The Application Of Complex Network In Marine Engine Failure Detection

Marine power plant is a collection of machine, electricity, heat and liquid for the integration of complex system, there are influence factors included many faults, the large number of thermal parameters, much complexity, and the relationship by hull status, navigational conditions and main engine running condition, so comprehensive utilization of thermal parameter is very difficult to diagnose. All the fault factors like a complex network, researching the fault reasons. In fact, it is the study of the complex network. This paper introduces the definition of complex networks, which describes a complex network of research and complexity of its characteristics and explains the complex network of three basic concepts:characteristic path length, clustering coefficient and degree distribution. Then the paper introduces the development platform (c#) and the main database and language usage. It also describes the fault diagnosis method of the main engine, Fault diagnosis, whose essence is pattern recognition of object's operation state, which can be accomplished through clustering methods. The network model is used to represent the fault data structure and thus the clustering problem is converted into the detection task for sub-network structures. Thereby, a fault diagnosis strategy based on complex network structure analysis is proposed.Standard data set through the algorithm in classification and diagnosis of turbine simulator application of similarity and accuracy of the algorithm was verified.The results show that the proposed algorithm can calculate fastly, which is less time-consuming and extract fault features effectively, we could get a good diagnostic accuracy. This program is one for the engine fault diagnosis, knowledge entry and management, providing maintenance advice and assisted learning, and interactive integration of diagnostic procedures. This procedure uses the C# as a development platform, the system knowledge base with things Access database, through SQL to achieve.This program can be run on most Windows platforms and in a short time, for fault diagnosis it can be more accurate, and cost less manpower and money.