Seminar Winter 2007

Le Laboratoire d’Automatique de l’EPFL a le plaisir de vous inviter aux séminaires selon la liste ci-après. Une mise à jour régulière des informations concernant ces séminaires est disponible à l’adresse sur cette page. En particulier, il est conseillé aux visiteurs externes de vérifier que les séminaires soient dispensés comme prévu ci-dessous.

Where: Salle de séminaire LA-EPFL, ME C2 405 (2è étage), 1015 Lausanne

When: Friday at 10h15

Fall 2007 seminars

System Design and Optimal Control of Polygeneration System

28.09.2007 – Dr. F. Maréchal – Industrial Energy Systems Laboratory (LENI), EPFL.

The LENISYSTEM group is developing optimal process design methodologies for application to industrial processes, energy conversion and urban systems. After a brief overview of the LENISYSTEM group activities, the presentation will focus on the optimal design of polygeneration systems in urban areas, such as combined heat and power and heat pumping systems. The thermo-economic design of polygeneration systems in urban or building systems includes the sizing of both conversion and storage equipments, whose performance depends on the selected optimal management strategy. A thermo-economic optimization strategy has been developed to design such systems. The problem is a mixed-integer nonlinear programming (MINLP) problem that is solved upon combining a multiobjective optimization algorithm with a mixed-integer linear programming (MILP) formulation for the optimal management strategy. In order to extend the validity of this study, the optimal management strategy has been included into an optimal control system that allows operating the system by accounting on weather and market forecasts, together with a model-based optimal management algorithm based on MILP models.

Dissipation Inequalities in Systems Theory: Past, Present and Future

12.10.2007 – Prof. F. Allgöwer – Institute for Systems Theory and Automatic Control, University of Stuttgart, Germany.

Dissipation inequalities play a fundamental role in systems and control theory and dissipativity is a very useful concept in the analysis and design of nonlinear control systems. They were introduced in the early 1970s as a generalization of Lyapunov inequalities to systems having inputs and outputs. Similar to Lyapunov theory, the biggest problem in applications is the construction of the storage function, which is the generalization of the Lyapunov function. However, for certain class of systems, storage functions can be constructed systematically, e.g., based on back-stepping techniques or recently developed tools from the area of computational semi-algebraic geometry, namely semi-definite programming and the sum of squares decomposition. In this talk we will give a brief historical perspective and an introduction to the system theoretic concept of dissipation inequalities. We will present some recent results on dissipation inequalities, e.g. minimum phase analysis, stability analysis of nonlinear differential algebraic equation (DAE) systems, and nonlinear feedback and observer design that are based on novel dissipation inequalities and we will discuss questions concerning the computation of the storage functions. The methods will be demonstrated and critically assessed with various examples from engineering and systems biology.

Optimizing NMPC – Concept and Application to Chromatographic Separations

16.11.2007 – Prof. S. Engell – Process Control Laboratory, Department of Biochemical and Chemical Engineering, University of Dortmund, Germany.

Abstract not yet available.

Teaching Hints for Engineering Educators – A Baker’s Dozena

23.11.2007 – Prof. O. D. Crisalle – Chemical Engineering Department, University of Florida, Gainesville, FL.

Twelve teaching hints that are useful for enhancing the effectiveness and efficiency of engineering educators are presented in this seminar. Effectiveness is interpreted in the sense of how much knowledge the teacher is able to impart to the students during a typical lecture interaction, and efficiency in terms of how much effort must be invested by the teacher to ensure an effective lecture delivery. The hints covered include techniques for learning student names in relatively large classes, organizing the delivery of lecture materials, optimizing voice projection and body language, and avoiding common traps in the adoption of computer-based teaching support tools.

A Geometrical Approach to Model Variance Analysis in System Identification

30.11.2007 – Prof. H. Hjalmarsson – Centre de Automatic Control Signals, Sensors and Systems, Royal Institute of Technology (KTH) Stockholm, Sweden.

