arXiv subjects

Laurent Dubus

Publications and source records attributed to Laurent Dubus.

12 recordsLinked to original sources

arXiv2026

Statistics of Temperature Extremes and Implications for Electrical Power Infrastructure in Continental France

In this article, we present extreme value statistics of temperature extremes over continental France with a focus on their implications for electrical power infrastructure. These are obtained by fitting a non-stationary generalised extreme value distribution using a Bayesian setup, which also provides errors or uncertainty values for all estimates. Within this method, a combination of simulated data from a collection of 28 CMIP6 models and measured records from the E-OBS dataset is used. The investigated climate scenarios are SSP2-4.5, SSP3-7.0 and SSP5-8.5, considering both historical and future climates spanning the years from 1850 to 2099. The method provides full spatial resolution on a 0.25 degree grid, allowing to asses extreme temperatures at arbitrary locations. Leveraging this particular aspect, various maps revealing the spatial structure of annual maximum high temperature extremes over continental France for the median and the statistical upper bound are shown together with summary information for the administrative regions in France towards the end of the century. It is statistically possible that the south west region of Occitanie could reach temperatures of up to 57°C under a high emission scenario by 2080, which is the highest compared to all other regions in continental France. We also perform a comparison to the reference climate adaptation trajectory for France (TRACC - Trajectoire de réchauffement de Référence pour l'Adaptation au Changement Climatique), showing that it potentially underestimates the stated maximum temperatures which could be surpassed by up to +8°C. Furthermore, five reference electrical power infrastructure locations are investigated on how they are potentially affected by temperature extremes with the quantified intensities.

physics.ao-ph↗

arXiv2026

AI-boosted rare event sampling to characterize extreme weather

Weather extremes pose major societal risks, especially in a changing climate, but due to their rarity, they are difficult to study using limited observations or complex climate models. We introduce AI+RES, a framework coupling fast AI weather forecasts with a high-fidelity physics model using a rare-event algorithm to efficiently characterize extremes. This approach enables the study of the statistics and physics of very rare events, such as once per millennium heatwaves at two orders-of-magnitude lower computational cost. AI+RES can be applied broadly across climate science and other fields concerned with rare events.

physics.ao-ph↗

arXiv2026

Leveraging Climate Services to Build Climate Resilient Power Systems

We explore the crucial interplay between climate change and power system planning, highlighting the urgent need to systematically integrate climate information into energy system studies. Climate change impacts the energy sector on multiple fronts. Short-term weather variability drives daily and seasonal fluctuations in supply and demand. Long-term trends and increased frequency of extremes pose risks to infrastructure performance, asset lifetimes, and system adequacy. Representing compound events and spatial correlations across borders is a complex challenge, and uncertainties persist due to uncertainties from different models, scenarios, and downscaling methodologies. The Pan-European Climate Database (PECD4.2), developed in partnership between ENTSO-E and C3S, marks a change in how energy system planning is conducted. The PECD4.2 integrates historical reanalysis and six climate models across four SSP's, providing harmonised, openly available datasets tailored for power system studies. The physical conversion models for wind and solar energy better reflect technological progression than machine learning methods trained on historical data, improving robustness under changing future conditions. Despite these advances, challenges remain. Particularly in hydropower modelling and the lack of public harmonised energy datasets that are required to train these models. Complex processing chains from raw climate data to actionable insights and the lack of standardized integration of climate information lengthen lead times for energy-sector adoption. This leads to diverging approaches and variable consideration of climate risks. Closer, more generalised collaboration and communication between climate service providers and energy stakeholders are therefore necessary, as are the development of user-friendly tools for data manipulation and analysis and robust feedback loops.

physics.ao-ph↗

arXiv2026

Non-stationary GEV models for estimating design sea-states in a changing climate. Applications to offshore wind farms along the French coasts

The rapid expansion of the French offshore wind sector requires a critical reassessment of structural durability in the face of evolving marine conditions driven by climate change. Traditional design methodologies, which rely on the assumption of stationary environmental conditions, are no longer adequate. This study introduces a novel statistical framework to assess future changes in significant wave height by employing non-stationary Generalized Extreme Value (GEV) models applied to monthly maxima. This approach aims to reduce uncertainty and provide robust design tools adapted to the non-stationary conditions of the future. Based on CMIP6 climate models and reanalysis data, results reveal a projected trend towards a more pronounced seasonal contrast along the French Atlantic and English Channel coasts under future scenarios (SSP1-2.6 and SSP5-8.5), whereas the French Mediterranean Sea exhibits results that are more difficult to interpret, due to a weaker increase of extremes and large uncertainties (inter-model spread). Projections indicate more intense winters and calmer summers, along with a shift in the seasonal cycle. Overall, the multi-model ensemble suggests an increase in the design levels for extreme sea states. The research concludes by defining a new methodology for calculating an equivalent design level over the structure's operational lifespan. This tool is deemed essential for ensuring the resilience and economic viability of future offshore wind farms in a changing climate.

