Organizers
Lead Organizer
Assoc. Prof. Dr. Youssef Jouane
CESI École d'Ingénieurs – CESI LINEACT Laboratory (UR 7527),
Parc Club des Tanneries, 2 allée des Foulons, 67380 Strasbourg, France
yjouane@cesi.fr
Co-Organizer
Assoc. Prof. Dr. Ilyass Abouelaziz
CESI LINEACT Laboratory (UR 7527),
7 bis Av. Robert Schuman, 51100 Reims, France
iabouelaziz@cesi.fr
Co-Organizer (pending confirmation)
Prof. Dr. Djaffar Ould Abdeslam
Université de Haute-Alsace, IRIMAS Laboratory, Mulhouse, France
Abstract and Scope
The integration of solar energy into the built environment is undergoing a major transition.
Building-Integrated Photovoltaics (BIPV) and smart microgrids require advanced operational
strategies to handle variable weather conditions, complex urban shading, and dynamic loads.
While Artificial Intelligence (AI) and data-driven methods offer promising tools for PV
forecasting and system control, current challenges lie in model robustness, physical
consistency, and computational efficiency.
This Special Session focuses on advanced AI and data-driven solutions designed to
optimize the lifecycle of building energy systems, from initial potential assessment to
real-time grid integration. A key emphasis is placed on bridging the gap between the physical
constraints of PV devices and modern machine learning, ensuring that AI models remain
physically consistent, computationally sustainable (frugal AI), and highly integrable with
smart microgrids and Electric Vehicle (EV) infrastructures.
Topics of Interest
Topics of interest include, but are not limited to:
-
Physics-Informed and Green AI for PV Systems: Integration of physical laws
(thermodynamics, PV electrical curves) into neural network architectures (PINNs); development
of lightweight, low-power AI models for edge-computing and embedded sensors.
-
Digital Twins, BIM, and Spatial AI for BIPV: Coupling of Building Information
Modeling (BIM), 3D photogrammetry, GIS, and deep learning for automated spatial potential
assessment and urban solar planning.
-
Advanced Multi-Horizon PV Forecasting and Fault Diagnostics: Data-driven,
high-resolution forecasting models; explainable AI (XAI) for anomaly detection, degradation
monitoring, and system diagnostics in smart buildings.
-
Smart Grid Integration, V2B/V2G, and Microgrid Control: AI-driven control and
optimization strategies for co-located BIPV, stationary battery storage (BESS), and EV charging
stations under dynamic grid signals.
-
Privacy-Preserving and Distributed Data Architectures: Applications of
federated learning, edge computing, and smart metering for decentralized energy management in
building portfolios.
Motivation and Timeliness
Three recent developments make this session particularly relevant in 2026.
Urban Solar Expansion
Global regulatory frameworks (such as the EU Energy Performance of Buildings Directive)
are mandating solar integration in new and renovated buildings. Predicting and managing
generation in dense, shaded urban environments has become an urgent power-engineering
priority.
Computational Sustainability
With the rising scrutiny over AI's own carbon and water footprint, the energy sector
must champion "Green AI." Developing lightweight, physics-constrained models that do not
rely on carbon-intensive cloud training is a technical and ethical necessity.
Dynamic Sector Coupling
The rapid integration of EVs and stationary batteries within building networks requires
decentralized, low-latency, and robust control algorithms. This session addresses these
multi-variable optimization challenges at the boundary of power electronics, power
systems, and computer science.
Organizer Biographies
Dr. Youssef Jouane is a researcher and faculty member at CESI École d'Ingénieurs,
affiliated with the CESI LINEACT laboratory (UR 7527). His research spans the physics of
organic and hybrid photovoltaic materials, building-integrated PV systems, and AI-based
energy prediction. He developed the BIM-AITIZATION methodology, which combines
photogrammetric point-cloud acquisition, BIM semantic modeling, and deep learning for
automated BIPV energy prediction and building decarbonization. He is currently
supervising a doctoral thesis on AI-driven energy management for industrial BIPV
buildings (co-funded by Région Grand Est and CESI). He has published in
Solar Energy, Energy and Buildings, Electric Power Systems
Research, Building Simulation, Optics Express, and
Organic Electronics, and has served as reviewer for Solar Energy,
IEEE Internet of Things, Journal of Building Engineering, and
Renewable Energy. He presented at ICRERA 2023 and ICRERA 2024.
Dr. Ilyass Abouelaziz is an Associate Professor at CESI LINEACT, Reims, France. His
research interests include photogrammetry, computer vision, and deep learning applied to
building simulation and energy systems. He is the co-developer of the BIM-AITIZATION
methodology and first author of the validation work published in Building
Simulation (2024).
Prof. Dr. Djaffar Ould Abdeslam is Full Professor at the Université de Haute-Alsace
(IRIMAS laboratory) and IEEE Senior Member. His research addresses artificial neural
networks for power system identification and control, smart metering, power quality,
smart buildings, microgrids, and the integration of EV charging with distributed
renewable generation. He has supervised more than twenty doctoral theses and is an
active participant in IEEE industrial applications and power electronics conferences.
Target Audience
Researchers and power systems engineers, control theorists, and computer scientists
working at the interface of renewable energy, machine learning, smart buildings, and
EV integration.
Expected Contributions
6–10 paper submissions are anticipated, drawing from research groups in France,
Germany, Morocco, Japan, and Turkey with whom the organizers have existing
collaborations. Depending on the quality and number of submissions, the organizers may
consider publishing a special post-conference issue in the conference proceedings or
elsewhere.
