The problem with renewable energy is its intermittency. In Ireland for example, it is not uncommon to see big variations in wind production. Those patterns are clearly visible on electricitymaps.com, as can be seen on this example taken from the 6th of February 2017:

6th of February 2017: Irish consumption mix goes from 80% fossil to 30% in a coupled of hours.

In this example, consuming electricity at midnight emitted 3 times more CO2 than a coupled of hours later, when the wind started to blow. Wouldn’t it make sense to build a society where our electric devices consume electricity at the right time?

We strongly believe so. As a proof of concept, we collaborated as part of the CITIES project in building a device that would react to changes in carbon intensity. The device receives a forecast through the Internet, and uses that forecast to plan ahead and chose the best time to turn on and off. In this case, the device controls a heat pump warming an indoor swimming pool, but it could control other things (air condition, electric vehicle charging etc..).

Predicted carbon intensity (green) and periods at which the heat pump is turned on (orange). The installation is located in Denmark.

Note that the device has to keep the controlled temperature within the defined comfort range. In order to do that, it will try to pre-heat in expectation of upcoming high-carbon periods.

This is what we observe during the evening of the 3rd of October, where the water is heated up (even though it was already quite warm) in order to allow the system to turn off until the low carbon period starts (around noon of the day after):

By consuming electricity at optimal times, this device has been able to reduce its emissions by 10% compared to a device that would have consumed at random times.

This system applies for every device that:

• has a consumption high enough to matter
• is located in countries with a high penetration of wind and solar power (see electricitymaps.com)
• has enough flexibility to move its consumption in time
• is connected to the internet

Electrical heating or cooling devices, heat pumps, electric vehicles and dishwashers or washing machines are therefore excellent targets.

Are you building one of those devices? Equip it with our forecast of carbon intensity (or amount of renewable energy), and enable it to consume at the right time.

Are you a developer? Check out our free co2signal API.

This work is part of a joint effort in the CITIES project together with DTU Compute, Novasol, FlexGrid and ENFOR. More information here.

‍