基于CUDA技术的海量电力负荷曲线聚类算法

随着用电信息采集、负荷控制等系统中用户负荷数据的爆炸式增长,传统计算框架与方法在处理海量用户负荷聚类、开展负荷特性分析等业务时面临着巨大的计算压力。着眼于计算精度日益提高、计算能力日渐强大的图形处理单元(graphic process unit,GPU),基于Nvidia的统一计算设备架构(compute uniform device architecture,CUDA)提出了一种负荷曲线快速并行K-means聚类算法,采用距离计算并行化、曲线数统计并行化、线程块分配合理化等多个并行加速策略,极大地提升了用户负荷曲线的聚类速度。多个测试算例表明,文中提出的基于CUDA的K-means电力负荷曲线聚类算法加速比高,适应性强,是解决海量负荷曲线聚类问题的好方法。

A Clustering Algorithm Based on CUDA Technology for Massive Electric Power Load Curves

With the explosive growth of user load data in power consumption information collection and load control systems,traditional computing frameworks and methods are faced with tremendous computational pressure when dealing with massive user load clustering and carrying out load characteristic analysis.In this paper,with a view to increasing accuracy and computational power of graphic process unit (GPU),the fast parallel K-means clustering algorithm for load curves is proposed based on Nvidia compute uniform device architecture (CUDA).This algorithm uses parallel acceleration strategies,such as parallelization of distance computing and curves counting,and rational allocation of thread blocks,which greatly improve the clustering speed of user load curves.A number of test examples show that the proposed clustering algorithm in this paper has a high acceleration ratio and strong adaptability,which is a good way to solve the problem of massive load curve clustering.