Position Sensorless Control Without Predefined Parameters And On-line Flux Linkage Profile Correction For Switched Reluctance Motors

Switched Reluctance Motor (SRM) is a novel AC driving system. It has many advantages such as simple structure, high reliability, low cost and so on. Besides, its efficiency is relatively high over a wide speed range. SRM has attracted even more attentions.To ensure proper operations of a SRM, a position sensor was required to offer the rotor position information. However, it increased the cost, reduced the reliability of the system and required extra space for installations. Therefore, the sensorless control techniques, in which the position can be obtained using the current and voltage signal indirectly, are the focus in SRM researches. Existing sensorless control method relies on the predefined motor model; however in practice, the magnetics profile of the motor varies during the manufacture and the daily operation, which may lead to errors or even failures. This limits the application of sensorless control in products. To solve the issue, wide speed range sensorless control methods without predefined magnetic parameters were investigated. For the standstill case, an initial position estimation method using bootstrap circuit was proposed. This method estimates the unsaturated inductance by analyzing the time required for the charging current of the bootstrap circuit to reach its maximum value. Then the geometry characteristics of the motor inductance profile were used to obtain the accurate initial position. For the low speed and high speed condition, first, a fast on-line flux linkage measurement method was proposed. The current pulses were applied to all phases after the rotor was moved to the aligned position. The pulse response and the2nd order Fourier series model was employed to obtain the flux linkage profile under different currents and different positions. Then the measured flux linkage profiles were used to implement the Flux/Current sensorless control method, with which the sensorless control of SRM from low speed to high speed can be achieved without predefined parameters.During the practice of product development, another important issue is the rotor eccentricity, which could lead to deterioration of the motor performance. The airgap length of SRM was relatively small; the influence of the rotor eccentricity can’t be neglected. Flux linkage profile measurement under rotor eccentricity condition is essential to solve the problem. To this end, an on-line flux linkage correction method considering the rotor eccentricity effects was proposed. The difference between actual flux linkage and estimated flux linkage was used to correct the amplitude and phase of the predefined2nd order Fourier series flux linkage model. This method can be used to improve the performance of SRM drive system, it can also be used to implement the real-time rotor eccentricity fault diagnosis algorithms.Experiment platforms were implemented to verify the feasibility and accuracy of the proposed methods. The causes of errors were analyzed by simulations. The guidelines were given for practical engineering applications.