AI – enabled smart storage services for future renewable district energy networks
Project Status: set-up
Start Date: January 2021
End Date: December 2023
Budget (total): 2162.5 K€
Name: Qian Wang
Company: Royal Institute of Technology, KTH (Kungliga Tekniska Högskolan)
AIT Austrian Institute of Technology, Austria
ARCbcn Consulting Engineers S.L., Spain
Geothermal Energy S.L., Spain
Maston AB, Sweden
Royal Institute of Technology, KTH (Kungliga Tekniska Högskolan), Sweden
Thyni &Veketoft Innovation AB (Tvinn), Sweden
The current electrification trends in society, fuelled by sustainability and renewable energy initiatives, will bring
with it a dramatic increase in peak power demand. The electric vehicle (EV) market alone has been estimated to
increase peak demand by as much as 20 % by 2050 in Scandinavian countries and Germany, without further
Already, several urban areas around the world are facing acute grid capacity shortages. This lack of capacity will
hamper adoption of EVs, and delay the renewable and sustainable transformation of our energy systems.
Moreover, increased shares of variable renewables (such as solar and wind) in the energy system will introduce
greater fluctuations in the energy supply. Flexibility, in terms of how we use energy, is high on the European
agenda, and will be critical for solving these issues in time.
Integrated energy storage solutions, with their ability to store and inject energy as needed, will have an
important role in enabling demand flexibility. However, to accomplish this goal, storage solutions must be properly
integrated with the energy system, and adapt operation according to the energy demand of the end-users.
Demand response, i.e., scheduling energy usage to better align with supply, will be an important part in future
control systems for storages and energy networks.
Digitalization will be important for enabling demand flexibility, and coordinate the different components in the
energy system. However, the current lack of standardization in the energy sector makes it difficult to adopt ICT
and IoT (internet of things) methods. New ICT solutions are needed, but they must be developed in collaboration
with industry representatives, to ensure widespread industry acceptance and adoption.
This project, AI-inStorage, is expected to result in new digital tools for designing and sizing energy storage
solutions, and new control algorithms for efficient operation. Machine learning will be used in these tools to adapt
according to the dynamic behaviours of the end-users. In the case of the control algorithms, this will be done in
real-time. The results will be deployed in real life environments and local energy networks, in collaboration with onboard
industry partners, and will be used to extend and improve the product and service portfolios of our