Design optimization of a 1/2-coil type contact based on NSGA-Ⅱ optimization algorithm and GABP neural network optimal

With the continuous growth of demand for high-voltage transmission networks, improving the breaking capacity of vacuum circuit breakers (VI) has become an important research frontier, and the core challenge is to maintain arc stability under extreme current conditions. Current research indicates that precise magnetic field regulation during arc ignition is crucial for maintaining the arc in a diffuse state. Among various methods, optimizing contact design to achieve axial magnetic field (AMF) control has become the main strategy. In this article, a three-dimensional model of a 1/2 turn AMF coil contact was established, and the transient magnetic field distribution was calculated using finite element simulation software. The BP neural network model optimized by genetic algorithm (GABP) was combined with non dominated sorting genetic algorithm II (NSGA-II) to optimize the contact structure, effectively improving the breakdown performance of vacuum insulators. The proposed framework not only advances the theory of contact design, but also provides practical guidance for manufacturing high-voltage vacuum insulator components.