A multilayer perceptron approach for condition monitoring of automobile suspension system through vibration signal

The suspension system is a vital component of car, with the aim improving stability and good road handling during vehicle movement in uneven conditions. A variety load conditions, extended operation, ongoing absorption shocks time-based deterioration internal parts can lead to defects. Faults destroy inner components that allow malfunction jeopardize safety cars. Condition inspection has now been an integral part identification defects device. This paper focuses on device status control by vibration signal analysis errors are classified using computer classification systems. test was conducted Ford EcoSport vehicle, which accelerometer mounted near system's stationary portion. In order derive data, captured used. An additional J48 decision tree algorithm used selection process classify most important features. type fault occurring from chosen features discovered Multilayer Perceptron (MLP) classifiers. MLP model provided 98% error rate 0.016 as prediction performance.