multi-vector-simulator 0.5.0rc10

Multi-vector Simulation Tool assessing and optimizing Local Energy Systems (LES) for the E-LAND project

MVS - Multi-Vector Simulator of the E-Land toolbox

Rights: Reiner Lemoine Institut (Berlin)

The multi-vector simulator (MVS) allows the evaluation of local sector-coupled energy systems that include the energy carriers electricity, heat and/or gas. The MVS has three main features:

• Analysis of an energy system model, which can be defined from csv or json files, including its costs and performance parameters.
• Near-future investments into power generation and storage assets can be optimized aiming at least-cost supply of electricity and heat.
• Future energy supply scenarios that integrate emerging technologies helping to meet sustainability goals and decrease adverse climate effects can be evaluated, e.g. through high renewable energy shares or sector-coupling technologies.

The tool is being developed within the scope of the H2020 project E-LAND (Integrated multi-vector management system for Energy isLANDs, project homepage HERE). A graphical user interface for the MVS will be integrated.

Latest release Check the latest release. Please check the CHANGELOG.md for past updates and changes.

Disclaimer As the MVS is still under development, changes might still occur in the code as well as code structure. If you want to try the MVS, please make sure to check this project regularly.

For advanced programmers: You can also use the dev branch that includes the latest updates and changes. You find the changelog HERE.

Getting started

If you are interested to try out the code, please feel free to do so! In case that you are planning to use it for a specific or a larger-scale project, we would be very happy if you would get in contact with us, eg. via issue. Maybe you have ideas that can help the MVS move forward? Maybe you noticed a bug that we can resolve?

We are still working on including a readthedocs for the MVS. Some information on this tool and code is already available here (stable version, latest developments here).

Setup and installation

To set up the MVS, follow the steps below:

• If python3 is not pre-installed: Install miniconda (for python 3.7: https://docs.conda.io/en/latest/miniconda.html)

• Open Anaconda prompt (or other software as Pycharm) to create and activate a virtual environment

  conda create -n [your_env_name] python=3.6 activate [your env_name]

• Install the latest MVS release

  pip install multi-vector-simulator

• Download the cbc-solver into your system from https://ampl.com/dl/open/cbc/ and integrate it in your system, ie. unzip, place into chosen path, add path to your system variables (Windows: “System Properties” -->”Advanced”--> “Environment Variables”, requires admin-rights).

  You can also follow the steps from the oemof setup instructions

• Test if that the cbc solver is properly installed by typing

  oemof_installation_test

  You should at least get a confirmation that the cbc solver is working

  *****************************
  Solver installed with oemof:
  
  cbc: working
  glpk: not working
  gurobi: not working
  cplex: not working
  
  *****************************
  oemof successfully installed.
  *****************************
  
  

• Test if the MVS installation was successful by executing

  mvs_tool

This should create a folder MVS_outputs with the example simulation's results

Using the MVS

To run the MVS with custom inputs you have several options:

Use the command line

Edit the json input file (or csv files) and run

`mvs_tool -i path_input_folder -ext json -o path_output_folder`

With path_input_folder: path to folder with input data,

ext: json for using a json file and csv for using csv files

and path_output_folder: path of the folder where simulation results should be stored.

For more information about the possible command lines options

`mvs_tool -h`

Use the main() function

You can also execute the mvs within a script, for this you need to import

from multi_vector_simulator.cli import main

The possible arguments to this functions are:

• overwrite (bool): Determines whether to replace existing results in path_output_folder with the results of the current simulation (True) or not (False) (Command line "-f"). Default: False.
• input_type (str): Defines whether the input is taken from the mvs_config.json file ("json") or from csv files ('csv') located within <path_input_folder>/csv_elements/ (Command line "-ext"). Default: json.
• path_input_folder (str): The path to the directory where the input CSVs/JSON files are located. Default: inputs/ (Command line "-i").
• path_output_folder (str): The path to the directory where the results of the simulation such as the plots, time series, results JSON files are saved by MVS E-Lands (Command line "-o"). Default: MVS_outputs/.
• display_output (str): Sets the level of displayed logging messages. Options: "debug", "info", "warning", "error". Default: "info".
• lp_file_output (bool): Specifies whether linear equation system generated is saved as lp file. Default: False.
• pdf_report (bool): Specify whether pdf report of the simulation's results is generated or not (Command line "-pdf"). Default: False.
• save_png (bool): Specify whether png figures with the simulation's results are generated or not (Command line "-png"). Default: False.

Edit the csv files (or, for devs, the json file) and run the main() function. The following kwargs are possible:

Default settings

If you execute the mvs_tool command in a path where there is a folder named inputs (you can use the folder input_template for inspiration) this folder will be taken as default input folder and you can simply run

`mvs_tool`

A default output folder will be created, if you run the same simulation several time you would have to either overwrite the existing output file with

`mvs_tool -f`

Or provide another output folder's path

`mvs_tool -o <path_to_other_output_folder>`

Generate pdf report or an app in your browser to visualise the results of the simulation

To use the report feature you need to install extra dependencies first

`pip install multi-vector-simulator[report]`

Generate a report after running a simulation

Use the option -pdf in the command line mvs_tool to generate a pdf report in the simulation's output folder (by default in MVS_outputs/report/simulation_report.pdf):

`mvs_tool -pdf`

Generate only the figures of a simulation's results

Use the option -png in the command line mvs_tool to generate png figures of the results in the simulation's output folder (by default in MVS_outputs/):

`mvs_tool -png`

post-processing

To generate a report of the simulation's results, run the following command after a simulation generated an output folder:

`mvs_report -i path_simulation_output_folder -o path_pdf_report`

where path_simulation_output_folder should link to the folder of your simulation's output, or directly to a json file (default MVS_outputs/json_input_processed.json) and path_pdf_report is the path where the report should be saved as a pdf file.

The report should appear in your browser (at http://127.0.0.1:8050) as an interactive Plotly Dash app.

You can then print the report via your browser print functionality (ctrl+p), however the layout of the pdf report is only well optimized for chrome or chromimum browser.

It is also possible to automatically save the report as pdf by using the option -pdf

`mvs_report -i path_simulation_output_folder -pdf`

By default, it will save the report in a report folder within your simulation's output folder default (MVS_outputs/report/). See mvs_report.py -h for more information about possible options. The css and images used to make the report pretty should be located under report/assets.

Contributing

If you want to contribute to this project, please read CONTRIBUTING.md. For less experienced github users we propose a workflow HERE.