Multiscale Solar Water Heating

Multiscale Solar Water Heating

Solar water heating system modeling and simulation for individual and community scale projects

Repository Content

Folder Content
mswh Python module to calculate solar irradiation on a tilted surface (mswh/system/

Python module with simplified component models (mswh/system/ for Converter (solar collectors, electric resistance heater, gas burner, photovoltaic panels, heat pump), Storage (solar thermal tank, heat pump thermal tank, conventional gas tank water heater), and Distribution (distribution and solar pump, piping losses) components.

Python module with preconfigured system simulation models (mswh/system/ for: base case gas tank water heaters, solar thermal water heaters (solar collector feeding a storage tank, with a tankless gas water heater backup in a new installation cases and a base case gas tank water heater in a retrofit case) and solar electric water heaters (heat pump storage tank with an electric resistance backup).

Database with component performance parameters, California specific weather data and domestic hot water end-use load profiles (mswh/comm/swh_system_input.db).

Modules to communicate with the database (mswh/comm/, unit conversion and plotting modules in mswh/tools.
scripts Jupyter notebooks with preconfigured models and any side analysis if applicable. Navigate to scripts in Binder to try them out quickly.
web Django web framework to configure project, parametrize components and run simulation from a web browser.
docs API documentation, including a short methodology documentation can be found here. To build HTML or LaTeX use make html or make latex. A pdf version of the Code Documentation can be viewed and downloaded here.

Statement of Need

We envision four main groups of users for the MSWH software:

  • Researchers and policy developers.
  • Solar water heating planners, designers and contractors.
  • Homeowners.
  • Educators.

The policy developers and researchers could utilize the existing MSWH software by embedding it into some larger analysis framework they construct such that it provides answers to their specific research questions.

The professional planners, designers, and contractors of solar thermal water heating systems might find it useful to have access to a freely available simulation tool such as the MSWH software, that they can use to evaluate alternative system designs.

Homeowners considering transitioning to a solar water heating system may be interested in doing the math before seeking further professional help, or just for their own education and curiosity about both solar water heating systems and system simulation in general.

Educators may wish and find it useful to utilize the MSWH simulation tool in the classroom when teaching the basics of energy simulation.


The fastest way to explore the preset simulations is to use the MSWH System Tool notebook. In the notebook the user provides a climate zone for a project, an occupancy for each household and whether any of the occupants stay at home during the day. The notebook can then load a set of example California specific hourly domestic hot water end-use load profiles from a database, size and locate the systems. The user can now simulate the hourly system performance over a period of one representative year, visualize and explore the simulation results using time-series plots for temperature profiles, heat and power rates, or look at annual summaries. Similarly the user can model individual household solar water heating projects and base case conventional gas tank water heater systems, such that the results can be compared between the individual, community and base case systems. All simulation and sizing parameters are exposed in the notebook and the user can easily change them if needed.

If you opt to use the web framework the shortest path to explore the simulaton results after setting up a local server is to:

  • Click on Configurations on the landing page.
  • Click on Simulate for any of the example preconfigured systems (Solar Thermal New or Solar Electric). This leads the user to a visualization page with hourly timeseries results for a representative year.
  • Play with sizes and performance parameters of preconfigured components.

To configure new system types in the web framework (such as Solar Thermal Retrofit) one would need to map it through the backend analogously to the currently preconfigured systems.

An example demonstrating usage of the simulation models for an additional climate outside of California, that is Banja Luka in Bosnia & Herzegovina, is provided in this notebook.

Setup and Installation

  1. Make sure that pip is installed.

  2. Unless you already have conda installed, please install the lightweight option Miniconda or Anaconda software.

Simple Installation Using Conda

  1. If you are familiar with conda and experienced with virtual environments you can perform the package installation using the following set of commands:

     conda create -n mswh -c conda-forge -c plotly python=3.8 pip git-lfs jupyterlab plotly-orca
     conda activate mswh
     git lfs install
     git clone
     cd MSWH
     pip install -e .

    To ensure functionality of the example notebooks install the following:

     python -m ipykernel install --user --name mswh
     jupyter labextension install jupyterlab-plotly

The examples are best explored using JupyterLab. Please check out the JupyterLab documentation for further help as needed.

Detailed Installation Steps

If for any reason a user encounters difficulties with the simple installation instructions, the user is encouraged to consult a more detailed installation guide that is posted with the code documentation.

