|本期目录/Table of Contents|

[1]范伟,林瑜阳,李钟慎.遗传算法优化的 BP 神经网络压电陶瓷蠕变预测[J].电机与控制学报,2018,22(07):91-96.[doi:10.15938/j.emc.2018.07.012]
 FAN Wei,LIN Yu-yang,LI Zhong-shen.Prediction model of the creep of piezoceramic based onBP neural network optimized by genetic algorithm[J].,2018,22(07):91-96.[doi:10.15938/j.emc.2018.07.012]
点击复制

遗传算法优化的 BP 神经网络压电陶瓷蠕变预测(PDF)
分享到:

《电机与控制学报》[ISSN:1007-449X/CN:23-1 408/TM]

卷:
22
期数:
2018年07
页码:
91-96
栏目:
出版日期:
2018-10-31

文章信息/Info

Title:
Prediction model of the creep of piezoceramic based onBP neural network optimized by genetic algorithm
作者:
 范伟 林瑜阳 李钟慎
 ( 华侨大学 机电及自动化学院,福建 厦门 361021)
Author(s):
  FAN Wei LIN Yu-yang LI Zhong-shen
  ( School of Mechanical Engineering and Automation,Huaqiao University,Xiamen 361021,China)
关键词:
压电陶瓷 蠕变 神经网络 遗传算法 预测
Keywords:
piezoelectric ceramics creep neural network genetic algorithm prediction
分类号:
TN 384
DOI:
10.15938/j.emc.2018.07.012
文献标志码:
A
摘要:
针对压电陶瓷驱动器的蠕变误差随时间呈现非线性变化,会严重影响其定位精度的问题,提出遗传算法优化 BP 神经网络的压电陶瓷蠕变预测算法。采用遗传算法优化了 BP 神经网络的权值和阈值,构建了基于遗传算法的 BP 神经网络( GA-BP 算法) 的蠕变预测模型。用 GA-BP 算法对压电陶瓷蠕变进行了预测仿真,并将结果与实测数据进行了对比。结果表明,获得的蠕变预测结果与实验数据的最大绝对误差均不超过 0. 2 μm,最大蠕变误差均小于 1. 5% ,最大均方误差仅为0. 004 6,因此,GA-BP 预测模型可作为预测压电陶瓷蠕变误差的一种有效手段。
Abstract:
Aiming at the creep errors of the piezoelectric ceramics showing nonlinear change with thetime,which affects the positioning accuracy of the piezoelectric ceramics badly,
a creep prediction ap-proach of back propagation neural network optimized by genetic algorithm is proposed for the piezoelectricceramics. The weight and threshold of the BP neural network were optimized by genetic algorithm,and acreep prediction model was constructed based on BP neural network. The creep of piezoelectric ceramicswas predicted by the model of GA-BP neural network. The results of BP neural network were comparedwith the measured data. The results show that,using this prediction mode,
the maximal absolute error isbelow 0. 2 μm,and the maximal creep error is below 1. 5% . The maximal mean square error is 0. 004 6.Thus,the GA-BP neural network prediction model can be applied into the creep prediction of the piezoe-lectric ceramics.

参考文献/References:

相似文献/References:

[1]王希花,郭书祥,叶秀芬,等.压电陶瓷迟滞特性的建模及复合控制[J].电机与控制学报,2009,(05):766.
 WANG Xi-hua,GUO Shu-xiang,YE Xiu-fen,et al.Modeling and feed-forward control based on piezoelectri cceramic hysteretic[J].,2009,(07):766.

备注/Memo

备注/Memo:
收稿日期: 2016 - 11 - 25
基金项目: 国家自然科学基金( 51475176) ; 2017 年福建省自然科学基金计划项目( 2017J01086) ; 中央高校基本科研业务费专项资金( 11J0234)
作者简介: 范 伟( 1980—) ,男,博士,讲师,研究方向为精密测试技术及仪器;
林瑜阳( 1991—) ,男,硕士研究生,研究方向为压电陶瓷特性建模及控制;
李钟慎( 1971—) ,男,教授,研究方向为控制理论与控制工程。
更新日期/Last Update: 2018-09-28