Research On State Assessment Method And Operation And Maintenance Strategy Of Transmission Lines Based On Data-driven

As the artery of power transmission in power system,transmission lines are the bridge connecting power supply and demand side.Its stable operation is directly related to the safety and stability of the entire power grid.Therefore,it is an important measure to ensure the stable operation of transmission line to formulate scientific operation and maintenance strategy and accurately evaluate the state of transmission line.With the development and application of transmission lines information management system and data mining technology,data drive provides new power for the study of transmission lines in state assessment and operation and maintenance strategy.The purpose of this thesis is to evaluate the state of transmission lines,determine the state level of transmission lines,and finally establish the differential operation and maintenance model of transmission lines according to the state level.In this thesis,in order to promote the pertinence and accuracy of transmission lines state evaluation,through the collection and sorting historical state information of transmission lines,patrol information based on the confidence of association rules and quantitative matrix,to quantify the importance degree of each state variable is applied principal component analysis dimensionality reduction for key state quantity,set up key state system,to meet state evaluation,on the basis of accurate,the efficiency of transmission line state assessment is improved.On the basis of the constructed key state quantity system,the state assessment samples are collected and sorted out.According to the unbalanced characteristics of the data samples,the transmission lines state assessment model based on Random forest is established,and the multi-classification performance evaluation indicators are set up.To deal with the problem of unbalanced data,the Random forest introduces the method of stratified resampling when selecting random feature subset,so as to improve the classification ability of the Random forest model to unbalanced data.In model parameter setting,OOB estimation was used as objective function to optimize Random forest parameters by improved particle swarm optimization algorithm,and the optimal parameters were selected.Through the example verification,the Random forest state assessment model constructed in this thesis has better classification performance,all the performance indicators have a better improvement compared with the traditional Random forest,and the classification of unbalanced data is more superior.The Random forest state assessment model is obtained by data-driven training,which reduces the influence of subjective factors and enables it to judge transmission line state grades more quickly,accurately and objectively,providing decision support for transmission operation and maintenance.For transmission line operations increasing needs of work and operational resources imbalance problem,a differential operation and maintenance model based on economy,reliability and workload balance is proposed,fully considering the operational time,sequence and resource constraints to maintenance strategy about various constraint conditions related to transmission line stability,multi-objective optimization model is established.The multi-objective criss-crossing algorithm was adopted to introduce pareto dominance principle,elite retention strategy and non-dominated ranking to solve the multi-objective optimization model to obtain the optimal solution set,and differentiated operation and maintenance strategy was formulated according to the optimal solution set.Through case analysis,the performance index of the operation and maintenance plan formulated by the transmission line differential operation and maintenance model based on the optimal solution set is better than that of the original operation and maintenance plan.On the premise of losing reliability,the indexes of economy and workload balance are optimized.