This research compares the frequency and duration of outages within distribution networks for main Nordic land countries (Sweden, Denmark, Finland, and Norway). In addition, this paper focuses on planned and unplanned outages for low- and medium-voltage networks; the consequences of outages for distribution networks and companies; and the level of discomfort experienced by consumers during both planned and unplanned outages. This study highlights the countries with the highest incidence of outages by collecting data from their official reports, compares the frequency and duration of unplanned outages, focuses on SAIFI (System Average Interruption Frequency Index), SAIDI (System Average Interruption Duration Index), and CAIDI (Customer Average Interruption Duration Index)-based outage indices, and examines their outage trends. 

Accurate short-term load forecasting is essential for energy-efficient building operations, however, many buildings lack sufficient historical data to train reliable predictive models. This paper investigates the use of instance-based transfer learning to improve forecasting performance in data-scarce buildings by reusing models trained on similar, data-rich buildings. We cluster buildings of various types based on their energy consumption profiles using dynamic time warping and hierarchical clustering. Within each cluster, we fine-tune XGBoost and LSTM models pretrained on source buildings (with sufficient data) to forecast the consumption in target buildings (with limited data) with access to increasing percentage of their original limited data of 7, 14, 21, 28, ... , 245 days. Our evaluation compares the performance against models trained from scratch using the same limited data. Our results show that transfer learning significantly improves forecasting accuracy in highly data-constrained settings, reducing RMSE by up to 40%. However, as more training data becomes available, its benefits diminish, and baseline models eventually outperform transferred ones beyond an average threshold of 140–168 days.

To support the integration of sustainable energy sources, accurate and reliable forecasting of photovoltaic (PV) power production and energy consumption is essential. Equally important is the ability to predict uncertainty bounds to enable robust scheduling in smart grids. Federated Learning (FL) is a privacy-preserving technique suitable for such forecasting, though applying it to tree-based models for multivariate time series remains challenging. We propose a novel multi-model tree-based FL architecture that supports client-side quantile regression to estimate uncertainty. By efficiently aggregating updates from multiple nodes, our approach improves forecast accuracy. Evaluations on grid data for PV production and consumption show substantial gains over non-federated models: CatBoost’s MAE for PV production dropped by 88.1%, LightGBM’s consumption error by 14.93%, and CatBoost’s consumption error by 79.5%. These gains are achieved while maintaining well-calibrated uncertainty bounds, with PICP values consistently above 0.96. Our reliable forecasts support more efficient energy scheduling and reduce costly imports from the main grid.

This study investigates the temporal and spatio-temporal correlations of solar power generation among different prosumers of Uppsala and Halmstad, Sweden. Using solar power generation data from seven prosumer in Uppsala and five in Halmstad, we evaluate the correlation of solar power production generation at specific locations correlates with itself over different time lags (autocorrelation).  In addition, we examine the spatiotemporal correlations of solar power production at various locations over a range of lags using time shifted cross correlation. These spatio-temporal correlations can facilitate the development of synchronized demand response strategies and dynamic energy pricing. Moreover, the time-shifted cross-correlation analysis assists in improving forecasting models for solar power generation. By identifying significant correlations between solar generation data from different locations and applying time shifts to account for variations in weather and sunlight exposure, operators can enhance the accuracy of their predictions. This methodology enables them to fill in missing data points by leveraging correlated information from neighboring regions. Consequently, more robust forecasting models can be developed, leading to better resource allocation, improved energy management, and reduced operational uncertainties in the grid. This research highlights the evaluation and potential of utilizing spatio-temporal and temporal correlations in solar power data to enhance energy management and planning.

Renewable Energy Communities (RECs), characterized by localized energy generation and consumption, are a key enabler for enhancing renewable energy utilization, cost-efficient planning and clean energy transition. However, optimizing RECs operations is challenging due to the complex interplay of different stakeholders with conflicting requirements. The complexity of managing such systems often leads to a lack of transparent and reliable decision-making, creating barriers for actors within the community. This paper explores the integration of Generative AI into Renewable Energy Communities (RECs) to enhance the transparency, explainability and accessibility of energy management systems (EMSs) that depend on solving optimization problems. We propose a novel framework, Chat-SGP, which uses generative AI to synthesize optimization modeling code to provide actionable, explainable insights for managing the REC operations. Our approach allows us to interact with the EMS through natural language queries, enhancing the system’s accessibility and user-friendliness. Our evaluation shows that using GPT-4 with in-context learning performs 96.72% accuracy on average in generating correct answers.

