基于BP神经网络的钢丝绳精密传动误差预测

发布时间：2018-01-11 22:06

本文关键词：基于BP神经网络的钢丝绳精密传动误差预测　出处：《机械传动》2017年09期 　论文类型：期刊论文

【摘要】：针对不同绳组数的钢丝绳精密传动方式,研究了基于神经网络的传动误差预测方法。以预紧力、负载、转角为输入,传动误差为输出,建立了双隐含层BP神经网络模型。在不同绳组数的传动条件下,测试了不同预紧力、负载和转角下的传动误差。将传动误差测试结果分为训练组和预测组,对网络进行训练和检验。测试结果显示,从动轮在0°~180°范围内转动时,传动误差随转角的增大而增大,最大值为5.6 mrad,随绳组数的增加而降低。在该转角范围内,模型预测误差小于0.294 mrad。表明采用双隐含层BP神经网络模型对传动误差进行预测是有效的。
[Abstract]:Aiming at the precision transmission mode of steel wire rope with different number of rope groups, the transmission error prediction method based on neural network is studied. The preload, load and angle of rotation are taken as input and transmission error as output. A BP neural network model with double hidden layers is established. The transmission errors under different preload, load and rotation angles are tested under different transmission conditions of different rope groups. The test results of transmission errors are divided into two groups: training group and prediction group. The network is trained and tested. The test results show that the transmission error increases with the increase of rotation angle when the driving wheel rotates in the range of 0 掳to 180 掳, with a maximum value of 5.6 mrad. The prediction error of the model is less than 0.294 mrad. it is shown that the double hidden layer BP neural network model is effective to predict the transmission error.
【作者单位】：西安应用光学研究所;陆军航空兵军事代表局驻西安地区军代室;
【基金】：“十二五”兵器预研支撑基金项目(62201070139) 中国兵器工业集团高等院校协同创新合作专项项目(KH201504,KH2016006)
【分类号】：TH132.3
【正文快照】： 0引言在高分辨率光电系统、小型机器人等领域,精密传动机构是不可缺少的组成部分[1-3]。随着系统性能的不断提高,控制系统对传动机构的质量和精度要求也越来越高。传统的齿轮传动空回较大,精度提高越来越困难。钢丝绳精密传动是一种新型的柔性传动技术,具有高精度、低空回、传

【相似文献】