改进模糊聚类的设备非结构化数据挖掘研究

Research on equipment unstructured data mining of improved fuzzy clustering

针对当前已有方法未能对电力设备非结构化数据进行去噪处理,导致运行时间增加、挖掘结果准确率下降的问题,提出一种基于改进模糊聚类的电力设备非结构化数据挖掘方法。先对数据进行集合经验模态分解,将分解后的各个模态分量按照排列熵的大小进行排列,确定降噪分量,同时引入小波包变换对数据进行去噪;采用改进的模糊聚类算法动态调整数据类别,同时构建电力设备非结构化数据挖掘模型,通过模型完成数据挖掘。仿真实验结果表明,所提方法可有效提升挖掘结果准确性,同时减少运行时间。

Based on the fact that the existing methods fail to denoise the unstructured data of power equipment,which leads to the increase of running time and the decrease of the accuracy of mining results,an unstructured data mining method of power equipment based on improved fuzzy clustering is proposed.Firstly,the data is decomposed by ensemble empirical mode,and the decomposed modal components are arranged according to the size of the arrangement entropy to determine the noise reduction component.At the same time,the wavelet packet transform is introduced to denoise the data.The improved fuzzy clustering algorithm is used to dynamically adjust the data categories,meanwhile,an unstructured data mining model of power equipment is constructed,and data mining is completed through the model.Simulation results show that the proposed method can effectively improve the accuracy of mining results and reduce the running time.