM converter, we have the well-known relationship:
𝑉𝑘
𝑣 𝑡 (𝑡) = 𝑀 cos(𝜔𝑡 + 𝛿). (4)
2

Fig. 3. Equivalent circuit of Z-source converter in the shoot-through state

𝑉𝑘 is the voltage of the dc side terminal of the converter, 𝑀 is the buck modulation index
which is considered constant during operation modes of the ZSC, and 𝛿 is the phase difference
Vol. 72 (2023) Generalized state space model and small signal stability analysis 177

with respect to the grid voltage. For ZSCs we have a relationship between the discontinuous input
voltage of the set of switches, (𝑉𝐾 ), and the voltage of the power source, (𝑉dc ):

1
𝑉𝑘 = 𝑉dc , (5)
1 − 2𝐷
1
where: 𝑉dc stands for the dc source voltage, 𝐷 and are the shoot through and boost factors,
1 − 2𝐷
respectively. In order to simplify modeling and analysis, all state equations in the abc domain are
transformed into a 𝑑-𝑞 rotating reference frame. The transformation matrix is given by (4) where
“ 𝑓 ” can be any electrical parameter.
   
cos(𝜔𝑡) cos 𝜔𝑡 − 2𝜋 2𝜋 

cos 𝜔𝑡 +
 𝑓𝑑   3 3   𝑓 𝑎 
  2       
2𝜋 2𝜋   𝑓  .
 𝑓𝑞  =  sin(𝜔𝑡) sin 𝜔𝑡 − (6)
 
  3 sin 𝜔𝑡 +   𝑏
 𝑓0  3 3   
 1 𝑓
  𝑐
   1 1 
 
 2 2 2 
It is assumed that the system is balanced and zero-sequence components do not exist. The
angle 𝜔𝑡 is provided by the phase-locked loop (PLL), arbitrarily assumed here to align the system
voltage with the 𝑑-axis. By using abc to the 𝑑-𝑞 transformation matrix for the terminal voltage
of the converter:
𝑉𝑘
𝑉𝑡𝑞 = 𝑀 sin(𝛿), (7)
2
𝑉𝑘
𝑉𝑡 𝑑 =𝑀 cos(𝛿). (8)
2
The current of inductances, the voltage of capacitors and the 𝑑-𝑞 components of the grid
current are defined as state variables.
 𝑇
𝑥(𝑡) = 𝑖 𝐿1 (𝑡) 𝑖 𝐿2 (𝑡) 𝑉𝐶1 (𝑡) 𝑉𝐶2 (𝑡) 𝑖 𝑑 (𝑡) 𝑖 𝑞 (𝑡) . (9)

The model of the converter and the ac grid are derived based on two modes of the ZSC
operation.

2.1. Shoot-through state


In mode 1, no energy is transferred from the source to the load, since the load and the source
sides are entirely separated by the shoot-through state (at least one leg of IGBTs are on) and
open status of the switch 𝑆. The duty ratio of the shoot-through state is defined as 𝑑. Differential
equations of the ac grid and impedance network are:

d𝑖 𝑑 (𝑡)
𝐿𝑠 = 𝐿 𝑠 𝜔𝑖 𝑞 (𝑡) − (𝑅𝑠 + 𝑟) 𝑖 𝑑 (𝑡) − 𝑉𝑠𝑑 (𝑡), (10)
d𝑡
d𝑖 𝑞 (𝑡)
𝐿𝑠 = −𝐿 𝑠 𝜔𝑖 𝑑 (𝑡) − (𝑅𝑠 + 𝑟) 𝑖 𝑞 (𝑡), (11)
d𝑡
178 Masoud Jokar Kouhanjani et al. Arch. Elect. Eng.

d𝑖 𝐿1 (𝑡) d𝑖 𝐿2 (𝑡)
𝐿1 = 𝑉𝐶1 (𝑡), 𝐿2 = 𝑉𝐶2 (𝑡), (12)
d𝑡 d𝑡
d𝑉𝐶1 (𝑡) d𝑉𝐶2 (𝑡)
𝐶1 = −𝑖 𝐿1 (𝑡), 𝐶2 = −𝑖 𝐿2 (𝑡). (13)
d𝑡 d𝑡
As shown in Fig. 3, the switch 𝑆 is opened and the dc current transferred to the impedance
network is zero; therefore, 𝑖 𝑘 (𝑡), the current that flows through IGBTs, should be zero, which is
an advantage for ZSCs.

