摘要
Abstract
In order to solve the problem of slow automatic updating of multi-type grid meter reading and accounting information,an automatic updating model of grid meter reading and accounting information based on fireworks algorithm optimizing extreme learning machine is constructed.The information acquisition layer of the model acquires meter reading and accounting informa-tion in real time through various terminals such as low-voltage centralized reading and specialized variable acquisition,and then sends it to the master station layer through the wireless private network of the information transmission layer.Intelligent power data management module of the master station layer uses the new meter reading and accounting information classification meth-od based on fireworks algorithm optimizing extreme learning machine,obtains the weight and bias parameters that can minimize the training error of extreme learning machine by fireworks algorithm,uses extreme learning with high-precision classification ability to classify the collected meter reading and accounting information,uses Kettle based non-destructive incremental infor-mation to automatically update the model to filter the duplicate information in classified information.The effective meter read-ing and accounting information is saved and used to update information and display it in the interface,and the automatic update of meter reading and accounting information is completed.The experimental results show that the constructed model can realize the updating of meter reading and accounting information,the update speed of various grid meter reading and accounting infor-mation is faster,50 pieces of information can be updated per millisecond.关键词
烟花算法/极限学习机/电网抄表核算信息/自动化更新/Kettle概念模型Key words
fireworks algorithm/extreme learning machine/grid meter reading and accounting information/automatic update/Kettle conceptual model分类
计算机与自动化
