A Multi-Objective Finite-Element Method Optimization That Reduces Computation Resources Through Subdomain Model Assistance, for Surface-Mounted Permanent-Magnet Machines Used in Motion Systems

Abstract

Surface-mounted permanent-magnet (PM) motors are widely used in motion control systems because of their high peak torque-to-inertia ratio. These motors exhibit high magnetic saturation to produce high peak torque. A precise finite-element model (FEM) is needed to optimize these motors. It requires high computation power, especially for multi-objective optimizations. On the contrary, subdomain models require low computing power but are inaccurate when magnetic saturation occurs. To solve this problem, we use a subdomain-assisted FEM optimization method (SAFEMOM) that combines subdomain models for preoptimization and FEM for refined optimization. We provide the quantitative measurement methods to compare SAFEMOM with FEM-only optimization. If the constraints in the magnetic fluxes in the subdomain part of SAFEMOM are based on the actual saturation value of the materials, then the advantage of SAFEMOM is not significant. The machines in this study use non-linear material with a knee point at 1.3 T and hence show heavy magnetic saturation above 1.5 T. In that case, contrary to what could be intuitively thought, we need to increase the magnetic flux density limit to 2.6 T -3.0 T in SAFEMOM to have a significant advantage. SAFEMOM reduces about 80% of computing time to obtain a slightly better convergence than the one using FEM only. Also, if limited computing resource is allowed, SAFEMOM gives an error reduced by a factor of nearly eight compared to the optimization error using FEM only. Those results are validated on a family of surface-mounted permanent-magnet machines with a wide range of design parameters.

Automatic Design System With Generative Adversarial Network and Vision Transformer for Efficiency Optimization of Interior Permanent Magnet Synchronous Motor

“…Das et al [22] performed a sensitivity analysis on the noise vibration performance of the permanent magnet synchronous motor (PMSM) for ten geometric parameters and performed design optimization for the highly sensitive parameters. Pfister et al [23] proposed a method to perform an optimization of a PMSM assuming linear magnetic material properties, followed by FEA optimization with a small number of generations. Although these methods effectively optimize the geometry, they determine the motor characteristics only at a single or a small number of current settings, making them unsuitable for IPMSMs operating over wide current ranges, such as those used in automotive applications.…”

A Novel Optimization Framework for High Dynamics Point-to-Point Direct Drive Motion Control System With a New Type of Surrogate Model

“…For the magnetic field models of the DDM, except the FEM model, a subdomain model is used to reduce the computation resource of optimization [26]. The subdomain model is a quick and precise semi-analytical model if the magnetic saturation effect is negligible [18].…”

“…The next optimization with FEM inherits the pre-optimal results as the initial design. The BCE-IBEA [29], NRBF, and robust filters [26] are used to get uniform sampling on the 5D Pareto front to generate the DDM optimization dataset.…”

“…They are further optimized using the FEM for magnetic field computation. This method reduces the computation resource by about 70% [26]. The population is 210, and the number of generations is 400 and 100 for the optimization using the subdomain model and FEM, respectively.…”

“…A single computation case involves computing the magnetic and thermal models for a single DDM design. The niche-radius-based filter (NSBF) and robust filter [26] are used to get uniform sampling on the Pareto front based on the optimization results from the different generations. Finally, the DDM optimization dataset is obtained.…”