基于改进FC-IRKM的电力用户负荷分类研究

为了提升电力用户负荷分类精度,从而为电力企业有效掌握用户用电规律,制定电价,合理评估用电需求,进行了基于改进的FC-IRKM的电力用户负荷分类研究。首先,模糊聚类综合了模糊思想与聚类方法,粗糙集则解决了数据分类不清晰的情况,将两种方法的优点进行结合,提出了模糊粗糙K均值算法;其次,对IRKM算法记进行了理论分析,结合模糊粗糙算法,提出了一种FC-IRKM算法,并通过遗传算法对其参数进行优化,获得改进后的FC-IRKM算法;然后通过仿真对比实验,验证了在处理相同数据的时候,改进的FC-IRKM算法比其他常见传统算法的分类精度更高,适用性更广;最后将改进的FC-IRKM算法应用于电力用户负荷分类上,分类结果与预期结果相互印证,验证了该方法的适用性。该研究为电力企业把握电力用户的用电规律、合理规划供电计划提供了可靠的理论支持。

Study on Power Load Classification Based on Improved FC-IRKM

In order to improve the accuracy of load classification for power users so that the power company is able to gasp the power consumption law and formulate electricity prices more effectively, and reasonably evaluate power demand, this paper proposes an improved FC-IRKM power user load classification study. Firstly, the fuzzy clustering integrates the fuzzy idea with the clustering method, while the rough set solves the problem of unclear data classification. The advantages of two methods are combined and a fuzzy rough k-means algorithm is proposed. Secondly, the theory of IRKM algorithm is analyzed, and a FC-IRKM algorithm is proposed based on fuzzy rough algorithm, and its parameters are optimized by genetic algorithm. Furthermore, through simulation and comparison experiments, it is verified that the improved FC-IRKM algorithm has higher classification accuracy and wider applicability than other traditional common algorithms when dealing with the same data. Finally, the improved FC-IRKM algorithm is applied to load classification of power users, and both the classification results and expected results verify applicability of the method. The study is expected to provides reliable theoretical support for power companies to grasp the power consumption law of power users and to plan power supply reasonably.