PID Controller based on BP Neural Network for Speed Control of Electric Vehicle

Abstract

In electric vehicles (EV), one or more electric motors are operated by energy stored in rechargeable batteries. In response to the increased interest in EVs, research into their modelling and simulation has operational variables alter depending on driving conditions, making it difficult to retain control. In the MATLAB/Simulink environment, the transfer function model of the EV is used for design and analysis purposes. In this work, the advanced Back Propagation Neural Network-based Proportional Integral Derivative (BPNN-PID) controller is designed to control the speed of the EV. To identify the effectiveness of the BPNN-PID controller the two conventional controllers fuzzy and PID are used. The error metrics are used to analyse the controller performance. The error metrics employed in this work are Integral Square Error (ISE), Integral Absolute Error (IAE), and Integral Time Absolute Frror (ITAE).