Classification of Troubleshooting in a Mechanical System for Fault Detection and Diagnosis with the aid of a Neural Network

The internal combustion engine (ICE) is widely used in applications such as automobiles, motorcycles and ships. After its long-term use faults occur that degrade its performance or cause it to malfunction. Therefore, ICE fault detection and diagnosis (FDD) research is important for preventing serious economic loss and even human injuries caused by undetected faults. The development of an ICE FDD for the prediction of faults is described in this work. The setup uses sensors to measure the ICE variables and LSTM techniques implemented in a computer program. The common ICE faults which are the common rail injection fault, the fuel consumption and the coolant temperature faults are carefully studied. The objective of the FDD is to determine if there is any of the above-mentioned faults in the Hyundai Santa fe 2008 model car. Several FDD algorithms are proposed, one category of which is based on data processing techniques such as the LSTM Recurrent Neural Network is implemented to arrive at our results. This category of FDD algorithms includes the LSTM-based FDD algorithm. The LSTM-based FDD algorithm introduces a new FDD index based on LSTM and statistics recorded from the car. According to the included experimental results, all of these algorithms are capable of detecting and locating these faults with 99.999% accuracy.