A Hybrid Optimization Method Based on Simplified Mechanical Model for High-Stress Magnets

Pulsed high-field magnets are important tools for modern scientific research. Given the current state of material development, stress-optimized design is a primary approach to achieving high magnetic fields. Nevertheless, the analysis and optimization of the structural mechanics for pulsed magnets are highly complex, involving various factors such as multi-physics coupling, elastic–plastic analysis, and conductor-fiber laminate structures. Traditional manual design methods are inefficient and unable to achieve the global optimal solution. Therefore, the development of automated optimization methods is important. However, magnet optimization involves hundreds of parameters, making intelligence optimization challenging. This article proposes a simplified analytical mechanical model for the reinforcement structure design of pulsed magnets, which reduces the number of variables to approximately ten and enables the application of heuristic optimization algorithms. Thus, this article proposes a hybrid optimization architecture based on the multi-objective particle swarm optimization (MOPSO). The analytical model and finite-element analysis (FEA) are deeply embedded in the MOPSO algorithm as a calculation module to achieve efficient and effective optimization. The obtained results show that the proposed method is valid and efficient for optimizing high-stress magnets.