Model Reference Adaptive Control of Marine Permanent Magnet Propulsion Motor Based on Parameter Identification

Marine permanent magnet synchronous propulsion motors have problems, such as low reliability and difficult maintenance in the traditional control. In this paper, a sensorless control system for motor (PMSM) based on parameter identification is proposed. According to mathematical model of two-phase rotating coordinate system, reference adaptation (MRAS) used estimate rotor speed position motor. Because MRAS highly dependent parameters, they will change with environment, working state, etc., Adaline neural network identify parameters online, then are corrected. The simulation results show that combined can reduce estimated error by about 50% compared method, reduces angle estimation 96%. It shows accurately rotational motor, it has high accuracy parameters.