1. Introduction
The global development in industrialization increases energy consumption, resulting
in the continued usage of fossil fuels. Rising carbon dioxide emissions from the combustion
of fossil fuels have accelerated global warming, causing major environmental concerns.
Electricity generated from renewable energy sources like photovoltaic (PV), wind,
or biomass might take the place of fossil fuel-based electricity generation. The design
of a grid-connected converter, as depicted in Fig. 1, is critical for such systems because it supplies stable AC power to the grid and
loads from the DC power of the renewable energy source.
The increasing use of power semiconductor devices is known to cause serious problems
in the electric power system. The power semiconductor devices play an important role
in the power converters but are the most fragile component (1). As a result, the power semiconductor components have a significant impact on the
dependability of power converters, which necessitates a high-reliability standard
for the power converter. According to (2), the failures in a power semiconductor device are mostly caused by thermal stress,
accounting for 60\%. ATC is a recently introduced technique for adjusting power losses
and thermal stress. The basic approach is to use thermal control to influence the
power losses within the power converter. The reduction of the power losses results
in a decrease in the average temperature and thermal cycles. Using a classification
approach based on converter types, the author in (3) examines the ATC techniques by classifying ATC according to the various converter
kinds. Adjusting converter's switching frequencies are the most common and easy method
for achieving thermal management, which has a direct influence on power losses without
significantly affecting the power system's operational state. By reducing the operation
switching frequency, the switching losses of the converter decrease, resulting in
the reduction of power losses (3). A different ATC scheme is to modify the modulation methods. This strategy is usually
applied in multilevel converters where the available redundant switching states reduce
the switching losses or conduction losses (3). The discontinuous PWM (DPWM) is a well-known strategy whose basic concept is to
clamp the voltage reference of the converter output to the upper or lower DC-link
potential for a predetermined interval so that the corresponding power device maintains
its status without switching, thereby mitigating switching loss in that interval.
Due to the reduced power device loss during the clamping time, the junction temperature
mean value and variance are also lowered. Switching frequency is typically used to
influence converter junction temperature, which can have a negative impact on output
performance in low-level systems. The ATC strategies using a modified modulation method
can reduce the power loss of the specific stressful device but are complicated. However,
these ATC approaches do not take into consideration that the phase legs of the converter
can have different aging conditions. For power semiconductors, manufacturing techniques
have been recognized as a possible source of failure. It leads to a typical distortion
in the quality of the components, which results in different aging conditions. In
order to improve the performance of the three-phase converter, phase legs are supplied
with varying amounts of electricity based on their respective aging conditions. The
goal is to avoid or postpone as much as possible any failures. The aging condition
can be observed (4), and the power is distributed accordingly, whereby more aged phase leg is supplied
less power or even no power. In this way, the power-dependent failures in the phase
legs are delayed. In the case that the phase leg can just supply part power to the
corresponding load, the grid will provide the rest power of the load, as shown in
Fig. 1. This is equivalent to unbalanced load condition in practical situations (5). As a critical component of power distribution, the three-phase converter should
be able to provide an unbalanced load in islanding mode and deliver regulated AC voltages
due to this evolution of the custom power concept for each phase leg.
Fig. 1. Typical configuration of the grid-connected system.
There are two primary topologies for a three-phase four-wire system that improves
a three-phase converter power quality while it is operating in an unbalanced operation.
In comparison to the three-leg converter, the three-phase four-leg converter feeds
a three-phase unbalanced load more efficiently because the midpoint of the additional
leg and the neutral point of the three-phase load is connected to a common point,
allowing the negative sequence current path to be controlled by dual current controllers,
and the topology is straightforward
(6). This topology, however, may pose an electromagnetic compatibility issue for the
neutral point, which is susceptible to high-frequency voltage transitions. In addition,
the fourth leg control system is sophisticated, and it cannot be operated separately.
