Design Optimization of Permanent Magnet Coupler Based on Physics-Informed Neural Networks

The non-contact transmission product permanent magnet coupler (PMC) has been widely used in industry due to its advantages such as low noise and vibration, high efficiency, high reliability, and overload protection. Owing to its complex electromagnetic behaviors, improving torque is crucial for the transmission performance of PMC. However, traditional optimization methods have problems with poor optimization effect and high computational costs. To address the above issues, this paper proposes a novel optimization method based on physics-informed neural networks (PINN) to optimize PMC performance with high accuracy and low computational costs. The design parameters of PMC are integrated into the typical PINN, and the gradient of the objective function relative to the design parameters is calculated and updated to obtain the optimal design parameters. We demonstrate the proposed method through single parameter optimization and multi-parameter optimization. The reliability, effectiveness, and accuracy of the PINN optimization method are confirmed through experiments. Overall, the PINN optimization method combines the data-driven model with prior knowledge, and first achieves the optimal design of PMC by PINN in electromagnetics.