Community Intelligence Based HydroElectric Plant Model And Its Applications

Power industry innovations bring more and more demands for hydropower enterprise. It can be summarized as"high automation, high reliability, high performance and high profit". With the utilization of the computer monitoring system, water automation monitoring system, management information system and cascaded hydropower scheduling system in hydropower plants, the automation level of hydropower plant will be higher, and the advanced automation system and information system will replace people to do more work. Meanwhile with the strengthening of function, expanding of the application area of automation system and information system, the relationship which is among the systems and between people and systems is becoming more and more complicated.Comparing with the human activities, this relationship exists between not only individuals, but also departments, and even enterprises. Meanwhile, we have to take into consideration on how those complicated systems could actually work with different people involved and with the corresponding organizations, which request is apparently beyond any existing intelligent techniques. Meanwhile the fusion of complicated system and human being or its organization must be considered. Now the disperser and operation alone hydropower plant automation systems can not deal with these relationships, so the integration of these automation systems in higher layer is very necessary and urgent.According to the demands and features of integration of hydropower plant automation systems, the Distributed Artificial Intelligence (DAI) will develop from Agent, MAS to Community Intelligence in this dissertation. The hydropower enterprise model has been set up based the CI concept. Furthermore, theory, method, build and application of the CI based on hydropower enterprise model are studied, and the researches carried out in the paper and the corresponding results are as follows.The problems, purposes and significance of this research are firstly presented. State of arts of the enterprise modeling methods and DAI technique are introduced. Then the status in quo and existing problems of hydropower plant automation systems are discussed in the articles. According to the demands of engineering, it is feasible and necessary to use the DAI technique on building the hydropower plant model to improve the integration automation level.How the CI concept is put forward and comes into being is introduced in details as follows. Firstly, the theory and method of CI are expounded. Moreover, the formalization description of individual microcosmic social behaviors has been given in the CI theory system. After the unit model CI-Agent based on CI theory is put forward, its topological framework and co operational model are illuminated. Three layers of CI based system: communication layer, cooperation layer and control layer are researched.Information expression and knowledge renew methods of CIMDSS are researched. CI-Agent and human-machine fusion technique are introduced into the plant maintenance system. CI-MAS data diagnosis set is built based on fuzzy Petri Net and Artificial Neural Network is used to analysis. FPN, ANN etc. can solve the information integration and knowledge expression difficulty. CI-based distributed cooperation maintenance environment improves the problem on knowledge capture and maintenance.According to the research results of previous chapters, monitoring and diagnosis of 19# unit excitation system of GEZHOUBA Hydropower Plants has been investigated. Its system schematic structure has been detailed, and then the research and develop the expert diagnosis workstation of excitation system. The methods of monitoring condition CI-MAS and performance analysis CI-MAS are further discussed, the application results are given. The test results demonstrated that CI-MAS can integrate different diagnosis intelligence method, condition monitoring and evaluation, and prompt diagnosis and maintenance of equipments and so on.The dissertation develops the DAI theory from individual, multi to community, and introduces the CI concept into the automation system field for the multi-system organic integration and man-system confusion, and makes enterprise mechanism intelligent, system integrated structure flexible and the overall optimized.