基于量子遗传和核模糊聚类的低压台区户变关系识别

User-transformer relation identification of low voltage area based on quantum genetic and kernel fuzzy clustering

户变关系对于营配融合、线损分析等业务的开展具有重要作用。为了得到准确的户变关系,提出一种基于量子遗传和模糊聚类的户变关系识别方法,该方法根据不同台区电能表的电压过零偏移特征不同,采用核模糊C均值聚类对电能表电压的过零偏移进行分类,通过与变压器端的过零偏移比较,实现户变关系识别。主要创新点为:采用量子遗传算法对模糊C均值聚类的聚类中心和核参数进行优化,提高聚类精度和效率;提出一种基于类间距离和类内距离的适应度函数构造方法;引入小生境协同进化策略、动态调整策略、Hadamard门变异策略,提高算法寻优能力。通过对Benchmark函数和UCI数据集特征的仿真测试,验证了所提出方法比标准核模糊C均值聚类具有更高的聚类精度和运算效率。采用文中方法对实际的台区变压器和电能表数据进行归属识别,结果表明,所提出方法得到的结果与真实户变关系一致,具有较好的应用效果。

User-transformer relation is very important to the service deployment of business distribution integration,line loss analysis,etc.To obtain the accurate affiliation relation,a nove user-transformer relation identification method based on quantum genetic algorithm(QGA)and kernel fuzzy C-means clustering(KFCM)is proposed in this paper.Since the electricity meter of different areas has different voltage zero-crossing shift features,the affiliation identification is realized through comparing the zero-crossing shift near transformer with the classification result of zero-crossing shift on electricity meter which is obtained by KFCM.The main innovation points are as follows,firstly,the QGA is introduced to optimize the cluster center and kernel parameters of KFCM to improve the accuracy and efficiency of clustering;secondly,a fitness function construction method based on inner-class distance and between-class distance is present;thirdly,the niche coevolution strategy,the dynamic adjustment strategy and the Hadamard gate mutation strategy are introduced to enhance the optimization ability of QGA.Then,the simulation tests on Benchmark function and UCI dataset character verify that the proposed method in this paper has better cluster accuracy and efficiency than standard KFCM.Finally,the proposed method is applied to solve the user-transformer relation identification of practical low voltage area.Result shows that the electricity meter affiliation obtained by proposed method is the same as the real affiliation,which has good application effect.