Improving energy efficiency in residential buildings is critical to combating climate change and reducing greenhouse gas emissions. Retrofitting existing buildings –that are major contributors to energy use– is therefore a key priority, particularly in regions with outdated building stock. Artificial Intelligence (AI) and Machine Learning (ML) can automate retrofit decision-making and find retrofit strategies. However, their implementation faces challenges of data availability, trust and alignment with trustworthiness guidelines, as well as compliance to AI regulations. This paper presents a trustworthy-by-design ML-based decision support framework that recommends energy efficiency strategies for residential buildings using minimal user-accessible inputs. The framework employs Conditional Tabular Generative Adversarial Networks (CTGAN) to augment limited and imbalanced data, while neural network-based multi-label classifier identifies potential combinations of retrofit measures. An Explainable AI (XAI) layer using SHAP is also incorporated to clarify the rationale behind recommendations, validate the model, and guide feature engineering. Two case studies on distinct datasets validate performance and replicability: i) a well-established, large Energy Performance Certificate (EPC) dataset for England and Wales; ii) an imbalanced post-retrofit dataset from Latvia (RETROFIT-LAT). Results demonstrate that the framework can handle diverse data conditions and improve performance up to 53% compared to the baseline model without XAI and synthetic data generation. Overall, the proposed framework provides a novel, user-friendly classification-based solution for building retrofit decision support that incorporates the trustworthiness aspects of transparency, human oversight, data governance, and fairness and aids stakeholders in achieving effective energy efficiency investments while aligning with AI regulation and ethical standards.

The rapid advancement of Artificial Intelligence (AI), particularly Machine Learning (ML) and Deep Learning (DL), has produced high-performance models widely used in various applications, ranging from image recognition and chatbots to autonomous driving and smart grid systems. However, security threats arise from the vulnerabilities of ML models to adversarial attacks and data poisoning, posing risks such as system malfunctions and decision errors. Meanwhile, data privacy concerns arise, especially with personal data being used in model training, which can lead to data breaches. This paper surveys the Adversarial Machine Learning (AML) landscape in modern AI systems, while focusing on the dual aspects of robustness and privacy. Initially, we explore adversarial attacks and defenses using comprehensive taxonomies. Subsequently, we investigate robustness benchmarks alongside open-source AML technologies and software tools that ML system stakeholders can use to develop robust AI systems. Lastly, we delve into the landscape of AML in four industry fields –automotive, digital healthcare, electrical power and energy systems (EPES), and Large Language Model (LLM)-based Natural Language Processing (NLP) systems– analyzing attacks, defenses, and evaluation concepts, thereby offering a holistic view of the modern AI-reliant industry and promoting enhanced ML robustness and privacy preservation in the future.

Renewable hydrogen can be produced with Power-to-Gas (PtG) systems, using renewable energy sources with minimal Greenhouse Gas (GHG) emissions. With this paper, the authors aim to provide clear insights into optimal PtG systems' techno-economic performance, in the case of renewable hydrogen blending gas grids such as that of the city of Athens. Two different production methods are investigated, namely which the investor owns the required energy generation units, and one in which the required energy procured from independent producers. Seven different underlying parametric scenarios are investigated, reflecting possible future costs developments according to scientific literature. The results show Levelized Costs of produced hydrogen ranging from 2.14 to 5 /kg. Production through own energy units is currently overall more economical. Energy and electrolysis costs comprise most of the costs, indicating the necessity considerable price reductions for green hydrogen to become competitive.

The DeployAI project develops a fully operational European AI-on-Demand Platform (AIoDP) to empower European industry-particularly SMEs and the public sector-with access to cutting-edge AI solutions. Focusing on European-developed, trustworthy, and ethical AI technologies, the AIoDP offers a comprehensive marketplace of AI products and services, including pre-trained models, tools, and reusable modules, supporting scalable deployment across Cloud and HPC infrastructures. The platform also features a Business Navigator, offering insights into the European AI ecosystem through interactive data and stakeholder mapping. Real-world use cases across Energy, Smart Cities, Earth Observation, and Logistics demonstrate the platform's potential to accelerate AI adoption, enhance productivity, and foster innovation within the European AI ecosystem.

