China Mechanical Engineering

Previous Articles     Next Articles

Hysteresis Modeling Method of Pneumatic Muscles Based on KP Model

XIE Shenglong1;LI Tiefeng2;WANG Binrui1;CHEN Dijian1   

  1. 1.College of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou,310018
    2.Department of Engineering Mechanics, Zhejiang University, Hangzhou,310027
  • Online:2020-05-25 Published:2020-06-28

基于KP模型的气动肌肉迟滞建模方法

谢胜龙1;李铁风2;王斌锐1;陈迪剑1   

  1. 1. 中国计量大学机电工程学院,杭州,310018
    2. 浙江大学工程力学系,杭州,310027
  • 基金资助:
    国家自然科学基金资助项目(51575503,U1613202);
    浙江省自然科学基金资助项目(LQ20E050017);
    之江国际青年人才基金资助项目(ZJ2019JS006)

Abstract: In order to identify the hysteretic nonlinearity of motion processes on PM, KP model and polynomial model were used to model the displacement/pressure hysteresis of PM based on the displacement/pressure hysteresis characteristics experiments of PM. The unknown parameters in the two models were identified by recursive least square (RLS) method, and the effects of the number of KP operators and the order of polynomial on the accuracy of the two models were studied. The parameter identification results show that the KP model is better than that of the polynomial model in modeling accuracy, the modeling accuracy of KP model increases significantly as the number of operators increases and modeling accuracy of polynomial model increases with the increase of polynomial order.

Key words: pneumatic muscle (PM), hysteresis modeling, Krasnosel'skii-Pokrovskii(KP) model, polynomial model, parameter identification

摘要: 为了辨识气动肌肉运动过程中的迟滞非线性特性,在气动肌肉位移/气压迟滞特性实验的基础上,分别采用KP模型和多项式模型对其位移/气压迟滞展开了建模研究,并采用递推最小二乘法分别对两种模型中的未知参数进行了辨识,研究不同算子个数和多项式次数对模型建模精度的影响。参数辨识结果发现,KP模型拟合精度明显优于多项式模型,随着算子个数的增加,KP模型的精度显著提高,多项式模型的建模精度也随着多项式次数的增加而提高。

关键词: 气动肌肉, 迟滞建模, Krasnosel'skii-Pokrovskii(KP)模型, 多项式模型, 参数辨识

CLC Number: