期刊文献+

基于LSTM的稳定平台振动分析

Vibration Analysis of Stable Platform Based on LSTM
下载PDF
导出
摘要 稳定跟踪系统是雷达伺服系统研究领域的热点。稳定跟踪要求跟踪平台在受到行进间各种随机振动影响的条件下,依然保持良好的跟踪精度。对于平台稳定的条件下,跟踪精度的保持已经有很多成熟的方法,但是对于在振动干扰条件下,如何保持平台稳定也成为了一个研究难点。由于平台受到的振动是随机的,频率和幅度都会不定性变化,规律性很差,很难进行传统建模分析。再加上传感器通信延时的存在,进一步导致平台保持稳定的误差增大,最终影响到跟踪精度。本文采用长短时记忆网络算法(Long Short Term Memory Network,LSTM)对平台振动数据进行深度学习训练,实现依据振动前五个时刻的数据预测后一个时刻数据的功能。经过交叉验证,预测误差可以达到0.3mil以内。 The stability tracking system is a hot topic in the research field of radar servo system.Stable tracking requires the tracking platform to maintain good tracking accuracy under the condition of being affected by various random vibrations during marching.There are many mature methods to maintain the tracking accuracy under the condition of platform stability,but how to maintain the stability of the platform under the condition of vibration interference was still difficult.Because the vibration is random,the frequency and amplitude will change indefinitely with relatively low regularity,so it is barely impossible to carry out traditional modeling and simulation.Coupled with the existence of sensor communication delay,the error of keeping the platform stable increased,and finally affected the tracking accuracy.In this paper,the long short term memory network(LSTM)algorithm was adopted to conduct deep learning and train the platform vibration data,so as to realize the function of predicting the data of the next time according to the data of the first five times of vibration.After cross validation,the prediction error can reach within 0.3mil.
作者 张博 何海龙 顾振海 卢珊珊 ZHANG Bo;HE Hailong;GU Zhenhai;LU Shanshan(Xi'an Institute of Electronic Engineering Research Institute,Xi'an 710100)
出处 《火控雷达技术》 2022年第4期97-102,共6页 Fire Control Radar Technology
关键词 平台稳定控制 人工智能 深度学习 LSTM platform stability control artificial intelligence deep learning LSTM
  • 相关文献

参考文献2

二级参考文献19

共引文献62

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部