How to take into account general and contextual knowledge for interactive aiding design: Towards the coupling of CSP and CBR approaches

The goal of this paper is to show how it is possible to support design decisions with two different tools relying on two kinds of knowledge: case-based reasoning operating with contextual knowledge embodied in past cases and constraint filtering that operates with general knowledge formalized using constraints. Our goals are, firstly to make an overview of existing works that analyses the various ways to associate these two kinds of aiding tools essentially in a sequential way. Secondly, we propose an approach that allows us to use them simultaneously in order to assist design decisions with these two kinds of knowledge. The paper is organized as follows. In the first section, we define the goal of the paper and recall the background of case-based reasoning and constraint filtering. In the second section, the industrial problem which led us to consider these two kinds of knowledge is presented. In the third section, an overview of the various possibilities of using these two aiding decision tools in a sequential way is drawn up. In the fourth section, we propose an approach that allows us to use both aiding decision tools in a simultaneous and iterative way according to the availability of knowledge. An example dealing with helicopter maintenance illustrates our proposals.     

A modular Petri net to modeling and scenario analysis of a network of road traffic signals

The dynamic behavior of a group of traffic signals controlling a network of intersections is a complex discrete event system that can be modeled by Petri nets. The approach used in this paper proposes a components-based design, which increases modularity, reduces complexity and is a good practice according to modern Systems Engineering. The main system elements are specified based on the proposed Petri net component with time intervals associated to places. The specified models are simulated through the common token player algorithm, and formal analysis using invariants and theorem proving are applied to verify models' soundness and to reason on specific scenarios.     

Application de l'analyse spectrale autoregressive AU traitement de mesures de turbulence

The Autoregressive (AR) Spectral analysis (also called Maximum Entropy Method) is applied here to velocity turbulence measurements. The validity of such a spectral analysis is tested for classical turbulence data collected in a wind tunnel. The AR estimates are compared with classical Fourier estimates (average of short modified periodograms). Both Levinson's and Burg's algorithms are used for calculating the AR coefficient estimates, and Akaike's “FPE” criterion for choosing the length of the AR filter.Such a method shows much interest in processing some geophysical fluid dynamics phenomena which often are intermittent on a physical point of view. Power spectrum time evolution of some very short length (1 min.) data of oceanic turbulence — they cannot be analysed by Fourier techniques — is a fine example.

Abstract

L'Analyse Spectrale Autorégressive (AR) (dite aussi Méthode de l'Entropie Maximale) est appliquée ici à des mesures de turbulence de vitesse. La validité d'une telle analyse a été testée sur des données classiques de turbulence recueillies en soufflerie. On a comparé les estimations AR avec les estimations classiques (moyenne de courts périodogrammes modifiés). On a utilisé pour l'estimation des coefficients AR les algorithmes de Levinson et de Burg, et pour le choix de l'ordre du filtre AR le critère “FPE” d'Akaike.Ce genre de méthode présente un grand intérêt dans le cas du traitement de certains phénomènes de mécanique des fluides géophysiques qui sont souvent intermittents du point de vue physique. Le calcul de l'évolution spectrale de très courtes (1 mn) données de turbulence océanique en est un bon exemple (ces données ne peuvent être analysées par les techniques de Fourier).     

Onde quasi-TEM des guides cylindriques blindés

tem approximation technique is widely used in the study of waveguides characterized by the fundamental mode without cutoff frequency. It seems that exact justification of this approximation has never been delt with before to his knowledge. Here the author is trying to show the justification for a closed waveguide noting the difference between a quasi-TEM wave and an ordinary tem wave.     

A new sarpedobilin-containing butterflyPapilio graphium stresemani stresemani and its bioecological situation within the species

The doubly cyclized bile pigment sarpedobilin (3) represents a unique case in nature. In spite of being present in trace amounts in many butterfly and moth species containing pterobilin (1) or phorcabilin (2) as the main bilin, this substance occurs as the main pigment in a restricted number of species geographically limited to Southeast Asia. We have isolated and identified sarpedobilin3 (as the dimethyl ester) as the main bile pigment fromPapilio graphium stresemani stresemani (from Ceram island, Indonesia) in the amount of ca. 0.1 mg per individual, double that found inPapilio sar-pedon orP. weiskei. We conclude thatP. sarpedon (and subspecies)P. weis-kei, andP. stresemani represent, in terms of evolution (see article), a group of closely related species. By comparison of colored wing patterns and resistance of the chromoproteins to hydrolysis by MeOH-H₂SO₄,P. stresemani is nearP. weiskei. The bioecological relationships of the three sarpedobilin-containing species is discussed.     

