Study On Fault Diagnosis Of High Voltage Circuit-breaker Based On Rough Set Theory

High voltage circuit-breaker is important equipment in electric system. It takes duel task of control and protection in power system. The study on fault diagnosis of high voltage circuit-breaker would not only help us know the working condition of the breaker immediately, but also avoid unnecessary examine and repair work. It strengthens the security and stability of power system and increases the economic benefit of it.The core of the paper is analyzing the various data of fault samples of high voltage circuit-breaker by using the data mining and knowledge discovery function of rough set theory and getting diagnosis rules from the data. Then we could use the rules as the basis for diagnosing and determining faults of high voltage circuit-breaker. This paper analyzes and summarizes the relationship between high voltage circuit-breaker faults and its status messages at first. Then it puts forward the method to establish a fault diagnosis rules by using the knowledge reduction based on greed algorithm. At last by Rosetta software it makes the algorithm simulation. The result shows that the model is accurate and effective.The innovation of the paper is the knowledge reduction method based on greed algorithm which is chosen because of the character of the sample data. Not only we could get the diagnosis rules by the method, but also its steps are easy for programming, so that we could establish circuit-breaker fault diagnosis system based on it.The main contents of the paper are listed as followed:1.The paper introduces the basic concepts of rough set and fault diagnosis. It discusses the method of combining them.2.The paper makes research on the state data of high voltage circuit-breaker and analyzes the inner relationship between them.3.The paper chooses appropriate reduction method according to the character of the sample data and makes simulation of the method by software.