A Machine Learning Based Engine Error Detection Method

Nowadays the fault of automobile engines climb due to the growth of automobiles. Traditional mechanical automobile testing is not efficient enough. In this paper, the Machine Learning based Engine Error Detection method (MLBED) is proposed for the complex nonlinear relation and operation parameters of automobile engine operating parameters such as large scale data, noise, fuzzy nonlinear etc. This method is a fault diagnosis and early warning method designed on the basis of self-organizing neural network, Elman neural network and probabilistic neural network. The experimental results show that MLBED has a great advantage in the current fault detection methods of automobile engine. The method improves the prediction accuracy and efficiency.