支持向量机下基于机器学习优化的继电保护故障诊断技术研究

Research on fault diagnosis technology of relay protection based on machine learning optimization under support vector machine

传统的继电保护故障诊断技术对故障位置的判断精准度较低,因此,在支持向量机下,利用机器学习技术设计了一种新的继电保护故障诊断方法。利用机器学习方式寻找匹配程度较高的优化算法追踪故障数据信息,并强化数据操作力度、完善数据空间,从而获取继电保护故障数据。然后选取匹配度较高的内部监控数据监测故障数据状态,并标记其信号信息,以处理后的故障数据为基础,实施继电保护故障诊断。由于选择样本的有限性,结合机器学习的算法处理模式,加强对内部故障信息的整合操作,有利于有效检测继电故障。实验结果表明,该技术能够精准地诊断出故障位置,具有很强的故障检测能力。

Traditional relay protection fault diagnosis technology has a low accuracy in fault location judgment.Therefore,a new relay protection fault diagnosis method using machine learning technology is designed in support vector machine.The machine learning method is used to find the optimized algorithm with high matching degree to track the fault data information,strengthen the data operation strength and improve the data space,so as to obtain the relay protection fault data.The internal monitoring data with high matching degree is selected to monitor the status of the fault data,and the signal information is marked.Based on the processed fault data,relay protection fault diagnosis is implemented.Due to the limited selection of samples,combined with the algorithm processing mode of machine learning,strengthening the integration of internal fault information is conducive to the effective detection of relay faults.Experimental results show that this technology can accurately diagnose the fault location and has a strong ability of fault detection.