Workshop Program

Large-scale digital twins are becoming a key component of Computational Wind Engineering, enabling high-fidelity CFD simulations over complex environments with urban infrastructure. However, the reconstruction and preparation of such models remain challenging due to the scale, geometric complexity, heterogeneous data sources, and the demanding requirements of CFD meshing and simulation workflows. This lecture and hands-on tutorial present best-practice guidelines for the creation of digital twins for CFD applications, with a focus on the Municipality of Genova, Italy. The potential of such models for multiple applications will be discussed, including urban wind assessment and the identification of wind-hazard prone regions. The practical session will demonstrate how these guidelines can be implemented in open-source environments, covering the complete workflow from geometry preparation, to mesh generation and the execution of wind simulations in OpenFOAM. Participants will gain practical insight into the challenges, limitations, and computational requirements associated with large-scale CFD digital twins.

The application of Computational Wind Engineering to cases of practical interest is often hindered by the need to model phenomena for which the standard tools implemented in available software are not sufficiently developed. This is the case, for instance, for synthetic turbulence generation and the modelling of permeable surfaces, both of which are commonly encountered in applications. In this seminar, we briefly present the difficulties that arise when including these effects in numerical simulations. Specifically, we first introduce the main theoretical aspects underlying their modelling strategies and then show how two freely available boundary conditions for OpenFOAM, named SynInflow and Pressure-Velocity Jump, can be used to address the problem. Participants will be guided through a first use of both tools, with emphasis on their strengths and current limitations.

This workshop provides hands-on CFD training to model urban wind flow and pollutant dispersion, accounting for air conditioning (AC) heat rejection and building greenery systems. It will introduce Large-eddy simulation for urban microclimates, considering realistic weather conditions (wind, solar radiation, humidity), transpiration by vegetation, and AC units as sensible heat sources. Training cases will take street canyons and cubic buildings as examples.

You will simulate baseline scenarios followed by distinct intervention cases. You will activate building AC heat rejection to analyze how buoyant plumes modify turbulent mixing, thermal stratification, and dispersion of traffic pollutants. Separately, you will activate a building-integrated greenery system by defining leaf area density, stomatal resistance, and transpiration rate via a User-Defined Function (UDF). The session allows you to evaluate how building greenery and anthropogenic heat alter wind flow, induce evaporative cooling, and affect pollutant dispersion.

This session is divided into two complementary parts addressing the modelling of downbursts and tornadoes. The hands-on training focuses on Large Eddy Simulation (LES) of tornado-like vortices and thunderstorm downburst winds for wind engineering applications. A brief overview of the governing physics of these non-synoptic extreme winds will highlight their key differences from synoptic wind systems. For tornadoes, two representative cases will be examined: (i) a stationary vortex in an idealized domain to analyze the flow field, and (ii) a translating vortex to investigate wind loading effects. For downbursts, a straight impinging jet flow will be reproduced with emphasis on wind-field characterization. The session will cover practical aspects of numerical modeling, including domain setup, boundary conditions, and meshing strategies, with particular attention to how modeling decisions influence the physical interpretation of results.

Urban microclimate modeling captures the coupled heat and moisture exchange between buildings, vegetation, soil, and the atmosphere. These processes drive urban heat islands, building energy demand, and outdoor thermal comfort. This full-day hands-on workshop introduces urbanMicroclimateFoam, an open-source multi-region OpenFOAM solver for high-resolution urban modeling. The solver combines a CFD model for turbulent airflow with coupled heat and moisture transport in porous materials, shortwave and longwave radiation, and a vegetation model for trees and green surfaces.

Through guided walkthroughs and live exercises, participants learn the solver structure, region coupling, boundary conditions, and material and vegetation property setup. The workshop progresses from a single building case to a street canyon comparing impervious, grass, and tree-lined ground covers, then through the complete workflow for a realistic urban geometry.

Designed for researchers with basic OpenFOAM or Linux experience, the workshop equips attendees to build, run, and analyze their own urban microclimate cases.

This session offers a hands-on introduction to Machine Learning (ML), progressing from foundational concepts to advanced practical applications. We begin by exploring common ML architectures to provide a self-contained overview of key concepts, then transition into a hands-on real-world engineering application focused on surrogate modeling to approximate CFD simulations using deep learning models.

Participants will follow an end-to-end pipeline, starting with data preprocessing, exploratory analysis, feature engineering, and model selection. The session covers critical stages including training, hyperparameter tuning, and rigorous performance evaluation, along with practical tips at each step. Attendees will build a functional tool capable of reducing hour-long simulations to milliseconds.

By the end of the workshop, participants will understand the end-to-end workflow for developing surrogate models and how to apply it across a range of engineering and scientific use cases where fast approximations of complex simulations are needed.

This two-hour workshop introduces practical methods for generating complex geometries and high quality computational grids for CFD simulations in sport aerodynamics using ANSYS SpaceClaim and ANSYS Fluent Meshing. Focusing on real applications such as cycling and running, the workshop emphasises the importance of accurately representing geometric complexity in order to obtain reliable aerodynamic predictions. Participants will work with an athlete geometry, exploring different approaches to geometry preparation, including manual cleaning tools and the shrink wrap functionality available in Fluent Meshing. Several meshing strategies suitable for external high Reynolds number flows will be introduced and discussed, highlighting their advantages and limitations for sport aerodynamics applications. The workshop concludes with selected examples and best practice guidelines for producing efficient and robust meshes, enabling reliable CFD analyses in sports aerodynamics.

This hands-on workshop guides participants through the end-to-end process of evaluating wind loads on structures using CFD, with explicit reference to established codes and standards in wind engineering. Beginning from a representative building geometry, attendees will progress through each stage of the simulation workflow, including domain sizing, boundary condition specification, turbulence modelling, and post-processing, while examining how decisions made at each step influence the resulting pressure distributions and integrated load quantities.

A central theme is the dialogue between numerical practice and codified design requirements. Participants will compare CFD-derived wind load estimates against values prescribed by relevant standards, and explore how choices such as mesh resolution, inflow turbulence intensity, and averaging time affect predicted loads and their code-compliant interpretation.

By the end of the workshop, participants will have a practical framework for conducting defensible CFD-based wind load evaluations and communicating results in the language of structural wind engineering practice.