Descripción del proyecto
To achieve the EU's green transition goal, Building-integrated photovoltaics (BIPV) is widely adopted. However, the efficiency of most BIPV systems is limited as they primarily utilize the rooftop space of buildings. Using both the rooftop and facade surfaces of buildings can improve the surface utilization of BIPV systems, but it also poses new challenges, including the heterogeneity and incompleteness of sensor data for semantic 3D reconstruction of buildings, various unstable factors affecting BIPV design and layout, and the necessity to coordinate and integrate BIPV systems with other energy systems. This project aims to develop a Semantic 3D Energy Model (S3EM) for BIPV systems, which leverages multi-modal data to construct accurate 3D representations of solar radiation and electrical potential across building surfaces, enabling the optimal placement of BIPV components and their smooth integration with other energy systems. To achieve the research goal, this project proposes three work packages, including:1) Constructing high-precision and complete 3D semantic models of buildings by leveraging the complementarity of multi-modal data, 2) Optimizing the design and placement of BIPV components on building surfaces by considering multiple factors, 3) Integration and coordination of BIPV systems with other energy systems using multi-agent reinforcement learning methods. S3EM specifies the resources needed for this project, including the quality and capacity of the host, mentors, data, and experimental facilities. This project will enhance the scientific skills and innovation capability, expand research horizons, and establish research collaborations of the applicant. A two-way knowledge transfer approach in energy big data, building modeling and energy system analysis is proposed to ensure benefits between the applicant and the host. S3EM will make a significant contribution to the state of the art in renewable energy production forecasting and the EU climate goals.