Power Quality Improvement based on VSHDE Algorithm Incorporating Shunt Active Power Filter

The usage of the word Power quality in recent times acquired intensified interest due to the complex industrial processes. The usage of intelligent tools to improve power quality is increasing day by day, as assumption of present day power system as a linear model is unsatisfactory. This paper deals with analysis of Differential Evolution (DE), Hybrid Differential Evolution (HDE) and Variable Scaling Hybrid Differential Evolution for harmonic reduction in the source current with optimal tuning of PI controller gain values. Shunt Active power Filter is one of the better solution to suppress the source current harmonics which are induced into power system because of nonlinear loads. Current controller called HBCC is considered for gating operation of switches in Voltage Source Inverter. The Intelligent tuned PQ theory is used for reference current generation. The then obtained compensating currents are injected at point of common coupling for current disturbance mitigation. Simulations of MATLAB/SIMULINK environment of the present work shows the efficacy.


Introduction
The widespread use of switching equipment in domestic and industrial purpose is very much noticeable.This electronic equipment, driving energy transit equipment is the major source of voltage and current variations which might lead the power quality issues like harmonics and sags/swells in the power utility systems.The presence of these PQ issues may diminish and cause low power factor, flicker voltage distortion and electromagnetic interference etc.Such adverse effects are usually addressed by power engineers and researchers [1], [2].If these PQ not addressed at right time then these may often make system to malfunction and finally tend to complete power system shutdown.
The power quality issues under current variations are first discussed by L. Gyugyi [1] in 1970's.The current variation issue can be addressed using power quality analyzers and synthesizers, i.e. shunt active power filters.H. Akagi [2] is other prominent researcher who stressed active filters dealing power quality variations and their measurement in processed control and power utility systems.With the advent of flexible AC transmission devices in the transmission line, which reduces the operational cost without increasing the installation cost the power engineers and researchers are also looking for such transit solutions to take care of the system in distributed side.These are usually called as custom power devices (CPDs).One such cost effective solution to reduce current variations is shunt active power filter.The researchers have looked into various factors of topological structures, sizing factors, optimal placement and control factors of SAPF.These have left few gaps like inappropriate compensation, switching related problems in this direction of work.Hence researchers focused on better filter architecture to sort of this problem.The performance of the system is measured and standardized universally by different international bodies like IEEE, IEC which have set the limits for various power quality issues and events [3].As conventional low power and medium power SAPF is not fit to resolve present day power quality problems regarding effective compensation [4] and appropriate switching.Hence, engineers looked into MLI based architecture [5] instead of concentrating on control and placement issues, which sorted out compensation related issues.The present paper focus on the further enhancement capabilities of such developed cascaded MLI based SAPF (MLISAPF).The reference current generation and respective switching mechanism is discussed by PQ theory and Hysteresis Band Current Controller (HBCC) [6].The control strategy is optimized with Particle swarm Optimization [7]- [9] which is good in search space problems.The self-tuned PSO called APSO is proposed and applied for enhancement of the concerned issues further.The control variables are gains of the PI controller which makes the conventional PQ theory evolved as Intelligent Computing active and reactive power (ICPQ) theory.Using this technique, the current variations aspect of the system can be improved.In MATLAB/SIMULINK, dynamic simulations are conducted to compare the results with those from other cases (other artificial intelligence techniques).The MLISAPF working phenomenon is illustrated in Figure 1.The APF's developed by Akagi, are good enough to ease harmonics in the utility system with improvised mechanism [2], [5].
The PI controller in the control strategy addresses harmonic content mitigation.The PI controller needs the calculation of proportional and integral gain values.Usually for power systems with linear loads, methods like Ziegler-Nichols, Cohen-Coon help in findig the KP and KI values emphirically.These methods of linear control do not work for system with nonlinear loads.
Later, Storn and Price [11] introduced another evolutionary computing based algorithm called DE to solve engineering electric system problems.M.Basu [15] applied DE to solve PQ issues incorporating FACTS devices.In later recent years, some hybrid algorithms with extended features have been developed to enchance and explore search efficiency.Thus developed HDE Algorithm [16], [17] sorted out large capacitor placement problem in utility system.The research in this direction is promising and encouraging as far as with obtained results.This paper employs an improved Hybrid algorithm called Variable Scaling Hyrid differential Evolution (PSODE) algorithm [18], [19] to solve the power quality issues discussed.This algorithm is the hybridization and modification done to DE algorithm.The SAPF as a FACTS device which is incorparated to adress the power issues.A SAPF connected to utility power system mitigates the concerned power quality disturbance.The harmonic disturbances, which are introduced into the system, are eased with the help of SAPF by injecting reverse harmonic currents into the power system.The PQ theory is used for reference currents and voltage generation.These reference currents and voltages are utilized for generation of gating pulses by the Hystersis Band Current Controller.The PI controller in PQ theory with intelligent and optimal tuning take care of harmonic mitigation.The optimal tuning of PI controller gains is done with the proposed VSHDE optimization technique.Later is compared with conventional PI controller [20].The active filter is further planned to enhanced with photovoltaic's as discussed in [21].The single objective optimization is planned to enhanced as multi objective optimization as discussed in [22].

