基于数据挖掘的电气设备状态自动检测方法

Automatic Detection Method of Electrical Equipment State Based on Data Mining

传统的电气设备状态检测方法易受到冗余信息影响,导致检测结果误差较大,为此,提出基于数据挖掘的电气设备状态自动检测方法。通过建立自适应核函数提取电力设备的状态特征,将电气设备的运行状态分割成时间序列,分析测量距离值与序列匹配之间的映射关系,建立自动检测模型,采用蒙特卡洛方法预测设备寿命周期内的故障概率,设计电气设备状态自动检测流程。仿真实验结果可得,该方法检测三相电流与仿真数据最大误差为0.15 A,证明其具有精准的检测效果。

The traditional state detection method of electrical equipment is easy to be affected by redundant information,resulting in large error of detection results.Therefore,this paper proposes an automatic state detection method of electrical equipment based on data mining.The state characteristics of power equipment are extracted by establishing adaptive kernel function,the operation state of electrical equipment is divided into time series,the mapping relationship between measured distance and sequence matching is analyzed,the automatic detection model is established,and the Monte Carlo method is used to predict the fault probability in the equipment life cycle,so as to design the automatic detection process of electrical equipment state.According to the simulation experiment,the maximum error between the three-phase current detection method and the simulation data is 0.15A,which proves that the method has accurate detection effect.