2.2. Non-shoot-through state


In mode 2, real energy is transferred between the source and load. As 𝑖 𝑝 is assumed constant,
we do not consider it in a dynamic model of the converter. The duty ratio of the shoot-through
state is defined as 𝑑 0. Similarly, the state equations of the ac grid and impedance network can be
written in mode 2 as follows, respectively:
𝐿 𝑠 d𝑖 𝑑 (𝑡)
= 𝐿 𝑠 𝜔𝑖 𝑞 (𝑡) − (𝑅𝑠 + 𝑟) 𝑖 𝑑 (𝑡) − 𝑉𝑠𝑑 + 𝑉𝑡 𝑑 (𝑡), (14)
d𝑡
𝐿 𝑠 d𝑖 𝑞 (𝑡)
= −𝐿 𝑠 𝜔𝑖 𝑑 (𝑡) − (𝑅𝑠 + 𝑟) 𝑖 𝑞 (𝑡) + 𝑉𝑡𝑞 (𝑡), (15)
d𝑡
𝐿 1 d𝑖 𝐿1 (𝑡) 𝐿 2 d𝑖 𝐿2 (𝑡)
= 𝑉𝑑𝑐 (𝑡) − 𝑉𝐶2 (𝑡), = 𝑉dc (𝑡) − 𝑉𝐶1 (𝑡), (16)
d𝑡 d𝑡
𝐶1 d𝑉𝐶1 (𝑡) 𝐶2 𝑑𝑉𝐶2
= 𝑖 𝐿2 (𝑡) − 𝑖 𝑘 (𝑡), = 𝑖 𝐿1 (𝑡) − 𝑖 𝑘 (𝑡). (17)
d𝑡 𝑑𝑡
Finally, the ac and dc circuit equations have to be coupled. The coupling equation follows
from the active power balance between the ac and dc sides of the converter. It allows calculating
the dc current of the converter:
3 
𝑉𝑘 (𝑡)𝑖 𝑘 (𝑡) = 𝑉𝑡 𝑑 (𝑡)𝑖 𝑑 (𝑡) + 𝑉𝑡𝑞 (𝑡)𝑖 𝑞 (𝑡) . (18)
2
From KVL in impedance network:
3 
𝑉𝑘 (𝑡)𝑖 𝑘 (𝑡) = 𝑉𝑡 𝑑 (𝑡)𝑖 𝑑 (𝑡) + 𝑉𝑡𝑞 (𝑡)𝑖 𝑞 (𝑡) . (19)
2

3. Control system
The purpose of designing the controller is to achieve good output-active and reactive power
tracking which are exchanged with the ac network at PCC. Among various control methods found
in the literature: internal model control in [28]; 𝐻∞ controllers in [28, 29]; a control strategy
based on the Lyapunov function in [30]; optimal coordinated control of the VSC and ac line in
parallel [31], two decouple-loop controllers are proposed for the Z-source inverter [32] through the
decoupled current-control method applied in this study, [25–27]. However, the control system is
modified so as to accommodate it in the ZSC system, Fig. 4 without 𝑑 0 in the current controller part.
In the case of disregarding variations of the dc source, two individual cases could have
occurred.
Vol. 72 (2023) Generalized state space model and small signal stability analysis 179

i) DC source represents an energy source that can meet the active power demand at various
ranges of dc voltages by changing the dc current and persisting output dc energy. More
profoundly, it can be concluded from Fig. 4 that any changes in 𝑉dc (𝑡) adversely contribute
to changes in 𝐷, leading to no effect in the tracking process of 𝐼 𝑑ref and 𝐼𝑞ref values.
ii) The power source is a capacitor which is a part of the dc line or back-to-back converter.
Consequently, any changes in dc voltage follow from energy equation disturbance held
between input power from the source of energy behind the proposed system, mentioned
capacitor and output energy to the Z-source impedance network.
Therefore, the presence of a dc voltage regulator in this case is unavoidable [13]. However,
the function of the dc controller is to change the set point of active power demand, as illustrated
in Fig 1. In the case of dynamic equations, it causes an extra state which is ignored in this study
for simplicity.The real and reactive powers delivered to the ac system at PCC are:
3
𝑃𝑠 (𝑡) = 𝑉𝑠𝑑 𝑖 𝑑 (𝑡), (20)
2
3
𝑄 𝑠 (𝑡) = − 𝑉𝑠𝑑 𝑖 𝑞 (𝑡). (21)
2
From (20) and (21) the reference value of dq-components of the current (𝐼 𝑑ref 𝐼𝑞ref ) can be
calculated based on the desired value of active and reactive  powers.  Therefore, a proportional-
𝑘𝑖
integral (PI) compensator of the generic form 𝐾 (𝑠) = 𝑘 𝑝 𝑠 + is sufficient for the control.
𝑠
Due to the presence of 𝐿𝜔 in (14) and (15), dynamics of 𝑖 𝑑 (𝑡) and 𝑖 𝑞 (𝑡) are coupled. To decouple
(14) and (15), 𝑚 𝑑 (𝑡) and 𝑚 𝑞 (𝑡) are determined as follows:

