Browsing
     by title


0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

or enter first few letters:   
OK
Full Text
Peer Reviewed
See detailApprentissage par renforcement appliqué à la commande des systèmes électriques
Dai, Jing; Phulpin, Yannick; Vannier, Jean-Claude et al

in Proceedings of "Les Journées Electrotechnique du Futur 2009" (2009)

Cet article propose une revue de littérature concernant les applications de l’apprentissage par renforcement à la commande des systèmes électriques. L'apprentissage par renforcement a pour caractéristique ... [more ▼]

Cet article propose une revue de littérature concernant les applications de l’apprentissage par renforcement à la commande des systèmes électriques. L'apprentissage par renforcement a pour caractéristique principale de résoudre des problèmes de commande optimale à partir de la seule observation des trajectoires du système. Il présente l’intérêt de ne pas requérir de connaissance à priori sur la dynamique du système à commander et convient ainsi aux problèmes de commande des systèmes complexes. Dans un premier temps, l’article détaille les caractéristiques des problèmes auxquels l’apprentissage par renforcement s’applique, puis cette technique est décrite. Ensuite, deux exemples classiques d’application aux systèmes électriques sont présentés. [less ▲]

Detailed reference viewed: 45 (2 ULg)
See detailL'apprentissage précoce des langues étrangères par la méthode 'immersive'
Comblain, Annick ULg; Rondal, Jean-Adolphe ULg

in Rééducation Orthophonique (1993), 31

Detailed reference viewed: 99 (15 ULg)
Full Text
Peer Reviewed
See detailL'apprentissage sans erreur
Richelle, Marc ULg

in Année Psychologique (L') (1966), 66(2), 535-543

Detailed reference viewed: 20 (2 ULg)
See detailL'apprentissage télématique des langues
Defays, Jean-Marc ULg; Doppagne, Véronique ULg; Thonard, Audrey ULg

Scientific conference (2007, May)

Detailed reference viewed: 22 (7 ULg)
See detailApprentissage, enseignement: pile ou face?
Defays, Jean-Marc ULg

Scientific conference (2003, June)

Detailed reference viewed: 15 (3 ULg)
See detailL'apprentissage
Richelle, Marc ULg

in Pelicier, Yves (Ed.) L'univers de la psychologie: La vie psychologique normale (1977)

Detailed reference viewed: 2 (0 ULg)
Full Text
See detailApprentissages automatiques supervisés pour le monitoring environnemental et énergétique d'une chaudière de régénération
Sainlez, Matthieu ULg

Doctoral thesis (2012)

The broad objective of this thesis is to apply and compare supervised learning techniques for prediction of nitrogen oxide pollutant emission from the recovery boiler of a Kraft pulp mill. In this task ... [more ▼]

