Fault Detection Method for Ship Equipment Based on BP Neural Network

Abstract

Fault detection is of great importance for ship equipment's maintenance and repair, therefore, in this paper, we propose a novel fault detection method for ship equipment based on BP neural network. As the net error estimated by the ANN is lower than the current iteration, we use the back propagation algorithm to solve this problem. Hence, we introduce the BP neural network to detect fault for ship equipment. We take the VTC254P turbocharger as an example test the effectiveness, and six failures of turbocharger are utilized, such as oil leakage, surge, high temperature, abnormal vibration and noise, high pressure and insufficient pressure. Experimental results demonstrate that the proposed method is able to achieve higher performance on the accuracy of fault detection for turbocharger than other methods.