2 𝑢 𝑑 (𝑡) + 𝑉𝑠𝑑 − 𝐿 𝑠 𝜔𝑖 𝑞 (𝑡)
𝑚 𝑑 (𝑡) = , (22)
𝑉dc (𝑡)

2 𝑢 𝑞 (𝑡) + 𝑉𝑠𝑞 + 𝐿 𝑠 𝜔𝑖 𝑑 (𝑡)
𝑚 𝑞 (𝑡) = , (23)
𝑉dc (𝑡)
where 𝑢 𝑑 (𝑡) and 𝑢 𝑞 (𝑡) are:
 
𝑘𝑖
𝑢 𝑑 (𝑡) = 𝑘 𝑝 + (𝑖 𝑑ref − 𝑖 𝑑 (𝑡)) , (24)
𝑠
 
𝑘𝑖 
𝑢 𝑞 (𝑡) = 𝑘 𝑝 + 𝑖 𝑞ref − 𝑖 𝑞 (𝑡) . (25)
𝑠
We introduce two additional state variables, 𝑧 𝑑 (𝑡) and 𝑧 𝑞 (𝑡), such that:

d 𝑧 𝑑 (𝑡)
= 𝑖 𝑑ref − 𝑖 𝑑 (𝑡), (26)
d𝑡
d 𝑧 𝑞 (𝑡)
= 𝑖 𝑞ref − 𝑖 𝑞 (𝑡). (27)
d𝑡
The outputs of the current controllers are dq-components of the modulation index, 𝑚 𝑑 and
𝑚 𝑞 , that include buck and boost factors simultaneously. By transforming the modulation index
180 Masoud Jokar Kouhanjani et al. Arch. Elect. Eng.

from dq to the abc-frame, two relations are derived according to the following trigonometric
relationships:

𝑚 𝑞 (𝑡) 𝑢 𝑞 (𝑡) + 𝐿 𝑠 𝜔𝑖 𝑑 (𝑡)


𝛿 = tan−1 = tan−1 , (28)
𝑚 𝑑 (𝑡) 𝑢 𝑑 (𝑡) + 𝑉𝑠𝑑 (𝑡) − 𝐿 𝑠 𝜔𝑖 𝑞 (𝑡)
√︃
𝑀
= 𝑚 2𝑑 (𝑡) + 𝑚 2𝑞 (𝑡) . (29)
1 − 2𝑑 (𝑡)

It can be concluded from (28) that the amplitude of the modulation signal, 𝑚 𝑎𝑏𝑐 (blue arrows),
is more than unity which causes over-modulation; moreover, the control system should calculate
a shoot-through coefficient appropriate to the desired value of instantaneous active and reactive
power that the ZSC system exchanged with the ac system. Therefore, the shoot-through calculation
part is proposed as illustrated in Fig. 4. The buck modulation index (𝑀) is arbitrarily chosen to
be less than unity. Finally, the abc-frame of the modulation signal obtained with an amplitude is
equal to 𝑀.
The impedance network is considered symmetric, that is 𝐿 1 = 𝐿 2 = 𝐿 and 𝐶1 = 𝐶2 = 𝐶. All
differential equations can be written in the form of a nonlinear state space equation,

¤ = 𝐴𝑥(𝑡) + 𝐵𝑢(𝑡) + 𝑁 (𝑥, 𝑢),


𝑥(𝑡)

that is:

𝑑 − 𝑑0
 0 0 0 0 0
 
 𝐿 
 𝑖 𝐿 (𝑡)   𝑑 0 − 𝑑
 

   0 0 0 0 0
𝑉𝐶 (𝑡)   𝐶 
   
d  𝑖 𝑑 (𝑡)   0 𝑅+𝑟 
= 0 − −𝜔 0 0
d𝑡  𝑖 𝑞 (𝑡)  
 𝐿 
  
 𝑧 𝑑 (𝑡)   0 𝑅+𝑟 
0 −𝜔 − 0 0
  
 𝑧 𝑞 (𝑡)   𝐿 
 0 0 −1 0 0 0
   
 
 0
 0 0 −1 0 0
 𝑑0   0 
0 
 3 0
   
 𝑖 𝐿 (𝑡)   𝐿 
 − 𝑑 𝑀 (𝑖 𝑑 (𝑡) cos 𝛿 (𝑡) + 𝑖 𝑞 (𝑡) sin (𝛿(𝑡)) 
𝑉𝐶 (𝑡)   0 0   4𝐶
  
   
𝑑 0𝑀 
 𝑖 (𝑡)   1  𝑉 (𝑡) − 𝑉
   
cos (𝛿(𝑡)) (2𝑉𝐶 (𝑡) − 𝑉dc (𝑡))