AI4EF (Artificial Intelligence for Energy Efficiency) is an advanced, user-centric tool designed to support decision-making in building energy retrofitting and efficiency optimization. Leveraging machine learning (ML) and data-driven insights, AI4EF (Artificial Intelligence for Energy Efficiency) enables stakeholders such as public sector representatives, energy consultants, and building owners—to model, analyze, and predict energy consumption, retrofit costs, and environmental impacts of building upgrades. Featuring a modular framework, AI4EF includes customizable building retrofitting, photovoltaic installation assessment, and predictive modeling tools that allow users to input building parameters and receive tailored recommendations for achieving energy savings and carbon reduction goals. Additionally, the service incorporates a Training Playground for data scientists to refine ML models used by said framework. Finally, AI4EF provides access to the Enershare Data Space to facilitate seamless data sharing and access within the ecosystem. AI4EF’s compatibility with open-source identity management (Keycloak) enhances security and accessibility, making it adaptable for various regulatory and organizational contexts. This paper presents an overview of AI4EF’s architecture, its application in energy efficiency scenarios, and its potential for advancing sustainable energy practices through artificial intelligence (AI).

Short-term load forecasting (STLF) is crucial for the daily operation of power grids. However, the non-linearity, non-stationarity, and randomness characterizing electricity demand time series renders STLF a challenging task. Various forecasting approaches have been proposed for improving STLF, including neural network (NN) models which are trained using data from multiple electricity demand series that may not necessarily include the target series. In the present study, we investigate the performance of a special case of STLF, namely transfer learning (TL), by considering a set of 27 time series that represent the national day-ahead electricity demand of indicative European countries. We employ a popular and easy-to-implement feed-forward NN model and perform a clustering analysis to identify similar patterns among the load series and enhance TL. In this context, two different TL approaches, with and without the clustering step, are compiled and compared against each other as well as a typical NN training setup. Our results demonstrate that TL can outperform the conventional approach, especially when clustering techniques are considered.

Since its conceptualization in 2008, blockchain technology has advanced rapidly and been applied in multiple domains. In higher education, blockchain can be applied to develop ICT systems that can revolutionize student accreditation through certificate verification and micro-accreditations, which represent skills and other learning outcomes, in the form of digital/smart badges. While there are multiple studies that highlight the significance of blockchain in higher education and propose digital systems, few of those studies include the evaluation of such proposed systems by real users. As such, the research question of how useful a higher education blockchain system would be for its relevant stakeholders remains largely unanswered. In the research publication at hand, a blockchain-powered higher education platform was applied in the School of Electrical and Computer Engineering of the National Technical University of Athens, where it was used and evaluated by students and professors at the school. The evaluation of the platform was positive, and participants found that the smart badge functionality was among the most useful. Finally, the execution and evaluation of the pilot led to several lessons learned and policy recommendations towards dealing with existing barriers and further promoting blockchain in higher education.

This paper presents DeepTSF, a comprehensive machine learning operations (MLOps) framework aiming to innovate time series forecasting through workflow automation and codeless modeling. DeepTSF automates key aspects of the machine learning (ML) lifecycle, making it an ideal tool for data scientists and MLops engineers engaged in ML and deep learning (DL)-based forecasting. DeepTSF empowers users with a robust and user-friendly solution, while it is designed to seamlessly integrate with existing data analysis workflows, providing enhanced productivity and compatibility. The framework offers a front-end user interface (UI) suitable for data scientists, as well as other higher-level stakeholders, enabling comprehensive understanding through insightful visualizations and evaluation metrics. DeepTSF also prioritizes security through identity management and access authorization mechanisms. The application of DeepTSF in real-life use cases of the I-NERGY project has already proven DeepTSF’s efficacy in DL-based load forecasting, showcasing its significant added value in the electrical power and energy systems domain.

Hydrogen and methane, when produced via water electrolysis with the usage of electricity from renewable energy sources (RES), could be used to help decarbonize sectors of the economy where other alternatives are either unfeasible to apply, or more expensive. Given Greece's favorable potential with respect to RES, such as wind and solar power, said power-to-gas (PtG) renewable products are considered crucial in the country's effort to meet its national environmental and energy policy targets. Meanwhile, they are also in accordance with the country's commitment to the EU's broader 2030 and 2050 climate and energy targets in plans such as the Green Deal. The present study proposes several key scenarios involving variants of PtG investments for the case of the Greek energy system, examining the coupling potential with both the natural gas and electricity transmission systems. The analysis emphasizes on setting up the basis for realistic mathematical simulations and their latter accurate techno-economic investigation and optimization, while indicative key performance indicators are identified towards the consistent evaluation and benchmarking of the different investments in question.

