This paper aims to propose an Improved Newton–Raphson-Based Optimizer (INRBO) to enhance the accuracy and convergence speed of Surface-mounted Permanent Magnet Synchronous Motor parameter identification.
A parameter identification model based on intelligent algorithms was established. This paper integrates a Piecewise Chaotic Dynamic Opposition-Based Learning Initialization, a Sine–Cosine Synergized Newton–Raphson Search Rule and a Dual-Mode Adaptive Mutation Strategy with the basic Newton–Raphson-Based Optimizer (NRBO). This integration addresses the limitations of the basic NRBO, creating the INRBO. The superiority of INRBO was validated using four subfunctions from the IEEE CEC2005 test suite and applied to Permanent Magnet Synchronous Motor parameter identification through simulation and experiment.
INRBO shows significant improvements in convergence speed and solution accuracy. Simulation and experimental results confirm its ability to rapidly and accurately identify the stator resistance, inductance and permanent magnet flux linkage of the motor.
The proposed INRBO can maintain high-accuracy electrical parameter identification results even with a small population size (50) and a limited number of iterations (50), making it highly applicable to systems with very limited control capabilities or scenarios with restricted computational resources.