 𝑑  0
+ −  dc 𝐷
+

 2𝐿

 . (30)
𝐿𝑠  𝑠
 𝑖 𝑞 (𝑡)   𝑉𝑠𝑑 (𝑡)
  
0𝑀
 𝑧 𝑑 (𝑡)   0 0   𝑑 
sin (𝛿(𝑡)) (2𝑉𝐶 (𝑡) − 𝑉dc (𝑡))
   
    2𝐿 𝑠 
 𝑧 𝑞 (𝑡)   0 0   
   

 0 

0 0   0 
  

where 𝑁 (𝑥, 𝑢) is the nonlinear part depending on both state variables and input.
Vol. 72 (2023) Generalized state space model and small signal stability analysis 181

Fig. 4. Control block diagram of current-controlled ZSC system


182 Masoud Jokar Kouhanjani et al. Arch. Elect. Eng.

4. Simplified model

So far, a comprehensive dynamic model and ac-side current-controller are proposed. The
proposed model is totally nonlinear and undeniably hard to work for stability consideration and
to drive related transfer functions for the controller design. Also, it is impossible to opt the value
of decoupling feed-forward links in the control system to separate the dynamic behavior of 𝑖 𝑑 (𝑡)
and 𝑖 𝑞 (𝑡) theoretically complete. Analytically and according to (30), the dynamic behavior of the
ac and dc sections is coupled. The chief component responsible for this fact is the impedance
network inductance 𝐿 which provides the main path of the current to the ac-section. More
elaborately, in the process of varying the active power setpoint, the dc current has to adapt to
the desired value, while the inductance (𝐿) is inherently reluctant to rapid changes, contributing
to capacitor participation in providing demanded energy at the first moment of changes. On the
other hand, the converter deals with reactive power variations more smoothly, since there is no
need to change the amount of produced dc energy by means of dc current. The desired amount of
reactive power relating tightly to the amplitude of ac terminal voltage of the converter is provided
by changes in the capacitor voltage with respect to (19). In addition, as 𝛿 is mostly small, the
dynamic response of 𝑖 𝑞 (𝑡) is less dependent on the impedance network than 𝑖 𝑑 (𝑡). Regarding this
explanation, the dependence of ac-part states on dc network components has been ignored for
the sake of simplicity; nevertheless, the dc-section dynamic equations are still inevitably coupled
to ac-side parameters. This dependence results from how to define 𝑉𝑡 𝑑 (𝑡) and 𝑉𝑡𝑞 (𝑡). Unlike the
previous form defined based on combination of electrical circuit and control system relations in
a simplified form, just control system equations, which mostly affect, are considered. Therefore,
we have:
𝑉dc
𝑉𝑡 𝑑 (𝑡) = 𝑚 𝑑 (𝑡) , (31)
2
𝑉dc
𝑉𝑡𝑞 (𝑡) = 𝑚 𝑞 (𝑡) . (32)
2
That is leading to:

