Modeling of Optimized Neuro-Fuzzy Logic Based Active Vibration Control Method for Automotive Suspension

Abstract

In this thesis, an active vibration control system was developed. The control system was developed and tested using a quarter car model of an adaptive suspension system. For active vibration control, an actuator was implemented in addition to the commonly used passive spring damper system. Due to nature of unpredictability of force required two different fuzzy inference system (FIS) were developed for the actuator. First a sequential fuzzy set was built, that resulted lower vertical displacement compared to basic damper spring model, but system had limited effect with disturbances of higher magnitude and continuous vibrations (rough road). To improve the performance of the sequential fuzzy set, the main fuzzy set was improved using an adaptive neuro fuzzy inference system (ANFIS). This model increased the performance substantially, especially for rough road and high magnitude disturbance scenarios. Finally, the suspension’s spring constant and damping co-efficient was optimized using a genetic algorithm to further improve the vibration control properties to achieve a balance of both ride stability and comfort. The final result is improved performance of the suspension system.