As mentioned in the literature, the corresponding controller is likewise challenging
to develop. The split dc-link capacitor three-phase converter can also accomplish
a three-phase four-wire structure, enabling a loop for the neutral current, which
increases the converter capability to feed unbalanced loads
(7). When the neutral point voltage is steady, the split dc-link capacitor three-phase
converter can be considered equivalent to three single-phase half-bridge converters,
which can be used to decouple the three-phase converter. Thus, the three-phase can
be controlled separately, effectively inhibiting the output voltage asymmetry induced
by an unbalanced load
(8). However, the existence of the capacitor voltage division problem caused by the neutral
current flowing into the capacitor or the difference between the two capacitors limits
the range of applications for this split dc-link capacitor, and a higher capacitance
is required as the load unbalanced degree increases. It can be seen that a three-phase
four-leg converter and a three-phase four-wire converter have both advantages and
disadvantages characteristics in the power system. In order to resolve the disadvantages
mentioned above of previous three-phase four-wire converters, a balance circuit is
introduced to the two-level three-phase four-wire converter, combining the former
two, i.e., split DC link with the additional fourth leg. The balance circuit is controlled
to balance the capacitor voltages and reduce the capacitor voltage ripple by forcing
the neutral current to flow through the balance circuit inductor
(9). This allows significantly reducing the capacitance of DC-link capacitors. Leakage
current passes through the ground and the photovoltaic array in a solar system, resulting
in increased radiated electromagnetic emissions, increased current harmonics and losses,
and reduced reliability of converter topologies. The neutral connection between the
neutral load and the mid-point of DC-link capacitors contributes to the reduction
of terminal voltage ripple to the ground. It means that with this topology, the leakage
current is significantly reduced
(10,11).
Fig. 2. Configuration circuit of the two-level three-phase four-wire converter with
balance circuit.
The three-phase legs of the two-level three-phase four-wire converter are totally
decoupled, allowing for independent control of each phase converter. This allows for
the implementation of ATC through unevenly loading each phase of the converter. By
changing the output power of each phase in accordance with the deterioration of power
semiconductor components, the lifetime of converter can be prolonged, as indicated
earlier. In this paper, an ATC approach based per-phase power control in the two-level
three-phase four-wire converter with balanced circuit is presented. An MPC approach
for the two-level three-phase four-wire converter with balanced circuit is investigated
in terms of per-phase control capability. An improved balance circuit control scheme
is proposed using a modified cost function with a neutral current feed-forward term
to reduce the capacitor voltage ripple. Additionally, a CBPWM-based controller using
per-phase voltage and current control is also presented. Both MPC and CBPWM-based
controllers are investigated in this paper regarding regulating power capability in
each phase leg. This paper presents the comprehensive study of the two-level three-phase
four-wire converter with balance circuit. The ATC method, including MPC and CBPWM-based
controllers, is investigated to evaluate the capability of regulating phase leg output
power depending on its aging condition. The simulation and experiment are conducted
to confirm the operation of the ATC strategy based per-phase control in the two-level
three-phase four-wire converter with balance circuit.
2. Description of the two-level three-phase four-wire converter with balance circuit
2.1 Structure and Operation
The topology of the two-level, three-phase, four-wire converter with a balance circuit
examined in this study is depicted in Fig. 2. This system contains a DC source, a two-level three-phase converter, and a balance
circuit. Compared with the conventional two-level three-phase converter, the neural
point of load is connected to the midpoint of the DC-link capacitors. As shown in
Fig. 2, the converter is supplied by a DC voltage $V_{dc}$ energized by any distributed
sources, such as fuel cells, photovoltaic devices, and wind turbines. The converter
outputs are connected to a three-phase $L$ filter and $R$ load. The three-phase converter
line-to-neutral voltages are denoted by $v_{a}$, $v_{b}$, and $v_{c}$. The three-phase
converter output currents are $i_{a}$, $i_{b}$, and $i_{c}$.
According to different switching patterns, the two-level three-phase four-wire converter
has two operation statuses per phase, as in the conventional two-level three-phase
converter. The voltage of the phase leg $v_{x}(x=a,\:b,\:c)$ respected to the neutral
point can be obtained as where $S_{x}$ denotes the phase leg
Table 1. Available switching states and corresponding output voltage of the two-level
three-phase four-wire converter.
