【摘要】：工程系統(tǒng)會(huì)存在許多跟蹤期望軌跡的問(wèn)題,為解決此類的問(wèn)題,許多的學(xué)者做了大量的研究,迭代學(xué)習(xí)控制也是其中被提出的一種解決方案。迭代學(xué)習(xí)控制適用于重復(fù)操作的工業(yè)控制過(guò)程,是智能控制領(lǐng)域的分支。系統(tǒng)在運(yùn)行的時(shí)候不可避免的會(huì)發(fā)生各種故障,而故障的發(fā)生有時(shí)也會(huì)在特定的頻域范圍內(nèi)。因此,針對(duì)特定頻域的故障重復(fù)控制系統(tǒng),迭代學(xué)習(xí)容錯(cuò)控制器是很好的一種選擇,可以維持系統(tǒng)比較理想的性能。本文結(jié)合重復(fù)過(guò)程控制的穩(wěn)定性理論,針對(duì)故障重復(fù)系統(tǒng),在有限頻域范圍設(shè)計(jì)迭代學(xué)習(xí)容錯(cuò)控制器,利用線性矩陣不等式(LMI)技術(shù)分析重復(fù)系統(tǒng)的單調(diào)收斂條件,并且求出控制器參數(shù)。本文的研究?jī)?nèi)容如下(1)針對(duì)執(zhí)行器故障線性離散重復(fù)過(guò)程,設(shè)計(jì)一種有限頻率范圍的迭代學(xué)習(xí)容錯(cuò)控制器。以LMI形式分別給出基于Kalman Yakubovich Popov(KYP)引理的全頻、分頻區(qū)域重復(fù)控制系統(tǒng)穩(wěn)定的充分條件,同時(shí)保證了故障系統(tǒng)在時(shí)域和頻域范圍內(nèi)的容錯(cuò)性能,重復(fù)注塑過(guò)程的注射速度控制仿真驗(yàn)證了分頻控制算法的有效性。(2)對(duì)于執(zhí)行器故障離散重復(fù)過(guò)程的系統(tǒng)穩(wěn)定問(wèn)題,設(shè)計(jì)一種有限頻率范圍的動(dòng)態(tài)迭代學(xué)習(xí)容錯(cuò)控制算法。定義故障系數(shù)矩陣,輸出跟蹤系統(tǒng)以及動(dòng)態(tài)迭代學(xué)習(xí)控制律,然后給出等價(jià)重復(fù)過(guò)程模型,沿故障系統(tǒng)的時(shí)間軸以及批次軸設(shè)計(jì)動(dòng)態(tài)迭代學(xué)習(xí)被動(dòng)容錯(cuò)控制器,基于KYP引理,給出系統(tǒng)穩(wěn)定的充分條件,多軸桁架式機(jī)器人控制仿真驗(yàn)證了頻域控制算法的有效性。(3)研究一類具有執(zhí)行器故障以及狀態(tài)時(shí)滯的線性離散系統(tǒng)的迭代學(xué)習(xí)容錯(cuò)控制問(wèn)題。設(shè)計(jì)一種有限頻域的迭代學(xué)習(xí)容錯(cuò)控制算法,借助于KYP引理,得出重復(fù)控制系統(tǒng)穩(wěn)定的充分條件,保證故障系統(tǒng)在頻域范圍內(nèi)的容錯(cuò)性能。重復(fù)注塑過(guò)程注塑速度過(guò)程仿真驗(yàn)證了設(shè)計(jì)的算法有效性和可行性。
[Abstract]:There will be many problems in the engineering system to track the desired track. in order to solve this kind of problems, many scholars have done a lot of research, and iterative learning control is also one of the solutions proposed. Iterative learning control is suitable for repeated industrial control process and is a branch of intelligent control field. It is inevitable that all kinds of faults will occur when the system is running, and the faults will sometimes occur in a specific frequency domain. Therefore, iterative learning fault-tolerant controller is a good choice for fault repetitive control systems in specific frequency domain, which can maintain the ideal performance of the system. In this paper, based on the stability theory of repetitive process control, an iterative learning fault-tolerant controller is designed for fault-repetitive systems in finite frequency domain, and the monotone convergence conditions of repetitive systems are analyzed by using linear matrix inequality (LMI) (LMI) technique. The parameters of the controller are obtained. The research contents of this paper are as follows: (1) an iterative learning fault-tolerant controller with finite frequency range is designed for linear discrete repetitive process of actuator fault. In the form of LMI, the sufficient conditions for the stability of the full frequency and frequency division region repetitive control system based on Kalman 鈮，

本文編號(hào)：2496795