Optimized Design of Permanent Magnet Trip Device Based on Orthogonal Experiments and BP-NSGA-II

To address the issues of slow operation and high variability in traditional electromagnetic trip devices, this paper proposes a magnetic trip device based on the principle of “permanent magnet holding, spring driving, and electric reset,” thereby reducing the circuit breaker tripping time. However, its high material and manufacturing costs have limited its widespread adoption. To address this issue, this paper employs a method combining orthogonal experiments with BP-NSGA-II. Using permanent magnet dimensions, coil wire gauge, moving component mass, and spring initial force as variables, the number of simulations is reduced through orthogonal experiments, and transient electromagnetic simulation is utilized to analyze the trip mechanism’s dynamic performance; a BP neural network surrogate model was constructed to replace finite element simulation, and the NSGA-II algorithm was employed to perform weightless Pareto optimization, with the volume of the permanent magnet and the amount of copper used in the coil as the cost optimization objectives. Under the constraint of a trip time ≤ 15 ms, the permanent magnet dimensions were reduced from 8 × 44 × 5 mm to 6.5 × 44 × 5 mm (a 18.7% reduction in volume), and the coil wire diameter was optimized from 0.18 mm (2000 turns) to 0.14 mm (1400 turns) (a 22.3% reduction in copper usage). Test results show that the optimized trip time was reduced from 16.0 ms to 14.1 ms, with the prototype measuring 14.56 ms in actual testing; the discrepancy between the simulation and experiment was less than 5%. This method provides a reference for the design of a permanent magnet trip device and holds significant engineering value.