摘要
针对压电型三维力传感器由于结构制造而产生的维间耦合问题,设计了一种基于改进的开普勒优化算法优化极限学习机(EKOA-ELM)的解耦算法。首先,阐述了压电型三维力传感器的耦合特性;然后,对压电型三维力传感器构建标定实验进行标定,得出三维力传感器的测力数据;最后,建立极限学习机非线性解耦模型,并利用混沌Cat映射与基于余弦规律变化的收敛因子对KOA(Kepler optimization algorithm)算法进行优化。实验结果表明:平均解耦Ⅰ类误差控制在0.38%以内,平均解耦Ⅱ类误差控制在0.32%以内,解耦时间为0.071 s,该算法有较好解耦精度的同时,保持较好的解耦效率。
To address the issue of inter-dimensional coupling in piezoelectric three-dimensional force sensors due to structural manufacturing,a decoupling algorithm based on an enhanced kepler optimization algorithm optimized extreme learning machine(EKOA-ELM)has been developed.This paper first explains the coupling characteristics inherent to piezoelectric three-dimensional force sensors.It then proceeds to calibrate these sensors through a designated calibration experiment,yielding force measurement data for further analysis.Subsequently,a nonlinear decoupling model for the extreme learning machine is established.To refine the effectiveness of this model,the Kepler optimization algorithm(KOA)is optimized using a chaotic cat map and a convergence factor that adapts based on the cosine law.Experimental results highlight that the average decoupling errors are maintained within 0.38%for TypeⅠerrors and 0.32%for TypeⅡerrors,with a rapid decoupling time of only 0.071 s.This demonstrates that the algorithm not only effectively minimizes the issue of inter-dimensional coupling in piezoelectric three-dimensional force sensors but also showcases superior efficiency and accuracy in nonlinear decoupling tasks.
作者
何祥
刘勇
刘诚
陈思涵
HE Xiang;LIU Yong;LIU Cheng;CHEN Sihan(School of Automation and Information Engineering,Sichuan University of Science&Engineering,Yibin 644000,China;Sichuan Provincial Key Laboratory of Artificial Intelligence,Sichuan University of Science&Engineering,Yibin 644000,China)
出处
《组合机床与自动化加工技术》
北大核心
2025年第4期32-36,42,共6页
Modular Machine Tool & Automatic Manufacturing Technique
基金
四川省科技计划项目(22ZFSHFZ0001)。
关键词
压电型三维力传感器
维间耦合
极限学习机
开普勒优化算法
混沌cat映射
收敛因子
piezoelectric three-dimensional force sensor
interdimensional coupling
extreme learning machine
Kepler optimization algorithm
chaotic cat mapping
convergence factor
