Constraint based approach for the steady-state simulation of complex systems: Application to ship control

When the steady states are largely predominant with respect to transitional phases, steady-state simulation seems sufficient to predict the behavior of a complex system. Over the past 20 years, different modeling languages and dedicated tools have been developed to improve steady state simulation.In this paper, focus is made on steady-state simulation for system control and design. A model combining an emission sub-model with a ship propulsion sub-model was implemented in a constraint programming (CP) approach. It will help to determine the efficiency (i.e. the ability to model and solve the problem) and complexity of implementation (i.e. difficulties encountered during the implementation) of this approach.First, requirements for the steady-state simulation of complex systems are defined. Then, CP approach is shown to be able to answer these issues through experiments. This approach is then compared to one of the main simulation languages: Modelica.Although the two approaches (i.e Modelica and CP) are able to reverse models, the study shows that the use of Modelica principles for steady-state simulation involves some crippling limitations, such as the non-management of under/over-constrained systems, or inequalities.This study also shows that the constraint programming approach permits to meet some needs for steady-state simulation not yet covered by current approaches.     

Time series classification for the prediction of dialysis in critically ill patients using echo statenetworks

Objective: Time series often appear in medical databases, but only few machine learning methods exist that process this kind of data properly. Most modeling techniques have been designed with a static data model in mind and are not suitable for coping with the dynamic nature of time series. Recurrent neural networks (RNNs) are often used to process time series, but only a few training algorithms exist for RNNs which are complex and often yield poor results. Therefore, researchers often turn to traditional machine learning approaches, such as support vector machines (SVMs), which can easily be set up and trained and combine them with feature extraction (FE) and selection (FS) to process the high-dimensional temporal data. Recently, a new approach, called echo state networks (ESNs), has been developed to simplify the training process of RNNs. This approach allows modeling the dynamics of a system based on time series data in a straightforwardway.The objective of this study is to explore the advantages of using ESN instead of other traditional classifiers combined with FE and FS in classification problems in the intensive care unit (ICU) when the input data consists of time series. While ESNs have mostly been used to predict the future course of a time series, we use the ESN model for classification instead. Although time series often appear in medical data, little medical applications of ESNs have been studiedyet.Methods and material: ESN is used to predict the need for dialysis between the fifth and tenth day after admission in the ICU. The input time series consist of measured diuresis and creatinine values during the first 3days after admission. Data about 830 patients was used for the study, of which 82 needed dialysis between the fifth and tenth day after admission. ESN is compared to traditional classifiers, a sophisticated and a simple one, namely support vector machines and the naive Bayes (NB) classifier. Prior to the use of the SVM and NB classifier, FE and FS is required to reduce the number of input features and thus alleviate the curse dimensionality. Extensive feature extraction was applied to capture both the overall properties of the time series and the correlation between the different measurements in the time series. The feature selection method consists of a greedy hybrid filter-wrapper method using a NB classifier, which selects in each iteration the feature that improves prediction the best and shows little multicollinearity with the already selected set. Least squares regression with noise was used to train the linear readout function of the ESN to mitigate sensitivity to noise and overfitting. Fisher labeling was used to deal with the unbalanced data set. Parameter sweeps were performed to determine the optimal parameter values for the different classifiers. The area under the curve (AUC) and maximum balanced accuracy are used as performance measures. The required execution time was also measured.Results: The classification performance of the ESN shows significant difference at the 5% level compared to the performance of the SVM or the NB classifier combined with FE and FS. The NB+FE+FS, with an average AUC of 0.874, has the best classification performance. This classifier is followed by the ESN, which has an average AUC of 0.849. The SVM+FE+FS has the worst performance with an average AUC of 0.838. The computation time needed to pre-process the data and to train and test the classifier is significantly less for the ESN compared to the SVM andNB.Conclusion: It can be concluded that the use of ESN has an added value in predicting the need for dialysis through the analysis of time series data. The ESN requires significantly less processing time, needs no domain knowledge, is easy to implement, and can be configured using rules ofthumb.     

