DTCC Sim

DTCC Sim is a simulation package for DTCC Platform. It provides simulation utilities and example workflows built around FEniCSx.

This project is part of the Digital Twin Platform (DTCC Platform) developed at the Digital Twin Cities Centre supported by Sweden’s Innovation Agency Vinnova under Grant No. 2019-421 00041.

Documentation

This project is documented as part of the DTCC Platform Documentation.

Dataset Integration

dtcc-sim registers simulation datasets into the dtcc_core.datasets registry when dtcc_sim.datasets is imported:

import dtcc_core.datasets as datasets
import dtcc_sim.datasets

result = datasets.urban_wind_simulation(bounds=[xmin, ymin, xmax, ymax])

Simulation datasets follow the same descriptor contract as core datasets:

• format=None returns a Python simulation result.
• format=<value> returns serialized bytes for direct download paths.
• describe() exposes data_category="simulation", result_kind, python_return_type, supported_formats, and timeout_hint.

Current simulation datasets:

Dataset	Python result	Formats	Notes
urban_heat_simulation	dolfinx.fem.Function	xdmf	Steady-state heat equation. xdmf is multi-file.
air_quality_field	dtcc_core.model.VolumeMesh	xdmf	PDE-smoothed sensor field. xdmf is multi-file.
urban_wind_simulation	dtcc_core.model.VolumeMesh	pb	CFD wind result with velocity, pressure, and speed fields.
traffic_simulation	dtcc_core.model.RoadNetwork	pb	Static user-equilibrium road assignment with synthetic DeSO demand.

The service wrapper strips format before running a dataset, then serializes the returned Python object in service.results.handle_result(). This is intentional: it preserves companion files for multi-file outputs such as xdmf.

Installation

Install Conda

The first step is to install (mini) Conda. Follow these instructions:

https://www.anaconda.com/docs/getting-started/miniconda/

Install FEniCSx

The next step is to create and activate a new environment fenicsx-env:

source ~/miniconda3/bin/activate
conda create -n fenicsx-env python=3.12
conda activate fenicsx-env

Note the use of Python v3.12 which is required for installation of DTCC (below).

Next, we install FEniCSx and some other packages:

conda install -c conda-forge fenics-dolfinx mpich pyvista

Install DTCC Platform

Install DTCC Platform via PyPi:

pip install dtcc dtcc-viewer

Install DTCC TetGen wrapper

Install the DTCC TetGen wrapper. This needs to be installed from source since we don't yet provide a PyPi package:

git clone git@github.com:dtcc-platform/dtcc-tetgen-wrapper.git
cd dtcc-tetgen-wrapper
bash vendor_tetgen.sh
pip install .

Activating the environment (in new sessions)

If you have followed all the above instructions, you should have a Conda environment that has both FEniCSx and DTCC Platform.

To activate the environment in new sessions (terminals), run the commands

source ~/miniconda3/bin/activate
conda activate fenicsx-env

Testing the environment

To test the environment, run the following commands from inside the dtcc-repository:

cd sandbox
python build_volume_mesh_gbg.py
python solve_poisson.py

Authors (in order of appearance)

• Anders Logg

License

This project is licensed under the MIT license.

Copyrights are held by the individual authors as listed at the top of each source file.

Community guidelines

Comments, contributions, and questions are welcome. Please engage with us through Issues, Pull Requests, and Discussions on our GitHub page.