d𝑖 𝑑 (𝑡)
= − 𝑅𝑠 + 𝑑 0 𝐾 𝑝 𝑖 𝑑 (𝑡) + 𝑑 0 𝐾𝑖 𝑧 𝑑 (𝑡) + 𝑑 0 𝐼 𝑑ref 𝐾 𝑝 ,

𝐿𝑠 (33)
d𝑡
d𝑖 𝑞 (𝑡)
= − 𝑅𝑠 + 𝑑 0 𝐾 𝑝 𝑖 𝑞 (𝑡) + 𝑑 0 𝐾𝑖 𝑧 𝑞 (𝑡) + 𝑑 0 𝐼𝑞ref 𝐾 𝑝 ,

𝐿𝑠 (34)
d𝑡
d𝑖 𝐿 (𝑡)
𝐿 = (𝑑 − 𝑑 0) 𝑣 𝐶 (𝑡) + 𝑑 0𝑉dc (𝑡) − 𝑅 𝐿 𝑖 𝐿 (𝑡), (35)
d𝑡
d𝑣 𝐶 (𝑡) 3𝑑 0 (1 − 2𝑑) 
𝐶 = (𝑑 0 − 𝑑)𝑖 𝐿 (𝑡) − 𝑖 𝑑 (𝑡)𝐾 𝑝 (𝐼 𝑑ref − 𝑖 𝑑 (𝑡)) + 𝑖 𝑑 (𝑡)𝑧 𝑑 (𝑡)𝐾𝑖
d𝑡 2𝑉dc (𝑡)
𝑉𝑠𝑑 𝑖 𝑑 (𝑡) 
+ 0
+ 𝑖 𝑞 (𝑡)𝑧 𝑞 (𝑡)𝐾𝑖 + 𝑖 𝑞 (𝑡)𝐾𝑞 (𝐼𝑞ref − 𝑖 𝑞 (𝑡)) , (36)
𝑑
d 𝑧 𝑑 (𝑡)
= 𝐼 𝑑ref − 𝑖 𝑑 (𝑡), (37)
d𝑡
d 𝑧 𝑞 (𝑡)
= 𝐼𝑞ref − 𝑖 𝑞 (𝑡). (38)
d𝑡
Vol. 72 (2023) Generalized state space model and small signal stability analysis 183

It should be noted that (33), (34) are completely decoupled. In other words, in the simplified
model, there is a capability to choose a proper value for decoupling feed-forward links, shown in
the controller part of Fig. 4. Moreover, a series resistance is considered in the dynamic equation
of the inductance current, due to avoiding instability of dc-side states. In Fig. 4, an impedance
network dynamic part is obtained from (35), (36).

5. Small signal stability

The small-signal stability of the test system is assessed using eigenvalue analysis in order to
find out the nature and root reason for the dynamic response of the ZSC. The simplified model
by differential Eqs. (33)–(38) is linearized around a set point of Table 1.
For analyzing the impact of different parameters, root-loci of eigenvalues are sketched ac-
cording to changes in the value of impedance network elements and current controller parameters
independently. As shown in Fig. 5, a pair of poles is under influence of the capacitance and
inductance of the dc-side network, while 𝐾 𝑝 and 𝐾𝑖 affect only the other two pair of poles. It can
be concluded that:
– High value of impedance network elements leads to decreasing oscillation of dynamic
response of all states, but mostly dc-side ones.
– It is potentially achievable to find an appropriate value of current controller parameters,
contributing to the order reduction of the system to the dominant poles.

(a) (b)

Fig. 5. Effect of parameters on poles of the system: different inductances and 𝐾 𝑝 (a);
different capacitances and 𝐾𝑖 (b)

6. Time domain validation

The studies reported in this section were performed on the Z-source converter of Fig. 1 using
parameters of Table 1. The studies are conducted:
– To verify the accuracy of the averaged models.
– To demonstrate the performance of the active and reactive power compensator proposed in
Section 3.
184 Masoud Jokar Kouhanjani et al. Arch. Elect. Eng.

Three study models of the system of Fig. 1 are simulated. The first one is the switching circuit
of the system including the 𝑑-𝑞 frame current controller and shoot-through calculation parts of
Fig. 4. This model is implemented in the PSCAD/EMTDC environment. The second and third
models are exact and simplified models derived in Sections 2 and 4, respectively. They can be
developed in an equation solver environment, e.g. MATLAB.

6.1. Models walidation


Initially the converter is delivering 0.1 MW active power to a local load, then at 𝑡 = 0.6 s the
breaker of the ac grid is closed and the reference signal of active power changed to 3.12 MW
corresponds to 4240 A of 𝐼 𝑑 . In order to analyze the dynamic response of the converter more
profoundly, the 1.37 Mvar reactive reference signal is performed at 𝑡 = 1.1 s equal to 1865 A of
𝐼𝑞 , as shown in Fig. 6, including the reference signals, exact switching circuit, averaged model
and simplified model responses.

(a) (b)

Fig. 6. 𝑑-𝑞 components of ac-side current: 𝑖 𝑑 (𝑡) (a); 𝑖 𝑞 (𝑡) (b)

The behavior of ac-side states, the dynamic response of 𝑖 𝑑 (𝑡) and 𝑖 𝑞 (𝑡), are different for both
the averaged model and exact switching circuit, as expected from (30). The various value of
overshoots with respect to reference signals are justified by the presence of 𝐿. As a matter of
fact, the inductance 𝐿 is reluctant to rapid set point changes. It causes more overshoot in the
dynamic response of 𝑖 𝑑 (𝑡), which is the representative of active power, while the response of 𝑖 𝑞 (𝑡)
encounters less overshoot, because of the fact that there is no need to provide energy from the dc
source. Therefore, the inductance 𝐿 does not participate in reactive power provision. It follows
the reference signal smoother. The differential equation of 𝑖 𝑞 (𝑡) from (30) is obviously proving
this fact by noting the nonlinear part. 𝛿 is mostly small, which results in neglecting its nonlinear
part responsible for the effect of dc-side states on 𝑖 𝑞 (𝑡).

