Research On Remote Fault Diagnosis Expert System Of Voltage Transformer

With the development of modern technology, artificial intelligence is not only a research field of computer. It has been applied in various related disciplines. Expert system is an important branch of artificial intelligence, has also been considerable development in application and practice, And it's widely applied to the fields of prediction, interpretation, diagnosis, failure analysis etc.. However, the expert system at this stage is difficult to meet the needs of large-scale commercial applications, it is difficult to achieve the needs of big data processing. The cloud computing is a new computing service, can provide efficient and cheap computing services stable, so the application of cloud computing to the expert system has important theoretical significance and application value.According to the expert system inference efficiency is low, combined with cloud computing characteristics of expert system based on cloud computing for the in-depth study and the research achievement has been applied to remote fault analysis expert system. The main research work is as follows:From the theory of expert system, cloud calculation theory and fault diagnosis theory three aspects analyzes the present remote expert system involved in the key issues, as well as remote fault analysis expert system for the process, illustrate the application and advantages of cloud computing technology in the remote expert system,the objectives of this study were to determine;Starting from the system structure of the expert system of remote fault analysis expert system knowledge base, inference engine, explanation mechanism, knowledge acquisition and other key module of detailed design;Aiming at the problem of expert system inference efficiency is low, the result is not accurate, the uncertain reasoning mechanism for cloud computing. The mechanism on the original basis for determining the reasoning mechanism on the uncertain reasoning function, which can also make corresponding diagnosis of uncertain events.Given the specific inference rules and reasoning methods, this paper puts forward the feasibility and mechanism of proof by examples;On the basis of the existing theory, using B/S architecture, design and initial implementation of the cloud computing based remote failure analysis the basic function of the expert system, the test results show that the proposed algorithm and architecture has good feasibility and practicability.