The Research Of Control System Of Synchronous Motors Applied In Ship Electric Propulsion

To control the submarine's navigation position and gesture quickly and effectively is the key to ensure its confidentiality and flexibility. Once there is something wrong with the gesture control system during the war, it might not only be hard for the submarine to carry on effective strategic attack on the enemy, but also expose itself under the enemy's fire and cause significant personnel casualty and economic loss. Therefore, it is an important task to find the fault parts of the whole system fleetly and accurately.In order to prevent the control performance of the submarine from declining and enhance its survivability in modern warfare, it is better to carry out intellective fault diagnosis if there is something wrong with submarine's gesture control system. And it is just the purpose of this article.First, taking the indigenous torpedo attack submarine, which has two axles and two screws, as study object, the article introduces the structure of its gesture control system based on analyzing the characteristic and stress which it is undertaken. According to the experience of the related domain experts, it analyzes kinds of fault instance that may happen during the navigation period. And it is the foundation to establish a fault tree model. Then, according to the basic principle of fault tree analytic method, the basic step of fault tree modeling and the basic structure of the submarine's gesture control system, it establishes the fault tree model of the system with the help of deductive method. After that, the article uses production rules to organize the fault diagnosis knowledge, based on which the repository of the fault diagnosis expert system is built. It uses object-oriented programming method and takes Windows XP as platform and VC++ 6.0 MFC frame as software development tool to compile the program of the fault diagnosis expert system. In this way, the combination of the fault tree analytic method and fault diagnosis expert system comes true. Finally, with the help of the fuzzy neural network technology, the article takes a sub-tree of the fault tree model as a sample to fetch up the disadvantages, such as unable to learn by itself or lack of online diagnosis ability, of the fault diagnosis expert system.The results of the simulation indicate that the fault diagnosis expert system, which combine with the fault tree model in this article, could discover faults that exist in the gesture control system quickly and accurately. It could also demonstrate the fault path vividly and provide convenience for the technologists. And the combination of the fault diagnosis expert system and fuzzy neural network could help the fault diagnosis system study and train itself automatically based on new fault diagnosis knowledge. Then it can carry on the process of fault diagnosis online. The whole fault diagnosis system has certain project application value.