HAL-MINES ParisTech
Not a member yet
27348 research outputs found
Sort by
On the Assessment of Hourly Means of Solar Irradiance at Ground Level in Clear-Sky Conditions by the ERA5, JRA-3Q, and MERRA-2 Reanalyses
No full textInternational audienceMeteorological reanalyses are one of the means to assess the solar irradiance reaching the ground. This paper deals with estimates of the hourly means of irradiance in clear-sky conditions provided by the ERA5, JRA-3Q, and MERRA-2 reanalyses. They are compared to coincident ground-based measurements from 28 BSRN stations located worldwide, selected by a new algorithm for detecting cloud-free instants. Although ERA5 most often underestimates measurements, it is quite reliable over time because it captures the temporal variability of measurements well and provides a constant level of uncertainty. JRA-3Q offers a complex pattern with negative and positive biases depending on station and season. It captures well the temporal variability but, as a whole, is not reliable over time. None of the three reanalyses is reliable in space. Because of its use of the mean solar time instead of the true solar time, MERRA-2 suffers many drawbacks over intraday scales. Its statistical indicators exhibit marked patterns depending on the season and station. Its assimilation of aerosol properties offers advantages when compared to the climatologies used in ERA5 and JRA-3Q. This work exposes the strengths and weaknesses of each reanalysis in clear-sky conditions and formulates suggestions to providers for further improvements.</div
Est-on entré dans un âge de la maintenance des infrastructures ?
No full textInternational audienceFace à la fin de l’ère de l’équipement, un nouveau défi émerge : gérer l’existant. L’investissement massif dans de nouvelles infrastructures cède la place à une approche sur la maintenance et la durabilité. Ce virage, perceptible chez les décideurs, invite à repenser les politiques publiques et à privilégier le soin des infrastructures existantes
Empreinte carbone : tous les barils de pétrole ne se valent pas et cela a son importance pour la transition énergétique
No full textNational audienc
Technical note: High-frequency, multi-elemental stream water monitoring – experiences, feedbacks and suggestions from 7 years of running three French field laboratories (Riverlabs)
No full textInternational audienceHigh-frequency and multi-elemental stream water monitoring are acknowledged as necessary to address data limitation in the fields of catchment sciences and freshwater biogeochemistry. In recent years, the development of stream bank analyzers and on-site field laboratories to measure various solutes and/or isotopes at sub-hourly measurement intervals has been in progress at an increasing number of sites. This trend should likely persist in the future as the technologies are still improving. Here we share our experiences of running three innovative lab-in-the-field prototypes, called Riverlabs, which consist of a field deployment involving continuous sampling and filtration of stream water and its analysis using laboratory instruments such as ion chromatographs. This note gives an overview of the technical and organizational points that we identify as critical because we claim that such practical considerations are generally missing in the literature in order to provide guidelines for the successful implementation of future projects running such or similar field-laboratory setups. We share the main stages in the deployment of this tool in the field, the difficulties encountered and the proposed solutions. Our two main conclusions for a successful, long-term functioning of these types of field laboratories are, first, the necessity to adapt several central components of the field laboratory to the local conditions (climate, river geometry, topography, physico-chemical characteristics of water, power supply) and, second, the need of diverse and in-depth technical skills within the engineering team. The critical aspects discussed here relate to (1) supply of the field laboratory – basic functioning of the pumping, filtration and analytical systems; (2) data quality control and assurance via maintenance services and operations; (3) data harmonization and coordination of the laboratory components; and (4) team structure, skills and organization. We believe that sharing these experiences, combined with providing some practical suggestions, might be useful for colleagues who are starting to deploy such or similar field laboratories. These considerations will save time, improve performance and ensure continuous field monitoring
Unlocking engineering expertise frozen by grand challenges: dealing with transitions' unknown
No full textInternational audienceThe contemporary transition of climate change, technology evolution, and social inequality present profound challenges in project management. It requires multitude of stakeholders who are facing many unknowns on the solution to develop. This study explores how grand challenges impact the nature of unknowns encountered by engineering actors in project management. Through a historical analysis of innovation projects within a century-old automotive manufacturer, we identify distinct types of unknowns—desirable, undesirable, endogenous, and exogenous—and examine the processes required to manage them, including endogenization and desirabilization. We demonstrate that the unknowns generated by grand challenges represent a densification and complexity beyond those traditionally managed, necessitating new knowledge management processes. From a scientific perspective, this research enriches the understanding of unknowns in design activities and adapts this concept to the context of grand challenges. From a practical perspective, it provides actionable insights for engineering departments to address these unknowns independently, without relying on systemic coordination
