基于博弈对抗的鱼雷抗干扰攻击建模与优化方法研究

Research on Modeling and Optimization Method of Torpedo Anti-jamming Attack Based on Game Confrontation

针对水下攻防作战强对抗性的特点和蒙特卡罗搜索耗时长的问题,建立基于博弈对抗的鱼雷抗干扰攻击模型,提出一种改进的遗传模拟退火算法用于优化模型求解。采用仿真分析的研究方法,在双雷齐射攻击和潜艇声抗防御模型的基础上,依据零和博弈下的纳什均衡理论构建攻防对抗模型;借鉴粒子群算法自适应放缩适应度和模拟退火算法Metropolis接受准则等思想,对遗传算法中的初始种群、适应度函数、进化策略进行改进。通过典型阵位下的仿真实验,对改进遗传模拟退火算法的性能进行验证,并定量分析了射击提前角和展开航程对鱼雷攻击效果的影响。

Aiming at the strong adversarial characteristics of underwater attack and defense operations and the time-consuming problem of traditional Monte Carlo method,a model of torpedo anti-jamming attack based on game confrontation is designed and an improved genetic simulated annealing algorithm for optimal model is proposed.Through the research method of simulation analysis,on the basis of the models of two torpedo salvo attack and submarine acoustic resistance defense,the attack-defense confrontation model is constructed according to the Nash equilibrium theory under zero-sum game.The initial population,fitness function,and evolutionary strategy of GA are improved by the ideas of the adaptive deflation fitness of PSO algorithm and the metropolis acceptance criterion of SA algorithm.Through simulation experiments under typical shooting arrays,the performance of ISAGA algorithm is verified and the impact of shooting advance angle and launch range on torpedo attack effect is quantitatively analyzed.