丁淑芬 劉凱 楊榮妮
摘要
本文針對分布式傳感器網(wǎng)絡(luò)系統(tǒng)的Fornasini-Marchesini (FM)狀態(tài)空間模型,對系統(tǒng)的預(yù)測控制器設(shè)計(jì)問題進(jìn)行研究.特別是針對所考慮的二維FM傳感器網(wǎng)絡(luò)系統(tǒng),提出了一種新的網(wǎng)絡(luò)預(yù)測控制方案來補(bǔ)償通信時(shí)滯.首先,根據(jù)李雅普諾夫穩(wěn)定性理論,給出了二維系統(tǒng)保持穩(wěn)定的充分條件;然后利用穩(wěn)定性條件,提出了一種新的預(yù)測控制器設(shè)計(jì)策略并保證系統(tǒng)的控制性能;最后,通過一個(gè)數(shù)值實(shí)例驗(yàn)證了所設(shè)計(jì)控制器的有效性.關(guān)鍵詞
分布式傳感器網(wǎng)絡(luò);FM狀態(tài)空間模型;網(wǎng)絡(luò)預(yù)測控制;通信時(shí)滯
中圖分類號 TM464
文獻(xiàn)標(biāo)志碼 A
0 引言
隨著計(jì)算機(jī)網(wǎng)絡(luò)和傳感器技術(shù)的發(fā)展,人們對分布式傳感器網(wǎng)絡(luò)產(chǎn)生了濃厚的興趣[1].分布式傳感器網(wǎng)絡(luò)由大量位于不同區(qū)域的智能傳感器節(jié)點(diǎn)組成[2],巨大的智能傳感器網(wǎng)絡(luò)在生活中有著廣泛的應(yīng)用,如農(nóng)業(yè)、制造業(yè)、土木工程和其他許多領(lǐng)域.值得注意的是,分布式傳感器網(wǎng)絡(luò)在實(shí)際中常常建模為二維系統(tǒng)[3],其動態(tài)更新發(fā)生在兩個(gè)獨(dú)立的方向.二維系統(tǒng)作為一個(gè)新興的領(lǐng)域,在實(shí)踐和理論中都具有大量的應(yīng)用,如圖像與信號處理、迭代學(xué)習(xí)、二維數(shù)字?jǐn)?shù)據(jù)濾波等方面,因此受到了廣泛的關(guān)注[4-6],對其進(jìn)行研究具有重要的意義[7-9].眾所周知,兩個(gè)變量使得二維系統(tǒng)的穩(wěn)定性分析和控制設(shè)計(jì)問題比一維系統(tǒng)更加困難和復(fù)雜.雖然二維系統(tǒng)可以看作是一維系統(tǒng)的擴(kuò)展,但是由于兩個(gè)變量之間的耦合,一些穩(wěn)定性分析和控制器設(shè)計(jì)的結(jié)果并不能直接應(yīng)用到二維系統(tǒng).因此,研究人員對二維系統(tǒng)的穩(wěn)定性分析產(chǎn)生了濃厚的興趣,Paszke等[10]研究了二維離散系統(tǒng)的魯棒穩(wěn)定性,Ahn等[11]實(shí)現(xiàn)了二維系統(tǒng)的耗散控制.
網(wǎng)絡(luò)化控制系統(tǒng)通過引入通信網(wǎng)絡(luò),具有了閉環(huán)反饋控制結(jié)構(gòu),包含傳感器、控制器和執(zhí)行器等部件.與傳統(tǒng)的點(diǎn)對點(diǎn)控制系統(tǒng)相比,網(wǎng)絡(luò)控制系統(tǒng)具有布線少、資源共享、可靠性高等優(yōu)點(diǎn),在模型工程中有著廣泛的應(yīng)用[12].網(wǎng)絡(luò)化控制系統(tǒng)在促進(jìn)工業(yè)制造業(yè)發(fā)展的同時(shí),也帶來了通信網(wǎng)絡(luò)的一些后續(xù)問題和挑戰(zhàn),如數(shù)據(jù)包丟失、網(wǎng)絡(luò)引起的延遲和量化.通信時(shí)滯和數(shù)據(jù)丟失是網(wǎng)絡(luò)通信系統(tǒng)最重要的問題,造成了網(wǎng)絡(luò)通信系統(tǒng)性能下降.到目前為止,已經(jīng)有許多和網(wǎng)絡(luò)化控制系統(tǒng)相關(guān)的結(jié)論[13-17],如Lin等[13]研究了具有擾動和時(shí)滯的切換系統(tǒng),Wang等[15]研究了故障估計(jì)和有限層位隨機(jī)最優(yōu)控制.
