基于概率盒的电网故障诊断方法研究

针对电网故障信息存在丢失、误动、拒动等不确定性问题,文章采用概率盒理论和支持向量机相结合的方法对电网故障进行诊断,充分利用概率盒在处理不确定问题上的优势。首先利用概率盒对故障录波、电气量等数据建模,然后利用融合规则将得到的多个概率盒进行融合,并提取特征向量。最后,利用支持向量机进行分类,并得出诊断结果。为了验证方法的有效性,采用仿真线路进行概率盒的故障诊断,实验验证该方法合理可行,且有较高的诊断率。

Fault Diagnosis of Power System based on Probability-box Theory

The fault information of power system often has the problems of missing information, malfunction and failure to operation, in order to solve the problems and improve the diagnosis ability, a new method based on probability-box theory and support vector machine was proposed. Firstly, the data of fault record and electrical quantities were used as the information source to construct the p-box. Then to take advantage of the complementation of the information source, the different p-box were fused and extract the feature from the p-boxes. Finally, using SVM classification and obtain the result. In order to verify the validity of the method, the simulation circuit is adopted to improve the probability box of fault diagnosis, the experimental shows that the p-box theory has high degree of diagnosis accuracy and is characterized by fast diagnosis and good real-time.