Sensorless Vector Control of Three-Phase Permanent Magnet Synchronous Motor Based on Model Reference Adaptive System

The speed sensorless vector control system not only saves the speed sensor but also reduces the size of the system device. It reduces the cost and improve the system's operating accuracy and reliability, which has strong practical value and economic benefits. Now many scholars have engaged in research in this area and achieved some results in simulation research, laying a solid foundation for practical applications basis. The model reference adaptive system(MRAS) was developed from the late 1950s which belongs to a type of adaptive system. From the structure, MRAS can be divided into three parts: adjustable model, reference model and adaptive law. The idea of MRAS identification is to use an expression that does not contain unknown parameters as the expected model, and an expression that contains the parameters to be identified is used in the adjustable model. And the two models have the same physical significance of output. Using two models of the difference in the output amount is achieved through the appropriate adaptive law to identify the parameters of the PMSM (Permanent Magnet Synchronous Motor). This paper designs a three-phase PMSM sensorless vector control system based on MRAS. The simulation model and results are given and analyzed.