Robust gain-scheduled blending control of raw-mix quality in cement industries

Michael Amrhein, Vinicius De Oliveira, Alireza Karimi
Ieee Service Center, 445 Hoes Lane, Po Box 1331, Piscataway, Nj 08855-1331 Usa, 2011
Article de conférence

Résumé

This paper addresses the control of the blending process in cement industries. This process can be modeled by a nonlinear multivariable system with large parametric uncertainty. Using a specific transformation, a linear parameter varying (LPV) model with set-points as scheduling parameters is developed. Moreover, the model uncertainty originated from the stochastic variation of the composition of the input materials is represented as a polytopic multimodel uncertainty. Then a multivariable gain-scheduled robust controller is designed by convex optimization to control the quality of the raw mix in the blending process. The control performance is illustrated by simulation and compared with a robust controller based on a nominal model.

Official source

https://infoscience.epfl.ch/record/169885?ln=fr

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Concepts associés

Chargement

Publications associées

Chargement

The scope of this thesis encompasses two main subjects: fixed-structure data-driven control design on one side, and control design in power systems on the other. The overall goal is to identify challenging and relevant problems in power systems, to express them as rigorous specifications from the viewpoint of control systems, and to solve them by developing and applying advanced methods in robust control. This work aims to combine expertise from both fields to open up a holistic perspective and bridge the gap between control and power systems. First, the derivation of a novel fixed-structure, data-driven frequency-domain control design method for multivariable systems is described. A key feature of the method is that only the frequency response of the plant is required for the design, and no parametric model is required. The designed controllers are fully parametrized in terms of matrix polynomial functions and can take a centralized, decentralized or distributed structure. The controller performance is formulated as H_2 and H_infinity constraints on any loop transfer function. A convex formulation of the optimization problem is derived, and it is shown that the solution converges to a locally optimal solution of the original problem. The versatility of the design method is demonstrated in various simulation examples, as well as in experiments on two electromechanical setups. Next, a frequency-domain modeling approach for power grids is discussed. A model based on dynamic phasors is developed that represents the electromagnetic and electromechanic dynamics of lines, inverters, synchronous machines and constant power loads. It also offers a modular structure that makes it straightforward to combine white-, grey- and blackbox models in a single framework. Then, the control design method and dynamic phasor model are applied in two relevant power systems case studies. First, the design of a decentralized current controller for parallel, grid-connected voltage source inverters in a typical distribution grid is considered. It is shown how performance specifications can be formulated as frequency-domain constraints in order to attenuate the resonances introduced by the output filters and coupling effects, and to provide robustness against model uncertainties and grid topology changes. The controllers for all VSIs are designed in a single step, and stability and performance is guaranteed by design. Furthermore, an approach for plug-and-play control design is presented. The results are validated in numerical simulation as well as in power-hardware-in-the-loop experiments. The second study concerns the design of a distributed controller that combines primary and secondary frequency and voltage control for an islanded, meshed low-voltage grid with any number of voltage source inverters and synchronous generators in a single framework. No assumption on the R/X-ratio of the lines is made, and it is shown how advanced control specifications such as proportional active power sharing, zero frequency steady-state error and decoupling can be formulated as constraints on the norm of weighted sensitivity functions. Furthermore, the communication delays of the distributed controller are considered exactly during the design. The controller is implemented in numerical simulation, and results show significantly improved performance as compared to the classical hierarchical structure.

EPFL2018

Chargement

H∞ and μ-Synthesis for Nanosatellites Rendezvous and Docking

Volker Gass, Camille Sébastien Pirat

In this brief, the nanosatellite rendezvous and docking problem is tackled. It was never attempted for small spacecraft, as critical technologies, such as six-degree-of-freedom (DoF) micropropulsion systems, have only recently become available due to advances in MEMS. The typical level of noise in nanosatellites' sensors and actuators combined with the dynamics uncertainties, low actuation capabilities, and reliability requirements makes the use of robust control appropriate. The system is described by a linearized rotation/translation, six DoFs, and coupled dynamics, including fuel sloshing. An H∞ controller is first designed, in which robust stability and performance are assessed using structured singular values. The controller robustness is then improved using μ-synthesis. Nonlinear Monte Carlo simulations for both controllers, including realistic sensors and actuators models, are provided allowing a thorough assessment of the complete guidance, navigation and control (GNC). The sought GNC schemes are shown to be robust to the modeled uncertainties and to satisfy the docking requirements.

2019

Chargement

Sample Complexity of Linear Quadratic Gaussian (LQG) Control for Output Feedback Systems

Luca Furieri, Na Li

This paper studies a class of partially observed Linear Quadratic Gaussian (LQG) problems with unknown dynamics. We establish an end-to-end sample complexity bound on learning a robust LQG controller for open-loop stable plants. This is achieved using a robust synthesis procedure, where we first estimate a model from a single input-output trajectory of finite length, identify an H-infinity bound on the estimation error, and then design a robust controller using the estimated model and its quantified uncertainty. Our synthesis procedure leverages a recent control tool called Input-Output Parameterization (IOP) that enables robust controller design using convex optimization. For open-loop stable systems, we prove that the LQG performance degrades linearly with respect to the model estimation error using the proposed synthesis procedure. Despite the hidden states in the LQG problem, the achieved scaling matches previous results on learning Linear Quadratic Regulator (LQR) controllers with full state observations.

PMLR2021

Chargement

EPFL2018