Smart Meter Data Analytics is a solution project aiming to investigate data mining of smart meter electricity consumption data in order to improve forecasting and money-saving opportunities for customers.
The aim of the solution on Smart meter data analytics, is to gather socioeconomic, demographic and weather data and insight into the profiles and classifications of consumption to develop them even further by investigating different approaches to analysis of smart meter data.
The hope is to develop a method that can characterize an average consumption for each profile and indicate if a specific consumption realization is above or below the average profile. This helps identify consumers who have a high or low consumption and quantify consumption in a monetary setting directly to the consumer or for energy optimization consultancy – but also to help reduce peak demands in energy consumption.
Impact
Enriched insights into customers consumption profiles will give an opportunity to reduce consumption – and thereby costs – and increase the potential amount of renewable energies which is supplied to a given consumer.
Solution insights
So far, the demonstration project has identified different approaches for analyzing smart meters readings. These counts i.a. 1) energy signatures/labels of building, 2) advanced knowledge about energy savings, 3) the opportunity to detect and deal with uncertainties and 4) splitting total readings into space heating and domestic hot tap water usage.
Further information
The project of CITIES Research Center is expected to continue on improving the solution throughout 2017.
By 2020 every electricity consumers in Denmark will be equipped with digital smart meters that automatically reports energy consumption at least every 15 minutes, which expands the observations tremendously and makes it possible to profile consumption on a much more detailed level. Current classifications have more than 130 different profiles and by using these technological classification techniques the number of profiles can be reduced and specified.
The solution is connected to the projects on analysis of heating data and energy demand modelling in buildings.
