Modeling and Performance Analysis of A DVR For Non-Linear Load
Modeling and Performance Analysis of A DVR For Non-Linear Load
2
H.O.D EPS, M.S.S CET, Jalna
Chandra6279@gmail.com
3
Asst. Professor,M.S.S.CET,Jalna
Pankajbhakre11@gmail.com
ABSTRACT: Dynamic voltage restorer (DVR) can provide the most commercial solution to mitigate voltage
sag by injecting voltage as well as power into the system. This paper describes the effectiveness of using dynamic
voltage restorers (DVRs) in order to mitigate voltage sags in power distribution systems at critical loads. The
DVR is a power electronic based device that provides three-phase controllable voltage source, whose voltage
vector (magnitude and angle) adds to the source voltage during sag event, to restore the load voltage to pre-sag
conditions. The DVR can restore the load voltage within few milliseconds. A control technique based on a
proportional–integral (PI) controller and a selective controller is used. The controller is designed in
asynchronously-rotating reference frame. In fact, three independent controllers (homo polar component, d-axis
component and q-axis component) have been used to tackle balanced and unbalanced voltage supplies.
Simulation results using Simulink’ Sim Power System Toolbox is presented to illustrate the principle and
performance of a DVR operation in load voltage compensation.
1. INTRODUCTION
At the present, with the increasing demand for electrical energy and rapidly growing number of new
production technologies, the voltage quality requirements are becoming stricter. In order to evaluate the level of
power quality, STN EN 50160 standard was introduced, which stipulates the limits for voltage quality [1].
Since the whole electrical power network represents a very complex structure, there is no way to control it
without any faults and disturbances. Thus, companies are often forced to save its facilities on their own. One of
the options for power quality and system stability improvement is to introduce FACTS devices [3].
FACTS controllers are able to control and regulate one or several key parameters in power transmission, such
as current, voltage, active, reactive power, frequency or phase angle. Reference [5] divides FACTS into four
basic types – series connected, shunt connected, combined series-series and combined series-shunt controllers.
The main disadvantage of implementing FACTS is very a high price of these devices and economic
requirements. The series controller can be a variable impedance, such as capacitor and reactor, or a power
electronics based variable voltage source. In general, all series controllers inject voltage in series with the line.
They are able to compensate voltage sags or swells and eliminate harmonic distortion as well. These are static
synchronous series controller (SSSC), thyristor-controlled series capacitor (TCSC) or dynamic voltage restorer
(DVR). As in the case of series controllers, the shunt controllers may be variable impedance, variable source, or
a combination of these. In basic principle, all shunt controllers inject current into the system at the point of
connection. These are static thyristor controlled reactor (TCR), synchronous controller (STATCOM) or static
var compensator (SVC). Combined series-shunt controllers are the most flexible and sophisticated FACTS
devices. They are able to regulate and affect many different parameters at the same time. One of these devices is
unified power flow controller (UPFC).
Dynamic Voltage Restorer (DVR) belongs to series connected FACTS controllers. The primary function of a
DVR is to compensate voltage sags and swells but it can also perform tasks such as harmonics elimination,
reduction of voltage transients and fault current limitation [6]. DVR is usually installed between a source and a
critical load that should be protected. Even the shortest voltage sag can cause serious equipment damage,
interruption of production cycles and thus financial losses as well.
In general a DVR consists of three parts (Fig. 1.):
measuring unit
control
power circuit
The measuring unit provides voltage and current measurements. The outputs are voltage and current analog
signals ( u, i), which enter the control unit. The control unit converts these signals to their digital representation
using A/D converter so that they can be processed by a microprocessor (DSP – digital signal microprocessor). Next
part is a voltage sags detection algorithm followed by compensating voltage calculation U_com, which is the
voltage needed to be injected into the system in order to remain the load side voltage of purely sinusoidal
waveform. Power section consists of a voltage source converter (VSC) equipped with a LC filter to smooth the
output voltage, a DC energy storage and an injection transformer (TR) – booster.
The basic principle of DVR function is to inject or draw the compensating voltage U_inj to or from the
supply voltage U_s in order to mitigate voltage sags or swells on the load side U_load. At every moment the
control algorithm compares desired voltage and actual measured voltage. The difference between these two
signals is considered as a compensating voltage signal (control signal) U_com, which is directly proportional to
compensating voltage U_inj (power circuit). U_com is a digital input signal for a pulse width modulation (PWM)
to control the voltage source converter. The VSC converts DC energy stored in an energy storage device (such
as batteries or supercapacitors) to injecting AC voltage that is to be superimposed to the source voltage. DVR
power output depends on the amount of energy that can be stored in the energy storage unit.
One of the most important requirements for DVR is that the controller should be able to operate in real-time
manner. It means that the whole compensating process is carried out immediately, after a failure occurs, without
any delay. The very important factor that influences the DVR speed most is the reaction time of the
implemented voltage sag detection algorithm. The best DVR systems are able to react within 1 ms.
There are several voltage sags detection techniques, which can be used in DVRs, such as:
peak value method,
RMS method,
DQ transformation,
Hybrid methods.
