Hysteresis Modeling Method for Pneumatic Muscle Based on Multi-Branch BP Neural Network
XIE Sheng-long1, ZHANG Wen-xin2, ZHANG Wei-min2, REN Guo-yin3, LU Yu-jun4, LU Qing5
1. School of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou, Zhejiang 310018, China
2. Zhejiang Xizi Heavy Machinery Co. Ltd., Jiaxing, Zhejiang 314423, China
3. National Institute of Metrology, Beijing 100029, China
4. Faculty of Mechanical Engineering and Automation, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China
5. Baowu Equipment Intelligent Technology Co. Ltd., Maanshan, Anhui 243000, China
Abstract：A novel modeling method based on multi-branch BP neural network is proposed to describe the displacement/pressure hysteresis of pneumatic muscle. Firstly, the displacement/pressure hysteresis characteristic test system was built to obtain the displacement/pressure hysteresis loops of pneumatic muscle. Then, the classical BP neural network, multi-branch BP neural network and Prandtl-Ishlinskii model are used to fit the hysteresis loop of pneumatic muscle, respectively. Finally, the comparative study found that the multi-branch BP neural network can effectively avoid the over-fitting phenomenon in the fitting process of classical BP neural network, and the modeling capacity is obviously better than the traditional Prandtl-Ishlinskii model. Compared with Prandtl-Ishlinskii model, the mean average, mean square and maximum errors of multi-branch BP neural network are reduced by 87.45%, 86.68% and 74.73%.
［1］谢建蔚, 陶国良, 周洪. 高速开关阀驱动的气动肌肉关节的滑模变结构跟踪控制［J］. 中国机械工程, 2007, 18 (5): 540-544.
Xie J W, Tao G L, Zhou H. Sliding mode tracking control of pneumatic muscle joint actuated by high-speed on-off solenoid valve［J］. China Mechanical Engineer-ing, 2007, 18 (5): 540-544.
［2］ Xie S L, Mei J P, Liu H T, et al. Hysteresis modeling and trajectory tracking control of the pneumatic muscle actuator using modified Prandtl-Ishlinskii Model［J］. Mechanism and Machine Theory, 2018, 120: 213-224.
［3］ 姜宁, 姚恩涛, 邹华章, 等. 基于智能气缸的功能可配置机械手的设计［J］. 计量学报, 2018, 39 (2): 173-177.
Jiang N, Yao E T, Zhou H Z, et al. A excogitation of function configurable manipulator based on the intelligent cylinder［J］. Acta Metrologica Sinica, 2018, 39 (2): 173-177.
［4］ Hassani V, Tjahjowidodo T, Do T N. A survey on hysteresis modeling, identification and control［J］. Mechanical Systems and Signal Processing, 2014, 49 (1): 209-233.
［5］ Lin C J, Lin C R, Yu S K, et al. Hysteresis modeling and tracking control for a dual pneumatic artificial muscle system using Prandtl-Ishlinskii model［J］. Mechatronics, 2015, 28: 35-45.
［6］ Zhong J, Fan J, Zhu Y, et al. One nonlinear PID control to improve the control performance of a manipulator actuated by a pneumatic muscle actuator［J］. Advances in Mechanical Engineering, 2014, 501 (8): 802-824.
［7］ Zhao J, Zhong J Z, Fan J. Position control of a pneu-matic muscle actuator using RBF neural network tuned PID controller［J］. Mathematical Problems in Engineering, 2015, (5):1-16.
［8］ Aschemann H, Schindele D. Comparison of model-based approaches to the compensation of hysteresis in the force characteristic of pneumatic muscles［J］. IEEE Transactions on Industrial Electronics, 2014, 61 (7): 3620-3629.
［9］ 谢胜龙, 梅江平, 刘海涛. McKibben型气动人工肌肉研究进展与趋势［J］. 计算机集成制造系统, 2018, 24 (5): 1065-1081.
Xie S L, Mei J P, Liu H T. Achievements and trends of research on McKibben pneumatic artificial muscles［J］. Computer Integrated Manufacturing Systems, 2018, 24 (5): 1065-1081.
