Condition Monitoring of Automotive Suspension System using Machine Learning and Deep Learning Techniques

Abstract

Modern cars, including hatchbacks, sedans, and electric vehicles (EVs), commonly newlineuse the McPherson suspension system as their primary suspension system. This independent newlinesuspension system is known for its compact design and lightweight components, newlinemaking it suitable for various front-wheel drive cars. However, despite its newlinesignificance in ensuring comfort, driving performance, and road safety, the McPherson newlinesuspension system lacks a sufficient monitoring system for early fault detection. newlineTo address this research gap, the current study focuses on developing a data-driven newlineapproach for online condition monitoring of the suspension system, with the aid of newlinemachine learning and deep learning technologies to classify faults based on unique vibration newlinesignal patterns specific to each fault type. By investigating the performance of newlinethese approaches under different conditions, the study aim to enhance the reliability and newlinesafety of automotive suspension systems. newlineTo conduct this investigation, a specially designed laboratory setup to simulate the newlineworking of a quarter car suspension model. The setup subjects the suspension system newlineto uniform loads, uniform speeds on a flat surface, and introduces various fault conditions. newlineVibration signals collected for each specific fault condition are subsequently used newlinefor further analysis and processing. By employing an intelligent fault diagnosis system newlineinvolving machine learning and deep learning techniques, the proposed approach can newlineeffectively monitor and detect faults in the suspension system. This study has the potential newlineto contribute to improving the overall reliability of the suspension system by newlinetimely detection faults in the suspension component there by improving the safety of newlineautomotive vehicles. The steps carried out in the study that helped in formulating this newlinethesis are provided below. newlineFaults in the suspension system - The study identified seven critical faults in the suspension newlinesystem, including strut wear, ball joint wear, strut mount damage, lower arm newlinebush wear, strut extern