In this paper the modeling of a DVR is proposed. Its control algorithm is based on DQ transformation. DQ
transformation (dq0 – direct-quadrature -zero) is a mathematical transformation used to simplify the analysis of
three-phase circuits [8]. d and q quantities represent rectangular two axis system, which rotates with angular
frequency ω [7]. In the case of symmetric three-phase system, introducing of the dq0 transformation reduces
three AC quantities (pu) to two DC quantities (d=1, q=0) . Any deviations from the steady state condition in
abc system reflect in changes of dq0 values in real-time. For unbalanced and asymmetric three phase system
applies d≠0, q≠0, 0≠0. According to this presumption it is possible to obtain the difference between desired
and instant values dynamically. Therefore, the output compensating voltage can be controlled by PID regulators
[3]. The resultant signal is converted back to abc values. DQ transformation can be applied in the case of three-
phase system and there is a phase-locked loop (PLL) required to lock the synchronization of the compensating
voltage in phase with the line voltage before the fault.
In order to confirm the correct function of the proposed control algorithm it was necessary to create
simulation model. For this purpose Simulink was used. The model is presented in Fig. 2. It consists of a 0.38 kV
electrical source, which represents a 0.38 kV distribution network, a power line, a distribution transformer ( TR 1)
which feds the sensitive load. There is DVR connected to the line right before the load that is protected. The
different faults are generated by the block “fault”. The variable U_bef is the voltage which is measured before the
point of connection of the DVR and voltage U_load is the load side voltage that is measured at the load
terminals. The lower part of the simulation model is DVR. It consists of measuring unit (U_load), control and
power unit (voltage source converter, LC filter, series transformer and energy storage). When DVR is idle, the
energy storage device is being recharged. The maximum amount of energy stored at the rated voltage 380 V.
Fig. 2 shows the proposed control algorithm that is used for DVR automatic regulation of compensating
voltage. It starts with measured three-phase load voltage U_load. These values are transformed to dq0 system
and they are compared with references values. The difference enters PID regulator.
PID controller regulates compensating voltage U_com, which is transformed back to abc system and the
resultant voltage is sent to PWM module. And thus VSC is controlled by means of IGBTs’ switching pulses.
5. SIMULATION RESULTS
The simulation model of DVR is designed to mitigate voltage sags and phase angle changes as well. It
implements compensation method pre-sag which means that DVR is able to correct phase angle changes during a
fault. Phase angle of voltage waveform is kept constant at the value as it was before the sag. Many different
simulations with various types of sags have been tested.
6. CONCLUSION
To sum up, the presented DVR simulation model is based on a DQ transformation control algorithm. The
main advantages of this control technique are high speed, simplicity and it can be used even during distorted
supply voltage waveform.
On the other hand the disadvantages are that DQ transformation can be only applied on a three phase
system and a reference sinusoidal signal is required (PLL) synchronized with the voltage before the fault. The
entire DVR model was created in Matlab/Simulink and its correct function was verified by several simulation
tests. The obtained results showed that, the proposed DVR mitigates voltage sags and harmonics very fast and
reliably. Many other parameters and conditions influence the DVR operation. They are: sampling frequency,
speed of the regulators, dynamic behavior of energy storage device, converter switching speed and the type of
the disturbance. The right choice of a reliable voltage sag detection method and a proper design of the whole
system is the very first step in the developing of a DVR. Thus, the proposed simulation model can support the
next research in the field.
REFERENCES
[1] STN EN 50160, “Charakteristiky napätia elektrickej energie dodávanej z verejnej distribučnej siete,” Slovak technical
standard, SÚTN, Bratislava, 2010.
[2] D. Chapman, “Introduction to power quality,” European Copper Institute publication, February 2012.
[3] P. Hečko, “Kompenzace krátkodobých poklesů a přerušení napětí v distribuční soustavě,” University of Žilina, faculty of
electrical engineering, 2010. Dissertation thesis.
[4] IEEE, “Proposed terms and definitions for flexible AC transmission system (FACTS),” IEEE Xplore digital library,
October 1997. ISSN0885-8977
[5] G. N. Hingorani, L. Gyugyi, “Understanding FACTS. Concepts and Technology of Flexible AC Transmission Systems,”
New York: IEEE Press, 2000. p. 432. ISBN 0-7803-3455-8
[6] M. N. Tandjaoui, et al., “Sensitive Loads Voltage Improvement Using Dynamic Voltage Restorer,” International Conference
on Electrical Engineering and Informatics, 2011. Conference publication. IEEE Xplore digital library
[7] M. Gonzalez, “DQ transformation development for single-phase systems to compensate harmonic distortion and reactive power,”
Power Electronics Congress, 2004. CIEP 2004. 9th IEEE International. p. 177-182
[8] R. A. Kantaria, “A novel technique for mitigation of voltage sag/swell by Dynamic Voltage Restorer,” Electro/Information
Technology (EIT), 2010 IEEE International Conference. Conference publication. p. 1-4
[9] M. Minarčík and A. Otčenášová, “Practical measurement of power quality and the possibility of its improvement,” 12-th
International Scientific Conference Electric Power Engineering 2011, Kouty nad Desnou, Czech Republic, 2011, p. 478-481,
ISBN 978-80-248-2393