Artificial intelligence (AI) can significantly enhance the decision-making processes in a plethora of key energy-related challenges such as energy supply and demand prediction, grid flexibility, asset maintenance and operation, and load forecasting. As such, it is expected to play a crucial role in the digitalisation and sustainability of the energy industry. Despite the potential benefits of AI and the wealth of research in the related fields, several factors, including a lack of in-house AI expertise, data interoperability issues, unclear regulations, and ethical concerns, prevent stakeholders from the Electric Power and Energy Systems (EPES) sector from taking a universal approach to the EPES value chain and fully utilizing AI's potential. This paper presents our work that aims to address the former challenges by proposing a holistic approach to utilizing, developing, and deploying state-of-the-art AI and data analytic services for the decision makers of the energy sector. The framework will be seamlessly integrated with the AI-on-Demand platform, i.e. Europe's one-stop-shop for AI assets. The paper introduces the I-NERGY framework and describes its bidirectional relation to the AIoD platform under the light of contributing to the EPES value chain optimal management, particularly for SMEs and non-tech industries.

The field of green finance has garnered growing international interest in recent years and can contribute to green development towards addressing climate change. This can be supported through emerging technologies in cross-domain data sharing and artificial intelligence (AI). The present study explores ways to leverage existing and innovative methods and technologies related to financial “green” products, such as green bonds alongside research programs aimed at promoting green investments. In this context, we investigate whether AI and cross-organisational and cross-domain data sharing can be effectively leveraged to address the lack of information on green investments. More specifically, we conduct a review on previous scientific research. Therefore, we demonstrate that by finding and exploiting data through various techniques, and by sharing the data among stakeholders, it is possible to optimise and solve issues that will promote sustainable development, thus paving an innovative and necessary path towards the green transition. Based on the derived common acceptance of the necessity to enhance green finance, we propose techniques and technologies for the use of cross-domain data sharing and AI. As proof-of-concept, we subsequently describe two ongoing case studies in the energy and green deal domains, supported by the emerging technology of European Data Spaces. Overall, the analysis and compilation of the derived results can be used by stakeholders including researchers, financial institutions, and investors toward the next steps for the development of green finance.

The present study aims to evaluate the current fuzzy landscape of Trustworthy AI (TAI) within the European Union (EU), with a specific focus on the energy sector. The analysis encompasses legal frameworks, directives, initiatives, and standards like the AI Ethics Guidelines for Trustworthy AI (EGTAI), the Assessment List for Trustworthy AI (ALTAI), the AI act, and relevant CEN-CENELEC standardization efforts, as well as EU-funded projects such as AI4EU and SHERPA. Subsequently, we introduce a new TAI application framework, called E-TAI, tailored for energy applications, including smart grid and smart building systems. This framework draws inspiration from EGTAI but is customized for AI systems in the energy domain. It is designed for stakeholders in electrical power and energy systems (EPES), including researchers, developers, and energy experts linked to transmission system operators, distribution system operators, utilities, and aggregators. These stakeholders can utilize E-TAI to develop and evaluate AI services for the energy sector with a focus on ensuring trustworthiness throughout their development and iterative assessment processes.

In order to support decision-making problems on the energy sector, like energy forecasting and demand prediction, analytics services are developed that assist users in extracting useful inferences on energy related data. Such analytics services use AI techniques to extract useful knowledge on collected data from energy infrastructure like smart meters and sensors. The big data value chain describes the steps of big data life cycle from collecting, pre-processing, storing and querying energy consumption data for high-level user-driven services. With the exponential growth of networking capabilities and the Internet of Things (IoT), data from the energy sector is arriving with a high throughput taking the problem of calculating big data analytics to a new level. This research will review existing approaches for big data energy analytics services and will further propose a framework for facilitating AI-enabled energy analytics taking into consideration all the requirements of analytics services through the entire big data value chain from data acquisition and batch and stream data ingestion, to creating proper querying mechanisms. Those query mechanisms will in turn enable the execution of queries on huge volumes of energy consumption data with low latency, and establishing high-level data visualizations. The proposed framework will also address privacy and security concerns regarding the big data value chain and allow easy applicability and adjustment on various use cases on energy analytics.

