1. Trang chủ
  2. » Kỹ Thuật - Công Nghệ

A Control Strategy for Dynamic Voltage Restorer

5 22 0
Tài liệu đã được kiểm tra trùng lặp

Đang tải... (xem toàn văn)

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Định dạng
Số trang 5
Dung lượng 1,46 MB

Các công cụ chuyển đổi và chỉnh sửa cho tài liệu này

Nội dung

This paper presents a configuration and a control strategy for dynamic voltage restorer (DVR). In order to compensate the voltage of each phase separately, a closedloop PI control law in the dq reference frame is proposed. The proposed method provides a fast response and effective sag compensation capabilities. In addition, in order to detect voltage sag, a linear Kalman filter is employed toestimate threephase voltages. By using Kalman filter, the voltage sag can be detected faster than other conventional methods. Therefore DVR can compensate voltage sag quickly and accurately. The obtained results that are simulated in Matlab Simulink indicate that the proposed method can mitigate the balanced and unbalanced voltage sag types efficiently in the distribution networks.

Trang 1

IEEE PEDS 2015, Sydney, Australia

9 – 12 June 2015

A Control Strategy for Dynamic Voltage Restorer

University of Science and Technology – The University of Danang Quynhon University

dtviet@ac.udn.vn, huuhieu019@yahoo.com, nglehoa79@gmail.com nmkhoaqnu@gmail.com

Abstract-This paper presents a configuration and a control

strategy for dynamic voltage restorer (DVR) In order to

compensate the voltage of each phase separately, a closed-loop

PI control law in the d-q reference frame is proposed The

proposed method provides a fast response and effective sag

compensation capabilities In addition, in order to detect voltage

sag, a linear Kalman filter is employed to estimate three-phase

voltages By using Kalman filter, the voltage sag can be detected

faster than other conventional methods Therefore DVR can

compensate voltage sag quickly and accurately The obtained

results that are simulated in Matlab/ Simulink indicate that the

proposed method can mitigate the balanced and unbalanced

voltage sag types efficiently in the distribution networks

I INTRODUCTION

According to the IEEE defined standard (IEEE Std

1159-1995), a voltage sag is defined as: “an rms variation with a

magnitude between 10% and 90% of nominal voltage and

duration between 0.5 cycles and one minute” An example of

voltage sag event is shown in Fig 1 Voltage sag is usually

caused by faults such as short circuit, heavy load

switching,… Voltage sags are the most important power

quality problem facing industrial customers today [1]

Equipment used in many modern industrial plants (process

controllers, programmable logic controllers, adjustable speed

drives) actually become more sensitive to voltage sags as the

complexity of the equipment increases and the equipment is

interconnected in sophisticated processes [1] Even relays and

contactors in motor starters can be sensitive to voltage sags,

resulting in shut down of the process

The use of custom power devices is one of the most

efficient methods to mitigate voltage sag There are many

custom power devices Each of which has its own benefits

and limitations [5] Among the several novel custom power

devices, DVR is now becoming more established in industry

to mitigate impact of voltage disturbances on sensitive loads

DVR is a series-connected device designed to maintain a

constant RMS voltage across a sensitive load [2], [4] The

basic principle of DVR is to inject a voltage of required

magnitude and frequency, so that it can restore the load

voltage to the desired amplitude and waveform even when the

source voltage is unbalanced or distorted Generally, it

employs a gate turn off thyristor solid state power electronic

switches in a pulse width modulated (PWM) inverter

structure DVR can generate or absorb independently

controllable real and reactive power at the load side In other

words, DVR is made of a solid state DC to AC switching power converter that injects a set of three phase AC output voltages in series and synchronism with the distribution line voltages

Fig 1 A voltage sag event.

In this paper, a new configuration and control strategy for DVR is proposed Its idea is based on the minimization of the energy supplied from DVR irrespective of the balance of the three-phase voltages supplied to the load In the control strategy of DVR, close-loop with PI control based on d-q reference frame in order to generate reference voltage and Kalman filter is used to detect volage sag quickly and accurately

II APROPOSED METHOD OF DVRCONTROL

A Configuration of DVR

In this study, a proposed configuration of DVR includes control unit, energy storage unit, voltage source converter (VSC), LC filter and series transformer as shown in Fig 2

Fig 2 Configuration of DVR

Trang 2

B Control Strategy of DVR

Control strategy is one of the most importance units of

DVR, it consists of voltage sag detection unit in order to

switch between standby and active modes accurately, and to

generate reference voltage based on Kalman filter to

compensate voltage sag in supply voltage A detail of the

proposed control strategy for DVR is presented as follows

1 Phase Locked Loop (PLL):

The PLL is a widely used mechanism which tries to track

the phase of the incoming signal Synchronizing with the

signal can be achieved by comparing the phase difference

between the incoming signal and reference signal generated

from a voltage-controlled oscillator via a phase detector and

through a loop filter A phase locked loop is shown in Fig 3

Where γ is an instant angle of the supply voltage and θ is a

angle of PLL The PLL tracks the positive sequence

component of the supply voltage and PLL angle is used to

transform from α-β reference frame to d-q reference frame

and vice versa

Fig 3 A PLL to synchronize DVR to supply voltage.

