Research Of The Assessment Model For Intelligent Circuit Breaker Operation Condition

The current most popular technology in electric power industry innovation is the intelligent on-line fault diagnosis technology for power transmission and transformation equipment.With the hardware design of online monitoring devices tend to be mature,the development of the intelligent online evaluation expert software will be the key to the intelligent on-line fault diagnosis technology for power grid equipment.In the future,the intelligent power grid equipment state overhaul will be a electric power equipment maintenance cloud service platform which is based on the analysis of the large data and the "Internet Plus".As the key apparatus in electric power system,the high voltage circuit breaker which can protect and control power equipment has been required to be intelligent immediately.So based on the theory of fuzzy neural network and genetic algorithm,this paper puts forward a new intelligent assessment model which base on the fuzzy neural network improved by genetic algorithm for high voltage circuit breaker,and the model could show the fault characteristics of high voltage circuit breaker when it quickly obtained the existing knowledge and learned from the historical diagnostic data.After studied the basic theory of fuzzy neural network and genetic algorithm,the paper puts forward a intelligent assessment model which uses the genetic algorithm to optimize the fuzzy neural network learning process based on the idea of BP neural network learning process.This method could enhance the adaptability of the model through combine the advantages of fuzzy neural network and genetic algorithm,and solve the problem which is the fuzzy neural network is easy to fall into local optimum.At the same time,this paper puts forward some improvement on the genetic operation for genetic algorithm which could make the model has better adaptability than before.And then,using the data from simulation experiment and the project cooperation company to verify the assessment model,the accuracy of the model in this paper is 97%,the standard fuzzy neural network model is 93%,the fuzzy logic is 87%,the BP neural network is 87%.The result shows the intelligent state evaluation model in this paper has higher diagnostic capacity and smaller error than any other intelligent algorithm models.Also,the model can effectively learn fault features of high voltage circuit breaker and can complete the online state evaluation accurately.This could have certain reference function for the future research.Finally,through the cooperation with the enterprise,this paper applies the assessment model for intelligent circuit breaker operation condition to the on-line monitoring expert system for the substation.Recently,the model has been run successfully on the expert system and could correct evaluate the mechanical and insulating properties of high voltagecircuit breaker.