6.2. Impact of mpedance network


As predicted before, the influence of the inductance 𝐿 on the behaviour of the system is
allocated to 𝑖 𝑑 (𝑡). For proving it, the previous scenario for implementing active and reactive
power signals is held for the ZSC system for different inductances, as shown in Fig. 7. The
Vol. 72 (2023) Generalized state space model and small signal stability analysis 185

overshoot of 𝑖 d (𝑡) for smaller inductance is larger, because the larger one is the more reluctant to
rapid changes in its current, At the same time, the response of 𝑖 𝑞 (𝑡) is completely independent of
the inductance.

(a)

(b)

Fig. 7. Impact of dc-side inductance value on: 𝐼 𝑑 (a); 𝐼𝑞 (b)

The effect of various capacitances and inductances on the behavior of capacitor voltage is
shown in Fig. 8. At 𝑡 = 0.6 s, the reference value of 𝐼 𝑑 is changed, the response of capacitor voltage

Fig. 8. Impact of capacitance and inductance on capacitor voltage dynamic response


186 Masoud Jokar Kouhanjani et al. Arch. Elect. Eng.

is significantly more intensive compared with 𝑡 = 1.1 s. It implies that at the first moments of
changing the active power set point, the inductance 𝐿 is obstructing the path of energy. Therefore,
the capacitor should provide the needed amount of energy, resulting in large dynamics and stresses
that directly depend mainly on the inductance, and secondly on capacitance.

Table 1. Parameters of the study system in Fig. 1

Parameter name Symbol Value Unit


Rated active power 𝑃𝑠 3.12 MW
Rated reactive power 𝑄𝑠 1.3 Mvar
AC grid nominal voltage 𝑉𝑠 600 V
AC equivalent inductance 𝐿𝑠 100 mH
AC equivalent resistance 𝑅𝑠 1.63 mΩ
Nominal frequency 𝐹 60 Hz
DC capacitance 𝐶 1000 µF
DC inductance 𝐿 1 mH
Nominal dc voltage 𝑉dc 1250 V
Current controller 𝐾𝑝 0.022 Ω
Current controller 𝐾𝑖 5 Ω/s
Carrier frequency 𝐹𝑠 1 kHz

7. Conclusions
The simulations on the circuit of Fig. 1 have proved that the dynamic behavior of the ZSC
connected to the ac grid system with any dc power can be persuasively simulated by the generalized
dynamic model proposed in this paper. Moreover, the property and accuracy of the proposed
controllers which are typical for VSC active and reactive current controllers are acquired through
the simulation of the exact switching model. At first, basic average equations of the converter and
its controllers are derived with regard to two individual operation modes, shoot-through and non-
shoot-through states. By combining all the equations, a general nonlinearity and simplification
are obtained, under some assumptions. In addition, in order to assess the impact of important
parameters such as PI coefficients and impedance network components on the stability of the
system, the small signal stability method is applied and the block diagram of the whole system
is proposed. The main contribution of this paper is that the well-accurate general model for
describing the dynamic response of the ZSC regardless of the type of dc power source and
the direction of active power flow is derived. The model is perfectly suited for the study of
AC/DC high-power equipment such as distributed generation and renewable energy systems. The
effectiveness of the ZSC can be assessed on HVDC-light and FACTS equipment for future works.
Vol. 72 (2023) Generalized state space model and small signal stability analysis 187

Acknowledgements
The authors are very grateful for the cooperation and financial support of Neyriz Ghadir Steel Com-
plex(NGHSCO), Shiraz University of Technology as well as Shiraz Electric Distribution Company.

