摘要
针对目前技术难以实现高射速状态下对高炮载车姿态进行实时测试的问题,提出一种基于BP神经网络的姿态预测方法。建立某自行高炮刚柔耦合动力学模型,通过试验数据修正模型以保证仿真结果的可靠性。针对射击条件均匀设计工况并仿真,提取姿态信息,经过数据预处理建立样本库。在matlab中设计BP神经网络结构参数,通过训练完成载车姿态预测模型的构建。结果表明,在不同的工况下,代理模型均可快速实现姿态预测,且满足精度要求。
To overcome the difficulty in testing the attitude of the anti-aircraft gun vehicle in real time at high firing rate with current technology,an attitude prediction method based on BP neural network was proposed.The rigid-flexible coupling dynamic model of an self-propelled anti-aircraft gun was established,and modified by test data to ensure the reliability of the simulation results.According to the uniform design and simulation of firing conditions,the attitude information was extracted,and the sample database was established through data preprocessing.BP neural network structure parameters were designed in matlab,and the vehicle attitude prediction model was built through training.The results show that the agent model can realize the attitude prediction quickly and meet the precision requirement under different working conditions.
作者
郭峰
谢立中
周成
于存贵
GUO Feng;XIE Lizhong;ZHOU Cheng;YU Cungui(Schol of Mechanical Engineering,Nanjing University of Science&Technology,Nanjing 210094,China;Bazooka Institute,Hubei Jiangshan Heavy Industry Limited Liability Company,Xiangyang 441057,China)
出处
《机械制造与自动化》
2021年第5期125-128,共4页
Machine Building & Automation
关键词
高射速
自行高炮
载车姿态
仿真
神经网络
刚柔耦合
rapid fire
self-propelled anti-aircraft gun
vehicle attitude
simulation
neural network
rigid-flexible coupling
