王永婧 張冬雯 于健騏
摘要:針對具有狀態(tài)時滯特性的非線性離散系統(tǒng),利用線性矩陣不等式的方法和Lyapunov穩(wěn)定性理論,研究了基于狀態(tài)反饋的非線性系統(tǒng)模型預(yù)測控制問題?;赥S模糊模型對非線性離散系統(tǒng)進(jìn)行描述并給出一種“minmax”預(yù)測控制算法,采用模型預(yù)測控制與模糊理論相結(jié)合的方法,利用平行分布補償?shù)脑?,通過在每一個采樣時刻優(yōu)化無窮時域的性能指標(biāo),來求解在范數(shù)有界條件下相應(yīng)的狀態(tài)反饋控制律,討論了預(yù)測控制器的設(shè)計問題,分析了此設(shè)計問題的可解性,給出了狀態(tài)反饋控制器基于線性矩陣不等式的設(shè)計算法,保證了系統(tǒng)的穩(wěn)定性,通過仿真實例證明了所提控制算法的有效性及系統(tǒng)的穩(wěn)定性。該方法能夠在化工、冶金、機械等具有時滯特性的工業(yè)生產(chǎn)過程中得到很好的應(yīng)用。
關(guān)鍵詞:自動控制理論;非線性離散系統(tǒng);TS模糊模型;預(yù)測控制;時滯
中圖分類號:TP13文獻(xiàn)標(biāo)志碼:Adoi: 10.7535/hbgykj.2018yx01007
Model predictive control for a class of nonlinear systems
with timedelay based on TS model
WANG Yongjing1, ZHANG Dongwen2, YU Jianqi3
(1. School of Electrical Engineering, Hebei University of Science and Technology, Shijiazhuang, Hebei 050018, China; 2. School of Information Science and Engineering, Hebei University of Science and Technology, Shijiazhuang, Hebei 050018, China; 3.Department of Information Engineering, Environmental Management College of China, Qinhuangdao, Hebei 066004, China)
Abstract:Aiming at a class of nonlinear discrete systems with timedelay, model predictive control problem based on state feedback is addressed by using linear matrix inequality and Lyapunov stability theory. The target system is described by the TS fuzzy model, and the "minmax" type of model predictive control algorithm is given. The model predictive control is combined with fuzzy model, and the principle of parallel distributed compensation is used. At each sampling time, the performance of the infinite time domain is optimized, and the design problem of the predictive controller in the case of the norm bounded is discussed. The solvability of the problem is analyzed and the stability of the system is guaranteed, and the state feedback controller design algorithm based on linear matrix inequality is given. Finally, a simulation example is given to prove the effectiveness of the proposed method and the stability of the systems. This method can be well applied in the model predictive control process in fields of chemistry, metallurgy, and machinery.
Keywords:automatic control theory; nonlinear discrete systems; TS fuzzy model; predictive control; timedelay
