Research On Key Technology Of Electric Equipment Condition Maintenance System Based On Hadoop Cloud Platform

With the improvement of the grid size and the rise in demand for electricity,electricity plays an important role in our daily life,it is the basis of national economic development.At present,power supply companies generally use schedule maintenance to ensure the safe operation of power system and reduce the occurrence of power outages.Regular maintenance carry out maintenance work in accordance with a fixed period of time on the power equipment,rather than built on the basis of the operation of the equipment.This approach has a blindness,greatly reducing the accuracy of maintenance,resulting in overhaul maintenance and leak Maintenance,resulting in economic losses and affect the service life of the equipment.Electric equipment will be stopped in regular maintenance work,Power failure caused the loss of electricity sales,and the indirect economic losses by power failure of about 10 times the loss of electricity bill.Therefore,the regular maintenance mode of operation can't meet the requirements of the development of smart grid.The condition maintenance of power equipment is a kind of on-demand maintenance,based on data acquisition and analysis,overcome the blindness of regular maintenance,become the current development trend.Power equipment condition maintenance refers to the equipment running data real-time monitoring,based on the data monitored in the field,the eigenvalues of the data are extracted to determine the running status of the equipment.But electric equipment is complex and huge,how to effectively store large amounts of data and data analysis is the two major problems of state maintenance.In recent years,Hadoop is an open source cloud computing platform,it is widely used in the field of large data analysis,which is introduced into power system,provided super computing capability for power system and improved the capacity of grid data storage and data processing.This paper provides a solution for large-scale data storage and data analysis by building a Hadoop cloud computing cluster.In this paper,we analyze the research hotspots and difficult problems of current electric power equipment,and establish a model of cable equivalent thermal path model and icing line mechanics.With the Hadoop cloud platform which high performance and reliability have the advantages of large data processing,a set of power equipment condition monitoring systems for cables and icing lines has been designed.The core functions of the system include power equipment status assessment,large-scale power equipment operation data storage,data preprocessing and data analysis,intelligent decision-making,etc.,provided decision support for power state maintenance.The system can improve the power supply company equipment management level and power supply reliability,and ensure the safety of the power grid,economic operationIn the system design,this paper carries on the research and the analysis to the system demand.On this basis,a set of system design framework is proposed,which uses filter algorithm to remove noise interference to ensure data quality,the BP neural network algorithm and the analytic hierarchy process are used to evaluate the state of the electric power equipment.In the aspect of system realization,the Hadoop cloud computing environment is set up,and the MapReduce parallel programming model is used to analyze the functional modules and complete the experimental analysis to achieve the intended purpose.Finally,the research work is summarized and the next work is forecasted.