System Identification and Neural Network Based PID Control of Servo - Hydraulic Vehicle Suspension System

This paper presents the system identification and design of a neural network based Proportional, Integral and Derivative (PID) controller for a two degree of freedom (2DOF), quarter-car active suspension system. The controller design consists of a PID controller in a feedback loop and a neural network feedforward controller for the suspension travel to improve the vehicle ride comfort and handling quality. Nonlinear dynamics of the servo-hydraulic actuator is incorporated in the suspension model. A SISO neural network (NN) model was developed using the input-output data set obtained from the mathematical model simulation. Levenberg-Marquardt algorithm was used to train the NN model. The NN model achieved fitness values of 99.98%, 99.98% and 99.96% for sigmoidnet, wavenet and neuralnet neural network structures respectively. The proposed controller was compared with a constant gain PID controller in a suspension travel setpoint tracking in the presence of a deterministic road disturbance. The NN-based PID controller showed better performances in terms of rise times and overshoots.