2 Using Kalman Filter to Detect Voltage Sag

The Kalman filtering has been recognized as a powerful

state estimation technique [3] It is a model based on the

optimal estimator with minimum error covariance Given

observed data, a Kalman filter is described by a set of

dynamic process (i.e., state) equations and a set of

measurement (i.e., observation) equations as follows:

k k k k

k k k k

where xk is the state vector and zk is the measurement at time

index k; Φk is the state transition matrix; Hk is the

measurement matrix; and wk and vk are the model and

measurement errors, respectively The model structure

required by the Kalman filter is flexible to allow the

measured signal to be represented in many ways based on

different assumptions

To overcome the disadvantages of the conventional

methods for detecting voltage sag based on d-q frame, an

algorithm to detect voltage sag based on Kalman filter [3] is

employed in this paper In this method, three Kalman filters

are used to track three-phase voltages, so that both balanced

and unbalanced voltage sags can be detected

Fig 4 shows a single phase diagram to detect voltage sag

by Kalman filter, where the input of Kalman filter is

one-phase voltage signal v(t), the outputs of Kalman filter consists

of the estimated phase angle θ(t) and the estimated voltage

magnitude Vm(t) To detect voltage sag, the estimated voltage

magnitude Vm(t) is compared to the reference voltage

Vref=1(pu) and then compared by hysteresis comparator to generate voltage sag detection signal

Fig 4 Voltage sag detection using Kalman filter

Fig 5 A balanced voltage sag detection; a) A 15% balanced voltage sag; b) Conventional dq-based method; c) Kalman filter-based method

Fig 6 An unbalanced voltage sag detection; a) A 15% unbalanced voltage sag; b) Conventional dq-based method; c) Kalman filter-based method

Fig 5 shows the capacity of recognising the 15% balanced voltage sag starting at the time of 0.2 second and 0.3 second

In this case, both methods can recognise voltage sag however

Trang 3

Kalman filter-based method is clearly better than

conventional dq-based method as the time for recognising is

shorter

However, in term of 15% unbalanced voltage sag, these

methods give completely different results which are showed

in Fig 6 It can be seen that conventional d-q method (Fig

6.b) cannot recognise the 15% unbalanced voltage sag

(“shallow” voltage sag) but Kalman filter method (Fig 6.c)

3 Control Methodology

A space vector control has been applied to generate the

reference voltage as shown in Fig 7 Where VSabc is the

supply voltage; VLabc is the load voltage; θPLL is phase angle

of PLL; Standby/Active is signal of standby or active mode

from Kalman filter algorithm VL.ref(dq) is reference voltage in

d-q set (VL.refd=1pu; VL.ref.q=0) G1÷G12 is pulse signal of

IGBTs 1 to 12

Space vector control is applied to control DVR, therefore

three-phase voltage are converted to d-q reference frame in

(2):

Fig 7 Control strategy of DVR

2

a d

b q

c

V V

V V

V

(2)

The DVR measures the three-phase supply voltages and

three-phase load voltages, then transform the voltages to the

dq-system The PLL uses the three-phase supply voltages and

generates the filtered PLL angle The angle of the PLL (θPLL)

is used in the transformation to the rotating dq-system The

DVR controls the load voltages to be in-phase with the supply

voltages by setting the d-reference voltage to 1pu and the

q-reference voltage to zero The error between the actual supply

voltages and the wanted load voltages is calculated and is

feedforward to PI control and the PWM signals to the three

full bridges is generated During normal supply voltages the

DVR is in standby mode, and the DVR-reference voltages are set to zero

III RESULTS AND DISCUSSIONS

A Simulation Model on Matlab/Simulink

DVR is modeled and simulated using Matlab/ Simulink software The power circuit and control system model of DVR is shown in Fig 8 and Fig 9, respectively In these diagrams, the load is supplied by a supply source that may appear voltage sag DVR's power circuit (including the serial transformers, filters, inverter) is connected between the source and the load The parameters of this system are chosen

as shown in Table I

Fig 8 Simulation model of DVR on Matlab/Simulink

Fig 9 Control system for DVR

TABLE I

PARAMETERS OF THE S IMULATION M ODEL

Supply source 22 kV; 50 Hz Distribution transformer 100 kVA; 22/0,4 kV;

