Dynamic Strategy Optimization Of Multi-Objectives The Laser Anti-Missile System Based On The Improved Intelligent Algorithms

This paper presents the solutions of the Strategy Optimization of the Dynamic Multi-objectives Laser Anti-missiles System (DMLAMS). DMLAMS has been applied in some western countries, like USA, Israel and etc. So it is meaningful to use the heuristic methods in the dynamic optimization problem(DOP).This paper constructs the dynamic strategy model of the DMLAMS, it separates an arrange into several parts, every part indicates the time of laser weapon shoot down one objective, at the same time the locations and the velocities massage of objectives are updated every time, then the separated parts are added into the model output.Genetic Algorithm can seek the best solution of the mathematic model, and can be used in many areas. First of all, the normal Genetic Algorithm is introduced, a group of sub-optimal solutions and operated individuals are presented, in order to improve the fitness of the individuals and search the best solution, the crossover and mutation rates are changed. In the first method the crossover and mutation rates are increased or decreased at the same time with the optimal times; the mutation rate is high at the beginning of the algorithm and it decreases with the generations increase, the crossover rate is opposite in the mutation first method. Both methods improve the fitness values of the individuals and increase the optimal solution rate.The paper verified that the Simulated Annealing Algorithm(SA) can search the optimal solution of the DMLAMS dynamic strategy. The annealing rate is modified, and the calculation time is reduced. The solution of the Ant Colony Optimization is carried out.The Hybrid Genetic Algorithm(HGA) is another direction of the Genetic Algorithm. This paper illustrates the Simulated Annealing Genetic Algorithm(SAGA) and the ant Colony Optimization Genetic Algorithm(ACOGA). They improve the fitness of the populations and increase the optimal solution rate. They improve the rate of optimal solution 10.24%, 10.24% respectively than the normal algorithm.