Study On Decision-making Expert System For Transmission Line

At present, the expert system has been applied in many fields. The development of expert system has brought much convenience for human work and life. The design of transmission line has relationship with Path, wire, terrain, weather condition and the cost control and so on. So the decision-making of transmission line has become complex work, which needs a large number of knowledge and experience of transmission line. Because of the above issues, it is necessary to use the artificial intelligence technology and computer technology to develop expert system for transmission line. This expert system can solve the problem of transmission line design through the expert knowledge and method.The paper develops the expert system for transmission line. The development of this system mainly includes interaction interface, backend database and inference engine. Inference engine module is realized by prediction method of BP neural network, support vector machine and multiple linear regression analysis. Stand-alone system can be realized by the remote access of database. The interaction of user and interface is completed through input and output device. And then the input data are transmitted to the inference engine. The results that calculated by inference module based on input data will transmitted to the interaction interface. The graphical display and result analysis is completed directly on the interaction interface.The test results of six samples show that the prediction of expert system is very reliable. BP neural network has higher prediction accuracy, and the prediction results are relatively stable. The analysis results of support vector machine are fluctuant, but they are reasonable. The prediction accuracy of multiple linear regression is inferior to BP neural network, but it has good stability.