Sensorless position estimation of switched reluctance motor at startup using quadratic polynomial regression

Sensorless position sensing of switched reluctance motor (SRM) has been of great interests to researchers for reducing costs and increasing reliability of the system. The startup position estimation is a difficult task. This study presents a new method to estimate motor phase positions during startup. It is based on the general magnetic characteristics of the inductance profile in an SRM. All phase positions are estimated without using any specific magnetic information. The calculation is simple and can be implemented easily and executed efficiently in a microcontroller.