Optical field concentration in low-index waveguides

We present a highly efficient optical field concentrator that is capable of confining optical field in nanometer-thin low-index media with very high optical confinement factor. The structure is made of multiple-layered low-index nanolayers embedded in high-index silicon waveguides. By creating multiple high-index-contrast interfaces, the normal field in the low-index nanolayer regions is significantly enhanced. It subsequently results in a very high optical confinement and power density in these regions. With the help of numerical simulation tools, the guiding and propagating characteristics of the new structure are studied and presented. The optimal structures have demonstrated confinement factors and normalized power densities in the range of 30%-60% and 20-160 /spl mu/m/sup -2/ for the 5-20-nm thin low-index multiple nanolayers.     

Progress in reliability research in the micro and nano region

Due to the rapid development of IC technology the traditional packaging concepts are making a transition into more complex system integration techniques in order to enable the constantly increasing demand for more functionality, performance and miniaturisation. These new concepts will have to combine smaller structures and layers made of new materials with even higher reliability. As these structures will more and more display nano-features, a coupled experimental and simulative approach has to account for this development to assure design for reliability in the future. A necessary “nano-reliability” approach as a scientific discipline has to encompass research on the properties and failure behaviour of materials and material interfaces under explicit consideration of their nano-structure and the effects hereby induced. It uses nano-analytical methods in simulation and experiment to consistently describe failure mechanisms on that length scale for more accurate and physically motivated lifetime prediction models for use on a larger (i.e. then the micro) scale. This paper deals with the thermo-mechanical reliability of microelectronic components and systems and methods to analyse and predict it. Various methods are presented to enable lifetime prediction on system, component and material level, the latter introducing the field of nano-reliability for nano-packaging in advanced electronics system integration.     

Two‐phase thermo‐mechanical and macrosegregation modelling of binary alloys solidification with emphasis on the secondary cooling stage of steel slab continuous casting processes

As an approach towards a better modelling of solidification problems, we introduce a thermo‐mechanical and macrosegregation model that considers a solidifying alloy as a binary mixture made of a liquid and a solid phase. Macroscopic conservation laws for mass, momentum and solute are obtained by spatial averaging of the respective microscopic conservation equations. Assuming local thermal equilibrium, a single equation for the conservation of the mixture energy is then written. A single equation can be obtained for the solute as well by invoking a proper microsegregation rule. The numerical implementation in a two‐dimensional finite element code is then detailed. Lastly, some examples of simulations of academic tests as well as industrial applications for continuous casting of steel slabs are discussed. They particularly enlighten the ability of the formulation to describe the formation of central macrosegregation during the secondary cooling of slab continuous casting processes. Copyright © 2006 John Wiley & Sons, Ltd.     

Methodology for the reliability evaluation of a passive system and its integration into a Probabilistic Safety Assessment

A methodology has been developed to evaluate the reliability of passive systems characterised by a moving fluid and whose operation is based on thermal–hydraulic (T-H) principles. The methodology includes:

• identification and quantification of the sources of uncertainties and determination of the important variables;

• propagation of the uncertainties through T-H models and assessment of T-H passive system unreliability;

• introduction of passive system unreliability in the accident sequence analysis.

Each step of the methodology is described and commented and a diagram of the methodology is presented. An example of passive system is presented with the aim to illustrate the possibilities of the methodology. This example is the Residual Passive heat Removal system on the Primary circuit (RP2), an innovating system supposed to be implemented on a 900 MWe Pressurized Water Reactor.     

Data-parallel tomographic reconstruction: A comparison of filtered backprojection and direct Fourier reconstruction

We consider the parallelization of two standard 2D reconstruction algorithms, filtered backprojection and direct Fourier reconstruction, using the data-parallel programming style. The algorithms are implemented on a Connection Machine CM-5 with 16 processors and a peak performance of 2 Gflop/s.