Trajectory Optimization Desigh Of Launching Rocket Based On Multi-objective Intelligent Algorithm

Ballistic optimization is an important part of the overall optimization design of the launch vehicle.Ballistic optimization design plays an extremely important role in the life cycle of the launch vehicle.In the process of ballistic optimization design,there are often more than one objective function selection.Under the premise of not sacrificing flight performance,various factors must be considered comprehensively to achieve the optimal cost-effective ratio under economical perspective.To this end,this paper takes the launching ballistic optimization of the launch vehicle as the research object,establishes the mathematical model of the rocket motion model and multi-objective optimization problem,and uses the discretization method to transform the multi-objective dynamic optimization problem into multi-objective parameter optimization problem,and then utilizes The multi-objective intelligent optimization algorithm solves the parameter optimization problem and analyzes and compares the results of different multi-objective intelligent optimization algorithms.The main research work of this paper is as follows:First,dynamics and kinematics are modeled according to the requirements of the ballistic optimization design problem.Using aerodynamic knowledge,analyze the form of the launch vehicle's force,establish the launch vehicle's centroid motion model and the atmospheric model,and give the launch vehicle's centroid dynamics equation,kinematics equation and necessary supplementary equations.Secondly,according to the research content of the subject,the type of multi-objective optimization problem is judged.Based on this,the mathematical model of the multi-objective optimization problem is established,the optimization variables and the objective function are determined,and the constraints are listed.The whole trajectory is discretized in time,and the fourth-order Runge-Kutta method can be used to calculate the flight parameters of each discrete point,and the whole discrete trajectory is obtained.The dynamic constraints and process constraints can be transformed into equality and inequality constraints at each discrete point,respectively,and the terminal constraint is the last point on the discrete trajectory.Since then,multi-objective dynamic optimization problems have been transformed into multi-objective parameter optimization problems.Then,according to the characteristics of each multi-objective intelligent algorithm,the multi-objective genetic algorithm and multi-objective particle swarm optimization algorithm which are more mature and proved feasible in many fields are selected for research.The classical algorithms are selected from the two types of multi-objective intelligent optimization algorithms,and the algorithm program is written according to the principle.The multi-objective parameter optimization problem is solved by MATLAB simulation,and the Pareto frontier can be obtained.Since the operations such as the initial generation and mutation of the intelligent algorithm are random,the Pareto fronts calculated each time are different.Run the program multiple times,select an optimal solution from each calculated Pareto front,observe and analyze the convergence of these optimal solutions,and verify the feasibility and correctness of the used algorithm.Finally,the multi-objective optimization algorithms are compared,and the solutions obtained by the two algorithms are observed.The calculation time,the convergence of the calculation results and the convergence speed are analyzed and evaluated respectively.The ability of the two algorithms to solve the problem is compared.