To coordinate the contradiction between artillery launching performance indexes under parameter uncertainty, this paper proposes an artillery structural dynamic responses optimization method based on robust Nash game theory. First, a multi-flexible body dynamic model for a 155 mm caliber artillery is established, which coupling the interior ballistic model, recoil force model, and balance mechanism model. Secondly, the live firing experiment is carried out to verify the accuracy of the established multi-flexible model. Then the muzzle vibration and maximum chamber pressure are selected as the players in the game. Because these two indexes can represent the most critical contradictory indexes of artillery, namely the firing accuracy and power. Afterward, to reduce the computational time, the BP neural network surrogate model is constructed to replace the original multi-flexible body dynamic model. Finally, the double-loop approach is adopted to search for the robust Nash equilibrium. The inner loop optimization is used to determine the worst-case scenario caused by the parameter uncertainty. The outer loop optimization is referred to as the robust Nash equilibrium solution process. The results show that the artillery structural dynamics responses have been significantly improved.

为协调参数不确定下火炮发射性能指标之间的矛盾，提出一种基于鲁棒纳什博弈论的火炮结构动力响应优化方法。首先，建立了155mm口径火炮多柔体动力学模型，耦合了内弹道模型、后坐力模型和平衡机构模型。其次，进行实弹射击实验，验证所建立的多柔性模型的准确性。然后枪口振动和最大膛压被选为游戏中的玩家。因为这两个指标可以代表火炮最关键的矛盾指​​标，即射击精度和威力。之后，为了减少计算时间，构建BP神经网络代理模型，替代原有的多柔体动力学模型。最后，采用双循环方法寻找鲁棒的纳什均衡。内环优化用于确定由参数不确定性引起的最坏情况。外环优化被称为鲁棒纳什均衡求解过程。结果表明，火炮结构动力学响应得到了显着改善。