模型預(yù)測控制因其能夠很好地處理控制和狀態(tài)的硬約束保證系統(tǒng)的穩(wěn)定性,而被廣泛應(yīng)用于工業(yè)系統(tǒng)中[16]。
在實際工業(yè)系統(tǒng)中,系統(tǒng)的非線性是普遍存在的,通常也會存在時滯現(xiàn)象。而時滯現(xiàn)象的出現(xiàn)可能會降低閉環(huán)系統(tǒng)的性能,甚至?xí)斐砷]環(huán)系統(tǒng)的不穩(wěn)定。文獻(xiàn)[7]針對一類具有多重狀態(tài)時滯和非線性擾動的不確定離散非線性系統(tǒng)構(gòu)造李雅普諾夫函數(shù),利用無限時域minmax優(yōu)化問題設(shè)計狀態(tài)反饋控制器。文獻(xiàn)[8]針對一類具有多重狀態(tài)和輸入時滯的不確定離散非線性系統(tǒng),充分利用時滯的上下界信息構(gòu)造改進(jìn)的二次李雅普諾夫泛函,從而降低系統(tǒng)的保守性。文獻(xiàn)[9]針對一類具有非線性擾動且同時存在多重狀態(tài)和輸入時滯的不確定系統(tǒng),通過滾動優(yōu)化來求解控制輸入,將無限時域二次性能指標(biāo)優(yōu)化問題轉(zhuǎn)化為具有LMI約束的線性目標(biāo)最小化問題,在線實時優(yōu)化性能指標(biāo),使得系統(tǒng)狀態(tài)平滑地趨于穩(wěn)定。文獻(xiàn)[10]針對一類輸入輸出受限的不確定離散時滯系統(tǒng),提出了輸出反饋魯棒預(yù)測控制方法,并給出了輸出反饋控制器存在的充分條件及構(gòu)造方法。文獻(xiàn)[11]將非線性預(yù)測控制與模糊系統(tǒng)理論相結(jié)合,基于Lyapunov方法設(shè)計了魯棒自適應(yīng)控制律,僅需在線調(diào)整2個參數(shù),無需調(diào)整權(quán)值,從而簡化了控制器的設(shè)計。文獻(xiàn)[12]研究了雙模MPC的設(shè)計方法,通過引入魯棒不變集構(gòu)造關(guān)于狀態(tài)估計的可行集和終端約束集,實現(xiàn)了輸出反饋雙??刂啤6鴮τ谟邢迺r域預(yù)測控制算法,在文獻(xiàn)[13]中,飽和局部控制律和終端加權(quán)矩陣均是通過LMI在線約束的最小化問題求解的,從而提高了閉環(huán)系統(tǒng)的性能。然而,范數(shù)有界不確定性也能很好地描述非線性系統(tǒng)的不確定性。TS模型能夠良好地描述非線性系統(tǒng)的特性,將其與預(yù)測控制的優(yōu)化算法相結(jié)合,能夠?qū)崿F(xiàn)對非線性系統(tǒng)的優(yōu)化控制,取得較好的控制效果。
將minmax無限時域預(yù)測控制的方法拓展到基于TS模糊模型描述的范數(shù)有界不確定非線性系統(tǒng)預(yù)測控制器設(shè)計當(dāng)中,利用相關(guān)的LyapunovKrasovskii函數(shù)結(jié)合線性矩陣不等式,實現(xiàn)該類非線性系統(tǒng)的預(yù)測控制,最后通過仿真實例驗證算法的有效性。
1問題描述
考慮一類由下述TS模糊模型描述的一般時滯非線性系統(tǒng):
4結(jié)語
將模型預(yù)測控制的方法應(yīng)用于具有輸入約束與狀態(tài)時滯的非線性離散系統(tǒng)中,將模糊理論與預(yù)測控制相結(jié)合,利用LyapunovKrasovskii泛函與LMI技術(shù)設(shè)計了模糊狀態(tài)反饋控制律,優(yōu)化了系統(tǒng)性能指標(biāo),給出了一種適用于此類系統(tǒng)的預(yù)測控制算法,最后通過仿真驗證,控制算法具有較好的控制性能。該方法仍存在不足之處,在時滯問題的處理上可以考慮時變時滯的情況,對系統(tǒng)進(jìn)行預(yù)測控制,從而更加全面的考慮系統(tǒng)的時滯情況。
參考文獻(xiàn)/References:
[1]席裕庚, 李德偉, 林姝. 模型預(yù)測控制——現(xiàn)狀與挑戰(zhàn)[J]. 自動化學(xué)報, 2013,39(3),222236.
XI Yugeng, LI Dewei, LIN Shu. Model predictive control:Status and challenges[J]. Acta Automatica Sinica, 2013,39(3):222236.
[2]席裕庚. 預(yù)測控制[M]. 北京:國防工業(yè)出版社, 2013.
[3]劉剛, 秦偉偉, 劉潔瑜,等. 運用區(qū)間算法的約束非線性系統(tǒng)系統(tǒng)魯棒模型預(yù)測控制[J]. 控制理論與應(yīng)用, 2014,31(6):735740.
LIU Gang, QIN Weiwei, LIU Jieyu, et al. RMPC for constrained nonlinear systems based on interval arithmetic[J]. Control Theory & Applications,2014,31(6):735740.
[4]何德峰,丁寶蒼,于樹友. 非線性系統(tǒng)模型預(yù)測控制若干基本特點與主題回顧[J].控制理論與應(yīng)用,2013,30(3):273287.
