基于数据挖掘的电力负荷脏数据动态智能清洗

Dynamic Intelligent Cleaning for Dirty Electric Load Data Based on Data Mining

来源于SCADA系统的负荷历史数据由于各种原因含有一定的脏数据,在进行高精度的电 力负荷预测或系统分析前必须仔细而合理地对历史数据进行清洗。文中基于数据挖掘理论提出一 种动态的智能清洗模型,先根据模糊软聚类思想对Kohonen神经网络进行了改进,改进后的 Kohonen神经网络能实现模糊软聚类的并行计算,提出的动态算法能根据样本集的更新而自动确 定新的聚类中心(即特征曲线),最后与径向基函数(RBF)网络一起构成脏数据的智能清洗模型。 模型的快速性和动态性特点使其宜于进行负荷数据的实时处理,对重庆江北负荷数据的实例分析 说明了该模型的高效性。

There is a number of dirty data in the load database produced by SCADA system, thus the data must be cleaned before it is used to forecasting electric load or performing power system analysis. This paper proposes a dynamic and intelligent model based on data mining theory. Firstly the Kohonen self-organization neural network is meliorated referring to fuzzy soft clustering arithmetic, the meliorated Kohonen network can realize the collateral calculation of fuzzy c-means soft clustering arithmetic and the approach we proposed can find dynamically the new clustering center, that is the character curve of data, according to the updating of swatch data. Then the RBF neural network is introduced to identify dirty data and compose the intelligent cleaning model. The rapidness and dynamic performance of model make it suits real-time calculation. Test results using actual data of Jiangbei power supply bureau in Chongqing demonstrate the validity and feasibility of the model.