The Design And Development Of Remote Diagnositic Center Based On Kowledge Service For Marine Power System

With the continuous development of computer network technology and communication technology, all kinds of network resources such as information, knowledge and so on have been shared in different ranges and degrees. Online monitoring and diagnosis of marine power system has gradually developed from stand-alone to remote distributed, and diagnosis knowledge sharing and network collaborative diagnosis have become irreversible trend. Especially in the knowledge economy area today, knowledge has become an important strategic resource. As the "brain" of fault diagnosis system for marine power system, the remote diagnostic center should expand to a knowledge service center whose position and function depends on whether to provide users with efficient knowledge service. But existing remote diagnostic center especially for marine power system and also using for actual intelligent fault diagnosis is not common, in addition, the problem of passive service, single function and conservative resource exist.The composition of the distributed resource environment for Marine remote knowledge service was proposed, and the role action of each component in the environment was defined. The basic functions that the remote diagnostic center as the service provider shall possess were discussed.For the knowledge acquisition in knowledge service, methods of acquiring data resources from the literature and enterprises'Word and Excel record were studied. The data extraction of curve graph describing variation trend was carried out, and the identification of data table from the Pdf literature was also realized. Then the way how to import data table in Word document into database and then automatically create form to save was explored. Above researches provided reliable knowledge resource for the knowledge acquisition.Knowledge representation is the key to knowledge service. Fault diagnosis domain knowledge of marine power system was divided into explicit knowledge and implicit knowledge. Production rule was used to represent explicit knowledge and neural network was used to represent implicit knowledge, and database technology was applied to design the storage structure of these two kinds of knowledge, which is more convenient to represent knowledge in knowledge service. The vb.net language was used to develop the neural network program which was independent of commercial software such as MATLAB. The implicit knowledge such as data case could be represented using neural network. Thus the knowledge without explicit expression could be accumulated effectively and rapidly, which laid the foundation of knowledge service.Aiming at the communication efficiency and security problem in marine remote knowledge service, the half-byte compression method was proposed to realize the compression and decompressing function, the method was easy to program and had higher compression ratio. A kind of plaintext encryption model which adopted the middleware was designed, data secure transmission and storage in the process of remote diagnosis was guaranteed to a certain extent, which provided a powerful guarantee for the reliability of knowledge service.Demand analysis to marine power system remote diagnositic center was carried on, and the design principle was formulated, and each functional module was analyzed, finally the service platform was established, which connected various resources and collaborated service and also set diagnosis knowledge management, remote knowledge service, remote information communication and other functions in one body. Once put into use, the plantform would provide the marine power equipment related users for cross-regional and networked technology support.