Intelligent machine learning models For speed control of bldc motor

Abstract

In brushless direct current (BLDC) motors speed control is a prominent operation based on which these motors are widely used in higher end industrial applications including robotics, aeronautics, disk drives, factory automation, consumer electronics, transport and military applications. This thesis is intended to develop novel intelligent machine learning models to accomplish effective speed control of the BLDC motor with the set specifications. The developed machine learning models are in the hybrid version of the neural network architectural models, type 1 and type 2 fuzzy modules and a stochastic population based optimization technique. The effective and significant features of all these individual models are brought out and combined together to carry out better speed regulation and perform varied application sectors of BLDC motor efficiently. The developed intelligent controllers based proportional integral derivative controller are simulated and analysed to attain the performance characteristics of the motor. newlineOn analysing the various conventional controller designs, it has been identified there is always a need and requirement for better controller models to accomplish enhanced control action for the considered specification of the BLDC motor. For all the developed modules, respective gain metrics are tuned with the novel intelligent hybrid techniques to carry out most operational speed regulation process for the motor mechanism. The significant research contributions made in this thesis are presented as follows. newline