含分布式电源的变电站运行设备异常状态监控方法

Abnormal State Monitoring Method of Substation Operating Equipment with Distributed Power Supply

传统方法在含分布式电源的变电站运行设备异常状态监控中的应用效果不佳,不仅错误监控次数较多,而且变电站运行设备回归稳态所需时间较长,无法达到预期的监控效果,因此提出含分布式电源的变电站运行设备异常状态监控方法。利用传感器采集变电站运行设备的温度、电流、电压等状态量信息,采用主成分分析算法降维处理状态量,利用深度神经网络技术提取设备异常特征,识别监测设备的异常状态,通过隔离保护异常状态设备,实现含分布式电源的变电站运行设备异常状态监控。实验证明,该设计方法错误监控次数占总监控次数的比例仅为0.6%,运行设备可以在1 s内回归到稳定状态,在含分布式电源的变电站运行设备异常状态监控方面具有良好的应用前景。

Due to the poor application effect of traditional methods in monitoring the abnormal status of substation operating equipment with distributed power sources,not only are there many times of error monitoring,but also the time required for the substation operating equipment to return to steady state is relatively long,which cannot achieve the expected monitoring effect.Therefore,a monitoring method for abnormal status of substation operating equipment with distributed power sources is proposed.Utilize sensors to collect temperature,current,voltage and other status information of substation operating equipment,use principal component analysis algorithm to reduce dimensionality and process status information,use deep neural network technology to extract abnormal features of equipment,identify abnormal status of monitoring equipment,and isolate and protect abnormal status equipment to achieve abnormal status monitoring of substation operating equipment containing distributed power sources.Experimental results have shown that the proportion of error monitoring times to the total monitoring times in this design method is only 0.6%,and the operating equipment can return to a stable state within 1 second.It has good application prospects in monitoring abnormal states of substation operating equipment with distributed power sources.