Study And Develop On Advanced Nuclear Power Plant Expert Evaluation System

As we know that at present the nuclear electricity installed capacity is about 8GW and is less than 2% of the total installed electricity capacity of China. But according to the development plan of government, this number will become 36-39GW and is about 4% of the temporal total installed electricity capacity in 2020. So the nuclear energy will develop fast recently and how to evaluate Nuclear Power Plants(NPPs) and select the optimal types becomes a pressing task. Based on such a background, an expert system is developed to evaluate advanced NPPs for the following several decades. The system is initiative in the nuclear field, although it is not so perfect, yet some preliminary evaluation results are reasonable and sound. In present dissertation, based on current status and developing trend of nuclear power technology, an advanced evaluation system is developed to assess reactors for NPPs systematically according to a set of pre-established evaluation indices, including the selection and classification of the indices, the determination of their weighting factors in applying AHP method. In this system three evaluation methods are used to get the evaluation result base on the data stored in this system. Considering the attributes of themselves, some advanced NPP are researched based on the results of evaluation. Then, a number of comprehensive evaluation-decision making methods, including the AHP-weighted summation method, the comprehensive fuzzy evaluation method and the fuzzy Borda number method are studied in detail. With some preliminary evaluation results, the evaluation methods mentioned-above are analysed, compared and discussed. By changing the weighting factors of evaluation indicese characterizing the economics, safety and technology of NPPs, the evaluation results under different weighting factors of such three evaluation indices mentioned above are studied and the sensitivities of these indices are analysed. Further more, the evaluation results are discussed and analysed depending on the marks graded by different experts grouped by age, major and job etc. The present thesis also illustrates the design and establish of database for this evaluation system. The system is modeled in UML(Unified Modeling Language), designed in OOAD(Object-Oriented Analysis and Design) and developed in MVC(Model, View, Controller) mode. In this thesis, the sequence diagrams for the implementation of several evaluation system's processes are illustrated and discussed. Further more, the MVC mode is introduced and analysed in details. The system is developed in architecture of the mixture framework C/S(Client/Server) and B/S(Browse/Server), thus is safe, advanced, extensible and easy for use. The present system is network based, so it is can be used not only in personal PC and Local Area but also in the Web Server which can be browsed by persons in different areas.