Shunt Active Power Filter (SAPF)
A sample case study with 3-ϕ, 415V, 50Hz supply is constructed to discuss the complicated process control system with nonlinear load.The SAPF intended to ease current disturbances needs the discussion of the following design parameters: 2.1.Specifications of SAPF A 3-Ø three leg six pulse shunt active filter with line to line voltage of 415V is considered.The DC bus Synchronizing voltage and its capacitor capacitance calculation is made as follows:

Synchronizing DC Bus Capacitance
The capacitance (C dc ) is calculated with: Where, Vdc and Vdc1 are the reference and minimum level of dc bus voltages respectively, 'a' loading factor set as 1.2, V and I are phase voltage and current respectively and t is recovery time of dc bus [5].

Synchronizing DC Bus Capacitor Voltage
Upon the measurement of three phase voltages (VLL) and with specified modulation index m the dc bus voltage can be obtained as follows [5]:

PQ Theory
The PQ theory is explained clearly with the help of the Fig. 2. PQ theory deals with conversion of three phase (a, b, c) voltages and currents to two-phase (α, β, 0) using the direct conversion and once again the inverse phenomenon backed with a filter [2][3][4][5].The Butterworth filter extract the fundamental component that are required for reference current or voltage generation.The source voltages and load currents are sensed and from which the required reference currents are developed.The illustration, development and design is clearly shown with expressions in Fig. 2.
The active and reactive power components are illustrated as follows: The reference currents (iabc*) and the actual injection harmonic currents (iabc) are sensed first, the error between these are supplied to HBCC which intern generates the gate pulses required by the VSC.The HBCC of the proposed work is as follows: Fig. 3. HBCC controlling switches of VSI.The outputs of HBCC (S1-6) are the gating signals to the six switches of VSC, which is shown in Fig. 3.
The concept of the VSHDE is to use the 1/5 success rule for the variable scaling factor.The updating scaling factor following the above rule is as follows [18], [19] : where t S ρ is the regularity of winning mutations measured.
Whenever the scale factor is too small then it is reset to find the improved result in the elucidation process.
The flow chart representation of VSHDE algorithm is as shown in Fig. 4.

Problem definition
The minimization objective function is defined as follow: subjected to gains of PI controller.
The PI controller mathematical model is given as : 6 ( ) The gains K P and K I of PI controller are tuned by the VSHDE algorithm as shown in Fig. 5.

Simulation Results
The power quality issue particularly concentrated in this paper are source current harmonics.The current harmonics are mainly caused due to nonlinear loads.The nonlinear load phenomenon is implemented with the help of power electronic converter (diode bridge rectifier) with RL load.The MATLAB/Simulink representation of SAPF for PQ enhancement is shown in Fig. 6.The compensating current generated by the IPQ theory is shown in Fig. 7(b).The graphical illustration of reduction of harmonics in the system is shown in the Fig. 7(c).3, Table 4 and Table 5 respectively.The fitness value of THD is reduced to 1.32% with DE, further to 1.31% with HDE and 1.30% with VSHDE (The value after 10 runs each) subjecting Kp and Ki ranging between 0 to 200 (By Expert system) varying CR and F in case of DE and HDE whereas varying CR and randomly choosing F in case of VSHDE .Where CR and F are crossover and mutation coefficients of proposed algorithms.The optimization is run for 100 iterations.The corresponding fitness convergence graphs with DE algorithm are shown in Fig. 8, HDE algorithm are shown in Fig. 9 and VSHDE algorithm are shown in Fig. 10 respectively.The suitable crossover rate (CR) for better optimization is 0.9 and Mutation constant (F) is varying between 0.5< F<1 which is evaluated by experimentation validation.Table 6 proves the superiority of VSHDE algorithm over other methods in reducing the THD value.Though the methods in Table 6 also got THD values less than 5 % as prescribed by IEEE standards [3], it is always desirable to reduce THD as much as possible.This minimizes the heating of various equipment and mechanical stress on bearings and shafts of motors.This reduction in THD is achieved with optimizing the PI parameters and without any additional device or cost.

Fig. 2 .
Fig.2.PQ Theory for Reference Current Generation.Where − − q p , : Fundamental active and reactive power components.

Fig. 7 .
Fig. 7.The simulation results with VSHDE Algorithm showing a) source current with harmonics before compensation.b) SAPF reverse injection compensating currents.c) source current after compensation.

Fig. 10 .
Fig.10.Convergence of fitness (source current THD) with VSHDE Algorithm making with CR Variable and F randomly chosen by the algorithm.

Table . 2
Control parameter convergence of source current THD in SAPF with DE Algorithm

Table . 3
Control parameter convergence of source current THD in SAPF with HDE Algorithm

Table 5
Control parameter convergence of source current THD with Different Algorithms

Table 6
The comparison of source current THD with other methods The performance analysis of SAPF based on DE, HDE and VSHDE techniques is presented and compared in the present work.The Comparative analysis of DE, HDE and VSHDE techniques shows the superiority of VSHDE over others.The limitations of DE and HDE are overcome in VSHDE.The rate of convergence and searching process is better in VSHDE when compared with DE and HDE.The stability of dc bus voltage maintaining constant voltage equal to the reference voltage is phenomenal.The efficacy of the proposed algorithm is shown with MATLAB simulation results.