【作者】 康忠健； 徐丽； 勾松波； 金华蓉； 李瑞生；

【Author】 Kang Zhongjian 1,Xu Li 1,Gou Songbo 2,Jin Huarong 3,Li Ruisheng 3 1.College of Information and Control Engineering,China University of Petroleum,Dongying 257061,P.R.China 2.Power Company of Shengli Petroleum Administrative Bureau,Dongying 257061,P.R.China 3.XJ Electric Co.,Ltd.,Xuchang 461000,P.R.China

【机构】 中国石油大学(华东)信控学院； 胜利石油管理局电力公司； 许继电气股份有限公司；

【摘要】 针对常规阻抗测距单端算法不能考虑线路分布电容的不足,本文提出了一种考虑分布电容的基于神经网络的单相接地故障测距算法。该算法是利用神经网络对传统故障测距算法的测距结果进行动态补偿,以提高故障测距精度。MATLAB仿真和动模实验结果表明,该算法具有较高的测距精度,可克服过渡电阻和分布电容对测距精度的影响。更多还原

【Abstract】 As for the distributed capacitor not taken into account in the conventional fault location algorithm,a single phase-ground fault location algorithm based on neural network with the distributed capacitor taken into account is proposed in this paper.The proposed algorithm uses neural network to correct the fault location distance dynamically.The MATLAB simulation results and dynamic modal experiments results suggest that the proposed algorithm have a high precision and could overcome the influence of the distributed capacitor and the ground resistance.更多还原