Voltage vectors
|
Switching state
|
Voltage
|
$S_{a}$
|
$S_{b}$
|
$S_{c}$
|
$v_{a}$
|
$v_{b}$
|
$v_{c}$
|
$v_{1}$
|
-1
|
-1
|
-1
|
$-\dfrac{Vdc}{2}$
|
$-\dfrac{Vdc}{2}$
|
$-\dfrac{Vdc}{2}$
|
$v_{2}$
|
-1
|
-1
|
1
|
$-\dfrac{Vdc}{2}$
|
$-\dfrac{Vdc}{2}$
|
$\dfrac{Vdc}{2}$
|
$v_{3}$
|
-1
|
1
|
-1
|
$-\dfrac{Vdc}{2}$
|
$\dfrac{Vdc}{2}$
|
$-\dfrac{Vdc}{2}$
|
$v_{4}$
|
-1
|
1
|
1
|
$-\dfrac{Vdc}{2}$
|
$\dfrac{Vdc}{2}$
|
$\dfrac{Vdc}{2}$
|
$v_{5}$
|
1
|
-1
|
-1
|
$\dfrac{Vdc}{2}$
|
$-\dfrac{Vdc}{2}$
|
$-\dfrac{Vdc}{2}$
|
$v_{6}$
|
1
|
-1
|
1
|
$\dfrac{Vdc}{2}$
|
$-\dfrac{Vdc}{2}$
|
$\dfrac{Vdc}{2}$
|
$v_{7}$
|
1
|
1
|
-1
|
$\dfrac{Vdc}{2}$
|
$\dfrac{Vdc}{2}$
|
$-\dfrac{Vdc}{2}$
|
$v_{8}$
|
1
|
1
|
1
|
$\dfrac{Vdc}{2}$
|
$\dfrac{Vdc}{2}$
|
$\dfrac{Vdc}{2}$
|
switching state, i.e., $S_{x}=1$ when the upper switch is ON, and $S_{x}=-1$ when
the upper switch is OFF. The three control signals $S_{a}$, $S_{b}$, $S_{c}$ form
a total of 8 (23) switching states. The available switching states with the corresponding
output voltage in each phase are presented in
Table 1.
The balance circuit is composed of two switches ($S_{r1}$ and $S_{r2}$) and a balance
circuit inductor $L_{N}$ and two split DC-link capacitors ($C_{1}$ and $C_{2}$), as
shown in the dotted box of Fig. 2. The balance circuit is controlled to stabilize the capacitor voltages. $S_{r1}$
and $S_{r2}$ are managed to balance capacitor voltages and cause the neutral current
to flow through the balance circuit inductor.
2.2 ATC for each phase leg of the two-level three-phase four-wire converter with balance
circuit
As stated previously, the two-level three-phase four-wire converter has to drive both
single-phase and three-phase loads. In a practical system, the three-phase load is
not completely balanced too. This leads to different aging conditions in each phase
leg and the lifetime of the converter. To overcome the problem of different remaining
lifetimes of phase legs, ATC strategy based per-phase control is needed to extend
the lifetime of the more aged phase legs. The objective is to allow the more aged
phase legs work properly until the next maintenance is implement by stressing them
less than the other ones. Fig. 3(a) shows the case that phase legs $b$ and $c$ are the most aged legs, meaning that two
phase legs $b$ and $c$ have low remaining lifetime with respect to phase $a$. To ease
the aging condition of phase leg, the corresponding output power in phases $b$ and
$c$ can be decreased to reduce the power losses, as shown in Fig. 3(a). It lowers the thermal stress of the power semiconductor devices in the phase legs
$b$ and $c$, resulted in increasing lifetime. Fig. 3(b) shows the case that phase $c$ is the most aged leg. The same procedure is realized
by reducing the output power of phase $c$ to increase the corresponding lifetime.
In Fig. 3(c), assuming that three phase legs have different aging condition, resulted in different
output power.
Fig. 3. ATC strategy by adjusting the output power of phase leg (a) Phases $b$ and
$c$ are the most aged leg, (b) Phase $c$ is the most aged leg, (c) Phases $a$, $b$,
and $c$ have different aging conditions.
3. MPC-based controller
The two-level three-phase four-wire converter with balance circuit, connected to $R-L$
load, shown in Fig. 2, is considered to generate the mathematical model of the converter. As indicated
previously, the relationship among the switching states and load voltage is expressed
in (1). The three control signals $S_{a}$, $S_{b}$, $S_{c}$ from a total of 8 (23) switching
states. The valid switching states with the corresponding output voltage in each phase
are presented in Table 1.
