Beta-label indicates that this is still an early version of the service, that contains many of the major features, but is not yet complete.
Costs include VAT, but you also pay for energy transfer and the margin defined in your electricity contract. Virtanen.ai does not know these extra charges, so you need to add them in yourself. Naturally, if you have an electricity contract with a fixed price, the costs do not apply.
Virtanen.ai finds the hours of the lowest consumption during the last year, when you are probably not at home and only constantly running devices such as the refrigerator are turned on. Base consumption is then estimated using these low consumption hours with some adjustments like ignoring zero consumption hours from potential power outages.
Virtanen.ai finds the correlation between the household electricity consumption and weather conditions like outside temperature. This requires electricity consumption data for a few years time period so that Virtanen.ai can observe how much electricity is consumed in different weather conditions.
The estimate might become less accurate, if you improve house insulation or lower your room temperature during the analysed time period as Virtanen.ai has no knowledge of these changes immediately. In the long term Virtanen.ai will adapt to the changes made.
The estimate for heat pump consumption is included in the heating consumption estimate. All heat pumps (air, air-water, geothermal) use electricity, although the total electricity consumption is less than with direct electrical heating. Finding the correlation between the household electricity consumption and weather conditions is still the same.
Virtanen.ai first estimates base and heating consumption. The remaining consumption is expected to result from using various electrical devices, and the device size category is estimated from the hourly total electricity consumption.
There are some limitations to the estimation due to the data that is available: since the data is hourly totals, it is difficult to detect if, for instance, you are running multiple devices at once, or for only a few minutes over two hours. However, we are still working on the methods to provide better estimates in the future. Stay tuned for future Virtanen.ai improvements!
After Virtanen.ai has estimated the electricity consumption for each device group at different hours, it finds the lowest market price 6 hours before or after the realised consumption. Saving potential is the cost difference between the market price at the realised consumption time and what the same amount of electricity cost during the lowest market price hour.
Virtanen.ai enriches the electricity consumption data (Fingrid Datahub) by historical electricity market prices (Nord Pool). Weather observations used by the analysis are from Finnish Meteorological Institute. Emission factors (CO₂ / kWh) for electricity consumption in Finland are from Fingrid.
Carbon Dioxide (CO₂) is a primary greenhouse gas emitted by human activities such as burning fossil fuels for energy, deforestation, and industrial processes.
Carbon Dioxide Equivalent (CO₂e) is a standardised unit used for carbon accounting. It combines the warming effects of various greenhouse gases, converting them into an equivalent amount of CO₂ based on their relative potency.
Virtanen.ai utilizes emission estimates sourced from Fingrid, which adopt CO₂ instead of the more common CO₂e. Virtanen.ai aligns with this approach for the time being.