| Power cables are an indispensable part of the power transmission and distribution system,and the safety and reliability of their operations are valued by power operation and maintenance personnel and researchers.In 2019,the State Grid put forward the strategic goal of "three types,two networks,world-class".The ubiquitous power Internet of Things is one of the "two networks".The focus of its construction is the collection and comprehensive processing of massive data in the power system.In order to provide a solution for the complete construction of the ubiquitous power Internet of Things system,this thesis designs the overall plan of the cable information comprehensive processing system based on the requirements of the cable information comprehensive processing system and combining the microservice architecture technology,and studies the operating cable partial discharge Signal processing method and realize the function of the system.This thesis mainly carried out the following work:First,based on the Spring Cloud framework,a comprehensive cable information processing system for the ubiquitous power Internet of Things was designed,and the design of a dedicated gateway in the ubiquitous power Internet of Things was completed.In order to solve the problems of high system coupling,low development efficiency and poor scalability brought about by the development of a ubiquitous power Internet of Things system with a monolithic architecture,this thesis designs a comprehensive cable information processing system based on the Spring Cloud microservice framework.According to the requirements of the ubiquitous power Internet of Things and the principle of dividing microservices,the cable information integrated processing system is vertically divided into 7 independent microservices,which realizes the independent design,independent development and independent deployment of each microservice.In order to solve the problem of different terminal equipment access protocols and data formats,the MQTT protocol is used to complete the design of a dedicated gateway in the Internet of Things,which realizes the unified management of data in the system.Secondly,this thesis designs an online monitoring system for partial discharge signals based on distributed operating cables,and constructs various microservices in the integrated cable information processing system.In order to realize real-time online monitoring of running cables,based on the study of the principle of partial discharge signal collection,a distributed partial discharge signal collection system based on running cables is designed.In order to better construct the various microservices in the cable information integrated processing system,My SQL relational database is used to store data,and the database and sequence diagrams of the cable information integrated processing system are designed in detail.Then,this thesis studies the binary tree multi-classification support vector machine algorithm based on DE-PSO to identify the four stages of electrical tree and its discharge,and realizes the evaluation of the insulation performance of the running cable in the ubiquitous power Internet of Things.The algorithm uses principal component analysis to extract feature parameters with high cumulative contribution rate to form the input data set of the support vector machine classifier,applies the binary tree multi-classification theory to construct the classifier,uses the DE-PSO algorithm to solve the optimal parameters of the classifier,and trains The classifier makes it achieve optimal recognition performance.Through data simulation verification,the results show that the applied algorithm can identify electrical tree branches and the four stages of their development,which improves the accuracy of the evaluation of the insulation performance of operating cables.Finally,this thesis describes the operation example of the cable information integrated processing system,and it has been successfully deployed and operated in the cooperative enterprise.Practice has proved that the system meets the needs of the ubiquitous power Internet of Things with large data volume,large request volume and high online analysis accuracy.Improve the automation level of cable operation and maintenance. |