| Permanent magnet synchronous motor(PMSM)has been widely used in many applications such as electric vehicle,aerospace and other fields due to its simple structure,reliable operation and good speed regulation performance.In the motor drive systems,rotor position information is very important,the accuracy of which has an impact on the control performance of the motor.Therefore,the resolvers―a kind of position sensor with strong anti-interference ability and convenient installation,are often used to detect the position information.In order to solve the problem of large quantization error in the resolver-to-digital conversion method under the traditional cosine excitation,a method based on square wave excitation and a new signal demodulation strategy are proposed.Meanwhile,the whole circuit based on MCU is designed for rotor position measurement.The proposed method takes square wave as the excitation signal,and uses the proportional relationship between the output signal and the sine and cosine of the rotor position to realize the demodulation,which can be completed by multiplying the excitation signal and the output signal.This method relieves the constraint that the sampling frequency is consistent with the excitation frequency of the resolver during the signal demodulation process under traditional cosine excitation method.The sampling frequency is independent from the excitation frequency,and the position accuracy of the resolver can be improved by increasing the sampling frequency.In addition,the method proposed in this paper is also applicable to the sine-cosine resolver and the proportional resolver in the brushless resolver.Finally,the rotor position detection experimental platform of PMSM Based on resolver is set up.By comparing the proposed method with the traditional resolver-to-digital conversion method under the cosine excitation,the advantages of the proposed method are demonstrated: The implementation of demodulation is independent of sampling,and the rotor position detection accuracy can be improved by increasing the sampling frequency;Experiments under different operating conditions show that the proposed method has good detection performance both in steady state and dynamic state.The effectiveness and feasibility of the method is verified by experimental results. |