韓麗麗+趙立英+白敬
摘要:針對(duì)具有時(shí)變時(shí)延的網(wǎng)絡(luò)化控制系統(tǒng)(networked control system,NCS),提出一種新的分散事件觸發(fā)通訊機(jī)制。該機(jī)制可以有效的節(jié)約網(wǎng)絡(luò)資源并且不依賴于狀態(tài)數(shù)據(jù)的完整信息,若干個(gè)空間分布式傳感器結(jié)點(diǎn)采集狀態(tài)數(shù)據(jù)的局部信息并根據(jù)子傳輸機(jī)制傳輸信號(hào)。本文首先建立包含網(wǎng)絡(luò)時(shí)延和分散事件觸發(fā)條件的閉環(huán)NCS模型,然后利用Lyapunov穩(wěn)定性理論和線性矩陣不等式方法給出閉環(huán)系統(tǒng)漸近穩(wěn)定的充分條件,進(jìn)而給出 控制器和分散事件觸發(fā)機(jī)制的協(xié)同設(shè)計(jì)方法,最后通過Matlab數(shù)值仿真驗(yàn)證該方法的有效性。
關(guān)鍵字:網(wǎng)絡(luò)控制系統(tǒng);分散事件觸發(fā)機(jī)制;時(shí)變時(shí)延;線性矩陣不等式
中圖分類號(hào):TP273 文獻(xiàn)標(biāo)志碼:A
Abstract: Aim at the networked control systems with time-varying delay (networked control system, NCS), in order to save the limited network resources, decentralized event-trigger communication mechanism is introduced. This mechanism can effectively save network resources and does not depend on the complete information of state data, many spatially distributed sensors collecte local information and transmit it according to sub-communication scheme.By establishing a closed-loop NCS model containing the delay and decentralized event-trigger, and using Lyapunov stability theory and liner matrix inequality approach, the sufficient condition is derived for the closed-loop system with asymptotic stability. Moreover, a method of co-design is given for controller and decentralized event-trigger mechanism. At last, a numerical simulation example is given to prove the effectiveness of the method.
Key word: Networked control systems;decentralized event-trigger; time-vary delay; liner matrix inequality
1.引言部分
網(wǎng)絡(luò)控制系統(tǒng)是由傳感器,控制器,執(zhí)行器等部件通過網(wǎng)絡(luò)構(gòu)成的閉環(huán)系統(tǒng),與傳統(tǒng)的控制系統(tǒng)相比,NCS具有成本低,接線少,便于維護(hù)與安裝等優(yōu)點(diǎn),但是由于網(wǎng)絡(luò)寬帶有限,該系統(tǒng)面臨著時(shí)延,丟包,失穩(wěn)等現(xiàn)象[1-3]。另外,由于NCS的大規(guī)模分布式構(gòu)架,工作環(huán)境和系統(tǒng)建模誤差的變化,其不確定因素也變多,所以針對(duì)同時(shí)具有參數(shù)不確定性和外部擾動(dòng)的系統(tǒng)有必要研究其魯棒穩(wěn)定性[4-5]。
為了有效利用網(wǎng)絡(luò)資源,事件觸發(fā)控制成為當(dāng)前研究的熱點(diǎn)之一。文獻(xiàn)[6]提出一種離散事件觸發(fā)機(jī)制,即在采樣時(shí)刻檢測(cè)系統(tǒng)的狀態(tài)并計(jì)算相應(yīng)誤差,然后與預(yù)先設(shè)定閾值比較,判斷系統(tǒng)狀態(tài)數(shù)據(jù)是否需要傳輸。文獻(xiàn)[7]在離散事件觸發(fā)基礎(chǔ)上研究了含參數(shù)不確定的時(shí)變時(shí)延控制系統(tǒng)的魯棒完整性。文獻(xiàn)[8]在文獻(xiàn)[7]的時(shí)變時(shí)延系統(tǒng)中又增加了外部擾動(dòng)因素并進(jìn)行了 控制器的設(shè)計(jì)保證系統(tǒng)具有 性能。文獻(xiàn)[9]和文獻(xiàn)[10]研究的分別是數(shù)據(jù)采樣系統(tǒng)和神經(jīng)網(wǎng)絡(luò)控制系統(tǒng),為了達(dá)到節(jié)約有限資源的作用而提出一種分散事件觸發(fā)機(jī)制。受此啟發(fā),把這一思想運(yùn)用在網(wǎng)絡(luò)控制系統(tǒng)中??紤]到現(xiàn)有的網(wǎng)絡(luò)控制系統(tǒng)中關(guān)于事件觸發(fā)的文獻(xiàn)都是假設(shè)用一個(gè)事件觸發(fā)條件判斷采集的數(shù)據(jù)是否可以通過網(wǎng)絡(luò)傳輸。而本文提出的分散事件觸發(fā)機(jī)制與現(xiàn)有文獻(xiàn)相比有以下幾點(diǎn)創(chuàng)新:第一,用若干個(gè)空間分布式傳感器采集狀態(tài)數(shù)據(jù)的局部信息,然后利用相應(yīng)的事件觸發(fā)條件判斷局部信息是否需要傳輸。顯然這樣設(shè)計(jì)的優(yōu)點(diǎn)在于不同的事件觸發(fā)機(jī)制提供不同的傳輸條件,每個(gè)觸發(fā)機(jī)制只是依賴狀態(tài)數(shù)據(jù)的局部信息。第二,改善以往的單通道數(shù)據(jù)傳輸方式,采用多通道傳輸?shù)乃枷?,盡可能緩解帶寬受限的問題。第三,分散事件觸發(fā)機(jī)制具有更廣泛的意義,文獻(xiàn)[6-8]提到的離散事件觸發(fā)機(jī)制都是分散事件觸發(fā)機(jī)制的特殊情況。
