Abstract
Metaheuristic optimization algorithms are increasingly used to reach near-optimal solutions for complex and large-scale problems that cannot be solved in due time by exact methods. Metaheuristics’ performance is, however, deeply dependent on their effective configuration and fine-tuning to align the algorithm’s search process with the specific characteristics of the problem that is being solved. Although the literature already offers some solutions for automatic algorithm configuration, these are usually either algorithm-specific or problem-specific, thus lacking the capability of being used for diverse metaheuristic models or diverse optimization problems. This work proposes a new approach for the automatic optimization of metaheuristic algorithms’ parameters based on a multi-agent system approach. The proposed model includes an automated fine-tuning process, which is used to optimize a given function in an algorithm- and problem-agnostic manner. Results show that the proposed model is able to achieve better optimization results than standard metaheuristic algorithms, with a negligible increase in the required execution time. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract
The share of renewable generation is growing worldwide, increasing the complexity of the grids operation to maintain its stability and balance. This leads to an increased need for designing new electricity markets (EMs) suited to this new reality. Simulation tools are widely used to experiment and analyze the potential impacts of new solutions, such as novel EM designs and power flow analysis and validation. This work introduces two web services for EMs' simulation and study, in addition to power flow evaluation and validation, namely the Elec-tricity Market Service (EMS) and Power Flow Service (PFS). EMS enables the simulation of two auction-based algorithms and the execution of three wholesale EMs. PFS creates and evaluates electrical grids from the transmission to distribution grids. Being published as web services facilitates their integration with other ser-vices, systems, or software agents. Combining them allows for the simulation of EMs from wholesale to local markets and testing if the results are compatible with a specific grid. This article presents a detailed description of each service and a case study of an electricity trading community participating in the MIBEL day-ahead market through an aggregator to reduce their energy bills. The results demonstrate the accuracy and usefulness of the proposed services.

Abstract
This paper proposes a novel approach for the provision of non-frequency ancillary service (AS) by consumers connected to low-voltage distribution networks. The proposed approach considers an asymmetric pool-based local market for AS negotiation, allowing consumers to set a flexibility quantity and desired price to trade. A case study with 98 consumers illustrates the proposed market-based non-frequency AS provision approach. Also, three different strategies of consumers' participation are implemented and tested in a real low-voltage distribution network with radial topology. It is shown that consumers can make a profit from the sale of their flexibility while contributing to keeping the network power losses, voltage, and current within pre-defined limits. Ultimately, the results demonstrate the value of AS coming directly from end-users.

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

Abstract