Periodic Regulation of Linear Time-Delay Systems via Youla-Kučera Parametrization This work was supported by the Czech Science Foundation project 24-10301S, and by the European Union under the project Robotics and advanced industrial production No. C ​ Z ​ .02.01.01 / 00 / 22 ​ _ ​ 008 / 0004590 CZ.

Can Kutlu Yüksel, Tomas Vyhlidal, Silviu-Iulian Niculescu · 2025

Read Periodic Regulation of Linear Time-Delay Systems via Youla-Kučera Parametrization This work was supported by the Czech Science Foundation project 24-10301S, and by the European Union under the project Robotics and advanced industrial production No. C ​ Z ​ .02.01.01 / 00 / 22 ​ _ ​ 008 / 0004590 CZ. by Can Kutlu Yüksel, Tomas Vyhlidal, Silviu-Iulian Niculescu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-17792.txt.

Introduction

Can Kutlu Yüksel, Tomas Vyhlidal, and Silviu-Iulian Niculescu proposed a new way in 2025 to make machines follow a repeating pattern, even when signals arrive late. Factories often need such steady control, but delays make the task hard and can make a system unstable.

Their method starts from a controller that already keeps the system stable, then adds a tuned part that lets it track or reject a repeating signal. This reverses the usual order of design steps. Tests on three kinds of delayed systems showed that even unstable ones could be guided into steady repeating control by upgrading a simple controller. The tuning step comes down to solving a set of plain equations.

Catalog record

CREATORCan Kutlu Yüksel, Tomas Vyhlidal, Silviu-Iulian Niculescu
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE24 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-17792

