Artificial Neural Network Controlled DSTATCOM for Power Quality Improvement

Abstract: The paper intends to develop the artificial neural network control algorithm for the control of DSTATCOM for the improvement of power quality. The presence of nonlinear loads makes the voltage to be deviated and current to be distorted from its sinusoidal waveform quality. Thus harmonics elimination, load balancing and voltage regulation is the heavy task that has to be accomplished to maintain the quality of the power. The performance of any device depends on the control algorithm used for the reference current estimation and gating pulse generation scheme. Thus the artificial neural network based Back Propagation (BP) algorithm has been proposed to generate the triggering pulses for the three phase H bridge inverter (DSTATCOM).The fundamental weighted value of active and reactive power components of load currents which are required for the estimation of reference source current is calculated by using BP-based control algorithm. Based on the difference of the target voltage and the generated voltage, the triggering pulse for the inverter is obtained by the BP algorithm. Then the voltage is injected at the point of common coupling to compensate the reactive power. Thus by regulating the voltage and compensation of reactive power, the power quality can be improved. The simulation modelling of the Back propagation algorithm controlled DSTATCOM and the PWM controlled DSTATCOM and the comparative analysis of the algorithms is presented in this paper.

Index Term: DSTATCOM, Artificial Neural Network, Back propagation (BP) control algorithm, Reference current Estimation, Power quality.