Abnormal Vibration Triangulation Modelling Methods in Internal Combustion Engines

Abstract

This paper presents a comprehensive review of
vibration analysis techniques for fault detection in internal combustion engines (ICEs). The use of vibro-acoustic signals has been pivotal in diagnosing complex issues related to ICE sub-components such as the pistons, bearings, and turbochargers. Traditional signal analysis methods, including Fourier transforms, wavelet analysis, and empirical mode decomposition (EMD), have been evaluated alongside advanced computational techniques like support vector machines (SVM) and artificial neural networks (ANN). The findings suggest that combining multiple domains of signal analysis (time, frequency, and angular domains) offers a robust mechanism for detecting and diagnosing faults. Furthermore, the potential integration of these techniques with real-time monitoring systems is discussed.