Pasar al contenido principal

NEXO: APLICACIONES A METEOROLOGIA Y CLIMA

PID2023-150663NB-C21

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2023
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD DE ALCALA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 7
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Otros (Other)

Identifying key drivers of heatwaves: A novel spatio-temporal framework for extreme event detection

  • Pérez-Aracil, J.
  • Peláez-Rodríguez, C.
  • McAdam, Ronan
  • Squintu, Antonello
  • Marina, Cosmin M.
  • Lorente-Ramos, Eugenio
  • Luther, Niklas
  • Torralba, Verónica
  • Scoccimarro, Enrico
  • Cavicchia, Leone
  • Giuliani, Matteo
  • Zorita, Eduardo
  • Hansen, Felicitas
  • Barriopedro, David
  • García-Herrera, Ricardo
  • Gutiérrez, Pedro A.
  • Luterbacher, Jürg
  • Xoplaki, Elena
  • Castelletti, Andrea
  • Salcedo-Sanz, S.
Heatwaves (HWs) are extreme atmospheric events that produce significant societal and environmental impacts. Predicting these extreme events remains challenging, as their complex interactions with large-scale atmospheric and climatic variables are difficult to capture with traditional statistical and dynamical models. This work presents a general method for driver identification in extreme climate events. A novel framework named Spatio-Temporal Cluster-Optimized Feature Selection (STCO-FS) is proposed to identify key immediate (short-term) HW drivers by combining clustering algorithms with an ensemble evolutionary algorithm. The framework analyzes spatio-temporal data, reduces dimensionality by grouping similar geographical grid cells for each variable, and develops driver selection in spatial and temporal domains, identifying the best time lags between predictive variables and HW occurrences. The proposed method has been applied to analyze HWs in the Adda river basin in Italy. The approach effectively identifies significant variables influencing HWs in this region. This research can potentially enhance our understanding of HW drivers and predictability., This work has been partially supported by the European Commission, project “CLImate INTelligence: Extreme events detection, attribution and adaptation design using machine learning, CLINT” (grant ref.: H2020-LC-CLA-2020-2, 101003876), by the “Agencia Estatal de Investigación (España)”, Spanish Ministry of Research and Innovation through NEXO project (grant ref.: PID2023-150663NB-C21), and the European Union through the European Regional Development Fund; “project IA-CONV” with ref. CM/DEMG/2024-039 Funded by the Community of Madrid through the grant agreement for the promotion and advancement of research and technology transfer at the University of Alcalá. Jürg Luterbacher and Elena Xoplaki acknowledge supported by the Horizon Europe project MedEWSa (grant no. 101121192). Verónica Torralba acknowledges the Beatriu de Pinós program (2022 BP 00227) and the Ministry of Research and Universities of the Government of Catalonia. All authors have read and approved the final manuscript., Peer reviewed
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Predicting weather-related power outages in large scale distribution grids with deep learning ensembles

