Whenever we start helping our customers, we build a digital twin. We collect all data of their energy systems digitally, and mimic the original system with data to be able to run simulations.
Similar to DHW, we optimize floor heating in various ways. Let’s highlight one that varies from the previous example.
Besides using a cost function for the optimal balance in ‘chance of discomfort versus costs’, and looking at dynamic energy prices, we also look at external factors such as weather predictions. If the weather predictions show a rise in temperature (36h ahead), do we still need to heat up? The only way to find out is to run simulations with our digital twin. With all data we collect, we can exactly determine what effect an X-degree rise in outside temperate has on the inside temperature, and manage our floor heating accordingly.
It's always a matter of looking at value and effort. If the effort required doesn’t outweigh the value it results in, don’t change anything. And the only way to find out is building a digital twin and running mathematical models to understand and decide.
Everyone is using electric devices at home, think of kitchen appliances, solar panels, or even heat pumps. Most of these devices are working optimally on their own, but looking at the entire system, a lot of efficiency can be gained. That is where our solution ‘EnTune’ comes in.
EnTune is an energy management system that optimizes energy of entire households. It looks at all devices that produce or consume energy, collects their data, gathers external data (energy prices, weather forecasts), and defines an optimal energy strategy based on these factors and user preferences.
And our secret is simple: mathematics.
Using mathematics to save energy
Let’s have a look at two examples: the optimizations for domestic hot water, and the optimizations for floor heating.
Domestic hot water
Domestic hot water (DHW) is the heated water we use while showering or doing dishes. We optimize DHW costs, using mathematics, in three ways: cost function, dynamic energy prices, and usage patterns.
Normally, DHW is kept at 60 degrees. However, do we really need it to be 60 degrees? Since everyone is unique, we all have our own temperature preferences. One might experience discomfort if the DHW is only 50 degrees, while others might never use such a high temperature. That’s why we optimized it using a cost function, which measures the performance of a machine learning model for given data. It continuously looks at the chance of experiencing discomfort versus costs. With a higher setpoint temperature, the chance of discomfort decreases, but costs increase. And since EnTune users are asked to give feedback in the EnTune app whenever they experience discomfort, we can find the optimal balance in the cost function.
The cost function uses dynamic energy prices, and we simply heat the water when energy prices drop, and create a buffer for when energy is expensive. Thereby, we also look at consumption patterns. Over time, it’s possible to predict when a user needs hot water, and when not. When a family is always out during the day, it’s of course wasted energy to keep the water at setpoint temperature. And with that, our cost function can save as much as 50% of the costs.
Back to overview
