基于Light-Resnet卷积神经网络的电力设备监测数值识别算法  

Numerical Recognition Algorithm for Power Equipment Monitoring Based on Light-Resnet Convolutional Neural Network

作　　者：孔志恒 谭冲[1] 唐培耀 胡成博 郑敏[1] KONG Zhiheng;TAN Chong;TANG Peiyao;HU Chengbo;ZHENG Min(Shanghai Institute of Microsystem and Information Technology,Chinese Academy of Sciences,Shanghai 200050,China;State Grid Jiangsu Electric Power Company Ltd.Research Institute,Nanjing 211103,China)

机构地区：[1]中国科学院上海微系统与信息技术研究所,上海200050 [2]国网江苏省电力有限公司电力科学研究院,江苏南京211103

出　　处：《中国电力》2024年第8期206-213,共8页Electric Power

基　　金：国家电网有限公司科技项目(5108-202218280A-2-201-XG)。

摘　　要：在智能电网中,精确监测输电、配电及供电关键设备的运行状态对在线运维至关重要。面对人工抄录和巡检的低效,以及监测装置数字化升级的复杂安装、高成本和长周期等挑战,结合图像采集装置与图像处理技术,根据计算资源合理分配任务,开发了一种基于Light-Resnet数值识别算法,该算法通过D-Add损失函数优化网络训练过程,实现电力设备监测数据的远程读取。实验表明:Light-Resnet以6090的参数量在MNIST数据集获得了98.8%的严格准确率,结合边端协同机制,终端侧能耗降低了20.73%。这一算法不仅证明了自身在资源受限环境下的适应性和高效性,同时D-Add损失函数的设计也显著提升了网络的准确度。In the smart grid,precise monitoring of the operational status of critical equipment for transmission,distribution and power supply is essential for effective online maintenance.Faced with the inefficiencies of manual recording and inspection,as well as the challenges associated with the complex installation,high cost and lengthy periods required for digital upgrades of monitoring devices,a novel approach that integrates image capture devices with image processing technology has been developed.This approach,leveraging the allocation of computational resources for task distribution,introduces a Light-Resnet-based numerical recognition algorithm,which enhances network training through the optimization of the D-Add loss function,enabling remote reading of electrical equipment monitoring data.Experiments have demonstrated that Light-Resnet achieves a rigorous accuracy rate of 98.8% on the MNIST dataset with only 6090 parameters.When combined with edge computing collaboration mechanisms,it resulted in a 20.73% reduction in power consumption on the terminal side.The proposed algorithm not only proves its adaptability and efficiency in resource-constrained environments but also significantly improves the network's accuracy with design of the D-Add loss function.

关 键 词：light-resnet D-add 边端协同机制 数值识别 智能电网 

分 类 号：TM50[电气工程—电器] TP274[自动化与计算机技术—检测技术与自动化装置] TP183[自动化与计算机技术—控制科学与工程]