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Periodic Regulation of Linear Time-Delay Systems via Youla-Kučera Parametrization This work was supported by the Czech Science Foundation project 24-10301S, and by the European Union under the project Robotics and advanced industrial production No. C ​ Z ​ .02.01.01 / 00 / 22 ​ _ ​ 008 / 0004590 CZ.02.01.01/00/22\_008/0004590 . The work of the first and last authors was also supported by a public grant overseen by the French National Research Agency (ANR) as part of the « Investissements d’Avenir » program, through the "ADI 2020" project funded by the IDEX Paris-Saclay, ANR-11-IDEX-0003-02. Preprint submitted to IFAC Can Kutlu Yüksel, Tomas Vyhlidal, Silviu-Iulian Niculescu Abstract The paper proposes an alternative way to achieve the Internal Model Principle (IMP) in contrast to the standard way, where a model of the signal one wishes to track/reject is directly substituted into the closed-loop. The proposed alternative approach relies on an already-existing stabilizing controller, which can be further augmented with a Youla-Kučera parameter to let it implicitly admit a model of a signal without altering its stabilizing feature. Thus, with the proposed method, the standard design steps of realizing IMP are reversed. The validity and potential of the proposed approach are demonstrated by considering three different types of time-delay systems. It is shown through simulations that all considered unstable systems, despite the infinite-dimensional closed-loop, can be straightforwardly periodically regulated by augmenting their stabilizing PI controllers. Thanks to a specifically chosen structure for the Youla-Kučera parameter, the required tuning can be done by solving a set of linear equations. keywords periodic control \sep time-delay \sep internal model control \sep vibration control \sep disturbance rejection \sep infinite-dimensional systems. 1 Introduction A popular way of achieving the famous Internal Model Principle (IMP) Francis and Wonham 1976 to make a closed-loop control system track/reject periodic signals requires the closed-loop to encapsulate the model of the signal and then its stability to be ensured. This lets the poles of the signal model become the zeros of the closed-loop sensitivity, which in return ensures the asymptotic tracking/rejection of the targeted signal by the control system. Such a design approach is adopted by the Repetitive Control method to achieve periodic regulation Hara et al. 1988 . However, for closed-loop systems involving infinite-dimensional models of both the signal and the system, guaranteeing stability can be a challenging task. In this paper, we propose and show an alternative approach made possible by the Youla-Kučera parametrization. The key benefit of this parametrization is that it allows for the swapping of the order of the two steps in the standard IMP realization. That is, one can alternatively first stabilize the system and then augment the controller to implicitly have the signal model. Such an approach can become especially convenient when taking into account the literature available for the stabilization of time-delay systems. In our opinion, based on this literature, periodic regulation of time-delay systems can be readily achieved without going through the challenging aspects of infinite dimensionality. Time-delay systems can be stabilized by considering various stability analysis methods and different controller structures. See, for instance, Michiels and Niculescu 2007 for eigenvalue-based methods, Sipahi et al. 2011 for graphical methods, Fridman 2001 for time-domain methods, and the references therein. Among them, PID controllers can still be relevant in the stabilization of time-delay systems Silva et al. 2005 or more recently Ma et al. 2022 ; Appeltans et al. 2022 . Furthermore, by letting the Youla-Kučera parameter take a particular structure involving delays, the essential regulation requirements can be posed as linear constraints and can be further optimized for additional constraints due to its optimization-friendly structure. Note that such similar methods based on parametrization were considered before for delay systems mainly in delay compensation Zitek et al. 2007 ; Pekař and Prokop 2007 and in the design of repetitive controllers Pipeleers et al. 2008 ; Chen and Tomizuka 2013 . The proposed work can be seen as the continuation and further extension of the Internal Model Control(IMC)-based periodic regulation approach previously proposed by the authors Yüksel et al. 2023 ; Yuksel et al. 2023 ; Yüksel et al. 2025 in the recent years. It is worth mentioning that the previous works were confined to stable and minimum-phase finite dimensional systems suffering from input delays. With the proposed extension, this restriction is no longer required and, in our opinion, it represents a novelty in the open literature. The structure of the paper is as follows: Section 2 introduces the Youla-Kučera parametrization for single-input-single-output (SISO) linear time-invariant (LTI) systems, whether with or without delay in the control loop. In addition, useful identities based on this parametrization are highlighted. In Section 3, a particular structure for the Youla-Kučera parameter based on time-delays is presented and is used to turn the periodic regulation requirements into linear constraints on the chosen (Youla-Kučera) structure. Section 4 demonstrates the effectiveness of the overall proposed approach via three examples, which are validated through simulations. Finally, Section 5 concludes the paper. The notations are standard and explained upon first used. 2 Youla-Kučera Parametrization for SISO LTI Systems Consider the classical feedback structure shown in Fig 1 , which aims to control the system G(s) via controller C(s) . As a fundamental requirement for this system to be regulated properly, a necessary condition is that the closed-loop should be stabilized by the controller. C(s) Controller G(s) System e y d r - Figure 1: Classical Feedback Control. Suppose that we have a controller, denoted by C_{p}(s) , which guarantees a stable closed-loop system, even if it does not fully satisfy the additional conditions imposed by the regulation requirements. Then, with respect to this controller, the Youla-Kučera parameter can capture all other stabilizing controllers, denoted generally by C(s) , for the system under consideration. Briefly stated, among these stabilizing controllers, our goal is to find a controller that can also satisfy our regulation requirements. The process of deriving these controllers in a parametric form begins by representing both the system and the controller in a coprime-factorized form in terms of stable and proper transfer functions, G(s)=\frac{N_{G}(s)}{D_{G}(s)}\quad\textrm{and}\quad C(s)=\frac{N_{C}(s)}{D_{C}(s)} (1) with N_{G}(s),N_{C}(s),D_{G}(s),D_{C}(s)\in\mathbb{R}_{ps}(s) . Note that \mathbb{R}_{ps}(s) is the vector space of all proper and stable rational functions of the Laplace variable s . Based on this factorization, the closed-loop transfer function writes as: \begin{split}G_{cl}(s)&=\frac{G(s)C(s)}{1+G(s)C(s)}\\ &=\frac{1}{D_{G}(s)D_{C}(s)+N_{G}(s)N_{C}(s)}N_{G}(s)N_{C}(s),\end{split} (2) Notice that, by factorization, N_{G}(s) , and N_{C}(s) are stable Hence, for the closed-loop to be stable (and proper), U^{-1}(s):=\frac{1}{D_{G}(s)D_{C}(s)+N_{G}(s)N_{C}(s)}, (3) must be stable (and proper). Consequently, this requirement leads to a Bézout identity, D_{G}(s)X(s)+N_{G}(s)Y(s)=1, (4) where X(s):=D_{C}(s)U^{-1}(s)\in\mathbb{R}_{ps}(s) and Y(s):=N_{C}(s)U^{-1}(s)\in\mathbb{R}_{ps}(s) . This Bézout identity implies that if we can find such X(s) and Y(s) , then we can use this to generate a controller that leads to a stable close-loop via C(s)=\frac{N_{C}(s)}{D_{C}(s)}=\frac{Y(s)}{X(s)}. (5) Thus, having C_{p}(s) at hand implies that we can find X_{p}(s),Y_{p}(s)\in\mathbb{R}_{ps} such that ( 4 ) is satisfied and C_{p}(s)=\frac{Y_{p}(s)}{X_{p}(s)} . Furthermore, if we happen to know such X_{p}(s) and Y_{p}(s) , notice that for an arbitrary Q(s)\in\mathbb{R}_{ps} having X(s)=X_{p}(s)-N_{G}(s)Q(s) and Y(s)=Y_{p}(s)+D_{G}(s)Q(s) also satisfies the Bézout identity. Hence, one can generally express all other stabilizing controllers based on C_{p}(s) as an affine rational function of Q(s) given by, C(s)=\frac{Y_{p}(s)+D_{G}(s)Q(s)}{X_{p}(s)-N_{G}(s)Q(s)}. (6) This arbitrary variable Q(s) is referred to as the Youla-Kučera

Read the complete text