Optimizing a Hybrid Controller for Automotive Active Suspension System by Using Genetic Algorithms With Two High Level Parameters

Abstract

This article presents the in-depth research results concerning the actively controlled suspension system following the Hybrid Active Suspension System (HASS) in vehicles. Initially, detailed introductions are provided for the controller and actuator of this suspension system model. Afterward, the HASS model is proposed for application to the Active Suspension System with integrating Skyhook and Groundhook control methods, wherein a coefficient $\alpha =0.3,\,0.6,\,0.9$ is utilized to adjust the correlation between these two models. Through the amplitude transfer function, the transition in enhancing ride comfort and road holding criteria in the HASS is demonstrated. Subsequently, the coefficients of the HASS are optimized through the Genetic Algorithm optimization with the parameter $\beta =0.1,\,0.5,\,0.9$ , aiming to enhance ride comfort and road holding corresponding to each value of $\alpha $ . The simulation and evaluation results utilizing the HASS in both frequency and time domains demonstrate significant improvements in ride comfort and road holding criteria compared to passive suspension systems. Specifically, the ride comfort has improved from 60% to 80%, while the road holding has improved from 10% to 15%. These findings facilitate the development of applications for the controlled suspension system, allowing adaptation to diverse real-world conditions at varying levels.

Keywords

• Vehicle dynamics and control
• active suspension system
• hybrid active suspension system
• Skyhook model
• Groundhook model
• genetic algorithm optimization