Torque estimation of sensorless SRM drive using adaptive-fuzzy logic control

This paper presents a novel sensor-less SRM drive based on adaptive-fuzzy logic control. In this method, SRM drive is model-based and current error between actual current and model current is from the proposed speed estimation. Speed estimation has been calculated in two stages; in first stage, current error is used for the compensation of SRM current and next stage obtains compensated torque using compensated current. Finally, compensated torque and electro mechanical differential equations are applied to speed estimation. This sensor-less SRM drive has been simulated using MATLAB/SIMULIK for two speed ranges. These results show that proposed sensor-less SRM drive has good performance in various speeds. Also, on-line adaptive-fuzzy logic improves compensated current and torque estimation. This sensor-less SRM drive has advantages of estimation accuracy, simple structure, fast computation, and low cost.