Dynamic Voltage Restorer for Mitigation of Voltage Sags Due to 3 Phase Motor Starts Based on Artificial Neural Networks

Abstract

The Direct On-Line (DOL) process of starting a high-power 3-phase induction motor causes voltage sags in the distribution system that is connected to one point of common coupling (PCC). Voltage sag can cause damage and failure of sensitive loads. This article analyzes and proposes a simulation of voltage sag recovery using a Dynamic Voltage Restorer (DVR) based on an Artificial Neural Network (ANN). ANN is used as a detector and regulator of the voltage compensation value. In this study, a 3-phase induction motor will be connected to a sensitive load, and the DVR will be placed in series with a voltage source or PCC with a sensitive load. The simulation test system uses Simulink-Matlab R2016a with different configurations of induction motor parameters. Based on the simulation results show that the parameters of the 3-phase induction motor cause the depth and duration of the voltage sag. DVR with ANN control can detect and compensate for a voltage sag of 0.5 pu so that the voltage will be normal to 1 pu.