Search results for "PREDICTION"
showing 10 items of 511 documents
Multi-physical modelling of reverse electrodialysis
2017
Abstract Reverse electrodialysis (RED) is an electrochemical membrane process that directly converts the energy associated with the concentration difference between two salt solutions into electrical energy by means of a selective controlled mixing. The physics of RED involves the interaction of several phenomena of different nature and space-time scales. Therefore, mathematical modelling and numerical simulation tools are crucial for performance prediction. In this work, a multi-physical modelling approach for the simulation of RED units was developed. A periodic portion of a single cell pair was simulated in two dimensions. Fluid dynamics was simulated by the Navier-Stokes and continuity …
Dip Phenomenon in High-Curved Turbulent Flows and Application of Entropy Theory
2018
The estimation of velocity profile in turbulent open channels is a difficult task due to the significant effects of the secondary flow. The present paper investigates the mechanism of the velocity-dip phenomenon, whereby the location of the maximum velocity appears to be below the free surface. Previous studies conducted in straight channels relate the mechanism of the velocity-dip phenomenon to secondary flow induced by anisotropy of turbulence. This work focuses on high-curved channels where the secondary motion, which is also induced by the channel’s curvature, evolves along the bend. The width-to-depth ratio, B/h, is one of the most important parameters that are affecting the secondary …
Application of LSTM architectures for next frame forecasting in Sentinel-1 images time series
2020
L'analyse prédictive permet d'estimer les tendances des évènements futurs. De nos jours, les algorithmes Deep Learning permettent de faire de bonnes prédictions. Cependant, pour chaque type de problème donné, il est nécessaire de choisir l'architecture optimale. Dans cet article, les modèles Stack-LSTM, CNN-LSTM et ConvLSTM sont appliqués à une série temporelle d'images radar sentinel-1, le but étant de prédire la prochaine occurrence dans une séquence. Les résultats expérimentaux évalués à l'aide des indicateurs de performance tels que le RMSE et le MAE, le temps de traitement et l'index de similarité SSIM, montrent que chacune des trois architectures peut produire de bons résultats en fon…
Event-Based Trajectory Prediction Using Spiking Neural Networks
2021
International audience; In recent years, event-based sensors have been combined with spiking neural networks (SNNs) to create a new generation of bio-inspired artificial vision systems. These systems can process spatio-temporal data in real time, and are highly energy efficient. In this study, we used a new hybrid event-based camera in conjunction with a multi-layer spiking neural network trained with a spike-timing-dependent plasticity learning rule. We showed that neurons learn from repeated and correlated spatio-temporal patterns in an unsupervised way and become selective to motion features, such as direction and speed. This motion selectivity can then be used to predict ball trajectory…
Climats tropicaux d'aujourd'hui à demain : variabilité et changements
2008
Present and future tropical climates : their climate variability and their climate changes The oriffttality of tropical climates is re-examined under the light of climate variability and climate change, both present and future . The interannual variability of rainfall is on average larger in the tropics than in the extra-tropics, but it is also spatially more contrasted. Human communities have often responded to this constraint through efficient coping strategies. The regions showing the most variable climate do not necessarily have lower population densities, an indication that in the long run variability is not decisive. However, the recent demographic growth of these regions is often low…
Agrégation de données spatiales hétérogènes pour l’évaluation d’un modèle
2017
National audience
Approche multi-critère pour la caractérisation des adventices
2022
The objective of this thesis is to develop a way to detect weeds in a field using multispectral images, in order to determine which weeds should be eliminated during the current crop cycle and more particularly at the early stages. The multi-criteria approach focuses on the spatial arrangement, the spec- tral signature, the morphology and the tex- ture of the plants located in the plots. This work proposes a method for selecting the best criteria for optimal discrimination for a given setup. Prior to the extraction of these crite- ria, a set of methods was developed in order to correct the errors of the acquisition de- vice, to precisely detect the vegetation and then to identify within the…
Les pourritures racinaires du pois potager, caractérisation et biocontrôle du complexe parasitaire d’origine tellurique incluant Aphanomyces euteiches
2021
Pea root rots are a major concern for pea growing regions around the world. The disease is caused by a parasitic complex composed of many species of fungi and oomycetes of soil origin. In France, the main pathogen is the oomycete Aphanomyces euteiches. The identity of the other components of the parasitic complex and their respective contributions to the disease have never been investigated. No means of control is currently available to effectively control the disease except for a predictive bioassay that is questioned by some users. However, this test allows the avoidance of infested plots, which limits the multiplication of the major pathogen A. euteiches in the soil. In this context, the…
La rétention de sûreté : la première mesure de sûreté privative de liberté depuis le nouveau Code pénal
2015
The French criminal law is characterised by a willingness to prevent recidivism. To give concrete expression to this will, security measures are implemented among which the secure detention. However, the secure detention is not an ordinary measure and gives rises to many debates which, in fact, take up old and recurrent controversies. This measure is based on the assumption that sexual and/or violent criminals show a disease which may suggest a risk of a second offence. As a result, by considering them as sick persons, the legislator asserts that they can be treated. Accordingly, medical care is advised as the only solution to prevent a second offence. However, this disease itself often ind…
Augmented context based on prediction for efficient knowledge reuse of business knowledge in design
2017
Works on the design context augmentation in product design, makes it possible to make awareness designers on their choice in the design for assembly. The presented approach will conclude on the method to develop a informatic system, based on case studies of mechanical design. The purpose of the project is to capture the design intent, in order to provide useful assistance to stakeholders, such as design support, decision making, verification/validation, and data structuring.