  • Prieto Godino, Luis
  • Peláez Rodríguez, César|||0000-0003-1260-8112
  • Pérez Aracil, Jorge|||0000-0002-4456-9886
  • Pastor Soriano, Jesús
  • Salcedo Sanz, Sancho|||0000-0002-4048-1676
Los eventos meteorológicos son los principales responsables de las interrupciones en el suministro eléctrico, lo que pone de manifiesto la necesidad de predecir con precisión estos cortes de energía relacionados con el clima. Este artículo se centra en la predicción de las incidencias diarias reportadas en redes eléctricas dentro de regiones específicas, utilizando las condiciones meteorológicas como variables predictivas clave. Para la selección de las ubicaciones óptimas de nodos de Reanálisis empleados como predictores, se ha considerado un enfoque de selección de características basado en optimización. Con el fin de superar el problema del desbalance de datos y mejorar la precisión de la predicción, se propone un algoritmo en conjunto basado en Aprendizaje Profundo (DL). Se consideran cinco arquitecturas DL distintas, generando múltiples aprendices individuales con hiperparámetros seleccionados aleatoriamente. La diversidad se garantiza entrenando cada modelo con conjuntos de datos ligeramente diferentes, obtenidos mediante muestreo aleatorio. Tres técnicas de fusión de información se utilizan para construir los modelos finales en conjunto. El enfoque propuesto ha sido evaluado con éxito en la predicción de incidencias diarias reales reportadas en líneas de distribución de dos provincias españolas, Valencia y Albacete, logrando una predicción precisa de los días con incidencias extremas, al tiempo que mantiene un buen rendimiento global y una baja tasa de falsas alarmas. Los modelos con mejor desempeño que emplean esta metodología alcanzan tasas de detección del 38 % y del 73 %, con tasas de falsas alarmas del 1 % y del 3 %, respectivamente. Este enfoque no solo mejora la precisión de la predicción en comparación con los aprendices individuales, sino que también incrementa la capacidad de generalización y la robustez de los modelos DL independientes. Además, reduce de manera efectiva el sobreajuste inherente a estos métodos, eliminando la necesidad de un proceso complejo de selección de hiperparámetros., Weather events are primarily contributors to electrical supply disruptions, prompting the need to accurately forecast these weather-related power outages. This paper focuses on predicting daily reported incidences in electrical grids within specific regions, leveraging weather conditions as specific predictive variables. An optimization-based feature selection approach has been considered for selecting the optimal Reanalysis node locations used as predictors. To overcome the data imbalance challenge and enhance prediction accuracy, we propose a Deep Learning-based (DL) ensemble algorithm. Five distinct DL architectures are considered, generating multiple individual learners with randomly selected hyperparameters. Diversity is ensured by training each model with slightly different randomly sampled data. Three information fusion techniques construct the final ensemble models. The proposed approach has been successfully evaluated in predicting real daily reported incidences in distribution lines across two Spanish provinces, Valencia and Albacete, achieving an accurate prediction of days with extreme incidences while maintaining a good overall performance and a low rate of false alarms. The top-performing models using this methodology achieve detection rates of 38% and 73%, with false alarm rates of 1% and 3%, respectively. This approach not only enhances prediction accuracy compared to individual learners but also improves the generalization ability and robustness of standalone DL models. Additionally, it effectively reduces the inherent overfitting of these methods, removing the necessity for a complex hyperparameter selection process., Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Detection and attribution of heat waves with the Multivariate Autoencoder Flow-Analogue Method (MvAE-AM)

  • Marina, Cosmin Madalin|||0000-0002-5849-6673
  • Pérez Aracil, Jorge|||0000-0002-4456-9886
  • McAdam, Ronan
  • Lorente Ramos, Eugenio
  • Luther, Niklas
  • Zorita, Eduardo
  • Scoccimarro, Enrico
  • Luterbacher, Jürg
  • Xoplaki, Elena
  • Salcedo Sanz, Sancho|||0000-0002-4048-1676
Heat waves (HWs) are complex, multivariate, extreme weather events that cause significant harm to human health, ecosystems, and economies. Correct detection and attribution of HWs to anthropogenic climate change is important to better understand the underlying mechanisms and to improve predictions. In this work, we address this issue and propose a multivariate version of a hybrid approach to reconstruct heat waves, consisting of the AM and deep Autoencoders (MvEA-AM algorithm), improving existing less effective methods used until now, such as the multivariate Analogue Method (MvAM). The proposed hybrid approach produces a more reliable representation of the event than the classical MvAM for reconstructing and attributing HWs in Europe. The explainable and interpretable analysis of the obtained results is based on leveraging the SHapley Additive exPlanations (SHAP) method to explain deep learning algorithms, a capability that is not achievable with the MvAM. This explainability analysis shows that our model learns useful features during the training of the algorithm, which are aligned with the Physics of the problem, and employs the correct features during reconstruction and attribution analysis of the HWs considered., European Commission, Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Efficient design of electromagnetic field exposure maps with multi-method evolutionary ensembles

