Abstract

Abstract
Deploying renewable energy communities, self-consumption and local energy markets are one of the ways to contribute to the energy system decarbonization by increasing the renewable energy share in the production mix and contributing to a better local balancing. However, how collective self-consumption structures are regulated has a direct impact on the flexibility of the energy sharing mechanisms and business models that can be set up. This paper compares and discusses how the European Union directives on self-consumption have been transposed to the national regulations of Portugal, Spain and France, providing a detailed regulatory discussion on the definition of basic concepts such as individual and collective self-consumption and renewable energy communities, proximity rules among members, energy sharing mechanisms and energy allocation coefficients, how the energy surplus is managed in each case, or how the grid access tariffs are modified to account for the self-consumed energy. The study highlights that dynamic allocation coefficients provide significant advantages for collective self-consumption by improving energy allocation efficiency, enabling advanced business models, and facilitating the integration of local energy markets, as it is the case in Portugal and France, while their absence in Spain limits these opportunities. The work also highlights the trade-off between flexible energy sharing and implementation complexity, and the role of digital tools to operationalize energy communities. Suggestions on potential regulatory improvements for all countries are also proposed. © 2025

Abstract

Abstract
Explainable AI (xAI) emerged as one of the ways of addressing the interpretability issues of the so-called black-box models. Most of the xAI artifacts proposed so far were designed, as expected, for human users. In this work, we posit that such artifacts can also be used by computer systems. Specifically, we propose a set of metrics derived from LIME explanations, that can eventually be used to ascertain the quality of each output of an underlying image classification model. We validate these metrics against quantitative human feedback, and identify 4 potentially interesting metrics for this purpose. This research is particularly useful in concept drift scenarios, in which models are deployed into production and there is no new labelled data to continuously evaluate them, becoming impossible to know the current performance of the model.

Abstract
Accurate evaluation of forecasting models is essential for ensuring reliable predictions. Current practices for evaluating and comparing forecasting models focus on summarising performance into a single score, using metrics such as SMAPE. While convenient, averaging performance over all samples dilutes relevant information about model behaviour under varying conditions. This limitation is especially problematic for time series forecasting, where multiple layers of averaging-across time steps, horizons, and multiple time series in a dataset-can mask relevant performance variations. We address this limitation by proposing ModelRadar, a framework for evaluating univariate time series forecasting models across multiple aspects, such as stationarity, presence of anomalies, or forecasting horizons. We demonstrate the advantages of this framework by comparing 24 forecasting methods, including classical approaches and different machine learning algorithms. PatchTST, a state-of-the-art transformer-based neural network architecture, performs best overall but its superiority varies with forecasting conditions. For instance, concerning the forecasting horizon, we found that PatchTST (and also other neural networks) only outperforms classical approaches for multi-step ahead forecasting. Another relevant insight is that classical approaches such as ETS or Theta are notably more robust in the presence of anomalies. These and other findings highlight the importance of aspect-based model evaluation for both practitioners and researchers. ModelRadar is available as a Python package.

Abstract