基于聚类思想的电能质量综合评估  

Power quality comprehensive evaluation based on the clustering method

作　　者：蒋德珑[1,2] 王克文[1] 杨平[3] 崔惟[1] 

机构地区：[1]郑州大学电气工程学院,河南郑州450001 [2]71781部队,河南洛阳471100 [3]安庆供电公司,安徽安庆246000

出　　处：《电力系统保护与控制》2012年第7期105-111,共7页Power System Protection and Control

摘　　要：针对目前电能质量综合评估方法中人为因素影响较大和评估模型较为复杂的问题,提出了基于聚类思想的电能质量综合评估方法。采用模糊聚类分析法进行电能质量的综合评估,以电能质量的各项评估指标作为特征值,将具有已知电能质量等级的数据源加入样本数据集进行动态聚类,根据'物以类聚'的原则,通过选取最佳阈值,得到包含各个电能质量等级的聚类结果,实现对样本数据的综合评估。通过对两组实例数据的计算,验证了方法的有效性。该评估过程无人为因素干扰,且模型简单、扩展性好,适应区域电网电能质量的统一评估,具有很好的应用前景。To solve the problems of anthropic factors and complex evaluation model, the idea of clustering is applied to power quality comprehensive evaluation. Power quality is estimated synthetically by the fuzzy cluster analysis. In clustering process, the indexes of power quality comprehensive evaluation are regarded as the characteristic values, and the data points with known power quality level are included in sample data set. According to the principle of ＂like attracts like＂ and choosing the optimal threshold, the power quality levels of the sample data points are found, and then the comprehensive evaluation of power quality is realized. Practical effect in two examples proves that the proposed method is effective. In this evaluation, the model is simple and has better expansibility without any personal factor interference. Besides that, it is better to be used for the power quality comprehensive evaluation of regional grid together, which has a good application prospect.

关 键 词：电能质量 综合评估 聚类分析 模糊等价关系 最佳阈值 

分 类 号：TM711[电气工程—电力系统及自动化]