HiDALGO2 D5.5 Research Advancements of Pilots

HiDALGO2 focuses on the development of high-fidelity simulations to address critical global challenges. This document builds on the foundations of D5.4 by highlighting the research advances achieved within the HiDALGO2 project between M24 and M35 in the five pilot applications: Urban Air Project, Urban Buildings, Renewable Energy Sources, Wildfires and Material Transport in Water. It emphasizes enhancements in solver modelling, accuracy, and versatility, alongside adjustments to the roadmaps based on unforeseen challenges and newly found insights.
For each pilot, specific advancements have been achieved:
•    Urban Air Project (UAP): Upgraded solvers, including the integration of RedSim with multi-node, multi-GPU-accelerated configurations, which have significantly improved the simulation of urban airflow and pollutant dispersion. Enhanced pre-processing tools utilizing OpenStreetMap to further support detailed urban environmental modelling. However, to incorporate atmospheric physics features into this pilot and due to a shift in focus, the integration of the digital twin and reduced models using MPI with GPU support and the overall development of the UAP-FOAM workflow have been postponed to the next reporting period.

•    Urban Buildings (UB): Progress was made in occupancy modelling, weather and geolocation integration, and energy output simulation, enhancing UB’s capacity to model energy consumption and indoor environmental conditions. A real-time data validation framework has been implemented, along with a user interface that supports interactive weather and scenario configurations. Planned future additions, such as heating regulation models, aim to further enhance the pilot’s simulation of urban building energy and comfort metrics. Challenges in dealing with specific geometric complexities pose a challenge in the generation of watertight meshes.

•    Renewable Energy Sources (RES): This pilot is advancing energy prediction models for solar and wind power generation, with progress in coupling models for enhanced accuracy in prediction. The lack of real-world data remains a bottleneck for certain aspects; however, advancements in small and large-scale photovoltaic models provide alternative data sources. Future work includes further refinement of prediction models using uncertainty quantification and visualization improvements.

•    Wildfires (WF): The WF pilot has made substantial advancements in simulating wildfire spread in both rural and urban settings, with updated preprocessing for historical wildfire events and coupling of the WRF model to simulate atmospheric interactions. Work is underway on GPU-enabled ensemble simulations, validation of surrogate models, automated vegetation mapping, and VR-ready visualization of fire spread scenarios to inform wildfire risk assessments at varying landscape scales.

•    Material Transport in Water (MTW): This new pilot focuses on modelling temperature transport in aquatic environments. Notable achievements include successful fluid-particle coupling and inclusion of temperature transport inside fluid and particles. This model development helps to better understand the fundamental aspects of heat transfer in water and aquatic environments.

This report provides a comprehensive assessment of each pilot’s progress, emphasizing the technical and operational challenges encountered and contributes to WP3 - Exascale Support for Global Challenges and WP4 - Data Exploration and Visualisation. The reported results showcase significant advancements, lay the groundwork for future deliverables, and reinforce HiDALGO2's goal of addressing global challenges through cutting-edge HPC solutions.