Comparative Study of Feature Selection Techniques for Machine Learning-Based Solar Irradiation Forecasting to Facilitate the Sustainable Development of Photovoltaics: Application to Algerian Climatic Conditions
No full textInternational audienceForecasting future solar power plant production is essential to continue the development of photovoltaic energy and increase its share in the energy mix for a more sustainable future. Accurate solar radiation forecasting greatly improves the balance maintenance between energy supply and demand and grid management performance. This study assesses the influence of input selection on short-term global horizontal irradiance (GHI) forecasting across two contrasting Algerian climates: arid Ghardaïa and coastal Algiers. Eight feature selection methods (Pearson, Spearman, Mutual Information (MI), LASSO, SHAP (GB and RF), and RFE (GB and RF)) are evaluated using a Gradient Boosting model over horizons from one to six hours ahead. Input relevance depends on both the location and forecast horizon. At t+1, MI achieves the best results in Ghardaïa (nMAE = 6.44%), while LASSO performs best in Algiers (nMAE = 10.82%). At t+6, SHAP-and RFE-based methods yield the lowest errors in Ghardaïa (nMAE = 17.17%), and RFE-GB leads in Algiers (nMAE = 28.13%). Although performance gaps between methods remain moderate, relative improvements reach up to 30.28% in Ghardaïa and 12.86% in Algiers. These findings confirm that feature selection significantly enhances accuracy (especially at extended horizons) and suggest that simpler methods such as MI or LASSO can remain effective, depending on the climate context and forecast horizon
Modéliser pour écologiser. Configurations sociotechniques et devenir-algorithme des experts agronomes
No full textInternational audienceModéliser pour écologiser. Configurations sociotechniques et devenir-algorithme des experts agronomes
Méthodes avancées pour améliorer l'interprétabilité des outils d'IA avec application au secteur de l'énergie
No full textDecision-making in energy systems traditionally relies on complex models integrating multiple uncertainties. The growing share of Renewable Energy Sources (RES) adds complexity due to weather dependence. While Artificial Intelligence (AI) tools improve forecasting, optimization, and decision-making, their lack of interpretability hinders adoption. This research enhances AI interpretability in energy applications by distinguishing between explainability methods and inherently interpretable models. It develops a symbolic AI approach for renewable energy bidding, applies rule-based models for grid adequacy forecasting, and utilizes Information Flow Fuzzy Cognitive Maps (IF-FCMs) to predict grid overloads. Additionally, it assesses AI regulations in the U.S., China, and Europe. The findings aim to improve trust, transparency, and regulatory compliance in energy AI applications.La prise de décision en énergie repose sur des modèles complexes intégrant de nombreuses incertitudes. L'essor des énergies renouvelables (ENR) accroît cette complexité en raison de leur dépendance météorologique. Bien que l'IA améliore la prévision et l'optimisation, son manque d'interprétabilité freine son adoption. Cette recherche clarifie l'explicabilité des modèles, développe une IA symbolique pour les enchères ENR, applique des modèles à règles pour la prévision de l'adéquation réseau et utilise les cartes cognitives floues à flux d'information (IF-FCMs) pour la prévision des surcharges du réseau. Enfin, elle évalue les réglementations IA aux États-Unis, en Chine et en Europe. Les résultats visent à renforcer confiance, transparence et conformité réglementaire dans l'IA énergétique
News and Load: Social and Economic Drivers of Regional Multi-horizon Electricity Demand Forecasting
No full textInternational audienceThe relationship between electricity demand and variables such as economic activity and weather patterns is well established.However, this paper explores the connection between electricity demand and social aspects. It further embeds dynamic information about the state of society into energy demand modelling and forecasting approaches.Through the use of natural language processing on a large news corpus, we highlight this important link.This study is conducted in five regions of the UK and Ireland and considers multiple time horizons from 1 to 30 days. It also considers economic variables such as GDP, unemployment and inflation.The textual features used in this study represent central constructs from the word frequencies, topics, word embeddings extracted from the news.The findings indicate that: 1) the textual features are related to various contents, such as military conflicts, transportation, the global pandemic, regional economics, and the international energy market.They exhibit causal relationships with regional electricity demand, which are validated using Granger causality and Double Machine Learning methods.2) Economic indicators play a more important role in the East Midlands and Northern Ireland, while social indicators are more influential in the West Midlands and the South West of England.3) The use of these factors improves deterministic forecasting by around 6%
Unlock Data Sharing in Wind Power Forecasting through Privacy-preserving Federated-Learning: Benchmarking of Mixed and Fully Encrypted Frameworks
No full textInternational audienceFederated learning is a technology enabling the privacy-preserving sharing of spatio-temporal data. Most works are only evaluated on a small-scale dataset. We analyse the scaling of a previous model developed by us. Showing that the idea of mixed encryption is able to allow efficient scaling and address this issue