然而,這些控制方法只能被動地處理網(wǎng)絡(luò)引起的時(shí)滯,而不能有效地補(bǔ)償它們.為了消除通信時(shí)滯對系統(tǒng)穩(wěn)定性的不利影響,Liu提出了一種有效的控制方案,即網(wǎng)絡(luò)預(yù)測控制方法[18].上述文獻(xiàn)的共同之處在于它們都關(guān)注于一維系統(tǒng),而需要指出的是,二維系統(tǒng)的復(fù)雜特性使得網(wǎng)絡(luò)預(yù)測控制方法與一維系統(tǒng)有很大的不同.分析二維系統(tǒng)最常用的方法是基于模型的狀態(tài)空間方法,二維系統(tǒng)的模型包括FM模型、Roesser模型、Attasi模型等.本文采用了更具一般性的FM 狀態(tài)空間模型,由于基于FM模型的網(wǎng)絡(luò)化預(yù)測控制問題還沒有解決,因此本文的研究具有重要的理論和實(shí)踐意義.
情形2.具有通信時(shí)滯的網(wǎng)絡(luò)控制.假設(shè)反饋通道中從傳感器到控制器存在兩步時(shí)滯,即d=2.首先,我們使用真實(shí)傳輸?shù)臄?shù)據(jù)來代替補(bǔ)償?shù)臄?shù)據(jù).從圖4—6可以看出,系統(tǒng)不再穩(wěn)定,意味著通信時(shí)滯使系統(tǒng)不穩(wěn)定,降低了控制性能.
情形3.具有時(shí)滯補(bǔ)償?shù)木W(wǎng)絡(luò)預(yù)測控制.引入網(wǎng)絡(luò)化預(yù)測控制方案來補(bǔ)償時(shí)滯,進(jìn)一步穩(wěn)定系統(tǒng).控制器是由u(n,t)=K(n,t)給出.具有時(shí)滯補(bǔ)償?shù)亩S網(wǎng)絡(luò)化控制系統(tǒng)的狀態(tài)軌跡如圖7—9所示.結(jié)果表明,該方案可以有效地補(bǔ)償通信時(shí)滯.
6 主要結(jié)論
本文主要研究了以二維FM模型為代表的分布式傳感器網(wǎng)絡(luò)的預(yù)測控制問題.在對閉環(huán)二維系統(tǒng)進(jìn)行分析的基礎(chǔ)上,采用一種新的二維網(wǎng)絡(luò)預(yù)測控制方案對通信時(shí)滯進(jìn)行補(bǔ)償,保證了系統(tǒng)的穩(wěn)定性,達(dá)到了預(yù)期的系統(tǒng)性能.最后通過實(shí)例驗(yàn)證了所提方法的有效性.
