Mutual Inductance Identification Of Wireless Power Transfer System Applied To The Modern Tram

The combination of wireless power transfer technology and EDLC(Electrical Double Layer Capacitor)energy storage technology is a feasible scheme of catonery-free power supply for modern tram,which possesses features of safety,reliability and flexibility in selecting charging modes.However,the uncertainty of parameters,susch as coupling coils mutual inductance,would extremely impact on the control performance of wireless power transfer system.Therefore,to reduce the uncertainty of parameters is considered as a hot spot of research in this field.A real-time coupling coils mutual inductance identification method is proposed in this thesis to reduce the uncertainty of coupling coils mutual inductance.Besides,an equivalent sampling frequency rising method based on sliding discrete Fourier transform algorithm is proposed to sensor noise and high-order harmonics upon the detected signals.Finally,simulations and experiments are conducted to verify the effectiveness of the mentioned method.The main contents of this thesis are as follows.First of all,the topology of main circuit and coupling system are given to meet the requirement of wireless power transfer system applied in modern tram power supply system.Based on the proposed circuit topology,two modeling method of wireless power transfer is proposed in this thesis,such as the static model based on mutual inductance model and the dynamic model based on coupled modes theories.Then,a mutual inductance identification method for wireless power transfer coupling coils is designed in this thesis,which includes two stages,namely transmitter parameters off-line identification and coupling coil mutual inductance on-line identification.Identification models of transmitter parameters and coupling coils mutual inductance are derived based on mutual inductance model.Then the feasible region of mutual inductance identification model is also derived.Next,during the process of mutual inductance identification,a filter to extract the fundamental amplitude of input signals is designed based on sliding discrete Fourier transform algorithm to reduce sensor noise and high-order harmonics upon detected signals.Then a sampling method is proposed to achieve a higher equivalent sampling frequency which improves the parameter detection precision during mutual inductance identification.In this method,sampling points in several periods are sequenced and considered as points in one period.Finally,the accuracy and the feasibility of mutual inductance identification model and sampling and filtering method.