Increasing energy efficiency: Virtual assistant helps private and commercial consumers

Energiesparen schon im Geschäft: So hilft der virtuelle Assistent "EcoBot" Verbraucherinnen und Verbrauchern beim Kauf von Elektrogeräten

Saving energy at the push of a button: An EU-funded project is developing Eco-Bot, a personalised virtual energy assistant that provides real-time information about the energy-saving potential of electronic appliances.


Physical and technological factors influence energy usage and CO2 emissions. However, individual energy consumption contributes substantially to global CO2 emissions. Exactly how individual consumption could be reduced is not always evident.

This is where the “Eco-Bot” project comes in. It is developing specialised chatbots that interact directly with users and facility managers to inform them about the energy-saving potential of various appliances and systems with the help of artificial intelligence. The Eco-Bot draws on existing smart meter-measuring techniques and uses complex algorithms to analyse the energy usage and energy-saving potential of individual devices. This virtual energy assistant can advise both private and commercial users in households or buildings on how to save energy by analysing additional information about their energy consumption patterns.

Lena Domröse, Eco-Bot Project Manager at adelphi, explains that: „This virtual assistant can recognise energy consumption patterns in a household and thus deduces opportunities for saving energy that may not be directly evident. Thus, the Eco-Bot can suggest small behavioural changes to users, which will have a significant impact on the environment and the energy costs borne by consumers!“

This solution combines modern technologies, including natural language processing, advanced signal processing (energy disaggregation) and multifactorial behavioural modelling. Modern information and communication technologies, including knowledge engineering, machine learning and expert systems, are also used. This allows multi-factorial models for energy saving to be transformed into personalised and targeted recommendations for saving energy and increasing energy efficiency.

The project is tested in three different uses in order to validate the Eco-Bot system under realistic and diverse conditions. Testing was thus conducted at an energy provider (Estabanell), a provider of a building management system (Dexma), and an energy-saving platform (SEnerCon).

The Eco-Bot Project is jointly implemented by universities, technology developers and energy providers, as well as hightech-SMEs (small and medium enterprises) in Europe and Turkey. Funded by the EU Research and Innovation Program Horizon 2020, the project was officially launched in October 2017. adelphi, as part of the research consortium, has been tasked with conducting a market analysis and creating an exploitation strategy to allow for commercial use of the Bots after the project concludes.