  • Guillén Pina, Jorge|||0000-0003-2153-5656
  • Pérez Aracil, Jorge|||0000-0002-4456-9886
  • Chocano del Cerro, Ricardo
  • Sánchez Montero, Rocío|||0000-0002-3135-8828
  • López Espí, Pablo Luis|||0000-0003-3750-255X
  • Salcedo Sanz, Sancho|||0000-0002-4048-1676
Radio-frequency electromagnetic field (EMF) exposure is a growing concern among the population. This concern has led to a need for practical tools to contribute to an adequate risk perception. Representing spatial variations af-ter measurements from fixed sites and interpolating using different techniques is the most suitable method for obtaining EMF high-quality exposure maps. This paper uses evolutionary computation to obtain the optimal set of points to construct high-quality electromagnetic field exposure maps (minimizing an error measure with respect to a reference exposure map). A multi-method en-semble evolutionary approach, able to combine different search operators in a single population (PCRO-SL), is introduced for this particular problem, and it has been tested over actual measurements at Meco town, Madrid, Spain, obtaining good quality electromagnetic field exposure map reconstructions in terms of the differences with a reference EMF exposure map. The results obtained show that reducing the number of measurement points necessary to obtain significant exposure maps, while maintaining their representativeness, is possible., Comunidad de Madrid, Universidad de Alcalá, Ministerio de Ciencia, Innovación y Universidades
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

A probabilistic alert system for extreme wind events prediction using quantile regression ensembles

  • Peláez Rodríguez, César|||0000-0003-1260-8112
  • Pérez Aracil, Jorge|||0000-0002-4456-9886
  • Cruz de la Torre, Carlos|||0000-0001-6937-9838
  • Cornejo Bueno, Laura María|||0000-0002-4126-8041
  • Prieto Godino, Luis
  • Alexandre Cortizo, Enrique|||0000-0003-1299-2345
  • Salcedo Sanz, Sancho|||0000-0002-4048-1676
Anticipar y mitigar el impacto de los eventos de viento extremo es cada vez más crítico a medida que la energía eólica se convierte en un componente central de los sistemas energéticos modernos. Sin embargo, los enfoques predictivos existentes a menudo tienen dificultades para capturar la incertidumbre y la variabilidad inherentes a los datos de viento, lo que limita su eficacia en la gestión del riesgo. Esta investigación tiene como objetivo desarrollar un sistema de alerta probabilístico que permita predecir de manera eficaz la ocurrencia de dichos eventos extremos. Para ello, se propone un marco novedoso que combina regresión cuantílica y estimación de densidad mediante núcleos, con el fin de construir un sistema de predicción en conjunto robusto. Al integrar las predicciones individuales de regresión cuantílica a través de múltiples cuantiles, el marco propuesto captura la variabilidad y la incertidumbre inherentes a los datos de viento. Además, las salidas probabilísticas del modelo en conjunto se calibran mediante regresión isotónica, lo que da lugar a distribuciones refinadas que se ajustan estrechamente a las tasas observadas de ocurrencia de eventos extremos. El marco fue validado utilizando datos reales de un parque eólico en España, mostrando mejoras sustanciales con respecto a los clasificadores binarios probabilísticos convencionales, tanto en precisión como en calibración de las probabilidades de eventos extremos. Estos resultados ponen de relieve el potencial del sistema propuesto para mejorar la toma de decisiones operativas y la resiliencia de la infraestructura de energía eólica frente a condiciones meteorológicas extremas., Anticipating and mitigating the impact of extreme wind events is increasingly critical as wind power becomes a central component of modern energy systems. However, existing predictive approaches often struggle to capture the uncertainty and variability inherent in wind data, limiting their effectiveness in risk management. This research aims to develop a probabilistic alert system to predict the occurrence of such extreme events effectively. To achieve this, a novel framework is proposed, combining quantile regression and kernel density estimation, to construct a robust predictive ensemble system. By integrating individual quantile regression predictions across multiple quantiles, the proposed framework captures the inherent variability and uncertainty of wind data. Additionally, the ensemble model’s probabilistic outputs are calibrated using isotonic regression, yielding refined distributions that closely align with observed extreme event occurrence rates. The framework was validated using real-world data from a wind farm in Spain, showing substantial improvements over conventional probabilistic binary classifiers in both accuracy and calibration of extreme event probabilities. These findings highlight the potential of the proposed system to enhance operational decision-making and resilience in wind power infrastructure under extreme weather conditions., Agencia Estatal de Investigación, Comunidad de Madrid
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Evolutionary optimization of spatially-distributed multi-sensors placement for indoor surveillance environments with security levels