The broad objective of this thesis is to apply and compare supervised learning techniques for prediction of nitrogen oxide pollutant emission from the recovery boiler of a Kraft pulp mill. In this task, we want to highlight a technique that is most suited and self-adapted to boiler transient operating conditions. The Kraft process is an alkaline process to produce chemical pulp; cellulose fibers are dissociated from lignin by cooking the chips in a solution of sodium hydroxide (NaOH) and sodium sulfide (Na2S), called white liquor. The residual black liquor is concentrated and burned in a recovery furnace to yield an inorganic smelt of sodium carbonate (Na2CO3) and Na2S. The recovery boiler both regenerates the cooking chemicals and produces high pressure steam to the pulp mill, but the boiler is a major source of atmospheric pollutants in the mill. In particular nitrogen oxide formation is very complex because of several chemicals and dynamic mechanisms: thermal NOx, prompt NOx and fuel NOx. Nowadays, there is an increasing demand in such industries for efficient data analysis tools, especially for pollutant monitoring and/or energy management. Literature reviews refer mainly on numerical solutions where a complete description of the process is needed and where stationary condition is often a working hypothesis. This is the case with the advanced data validation and reconciliation techniques that we evaluate. This technique is based on thermodynamic models, chemical and physical relationships within process parameters and equipment. This is helpful to highlight some lack of information about the process, but this approach failed to model accurately steam and fumes utilities operating points. Indeed, in a Kraft recovery boiler, the total nitrogen oxide emission is dependent on several operating factors and heterogeneous conditions, e.g. operating fuels (black liquor or heavy fuel), furnace load, droplet size, air system operation, retention time, biomass characteristics,... For such a complex problem, machine learning techniques may be used as alternative methods in engineering analysis and predictions. They involve algorithms that improve automatically through experience collected in historical databases. Among supervised learning techniques, we focus mainly on neural networks methods (static and dynamic architectures) and additionally on tree-based (regression tree and random forests) and linear ones. For each method, we evaluate its ability to predict NOx pollutant emission in varying conditions. A random forest is a collection of uncorrelated regression trees, induced from bootstrap samples of the training data. Its internal estimates are also used to measure variable importance and allow us to classify relevant variables for a model inputs selection task. Note that we need some additional a priori knowledge to select the final inputs set. Among static neural network structures, the multilayer perceptron is the most widely used, particularly the two-layer structure in which the input units and the output layer are interconnected with an intermediate hidden layer. The model of each neuron in the network includes a nonlinear activation function that is differentiable; this network can perform static mapping between an input space and an output space. Within dynamic architectures, we distinguish those that have only feed-forward connections and those that have feedback (recurrent) connections. In this work, we focus mainly on NARX network (Nonlinear AutoRegressive model with eXogenous inputs) and additionally on Elman recurrent neural network. This last one incorporates an additional layer, called context layer, the nodes of which are the one-step delay elements embedded into the local feedback paths. Nevertheless, Elman's approach has some drawbacks associated with learning parameters scheme and temporal gradient approximation. Particularly, the NARX network is used for input-output modeling of nonlinear dynamical systems. It is a recurrent model: model inputs are applied to a tapped-delay-line memory of n units and outputs are fed back to the input layer through another line of m units. The total model order s=n+m is therefore a key parameter and the method of Lipschitz numbers is a tool for estimating it. An advantage of NARX is that we can use standard backpropagation algorithm for neural network learning scheme. Furthermore, to increase model robustness, we average neural predictions over a set of individual neural predictors, this is helpful for reducing variance prediction across trials. Despite the fact that generalization is done on the worst case configuration possible, we see that ensemble of NARX networks perform well on predicting NOx emissions during transient operations and Lipschitz numbers are very helpful for system orders estimations. We illustrate the potential of a dynamic neural approach compared to the others in the nitrogen oxide prediction task. It is more suited to practical modeling needs and offers a modeling of time and memory. It allows us to monitor NOx pollution and possibly adjusting control variables and performing diagnostics. The thesis is divided into seven chapters covering several publications. Chapter 1 is about the Kraft process and its recovery boiler. We start with a short description of the Kraft pulp mill. Then we describe the Kraft recovery boiler, some chemical reactions in the furnace, the steam production equipments and the atmospheric pollutants. Finally we discuss about nitrogen oxide formation in the furnace, the effects of several operating conditions on its production. Chapter 2 is about data mining, on what it is, on what it is used for and which are the main modeling cultures. This chapter deals with system identification, modeling approaches (white box, grey box, black box), some definitions about learning and modeling, and finally some links between modeling and optimization techniques. Chapter 3 starts with a state-of-the-art about numerical simulation of a Kraft recovery boiler, then we apply and evaluate a data validation scheme for steam and fumes utilities modeling. Finally we discuss the application of artificial intelligence techniques within the framework of a recovery boiler. Chapter 4 aims at selecting model inputs, starting with a supervised selection approach based on random forests. We introduce some methodological insights about tree-based methods, from a simple regression tree to random forests. Random forests internal estimates are used to measure the relative importance of each input variable in predicting a response, i.e. nitrogen oxide emission or high pressure steam production. Finally we discuss about some useful extra knowledge to take in account for the selection of final inputs. Chapter 5 is about neural networks modeling, we introduce the perceptron, the multilayer perceptron, and the associated backpropagation algorithm. We discuss about static and dynamic architectures, especially the Elman recurrent neural network. Finally, we apply a multilayer perceptron and an Elman recurrent neural network for predicting the high pressure steam flow rate from the Kraft recovery boiler. Chapter 6 presents some insights about input-output modeling of nonlinear dynamical systems, especially with NARX network. At the end, we explain the Lipschitz method that is applied for system orders estimation. Chapter 7 summarizes some comparison results about supervised learning techniques applied to predict nitrogen oxide pollutant emission from the recovery boiler. This comparison involves neural network techniques, tree-based methods and multiple linear regression. Finally, some research perspectives are presented and some conclusions are drawn. [less ▲]

Detailed reference viewed: 38 (7 ULg)
See detailApprentissages élémentaires
Ferrara, André ULg

in Rondal, Jacques (Ed.) Introduction à la psychologie scientifique (1999)

Detailed reference viewed: 31 (4 ULg)
See detailApprentissages et multimédia
Leclercq, Dieudonné ULg; Denis, Brigitte ULg

Scientific conference (1994, April)

Detailed reference viewed: 5 (0 ULg)
Peer Reviewed
See detailApprentissages et multimédia
Denis, Brigitte ULg; Leclercq, Dieudonné ULg

in Noirhomme, Monique (Ed.) Actes de la journée d'information sur le Multimédia (1995)

Detailed reference viewed: 23 (8 ULg)
Full Text
See detailApprentissages et multimédias
Leclercq, Dieudonné ULg; Denis, Brigitte ULg

in Les centres de ressources en langues : méthodologies, mise en valeur, optimisation de leur utilisation (1994, April)

Detailed reference viewed: 25 (4 ULg)
See detailApprentissages et multimédias
Leclercq, Dieudonné ULg; Denis, Brigitte ULg

in Slangen, L. (Ed.) Actes du colloque 'Les centres de ressources en langues. Méthodologies, Mises en valeur, optimisation de leur utilisation' (1995)