stat.AP↗

arXiv2024

Evaluation of 'Dunkelflaute' event detection methods considering grid operators' needs

Weather conditions associated with low electricity production from renewable energy sources (RES) can result in challenging 'dunkelflaute' events, where 'dunkel' means dark and 'flaute' refers to low windspeeds. In a power system relying significantly on RES, such events can pose a risk for maintaining resource adequacy, i.e. the balance between generation and demand, particularly if they occur over a large geographical area and for an extended period of time. This risk is further emphasized in periods of cold ('kalte') temperature, known as 'kalte dunkelflaute'. In this paper, we perform a literature review of different methods to identify dunkelflaute events from hourly RES production and load data alone. We then validate three of these methods by comparing their results with periods of shortage identified from a detailed power system simulation model used by grid operators (ERAA2023). Strengths and weaknesses of these methods are discussed in terms of their data requirements, ease of application, and skill in detecting dunkelflaute events. We find that all three 'dunkelflaute' event detection methods have some ability to identify potential energy shortages, but none are able to detect all events. Most likely other factors such as the presence of energy storage capacity, non-weather-dependent outages, and model-related factors limit the skill of these methods. We find that all three methods perform best if the residual load is used as input, rather than hourly RES production or load alone. Overall, we find that Otero'22 is the method that yields the best results while being straightforward to implement and requiring only data with daily resolution. The results hold for countries relying on a small or a large share of RES production in their electricity mix.

physics.soc-ph↗

arXiv2024

Space-scale Exploration of the Poor Reliability of Deep Learning Models: the Case of the Remote Sensing of Rooftop Photovoltaic Systems

Photovoltaic (PV) energy grows rapidly and is crucial for the decarbonization of electric systems. However, centralized registries recording the technical characteristifs of rooftop PV systems are often missing, making it difficult to accurately monitor this growth. The lack of monitoring could threaten the integration of PV energy into the grid. To avoid this situation, the remote sensing of rooftop PV systems using deep learning emerged as a promising solution. However, existing techniques are not reliable enough to be used by public authorities or transmission system operators (TSOs) to construct up-to-date statistics on the rooftop PV fleet. The lack of reliability comes from the fact that deep learning models are sensitive to distribution shifts. This work proposes a comprehensive evaluation of the effects of distribution shifts on the classification accuracy of deep learning models trained to detect rooftop PV panels on overhead imagery. We construct a benchmark to isolate the sources of distribution shift and introduce a novel methodology that leverages explainable artificial intelligence (XAI) and decomposition of the input image and model's decision in terms of scales to understand how distribution shifts affect deep learning models. Finally, based on our analysis, we introduce a data augmentation technique meant to improve the robustness of deep learning classifiers to varying acquisition conditions. We show that our proposed approach outperforms competing methods. We discuss some practical recommendations for mapping PV systems using overhead imagery and deep learning models.

cs.CV↗

arXiv2023

Assessment of the Reliablity of a Model's Decision by Generalizing Attribution to the Wavelet Domain

Neural networks have shown remarkable performance in computer vision, but their deployment in numerous scientific and technical fields is challenging due to their black-box nature. Scientists and practitioners need to evaluate the reliability of a decision, i.e., to know simultaneously if a model relies on the relevant features and whether these features are robust to image corruptions. Existing attribution methods aim to provide human-understandable explanations by highlighting important regions in the image domain, but fail to fully characterize a decision process's reliability. To bridge this gap, we introduce the Wavelet sCale Attribution Method (WCAM), a generalization of attribution from the pixel domain to the space-scale domain using wavelet transforms. Attribution in the wavelet domain reveals where and on what scales the model focuses, thus enabling us to assess whether a decision is reliable. Our code is accessible here: \url{https://github.com/gabrielkasmi/spectral-attribution}.

cs.CV↗

arXiv2023

Can We Reliably Improve the Robustness to Image Acquisition of Remote Sensing of PV Systems?