Assuming that the capacitor voltages are balanced at nominal value $V_{dc}/2$. Each
phase of the two-level three-phase four-wire converter can be analyzed separately.
To implement ATC by adjusting the output power in each phase leg, the MPC scheme is
realized in $abc-$ frame instead of $dq-$ frame as conventional MPC. By applying the
Kirchhoff voltage law to Fig. 2, the output voltages of the converter can be described corresponding to $R-L$ load
and the output currents ($i_{a}$, $i_{b}$, $i_{c}$).
The continuous-time expression for the output current can be derived from
(2) as
Based on this,
eq. (3) can be expressed in a discrete-time model by predicting the next sampling instant
of output current following
As shown in
(4), the predicted output current in $k+1$ sampling instant requires the output current
and output voltage in $k$th instant. The output voltage prediction is generated using
the 8 switching states and the DC-link voltage, as indicated in
(1).
In the final stage, the predicted output currents in $i_{x}(k+1)(x=a,\:b,\:c)$ are
compared to the corresponding reference $i_{a}^{*}(k+1)(x=a,\:b,\:c)$, in a natural
reference frame making use of cost function $g_{1}$ as follows:
The optimal switching state is one that minimizes the cost function $g_{1}$. The cost
function $g_{1}$ includes three terms regarding three phases output currents. Different
current references can be used to regulate the output power in the corresponding phase
leg. It is the same as in indicated per-phase power current control method, the current
reference for the most aged phase leg can be even reduced to zero to decrease the
power losses. The reliability and lifetime of the converter will be improved.
Fig. 4. Structure of MPC of the two-level three-phase four-wire converter with balance
circuit.
The $k+1$ sampling instant of the output current references, is obtained by making
use of the fourth-order Lagrange extrapolation formula as:
The balance circuit can be described using two differential equations. One is the
balance circuit inductor current dynamics, and the other is capacitor voltage dynamics.
They are expressed as follows:
where $L_{N}$ is the balance circuit inductance, $C_{1}$ and $C_{1}$ are the upper
and lower DC-link capacitors, respectively. The balance circuit inductor current $i_{\ln}$,
the capacitor voltages $v_{C1}$ and $v_{C2}$, the capacitor current $i_{C}$, and neutral
current $i_{N}$ correspond to variables defined in
Fig. 2. The neutral voltage $v_{N}$ respected to negative DC-link can be obtained as $v_{N}=V_{dc}\times
S_{r}$, where $S_{r}$ denotes the balance circuit switching states, i.e., $S_{r}=1$
when the upper switch is ON, and $S_{r}=0$ when the upper switch is off. Considering
the constraint of DC-link voltage $v_{C1}+v_{C2}=V_{dc}$, $i_{\ln}$ and $v_{C2}$ are
selected as the state variables. Equations
(7) and
(8) can be expressed as:
The discrete-time equations of the balance circuit inductor current and lower capacitor
voltage can be achieved from
(9), as demonstrated below:
It should be noted that the goal of controlling balance circuit is to maintain the
upper and lower capacitor voltage at nominal value. The lower capacitor voltage $v_{C2}$
is compared with the corresponding reference $v_{C}^{*}=V_{dc}/2$, using the cost
function
Fig. 5. Structure of MPC for balance circuit control with neutral current feed-forward
term.
The MPC scheme of the two-level three-phase four-wire converter with balance circuit
is depicted in
Fig. 4.
Regarding balance circuit control, it is noticed that the neutral current $i_{N}$
impacts on the capacitor voltages. Consequently, a neutral current feed-forward term
is presented and added to the cost function in (14) to form a modified cost function as follows:
where $K_{i}$ is the gain of the neutral current feed-forward term. Different value
of $K_{i}$ will be to evaluate the effect of neutral current feed-forward term on
the capacitor voltage ripple. The structure of MPC for balance circuit control with
neutral current feed-forward term is depicted in
Fig. 5.
