Fault Detection and Classification of Spark Ignition Engine Based on Acoustic Signals and Artificial Neural Network

Internal combustion (IC) engines have been used in many transportation and manufacturing applications. Therefore, early detection of malfunctions in engines is the most significant issue to help to avoid causalities and further damage. The present research analysis is aimed to detect the faults of the spark ignition engines using acoustic signals. Initially, the fault detection with the aid of digital computers and an automated diagnosis based on a system using the artificial neural network (ANN) is proposed. The acoustic signal are picked up from the engine as features to feed the network at four speeds 1000, 2000, 3000, and 4000 rpm. Then, acoustic signals from the spark-ignition engine with four strokes four cylinders have the misfire with one spark plug and two misfire fault simulated and are tested with the proposed network. Finally, the ANN system evaluated and classified the faults of the spark-ignition engine. It is found that the results are proved an excellent potential for using acoustic signal from the engine with ANN.