基于深度学习的非线性广义预测控制

doi:10.13976/j.cnki.xk.2023.2064

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (1148 KB) HTML ( )
输出: BibTeX | EndNote (RIS)

摘要针对一类离散时间单输入-单输出(single-input single-output， SISO)非线性动态系统，将非线性切换控制与基于深度学习的未建模动态估计方法相结合，提出了一种新的非线性广义预测控制方法。该方法针对未建模动态的未知增量，通过使用基于深度学习技术的长短记忆神经网络(long short-term memory， LSTM)进行预估，设计了一种带有未建模动态增量估计的非线性广义预测控制器，增强控制性能。对所提的控制算法进行了稳定性和收敛性分析，最后通过数值仿真实验验证了所提方法的有效性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	李旭生
	牛宏
	陶金梅

关键词 ：深度学习, 长短记忆神经网络(LSTM), 未建模动态, 非线性系统, 广义预测控制

Abstract：In this study， we propose a new nonlinear generalized predictive control method for a class of discrete-time single-input single-output nonlinear dynamic systems. For this， we combine nonlinear switching control and unmodeled dynamics estimation methods based on deep learning technology. In the new algorithm， we use long short-term memory neural networks to estimate the unknown increment of unmodeled dynamics based on deep learning technology. We then enhance control performance by designing a nonlinear generalized predictive controller with unmodeled dynamic increment estimation. We also analyze the stability and convergence of the proposed control algorithm. The results of our simulation experiments verified the effectiveness of the proposed method.

Key words： deep learning long short-term memory (LSTM) unmodeled dunamics nonlinear system generalized predictive control

收稿日期: 2022-02-15 出版日期: 2023-05-06

ZTFLH:

TP273

基金资助:国家自然科学基金(61773107)；辽宁省自然基金指导计划(2019-ZD-0054)

通讯作者: 牛宏,niuhong@lnpu.edu.cn E-mail: niuhong@lnpu.edu.cn

作者简介: 李旭生(1993-)，男，硕士生。研究领域为广义预测控制和切换控制。
牛宏(1979-)，女，博士，副教授。研究领域为非线性系统的自适应控制和变结构控制。
陶金梅(1995-)，女，硕士。研究领域为非线性系统的自适应控制与系统辨识。

引用本文:

李旭生, 牛宏, 陶金梅. 基于深度学习的非线性广义预测控制[J]. 信息与控制, 2023, 52(2): 202-210.
LI Xusheng, NIU Hong, TAO Jinmei. Nonlinear Generalized Predictive Control Based on Deep Learning. Information and control, 2023, 52(2): 202-210.

链接本文:

https://xk.sia.cn/CN/10.13976/j.cnki.xk.2023.2064 或 https://xk.sia.cn/CN/Y2023/V52/I2/202