參考文獻(xiàn)
References
[1] Qi H R,Iyengar S,Chakrabarty K.Multiresolution data integration using mobile agents in distributed sensor networks[J].IEEE Transactions on Systems,Man and Cybernetics,Part C (Applications and Reviews),2001,31(3):383-391
[2] Aboelfotoh H M F,Iyengar S S,Chakrabarty K.Computing reliability and message delay for cooperative wireless distributed sensor networks subject to random failures[J].IEEE Transactions on Reliability,2005,54(1):145-155
[3] Lin Z Y,Han T R,Zheng R H,et al.Distributed localization for 2-D sensor networks with bearing-only measurements under switching topologies[J].IEEE Transactions on Signal Processing,2016,64(23):6345-6359
[4] Liang J L,Wang Z D,Liu X H.Robust state estimation for two-dimensional stochastic time-delay systems with missing measurements and sensor saturation[J].Multidimensional Systems and Signal Processing,2014,25(1):157-177
[5] Wei Y L,Peng X Y,Qiu J B,et al.H∞ filtering for two-dimensional continuous-time Markovian jump systems with deficient transition descriptions[J].Neurocomputing,2015,167:406-417
[6] Duan Z X,Xiang Z R,Karimi H R.Stability and l1-gain analysis for positive 2D T-S fuzzy state-delayed systems in the second FM model[J].Neurocomputing,2014,142:209-215
[7] Rogers E,Galkowski K,Paszke W,et al.Multidimensional control systems:case studies in design and evaluation[J].Multidimensional Systems and Signal Processing,2015,26(4):895-939
[8] Sumanasena M G B,Bauer P H.Realization using the Fornasini-Marchesini model for implementations in distributed grid sensor networks[J].IEEE Transactions on Circuits and Systems I:Regular Papers,2011,58(11):2708-2717
[9] Knorn S,Middleton R H.Two-dimensional analysis of string stability of nonlinear vehicle strings[C]∥52nd IEEE Conference on Decision and Control,2013,DOI:10.1109/CDC.2013.6760814
[10] Paszke W,Lam J,Gakowski K,et al.Robust stability and stabilisation of 2D discrete state-delayed systems[J].Systems & Control Letters,2004,51(3/4):277-291
[11] Ahn C K,Shi P,Basin M V.Two-dimensional dissipative control and filtering for roesser model[J].IEEE Transactions on Automatic Control,2015,60(7):1745-1759
[12] Walsh G C,Ye H,Bushnell L G.Stability analysis of networked control systems[J].IEEE Transactions on Control Systems Technology,2002,10(3):438-446
[13] Lin H,Antsaklis P J.Stability and persistent disturbance attenuation properties for a class of networked control systems:switched system approach[J].International Journal of Control,2005,78(18):1447-1458
[14] Mahmoud M S.Generalized control of switched discrete-time systems with unknown delays[J].Applied Mathematics and Computation,2009,211(1):33-44
[15] Wang Y,Ding S X,Xu D M,et al.An H∞ fault estimation scheme of wireless networked control systems for industrial real-time applications[J].IEEE Transactions on Control Systems Technology,2014,22(6):2073-2086
[16] Xu H,Jagannathan S.Neural network-based finite horizon stochastic optimal control design for nonlinear networked control systems[J].IEEE Transactions on Neural Networks and Learning Systems,2015,26(3):472-485
[17] Ren M F,Zhang J H,Jiang M,et al.Minimum entropy control for non-Gaussian stochastic networked control systems and its application to a networked DC motor control system[J].IEEE Transactions on Control Systems Technology,2015,23(1):406-411
[18] Liu G P.Predictive controller design of networked systems with communication delays and data loss[J].IEEE Transactions on Circuits and Systems,2010,57(6):481-485
[19] Rogers E,Galkowski K,Paszke W,et al.Multidimensional control systems:case studies in design and evaluation[J].Multidimensional Systems and Signal Processing,2015,26(4):895-939
Networked predictive control of the distributed sensor networks
DING Shufen1 LIU Kai2 YANG Rongni1
1
School of Control Science and Engineering,Shandong University,Jinan 250061
2 Shandong Academy of Occupational Health and Occupational Medicine,Shandong First Medical University &
Shandong Academy of Medical Sciences,Jinan 250002
Abstract This paper is concerned with the networked predictive controller design problem for the distributed sensor network represented by the well-known discrete-time two-dimensional (2-D) Fornasini-Marchesini (FM) state-space model.Particularly,a novel networked predictive control (NPC) scheme is employed for the considered 2-D distributed sensor networks in order to compensate for the communication delay actively.Firstly,by using the Lyapunov stability theory,sufficient conditions are established for such distributed sensor networks.Based on the stability analysis results,a new predictive controller design strategy is proposed to stabilize the addressed systems and also achieve the desired control performance.Finally,an example is given to demonstrate the effectiveness of the newly developed NPC strategy for the distributed sensor networks.
Key words distributed sensor networks;Fornasini-Marchesini (FM) state-space model;networked predictive control (NPC);communication delay
收稿日期 2020-02-09
資助項(xiàng)目 國家自然科學(xué)基金(61873147);山東大學(xué)青年學(xué)者未來計(jì)劃(2017WLJH27)
作者簡介丁淑芬,女,碩士生,研究方向網(wǎng)絡(luò)化控制.shufending@163.com
楊榮妮(通信作者),女,博士,副教授,碩士生導(dǎo)師,研究方向?yàn)榫W(wǎng)絡(luò)化控制及多維系統(tǒng).rnyang@sdu.edu.cn