Research On Sensorless Control Of Switched Reluctance Motor Based On Full Cycle Inductance Model

Switched reluctance motors(SRM)are widely used in all kinds of harsh working conditions because of their simple structure,durability and low cost.During normal operation,the switched reluctance motor needs rotor position information to determine the appropriate conduction time,so it is necessary to install position sensor.The position sensor is easy to be damaged under serious working conditions,and the cost of the motor is also increased.Therefore,it is of great significance to study sensorless technology of switched reluctance motor.For better sensorless control of switched reluctance motors,an accurate motor model must be established.In this paper,the traditional linear model and the improved piecewise linear model are analyzed.The two models are not accurate enough for sensorless control.In order to ensure the accuracy of angle estimation,a finite element model with high accuracy was established.And the motor model was built in Simulink based on the data obtained from FEA and the basic mathematical model of the motor.After verifying the traditional phase splitting estimation strategy,the shortcomings are analyzed,and the virtual unit inductance vector method is proposed to estimate the rotor position of the motor according to the existing problems.The inductance space vector is synthesized by the three-phase inductance of the motor,and the specific position of the rotor is determined according to the position of the space vector.The functions of various coefficients under different current are constructed by fitting,and the inductance is unitized and the effect of magnetic saturation is eliminated by calculating the coefficients in the angle estimation algorithm.The inductance is divided into first order series and quadratic series.It is relatively simple to use the first-order series,and after the comparative analysis,the error is only slightly increased than that of the second-order series model,so the first-order series model of inductance is selected.Without considering the complexity of the algorithm,a hybrid Angle estimation algorithm is proposed.The estimates are more accurateFinally,the hardware experimental platform is introduced,and the feasibility of the sensorless control strategy is verified by experiments.