A5090648505- Meteorological Phenomena and Simulations
- Energy Load and Power Forecasting
- Wind and Air Flow Studies
- Ship Hydrodynamics and Maneuverability
- Hydrological Forecasting Using AI
- Structural Health Monitoring Techniques
- Advanced Multi-Objective Optimization Algorithms
- Time Series Analysis and Forecasting
- Flow Measurement and Analysis
- Metaheuristic Optimization Algorithms Research
- Radiative Heat Transfer Studies
- Nuclear Engineering Thermal-Hydraulics
- Solar Radiation and Photovoltaics
- Retinal Imaging and Analysis
- Cryospheric studies and observations
- Probabilistic and Robust Engineering Design
- Anomaly Detection Techniques and Applications
- Calibration and Measurement Techniques
- Cerebral Venous Sinus Thrombosis
- Retinal and Optic Conditions
- Heat transfer and supercritical fluids
- Hydrology and Drought Analysis
- Infrastructure Maintenance and Monitoring
- Climate variability and models
- Neural Networks and Applications
Universidad de Alcalá
2023-2024
University of Maribor
2024
University of Molise
2024
Some materials, such as reinforced and prestressed concrete, involve non-linear constitutive relationships in elasticity problems defined on them. In particular, the shear strength of a concrete beam may be calculated by considering diagonal struts field context so-called "Compression Field Theories" (CFTs). This work presents an efficient Machine Learning method alternative to numerical methods for obtaining full response beams based CFT regarding stresses, strains, crack angles. For that,...
This paper presents a method for providing explainability in the integration of artificial intelligence (AI) and data mining techniques when dealing with meteorological prediction. Explainable (XAI) refers to transparency AI systems explanations their predictions decision-making processes, contribute improve prediction accuracy enhance trust systems. The focus this relies on interpretability challenges ordinal classification problems within weather forecasting. Ordinal involves predicting...
Heat stress represents a major risk to human health, making the development of advanced warning systems essential for safeguarding individuals and communities. Data-driven models, such as FourCastNet, PanguWeather, GraphCast, provide rapid, accurate, publicly accessible forecasts meteorological variables. However, these models do not all variables required calculate thermal indices, Universal Thermal Climate Index (UTCI). To address this limitation, study proposes method estimate UTCI...
Wind extremes, encompassing both high-intensity wind events and periods of diminished activity, pose multifaceted challenges across sectors such as renewable energy production, infrastructure resilience, environmental risk management. These phenomena, driven by complex interactions within atmospheric systems, demand innovative analytical predictive approaches. This study explores the application artificial intelligence (AI) to address these challenges, focusing on its potential enhance...
This study introduces an innovative preprocessing technique utilizing Autoencoder (AE) as alternative to the traditional multivariate Analogue Method (AM). The newly proposed method, MvAE-AM, is employed reconstruct historical heat wave events: France in 2003, Balkans 2007, Russia 2010, and Spain 1995. AE effectively extracts critical information from variables such soil moisture (SM), potential evaporation (PEva), mean sea level pressure (MSL), geopotential height at 500 hPa (Z500) into a...
In this paper, we propose different explicable forecasting approaches, based on inductive and evolutionary decision rules, for extreme low-visibility events prediction. Explicability of the processes given by rules is in core proposal. We two methodologies: first, apply PRIM algorithm evolution to obtain induced evolved subsequently these boxes are used as a possible simpler alternative ML/DL classifiers. Second, integrate information provided induced/evolved techniques, extra inputs, order...
In this paper, new probabilistic and dynamic (adaptive) strategies for creating multi-method ensembles based on the coral reef optimization with substrate layers (CRO-SL) algorithm are proposed. CRO-SL is an evolutionary-based ensemble approach that able to combine different search procedures a single population. work, two improve analyzed. First, (PCRO-SL) presented, which substitutes substrates in population tags associated each individual. Each tag represents operator will modify...
In the context of reinforced concrete members subjected to shear, steel behaviour, assumed as embedded in concrete, has been modelled through different strategies. One them, so-called Refined Compression Field Theory (RCFT), is based on concept tension stiffening area, and shows a better fitting with experimental results than other shear theories. However, for certain standard design conditions, RCFT non-linear formulation does not throw real physical solution yield strain, what hinders its...
Abstract This paper presents a novel hybrid approach for the probabilistic reconstruction of meteorological fields based on combined use analogue method (AM) and deep autoencoders (AEs). The AE–AM algorithm trains AE in predictor fields, which encoder filters towards compressed space reduced dimensionality. AM is then applied this latent to find similar situations (analogues) historical record, from target field can be reconstructed. compared classical AM, flow analogues are explicitly...
Heatwaves (HWs) are extreme atmospheric events that produce significant societal and environmental impacts. Predicting these remains challenging, as their complex interactions with large-scale climatic variables difficult to capture traditional statistical dynamical models. This work presents a general method for driver identification in climate events. A novel framework (STCO-FS) is proposed identify key immediate (short-term) HW drivers by combining clustering algorithms an ensemble...
The Analogue Method (AM) is a classical statistical downscaling technique applied to field reconstruction. It widely used for prediction and attribution tasks. method based on the principle that two similar atmospheric states cause local effects. core of AM K-nearest neighbor methodology. Thus, different have similarities according analogy criterion. has remained unchanged since its definition, although some attempts been made improve performance. Machine learning (ML) techniques recently...
This paper proposes two hybrid approaches based on Autoencoders (AEs) for long-term temperature prediction. The first algorithm comprises an AE trained to learn patterns, which is then linked a second AE, used detect possible anomalies and provide final proposed approach involves training using the resulting latent space as input of neural network, will prediction output. Both are tested in air European cities: seven locations where major heat waves occurred have been considered. entire year...
In this paper we propose new probabilistic and dynamic (adaptive) strategies to create multi-method ensembles based on the Coral Reefs Optimization with Substrate Layers (CRO-SL) algorithm. The CRO-SL is an evolutionary-based ensemble approach, able combine different search procedures within a single population. work discuss two improve First, defined Probabilistic (PCRO-SL), which substitutes substrates in population by {\em tags} associated each individual. Each tag represents operator...