Detailed reference viewed: 14 (4 ULg)
See detailL'apprivoisement de la méthode en Belgique, exploration par une équipe Sos-enfants.
Bullens, Quentin ULg

in Alföldi, Françis (Ed.) 18 cas pratiques d'évaluation en action sociale et médico-sociale. (2008)

Detailed reference viewed: 23 (3 ULg)
Full Text
Peer Reviewed
See detailAn approach for dividing models of biological reaction networks into functional units
Ederer, Michael; Sauter, Thomas; Bullinger, Eric ULg et al

in Simulation: Trans. Society for Modeling and Simulation International (2003), 79(12), 703-716

Biological reaction networks consist of many substances and reactions between them. Like many other biological systems, they have a modular structure. Therefore, a division of a biological reaction ... [more ▼]

Biological reaction networks consist of many substances and reactions between them. Like many other biological systems, they have a modular structure. Therefore, a division of a biological reaction network into smaller units highly facilitates its investigation. The authors propose an algorithm to divide an ordinary differential equation (ODE) model of a biological reaction network hierarchically into functional units. For every compound, an activity function dependent on concentration or concentration change rate is defined. After performing suitable simulations, distances between the compounds are computed by comparing the activities along the trajectories of the simulation. The distance information is used to generate a dendrogram revealing the internal structure of the reaction network. The algorithm identifies functional units in two models of different networks: catabolite repression in Escherichia coli and epidermal growth factor (EGF) signal transduction in mammalian cells. [less ▲]

Detailed reference viewed: 11 (1 ULg)
Full Text
Peer Reviewed
See detailApproach of regionalization of low flow of the Walloon Region
Gailliez, Sébastien; Degre, Aurore ULg

Poster (2010, April)

The walloon part of the Meuse’s watershed represents 12283 km² and 17694 km of river. The anthropogenic pressure is important (population density is more or less 200 inhabitants/km²). In low flow period ... [more ▼]

The walloon part of the Meuse’s watershed represents 12283 km² and 17694 km of river. The anthropogenic pressure is important (population density is more or less 200 inhabitants/km²). In low flow period, water scarcity can touch both the water users (producer of drinking water and hydropower, tourism and pleasure activities and kayaking) and the river itself affecting the ‘good environmental state’ (context of Water Framework Directive 2000/60/CE). The operational management of rivers during low flow periods needs a deep knowledge of this drought phenomenon including an analysis of low flow severity and its occurrence probability. It also needs the computation of low flow discharge at any point of a river based on available hydrologic variables. The aim of this study is the low flows’ regionalization in the Walloon Region. First of all, the time series of flow data are filled in and validated. The quality is then controlled. The different tests are the determination of the minimum year requirement for a monitoring site, homogeneity tests, verification of presence or absence of summer alga and the proportion of extrapolation of the discharge rating curve Secondly, homogeneous regions will be defined and regression equations will be build. These equations will establish the relation between low flow and physical parameters (watershed area, pedology, slope), climate ones and/or ground water ones. The regression model will permit the computation of low flow discharge at any point of an ungauged river. [less ▲]

Detailed reference viewed: 28 (9 ULg)
Full Text
Peer Reviewed
See detailAn approach to assess the quality of collaboration in technology-mediated design situations.
Burkhardt, Jean-Marie; Détienne, Françoise; Hebert, Anne-Marie et al

in Proceedings of ECCE 2009 : European Conference on Cognitive Ergonomics (2009, September)

ur objective is to measure and compare the quality of collaboration in technology-mediated design activities. Our position is to consider collaboration as multidimensional. We present a method to assess ... [more ▼]

ur objective is to measure and compare the quality of collaboration in technology-mediated design activities. Our position is to consider collaboration as multidimensional. We present a method to assess quality of collaboration which is composed of seven dimensions concerning communication processes such as grounding, coordination processes, task-related processes, symmetry of individual contributions as well as motivational processes. This method is used in a study aiming to compare the quality of collaboration in architectural design. In this experimental study, design situations vary according to technology-mediation - co-presence with an augmented reality (AR) environment versus distance with AR and visio-conferencing -, and according to number of participants - pairs versus groups of four architects -. Our results show that distinctive dimensions of collaboration are affected by the technology mediation and/or the number of co-designers. We discuss these results with respect to technology affordances such as visibility and group factors. [less ▲]

Detailed reference viewed: 69 (14 ULg)
Peer Reviewed
See detailAn approach to corrective control of voltage instability using simulation and sensitivity
Van Cutsem, Thierry ULg

in IEEE Transactions on Power Systems (1995), 10

Detailed reference viewed: 20 (1 ULg)
See detailAn approach to corrective control of voltage stability using simulation and sensitivity
Van Cutsem, Thierry ULg

in Proc. IEEE NTUA Athens Power Tech conference (1993, September)

Detailed reference viewed: 16 (0 ULg)
Full Text
See detailAn approach to desication-tolerant bacteria in starter culture production
Weekers, F.; Jacques, Ph.; Mergeay, M. et al

in Engineering and manufacturing for biotechnology (2001)

Detailed reference viewed: 3 (1 ULg)