人工肌肉
普朗特数
气动人工肌肉
流离失所(心理学)
非线性系统
控制理论(社会学)
粒子群优化
磁滞
联轴节(管道)
计算机科学
执行机构
数学
人工智能
物理
工程类
机械
算法
机械工程
传热
量子力学
心理治疗师
控制(管理)
心理学
作者
Yeming Zhang,Chuangchuang Liu,Maolin Cai,Yan Shi,Sanpeng Gong,Hui Zhang,Shuaijie Zhang
标识
DOI:10.1177/09544062241300810
摘要
As an advanced actuator that simulates the characteristics of biological muscle, the internal structure and working principle of pneumatic artificial muscle imply significant complexity and highly nonlinear characteristics, which are particularly prominent in the theoretical modelling process, and pose many challenging problems for researchers. The classical Prandtl-Ishlinskii model fails to provide ideal mathematical descriptions and accurate models when dealing with the nonlinear characteristics of pneumatic artificial muscle, resulting in the construction of models for such nonlinear systems that still require further research. Based on the above, this paper proposes a generalised Prandtl-Ishlinskii model based on the improved force-displacement hysteresis characteristics of pneumatic artificial muscle, which employs an improved particle swarm optimisation to identify the unknown parameters of the model, and applies the identified parameters to construct the force-displacement hysteresis curves of pneumatic artificial muscle. Comparing the model fitting results with the experimental data, it is found that the maximum error of the improved generalised Prandtl-Ishlinskii model is reduced by more than 50% and the average error is reduced by more than 75% compared with the classical Prandtl-Ishlinskii model when the internal pressure of the pneumatic artificial muscle is 0.35, 0.4 and 0.45 MPa. The improved generalised Prandtl-Ishlinskii model can better characterise the asymmetric hysteresis property, which improves the model accuracy of the force-displacement coupling of pneumatic artificial muscle, and provides a certain theoretical basis for the later realisation of the precise control of the force-displacement coupling of pneumatic artificial muscle.
科研通智能强力驱动
Strongly Powered by AbleSci AI