Research on Adaptive Design and Optimization of High-Speed Coil Launcher

With the increase of coil stages of the high-speed electromagnetic coil launcher, the mutual coupling between the parameters and the increase of the variables to be designed make the design of the multistage high-speed electromagnetic coil launcher more difficult. In this article, an adaptive design method of electromagnetic coil launcher based on the current filament model is proposed. Its feasibility is verified by a 25-stage coil launcher design scheme calculation example. However, the coil temperature rise cannot be restricted with the traditional step-by-step trial method and adaptive design method alone. The use of traditional optimization algorithms will consume a lot of time to get a result. Therefore, this article adopts the genetic algorithm (GA) based on the predictive model to obtain constrained optimization results. With the predictive model of support vector regression (SVR), the average prediction error of the armature peak speed can be controlled below 5%. Using the established predictive model, GA is used to optimize the adjustable parameters in the adaptive design. The peak velocity of the predictive model output is selected as the fitness function, and the coil temperature rise and acceleration stability are constrained. Sixteen sets of optimization results are obtained by repeated optimization calculation. The peak speed of the armature can reach 1083 m/s, the launch efficiency of the launcher is 25.78%, and the maximum temperature rise of the coil is 76.88°C. Compared with ordinary optimization algorithms, the calculation time is greatly reduced.