Adapting groupware systems to changes in the collaborator’s context of use

Owing to the dynamic nature of collaborative environments, the software intended to support collaborative work should adapt itself to the different situations that may occur. This requirement is related to the concept of “context of use”, which has been considered as an important aspect in the design of interactive systems. Nevertheless, two main problems about this concept have been identified by current research in context-aware computing: (1) most of the studies have mainly focused on the context of a single user, so the context of multiple users involved in a common endeavor remains little explored, and (2) adaptability in context-aware systems generally takes into account a reduced number of contextual variables (mainly the user’s location and platform). In this paper, we firstly re-conceptualize the notion of “context of use”, in order to consider the main characteristics of collaborative environments. Based on this new notion, we then design and implement a framework that allows application developers to specify the adaptability of groupware systems in terms of the state of activities, roles, collaborators’ location, available resources, and other typical variables of working groups. This framework has been generalized from scenarios that highlight dynamic situations presented in real collaborative settings. Finally, we validate our proposal by a set of applications that are able to adapt their user interface and functionality, when significant changes are produced in the environment, the working group, and/or the used devices.     

High and very high temperature reactor research for multipurpose energy applications

Ten years ago, the European High Temperature Reactor (HTR) Technology Network (HTR-TN) launched a programme for developing HTR Technology, which expanded so far through 4 successive Euratom Framework Programmes. Many projects have been performed - in particular the RAPHAEL project in the 6th Euratom Framework Programme and presently ARCHER in the 7th - in line with the Network strategy that identified cogeneration of process heat and power as the main specific mission of HTR. HTR can indeed address the growing energy needs of industry presently fully relying on fossil fuel combustion with a CO₂-lean generation technology, thanks to its high operating temperature and to its unique flexibility obtained from its large thermal inertia and its low power.Relying on the legacy of the former European leadership in HTR technology, this programme has addressed specific developments required for industrial process heat applications and for increasing HTR performances (higher temperatures and fuel burn-up). Decisive achievements have been obtained concerning fuel manufacturing and irradiation behaviour, key components and their materials, safety, computer code validation and specific HTR waste (fuel and graphite) management. Key experiments have been performed or are still ongoing: irradiation of graphite, fuel and vessel materials and the corresponding post-irradiation examinations, safety tests and isotopic analyses; thermal-hydraulic tests of an Intermediate Heat Exchanger mock-up in helium; air ingress experiments for a block type core, etc. Through Euratom participation in the Generation IV International Forum (GIF), these achievements contribute to international cooperation.HTR-TN strategy has been recently integrated by the “Sustainable Nuclear Energy Technology Platform” (SNE-TP) as one of the 3 “pillars” of its global nuclear strategy. It is also in line with the orientations and the timing of the “Strategic Energy Technology Plan (SET-Plan)” for the development of CO₂-lean energy technologies, and thus strengthens the nuclear option in a future European energy mix.Nuclear cogeneration for industrial process heat applications is a major innovation and a major challenge, requiring large-scale demonstration to prove its industrial viability. To enable demonstration, it is necessary not only to develop an appropriate nuclear heat source, but also to develop coupling technologies and to adapt industrial processes to the coupling with a HTR. This requires a close partnership between the conventional and the nuclear technology holders as the base of a Nuclear Cogeneration Industrial Initiative.Recently the project EUROPAIRS initiated by HTR-TN together with process heat user industries has set the bases of such a strategic partnership.     

Structure and Reactivity of Spruce Mechanical Pulp Lignins Part I. Bleaching and Photoyellowing of in situ Lignins

This paper reports the structural investigation of a series of in situ spruce lignins in wood, thermomechanical pulp (TMP) and the corresponding bleached (BTMP) and photoyellowed (YBTMP) samples. This was achieved by using an original two-step degradative technique, thioacidolysis followed by Raney nickel desulfuration. The determination of thioacidolysis monomeric and dimeric products allowed an estimate of various lignin building units and interunit bonds, respectively. It was observed that the thermomechanical and bleaching treatments did not affect spruce lignin structure to an appreciable extent. On the contrary, the photoyellowing treatment caused marked structural changes in lignin, particularly when run in severe conditions. These changes were essentially a decrease in the amount of β-O-4 and β-1 interunit bonds and a relative increase in catechol units and in vanillin end-groups.     

