电器与能效管理技术 ›› 2024, Vol. 0 ›› Issue (12): 69-76.doi: 10.16628/j.cnki.2095-8188.2024.12.012

• 充电桩技术 • 上一篇    下一篇

基于V-I轨迹的非侵入式电动自行车充电行为在线辨识

段佳其1, 鲍光海1, 方艳东2   

  1. 1.福州大学 电气工程与自动化学院, 福建 福州 350108
    2.浙江天正电气股份有限公司, 浙江 乐清 325600
  • 收稿日期:2024-06-24 出版日期:2024-12-30 发布日期:2025-01-02
  • 作者简介:段佳其(1996—),男,硕士研究生,研究方向为电气装备测试与在线监测技术。|鲍光海(1977—),男,教授,博士生导师,研究方向为电器及其系统智能化与故障诊断。|方艳东(1973—),男,工程师,主要从事低压电器检测和装配设备自动化开发设计。
  • 基金资助:
    福建省科技计划项目(2023H0007)

Online Identification of Non-Invasive Electric Bicycle Charging Behavior Based on V-I Trajectory

DUAN Jiaqi1, BAO Guanghai1, FANG Yandong2   

  1. 1. College of Electrical Engineering and Automation, Fuzhou University, Fuzhou 350108, China
    2. Zhejiang Tianzheng Electric Co.,Ltd, Yueqing 325600, China
  • Received:2024-06-24 Online:2024-12-30 Published:2025-01-02

摘要:

为杜绝安全隐患,利用V-I轨迹和改进MobileNetv2模型对入户充电行为进行在线辨识。设计实验场景,从采样率选取、迁移学习、泛化性和不同网络对比4个方面验证模型性能,最后把模型部署到上位机和K210芯片上。上位机系统在电动自行车单独充电时准确识别,当充电行为和常用家庭负载混合运行时,识别准确率达到98%以上。

关键词: V-I轨迹, 改进MobileNetv2模型, 迁移学习, 在线识别

Abstract:

To prevent the safety hazards,the V-I trajectory features and improved MobileNetv2 model are used for the online identification of the household charging behavior.The experimental scenarios are designed to validate the model performance from four aspects:sampling rate selection,transfer learning,generalization,and comparison of different networks.Finally,the model is deployed to the computer and the K210 chip.The online recognition system based on the upper computer can accurately identify electric bicycles when charging separatly,and the recognition accuracy is over 98% when charging behavior is mixed with commonly used household loads.

Key words: V-I trajectory, improved MobileNetv2 model, transfer learning, online recognition

中图分类号: