Abstract
Complex optimization problems are often associated to large search spaces and consequent prohibitive execution times in finding the optimal results. This is especially relevant when dealing with dynamic real problems, such as those in the field of power and energy systems. Solving this type of problems requires new models that are able to find near-optimal solutions in acceptable times, such as metaheuristic optimization algorithms. The performance of these algorithms is, however, hugely dependent on their correct tuning, including their configuration and parametrization. This is an arduous task, usually done through exhaustive experimentation. This paper contributes to overcome this challenge by proposing the application of sequential model algorithm configuration using Bayesian optimization with Gaussian process and Monte Carlo Markov Chain for the automatic configuration of a genetic algorithm. Results from the application of this model to an electricity market participation optimization problem show that the genetic algorithm automatic configuration enables identifying the ideal tuning of the model, reaching better results when compared to a manual configuration, in similar execution times.

Abstract
Real Estate Agents perform the tedious job of selecting and filtering pictures of houses manually on a daily basis, in order to choose the most suitable ones for their websites and provide a better description of the properties they are selling. However, this process consumes a lot of time, causing delays in the advertisement of homes and reception of proposals. In order to expedite and automate this task, Computer Vision solutions can be employed. Deep Learning, which is a subfield of Machine Learning, has been highly successful in solving image recognition problems, making it a promising solution for this particular context. Therefore, this paper proposes the application of Vision Transformers to indoor room classification. The study compares various image classification architectures, ranging from traditional Convolutional Neural Networks to the latest Vision Transformer architecture. Using a dataset based on well-known scene classification datasets, their performance is analyzed. The results demonstrate that Vision Transformers are one of the most effective architectures for indoor classification, with highly favorable outcomes in automating image recognition and selection in the Real Estate industry.

Abstract
Agriculture being an essential activity sector for the survival and prosperity of humanity, it is fundamental to use sustainable technologies in this field. With this in mind, some statistical data are analyzed regarding the food price rise and sustainable development indicators, with a special focus on the Portugal region. It is determined that one of the main factors that influences agriculture's success is the soil's characteristics, namely in terms of moisture and nutrients. In this regard, irrigation processes have become indispensable, and their technological management brings countless economic advantages. Like other branches of agriculture, the wine sector needs an adequate concentration of nutrients and moisture in the soil to provide the most efficient results, considering the appropriate and intelligent use of available water and energy resources. Given these facts, the use of renewable energies is a very important aspect of this study, which also synthesizes the main irrigation methods and examines the importance of evaluating the evapotranspiration of crops. Furthermore, the control of irrigation processes and the implementation of optimization and resource management models are of utmost importance to allow maximum efficiency and sustainability in this field.

Abstract
In recent years, there has been a significant growth in the use of technologies that rely on natural resources (wind, solar, etc.) as primary sources of energy. The generation originating from renewable sources brings an increased need for adaptation in power electrical systems. Predicting the amount of energy produced by these technologies is a complex task due to the uncertainty associated with natural resources. This uncertainty hinders decision-making, both at the system level and for consumers themselves who are increasingly using this type of technology for self-consumption. This study focuses on classifying solar intensity using imbalanced data, which means that some of the data categories are more prevalent than others. Oversampling techniques are be employed to increase the amount of data, thereby allowing for balanced training data and improving the performance of prediction models. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

Abstract
This article presents a comprehensive state of the art analysis of the challenging domain of synthetic data generation. Focusing on the problem of synthetic data generation, the paper explores various difficulties that are identified, especially in real-world problems such as those is the scope of power and, energy systems, including the amount of data, data privacy concerns, temporal considerations, dynamic generation, delays, and failures. The investigation delves into the multifaceted nature of the challenges presented by these factors in the synthesis process. The review thoroughly examines different models used in synthetic data generation, covering Generative Adversarial Networks (GANs), Variational Autoencoder (VAE), Synthetic Minority Oversampling Technique (SMOTE), Data Synthesizer (DS) and E. Non-Parametric SynthPop (SP-NP). Each model is dissected with respect to its advantages, disadvantages, and applicability in different data generation scenarios. Special attention is paid to the nuanced aspects of dynamic data generation and the mitigation of challenges such as delays and failures. The insights drawn from this review contribute to a deeper understanding of the landscape around synthetic data generation, providing a valuable resource for researchers, practitioners, and stakeholders who aim to harness the potential of synthetic data in addressing real-world data challenges. The paper concludes by outlining possible avenues for future research and development in this ever-evolving field. © 2024 IEEE.

Abstract
This study explores models for synthetic data generation of time series. In order to improve the achieved results, i.e., the data generated, new ways of improvement are explored and different models of synthetic data generation are compared. The model addressed in this work is the Generative Adversarial Networks (GANs), known for generating data similar to the original basis data through the training of a generator. The GANs are applied using the datasets of Quinta de Santa Bárbara and the Pinhão region, with the main variables being the Average temperature, Wind direction, Average wind speed, Maximum instantaneous wind speed and Solar radiation. The model allowed to generate missing data in a given period and, in turn, enables to analyze the results and compare them with those of a multiple linear regression method, being able to evaluate the effectiveness of the generated data. In this way, through the study and analysis of the GANs we can see if the model presents effectiveness and accuracy in the synthetic generation of meteorological data. With the proper conclusions of the results, this information can be used in order to improve the search for different models and the ability to generate synthetic time series data, which is representative of the real, original, data. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.