Django Web Framework Deployment

1. Local

If the installation succeeded, to run the Django application navigate to the web folder (there should be a file) and start the development server on your local machine with:

    python runserver

Now you can open your browser and type in localhost:8000 (or if you are on a Windows machine) to start the web interface.

Note that to build python extensions one needs to have python3.x-dev installed.

Make sure that DEBUG = True in, this will ensure that the development server is able to serve local static files.

2. Public

Override settings locally

To deploy publicly, rename the file to and update the constants.

  • SECRET_KEY = '<random_string>'

    The random string should be 50 characters long and can created (on Linux) by using the following command as super user:

      </dev/urandom tr -dc '1234567890!#$?*#-.,+qwertyuiopQWERTYUIOPasdfghjklASDFGHJKLzxcvbnmZXCVBNM' | head -c50; echo ""

    An example for a good secret key is this: SECRET_KEY = 'O&2aYmv%)0B5#U-'9qsLTpfItC9N*V?%3L#fOHxDO,zyUm*S,U'

  • DEBUG = False

    Keep the Debug constant set to True in to get more debugging info for local deployement (using the Django development server). For the public deployement, you should set it to False (in

Serving static files

For detailed documentation on how to serve static files, see the official Django documentation:

As the Django devlopment server is not meant for production and only serves static files if Debug is set to True, the static files used in the Django project need to be served another way.

At this point, two important aspects regarding how to deploy static files in production will be named:

  1. Running this command, will create a folder static that will contain a copy of all static files from different directories across the Django project.

    python collectstatic

    ⚠️ Run this command every time you update one of the static files in their respective location in the Django project folder.

  2. Configure nginx to serve static files from the generated static folder by adding a location /static block to the server block of the nginx config file for the domain you serve the Django app with. This is an example nginx server block:

    server {
      listen 80;
      server_name <domain>;
      access_log  /var/log/nginx/access.log;
      error_log  /var/log/nginx/error.log;
      location / {
        # For testing, using the django development server:
        # proxy_pass;
        # For production, using gunicorn:
        proxy_pass http://unix:/run/swhweb.sock;
      # Run 'python collectstatic' command in Django root project folder, so this folder will be created
      location /static {
        root <path_to_repo>/MSWH/web;
        try_files $uri $uri/ =404;

    Replace <path_to_repo> with the actual path to the MSWH repository and <domain> with your domain.


All are invited to contribute to the MSWH software through following the Guidelines for Contributors.

Automated tests

To run tests, from the MSWH folder use the following command modified according to the test module and method you intend to run:

python -m unittest mswh.{my_module}.tests.{test_my_module}.{MyModuleTests}.{test_my_method}


The code was used for the following publications:

  • Coughlin, Katie, Milica Grahovac, Mohan Ganeshalingam, Robert Hosbach, and Vagelis Vossos. 2020. Costs and Benefits of Community versus Individual End-use Infrastructure for Solar Water Heating. California Energy Commission. CEC-XXX-2020-XXX. (in press)

  • Grahovac, Milica, Katie Coughlin, Mohan Ganeshalingam, Robert Hosbach and Vagelis Vossos. 2020. Costs and Benefits of Community Scale Solar Water Heating. 2020 ACEEE Study on Energy Efficiency in Buildings. Pacific Grove, California. Link to the paper with a video presentation

  • Milica Grahovac, Katie Coughlin, Robert Hosbach, Hannes Gerhart, (2020). Multiscale Solar Water Heating. Journal of Open Source Software, 5(56), 2695, DOI

  • Gerhart, H. (2019). Implementation of a Flexible Web Framework for Simulating Python System Models (p. 82). Technical University of Munich; Technical University of Munich. Research performed at LBNL. Download link

  • Web deployed version of the Django app is under construction on this publicly available private website.


The software may be distributed under the copyright and a BSD license provided in

Milica Grahovac, Robert Hosbach, Katie Coughlin, Mohan Ganeshalingam and Hannes Gerhart created the contents of this repo in the scope of the CEC "Costs and Benefits of Community vs. Individual End-Use Infrastructure for Solar Water Heating" project.

To cite use format provided at the DOE CODE MSWH record.


This work was supported by the California Energy Commission, Public Interest Energy Research Program, under Contract No. PIR-16-022.

We thank the reviewers and the editor of The Journal of Open Source Software (JOSS), Bryn Pickering, Nithiya Streethran, and Stefan Pfenninger for their contributions in improving the code, the examples and the code documentation for the code release 2.0.0.

Related Searches

Get A Weekly Email With Trending Projects For These Categories
No Spam. Unsubscribe easily at any time.
Jupyter Notebook