Multi-layer Perceptron and Pruning


Nouvel article accepté dans la revue "Turkish Journal of Forecasting"
A Multi-Layer Perceptron (MLP) defines a family of artificial neural networks often
used in TS modeling and forecasting. Because of its “black box” aspect, many
researchers refuse to use it. Moreover, the optimization (often based on the exhaustive
approach where “all” configurations are tested) and learning phases of this artificial
intelligence tool (often based on the Levenberg-Marquardt algorithm; LMA) are
weaknesses of this approach (exhaustively and local minima). These two tasks must be
repeated depending on the knowledge of each new problem studied, making the process,
long, laborious and not systematically robust. In this short communication, a pruning
process is presented. This method allows, during the training phase, to carry out an
inputs selecting method activating (or not) inter-nodes connections in order to verify if
forecasting is improved. We propose to use iteratively the popular damped least-squares
method to activate inputs and neurons. A first pass is applied to 10% of the learning
sample to determine weights significantly different from 0 and delete other. Then a
classical batch process based on LMA is used with the new MLP. The validation is done
using 25 measured meteorological TS and cross-comparing the prediction results of the
classical LMA and the 2-stage LMA


Rédigé le Jeudi 22 Juin 2017 à 09:43

A Short Synthesis Concerning Biological Effects and Equivalent Doses in Radiotherapy


Nouvel article dans la revue "Journal of Radiology and Oncology"
The limits of classical equivalent computation based on time, dose, and fractionation (TDF) and linear quadratic models have been known for a long time. Medical physicists and physicians are required to provide fast and reliable interpretations regarding the delivered doses or any future prescriptions relating to treatment changes. In this letter, we propose an outline related to the different models usable for equivalent and biological doses that are likely to be the most appropriate. The used methodology is based on: the linear-quadratic-linear model of Astrahan, the repopulation effects of Dale, and the prediction of multi-fractionated treatments of Thames.


Rédigé le Jeudi 22 Juin 2017 à 09:19

-Tilos, the first autonomous renewable green island in Mediterranean: A Horizon 2020 project
-Solar potential for building integrated solar collectors: Application in Bulgaria, Romania & France
-Bounded global irradiation prediction based on multilayer perceptron and time series formalism


Participation au conngrés "15-th INTERNATIONAL CONFERENCE ON ELECTRICAL MACHINES, DRIVES, At Sofia (Bulgaria)"
Grâce a ce congrès nous avons pu développer et exposer nos dernières avancées concernant la prévision du rayonnement solaire global relatives au projet TILOS

Rédigé le Jeudi 22 Juin 2017 à 09:08

Conference: IRCC; 4th International Conference on Energy, Sustainability and Climate Change, At Santorini (Greece)


Participation à un congés en Grèce: "Kalman filtering and classical time series tools for global irradiation forecasting"
Cette conférence nous permet de présenter nos derniers travaux publier en 2017 relatifs à la prévision du l'irradiation solaire avec une méthode dite "trainless" qui ne nécessite pas d'historique d'apprentissage (à opposer aux méthodes de machine learning et de data driven)





Rédigé le Jeudi 22 Juin 2017 à 08:58

Global irradiation forecasting based on multilayer perceptron


European energy dependence costs 350 billion Euros annually. For that Europe built an European strategy for energy with three objectives that fall under the security of supply, competitiveness and sustainability. This aims is to ensure citizens and EU companies a safe and environmentally friendly energy at an affordable price. For this, the share of RES in Europe will rise from 14.1 % up to 27% by 2030 .

In this context, scientific research is directed towards an intelligent energy management at various levels from housing to city or regional scale.

At each level, you must have access to energy-efficient items, including intelligent management of stationary or mobile equipment, whether consumers, producers or storage of electricity included in renewable energy smart grids.

An example of the concept of intelligent building is the integration of energy management solutions in housing and business buildings, especially to achieve positive energy and consequently optimally manage this energy on a larger scale.

In this approach, the need to observe, analyze and control physical phenomena at these different levels requires environmental and urban metrology applications. This new way of looking metrology, by detecting a phenomenon at various points scattered on a site, inevitably leads to new technological issues (type and number of sensors, energy independence, need communications), new scientific issues (smart initiatives, sensor cooperation, humans interaction, connected objects).

Providing answers to these concepts opens up many opportunities for innovative applications to make smart building or smart cities. Not only the concept of smart building is a major concern for environmental issues (optimization of energy consumption) but the smart building must also meet the needs of user comfort by providing a set of individual services. Finally, the definition of uses, the design of an adequate network of sensors, data collection and fusion of sensor data, the implementation of a human interaction network interface are among the key parameters to be addressed to meet socio- economic issues.

This multidisciplinary and interdisciplinary approach is positioning itself on the border of two complementary areas. The innovation is the coupling between the BCIM (Building Information Modeling) and sensor networks for the development of business intelligence processes in Smart Building. This approach is part of a commitment in order to upgrading the technology platform “Paglia-Orba” implemented at Ajaccio (University of Corsica/CNRS).



