Axial leakage flux를 고려한 딥러닝 기반의 PMSM의 설계

초록

A sizing process is required to analyze the electromagnetic properties according to the major shape parameters in the initial design stage of a permanent magnet synchronous motor (PMSM). However, predicting the performance of PMSM by commonly used 2-D finite element analysis (FEA) is error-prone due to axial leakage flux. Therefore, the axial leakage flux should be considered in the sizing process. The most accurate way to consider the axial leakage flux is to perform 3-D FEA, but it has the disadvantage of being computationally expensive. From this perspective, this paper proposes a deep transfer learning-based surrogate modeling method to reduce the computational cost for calculating 3-D FEA-based motor parameters. Transfer learning is performed using a large amount of 2-D FEA-based and a small amount of 3-D FEA-based motor parameters. Using the proposed process, it is possible to accurately predict the motor characteristics according to the size-related variables that satisfy the required specifications with a small amount of 3D FEA-based motor parameters. The proposed method was verified through 3D FEA and experiments for PMSM designed from this method.

키워드