基于随机矩阵理论的地下电缆异常检测方法研究  

Research on Underground Cable Anomaly Detection Method Based on Random Matrix Theory

作　　者：段晨东[1] 张伟 代杰[1] Duan Chendong;Zhang Wei;Dai Jie(School of Electronic and Control Engineering,Chang’an University,Xi’an Shaanxi 710064,China)

机构地区：[1]长安大学电子与控制工程学院,陕西西安710064

出　　处：《电气自动化》2021年第5期115-118,共4页Electrical Automation

基　　金：中央高校基金(300102320105)。

摘　　要：为了及时准确地发现地下电缆异常,提出一种应用随机矩阵理论和奇异值分解理论的信号异常监测方法。首先利用地下电缆监测数据构建高维矩阵,再通过随机矩阵理论提取特征值,对这些特征值进一步处理建立特征指标矩阵。然后对特征指标矩阵奇异值分解降维以获取关键分量,在此基础上构造融合指标,以此判断地下电缆的异常状态。算例显示融合指标对异常更加敏感,可有效避免监测信号波动引起误判。试验结果表明,方法的抗干扰能力比较强。In order to discover timely and accurately the anomalies of underground cable, a signal anomaly monitoring method based on random matrix theory and singular value decomposition theory was proposed. When using this method, firstly use the underground cable monitoring data to construct a high-dimensional matrix, and then extract eigenvalues through random matrix theory, and further process these eigenvalues to establish a characteristic index matrix. After that the singular value decomposition of the characteristic index matrix was reduced to obtain the key components, and the fusion index was constructed on this basis to judge the abnormal state of the underground cable. The calculation examples show that the fusion index was more sensitive to anomalies, which can effectively avoid the misjudgment caused by the fluctuation of the monitoring signal. The test results show that the method has relatively strong anti-interference ability.

关 键 词：地下电缆 异常检测 随机矩阵理论 奇异值分解 融合指标 

分 类 号：TM75[电气工程—电力系统及自动化]