Optimization of Solenoid Valve Armature Structure Using a Surrogate Model-Based Approach

Solenoid valves, serving as critical actuators in automated control systems, require meticulous structural optimization to enhance operational efficiency. To address the computational inefficiency inherent in high-fidelity multiphysics simulations, this study proposes a Kriging surrogate modeling framework driven by Latin Hypercube Sampling (LHS) and finite element analysis (FEA). This approach reduces computational costs while maintaining prediction accuracy, with the Kriging model establishing a robust nonlinear mapping between armature geometry parameters (H and L) and transient response metrics. A NSGA-II (Nondominated Sorting Genetic Algorithm II) multi-objective optimization framework was subsequently deployed, identifying a Pareto-optimal armature configuration (H=O.77 mm, L=4.17 mm) that minimizes both valve opening time (Tt=4.1 ms) and closing time (T2=6.8 ms). FEA validation confirmed the model's reliability, achieving Tt=4.3ms and T2=7.2ms with relative errors of 4.65% and 5.56%, respectively — both within the industrial tolerance threshold «6%). The optimized design offers a scalable methodology for energy-efficient electromagnetic actuator design in automotive and robotic systems.