Δ/Y 0 -11 Load 48 Ω Series transformer Ratio 1:1; 1 kVA

LC filter 18 μF; 10 mH Voltage source converter H-bridge converter

DC source voltage 150 Vdc Switching frequency 10 kHz Sampling time 25 s

B Simulation Results

1 A Balanced Voltage Sag

We suppose that a balanced three-phase voltage sag in the source with 50% magnitude starts at 0.05s and ends at 0.15s The supply voltages are shown in Fig 10(a) and Fig 10(b) When t<0.05s, there is no voltage sag in the system, therefore the DVR is set in the standby mode In this mode, the voltage

on DVR (Fig 10.c, d) is the voltage drop in the serial transformer At t=0.05s the voltage sag appears in the supply

Trang 4

voltage, then Kalman filter will generate voltage sag

detection signal after a delay time (Fig 10.g, h) Therefore

DVR will change its operation mode to the active mode to

compensate the missing voltage at the load side so that load

voltages are maintained at the rated voltage (Fig 10.e, f)

Fig 10 A balanced voltage sag; a-b) Supply voltage; c-d) DVR voltage; e-f)

Load voltage; g-h) Signal of voltage sag detection of phase A by Kalman

filter

2 An Unbalanced Voltage Sag

In this case, we assume the voltage sag appears in

one-phase of 22kV source side with 50% magnitude starts at

t=0.05s and ends at t=0.15s The 22/0.4 kV distribution

transformer structured connection windings is Δ/Y0-11

therefore phase A and phase B are in low voltage while phase

C is still in normal voltage (Fig 11.a, b) In this case, Kalman

filter will detect voltage sag on phase A and phase B Then

DVR will change its operation mode to the active mode in

order to maintain the load voltage at the nominal voltage

In both cases of the balanced and unbalanced voltage sag

are mentioned above Using fast Fourier transform to analysis

the frequency spectrum of the compensated voltage signals at

load side by DVR for a period of one cycle of the load voltage signal The results of the frequency spectrum of phase

A load voltage when balanced voltage sag occurs at source side is THD=1.17%, as shown in Fig 12 Fig 13 shows the result of frequency spectrum of the phase A load voltage when unbalanced voltage sag occurs at supply side is THD=1.34% In both cases showed the second order frequency components appear in load voltage due to switching IGBT valves in the voltage source inverter but this value is not significant

Fig 11 An unbalanced voltage sag; a-b) Supply voltage; c-d) DVR voltage; e-f) Load voltage; g-h) Signal of voltage sag detection of phase A by Kalman filter

IV CONCLUSION

Kalman filter is used to quickly and accurately detect voltage sag compared with dq transformation traditional method used in many other studies In this paper, authors use three Kalman filter to detect the three-phase voltage sag individually Another advantage of the proposed DVR configuration that is capable of saving energy in compensation mode by just compensates phases that appear

Trang 5

voltage sag In standby mode, using inverter’s IGBT to

bypass circuit the secondary side of the serial transformer so

no need to use additional bypass switches

The results were verified in simulation model in

Matlab/Simulink The balanced voltage sag and unbalanced

voltage sag were simulated in this paper to examine the

performance of proposed DVR The simulation results show

that the proposed DVR ensure to compensate voltage sag to

maintain the load voltage at rated voltage

Fig 12 Frequency contents of phase A during balanced voltage sag

ACKNOWLEDGMENT

The authors would like to thank National Foundation for

Science and Technology Development of Vietnam and

Quynhon University for the financial support to carry out this

work

Fig 13 Frequency contents of phase A during unbalanced voltage sag

REFERENCES

[1] M.H Bollen, Understanding Power Quality Problems: Voltage Sags

and Interruptions, New York: IEEE Press, Vol 1, 2000

[2] J.G Nielsen and F Blaabjerg, “A detailed comparison of system

topologies for dynamic voltage restorers,” IEEE Transactions on

Industry Applications, vol 41, no 5, pp.1272-1280, 2005

[3] M.H.Bollen andI.Y.H.Gu, Signal Processing Of Power Quality

Disturbances, Wiley- IEEE Press, 2006

[4] F BadrkhaniAjaei, S.Afsharnia, A Kahrobaeian, and S Farhangi, “A fast and effective control scheme for the dynamic voltage restorer,”

IEEE Transactions on Power Delivery, vol 26, no 4, pp

2398-2406,2011

[5] C Benachaiba and B Ferdi, “Voltage quality improvement using

DVR,” Electrical Power Quality and Utilisation, Journal,vol XIV, no

1, 2008

Ngày đăng: 13/11/2021, 15:23

TỪ KHÓA LIÊN QUAN

TÀI LIỆU CÙNG NGƯỜI DÙNG

TÀI LIỆU LIÊN QUAN