［10］ Kosaki T, Sano M. Control of a parallel manipulator driven by pneumatic muscle actuators based on a hysteresis model［J］. Journal of Environment and Engineering, 2011, 6 (2): 316-327.
［11］ Kosaki T, Minesaki A, Sano M. Adaptive hyster-esis compensation with a dynamic hysteresis model for control of a pneumatic muscle actuator ［J］. Journal of Environment and Engineering, 2012, 7 (1): 53-65.
［12］ Xie S L, Liu H T, Wang Y. A method for the length-pressure hysteresis modeling of pneumatic artificial muscles［J］. Science China Technological Sciences, 2020, 63(5): 829-837.
［13］ Xie S L, Liu H T, Mei J P, et al. Modeling and compensation of asymmetric hysteresis for pneumatic artificial muscles with a modified generalized Prandtl-Ishlinskii model［J］. Mechatronics, 2018, 52: 49-57.
［14］ Liu Y X, Zang X Z, Lin Z K, et al. Modelling length/pressure hysteresis of a pneumatic artificial muscle using a modified Prandtl-Ishlinskii model ［J］. Strojniki ves-tnikJournal of Mechanical Engineering, 2017, 63 (1): 56-64.
［15］ Yang H, Chen Y, Sun Y, et al. A novel Kriging-Median inverse compensator for modeling and compens-ating asymmetric hysteresis of pneumatic artificial muscle［J］. Smart Materials and Structures, 2018, 27 (11). DOI: 10. 1088/1361-665X/aad758 (Published online).
［16］ Xu J H, Xiao M B, Ding Y. Modeling and compens-ation of hysteresis for pneumatic artificial muscles based on Gaussian Mixture models ［J］. Sci China Tech Sci, 2019, 62: 1094-1102.
［17］ 崔霞, 施光林, 沈伟. 基于分组数据处理神经网络气动人工肌肉迟滞特性［J］. 上海交通大学学报, 2012, 46 (6): 931-935.
Cui X, Shi G L, Shen W. Study on hysteresis pneumatic artificial muscle based on group method of data handling neural network ［J］. Journal of Shanghai Jiaotong University, 2012, 46 (6): 931-935.
［18］ 谢胜龙, 刘海涛, 梅江平, 等. 气动人工肌肉迟滞-蠕变特性研究现状与进展［J］. 系统仿真学报, 2018, 30 (3): 809-823.
Xie S L, Liu H T, Mei J P, et al. Achievements and developments of hysteresis and creep of pneumatic artif-icial muscles ［J］. Journal of System Simulation, 2018, 30 (3): 809-823.
［19］ 范伟, 林瑜阳, 李钟慎. 基于BP神经网络的压电陶瓷蠕变预测［J］. 计量学报, 2017, 38 (4): 429-434.
Fan W, Lin Y Y, Li Z S. Prediction of the creep of piezoelectric ceramic based on BP neural network optimized by genetic algorithm ［J］. Acta Metrologica Sinica, 2017, 38 (4): 429-434.
［20］ 张淑清, 任爽, 陈荣飞, 等. 基于大数据简约及PCA 改进ＲBF 网络的短期电力负荷预测［J］. 计量学报, 2018, 39 (3): 392-396.
Zhang S Q, Ren S, Chen R F, et al. Short-term power load forecast based on big data reduction and PCA-improved RBF network ［J］. Acta Metrologica Sinica, 2018, 39 (3): 392-396.
［21］ 王有贵, 吴双双, 陈红江. 称重传感器蠕变误差的神经网络补偿方法［J］. 计量学报, 2018, 39 (4): 510-514.
Wang Y G, Wu S S, Chen H J. Compensation method for creep error of load cell based on neural networks［J］. Acta Metrologica Sinica, 2018, 39 (4): 510-514.
［22］ Mei J P, Xie S L, Liu H T, et al. Hysteresis modeling and compensation of pneumatic artificial muscles using the generalized Prandtl-Ishlinskii model［J］. Strojniki vestnik-Journal of Mechanical Engineering, 2017, 63 (11): 657-665.
2021, Vol. 42 