Rédigé le Jeudi 22 Juin 2017 à 08:52

-Machine Learning methods for solar radiation forecasting: a review (Renewable energy)

-Uncertainties in global radiation time series forecasting using machine learning: The multilayer perceptron case (Energy)

-Forecasting method for global radiation time series without training phase: comparison with other well-known prediction méthodologies (Energy)


Projet TILOS: trois articles acceptés dans les revues "Energy" et "Renewable Energy"

Rédigé le Lundi 2 Janvier 2017 à 11:02

Mean absolute log-return and solar radiation forecastabity
Cyril Voyant, Ted Soubdhan, Philippe Lauret, Mathieu David, Marc Muselli


Papier accepté pour la conférence ISES Solar World Congress 2015
Lors de cette conférence nous exposerons les principaux résultats de nos travaux publiés dans l'article Statistical parameters as a means to a priori assess the accuracy of solar forecasting models. Le résumé est:
"In this paper we propose to determinate and to test a set of statistical parameters (20) to estimate the predictability of the global horizontal irradiation time series and thereby propose a new prospective tool indicating the expected error regardless the forecasting methods a modeller can possibly implement. The mean absolute log return, which is a tool usually used in econometry, proves to be a very good estimator. Some examples of the use of this tool are exposed, showing the interest of this statistical parameter in concrete cases of predictions or optimizations."
ises.pdf ISES.pdf  (483.2 Ko)

Rédigé le Vendredi 24 Juillet 2015 à 08:28

Statistical parameters as a means to a priori assess the accuracy of solar forecasting models
Cyril Voyant, Ted Soubdhan, Philippe Lauret, Mathieu David, Marc Muselli


Nouveau papier accepté dans la revue Energy Journal
Dans ce papier , nous démontrons qu'il existe des paramètres statistiques (tel le rendement logarithmique) qui permettent d'estimer l'erreur de prédiction relative à une série temporelle spécifique, et ce quelque soit la méthode de machine learning utilisée. A travers 3 exemples simples, nous proposons une méthodologie d'utilisation de ces paramètres. Le résumé est:

"In this paper we propose to determinate and to test a set of statistical parameters (20) to estimate the predictability of the global horizontal irradiation time series and thereby propose a new prospective tool indicating the expected error regardless the forecasting methods a modeller can possibly implement. The mean absolute log return, which is a tool usually used in econometry, proves to be a very good estimator. Some examples of the use of this tool are exposed, showing the interest of this statistical parameter in concrete cases of predictions or optimizations."

discriminent_param.pdf discriminent_param.pdf  (836.37 Ko)

Rédigé le Vendredi 24 Juillet 2015 à 08:18

Energétique, génie des procédés (62° CNU)


Nous recherchons des candidats pour une thèse débutant en septembre 2015. Cette thèse sera rattachée à l'UMR CNRS SPE 6134 (Projet Structurant ENR) de l'université de Corse. L'étudiant sera basé sur Ajaccio.

Modélisation stochastique et prévisions numériques du rayonnement solaire pour le pilotage de la plateforme hybride de R&D MYRTE.

Résumé: Lors de cette thèse, l’étudiant aura la tâche de manipuler des séries temporelles, des outils de modélisation stochastique (ANN, SVM, etc.), des sorties de modèles globaux (ECMWF, AROME, etc.), et des modèles numériques régionaux haute résolution (WRF ou Méso-NH) dans le but de développer un outil de prévision du rayonnement solaire à différents horizons, intégré au système de contrôle-commande de la plateforme PV MYRTE, pour la production d’énergie électrique sur un réseau non-interconnecté.

Contact:
-MUSELLI Marc, PRU, directeur de thèse
marc.muselli@univ-corse.fr / 06 08 07 05 01
-VOYANT Cyril, Dr, (PAST – adossement Projet ENR), co-directeur de thèse
voyant@univ-corse.fr


Rédigé le Dimanche 19 Avril 2015 à 21:18

A benchmarking of machine learning techniques for solar radiation forecasting in an insular context
Philippe Lauret, Cyril Voyant, Ted Soubdhan, Mathieu David, Philippe Poggi


Ce papier est le fruit de la collaboration de 3 universités Françaises présentant une caractéristique commune : l'insularité. En effet, que ce soit à La Réunion, la Guadeloupe ou encore la Corse, la problématique liée à l'approvisionnement énergétique est dominante. Dans ce papier une large revue des méthodes de prévision dites "Machine Learning" est proposée et des tests de prévision de rayonnement global solaire sont réalisés sur les trois iles. Le but étant de déterminer les modèles les plus performants et surtout d'essayer d'établir des règles de sélection de modèle en fonction du site et des caractéristiques de l'étude. Le résumé de cet article est :
"In this paper, we propose a benchmarking of supervised machine learning techniques (neural networks, Gaussian processes and support vector machines) in order to forecast the Global Horizontal solar Irradiance (GHI). We also include in this benchmark a simple linear autoregressive (AR) model as well as two naive models based on persistence of the GHI and persistence of the clear sky index (denoted herein scaled persistence model). The models are calibrated and validated with data from three French islands: Corsica (41.91°N; 8.73°E), Guadeloupe (16.26°N; 61.51°W) and Reunion (21.34°S ; 55.49°E). The main findings of this work are, that for hour ahead solar forecasting, the machine learning techniques slightly improve the performances exhibited by the linear AR and the scaled persistence model. However, the improvement appears to be more pronounced in case of unstable sky conditions. These nonlinear techniques start to outperform their simple counterparts for forecasting horizons greater than one hour."
Nouveau papier accepté dans la revue Solar Energy

Rédigé le Jeudi 1 Janvier 2015 à 16:07
1 2 3 4 5 » ... 7