本文在文獻(xiàn)[6-8]的基礎(chǔ)上,考慮不確定性網(wǎng)絡(luò)時(shí)延的NCS并通過分散事件觸發(fā)機(jī)制來提高網(wǎng)絡(luò)資源的利用率。首先提出包含網(wǎng)絡(luò)屬性,不確定參數(shù)和觸發(fā)條件的閉環(huán)系統(tǒng)漸近穩(wěn)定的充分條件,進(jìn)一步給出 狀態(tài)反饋控制器和分散事件觸發(fā)機(jī)制的協(xié)同設(shè)計(jì)方法。最后通過仿真算例驗(yàn)證了該方法的有效性。
5結(jié)論
本文針對(duì)一類具有時(shí)變時(shí)延的不確定NCS,提出了一種新穎的事件觸發(fā)機(jī)制,即分散事件觸發(fā)機(jī)制,使得系統(tǒng)在保持穩(wěn)定的前提下還有效的節(jié)約了網(wǎng)絡(luò)資源并降低了信號(hào)的傳輸率。同時(shí)給出了網(wǎng)絡(luò)控制系統(tǒng) 控制器和通訊協(xié)議協(xié)同設(shè)計(jì)的方法。最后通過數(shù)值仿真驗(yàn)證了本文所提方法的有效性。
參考文獻(xiàn)
[1] Huijun Gao,Xiangyu Meng,Tongwen Chen.Stab-
ilization of network control systems with a new
delay Characterization[J].IEEE Transaction on
Automatic Control,2008,53(9):2142-2148 .
[2] Zidong Wang,F(xiàn)uwen Yang,Daniel W.C.Ho,Xiao
hui Liu.Robust control for network systems
with random packet losses[J].IEEE Transactions
on Systems,Man,and Cybernetics,Part B(Cybern-
etics),2007,37(4):916-924.
[3] Liqian Zhang,Yang Shi,Tongwen Chen,Biao Hua-
ng.A new method for stabilization of networked
control systems with random delays[J].IEEE Tran-
sactions on AutomaticControl,2005,50(8):1177-
1181.
[4] Dong Yue,Qinglong Han,James Lam.Networked-
based robust control of systems with uncer-
tainty[J].Automatica,2005,41(6):999-1007.
[5] Hai Lin,Guisheng Zhai,P.J.Antsaklis.Robust stab-
ility and disturbance attenuation analysis of a class
of networked control systems[J]. In Proceedings
of the conference on decision and control,2003,2
(10):1182-1187.
[6] Peng Chen,QingLong Han,Dong Yue.To transmmit
or not transmit:a discrete event-triggered commu-
nication Scheme for networked Takagi-Sugeno fu-
zzy systems [J],IEEE Transactions on Fuzzy Syst-
Ems,,2013,21(1):164-170.
[7] 李煒,趙莉,蔣棟年等.基于事件觸發(fā)的NCS魯棒
完整性設(shè)計(jì)[J].蘭州理工大學(xué)學(xué)報(bào),2014,40(1):
74-79.
[8] Shen Yan,Huaichen Yan,Hongbo Shi,Hao Zhang.
controller design of event-triggered networ-
ked control systems[C].International Conference
on Mechatronics and Control,2014:1972-1976.
[9] Yanpeng Guan,Qinglong Han,Chen peng.Decentr-
alized event-triggered control for sampled-data
systems with asnchronous sampling[C].American
Conrol Conference.2013:6565-6570.
[10] Jin Zhang,Peng Chen.Synchronization of master-
slave neural networks with a decentralized event
triggered communication scheme[J].Neurocomp-
uting.2016,173:1824-1831.
[11] Xiaofeng Wang, Michael D. Lemmon. Self-
feedback control systems with finite-L2 gain
stability[J].IEEE Transactions on Automatic
Control.2009,54(3):452-467.
[12] Xiefu Jiang,Qinglong Han,Shirong Liu,Anke
Xue.A new stabilization criterion for net-
worked control systems[J].IEEE Transactions
on Automatica Control.2008,53(4):1025-1032.
[13] Park Poogyeon,Ko Jeong Wan,Jeong Changki.
Reciprocally convex approach to stability of
systems with timi-varying delay[J].Automatic.
2011,47(1):235-238.
[14] Xie Lihua.Output feedback control of syst-
ems with parameter uncertainty[J].International
Journal of Control.1996,63(4):741-750.
[15] Junlin Xiong,James Lam. Stabilization of netwo-
rked control systems with a logic ZOH[J].IEEE
Transactions on Automatic Control.2009,54(2):
358-363.