Research on energy consumption optimization of RetinaNet model based on lightweight improvement in edge intelligent terminals in power system
DOI:
https://doi.org/10.54097/vna6fq75Keywords:
Edge computing, RetinaNet, Structural design, Model pruningAbstract
With the increasing application of edge computing in power systems, intelligent terminal devices are facing new challenges in processing capabilities and energy optimization. As a highly efficient target detection model, RetinaNet's potential application in edge intelligent terminals of power systems has not been fully explored. This study proposes a lightweight model based on the convolutional neural network RetinaNet. Multiple lightweight models are used to replace the original backbone network ResNet for comparison. The best model is selected while ensuring accuracy. Redundant connections in the model network are pruned through channel pruning to reduce model size and improve detection speed. The results show that compared with the original algorithm, the RetinaNet model proposed in this study reduces parameter count by 73%, decreases computational load by 41.8%, reduces model size by 72.7%, and only decreases the mean average precision (mAP) value by 1.8 percentage points.
References
[1] Cui Hengzhi, Jiang Chengling, Miao Weiwei, et al. Design and implementation of intelligent electric power IoT system based on edge computing[J].Electric power information and communication technology,2020, 18(4): 33-41. CUI Hengzhi, JIANG Chengling, MIAO Weiwei, et al. Design and implementation of power intelligent IoT system based on edge computing[J]. Electric Power Information and Communication Technology, 2020, 18(4): 33-41(in Chinese).
[2] SUN Haoyang, ZHANG Jichuan, WANG Peng, et al. Edge computing technology for power distribution Internet of Things[J].Power Grid Technology, 2019, 43(12): 4314-4321. SUN Haoyang, ZHANG Jichuan, WANG Peng, et al. Edge computation technology based on distribution internet of things[J]. Power System Technology, 2019, 43(12): 4314- 4321(in Chinese).
[3] QIU Shuguang,PANG Chengxin,JIA Jia. Research on the application of LPWAN and edge computing fusion in power Internet of Things[J].Internet of Things technology,2019, 9(7): 63-66.
[4] LI Yuke. Design and application of wireless power monitoring and control system based on Internet of Things and PLC[J]. Electrical technology and economy,,2024,(06):376-379.
[5] Duan Jian,Wang Xinchao,He Xiaoyang. Power grid visualization construction based on embedded video surveillance technology [J]. Information technology, 2019, 43(11):169-172+176.DOI:10.13274/j.cnki.hdzj.2019.11.035.
[6] YANG Caiwei. Design of intelligent power terminal[D]. Changchun Institute of Technology, 2023. DOI: 10.27834 / d.cnki.ggccc.2023.000110.
[7] Gu Jian,Xin Zhanqiang,Luo Zhixiong. Application of temperature monitoring system in energy storage battery compartment[J]. Mass electricity,2024,39(01):44-45.
[8] LI Xin, LAI Ji, CHEN Zhongtao, et al. Intelligent fault detection system for state grid IoT equipment based on edge computing[C]//2019 IEEE 3rd International Electrical and Energy Conference (CIEEC). Beijing: IEEE, 2019 : 1434-1439.
[9] Qi Yincheng, Jiang Aixue, Zhao Zhenbing, et al. Transmission line inspection image fitting detection method based on improved SSD model[J].electrical measurement and instrumentation,2019,56(22):7-12+43. DOI:10.19753/j.issn1001-1390.2019.022.002.
[10] PENG Ziyang, CHEN Nuotian, YI Junfei, et al. Transmission line power devices and abnormal target detection based on improved RetinaNet algorithm[J].Hunan Electric Power, 2023, 43(05):79-84.
[11] ZHAO Qiang. Theory and application of transmission line fittings[M].Beijing: China Electric Power Press, 2013. 2-3.
[12] ZHOU Chunxin,HUO Yizhi,DU Youhai,JIANG Minlan, ZENG Lingguo, ZHANG Changjiang, SHI Xiaowei. Nondestructive testing of soybean appearance quality based on improved RetinaNet[J/OL]. Chinese Journal of Cereals and Oils..
[13] HE K, ZHANG X, REN S, et al. Deep Residual Learning for Image Recognition[C]. Proceedings of the IEEE conference on computer vision and pattern recognition, 2016: 770-778
[14] Lin T Y , Goyal P , Girshick R ,et al.Focal Loss for Dense Object Detection[J].IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, PP(99):2999-3007. DOI: 10.1109 / TPAMI.2018.2858826.
[15] Ma N , Zhang X , Zheng H T ,et al.ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design[C]//European Conference on Computer Vision. Springer, Cham, 2018. DOI:10.1007/978-3-030-01264-9_8.
[16] Howard, A., Zhu, M., Chen, B., & Kalenichenko, D. (2017). Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861.
[17] LIU Z, LI J, SHEN Z, et al. Learning efficient convolutional networks through network slimming[C]// Proceedings of the 2017 IEEE international conference on computer vision. Piscataway: IEEE, 2017: 2736-2744.
[18] WU D, LV S, JIANG M, et al. Using channel pruning-based YOLOv4 deep learning algorithm for the real-time and accurate detection of apple flowers in natural environments[J]. Computers and Electronics in Agriculture. 2020, 178: 105742.
[19] WANG D, HE D. Channel pruned YOLO V5s-based deep learning approach for rapid and accurate apple fruitlet detection before fruit thinning[J]. Biosystems Engineering, 2021, 210: 271-281.
[20] FANG G, MA X, SONG M, et al. Depgraph: Towards any structural pruning[C]// Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2023: 16091-16101.
[21] CHEN X, ZHU J, JIANG J, et al. Tight compression: compressing CNN model tightly through unstructured pruning and simulated annealing based permutation[C]// Proceedings of the 57th ACM/IEEE Design Automation Conference (DAC). Piscataway: IEEE, 2020: 1-6.
[22] SITU Z, TENG S, LIAO X, et al. Real-time sewer defect detection based on YOLO network, transfer learning, and channel pruning algorithm[J]. Journal of Civil Structural Health Monitoring, 2024, 14(1): 41-57.
[23] YANG F. An improved YOLOv3 algorithm for remote Sensing image target detection[C]// Proceedings of the 2021 Journal of Physics: Conference Series. Bristol: IOP Publishing, 2021, 2132(1): 012028.
[24] CHENG X , ZHANG Y , CHEN Y ,et al. Pest identification via deep residual learning in complex background[J].Computers and Electronics in Agriculture, 2017, 141:351-356
[25] SUH H K , JORIS I , WILLEM H J ,et al. Transfer learning for the classification of sugar beet and volunteer potato under field conditions[J].Biosystems Engineering, 2018, 174:50-65
[26] MILELLA A , MARANI R , PETITTI A ,et al. In-field high throughput grapevine phenotyping with a consumer-grade depth camera[J]. Computers and Electronics in Agriculture, 2019, 156:293-306
[27] YAHYA A,ZAFER C, KOCAMAZ A F .Identification of haploid and diploid maize seeds using convolutional neural networks and a transfer learning approach[J].Computers and Electronics in Agriculture, 2019, 163(2019):1-11.
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.