Vibration Analysis and Optimization of Iron-Core Reactors Based on Fe-Based Soft Magnetic Composite Materials

To effectively reduce the vibration of iron-core reactors, a vibration optimization method considering the magnetostrictive properties of Fe-based soft magnetic composite materials is proposed. First, an improved magnetostrictive model incorporating stress effects is established based on the classical Jiles–Atherton (J-A) model and the quadratic domain rotation theory. The characteristic parameters of the improved model are identified using the particle swarm optimization–simulated annealing (PSO-SA) algorithm, with the identified root mean square error not exceeding 3.5, verifying the model’s accuracy. Then, an electromagnetic-structural coupled simulation model of the iron-core reactor is developed to calculate the magnetic field and vibration distribution. Based on multiphysics simulation and Latin hypercube sampling, combined with sensitivity analysis techniques, the influence of each parameter on vibration is identified, and optimization objectives and variables are hierarchically classified. Finally, response surface methodology (RSM) and Kriging methods are employed for the parameter optimization design of the reactor, yielding the optimal structural parameters under different optimization strategies. The results show that, compared to the initial parameters, the maximum vibration displacement of the iron core is reduced by 13.93% and 24.64% using the RSM and Kriging methods, respectively. Additionally, both core loss and conductor consumption are significantly reduced. Therefore, under the premise of meeting performance requirements, the Kriging optimization method can significantly reduce the vibration displacement of the iron-core reactor, providing valuable guidance for its vibration reduction and optimization.