Frequency tracking control of the WPT system based on fuzzy RBF neural network
Yuanyuan Liu,Fei Liu,Hongwei Feng,Guoxin Zhang,Lu Wang,Ronghua Chi,Kexin Li
DOI: https://doi.org/10.1002/int.22706
IF: 8.993
2021-10-13
International Journal of Intelligent Systems
Abstract:With the application of electrical equipment, magnetically coupled resonant (MCR) wireless power transfer (WPT) technology has become an effective means to improve equipment intelligence. The MCR-WPT system is a loosely coupled system, and the resonant frequency may be split or detuned due to the changes of load or transferring distance, resulting in the system transfer efficiency (TE) greatly reduced. To solve the problems of limited speed and accuracy in the existing frequency tracking methods, this paper analyzes the relation between the detuning rate and the system TE, proposing an adaptive frequency tracking control method based on fuzzy radial basis function neural network control. The neural network outputs proportion–integration–differentiation parameters to adjust the inverter drive circuit, and the frequency of inverter drive circuit is adjusted nonlinearly in real time to ensure the accurate frequency tracking of the MCR-WPT system. The simulation and experimental results show that the proposed method can enhance the tracking ability of the resonant frequency, and effectively improve the system TE.
computer science, artificial intelligence