Freeing 1.5 million Dutch Homes From Gas Heating by Using AI

The Energy Act is now operative. This means that we need to switch from fossil fuels as a primary heating source to alternative renewables as soon as possible. To hit the goal of reducing our emissions by 40% before 2030 in The Netherlands alone, about 1.5 million homes need to be modified as they are currently heated by mainly natural gas-powered heaters. But how are we going to do that?

First, we need to look at how we got here. Natural gas has worked out very well for a long time. It has proven to be a reliable energy source with a high energy density. For years, all newly built houses in the Netherlands were connected to the Dutch gas grid. By getting the natural gas from Dutch soil (the Groningen gas field), residents could keep their homes cosy at a reasonable price.

On the surface, natural gas seems to be the perfect solution. However, there is a problem. Natural gas produces a lot of CO2, and that is precisely what we are trying to reduce. And, in the Netherlands, this is about more than just emissions. We also need to take into account the damage earthquakes have caused as a result of pumping gas from the ground in Groningen over the last several decades. These two reasons alone demonstrate the urgency of stepping away from natural gas as quickly as possible, not just in the Netherlands, but worldwide.

Image for post
Image for post
One of many affected houses in Groningen that needed reinforcement to withstand future earthquakes from surrounding gas wells (source: Elsevier Magazine)

However, that may not be as easy as it sounds. Remember, the use of natural gas is driven by cost and effectiveness. Current alternatives like electric heat pumps are expensive and are not always suitable to heat an entire house. This leads us to where we are now — with an urgent need to reduce the cost of heating alternatives like heat pumps while making them as comfortable as current gas-powered heaters. But how do we get there?

We can outline three major trends when it comes to spotting innovative opportunities related to heating & cooling houses:

  • Home control with the smart-home revolution — Houses are becoming smarter with more lights, appliances, and systems controlled by smart devices. For example, with devices like the Google Nest, an AI algorithm can notice that you are not home and turn off your heater. The intelligence behind it eventually learns when what days you are home and make sure the house is automatically heated when you are at home.
  • Data collection with smart- meters and heaters — Smart energy meters and heater devices these days are equipped with near real-time sensoring equipment that generates large data streams. Having this data makes more advanced data analytics and discoveries possible. This leads to even more significant improvements in efficiency and comfort.
  • The advent of deep learning in itself —Using deep learning models we can find hidden patterns we never saw before. Deep learning models tend to work better as the volume of data increases. These models can predict heater failures or detect efficiencies before they happen.
Image for post
Image for post
The performance (=value) of deep learning algorithms increases as the volume of data scales.

Feenstra, an organization that installs and maintains heaters and coolers in the Netherlands, is already at work in smoothing the energy transition. They are working to leverage these opportunities in home heating to help reduce our dependence on natural gas while keeping homes warm and comfortable.

Feenstra has collected years of data using its smart rental heater program. They used this data to monitor how well the heater is operating. This means that they may know there is a problem even before the homeowner knows. When the heater isn’t functioning correctly, technicians can remotely service the heater from a distance or send a service engineer right away. With about 700 service engineers on the road daily, people are never stuck in the cold.

Image for post
Image for post
Home heating & cooling service by Feenstra (source: Feenstra)

With the advent of machine learning and data, Feenstra has invested extensively in its data landscape. Currently, Feenstra has thousands of smart heaters rented out that produce millions of data points every week. This all needs to be stored properly somewhere. On top of that, Feenstra saw the importance of investing in data analytics and is now obtaining valuable information from the data.

With this valuable information, Feenstra can better plan the service intervals of engineers, optimize heaters at a distance to lower gas bills. This data can also be used to help customers be more comfortable and even detect heater failures before they happen.

This is all good news. But how can Feenstra help us take the next steps toward reducing carbon emissions? There are two main challenges to stepping away from gas. First, we must reduce the cost of alternative heating installations. Most people are not ready to observe a substantial cost increase in the name of reducing emissions. Second, it is essential to know with utmost precision if a heater is suitable for a given home. If new systems do not adequately heat homes, people will become more resistant to changes.

Lowering costs is simply a matter of increasing volume. But volume will only increase when pricing makes it attractive for people to switch. To avoid this catch 22, we have to look to other advantages. This is where AI comes in. Combining real-time heating data, house characteristics, human preferences, and weather data, it is possible to predict which alternative heater is powerful enough to heat a given home.

Image for post
Image for post
Sustainable heating alternatives are already available in the market i.e. like (electrical) heat pumps, but often come with a price tag (source: Elga)

With a high level of confidence in this prediction, people will understand that they can decarbonize their home without any loss of comfort. In the beginning, this will certainly not apply to all houses. However, if we start installing new systems in homes that are suitable, especially new and well-insulated homes, enough volume may be created to lower the costs of the next-generation heat pumps. With more demand, cheaper and more powerful pumps will follow.

At Hemisphere, we are helping Feenstra in this transition and building the required AI algorithms. We focus on helping our clients streamline the work of innovation through collaboration. Through this work, organizations can find new business and new revenue streams to move ahead of the competition. Want to know more and see what we can do for you? Or do you want to work for us? Contact info@hemisphere.ai

Written by

AI Concept developer @ Hemisphere

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store