Fig. 6. Structure of per-phase voltage current control
4. CBPWM-based controller
4.1 Three-phase converter using per-phase voltage current control
The two-level three-phase four-wire converter can be treated as three equivalent single-phase
half-bridge circuits, and a
Table 2. Parameters of the two-level three-phase four-wire converter with balance
circuit in simulation.
Parameters
|
Simulation
|
DC-link voltage $V_{dc}$ (V)
|
720
|
Balance circuit inductor (H)
|
2.7m
|
Split DC-link capacitor (F)
|
50u
|
Load inductor (H)
|
10m
|
Load resistor (Ω)
|
15
|
Sampling frequency (kHz)
|
50
|
Fig. 7. Structure of balance circuit control (a) Voltage control, (b) Voltage current
control.
per-phase control approach can be straightly employed. The per-phase control offers
flexible power control to allow resilience to variable power demand. Additionally,
the output power in each phase can be adequately adjusted to extend the lifetime of
each phase leg based on the aging condition. In this study, a per-phase voltage current
control for the two-level three-phase four-wire converter is illustrated in
Fig. 6, where the generic $x$-th phase, $x=a,\:b,\:c$, is presented. The control scheme
includes two parts: voltage and current based PI controllers. The error signal between
measured voltage and voltage reference is fed into a modified PI controller to generate
the corresponding phase current reference. The modified PI controller is produced
by multiplying the integral part by the sign and the unity of corresponding phase
voltage reference, as shown in
Fig. 6. The output current control is achieved with a modified PI controller to generate
the switching pattern for each phase, as shown in
Fig. 6.
4.2 Balance circuit control
As stated earlier, the control of capacitor voltages is decoupled from the control
of the three-phase converter. Maintaining the capacitor voltages balance at a nominal
voltage value $V_{dc}/2$ is the crucial task, especially in unbalanced load conditions
or different output powers for ATC implementation. A PI controller controls the balance
circuit to generate a PWM pattern for $S_{r1}$ and $S_{r2}$. As for voltage control
only, shown in Fig. 7(a), the voltage error between the reference capacitor voltage and the lower capacitor
voltage sensed signal is regulated by using a PI controller to keep the capacitor
voltages balanced generate the switching pattern for the balance circuit switches.
In voltage and current control for balance circuit, shown in Fig. 7(b), the output of the voltage control part is used as a reference for the current control
part. The sensed signal of balance circuit current is controlled using a PI controller
and generates a switching pattern for the balance circuit switches.
5. Verification results
The simulation results verify the MPC scheme for the two-level three-phase four-wire
converter with balance circuit is presented in this section. As depicted in the previous
section, the two-level three-phase four-wire converter with balance circuit is connected
to $R-L$ load, as shown in Fig. 2, which is modeled in PSIM software. The MPC is implemented following the given structure
in Fig. 4. The converter system is characterized by the following parameters listed in Table 2.
Fig. 8. Simulation waveforms of output currents, capacitor voltages and neutral current
using MPC in unbalanced operation. (a) Pa = Pb = 3.3kW, Pc = 1kW, (b) Pa = 3.3kW,
Pb = Pc = 1kW, (c) Pa = 3.3kW, Pb = 2kW, Pc = 1kW.
In this simulation, three unbalanced operation cases are considered regarding ATC
for each converter phase leg. The simulation results of the two-level three-phase
four-wire converter with balance circuit using MPC under unbalanced operation where
the output power in each phase is different, are presented in
Fig. 8. In the first case, assuming that phase legs $c$ is the most aged, the output power
of that phase $c$ should be decreased to reduce the power losses and extend the corresponding
lifetime.
Fig. 8(a) shows the waveforms of output currents, capacitor voltages, neutral current, and
measured output power in the first case. In this case, the output power in phase $a$
and $b$ is kept the same at 3.3kW, whereas the output power in phase $c$ is reduced
to 1kW. The output current waveforms are sinusoidal with the correct magnitude and
phase following the output power. The capacitor voltages are kept balanced at the
nominal value $V_{dc}/2$. The simulation waveforms of the second case, where phase
$b$ and $c$ are the most aged, is presented in
Fig. 8(b). In this case, the output power in phase $b$ and $c$ is reduced to 1kW, whereas the
output power in phase $a$ is 3.3kW. As can be observed. the output currents are sinusoidal
with the correct magnitude and phase following the output power. The capacitor voltages
are kept balanced at the nominal value $V_{dc}/2$. In the third case, assuming that
three phase legs have different aging conditions, resulted in different output power
as shown in
Fig. 8(c). Apparently, the output currents are sinusoidal with the correct magnitude and phase
following the output power where Pa = 3.3kW, Pb = 2kW, and Pc = 1kW.
