| This paper explores Explainable Artificial Intelligence (XAI) in the energy sector, highlighting its importance for enhancing transparency, user adoption and engagement. It introduces a User-Centric Explainable By-Design Framework for AI services, aimed at overcoming the ``black-box" challenges of AI models by prioritizing explainability and user interaction throughout the AI development process. The framework addresses the deployment of AI in applications such as predictive maintenance, anomaly detection, and photovoltaic (PV) forecasting, emphasizing the need for clear, understandable AI decisions to foster user trust and promote sustainable energy practices. By advocating for a systematic incorporation of XAI principles from data collection to model training and user feedback, the framework seeks to enhance the effectiveness and user experience of AI in the energy sector, supporting a transition towards more efficient and environmentally friendly energy systems. In general, the proposed framework provides a structured approach to achieving these goals, while future research and development in this area should focus on refining and testing the framework across various applications of the energy sector, from smart grids and renewable energy systems to energy efficiency services in residential and commercial buildings. |
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.