Design And Development Of XLPE Cable Partial Discharge Data Analysis System Based On Feature Map And Recognition Technology

With the XLPE cable in the distribution network is widely used,partial discharge detection as a cable insulation diagnosis of a most effective method is generally applicable,but the partial discharge of the measured data susceptible to environmental,magnetic interference,easy to cause the grid operators on the PD Data misjudgment,the existence of the cable every year due to partial discharge data anomalies are not timely analysis and damage caused by the accident occurred.With the increasing demand of the reliability of power supply,it is necessary to identify the fault and discharge characteristics of XLPE cable in time,analyze it,and deal with the data in time,so as to avoid further cable failure and secondary disasters.It can be seen that the research prospect of cable partial discharge data assist system is broad.In this paper,by combing the fault and discharge characteristics of power cable,the requirements of the data aided analysis system based on the characteristic spectrum and identification technology are proposed,and the overall scheme of the system design is determined.In this paper,the auxiliary analysis system of two modules of feature map and recognition technology is analyzed in detail.the principle of the feature spectrum is expounded.the competitive clustering method is used to divide the fuzzy interval dynamically,and the continuous eigenvalue is merged into the corresponding discrete interval.The association rules method is applied to extract the relationship between discrete eigenvalues and mining classification rules,and then the rules are fuzzy and the final classifier is obtained.The function of data import,data display,diagnostic analysis algorithm,diagnosis result display and so on are required for the diagnosis and analysis function of the local test data file.Through the live detection and analysis of the cable of 35 kv yuanwu 7404 in Guangdong 220 kv spark substation,the effectiveness of the method is verified.