Study On Remote Monitoring For Ships Based On BP Neural Network

Safety is the eternal theme of the shipping industry. Recent years, safety accident occurred frequently, and fault diagnosis must be taken to reduce it. However, it is inadequate that fault diagnosis on ship was just taken by the crew according to their experience. And it makes against the fault information exchange between the people on ship and shore. People on shore can use remote fault diagnosis technology get hold of the working condition of the equipments and make the crew know the abnormal situation and take action before the fault in order to ensure the safety of the ship and reduce the economic loss. Real time commucation and data sharing come true as the rapid improvement of the computer network technology and wireless commucation. And remote fault diagnosis combining computer communication technology and artificial neural network is achieving step by step.This paper is based on the program The Online Monitoring System for Ocean-going Vessel and Cargo Transportation (SCOM) of COSCO.Firstly, it introduces the background of this topic and the significance of using remote monitoring system on ship. Secondly, it gives the structure of the remote monitoring system, introduction of SCOM, fault diagnosis of mechanical and electrical equipments on ship including compositon and strcture taking marine diesel engine as an example. Thirdly, it does the research of the utilization of the BP neural network in remote monitoring of ship including the stategy of the marine diesel engine diagnosis, fault diagnostic flow and the normalization operation of the samples, the process of fault diagnosis with the help of MATLAB. At last, it studies the process from getting fault samples with the fault emulation program of the fault diagnosis system based on BP neural network and training and establishing the network after the normalization operation to testing the network and analyzing the fault using the simulation test data and online data separately in detail.The potential of using artificial neural network in fault diagnosis techonology is great. It will make more effects to the development of the fault diagnosis of the marine diesel engine as the development and improvement of the neural network theory.