Fig. 9. Simulation waveforms of output currents, capacitor voltages and neutral current
using MPC with modified cost function of balance circuit control $K_{i}=-0.2$ in unbalanced
operation. (a) Pa = Pb = 3.3kW, Pc = 1kW, (b) Pa = 3.3kW, Pb = Pc = 1kW, (c) Pa =
3.3kW, Pb = 2kW, Pc = 1kW.
Fig. 10. Simulation waveforms of output currents, capacitor voltages and neutral current
using CBPWM based voltage current control in unbalanced operation. (a) Pa = Pb = 3.3kW,
Pc = 1kW, (b) Pa = 3.3kW, Pb = Pc = 1kW, (c) Pa = 3.3kW, Pb = 2kW, Pc = 1kW.
Fig. 11. Experimental waveforms of output current and capacitor voltage using CBPWM
based voltage current control in unbalanced operation. (a) Pa = Pb = 500W, Pc = 250W,
(b) Pa = Pb = 250W, Pc = 500W, (c) Pa = 500W, Pb = 300W, Pc = 50W.
As shown in
(12), the modified cost function is used to decrease the capacitor voltage peak-to-peak
value, as indicated earlier.
Fig. 9(a) - (c) show the simulation results using a modified cost function with the gain $K_{i}=-0.2$.
As can be observed, the output currents are the same as using the conventional cost
function in
(11). It can be seen that the capacitor voltage peak-to-peak value is reduced by about
20\% in both three cases of unbalanced operation. Meanwhile, the output currents are
sinusoidal with correct phase and magnitude as using the conventional cost function.
Therefore, the modified cost function of balance circuit control reduces the capacitor
voltage ripple without deteriorating the converter performance.
The simulation results show that the ATC approach based MPC scheme can solve the unbalanced
operation and extend the lifetime of a specific phase leg by adjusting the corresponding
output power. The two-level three-phase four-wire converter with balance circuit operates
correctly in both three cases of unbalanced operation.
The simulation results of the two-level three-phase four-wire converter with balance
circuit using CBPWM based per-phase voltage current control under unbalanced operation
where the output power in each phase is different, are presented in Fig. 10(a) – (c). Similar to the MPC scheme, in both three cases of unbalanced operation, the output
currents are sinusoidal with correct phase and magnitude corresponding to the output
power.
The experiment is conducted using a down-scaled prototype of the two-level three-phase
four-wire converter with balance circuit. Fig. 11(a) – (c) shows the experimental waveforms of output currents and lower capacitor voltage in
different cases of output power. As can be observed, the output currents are sinusoidal
with correct magnitude and phase following the output power. The capacitor voltage
is maintained at the nominal value $V_{dc}/2$.
The simulation and experimental results show that the per-phase power current control
scheme can solve the unbalanced operation and serve as an ATC strategy to extend the
lifetime of a specific phase leg by adjusting the corresponding output power.
6. Conclusion
This paper presents the ATC approach to regulate the output power in each phase leg
depending on the aging condition for the two-level three-phase four-wire converter
with balance circuit. The output power in each phase is controlled to reduce the stress
for the most aged phase leg. The MPC strategy is for the two-level three-phase four-wire
converter with balance circuit. The three-phase converter cost function includes three
separate current control terms, which allows independently regulating the output power
in each phase by changing the current reference. This contributes to increasing the
lifetime of desired phase leg by adjusting the corresponding power losses. Using the
neutral current feed-forward term, the modified cost function is proposed to improve
the capacitor voltage control capability by reducing the capacitor voltage ripple
without deteriorating the converter performance. The CBPWM based per-phase voltage
current control is investigated in terms of regulating output power in each phase
depending on the aging condition. The simulation and experiment results show that
ATC approaches based MPC and CBPWM operated correctly in unbalanced operation.
