Coefficient Optimization Based on Genetic Algorithm for Hybrid Control of Semi-active Suspension System

Abstract

Composed of sky-hook control and ground-hook control, hybrid control used for a semi-active suspension system is of great significance for enhancing vehicle comfort, stability, and safety. The former hybrid control is highly difficult in coefficient selection to lead to has a low degree of improvement in suspension performance. In this paper, based on a genetic algorithm, a coefficient optimization method (COGA) is presented for hybrid control of a semi-active suspension system. A genetic algorithm is used to optimize the selection of suspension control system coefficients. Computer simulation of the quarter-vehicle suspension system model is conducted by MATLAB/SIMULINK software. The vehicle comfort and handling characteristics of the suspension systems are predicted under a random road input simulation model. Passive suspension system performance is compared with the semi-active suspension system using hybrid control. The experimental results show that the semi-active suspension hybrid control system with COGA has a minimal root-mean-square value of sprung mass acceleration, suspension dynamic deflection, and dynamic tire load. It can be proved that COGA can effectively coordinate and ameliorate the function of hybrid control.