基于多层前馈神经网络的反窃电系统研究  

Research on Anti-power theft System Based on Multi-layer Feedforward Neural Network

作　　者：张合川 石盼 王磊 史辉 徐相波 ZHANG Hechuan;SHI Pan;WANG Lei;SHI Hui;XU Xiangbo(State Grid Jibei Electric Power Company Limited Skills Training Center,Baoding 071051,China)

机构地区：[1]国网冀北电力有限公司技能培训中心,河北保定071051

出　　处：《微型电脑应用》2022年第8期58-61,78,共5页Microcomputer Applications

摘　　要：面对国家电网中的窃电问题,充分考虑产生窃电问题原因,研究基于多层前馈神经网络学习算法的反窃电系统,提升电网经济效益。通过系统采集层的上位机采集卡、智能电表、计量设备等采集用电原始数据,经通信层的光纤网或无线网将采集层采集的用电数据交互传输至主站层,通过主站层汇总、存储采集数据后,采用多层前馈神经网络学习算法分析数据得出窃电行为分析结果,通过表现层为操作者提供展示界面,便于操作者及时发现窃电具体位置并下发核查单,实现反窃电检测。实验结果表明该系统可分析窃电数据并及时上报客户端,系统应用后,可节省用电量,提高供电公司的经济效益、电网公司管理效率和网络资源利用率,反窃电效果较好。Facing on the problem of power theft in the state grid,we fully consider the reasons for the power stealing,and the anti-power theft system based on the multi-layer feedforward neural network learning algorithm to improve the economic benefits of the power grid.We collect raw data of electricity consumption through the host computer acquisition card,smart meter,metering equipment,etc.The system acquisition layer interactively transmits the electricity consumption data and send to the master station layer through the optical fiber network or wireless network of the communication layer.After collecting and storing the data,the multi-layer feedforward neural network is used to analyze the data to obtain the analysis results of the stealing behavior,and the display interface is provided for the operator through the presentation layer,so that the operator can find the specific location of the stealing in time and issue a checklist,which realizes anti-theft detection.The experimental results show that the system can analyze the power theft data and report it to the client in time.After the system is applied,it can save electricity consumption,improve the economic benefits of the power supply company,the management efficiency of the grid company and the utilization of network resources,and the anti-power theft effect is better.

关 键 词：多层前馈 神经网络 学习算法 反窃电系统 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]