Study on Power Transformer Protection Based on Chaos Particle Swarm Optimazation

摘要：

Power transformers are important parts in power supply systems. The stability, safety and reliability of transformers are especially important for electricity systems. This paper uses dissolved gas value as a feature parameter and creates a power transformer fault detection model by support vector machine classifier. In order to address the difficulty of determining the parameters for support vector machine, an optimization algorithm based on chaos particle swarm is proposed. The approach is not easy to be trapped into local optimum and improves population diversity and particle ergodicity. The experiment results prove that the support vector machine using chaos particle swarm optimization can achieve classification with higher precision and faster speed in power transformer fault diagnosis, compared to the support vector machine using traditional optimization algorithm.

关键词： Transformer CPSO SVM Parameter optimization DGA.

作者: Han Han Wang Houjun

作者单位: College of Automation Engineering University of Electronic Science and Technology of China,Chengdu 611731,China

会议类型: 国际会议

会议名称: 2011 10th International Conference on Electronic Measurement & Instruments(第十届电子测量与仪器国际会议 ICEMI2011)

会议地点: 成都

会议语种:英文

页码: 1347-1350

在线出版日期: 2011-08-16（万方平台首次上网日期，不代表论文的发表时间）