For WPT systems, when the distance between the receiving coil and the transmitting coil shifts greatly, the efficiency of wireless charging is affected because the magnetic fields between the coils cannot be perfectly matched. In addition, WPT coil excursion can also lead to electromagnetic leakage and heat generation, which can damage the device or reduce the life of wireless charging. Therefore, in view of the influence of coil offset, this paper establishes a coupling model of the multi-transmitter coil wireless charging system, studies the relationship between the current of each transmitter coil and the load position, explores the relationship between the current of the transmitter coil and the position of the receiving coil, and proposes a region-based modeling method based on BP neural network, and the prediction error of the optimized region-based positioning model is less than 10 mm, so as to achieve high-precision coordinate prediction effect. The above concepts are simulated and experimentally verified. The results show that the average experimental prediction error is 10.07 mm, which basically meets the set 10 mm positioning accuracy, and the simulation and experimental verification meet the set requirements.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Load Positioning Strategy of WPT System Based on BP Network

  • Feng Wen,
  • Ge Yu,
  • Jiaqi Guo,
  • Jiangtao Gao,
  • Qiang Li,
  • Zhijun Yao

摘要

For WPT systems, when the distance between the receiving coil and the transmitting coil shifts greatly, the efficiency of wireless charging is affected because the magnetic fields between the coils cannot be perfectly matched. In addition, WPT coil excursion can also lead to electromagnetic leakage and heat generation, which can damage the device or reduce the life of wireless charging. Therefore, in view of the influence of coil offset, this paper establishes a coupling model of the multi-transmitter coil wireless charging system, studies the relationship between the current of each transmitter coil and the load position, explores the relationship between the current of the transmitter coil and the position of the receiving coil, and proposes a region-based modeling method based on BP neural network, and the prediction error of the optimized region-based positioning model is less than 10 mm, so as to achieve high-precision coordinate prediction effect. The above concepts are simulated and experimentally verified. The results show that the average experimental prediction error is 10.07 mm, which basically meets the set 10 mm positioning accuracy, and the simulation and experimental verification meet the set requirements.