Research On Data Mining And Condition Assessment Method Of Mass Equipment In Substation

Electric energy transmission has the characteristics of high voltage,large capacity and cross-regional transmission.The substation is an important hub for long-distance power transmission of the power grid.The operating status of the power equipment that makes up the substation is directly related to the safe and reliable operation of the power grid.In order to ensure the safe and reliable operation of the power grid,power companies continue to promote the construction of smart grids and obtain status data of power equipment through monitoring,testing,testing,and inspections to improve the perception of power grids and the efficiency of power equipment management.In this case,the data generated by the power system presents the characteristics of multi-source heterogeneous big data.The application of big data mining technology to data mining of electric power big data has high research value.In addition,the current status of the power equipment can be understood in time through the status assessment,and the corresponding maintenance strategy can be formulated in a timely manner based on the grid load situation to reduce the probability of equipment failure,which has important practical significance.In response to the above research needs,this paper carries out research work in two aspects:power big data mining and power equipment state evaluation:Sort out the characteristics of power big data from multiple angles and classify them in detail.It focuses on analyzing the characteristics of text data of power equipment,and introduces the process of cleaning the missing data and wrong data in the text data of power equipment defects.The process of structuring the unstructured information in the family defect text data is specifically explained.The array-based Apriori algorithm is used to find the key information in the text data of electrical equipment family defects and the association rules between the family defects,and analyze them with examples.The status level of power equipment is divided into 4 levels,and the specific meaning is explained.Constructed a streamlined and comprehensive indicator evaluation system.The hierarchical evaluation framework system is divided into target layer,middle layer and indicator layer.Taking transformers and circuit breakers as examples,the index level indexes are divided into numerical data indexes and non-numerical data indexes according to data types.Use the analytic hierarchy process to unify the multi-source heterogeneous state quantity index data into the same evaluation system,play the role of industry experience,establish a judgment matrix,calculate the initial weight phasor,and then introduce the fuzzy mathematics membership function to calculate the measured value of the state quantity index The degree of membership belonging to the state level is modified to modify the initial weight of the state quantity index to obtain the comprehensive weight of the current state quantity index of the equipment to the middle layer element,and finally calculate the equipment degradation value to quantify the current state of the equipment.The research shows that: using the array-based Apriori algorithm to conduct data mining on the defect text of power equipment,implicit knowledge can be obtained.According to different data types,different status evaluation methods have been adopted to achieve good evaluation results.It avoids the incompatibility of evaluation data and evaluation methods caused by the adoption of a unified evaluation method for all types of data,resulting in inaccurate evaluation results.