Calculation of Iron Loss in Permanent Magnet Synchronous Motors Based on PSO-RNN

This paper proposes a Particle Swarm Optimization (PSO) Recursive Neural Network (RNN) method independent of the empirical iron loss formula to analyze the iron loss in a permanent magnet synchronous motor (PMSM). This model establishes an iron loss calculation model considering high-order harmonic, rotating magnetization, and temperature factors. Considering the influence of multi-factors, the model studies the law of loss change under different magnetic flux densities, frequencies, and temperature conditions. To avoid the deviation problem caused by traditional polynomial fitting, the multi-layer RNN and PSO are used to train and optimize the neural network. Using the machine learning method, the iron loss model versus input parameters is established. The iron loss in complex cases beyond the measurement range can be estimated accurately. The proposed method helps achieve a PMSM iron loss calculation model with broad applicability and high accuracy.