Design and implementation of security inspection system for intelligent distribution station based on field programmable gate array
CHEN Biaofa1, CHEN Chuandong1, WEI Rongshan1, LUO Haibo2
1. School of Physics and Information Engineering, Fuzhou University, Fuzhou 350108; 2. School of Computer and Control Engineering, Minjiang University, Fuzhou 350121
Abstract：In Chinese power system, distribution station belongs to the edge node of power network and is a key link in power system. However, manual patrol inspection or data acquisition methods of traditional hardware equipment can not match the current security patrol inspection requirements due to problems such as cost and efficiency. To solve this problem, this paper proposes a security inspection solution of intelligent distribution station based on FPGA. Firstly, YOLOv4 tiny network is used to realize the functions of helmet wearing detection, work clothes wearing detection and cross-border early warning identification, with an accuracy of 93.5%. Secondly, for the application scenario of power distribution station, this paper uses FPGA to realize the effect of real-time detection on edge devices, and optimizes from the aspects of parallel deployment and pipeline. The results of the experiment show that the system realizes the detection speed of 68 frames per second on ZCU102 platform, and the overall average performance reaches 228 giga operations per second.
陈标发, 陈传东, 魏榕山, 罗海波. 基于现场可编程门阵列的智能配电站安防巡检系统设计与实现[J]. 电气技术, 2022, 23(5): 34-38.
CHEN Biaofa, CHEN Chuandong, WEI Rongshan, LUO Haibo. Design and implementation of security inspection system for intelligent distribution station based on field programmable gate array. Electrical Engineering, 2022, 23(5): 34-38.
[1] 朱志伟. 基于大数据的35kV变配电站智能监控整体解决方案的实现[J]. 电气技术, 2018, 19(7): 63-67. [2] 赵凯, 孙豪文. 10kV配电物联网智慧台区关键技术与实践[J]. 电气技术, 2020, 21(7): 80-84. [3] 周国亮, 朱永利, 王桂兰, 等. 实时大数据处理技术在状态监测领域中的应用[J]. 电工技术学报, 2014, 29(增刊1): 432-437. [4] DING Ru, SU Guangda, BAI Guoqiang, et al.A FPGA-based accelerator of convolutional neural network for face feature extraction[C]//2019 IEEE International Conference on Electron Devices and Solid-State Circuits (EDSSC), Xi’an, China, 2019. [5] ZHANG Yang, SONG Chenglong, ZHANG Dongwen.Deep learning-based object detection improvement for tomato disease[J]. IEEE Access, 2020, 8: 56607-56614. [6] ALGABRI M, MATHKOUR H, BENCHERIF M A, et al.Towards deep object detection techniques for phoneme recognition[J]. IEEE Access, 2020, 8: 54663-54680. [7] REDMON J, DIVVALA S, GIRSHICK R, et al.You only look once: unified, real-time object detection[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016: 779-788. [8] JIANG Zicong, ZHAO Liquan, LI Shuaiyang, et al.Realtime object detection method based on improved YOLOv4-tiny[EB/OL]. [202011]. https://arxiv.org/abs/ 2011.04244v1. [9] JACOB B, KLIGYS S, CHEN B, et al.Quantization and training of neural networks for efficient integer- arithmetic-only inference[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, US, 2018. [10] 黄兆伟. 基于FPGA的深度卷积神经网络硬件加速方法研究[D]. 长春: 东北师范大学, 2020. [11] 朱建鑫, 胡海兵, 陆道荣, 等. 应用于级联STATCOM的高精度低成本全FPGA实时仿真模型研究[J]. 电工技术学报, 2019, 34(4): 777-785. [12] ZHANG Shiguang, CAO Jian, ZHANG Quan, et al.An FPGA-based reconfigurable CNN accelerator for YOLO[C]//2020 IEEE 3rd International Conference on Electronics Technology (ICET), Chengdu, 2020: 74-78. [13] ADIONO T, PUTRA A, SUTISNA N, et al.Low latency YOLOv3-tiny accelerator for low-cost FPGA using general matrix multiplication principle[J]. IEEE Access, 2021, 9: 141890-141913. [14] PESTANA D G, MIRANDA P R, LOPES J D, et al.A full featured configurable accelerator for object detection with YOLO[J]. IEEE Access, 2021, 9: 75864-75877.
2022, Vol. 23 