References
[1] Peng F.Z., Z-source inverter, IEEE Trans. Ind Appl., vol. 39, no. 2, pp. 504–510 (2003), DOI: 10.1109/
IAS.2002.1042647.
[2] Siwakoti Y.P., Peng F.Z., Blaabjerg F., Loh P.C., Town G.E., Impedance-source networks for electric
power conversion part I: a topological review, IEEE Trans. Power Electron., vol. 30, no. 2, pp. 699–716
(2015), DOI: 10.1109/TPEL.2014.2313746.
[3] Siwakoti Y.P., Peng F.Z., Blaabjerg F., Loh P.C., Town G.E., Yang S., Impedance-source networks for
electric power conversion part II: review of control and modulation techniques, IEEE Trans. Power
Electron., vol. 30, no. 4, pp. 1887–1906 (2015), DOI: 10.1109/TPEL.2014.2329859.
[4] Peng F.Z., Yuan X., Fang X., Qian Z., Z-source inverter for adjustable speed drives, IEEE Trans. Power
Electron., vol. 1, no. 2, pp. 33–35 (2003), DOI: 10.1109/LPEL.2003.820935.
[5] Peng F.Z., Joseph A., Wang J., Shen M., Chen L., Pan Z., Rivera E.O., Huang Y., Z-source inverter
for motor drives, IEEE Trans. Power Electron., vol. 20, no. 4, pp. 857–863 (2005), DOI: 10.1109/
TPEL.2005.850938.
[6] Shen M., Joseph A., Wang J., Peng F.Z., Adams D.J., Comparison of traditional inverters and Z-
source inverter for fuel cell vehicles, IEEE Trans. Power Electron., vol. 22, no. 4, pp. 1453–1463
(2007), DOI: 10.1109/TPEL.2004.1393815.
[7] Guo F., Fu L., Lin C.H., Li C., Choi W., Wang J., Development of an 85-kW bidirectional quasi-Z-
source inverter with DC-link feed forward compensation for electric vehicle applications, IEEE Trans.
Power Electron., vol. 28, no. 12, pp. 5477–5488 (2013), DOI: 10.1109/TPEL.2012.2237523.
[8] Peng F.Z., Shen M., Holland K., Application of Z-source inverter for traction drive of fuel cell-
battery hybrid electric vehicles, IEEE Trans. Power Electron., vol. 22, no. 3, pp. 1054–1061 (2007),
DOI: 10.1109/TPEL.2007.897123.
[9] Dehghan S.M., Mohamadian M., Yazdani A., Hybrid electric vehicle based on bidirectional z-source
nine-switch inverter, IEEE Trans. Vehicular Tech., vol. 59, no. 6, pp. 2641–2653 (2010), DOI: 10.1109/
TVT.2010.2048048.
[10] Gajanayake C.J., Vilathgamuwa D.M., Loh P.C., Teodorescu R., Blaabjerg F., Z-source-inverter-based
flexible distributed generation system solution for grid power quality improvement, IEEE Trans. Energy
Convers., vol. 24, no. 3, pp. 695–704 (2009), DOI: 10.1109/TEC.2009.2025318.
[11] Jokar Kouhanjani M., Seifi A.R., Mehrtash M., Dynamic Model and Small Signal Analysis of Z-
Source Inverter, IETE Journal of Research, vol. 66, no. 65:3, pp. 342–350 (2018), DOI: 10.1080/
03772063.2018.1432421.
[12] Jokar Kouhanjani M., Seifi A.R., Comparison of VSC and Z-source converter: power system appli-
cation approach, Advances in Electronic and Electric Engineering, vol. 15, no. 1, pp. 12–18 (2017),
DOI: 10.15598/aeee.v15i1.1766.
[13] Ge B., Abu-Rub H., Peng F.Z., Li Q., de Almeida A.T., Ferreira F.J.T.E., Sun D., Liu Y., An energy
stored quasi-Z-source inverter for application to photovoltaic power system, IEEE Trans. Ind. Electron.,
vol. 60, no. 10, pp. 4468–4481 (2013), DOI: 10.1109/TIE.2012.2217711.
[14] Dehghan S.M., Mohamadian M., Varjani A.Y., A new variable-speed wind energy conversion system
using permanent-magnet synchronous generator and z-source inverter, IEEE Trans. Energy Convers.,
vol. 24, no. 3, pp. 714–724 (2009), DOI: 10.1109/TEC.2009.2016022.
[15] Jung J.W., Keyhani A., Control of fuel cell based z-source converter, IEEE. Trans. Energy Convers.,
vol. 22, no. 2, pp. 467–476 (2007), DOI: 10.1109/TEC.2006.874232.
188 Masoud Jokar Kouhanjani et al. Arch. Elect. Eng.

