DIAGNOSTICS AND CLASSIFICATION OF FAULTS IN DIESEL ENGINE COMPONENTS USING TIME-FREQUENCY APPROACH AND MACHINE

Abstract

Diagnosing engine component faults is a challenging task for every researcher due to the complexity of engine operation. Developed faults in engine components subsequently reduce their performance and lead to increased maintenance costs. Therefore, it is necessary to implement an effective condition monitoring method to diagnose faults in engine components. Therefore, this work presents potential fault diagnosis techniques for detecting and diagnosing scuffing defects occurring in diesel engine components. Condition monitoring techniques such as vibration and acoustic emission analysis were used to obtain signals associated with faults. These signals were analyzed in time domain, frequency domain and time-frequency domain using signal processing techniques viz. fast Fourier transform ( FFT ) and short time Fourier transform ( STFT ). Statistical feature parameters were also extracted from the received signals to diagnose fault severity. Additionally, artificial neural network (ANN) models have been developed to predict and classify scuffing defects occurring in engine components. The results showed that FFT and STFT methods provided better diagnostic information. The developed neural network models effectively classified scoring defects on engine components with 100% accuracy.