[1] CHEN L J, NARENDRA K S. Nonlinear adaptive control using neural networks and multiple models[J]. Automatica, 2001, 37(8):1245-1255.
[2] ZHANG Y J, CHAI T Y, HONG W, et al. An adaptive generalized predictive control method for nonlinear systems based on ANFIS and multiple models[J]. IEEE Transactions on Fuzzy Systems, 2011, 18(6):1070-1082.
[3] 柴天佑, 张亚军. 基于未建模动态补偿的非线性自适应切换控制方法[J]. 自动化学报, 2011, 37(7):773-786. CHAI T Y, ZHANG Y J. Nonlinear adaptive switching control method based on unmodeled dynamics compensation[J]. Acta Automatica Sinica, 2011, 37(7):773-786.
[4] CHAI T Y, ZHANG Y J, WANG H, et al. Data-based virtual unmodeled dynamics driven multivariable nonlinear adaptive switching control[J]. IEEE Transactions on Neural Networks, 2011, 22(12):2154-2172.
[5] 张亚军, 牛宏, 柴天佑. 数据与未建模动态驱动的非线性PID切换控制[J]. 控制理论与应用, 2020, 37(3):481-491. ZHANG Y J, NIU H, CHAI T Y. Data-based un-modeled dynamics driven nonlinear PID[J]. Control and Decision, 2020, 37(3):481-491.
[6] 牛宏, 陶金梅, 张亚军. 一种新的数据驱动的非线性自适应切换控制方法[J]. 自动化学报, 2020, 46(11):2359-2366. NIU H, TAO J M, ZHANG Y J. A new data-driven nonlinear adaptive switching control method[J]. Acta Automatica Sinica, 2020, 46(11):2359-2366.
[7] NIDOZI I, RAHMANI M. LMI-based model predictive control for switched nonlinear systems[J]. Journal of Process Control, 2017, 59(9):49-58.
[8] ZHANG L, ZHANG Y, WANG D, et al. Multiple models generalized predictive control for superheated steam temperature based on MLD model[C]//IEEE International Conference on Automation & Logistics. Piscataway, USA:IEEE, 2007:2740-2743.
[9] FOROUZ B S, SEDIGH A K, MANZAR M N. Multiple model unfalsified adaptive generalized predictive control based on the quadratic inverse optimal control concept[J]. Optimal Control Applications and Methods, 2021, 42(3):769-785.
[10] 陶金梅, 牛宏, 张亚军, 等. 基于随机配置网络的非线性系统智能建模方法[J/OL]. 控制与决策, 2022, 37(10):2559-2564.TAO J M, NIU H, ZHANG Y J, et al. An intelligent modeling method for nonlinear systems based on random Configuration networks[J/OL]. Control and Decision, 2022, 37(10):2559-2564.
[11] TONG S C, SUI S, LI Y M. Observed-based adaptive fuzzy tracking control for switched nonlinear systems with dead-zone[J]. IEEE Transactions on Cybernetics, 2015, 45(12):2816-2826.
[12] TONG S C, LI Y M. Adaptive fuzzy output feedback Control for switched nonlinear systems with unmodeled dynamics[J]. IEEE Transactions on Cybernetics, 2016, 47(2):295-305.
[13] LI Y M, SUI S, TONG S C. Adaptive fuzzy control design for stochastic nonlinear switched systems with arbitrary switchings and unmodeled dynamics[J]. IEEE Transactions on Cybernetics, 2017, 47(2):403-414.
[14] PAWLOWSKI A, CERVIN A, GUZMÁN J L, et al. Generalized predictive control with actuator deadband for event-based approaches[J]. IEEE Transactions on Industrial Informatics, 2014, 10(1):523-537.
[15] ZHANG L X, ZHUANG S L, BRAATZ R D. Switched model predictive control of switched linear systems:Feasibility, stability and robustness[J]. Automatica, 2016, 67(1):8-21.
[16] 胡杰, 杜胜, 吴敏, 等. 铁前炉料制备过程先进控制与智能优化[J]. 信息与控制, 2018, 47(4):411-420. HU J, DU S, WU M, et al. Advanced control and intelligent optimization of raw-materials preparation process in ironmaking[J]. Information and Control, 2018, 47(4):411-420.
[17] 郭冠呈, 刘书明, 李俊禹, 等. 基于双向长短时神经网络的水量预测方法研究[J]. 给水排水, 2018, 54(3):123-126. GUO G C, LIU S M, LI J Y, et al. Study on water quantity prediction method based on bidirectional long and short time neural network[J]. Water & Wastewater Engineering, 2018, 54(3):123-126.
[18] 李爱莲, 张帅. 基于LSTM网络的焦炉火道温度预测建模仿真[J]. 计算机仿真, 2020, 37(6):466-470. LI A L, ZHANG S. Modeling and simulation of coke oven tunnel temperature prediction based on LSTM network[J]. Computer Simulation. 2020, 37(6):466-470.
[19] 任俊超, 刘丁, 万银. 基于混合集成建模的硅单晶直径自适应非线性预测控制[J]. 自动化学报, 2020, 46(5):1004-1016. REN J C, LIU D, WAN Y. Hybrid integrated modeling based adaptive nonlinear predictive control of silicon single crystal diameter[J]. Acta Automatica Sinica, 2020, 46(5):1004-1016.
[20] 孙俊峰, 李志斌. 基于LSTM的滚动预测算法的电缆缆芯温度的研究[J]. 电子测量技术, 2021, 44(21):84-88. SUN J F, LI Z B. Research on cable core temperature based on rolling prediction algorithm of LSTM[J]. Electronic Measurement Technology, 2021, 44(21):84-88.
[21] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8):1735-1780.