Acknowledgements
이 논문은 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구임 (2020R1A2C1013413).
References
S. Yang, A. Bryant, P. Mawby, D. Xiang, L. Ran, P. Tavner, May 2011, An Industry-Based
Survey of Reliability in Power Electronic Converters, in IEEE Transaction on Industrial
Application, Vol. 47, No. 3, pp. 1441-1451
L. Gopi Reddy, L. Tolbert, B. Ozpineci, May 2015, Power Cycle Testing of Power Switches:
A Literature Survey, in IEEE Transaction on Power Electronics, Vol. 30, No. 5, pp.
2465-2473
M. H. Nguyen, S. Kwak, Dec 2020, Enhance Reliability of Semiconductor Devices in Power
Converters, in Electronics, Vol. 9, No. 12, pp. 2068
S. Yang, D. Xiang, A. Bryant, P. Mawby, L. Ran, P. Tavner, Nov 2010, Condition Monitoring
for Device Reliability in Power Electronic Converters: A Review, in IEEE Transaction
on Power Electronics, Vol. 25, No. 11, pp. 2734-2752
G. Carrasco, C. A. Silva, R. Pena, R. Cardenas, Jul 2015, Control of a Four-Leg Converter
for the Operation of a DFIG Feeding Stand-Alone Unbalanced Loads, in IEEE Transaction
on Industrial Electronics, Vol. 62, No. 7, pp. 4630-4640
Z. Liu, J. Liu, J. Li, May 2013, Modeling, Analysis, and Mitigation of Load Neutral
Point Voltage for Three-Phase Four-Leg Inverter, in IEEE Transaction on Industrial
Electronics, Vol. 60, No. 5, pp. 2010-2021
P. Verdelho, G. D. Marques, Oct 1998, Four-wire current-regulated PWM voltage converter,
in IEEE Transaction on Industrial Electronics, Vol. 45, No. 5, pp. 761-770
A. Mohd, 2010, Control strategy and space vector modulation for three-leg four-wire
voltage source inverters under unbalanced load conditions, in IET Power Electronics,
Vol. 3, No. 3, pp. 323
J. Liang, T. C. Green, C. Feng, G. Weiss, 2009, Increasing Voltage Utilization in
Split-Link, Four-Wire Inverters, in IEEE Transaction on Power Electronics, Vol. 24,
No. 6, pp. 1562-1569
T. Kerekes, R. Teodorescu, M. Liserre, C. Klumpner, M. Sumner, Sep 2009, Evaluation
of Three-Phase Transformerless Photovoltaic Inverter Topologies, in IEEE Transaction
on Power Electronics, Vol. 24, No. 9, pp. 2202-2211
W. Zhao, X. Ruan, D. Yang, X. Chen, L. Jia, Apr 2017, Neutral Point Voltage Ripple
Suppression for a Three-Phase Four-Wire Inverter With an Independently Controlled
Neutral Module, in IEEE Transaction on Industrial Electronics, Vol. 64, No. 4, pp.
2608-2619
저자소개
응웬 민 호앙(Minh Hoang Nguyen)
Minh Hoang Nguyen received the B.S. degree in electrical and electronics engineering
from Hanoi University of Science and Technology, Vietnam, in 2016.
He is currently pursuing the M.S and PhD combined degree in electrical and electronics
engineering with Chung-Ang University, Seoul, South Korea.
His research interests are control for multilevel converters.
Sang-Shin Kwak received the Ph.D. degree in electrical engineering from Texas A&M
University, College Station, TX, USA, in 2005.
From 1999 to 2000, he was a Research Engineer with LG Electronics, Changwon, South
Korea.
From 2005 to 2007, he was a Senior Engineer with Samsung SDI R&D Center, Yongin, South
Korea.
From 2007 to 2010, he was an Assistant Professor with Daegu University, Gyeongsan,
South Korea.
Since 2010, he has been with Chung-Ang University, Seoul, South Korea, where he is
currently a Professor.
His current research interests include the design, modeling, control, and analysis
of power converters for electric vehicles and renewable energy systems as well as
the prognosis and fault tolerant control of power electronics systems.