Abstract
Actual integration of high-tech devices brings opportunities for better monitoring, management and control of low voltage networks. In this new paradigm, efficient tools should cope with the great amount of dispersed and considerably distinct data to support smarter decisions in almost real time. Besides the use of tools to enable an optimal network reconfiguration and integration of dispersed and renewable generation, the impact evaluation of integrating storage systems, accurate load forecasting methods must be found even when applied to individual consumers (characterized by the high presence of noise in time series). As this effort becomes providential in the smart grids context, this article compares three different approaches: one based on Kernel Density Estimation, an alternative based on Artificial Neural Networks and a method using Support Vector Machines. The first two methods revealed unequivocal benefits when compared to a Naive method consisting of a simple reproduction of the last available day.

Abstract
While the transmission system benefits from a high observability, the distribution system has a relatively low level of observability. This problem is already being addressed with the deployment of smart meters, in an effort to make the smart grid concept a reality. Nevertheless, as observability increases, so too does the volume of data, which makes the development of advanced software tools a very important subject. In this paper, the application of image analysis techniques to a low voltage grid is explored, by converting voltage data into an image format, using a cognitive network to evaluate and cluster grid operating modes. The proposed method is applied to a 33-bus low voltage grid to evaluate voltage profiles at each bus and the associated technical limits (voltage limits alarms).

Abstract
The decarbonization of the economy, for which the contribution of power systems is significant, is a growing trend in Europe and in the world. In order to achieve the Paris Agreement's ambitious environmental goals, a substantial increase in the contribution of renewable sources to the energy generation mix is required. This trend brings about relevant challenges as the integration of this type of sources increases, namely in terms of the distribution system operation. In this paper, the challenges foreseen for future power systems are identified and the most effective approaches to deal with them are reviewed. The strategies include the development of Smart Grid technologies (meters, sensors, and actuators) coupled with computational intelligence that act as new sources of data, as well as the connection of distributed energy resources to distribution grids, encompassing the deployment of distributed generation and storage systems and the dissemination of electric vehicles. The impact of these changes in the distribution system as a whole is evaluated from a technical and environmental perspective. In addition, a review of management and control architectures designed for distribution systems is conducted. This article is categorized under: Energy Infrastructure > Systems and Infrastructure Energy Infrastructure > Economics and Policy

Abstract
Urban wastewater sector is being pushed to optimize processes in order to reduce energy consumption without compromising its quality standards. Energy costs can represent a significant share of the global operational costs (between 50% and 60%) in an intensive energy consumer. Pumping is the largest consumer of electrical energy in a wastewater treatment plant. Thus, the optimal control of pump units can help the utilities to decrease operational costs. This work describes an innovative predictive control policy for wastewater variable-frequency pumps that minimize electrical energy consumption, considering uncertainty forecasts for wastewater intake rate and information collected by sensors accessible through the Supervisory Control and Data Acquisition system. The proposed control method combines statistical learning (regression and predictive models) and deep reinforcement learning (Proximal Policy Optimization). The following main original contributions are produced: (i) model-free and data-driven predictive control; (ii) control philosophy focused on operating the tank with a variable wastewater set-point level; (iii) use of supervised learning to generate synthetic data for pre-training the reinforcement learning policy, without the need to physically interact with the system. The results for a real case-study during 90 days show a 16.7% decrease in electrical energy consumption while still achieving a 97% reduction in the number of alarms (tank level above 7.2 m) when compared with the current operating scenario (operating with a fixed set-point level). The numerical analysis showed that the proposed data-driven method is able to explore the trade-off between number of alarms and consumption minimization, offering different options to decision-makers.

Abstract
The increasing penetration of Distributed Energy Resources is changing the energy system by empowering consumers with the capacity to generate the electrical energy they need and sell its excess. This trend follows the EU strategy towards increasing competition and flexibility on the electricity market, as well as pushing the role of customers, expanding their rights and their involvement in energy communities (ECMs). Peer-to-Peer (P2P) energy markets appear as one of the possible solutions to accomplish these goals by providing direct energy trading between peers. Although P2P are being extensively addressed in the literature (e.g., market structures and platforms, experimental projects), few works offer a broad perspective of the different aspects involved in the actual implementation of these structures, as well as the real benefits that this type of markets can have for the players and for the system itself. This paper reviews business models related with ECMs and P2P markets, and the system benefits and main regulatory issues.

Abstract
The penetration of distributed generation in medium (MV) and low (LV) voltage distribution grids has been steadily increasing every year in multiple countries, thus creating new technical challenges in grid operation and motivating developments in distributed optimization for flexibility management. The traditional centralized optimal power flow (OPF) algorithm can solve technical constraints violation. However, computational efficiency, new technologies (e.g., edge computing) and control architectures (e.g., web-of-cells) are demanding for distributed approaches. This work formulates a novel distributed multi-period OPF for three-phase unbalanced grids that is essential when integrating energy storage units in operational planning (e.g., day-ahead) of LV or local energy community grids. The decentralized constrained optimization problem is solved with the alternating direction method of multipliers (ADMM) adapted for unbalanced LV grids and multi-period optimization problems. A 33-bus LV distribution grid is used as a case-study in order to define optimal battery storage scheduling along a finite time horizon that minimizes overall grid operational costs, while complying with technical constraints of the grid (e.g., voltage and current limits) and battery state-of-charge constraints.