The NOWATT project aims to advise residents on the optimal time of day to run their washing machine. Although a washing machine is not really a heavy energy user (compared to the usual suspects heat pump, electric car and airco), the washing machine is chosen as a representative appliance, found in almost every household. The commonality makes it the ideal equipment for field-testing. The research question to answer is: How can we leverage existing technological developments to (automatically) save energy while ensuring that residents still have influence and remain in control of their appliances?
For the curious who understand Dutch, NOWATT is the abbreviation for:
Nieuw Overkoepelend Wijkgericht Aanvalsplan voor Toekomstgerichte energieTransitie.
For the non-Dutch the CoPilot translation: New Overarching Neighborhood-Oriented Action Plan for Future-Oriented Energy Transition.
The objectives of the NOWATT application are:
- To use as much self-generated energy as possible.
- To prevent grid congestion on the local microgrid.
Grid congestion is defined here as exceeding a local grid voltage of 253 Volt, at which point the solar panels are automatically shut down by the solar equipment. Once the voltage has dropped sufficiently (and with a randomized delay), the panels are switched on again.
The application is used in a small field test, involving four households. These households are considered to be connected to a microgrid: a small-scale electricity network where energy is shared locally among users. All homes in the test have solar panels, though specific details about the installation like peak output or orientation are not collected. The composition of the household (e.g., number of residents) is known to the researchers but is hidden from the application.
- Each household receives a Siemens washing machine that can be remotely monitored and controlled via HomeConnect. The backend of the NOWATT app can interface with HomeConnect to read and control the washing machine.
- Each home is equipped with a Hanzebox electricity measurement logger connected to the P1 port of the smart electricity meter. The Hanzebox measures the net electricity flow and voltage. It does not measure the solar panel output directly, only the net difference between consumption and production.
- The application has a subscription on the data from Solar Forecast. This datastream can give a forecast of the solar production in the next 24 hours.
- The IoT platform OpenRemote is used for collecting the datafeeds of Solar Forecast, the Hanzeboxes and the washing machines connected through Homeconnect.
The architecture is explained using diagrams developed in Enterprise Architect.
The source code consists of the following folders in this monorepo:
- backend - The backend code, written in Python, using the FastAPI framework.
- docs - Documentation of the system architecture.
- front - The frontend code, written in Javascript, using the React framework.
- homeConnectTokenTool - A tool to collect the homeconnect tokens for this project
- mqtt_csv_ingestion_automating_token_refresh - A tool for ingesting the HomeConnect events and converting them to MQTT messages that are consumed and stored by OpenRemote for further use
- openremote_docker_config Configuration to run OpenRemote in a docker environment.
- washer_forecast_scheduler First attempt to forecast the optimal washing moment.