[16] Liu J., Hue J., Xu L., Dynamic modelling and analysis of z source converter-derivation of ac small
signal model and design-oriented analysis, IEEE Trans. Power Electron., vol. 22, no. 5, pp. 1786–1796
(2007), DOI: 10.1109/TPEL.2007.904219.
[17] Yu K., Luo F.L., Zhu M., Study of an improved z-source inverter: small signal studies, in Proc., 5th
Annual IEEE Industrial Electronics and Application. Conf., pp. 2169–2174 (2010), DOI: 10.1109/
ICIEA.2010.5515153.
[18] Gajanayake C.J., Vilathgamuwa D.M., Loh P.C., Small- signal and signal-flow-graph modelling of
switched z-source impedance network, IEEE Lett. Power Electron., vol. 3, no. 3, pp. 111–116 (2005),
DOI: 10.1109/LPEL.2005.859771.
[19] Loh P.C., Vilathgamuwa D.M., Gajanayake C.J., Lim Y.R., Teo C.W., Transient modelling and analysis
of pulse-with-modulated z-source inverter, IEEE Trans. Power Electron., vol. 22, no. 2, pp. 498–507
(2007), DOI: 10.1109/TPEL.2006.889929.
[20] Galigekere V.P., Kazimierczuk M.K., Analysis of PWM z-source DC-DC converter in CCM for steady
state, IEEE Trans. Circuits and Systems, vol. 59, no. 4, pp. 854–863 (2012), DOI: 10.1109/TCSI.
2011.2169742.
[21] Galigekere V.P., Kazimierczuk M.K., Small-signal modelling of open-loop PWM z-source converter
by circuit-averaging technique, IEEE Trans. Power Electron., vol. 28, no. 3, pp. 1286-1296, Mar. 2013,
DOI: 10.1109/TPEL.2012.2207437.
[22] Peng F.Z., Shen M., Qian Z., Maximum Boost Control of the Z-Source Inverter, IEEE Trans. Power
Electron, vol. 20, no. 4, pp. 833–838 (2005), DOI: 10.1109/TPEL.2005.850927.
[23] Shen M., Wang J., Joseph A., Peng F.Z., Tolbert L.M., Adams D.J., Maximum constant boost control
of the z-source inverter, IEEE Indus. Appl. Conf., pp. 3–7 (2004), DOI: 10.1109/IAS.2004.1348400.
[24] Ding X., Qian Z., Xie Y., Peng F.Z., A Novel ZVS Z-Source Rectifier, IEEE Power Electron. Specialists
Conf., pp. 1–5 (2006), DOI: 10.1109/APEC.2006.1620653.
[25] Yazdani A., Irevani R., Voltage-sourced converters in power systems, 1𝑠𝑡 ed., IEEE Press, John Wiley
& Sons, pp. 204−244 (2010).
[26] Yazdani A., Irevani R., An accurate model for the DC-side voltage control of the neutral point diode
clamped converter, IEEE Trans. Power Del., vol. 21, no. 1, pp. 185–193 (2006), DOI: 10.1109/TP-
WRD.2005.852342.
[27] Yazdani A., Iravani R., A generalized state-space averaged model of the three-level NPC converter for
systematic DC-voltage-balancer and current-controller design, IEEE Trans. Power Del., vol. 20, no. 2,
pp. 1105–1114 (2005), DOI: 10.1109/TPWRD.2004.834307.
[28] Liang H., Li G., Li P., Yin M., Analysis and design of H1 controller in VSC HVDC systems, in:
IEEE/PES Transmission and Distribution Conference and Exhibition: Asia and Pacific, p. 6 (2005),
DOI: 10.1109/TDC.2005.1547069.
[29] Zhang L., Nee H.-P., Multivariable feedback design of VSC-HVDC connected weak AC systems, in:
Power Tech., pp. 1–8 (2009), DOI: 10.1109/PTC.2009.5282110.
[30] Li G., Ma G., Zhao C., Li G., Research of nonlinear control strategy for VSC-HVDC system based on
Lyapunov stability theory, in Proc. 3rd International Conference on Electric Utility Deregulation and
Restructuring and Power Technologies, pp. 2187–2191 (2008), DOI: 10.1109/DRPT.2008.4523773.
[31] Mao C., Hu Z., Lu J., Chang D., Fan S., Application of an optimal coordinated control strategy to
VSC HVDC, in Proc. IEEE PES Power Systems Conference and Exposition, pp. 2141–2145 (2006),
DOI: 10.1109/PSCE.2006.296275.
[32] Tsang K.M., Chan W.L., Decoupling controller design for z-source inverter, IET Power Elec., vol. 8,
no. 4, pp. 536–545 (2015), DOI: 10.1049/iet-pel.2014.0207.

You might also like

pFad - Phonifier reborn

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

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:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy