摘要为了讨论神经网络运用于弹道预测的可行性,并构建实用的弹道预测工具,建立了基于神经网络理论的弹道预测模型。利用二自由度质点弹道模型,选取BP网络和Elman网络进行神经网络弹道预测仿真。基于误差反向传播理论,比较了带动量项算法与自适应学习率算法这两种网络权值训练速度。研究了不同算法中主要参数对预测误差的影响。对两种网络不同隐层节点数的学习误差和预测误差进行对比分析。数值仿真计算结果表明,神经网络具有较高的预测精度,36.7Km射程仅有不足100m的射程误差,12.3Km射高仅有不足70m的高度误差,预测结果满足要求,利用神经网络进行弹道预测是合理可行的。65234
毕业论文关键字 弹道预测 神经网络 误差反传 数值仿真
毕业设计说明书(论文)外文摘要
Title Research on the trajectory prediction method based on neural network algorithm
Abstract
In order to discuss the feasibility of neural network uses for trajectory prediction, and create a practical prediction trajectory tool, we built a trajectory predictive model based on neural network. It is using tow dof particle trajectory model to do neural network trajectory prediction simulation by selected BP network and Elman network. Based error back propagation theory, it compares two types of weight training velocities from momentum back propagation and adaptive learning rate back propagation. Studied the impact of the prediction error from the main parameters in different algorithms. By comparing the network structure, comparative analysis the learning error and prediction error of the two networks in different amount of hidden nodes. Numerical results show that it is less than 100m range error of 36.7Km and less than 70m height error of 12.3Km shot high. The neural network has high prediction accuracy,and its trajectory prediction results meet the requirements. The ballistic prediction using neural networks is reasonably practicable.
Keywords ballistic prediction neural network error back propagation numerical simulation
目 录
1 绪论 1
1.1 弹道预测的背景及意义 1
1.2 神经网络简介 2
1.3 基于神经网络算法的弹道预测方法 3
1.5 本文的主要研究内容 5
2 弹丸质心运动方程 6
2.1 仿真条件 6
2.2 弹丸质心运动方程组 6
2.3 弹道数值积分结果 7
3 神经网络基本概念 10
3.1 生物神经网络 10
3.2 人工神经网络 11
4 基于BP网络的弹道预测 17
4.1 BP网络 17
4.2 标准BP算法 19
4.3 BP网络的优势与局限 22
4.4 标准BP算法改进 23
4.5 BP网络弹道预测模型 24
5 基于Elman网络的弹道预测 44
5.1 Elman网络 44
5.2 Elman网络学习算法 45
5.3 Elman网络弹道预测模型