HE Defeng, DING Baocang, YU Shuyou. Review of fundamental properties and topics of model predictive control for nonlinear systems [J]. Control Theory & Applications, 2013,30(3):273287.
[5]張冬雯. 不確定系統(tǒng)的魯棒分析與綜合:矩陣不等式方法[M]. 北京:國防工業(yè)出版社, 2014.
[6]ZHANG Dongwen. Constrained robust model predictive control for timedelay descriptor systems with linear fractional uncertainty[J]. Engineering Review, 2015, 35(2):147155.
[7]周衛(wèi)東, 鄭蘭, 廖成毅,等. 多重狀態(tài)時滯系統(tǒng)的minmax魯棒預(yù)測控制[J]. 哈爾濱工程大學(xué)學(xué)報, 2016,37(12):16851690.
ZHOU Weidong, ZHENG Lan, LIAO Chengyi, et al. Minmax robust predictive control for multistate timedelay systems[J]. Journal of Harbin Engineering University, 2016,37(12): 16851690.
[8]周衛(wèi)東, 鄭蘭, 廖成毅,等. 多重時滯離散非線性系統(tǒng)的魯棒預(yù)測控制[J]. 哈爾濱工業(yè)大學(xué)學(xué)報, 2015,47(9):2430.
ZHOU Weidong, ZHENG Lan, LIAO Chengyi, et al. Robust prediction control for multiple time delay discrete nonlinear system[J]. Journal of Harbin Institute of Technology, 2015,47(9):2430.
[9]蘇成利, 趙家程, 李平. 一類具有非線性擾動的多重時滯不確定系統(tǒng)魯棒預(yù)測控制[J]. 自動化學(xué)報, 2013,39(5):644649.
SU Chengli, ZHAO Jiacheng, LI Ping. Robust predictive control for a class of multiple time delay uncertain systems with nonlinear distu rbance [J]. Acta Automatica Sinica, 2013,39(5):644649.
[10]陳秋霞, 俞立. 不確定離散時滯系統(tǒng)的輸出反饋魯棒預(yù)測控制 [J]. 控制理論與應(yīng)用, 2007,24(3):401406.
CHEN Qiuxia, YU Li. Robust model predictive control for uncertain discrete timedelay systems via dynamic output feedback[J]. Control Theory & Applications, 2007,24(3):401406.
[11]方煒, 姜長生. 一類基于模糊系統(tǒng)的非線性魯棒自適應(yīng)預(yù)測控制[J]. 西安交通大學(xué)學(xué)報, 2008,42(6):669673.
FANG Wei, JIANG Changsheng. Nonlinear robust adaptive predic tive control based on fuzzy systems[J]. Journal of Xian Jiaotong University,2008,42(6):669673.
[12]李志軍. 約束模型預(yù)測控制的穩(wěn)定性與魯棒性研究[D]. 北京:華北電力大學(xué),2005.
LI Zhijun. Research on Stability and Robustness of Constrained Model Predictive Control[D]. Beijing: North China Electric Power University,2005.
[13]LU Mei, JIN Chengbo, SHAO Huihe. An improved fuzzy predictive control algorithm and its application to an industrial CSTR process [J]. Chinese Journal of Chemical Engineering, 2009,17(1):100107.
[14] 梁華清, 張冬雯, 邢少光,等. 基于多層概率集的隨機系統(tǒng)預(yù)測控制[J]. 河北科技大學(xué)學(xué)報, 2016, 37(2): 205212.
LIANG Huaqing, ZHANG Dongwen, XING Shaoguang, et al. Predictive control for stochastic systems based on multilayer probabilistic sets[J]. Journal of Hebei University of Science and Technology, 2016,37(2):205212.
[15]CAO Yongyan, LIN Zongli, SHAMASH Y. Set invariance anlysis and gainscheduling control for LPV systems subject to actuator saturation[J].Systems & Control Letters,2002,46(2):137151.
[16]ZHANG Liqian, HUANG Biao. Robust model predictive control of singular systems[J]. Transactions on Automatic Control,2004,49(6): 10001006.