Mid-infrared spectroscopy as a tool for rapid determination of internal quality parameters in tomato

The objective of this study was to investigate the possibility of predicting the quality parameters of tomato by mid-infrared spectroscopy. For 2 years, tomato samples, representing a large variability in the chemical composition, were scanned using the attenuated total reflectance accessory of a Fourier transform spectrometer in the wavenumber region between 4000 and 400 cm⁻¹_. Calibration models were developed using partial least squares (PLS) regression method and were tested with internal validation sample set in the first year. Different spectral preprocessing techniques were investigated and different spectral regions were selected to optimise the calibration models. In addition, the models obtained in 2007 were used to predict the soluble solids, dry matter and total acidity in tomato harvested in 2008.     

Exploring the bibliometric and semantic nature of negative results

Negative results are not popular to disseminate. However, their publication would help to save resources and foster scientific communication. This study analysed the bibliometric and semantic nature of negative results publications. The Journal of Negative Results in Biomedicine (JNRBM) was used as a role model. Its complete articles from 2002–2009 were extracted from SCOPUS and supplemented by related records. Complementary negative results records were retrieved from Web of Science in “Biochemistry” and “Telecommunications”. Applied bibliometrics comprised of co-author and co-affiliation analysis and a citation impact profile. Bibliometrics showed that authorship is widely spread. A specific community for the publication of negative results in devoted literature is non-existent. Neither co-author nor co-affiliation analysis indicated strong interconnectivities. JNRBM articles are cited by a broad spectrum of journals rather than by specific titles. Devoted negative results journals like JNRBM have a rather low impact measured by the number of received citations. On the other hand, only one-third of the publications remain uncited, corroborating their importance for the scientific community. The semantic analysis relies on negative expressions manually identified in JNRBM article titles and abstracts and extracted to syntactic patterns. By using a Natural Language Processing tool these patterns are then employed to detect their occurrences in the multidisciplinary bibliographical database PASCAL. The translation of manually identified negation patterns to syntactic patterns and their application to multidisciplinary bibliographic databases (PASCAL, Web of Science) proved to be a successful method to retrieve even hidden negative results. There is proof that negative results are not only restricted to the biomedical domain. Interestingly a high percentage of the so far identified negative results papers were funded and therefore needed to be published. Thus policies that explicitly encourage or even mandate the publication of negative results could probably bring about a shift in the current scientific communication behaviour.     

Effects of Botrytis cinerea infection on Champagne wine foaming properties

Qualitative effects of Botrytis cinerea infection were studied by comparing the foaming properties of Champagnes obtained from healthy grapes with those from grapes infected by B cinerea. This is of particular interest when it is considered that the foaming properties of Champagne wines are important in terms of product attractiveness for the consumer. Experiments using artificial viewing equipment clearly showed the dramatic effects on Champagne foam characteristics when grape berries were highly infected. The speed at which liquid separated from the foam (expressed as liquid height L_S) in the glass depended largely on the level of infection for Chardonnary and Pinot noir wines (+268% for Chardonnary wines at 40% infection and +627% for Pinot noir wines at 20% infection). B cinerea infection reduced the time (L_T) at which liquid first appeared in the glass; for Pinot noir wines an infection level of 20% caused very rapid drainage. The time of pouring (P_T, ie the time to fill the glass under controlled conditions) also diminished considerably at an infection level of 20% for both Pinot noir and Pinot meunier wines (?74 and ?58% respectively). Mouldiness considerably altered the foam height observed 80 s (H₈₀) after the start of pouring (?83% for Pinot noir and ?89% for Pinot meunier at 20% infection). Champagne foamability was also quantified using a sparging technique to standardise effervescence. Foamability suffered considerably (?60 to ?65%) when the B cinerea infection level was 20% as compared to control Champagnes. All wines studied had low and very similar protein contents. Differences in the concentrations of these compounds, which are generally implicated in the formation of foam in sparkling wines, could not explain the differences in Champagne foaming properties observed here. © 2001 Society of Chemical Industry