Power-to-Gas (P2G) is an emerging technology aiming to contribute towards addressing the climate change andenvironmental degradation. Yet, numerous factors need to be taken into consideration to for practical P2G applications.Digital Twins (DT)

Due to the rapid development of Internet of Things (IoT) technology during the last decade, there has been recorded asignificant growth in the size of data collected by data warehouses, especially on the ones connected to sensors and meterson the

Effective storing and querying of building energy consumption data is crucial, since the energy sector accounts for almost 40% of the energy consumed worldwide. However, the rapid growth of networking capabilities has led to the storage needs for energy consumption data in data lakes and warehouses becoming unmanageable, making querying an inefficient task. To effectively store such vast amounts with regard to their dimensions and their semantic interpretation, many semantic ontology models have been developed. Data sharing mechanisms are then applied on top of data warehouses to extract useful knowledge for user related purposes. It is recognized, though, that distributed query engines are capable of querying only dynamic big data, while semantic mechanisms are confined to querying semantic ontologies. In this paper, we propose a framework that facilitates efficient querying of heterogenous data sources of dynamic and semantic data including graph ontologies for metadata representation. The framework also establishes high level visualization services by employing well-known visualization tools. Finally, particular emphasis is placed on securing the framework with proper resource identity management. The framework is developed in the context of DigiBUILD, an Horizon Europe funded project aiming at providing a digital logbook for analytic services on the building sector.

type. These characteristics are then used to distinguish the clusters that would be most likely to aid with the DR schemes would fit each cluster. Finally, we conceptualize a DR system that combines forecasting, clustering and a price-based demand projection engine to produce daily individualized DR recommendations and pricing policies for prosumers participating in the program. The results of this study can be useful for network operators and utilities that aim to develop targeted DR programs for groups of prosumers within flexible energy communities

Recently the energy sector undergoes a rapid transformation that revolves around digitalization, decentralization anddemocratization. This is because global energy crisis contributes to rising poverty, raising prices and slowing economies.At the

The present study proposes clustering techniques for designing demand response (DR) programs for commercial and residential prosumers. The goal is to alter the consumption behavior of the prosumers within a distributed energy community in Italy. This aggregation aims to: a) minimize the reverse power flow at the primary substation, occuring when generation from solar panels in the local grid exceeds consumption, and b) shift the system wide peak demand, that typically occurs during late afternoon. Regarding the clustering stage, we consider daily prosumer load profiles and divide them across the extracted clusters. Three popular machine learning algorithms are employed, namely k-means, k-medoids and agglomerative clustering. We evaluate the methods using multiple metrics including a novel metric proposed within this study, namely peak performance score (PPS). The k-means algorithm with dynamic time warping distance considering 14 clusters exhibits the highest performance with a PPS of 0.689. Subsequently, we analyze each extracted cluster with respect to load shape, entropy, and load types. These characteristics are used to distinguish the clusters that have the potential to serve the optimization objectives by matching them to proper DR schemes including time of use, critical peak pricing, and real-time pricing. Our results confirm the effectiveness of the proposed clustering algorithm in generating meaningful flexibility clusters, while the derived DR pricing policy encourages consumption during off-peak hours. The developed methodology is robust to the low availability and quality of training datasets and can be used by aggregator companies for segmenting energy communities and developing personalized DR policies.

In power grids, short-term load forecasting (STLF) is crucial as it contributes to the optimization of their reliability, emissions, and costs, while it enables the participation of energy companies in the energy market. STLF is a challenging task, due to the complex demand of active and reactive power from multiple types of electrical loads and their dependence on numerous exogenous variables. Amongst them, special circumstances, such as the COVID-19 pandemic, can often be the reason behind distribution shifts of load series. This work conducts a comparative study of Deep Learning (DL) architectures, namely Neural Basis Expansion Analysis Time Series Forecasting (N-BEATS), Long Short-Term Memory (LSTM), and Temporal Convolutional Networks (TCN), with respect to forecasting accuracy and training sustainability, meanwhile examining their out-of-distribution generalization capabilities during the COVID-19 pandemic era. A Pattern Sequence Forecasting (PSF) model is used as baseline. The case study focuses on day-ahead forecasts for the Portuguese national 15-minute resolution net load time series. The results can be leveraged by energy companies and network operators (i) to reinforce their forecasting toolkit with state-of-the-art DL models; (ii) to become aware of the serious consequences of crisis events on model performance; (iii) as a high-level model evaluation, deployment, and sustainability guide within a smart grid context.