Design and Simulation of a Unified Power Quality Conditioner by Fuzzy Logic for Egyptian Power Grid Connected to Zafarana Egypt Wind System

DOI: http://dx.doi.org/10.24018/ejers.2019.4.9.1459 1  Abstract—Wind energy system is lately receiving a lot of attention, because they are cost inexpensive, environmental safe and clean renewable energy source, as compared with nuclear and fossil fuel power generation. The operational characteristics of wind electric turbines has considerable dissatisfaction and stress on the quality of electric power system. Harmonics, variations of voltage and reactive power are most of power quality issues for grid connected with wind turbine. This paper introduces a design and simulation of unified power quality conditioner using a fuzzy controller to improve the power quality for Egyptian power grid connected to Zafarana Egypt wind system. The proposed performance of the unified power quality conditioner system is verified by simulating the model using MATLAB/SIMULINK environment. The simulation results showed that the proposed unified power quality conditioner provide efficient cancellation of both load current harmonics and supply voltage sag in addition to compensation of reactive power, and thus making the electrical grid connected wind energy system more efficient by improving the quality of power.


I. INTRODUCTION
The rising prices and harmful effect of fossil fuel on the environment have strengthened worldwide attentions to sustainable energy sources.One of alternative energy sources is wind energy as pollution free and permanent source [1].With the increase in wind power penetration into the power grid, the power quality of the power system might be affected due to fluctuations generated by random nature of wind resources.These power fluctuation cause voltage variations with consequences for the electrical power grid and the consumers [2].
The increased use of power electronics in wind power turbines has resulted in generated harmonics and poor of power quality at the point of common coupling [3][4].It can cause troubles such as automatic resets, equipment damage and data errors.Power quality problems and elimination methods have emerged as significant challenges and issues confronting electric utilities and customers.
FACTS devices and various control schemes can be used to mitigate power quality problems.There are several types of FACTS devices such as dynamic voltage restorers (DVRs), distribution static compensators (DSTATCOMs), and unified power quality conditioners (UPQC).The introduction of these devices to electric power systems has led to improved power quality of the power system [5].DVR is completely appropriate to protect sensitive load from sag and swell, but doesn't take care of load current harmonics [6].The device STATCOM is widely used for the elimination of load current harmonics in addition to the compensation of reactive power, but it doesn't take care of voltage problems [7].
UPQC is the only appliance widely used for reactive power compensation and elimination of both supply voltage sag and load current harmonics.The UPQC consists of shunt active power filter SHAPF and series active power filter SEAPF, which is responsible for the compensation power quality distortions in both load side and source side [8].
This paper proposes UPQC using a fuzzy controller (FC) to improve the power quality for Egyptian power grid connected to Zafarana Egypt wind system.The FC is better suited for nonlinear loads, as it works with linguistic variables set theory and it does not need any mathematical model [9].Simulation Model of the system under study and its power quality problems before and after connecting a UPQC verified using MATLAB/ SIMULINK.

II. SYSTEM CONFIGURATION
Al-Zafarana wind farm is located at Red Sea coast in south east of Cairo, Egypt.The farm was designed and structured through seven stages of 30, 33, 30, 47, 80, 85 and 120 MW respectively.In this paper, the fifth stage of the farm is taken into account for simulation purpose.It is made up 100 variable speed wind turbine generators WTGs (with 850KW DFIG units for each WTG).Each WTG is connected to 690V/22kV step-up transformer.The collected electrical power is then fed to the Egyptian power grid 220 kV through three 75 MVA, 22 / 220 kV main step-up transformers as shown in Fig. 1.

A. Wind turbine model
According to wind turbine generators (WTGs) characteristics, a mechanical power is given by the following relations: Where: ρ is air density, A is swept area of a wind turbine, Vω is wind speed, Cp is power coefficient, β is blade pitch angle and λ is the ratio between the turbine angular speed and the wind speed.
Where: R is the turbine radius; ωt is the turbine rotational speed.The power coefficient Cp value is approximated according to non-liner function The dynamic voltages of the stator (V ds , V qs ) and rotor (V dr , V qr ) in d and q reference frame are given by : V qs = R s I qs + Pφ qs + ω s φ ds V dr = R r I dr + Pφ dr − ω r φ qr V qr = R r I qr + Pφ qr + ω r φ qr (8) The fluxes of the stator (φ ds , φ qs ) and rotor (φ dr , φ qr ) in d and q are related to identical d and q currents I ds , I qs , I dr , and I qr : φ qs = L s I qs + MI qr (10) φ dr = L r I ds + MI ds (11) The electromagnetic torque T em is determined by: Where: R s , R r are the stator and rotor resistances for each phase, L s , L r are the stator and rotor inductances for each phase, M is the magnetizing inductances, ω s , ω r are synchronous and rotor speed.

III. POWER QUALITY ISSUES IN WIND POWER SYSTEM AND IMPROVEMENT TECHNIQUES
Power quality problem is any change displayed in power, current, voltage, or frequency that causes a failure in customer devices which is caused due to non-sinusoidal wave forms [9].The addition of wind energy system into any power system causing many technical problems such as: voltage regulation, instability, harmonic and other problems, these power quality problems are to be limited to IEC and IEEE standards as in Table I.In order to meet the problems of power quality FACTS devices can be efficiently used.Due to availability of power electronic devices such as IGBTs, GTOs and IGCTs, FACTS technologies can control of active and reactive power.FACTS devices can be divided into four basic types: series type, shunt type, combined series-series type and combined series shunt type.The comparison and detailed description of different types of FACTS are discussed in the literature [10].Table II shows the services and performance level of each type of FACTS technologies.From this comparison, we can see that UPQC can perform effectively all of the function of FACTS devices such as voltage control, harmonics suppression, and reactive power compensation.

