Performance Comparison of Slim Drive with ANFIS Controller

Abstract

Normally speed control of a Single-Sided Linear Induction Motor (SLIM) by an indirect vector control scheme is difficult because the motor's parameters are time-dependent and the performance depends on various factors such as end effect, saturation, location of primary losses, and iron losses. Traditional PI current regulators are commonly used in vector regulators, but there is a tuning problem due to the oscillation of an operating point. This problem can be overcome by substituting an adaptive neuro-fuzzy-based current controller, and this controller improves the operation of a SLIM, such as its motor speed and thrust force. In this adaptive neuro-fuzzy controller, the ID and IQ errors and the error delay are inputs, and its outputs are Vds and Vqs, respectively. It is trained based on available values. A SLIM's dynamic modelling is implemented by dividing current (I) and flux-linkages into two terms. In these two terms, one is dependent on the end effect, and the other is independent of the end effect. The function of a Voltage Source Inverter (VSI)-fed indirect vector-controlled SLIM drive is simulated in MATLAB/Simulink, and its operation under various operating conditions is studied using an adaptive neuro-fuzzy current controller. These results are compared to a traditional P-I controller. The Pulse Width Modulation (PWM) technology that is used for controlling the VSI is called Space Vector Modulation (SVM).