基于云计算平台Hadoop的线路参数并行辨识算法研究

Study of Parallel Line Parameter Identification Algorithm Based on Hadoop Platform

随着作为电网动态监测技术平台的广域量测系统(WAMS)在电网的应用普及,电网运行人员对于电网动态变化有了实时监测与分析的手段,但WAMS所产生的海量数据以及对于分析平台的高效率要求是WAMS应用的一大挑战。本文深入研究了基于Hadoop云计算平台的线路参数并行辨识算法,并提出算法的设计思路以及实现方法,为高效利用WAMS数据辨识线路参数给出了解决方法。对比实验证明基于云计算平台Hadoop的线路参数并行辨识算法大大提高线路参数辨识算法的计算效能,适合处理该应用中的WAMS海量数据。

As the power grid dynamic monitoring platform,the widely use of Wide-area Measurement System（ WAMS） will help the grid operator monitor and analyze the dynamic changes in power grid operation. However,the huge dataset generated by WAMS and requirement of high efficient analysis platform are the challenges of WAMS applications. This paper has conducted thorough study of the parallel line parameter identification algorithm based on Hadoop platform,and provided the parallel algorithm design and implementation methods. It provides a efficient solution for line parameter identification using WAMS data. The comparison test demonstrates that the Hadoop platform based parallel line parameter identification algorithm will boost the efficiency greatly,and it is suitable for processing huge datasets.