Photovoltaic (PV) energy is crucial for the decarbonization of energy systems. Due to the lack of centralized data, remote sensing of rooftop PV installations is the best option to monitor the evolution of the rooftop PV installed fleet at a regional scale. However, current techniques lack reliability and are notably sensitive to shifts in the acquisition conditions. To overcome this, we leverage the wavelet scale attribution method (WCAM), which decomposes a model's prediction in the space-scale domain. The WCAM enables us to assess on which scales the representation of a PV model rests and provides insights to derive methods that improve the robustness to acquisition conditions, thus increasing trust in deep learning systems to encourage their use for the safe integration of clean energy in electric systems.

cs.CV↗

arXiv2023

PyPVRoof: a Python package for extracting the characteristics of rooftop PV installations using remote sensing data

Photovoltaic (PV) energy grows at an unprecedented pace, which makes it difficult to maintain up-to-date and accurate PV registries, which are critical for many applications such as PV power generation estimation. This lack of qualitative data is especially true in the case of rooftop PV installations. As a result, extensive efforts are put into the constitution of PV inventories. However, although valuable, these registries cannot be directly used for monitoring the deployment of PV or estimating the PV power generation, as these tasks usually require PV systems {\it characteristics}. To seamlessly extract these characteristics from the global inventories, we introduce {\tt PyPVRoof}. {\tt PyPVRoof} is a Python package to extract essential PV installation characteristics. These characteristics are tilt angle, azimuth, surface, localization, and installed capacity. {\tt PyPVRoof} is designed to cover all use cases regarding data availability and user needs and is based on a benchmark of the best existing methods. Data for replicating our accuracy benchmarks are available on our Zenodo repository \cite{tremenbert2023pypvroof}, and the package code is accessible at this URL: \url{https://github.com/gabrielkasmi/pypvroof}.

eess.SP↗

arXiv2023

Towards unsupervised assessment with open-source data of the accuracy of deep learning-based distributed PV mapping

Photovoltaic (PV) energy is rapidly growing and key to mitigating the energy crisis. However, distributed PV generation, which amounts to half of the PV installed capacity, is typically unavailable to transmission system operators (TSOs), making it increasingly difficult to balance the load and supply and avoid grid congestions. To assess distributed PV generation, TSOs need precise knowledge regarding the metadata of distributed PV installations. Many remote sensing-based approaches have been proposed to map these installations in recent years. However, to use these methods in industrial processes, assessing their accuracy over the mapping area, i.e., the area covered by the model during deployment, is necessary. We define the downstream task accuracy (DTA) as the accuracy over the mapping area, automatically computed using publicly available data sources and the model's outputs and expressed in an interpretable way for operators. We benchmark existing models for distributed PV mapping and show how they perform in terms of DTA. We show that the accuracy computed on the test set overestimates by about 30 percentage points the accuracy on the mapping area. Our approach paves the way for safer integration of deep-learning-based pipelines for remote PV mapping. Code is available at https://github.com/gabrielkasmi/deeppvmapper.

cs.CV↗

arXiv2022

A crowdsourced dataset of aerial images with annotated solar photovoltaic arrays and installation metadata

Photovoltaic (PV) energy generation plays a crucial role in the energy transition. Small-scale PV installations are deployed at an unprecedented pace, and their integration into the grid can be challenging since public authorities often lack quality data about them. Overhead imagery is increasingly used to improve the knowledge of residential PV installations with machine learning models capable of automatically mapping these installations. However, these models cannot be easily transferred from one region or data source to another due to differences in image acquisition. To address this issue known as domain shift and foster the development of PV array mapping pipelines, we propose a dataset containing aerial images, annotations, and segmentation masks. We provide installation metadata for more than 28,000 installations. We provide ground truth segmentation masks for 13,000 installations, including 7,000 with annotations for two different image providers. Finally, we provide installation metadata that matches the annotation for more than 8,000 installations. Dataset applications include end-to-end PV registry construction, robust PV installations mapping, and analysis of crowdsourced datasets.

cs.CV↗

arXiv2019

A parametric model for wind turbine power curves incorporating environmental conditions

A wind turbine's power curve relates its power production to the wind speed it experiences. The typical shape of a power curve is well known and has been studied extensively; however, the power curves of individual turbine models can vary widely from one another. This is due to both the technical features of the turbine (power density, cut-in and cut-out speeds, limits on rotational speed and aerodynamic efficiency), and environmental factors (turbulence intensity, air density, wind shear and wind veer). Data on individual power curves are often proprietary and only available through commercial databases. We therefore develop an open-source model which can generate the power curve of any turbine, adapted to the specific conditions of any site. This can employ one of six parametric models advanced in the literature, and accounts for the eleven variables mentioned above. The model is described, the impact of each technical and environmental feature is examined, and it is then validated against the manufacturer power curves of 91 turbine models. Versions of the model are made available in MATLAB, R and Python code for the community.

eess.SP↗