Installation

Installation#

The subsequently described installation steps are demonstrated as shell commands, where the path before the % sign denotes the directory in which the commands following the % should be entered.

Option B: Manual installation (All Platforms)#

Clone the Repository#

First of all, clone the Open-TYNDP repository using the version control system git in the command line.

$ git clone https://github.com/open-energy-transition/open-tyndp.git

Install Python Dependencies#

Preferred method: pixi#

PyPSA-Eur, and consequently Open-TYNDP, relies on a set of other Python packages to function. We manage these using pixi. Once pixi is installed, you can activate the project environment for your operating system and have access to all the PyPSA-Eur dependencies from the command line:

$ pixi shell -e open-tyndp

Tip

You can also set up automatic shell activation in several popular editors (e.g. in VSCode or Zed). Refer to the pixi documentation for the most up-to-date options.

Note

We don’t currently support linux operating systems using ARM processors since certain packages, such as PySCIPOpt, require being built from source.

Legacy method: conda#

If you cannot access pixi on your machine, you can also install using conda (or mamba/micromamba). To do so, we highly recommend you install from one of our platform-specific environment files:

  • For Intel/AMD processors:

    • Linux: envs/default_linux-64.pin.txt

    • macOS: envs/default_osx-64.pin.txt

    • Windows: envs/default_win-64.pin.txt

  • For ARM processors:

    • macOS (Apple Silicon): envs/default_osx-arm64.pin.txt

    • Linux (ARM): Currently not supported via lock files; requires building certain packages, such as PySCIPOpt, from source

$ conda update conda

$ conda create -n open-tyndp -f envs/default_linux-64.pin.txt # select the appropriate file for your platform

$ conda activate open-tyndp

These platform-specific files have locked dependencies, to ensure reproducibility. If you are having difficulties with the above files, you can also install directly from the un-locked environment YAML file (not recommended):

$ conda update conda

$ conda env create -n open-tyndp -f envs/environment.yaml

$ conda activate open-tyndp

Install a Solver#

PyPSA passes the PyPSA-Eur network model to an external solver for performing the optimisation. PyPSA is known to work with the free software

and the non-free, commercial software (for some of which free academic licenses are available)

For installation instructions of these solvers for your operating system, follow the links above. Commercial solvers such as Gurobi and CPLEX currently significantly outperform open-source solvers for large-scale problems, and it might be the case that you can only retrieve solutions by using a commercial solver. Nevertheless, you can still use open-source solvers for smaller problems.

See also

Instructions how to install a solver in the documentation of PyPSA

Note

The rules cluster_network solves a mixed-integer quadratic optimisation problem for clustering. The open-source solvers HiGHS, Cbc and GlPK cannot handle this. A fallback to SCIP is implemented in this case, which is included in the standard environment specifications. For an open-source solver setup install for example HiGHS and SCIP in your conda environment on OSX/Linux.

To install the default solver Gurobi, run

$ conda activate open-tyndp
$ conda install -c gurobi gurobi"=12.0.1"

Additionally, you need to setup your Gurobi license.

To use Xpress, install the xpress Python package and ensure you have:

  • XPRESSDIR environment variable pointing to your Xpress installation

  • XPAUTH_PATH environment variable pointing to your license directory

  • A valid Xpress license file

Then configure the solver in your config file:

solving:
  solver:
    name: xpress
    options: xpress-default

For GPU-accelerated solving, use:

solving:
  solver:
    name: xpress
    options: xpress-gpu
Contents