Maria Francisca Almeida

Abstract
This paper presents an optimization model for electric vehicle (EV) fleet charging under MIBEL (Iberian Electricity Market). The model integrates EV charging with day-ahead forecasting for grid energy prices, photovoltaic (PV) generation, and local power demand, combined with a battery energy storage system (BESS) to minimize total charging costs, reduce peak demand, and maximize renewable use. Simulations across Baseline, Certainty, and Uncertainty scenarios show that the proposed approach would reduce total charging costs by up to 49%, lower carbon emissions by 73.7%, and improve SOC compliance, while smoothing demand curves to mitigate excessive contracted power charges. The results demonstrate the economic and environmental benefits of predictive and adaptive EV charging strategies, highlighting opportunities for further enhancements through real-time adjustments and vehicle-to-grid (V2G) integration.

Abstract
Reducing the gap between renewable energy needs and supply is crucial to achieve sustainable growth. Hydroelectric power production predictions in several Madeira Island catchment regions are shown in this article using Long Short-Term Memory, LSTM, networks. In order to foresee hydro reservoirs inflows, our models take into account the island's dynamic precipitation and flow rates and simplify the process of water moving from the cloud to the turbine. The model developed for the Socorridos Faja Rodrigues system demonstrates the proficiency of LSTMs in capturing the unexpected flow behavior through its low RMSE. When it comes to energy planning, the model built for the CTIII Paul Velho system gives useful information despite its lower accuracy when it comes to anticipating problems.

Abstract
The application of machine learning (ML) algorithms to electronic health records (EHR) data allows the achievement of data-driven insights on various clinical problems and the development of clinical decision support (CDS) systems to improve patient care. However, data governance and privacy barriers hinder the use of data from multiple sources, especially in the medical field due to the sensitivity of data. Federated learning (FL) is an attractive data privacy-preserving solution in this context by enabling the training of ML models with data from multiple sources without any data sharing, using distributed remotely hosted datasets. The Secur-e-Health project aims at developing a solution in terms of CDS tools encompassing FL predictive models and recommendation systems. This tool may be especially useful in Pediatrics due to the increasing demands on Pediatric services, and the current scarcity of ML applications in this field compared to adult care. Herein we provide a description of the technical solution proposed in this project for three specific pediatric clinical problems: childhood obesity management, pilonidal cyst post-surgical care and retinography imaging analysis.