AI-enabled basestations create virtual power plant in Finland
Elisa in Finland is using cellular basestation backup batteries as an AI-enabled virtual power station.
Using the Radio Access Network (RAN) to run a Virtual Power Plant could save telecoms operators around 50% of their current electricity costs by optimising their energy purchases as well balancing the grid with renewable energy at times of need says Elisa.
The company received pre-qualification for what it believes to be the world’s first distributed energy storage (DES) system for the Automatic Frequency Restoration Reserve (aFRR) market earlier this year and is replacing the lead acid batteries with lithium ion cells across a third of its basestation in Finland as well as in Estonia.
DES is a key capability to even out the supply of energy from renewables, particularly wind turbines. Elisa forecasts that there could be an estimated 15GWh of storage available from Europe’s telecoms networks.
To convert a telecoms network and battery storage to form the role of a VPP, Elisa’s AI-powered DES enables load shifting to purchase electricity from the grid during low-cost periods and store it for consumption when prices are higher. This power can be distributed throughout the network, or sold to the electricity grid when balancing is required.
Elisa predicts that through this functionality, telecoms operators will be able to save around 50% of their current electricity costs through a combination of cost savings and new revenue opportunities.
Elisa ran an initial trial of its DES solution in Finland across 200 base stations in 2022 as well as its network in Estonia. By 2025, the system will be rolled out to 2000 Elisa base stations in Finland, with that work well underway in 2023.
“Last summer we conducted testing with Fingrid (Finland’s electricity transmission systems operator) across 200 of our base stations. It was successful and as a result, in the summer of 2022, we received the technical pre-qualification acceptance from Fingrid for its Distributed Energy Storage solution to provide balancing services in the ‘aFRR’ balancing market,” said Ismo Räsänen, senior development manager at Elia.
“We are now rolling the system out across our network. Of course, that means there is some investment in the new battery capacity, but with continued volatility in the energy market we think we will get the return on that investment within three years.”
“The effective operation of highly distributed VPP solutions requires competencies in mathematical optimization, control systems and communication systems. Elisa’s heritage in telecommunications in combination with our expertise in data management, AI and machine learning solutions have been instrumental in allowing us to develop an effective solution. All the work happens invisibly so it’s very easy to use in the multivendor environment and it makes our network more resilient and our power supply more flexible.
The DES is composed of three layers of control intelligence powered by AI software, tapping into the electricity and power equipment data to provide actionable intelligence for grid optimization.
The top-level planner uses supply and demand data of different electricity balance markets to perform thousands of simulations that allow it to determine how much capacity to allocate at any given time to load-shifting or grid balancing.
The data-intensive middle layer supervises the flow of electricity and conditions of the base station power equipment to ensure the best use of the assets, including battery longevity over time.
The lowest level controller selects which specific units should be adjusted to respond to changes in load-shifting plans or to TSO grid balancing requests.
The combination of the three layers allows to DES to control the charging and discharging of batteries to optimum effect for the purposes of powering the network in the most cost-effective way, as well as supporting the grid with balancing services.
“After the successful roll out of the solution in Elisa’s networks in both Finland and Estonia, our current focus is to adapt the AI/ML control to allow it to be installed with other operators’ critical infrastructure,” said Räsänen.
“Elisa Polystar’s Intelligent Energy Saver has already harnessed the power of AI to deliver significant cuts to energy consumption by targeting savings at the cell level in the RAN. The DES takes control of energy consumption in a different direction with a solution that enables further deployment of cost effective renewable energy by optimizing storage usage, essentially transforming telecom infrastructure into a flexibility asset for electricity grid balancing,” said Jukka-Pekka Salmenkaita, VP of AI and Special Projects at Elisa.
Further technical work is underway to ensure compatibility in the widest possible multi-vendor environment for the power and battery hardware to create a wider virtual power plant.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
