Intelligent current-speed soft-starting controller for an induction motor drive system

Abstract

In industry, three-phase induction motors (IMs) are the most used motors. IMs’ high starting supply current and large accelerating torque make their starting process a potential source of unnecessary energy waste, especially when it is repeated. To overcome the starting problems of an IM, an intelligent soft-starter controller based on an artificial neural network (ANN) is proposed. The controller contains two soft starting control schemes namely, current and speed controls. That allows one to select the soft starting control scheme based on the load requirements and/or supply utility capability. Current and speed controls are to keep the motor's starting current and accelerating torque constant at a specified level, respectively. That is achieved by selecting the appropriate firing angles for the thyristors in the soft starter. The online determination of the thyristor firing angles is resolved through the application of the ANN approach. First, the ANN models for two control techniques are developed using a MATLAB Neural Network Toolbox for training. Based on the current-speed and torque-speed characteristics, two ANN models are trained for a range of thyristor firing angles. Secondly, the ANN models of two control schemes are implemented using MatLab/Simulink for a 3kW three-phase IM. A variety of simulation tests prove that a validation on control technique's effectiveness under various load conditions (no-load, pump, constant, and linear loads) and reference values exists. The least inrush current and smooth acceleration are obtained with two control schemes.