Design And Implementation Of Early Warning System Based On Insulator Pollution Flash Over Leakage Current

With the gradual development of modern industrial production in China,in recent years,ultra high voltage overhead transmission line in our country has got widespread application and of the power grid voltage developed rapidly,which leads to the pollution flashover phenomenon of the safety problems and more serious economic losses.Today,the power system has accumulated a lot of data in the long run,however,there is little people to dig out the underlying value of these data.Therefore,how to use the historical information and effectively forecast imminent flashover,and making early warning and timely treatment is particularly important before flashover is coming,so the early warning system of insulator pollution flashover is of great engineering significance.In order to reduce the loss caused by pollution flashover and to improve the utilization of large data of power systems,this paper completed the design and implementation of the early warning system based on insulator pollution flashover leaking current.Combined with the characteristics of flashover leaking current,the flashover leakage current is divided into stable zone and the shock region,the theory and method of application of compressed sensing,sparse basis selection of different signals in various regions were sparse processing,matching reconstruction tracking algorithm to complete the signal by orthogonal.After the analysis and processing of leaking current data,it provides real-time monitoring indicators for early warning systems,and completes the visual display of leaking current data.The advantages and feasibility of subsection reconstruction are proved by the method of contrast experiments.Secondly,using related theories and methods of data mining to discover pollution flashover leakage current history data and knowledge,to provide decision support for the early warning system,to reduce the false alarm rates and improve early warning accuracy rates.According to the morphological characteristics of leaking current of the pollution flashover,using adaptive feature extreme point representation to complete the leaking current,the experimental results show that using the method of leaking current characteristics of sequence data not only improves the accuracy of similarity measure in the mining process,but also reduce the computing time,can effectively complete found the historical data of the task of knowledge.Finally,based on the integration of experiment and algorithm,an early warning system design and implementation based on insulator contamination flashover leakage current is completed.