The Study Of Fault Diagnosis For Ship Fuel Oil Supplying System

Safety is the most important issue in the operation of the ship, and it is the basic requirement for all the ships. In the recent years, the ship security incidents usually occurred. So the fault diagnosis of the ship machinery and equipments is very necessary. The fuel oil supplying system of the main engine is one of the most important system of the ship power plant. The function of this system is to supply a certain pressure, viscosity fuel oil for the main engine combustion. To ensure the normal operation of this system is the precondition of ensuring the normal operation of the ship main engine. Therefore, it is very important to forecast and judge whether the fuel oil supplying system is working properly, whether the fault is existed and give the cause of the fault timely and accurately.Due to the automation and complication of the ship machinery and equipment, the traditional ship fault diagnosis technology can’t meet the practical requirements. And the artificial intelligence technology has been widely applied to various fields. Therefore, the smart ship fault diagnosis technology becomes the new field of the fault diagnosis research directions. With the research and development of decades, the fuzzy theory and BP neural network has matured in the theory field and the application in the fault diagnosis has achieved certain results. This paper introduces the fuzzy theory and BP neural network separately. It also analyzes the application of these two theories in the field of the fault diagnosis and the necessity of these two theories combination, and their connections. This paper establishes an intelligent fault diagnosis model using the traditional BP neural network. And the study of fault diagnosis is applied to the main engine fuel oil supplying system of the "YUKUN" vessel. According to the defects exposed, this paper brings up the optimization of the fault diagnosis. It can diagnose and identify the fault types of the "YUKUN" vessel main engine fuel oil supplying system. It proves the improvement as an effective optimization plan.Firstly, this paper gives a brief introduction of the ship fault diagnosis technology and the teaching-training vessel "YUKUN". And it also gives the failure analysis of the "YUKUN" vessel main engine fuel oil supplying system Then this paper introduces the basic theory of the fuzzy mathematics and neural networks, and focuses on the analysis of the BP neural network algorithm and fuzzy neural network. At last, the BP neural network fault diagnosis model is established based on the fault symptom parameters. The study of fault diagnosis is performed by dusing the MATLAB software. The problems of the fault diagnosis model are optimized.