DynANA

Dynamic Analysis of Gas Network Applications, developed at AIT Austrian Institute of Technology GmbH.

Overview

DynANA provides a dynamic simulation framework for gas networks with hydrogen injection.
It is component-based and includes models for sinks, sources, junctions, injections, and pipes, which can be connected to form different network configurations.

The simulated gas mixture consists of methane and hydrogen.
The underlying physical model is valid for a minimum methane mass fraction of 0.6, meaning it is not designed for pure hydrogen simulations.

A detailed description is available in the Master’s Thesis:
“Simulations of Hydrogen Blending in Gas Grids: Development of a Dynamic Framework for Multi-Node Injection” (link to be added).

---

Folder Structure

• src/ – Source code for network simulations
• test/ – Test files for individual components
• applications/ – Simulation projects
  • UseCase/ – Simple network with two injection setups. Results are described in the Master’s Thesis.
• template/ – Template folder for new simulation projects
  • Input/ – Files required to generate a network and define parameters
  • Results/ – Output data from DynANA simulations
  • Plots/ – Plots generated during data processing
  • DataProcessing.ipynb – Notebook for analyzing and plotting simulation results
  • Simulation.jl – Script for constructing and running simulations
  • run_simulation.sh – Shell script to run simulations on a cluster

---

Installation

Clone the repository:

git clone https://github.com/ait-energy/DynANA
cd DynANA

Activate the Julia environment:

julia --project=path/to/folder

NOTE: If running from the project directory, use --project=./

Instantiate the environment:

] instantiate

NOTE: Typing ] enters Julia’s package mode, shown as (DynANA) pkg>.

Install Python dependencies for notebook use:

uv sync

---

Usage

There are two main ways to construct a gas network with DynANA:

1. Using the Template Folder
   Modify the files in template/ to suit your project and run the simulation.

2. Manual Network Construction
   Read up on the ModelingToolkit, build your network manually, initialize the system and simulate it. (And maybe say a small prayer to the spaghetti monster for good measure.)

---

Running a Simulation

Either adapt and run:

./run_simulation.sh

or directly execute:

julia --project=./ Simulation.jl

---

Output Files

• Simulation.log – Logs information about the network construction and connection process.
• *_structure.csv – Defines pipe and node topology, parameters, and the mapping between element names and internal IDs used in simulations.
• *_results.csv – Contains results such as mass flow (q), pressure (p), and methane concentration (c) for each element.
  Pipe data may also include internal volume results if specified in the simulation file.
• simulation_time.csv – Lists the start time, simulation duration, and output time steps.
• Results_full_time.csv – Contains the complete set of unprocessed simulation data for all model variables.

---

Data Processing

Open and adapt the DataProcessing.ipynb notebook to analyze and visualize your results.

---

Citation

If you use DynANA.jl in your research, please cite this paper:

@unpublished{orsolits_advantages_2025,
  author    = {Orsolits, Dana and Maia-Novak, Ina and Strömer, Stefan and Reuter,  Stefan},
  title     = {On the Advantages of Dynamic Simulations when Modeling Multi-Node Hydrogen Blending},
  journaltitle = {Energy, Sustainability and Society},
  note      = {Manuscript under review at Energy, Sustainability and Society},
  year      = {2025}
}

Roadmap

Future development plans include:

• Extending the model to additional gas components (e.g., CO₂)
• Adding new components such as compressors and pressure-reduction valves