Search

SUMMARY The geographical transposition of the results obtained from representative and experimental watersheds is a matter of regressions based upon geomorphological caracteristics of these watersheds. Therefore, the lumped model for the transformation of precipitations into runoff, generally used for small sized basins, must be simplified and standardized in order to be defined, in any case, by the same parameters. Moreover, the elements of the hydrograph, especially the base time, are to be determined by a very strict processus providing with “consistent” estimates. These new necessary conditions lead to see again the methods used till now for scientific purposes in operating the data from small watersheds, to the so-called “standard hydrograph”.

Cet article résume les résultats d'enquêtes sur les difficultés rencontrées par les élèves dans le calcul des multiplications. Ces résultats corroborent les conclusions de la modélisation ergonomique. Il avance quelques propositions destinées à améliorer le calcul « à la plume » des multiplications, enseigné à l'école primaire. En se fondant sur une similitude formelle des algorithmes avec les théories mathématiques l'auteur propose d'étendre cette méthode à l'étude de l'enseignement des mathématiques elles-mêmes. Le lecteur intéressé trouvera de nombreux articles et textes de Guy Brousseau, publiés ou non, sur le site : http://www.guy-brousseau.com

Comment évaluer les ressources en eau emmagasinées dans les dizaines de milliers de petits réservoirs existant dans le Nord-Est du Brésil ? A partir d'un échantillon de 200 réservoirs dont toutes les caractéristiques au sol sont connues, on établit des relations permettant d'éliminer les facteurs "hauteur de retenue" et "formes du thalweg". On peut alors calculer les volumes emmagasinés au seul examen de la surface du réservoir et de la longueur du bassin versant observées sur la carte. Sur le volume total, l'erreur à craindre ne doit pas excéder 25%. (Résumé d'auteur)

The Petersen method was used to make growth assessments from experimental data collected during 1966-67 and 1969-70. The parameters K and L ∞ were calculated from the Von Bertalanffy growth curve. There was very little difference between the two years although growth in 1969 appeared slightly greater. A comparison of our results in Côte d'Ivoire with those from Senegal and Gulf of Mexico showed that the greatest growth occured on the west African coast and especially off the Côte d'Ivoire.

This report discusses the various conceptions underlying the design of deterministic precipitation/runoff transformation models. After a very brief review of the past development of such models, the author describes what he calls a general physical transformation scheme ; which refers to the various stages of the water circuit in the hydrological cycle of a catchment area. This scheme can be summed up as a number of exchanges taking place between reservoirs containing various storage levels likely to arise in real life, with which losses and flow lag can be associated as appropriate. A precipitation/runoff transformation model is simply the translation by means of methematical expressions of the relationships between some or all of the items forming part of the overall scheme. Irrespective of the form of the model, there is a series of successive operations which can be summed up as follows, the time step being taken as the operating unit : (i) determination of precipitation occurring during the time step ; (ii) reduction of precipitation ; (iii) construction of a unit hydrograph for each surface/element; (iv) "transporting" the hydrographs to the outlet (distributary); (v) summation of the element hydrographs reaching the outlet, giving the overall discharge. Considered from point of view of the vertical water cycle, the models can be subdivided into storage and runoff models. In the former, all the flows are governed by varyingly complex transfers between individual reservoirs. In the latter, runoff is considered to be the most important factor for the flood hydrograph, whilst the base flow is subjected 10 a much more summary analysis ; precipitation can be reduced either by considering a given absorption capacity or by application of a runoff coefficient. For the horizontal distribution of the transformation the catchment area can either be considered to react as a whole (overall model), or alternatively, each surface element can be taken as reacting individually, giving a hypothetical element hydrograph which is then "transported" to the outlet and added to those of the other surface elements to give the hydrograph for the whole catchment area. Outstanding among these models are the matrix models in which the catchment area is divided up on the basis of a characteristic matrix formed by the intersection of Thiessen polygons related to the various rain gauges and the isochronous network for the catchment area.

This thesis proposes a modeling of the corrective maintenance in stochastic universe. After having detailed the basic elements of this vast subject (reliability, maintainability, spare inventory management, etc.), modeling in the formof a network of queues is developed. An algorithm for calculating the availability of a fleet of systems is then developed. After defining a set of decision variables (number of repairers, levels allocated to spare stocks), an original algorithm is developed to provide the optimal solution minimizing the overall cost of maintenance under a minimum availability constraint to satisfy., Cette thèse propose une modélisation de la maintenance corrective en univers stochastique. Après avoir détaillé les éléments de base de ce vaste sujet (fiabilité, maintenabilité, gestion des stocks de rechange, etc), une modélisation sous forme de réseau de files d'attente est élaborée. Un algorithme de calcul de la disponibilité d'un parc de systèmes est ensuite développé. Après avoir défini un ensemble de variables de décision (nombre de réparateurs, niveaux alloués aux stocks de rechanges), un algorithme original est élaboré permettant de fournir la solution optimale minimisant le coût global de la maintenance sous une contrainte de disponibilité minimale à satisfaire.