Load forecasting is a crucial topic in energy management systems (EMS) due to its vital role in optimizing energy scheduling and enabling more flexible and intelligent power grid systems. As a result, these systems allow power utility companies to respond promptly to demands in the electricity market. Deep learning (DL) models have been commonly employed in load forecasting problems supported by adaptation mechanisms to cope with the changing pattern of consumption by customers, known as concept drift. A drift magnitude threshold should be defined to design change detection methods to identify drifts. While the drift magnitude in load forecasting problems can vary significantly over time, existing literature often assumes a fixed drift magnitude threshold, which should be dynamically adjusted rather than fixed during system evolution. To address this gap, in this paper, we propose a dynamic drift-adaptive Long Short-Term Memory (DA-LSTM) framework that can improve the performance of load forecasting models without requiring a drift threshold setting. We integrate several strategies into the framework based on active and passive adaptation approaches. To evaluate DA-LSTM in real-life settings, we thoroughly analyze the proposed framework and deploy it in a real-world problem through a cloud-based environment. Efficiency is evaluated in terms of the prediction performance of each approach and computational cost. The experiments show performance improvements on multiple evaluation metrics achieved by our framework compared to baseline methods from the literature. Finally, we present a trade-off analysis between prediction performance and computational costs.

For efficient energy exchanges in smart energy grids under the presence of renewables, predictions of energy production and consumption are required. For robust energy scheduling, prediction of uncertainty bounds of Photovoltaic (PV) power production and consumption is essential. In this paper, we apply several Machine Learning (ML) models that can predict the power generation of PV and consumption of households in a smart energy grid, while also assessing the uncertainty of their predictions by providing quantile values as uncertainty bounds. We evaluate our algorithms on a dataset from Swedish households having PV installations and battery storage. Our findings reveal that a Mean Absolute Error (MAE) of 16.12W for power production and 16.34W for consumption for a residential installation can be achieved with uncertainty bounds having quantile loss values below 5W. Furthermore, we show that the accuracy of the ML models can be affected by the characteristics of the household being studied. Different households may have different data distributions, which can cause prediction models to perform poorly when applied to untrained households. However, our study found that models built directly for individual homes, even when trained with smaller datasets, offer the best outcomes. This suggests that the development of personalized ML models may be a promising avenue for improving the accuracy of predictions in the future.

Climate change and its impact on global sustainability are critical challenges, demanding innovative solutions that combine cutting-edge technologies and scientific insights. Quantum machine learning (QML) has emerged as a promising paradigm that harnesses the power of quantum computing to address complex problems in various domains including climate change and sustainability. In this work, we survey existing literature that applies quantum machine learning to solve climate change and sustainability-related problems. We review promising QML methodologies that have the potential to accelerate decarbonization including energy systems, climate data forecasting, climate monitoring, and hazardous events predictions. We discuss the challenges and current limitations of quantum machine learning approaches and provide an overview of potential opportunities and future work to leverage QML-based methods in the important area of climate change research.

In this paper, we address the optimal operation of energy communities, under energy production and consumption uncertainties. In the nominal case, the operational problem is formulated as the maximization of the profit of the community over a given time horizon. Inspired by the regulation adopted in Italy since 2020, the profit includes an incentive for the self-consumption realized at the community level in each time period. In the presence of uncertainties, we use a robust formulation aiming at maximizing the worst profit achievable when energy production and consumption deviate from their nominal values. The model results in a robust scheduling policy of battery charging/discharging, guaranteeing feasible operation of the community in all scenarios of the uncertainty set. On the other hand, numerical results show that the nominal scheduling policy may suffer from a high constraint violation probability.

Energy management within microgrids under the presence of large number of renewables such as photovoltaics is complicated due to uncertainties involved. Randomness in energy production and consumption make both the prediction and optimality of exchanges challenging. In this paper, we evaluate the impact of uncertainties on optimality of different robust energy exchange strategies. To address the problem, we present AIROBE, a data-driven system that uses machine-learning-based predictions of energy supply and demand as input to calculate robust energy exchange schedules using a multiband robust optimization approach to protect from deviations. AIROBE allows the decision maker to tradeoff robustness with stability of the system and energy costs. Our evaluation shows, how AIROBE can deal effectively with asymmetric deviations and how better prediction methods can reduce both the operational cost while at the same time may lead to increased operational stability of the system.