Study Of The Multi-objective Optimization Algorithm For Pintle-controlled Solid Motor

The variable thrust solid rocket motor has become one of the important research directions due to its advantages of adjustable thrust characteristics and high energy efficiency.The design of the variable thrust solid rocket motor usually involves multiple objective functions and constraints.To improve its overall performance,the multiobjective optimization method must be adopted in the optimization design of the variable thrust solid rocket motor.However,the time-consuming refined simulation model restricts the optimization efficiency of the multi-objective optimization algorithm,which will bring unacceptable computational costs.In this paper,the researches of disciplinary modeling,multi-objective optimization algorithm,and case verifications are carried out to build a systematic multi-objective optimization framework,so as to realize the efficient optimization design of the variable thrust solid rocket motor.The simulation model of the variable thrust solid rocket motor is established to serve as the refined simulation model in the multi-objective optimization design.In view of the difference between variable thrust and traditional solid rocket motors,a calculation model is proposed based on the dichotomous search method and a zero-dimensional internal ballistic model of variable thrust solid rocket motor is established.The high-fidelity simulation model based on Fluent is established to calculate the accurate performance parameters of variable thrust solid rocket motor.Besides,a thrust matching design method for variable thrust solid rocket motor is proposed to quickly realize the specified thrust scheme by adjusting the position of the throat bolt.A multi-objective optimization algorithm based on the inaccurate search and the tabu vectors is proposed.To address the problem of trapping into local optima,the inaccurate search method is introduced into the auxiliary optimization during the multi-objective optimization process.With the inaccurate search,the auxiliary optimization is terminated when the Pareto front improvement reaches a specified threshold at the early stage of optimization when the accuracy of the surrogate model is poor,which can help the optimization avoid falling into local optima.As the optimization iteration proceeds,the search intensity on the surrogate model is gradually increased to ensure that the search gradually approaches the global optimum.To avoid over-exploitation,the tabu vector strategy is applied to effectively avoid the regions with poor exploitation prospects,so as to avoid the waste of computational resources and the pathological phenomena that may occur during the training of the surrogate model.Besides,the maximin distance criterion is used to enhance the global exploration capability of the algorithm and further improvement the optimization performance.Based on the above researches,a multi-objective optimization design framework of the variable thrust solid rocket motor is proposed.Based on the established high-fidelity simulation model of the variable thrust solid rocket motor and the requirements of actual design,the proposed multi-objective optimization algorithm is applied to perform the optimization design of the variable thrust solid rocket motor.The overall optimization design of variable thrust solid rocket motor,the optimization design of the pintle nozzle,and the integrated optimization design of internal and external ballistics of variable thrust solid motor are realized.The design results show that the proposed algorithm can efficiently deal with the real-world variable thrust solid rocket motor engineering problems.The range of the optimized variable thrust solid rocket motor performance is nearly doubled and its overall performance is greatly improved.