基于小波能量熵和支持向量机的高压输电线路故障选相方法研究

Research on a New Faulty Phase Selector of High Voltage Power Transmission Lines Based on Wavelet Energy Entropy and SVM

提出了一种基于小波能量熵和最小二乘支持向量机(LS-SVM)的高压输电线路故障选相的新方法。首先对采集到的故障后三相电流信号进行合适的小波分解,得到特定时间窗口内的三相小波能量熵的累加值以及各相之间的比值。利用各相的小波能量熵累加值以及比值作为表征不同故障类别的特征向量,并输入到LS-SVM分类器。采用支持向量机算法对数据样本进行训练,找出样本内潜在的规律,最终完成整个输电线路的故障选相。仿真结果表明,该方法不受系统运行方式、过渡电阻、故障位置以及故障初始角因素的影响,能够有效地识别故障类型,具有较强的通用性和实用性。

A new faulty phase selection method of high voltage power transmission lines is presented, based on the concept of wavelet energy entropy （ WEE ） and the least square support vector machine （LS-SVM）. First, appropriate wavelet decomposition is conducted on the collected three-phase post-fauh current signals, obtaining the accumulated value and ratio of three-phase WEE within the specified time. The accumulated value and ratio of WEE of every phase are taken as the variant faulty eigenvectors and then input into LS- SVM. By means of SVM technique, the data samples are trained ; the potential discipline is found and then the faulty phase of the whole transmission lines is selected. The simulation results of Matlab show that the proposed method is not affected by the operation mode of system, the transitional resistance, the fault location, and the fault initial angle. Moreover, it can effectively identify the type of the faulty, with better generality and practicality.