A £ 2 million ($ 2.7 million) machine learning tool creates a matrix-like simulation of the power grid in south east England. It has the potential to unlock an energy capacity of 1,371 new fast chargers for electric cars by 2028.
UK Power Networks’ Envision project simulates how electricity flows through its networks in London, the south and south east of England. This data-driven approach, which provides better “visibility” of everything that happens over hundreds of thousands of kilometers of network, enables the company to operate the network more efficiently and securely, enabling low-carbon technologies such as heat pumps and EV chargers to connect.
‘)} // ->’)} // ->
Experts predict that Envision could unlock nearly 70 MW of electricity capacity by 2028, freeing up more space for EV chargers or low-carbon heat pumps. This means engineers don’t have to physically upgrade the network to free up capacity, resulting in significant cost and time savings: a total of up to 4 million pounds ($ 5.4 million) over the next five years.
Connecting more electric car chargers and electric heat pumps is critical to tackling the climate crisis and meeting the UK’s net zero CO2 emissions target by 2050. There are currently more than 150,000 electric cars and 20,000 heat pumps in the areas served by UK Power Networks, but his analysis predicts over 2.6 million electric vehicles and 712,000 heat pumps by 2030. These require “space” in the power grid, which is why the rapid and efficient development of additional capacities is an essential part of the network operator’s strategy.
Envision is developing new predictive models that combine data from UK Power Networks with external and real-time data from monitoring equipment connected to substations. The machine learning algorithm will create a simulation of the electrical “load” in certain areas and extend it to the entire network. Engineers compare the simulation to real physical monitors; Feed the software more and better data over time so that the algorithm becomes more precise
“Our customers rightly expect us to do everything in our power to make the switch to electric cars and low carbon heating as cheap as possible,” said Ian Cameron, Head of Customer Service and Innovation at UK Power Networks. “With Envision we think outside the box and rethink traditional ways of working to make this a reality.”
“The goal of the Envision model is to generate a ‘virtual sensor network’ that uses advanced data capabilities and machine learning to simulate the behavior of the network on a scale and accurately estimate changing network load profiles,” says Simone Torino, Head of Product and business development at CKDelta working on the project. “In a world where the use of new decentralized energy resources and the increasing electrification of transport are affecting electricity demand and distribution network constraints like never before, this type of modeling and predictive analysis capabilities is critical to the utility sector and has potential. to reshape our approach to supply and demand in other sectors such as transport. “