A. Unified power quality conditioner
Unified power quality conditioner is a power electronics controller device which is responsible for the mitigation of power quality distortions in supply side and load side.The design configuration of a UPQC is based on the connection of SEAPF and SHAPF, tied back-to-back with sharing a common DC capacitor.The SEAPF of the UPQC is responsible for compensation of voltage supply disturbances, flicker, voltage harmonics and voltage unbalance.The SHAPF of the UPQC is responsible for compensation of load current distortions, lagging power factor, load harmonic current, load unbalance.Figure 3 shows the schematic diagram of UPQC.

B. Control strategy of series active power filter
The control strategy of SEAPF is based on extraction of unit vector templates from the three phase distorted supply.Three phase source distorted voltage (Vsa , Vsb, Vsc) are sensed and multiplied by gain K = 1 V m to get unit input voltage vectors (Va,Vb,Vc), where Vm is maximum amplitude fundamental input voltage.The unit input voltage vectors are applied to the phase locked loop (PLL), the output angle (ωt) from the PLL is used to obtain two quadrature unit vectors (sin ωt, cosωt) using equation: V c = sin (ωt + 120) The maximum amplitude fundamental input voltage V m is multiplied by unit vector templates of (14-16) to obtain the reference voltage signals.
The source voltage sensed are compared with the reference voltage signals and then the error signals are delivered to the pulse width modulation (PWM) to generate the required gate signals for series converter as shown in Fig. 4.

C. Control strategy of shunt active power filter
Control strategy of the proposed shunt converter controller is based on instantaneous power theory.In this method, the source voltages a, V, and  and load currents , , and  are transformed into - frames using Clark transformation (18-19).After that, the instantaneous active and reactive power p and q are calculated using (20).
The DC signal represents the instantaneous fundamental power () and (), while the ripple represents harmonic power components are (ℎ) and (ℎ).Active harmonic components are extracted by using LPF.The reference current is computed in - frames   , Inverse Clark transformation is then applied to obtain desired reference current   using ( 22).
The load current sensed are compared with the reference current signals and then the error signals are delivered to the PWM to generate the demanded gate signals for shunt converter.as shown in Fig. 5.The FLC in order to change the numerical variables into linguistic variables uses the following fuzzy levels, which are: NB (negative big), NM (negative medium), NS (negative small), ZE (zero area), PS (positive small), PM (positive medium), and PB (positive big) as shown in Figs.7-9.The control surface of the proposed FC is shown in Fig. 10.
The FC is described as follows: 1) Seven fuzzy sets for input and output with triangular membership functions.
2) Fuzzification using continuous universe of discourse; 3) Implication using Madman's 'min' operator; 4) De-fuzzification using'centroid'algorithm.Fuzzification: It may be defined as the process of changing continuous values variables to linguistic variables.
Defuzzification: is the process of converting linguistic variables to crisp output.
Database: Database saves the definition of the membership functions demanded by fuzzifier and defuzzifier.
Rule Base: the contents of rule base table are specified depends on the theory that in the sudden state, large errors need intense control, which requires accurate input/output variables; in stable condition, small errors need soft control, which requires smooth input/output variables.Based on this the contents of the rule table are obtained as shown in Table III, with'Vdc' and 'Vdc-ref' as inputs.

IV. SIMULATION MODEL AND RESULTS
The detailed model of wind turbine unit constructed by induction generator based on the synchronous machine in MATLAB Simulink as shown in Fig. 11.Wind energy generating system connected to the Egyptian grid will be presented to study influence of UPQC device on power quality and its improvement.The proposed control circuits of SEAPF and SHAPF are illustrated in Figs.12-15.simulation of power quality problems with and without implementation of UPQC are shown below.This paper focuses on improving power quality for Egyptian power grid connected to Zafarana Egypt wind system.UPQC using FLC is proposed to compensate both voltage sag, current harmonics and reactive power.The simulation results of the studied system before and after connecting UPQC are obtained using Matlab/simulink software.From the simulation results, proposed UPQC with FLC have been offered proficient compensation for voltage sag, harmonics load current and reactive power.The UPQC reduces the THD of the load current from 31.41% to 4.15% and reduces the THD of the supply voltage from 7.26% to 1.02%, thus making the electrical grid connected wind energy system more efficient.

Fig. 2 .
Fig. 2. Basic configuration of doubly fed induction generator based wind energy conversion systems

Fig. 5 .
Fig. 5.Control strategy of shunt active power filter

Fig. 6 .
Fig. 6.Control technique of voltage error by fuzzy logic controller

Fig. 11 .
Fig. 11.Detailed model of Egyptian Power Grid Connected to Zafarana Egypt wind System

Fig. 22 .
Fig. 22. Voltage waveforms before connecting a UPQC Simulation of a Unified Power Quality Conditioner by Fuzzy Logic for Egyptian Power Grid Connected to Zafarana Egypt Wind System

TABLE I :
IEC AND IEEE POWER QUALITY EVENTS

TABLE II :
PERFORMANCE LEVELS OF FACTS DEVICES

TABLE III :
FUZZY RULE BASE