Abstract
The decarbonisation of the building sector is crucial for Portugal's goal of achieving economy-wide carbon neutrality by 2050. To mobilize communities towards energy efficiency measures, it is important to understand the primary drivers and barriers that must be overcome through policymaking. This paper aims to review existing Energy Policies and Actions (EPA) in Portugal and assess their effectiveness in improving Energy Efficiency (EE) and reducing CO2 emissions in the building sector. The Local Energy Planning Assistant (LEPA) tool was used to model, test, validate and compare the implementation of current and alternative EPAs in the North of Portugal, including the national EE plan. The results indicate that electrification of heating and cooling, EE measures, and the proliferation of Renewable Energy Sources (RES) are crucial for achieving climate neutrality. The study found that the modelling of alternative EPAs can be improved to reduce investment costs and increase Greenhouse Gas (GHG) emissions reduction. Among the alternatives assessed, the proposed one (Alternative 4) presents the best returns on investment in terms of cost savings and emissions reduction. It allows for 52% investment cost savings in the residential sector and 13% in the service sector when compared to the current national roadmap to carbon neutrality (Alternative 2). The estimated emission reduction in 2050 for Alternative 4 is 0.64% for the residential sector and 3.2% for the service sector when compared to Alternative 2.

Abstract
Energy communities have been designed to empower consumers while maximizing the self-consumption of local renewable energy sources (RESs). Their presence in distribution systems can result in strong modifications in the operation and management of such systems, moving from a centralized operation to a distributed one. In this scope, this work proposes a distributed community-based local energy market that aims at minimizing the costs of each community member, accounting for the technical network constraints. The alternating direction method of multipliers (ADMM) is adopted to distribute the market, and preserve, as much as possible, the privacy of the prosumers' assets, production, and demand. The proposed method is tested on a 10-bus medium voltage radial distribution network, in which each node contains a large prosumer, and the relaxed branch flow model is adopted to model the optimization problem. The market framework is proposed and modeled in a centralized and distributed fashion. Market clearing on a day-ahead basis is carried out taking into account actual energy exchanges, as generation from renewable sources is uncertain. The comparison between the centralized and distributed ADMM approach shows an 0.098% error for the nodes' voltages. The integrated OPF in the community-based market is a computational burden that increases the resolution of the market dispatch problem by about eight times the computation time, from 200.7 s (without OPF) to 1670.2 s. An important conclusion is that the proposed market structure guarantees that P2P exchanges avoid the violation of the network constraints, and ensures that community agents' can still benefit from the community-based architecture advantages.

Abstract
A common problem in distribution planning is the scarcity of historic data (training examples) relative to the number of variables, meaning that most data-driven techniques cannot be applied in such situations, due to the risk of overfitting. Thus, the suitable regression techniques are restrained to efficient models, preferably with embedded regularization features. This article compares three of these techniques: LASSO, Bayesian and CMLR (Conditioned multi-linear regression - a new approach developed within the scope of a project with a distribution company). The results showed that each technique has its own advantages and limitations. The Bayesian regression has the main advantage of providing inherent confidence intervals. The LASSO is a very economic and efficient regression tool. The CMLR is versatile and provided the best performance.A common problem in distribution planning is the scarcity of historic data (training examples) relative to the number of variables, meaning that most data-driven techniques cannot be applied in such situations, due to the risk of overfitting. Thus, the suitable regression techniques are restrained to efficient models, preferably with embedded regularization features. This article compares three of these techniques: LASSO, Bayesian and CMLR (Conditioned multi-linear regression - a new approach developed within the scope of a project with a distribution company). The results showed that each technique has its own advantages and limitations. The Bayesian regression has the main advantage of providing inherent confidence intervals. The LASSO is a very economic and efficient regression tool. The CMLR is versatile and provided the best performance.

Abstract
This paper explores the application of Deep Learning techniques to forecast electricity market prices. Three Deep Learning (DL) techniques are tested: Dense Neural Networks (DNN), Long Short-Term Memory Networks (LSTM) and Convolutional Neural Networks (CNN); and two non-DL techniques: Multiple Linear Regression and Gradient Boosting (GB). First, this work compares the forecast skill of all techniques for electricity price forecasting. The results analysis showed that CNN consistently remained among the best performers when predicting the most unusual periods such as the Covid19 pandemic one. The second study evaluates the potential application of CNN for automatic feature extraction over a dataset composed by multiple explanatory variables of different types, overcoming part of the feature selection challenges. The results showed that CNNs can be used to reduce the need for a variable selection phase.

Abstract
In every critical infrastructure system, unexpected events and outages have the potential to cause massive impacts, affecting people and the economy, such as in the power power grid blackouts. To avoid similar incidents in the future, extensive research is necessary to improve resilience and reliability of power grids. This work presents a Transmission Expansion Planning (TEP) model that confronts the largely adopted deterministic security criteria N-1 versus an AC-Cascade Failure Model (AC-CFM) analysis. The main goal is to highlight the importance of cascade failure analysis to increase power system resilience. Tests over the NREL-118 system verify the AC-CFM coupling in TEP models, demonstrating its benefits for assuming a risky proneness behavior for reaching long-term power grid resilience.

Abstract
Energy storage systems, integrated in Renewable Energy Communities (REC), are enabling the development of operation strategies together with Photovoltaic (PV) systems. Additionally, Local Energy Markets (LEM) are emerging mechanisms to enable local energy trading in RECs, the integration of storage systems can increase the community energy savings and profits. In this context, a market environment was modelled as a Markov Decision Process (MDP). In this scope, an Agent Based Model (ABM) using the Q-Learning mechanism was used to implement and to simulate a LEM and its interaction with the Wholesale Market (WSM), also considering an architecture with storage systems. The developed model was tested considering real data regarding energy consumption and PV generation. The paper describes and discusses the obtained market strategy and the profits that can be obtained with this approach.