针对雷达阵地毁伤评估问题,提出了一种基于GA-动态BP神经网络的评估方法。该方法首先通过遗传算法(GA)寻求神经网络最优权重值,充分发挥了动态神经网络的泛化和非线性映射能力以及GA的全局寻优能力,提高了结果的精确度;再引入牛顿迭代法优化网络训练算法,克服了神经网络在训练过程中易陷入局部极小值及网络学习后期收敛速度慢的缺点。以某一时刻防空作战为想定,仿真实现了雷达阵地的毁伤评估,与现有算法相比,该算法在收敛速度、可靠性和准确性上都有明显提高。 Aiming at the damage assessment of radar positions, a method based on GA-dynamic BP neural network is proposed. Firstly, GA (Genetic Algorithm) is used to find the optimal weighted value of neural network, which gives full play to the generalization and nonlinear mapping ability of dynamic neural network and the global search ability of GA, and improves the accuracy of the result. Second, Newton iteration The method optimizes the network training algorithm, overcomes the shortcoming that the neural network tends to fall into the local minimum during training and the convergence speed is slow at the late stage of network learning. Considering the air defense operations at a certain moment, the damage assessment of the radar position is simulated, and compared with the existing algorithms, the algorithm has obviously improved the convergence speed, reliability and accuracy.