There are many results in the system identification literature on how various model structures and identification experiments affect the model accuracy, or the model variance. In this talk, we present a geometrical approach to analyze the variance of different estimated quantities, which aims to simplify this type of analysis. We provide a number of examples to illustrate the usefulness of the technique, ranging from estimation of properties of linear time-invariant (LTI) systems (e.g. impulse response coefficients, zeros, poles and system gains) to structural properties (e.g. the benefits of multiple inputs and/or sensors, open vs. closed loop identification), and finally to non-linear systems.

Systèmes plats, développements récents sur leur caractérisation, conséquences sur l’équivalence entre platitude différentielle et linéarisabilité par bouclage dynamique.

07.12.2007 – Dr. J. Lévine – Centre Automatique et Systèmes (CAS), Ecole des Mines de Paris, Fontaine-bleau, France.

Après un bref rappel sur la notion de platitude différentielle et ses applications, on commencera par montrer comment caractériser les sorties plates d’un système linéaire commandable. On introduira dans un second temps les outils algébriques et géométriques permettant d’obtenir la caractérisation cherchée en non linéaire. Cette dernière s’exprime sous la forme d’une généralisation des équations de structure du repère mobile de Cartan. On discutera de la complexité de ces équations du point de vue du calcul formel sur un exemple. On présentera enfin, comme une conséquence directe de ce résultat, la preuve de l’équivalence entre platitude et linéarisabilité par bouclage dynamique, question restée ouverte jusqu’à présent.

Elucidating the Control Mechanism for DNA Damage Repair with the p53-Mdm2 System: Single Cell Data Analysis and Ensemble Modeling

14.12.2007 – Prof. B. A. Ogunnaike – Department of Chemical Engineering, University of Delaware, Newark DE, USA.

The p53-Mdm2 system, which plays a crucial role in DNA damage repair, is one of the best-studied of the ‘negative feedback motifs’ known to be present in human cells. Recent experimental evidence about DNA damage response using this system has raised some fundamental questions about the control mechanism employed. In response to DNA damage, an ensemble of cells shows a damped oscillation in p53 expression whose amplitude increases with increased DNA damage-consistent with ‘analogue’ control. Recent experimental results, however, show that the single cell response is a series of discrete pulses in p53; and with increase in DNA damage, neither the height nor the duration of the pulses change, but the mean number of pulses increase-consistent with ‘digital’ control. In this seminar we present a comprehensive, systems engineering model that uses published data to elucidate this mechanism and resolve the dilemma. First, we develop a dynamic model of the p53-Mdm2 system that produces non-oscillatory responses to a stress signal. Next, from a careful analysis of the data we develop a probabilistic model of the distribution of pulses observed in a cell population. Finally, we combine the two with the simplest possible digital control algorithm to show how oscillatory responses whose amplitudes grow with DNA damage can arise from single cell behavior in which each single pulse response is independent of the extent of DNA damage.

Autonomous Aerial Refueling Technology: Demonstration for UAV Applications

21.12.2007 – Dr. N. Nabaa – Sierra Nevada Corporation (SNC), Sparks NV, USA.

The presentation describes a system, developed by Sierra Nevada Corporation (SNC), which has demonstrated the world’s first autonomous airborne refueling operations, using a NASA Dryden Flight Research Center F/A-18 as a surrogate UAV. The system includes an integrated sensor suite, fusing GPS/INS and visual imagery, with a high-gain precision guidance and control processor designed for fully autonomous in-flight refueling from rendezvous to trail positioning, drogue capture and hold with fuel transfer followed by unplug. The testing focused on the more challenging probe and drogue refueling method. The Autonomous Air-borne Refueling Demonstration (AARD) program completed over forty-five fully autonomous probe to drogue engagements behind a commercial B-707 tanker, under a variety of test conditions to include engagements and actual fuel transfer in a turn.