Performance Analysis of a PMBLDC Motor Drive based on ANFIS Controller and PI Controller

Abstract

Adaptive Neuro-Fuzzy Inference System (ANFIS) is an interesting choice of research that is a combination of two soft-computing methods of Artificial Neural Network (ANN) and Fuzzy Logic. This paper designs and describes a control system based on Adaptive Neuro-Fuzzy Inference System for Permanent Magnet Brushless DC (PMBLDC) Motor Drive. It is observed that PMBLDC motor is complex to handle for their multi-variable and nonlinear system. The motor speed and torque control are frequently needed for controlling various drives such as robotics, copter, electric vehicles and similar other drives applications. But it is complicated to control by using conventional PI controller and tuning is necessary to achieve desired performance. In order to overcome these problems, the ANFIS based controller is proposed and developed. The PI controller is also designed and tuned by Ziegler-Nichols method. The drive performance is tested under different operating conditions such as starting condition, sudden load torque changes, speed variation and parameter changes in C++ simulation environment. The results of ANFIS controller are compared with those obtained through PI controller.