基于广域测量系统的电网实时监控研究

摘要

In the face of the problem of real-time monitoring of power grid,this paper transforms the grid failure into voltage change trend,and uses the singular value decomposition to reduce the interference of the wide-area measurement data to remove the redundant interference. The matrix completes the bus voltage to determine the sensitivity of the load to evaluate the voltage break?down threshold. By constructing a wide-area measurement and analysis system for real-time monitoring of the grid,the reduced-di?mension data stream is transformed into the tidal data stream about the time series using the polar coordinate transformation. The to?pology of the processed wide-area measurement data is calculated using the Apache spark flow data processing platform Structure, and approximate the criterion of voltage stability to reflect the stability of the operation and maintenance of the power grid,and final?ly to achieve the purpose of real-time monitoring of the power grid. The experimental results show that the data processing method can reduce the dimensionality of the wide-area measurement data and can be used as the prediction index of the voltage collapse. With the increase of the number of CPUs,the running time has a significant decrease,which can improve the real time The efficien?cy of monitoring.%面对电网实时监控涉及众多变量参数问题,论文将电网故障发生转化为电压变化趋势,利用奇异值分解对广域测量数据进行降维处理以去除冗余性的干扰,结合潮流雅可比矩阵完成母线电压对负载的灵敏度判断以评估电压崩溃临界值指标.通过搭建针对电网实时监控的广域测量分析系统,使用极坐标变换将降维数据流转化为关于时间序列的潮汐数据流,运用Apache火花流大数据处理平台对处理后的广域测量数据计算拓扑结构,近似地得出电压稳定的判据以反映电网运维过程中的稳定性,最终达到对电网的实时监控目的.通过实验模拟结果表明:论文提出数据处理方式对广域测量数据降维效果较好,可以作为电压崩溃的预测指标,并且随着CPU数目的增多,运行时间有明显的下降,可提升电网实时监控的效率.