  • Moreno Saavedra, Luis Miguel|||0009-0004-9211-8491
  • Costa, Vinícius G.
  • Garrido Sáez, Adrián
  • Jiménez Fernández, Silvia|||0000-0002-2065-1754
  • Portilla Figueras, José Antonio|||0000-0001-6569-6780
  • Salcedo Sanz, Sancho|||0000-0002-4048-1676
The surveillance multi-sensor placement is an important optimization problem that consists of positioning several sensors of different types to maximize the coverage of a determined area while minimizing the cost of the deployment. In this work, we tackle a modified version of the problem, consisting of spatially distributed multi-sensor placement for indoor surveillance. Our approach is focused on security surveillance of sensible indoor spaces, such as military installations, where distinct security levels can be considered. We propose an evolutionary algorithm to solve the problem, in which a novel special encoding (integer encoding with binary conversion) and effective initialization have been defined to improve the performance and convergence of the proposed algorithm. We also consider the probability of detection for each surveillance point, which depends on the distance to the sensor at hand, to better model real-life scenarios. We have tested the proposed evolutionary approach in different instances of the problem, varying both size and difficulty and obtained excellent results regarding the cost of sensors? placement and convergence time of the algorithm., Agencia Estatal de Investigación, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Fuzzy-based ensemble methodology for accurate long-term prediction and interpretation of extreme significant wave height events

  • Peláez Rodríguez, César|||0000-0003-1260-8112
  • Pérez Aracil, Jorge|||0000-0002-4456-9886
  • Gómez Orellana, Antonio Manuel
  • Guijo Rubio, David
  • Vargas, Victor Manuel
  • Gutiérrez, Pedro Antonio
  • Hervás Martínez, C.
  • Salcedo Sanz, Sancho|||0000-0002-4048-1676
Providing an accurate prediction of Significant Wave Height (SWH), and specially of extreme SWH events, is crucial for coastal engineering activities and holds major implications in several sectors as offshore renewable energy. With the aim of overcoming the challenge of skewness and imbalance associated with the prediction of these extreme SWH events, a fuzzy-based cascade ensemble of regression models is proposed. This methodology allows to remarkably improve the predictive performance on the extreme SWH values, by using different models specialised in different ranges on the target domain. The method?s explainability is enhanced by analysing the contribution of each model, aiding in identifying those predictor variables more characteristic for the detection of extreme SWH events. The methodology has been validated tackling a long-term SWH prediction problem, considering two case studies over the southwest coast of the United States of America. Both reanalysis data, providing information on various meteorological factors, and SWH measurements, obtained from the nearby stations and the station under examination, have been considered. The goodness of the proposed approach has been validated by comparing its performance against several machine learning and deep learning regression techniques, leading to the conclusion that fuzzy ensemble models perform much better in the prediction of extreme events, at the cost of a slight deterioration in the rest of the samples. The study contributes to advancing the SWH prediction field, specially, to understanding the behaviour behind extreme SWH events, critical for various sectors reliant on oceanic conditions., European Commission, Ministerio de Economía y Competitividad, Agencia Estatal de Investigación