基于RS-IA数据挖掘的配电网故障定位模型

摘要

针对配电网故障定位的相关性分析过程中,由于馈线终端单元(FTU)运行环境较恶劣、元器件受损或信息丢失等导致故障信息变异的问题,提出基于粗糙集(RS)理论和免疫算法(IA)相结合的数据挖掘配电网故障定位相关性分析模型.首先,借助RS理论提取领域知识,将给定变异故障模式集合转换成RS中的决策表;然后,利用IA进行决策表的属性约简,并挖掘出该问题中输入矢量(条件属性)与输出矢量(决策属性)的相互关联性规则;最后,将该数据挖掘方法用于处理FTU实时输入信息的畸变,根据各分段开关的电流越限信息序列判断各段线路故障状态,实现配电网的故障定位,并通过算例验证了所提模型的可行性和有效性.%Aiming at the fault information variation due to bad FTU(Feeder Terminal Unit) operational environment,device damage and information loss during the correlation analysis for locating the distribution network faults,a correlation analysis model based on data mining is proposed,which combines RS (Rough Set)theory and IA(Immune Algorithm).The domain knowledge is extracted based on RS theory and the given variation fault patterns are converted to decision table of RS.The attributes of decision table are then reduced based on IA and the correlation rules between input vector(condition attribute) and output vector (decision attribute) are dug out.The proposed method of data mining is used to deal with the distortion of real-time input information of FTU,based on which,the fault state of each line is decided according to the current limit violation information sequence of segment switches and the distribution network fault is located.Case study verifies the feasibility and effectiveness of the proposed model.