Applied Research of BP Neural Network in Remote Marine Diesel Engine Fault Diagnosis System

Abstract

In order to detect and diagnose abnormal conditions of marine diesel engine and ensure its normal functioning, the present study adopts the BP neural network and related algorithms to determine the remote fault diagnosis process. Taking the design of exhaust gas temperature remote monitoring sub-system as an example, MATLAB programming was used for data simulation and verification. The applying of the system on board a real ship shows that it has a high working rate, a reliable and safe storage mode and a self- adaptive process.

Diesel Engine Inlet and Exhaust System Fault Detection Based on PSO-BP NN

“…Scholars have achieved certain results about this subject. Literature [1] proposes a method to detect an intake manifold leakage for a diesel engine with a dual loop EGR system; the neural network based on Levenberg-Marquardt is used for diesel engine leaks detection [2]; a hybrid classification algorithm, which inspired by combining nonnegative matrix factorization and artificial immune system, was applied to the fault classification of diesel valve train [3]; BP Neural Network (NN) was applied to to detect and diagnose abnormal conditions of marine diesel engine [4], however, there are poor convergence, long training time, and local optimum problem in BP NN.…”