“What we’re not doing is installing a whole load of low-carbon technologies,” says Miller. “We’re looking to find where there are communities who have an interest in this already, maybe already have some of this up and running, and see how we can facilitate that, how we can integrate it, get it to work together”.
The Community Distribution System Operator (DSO) project will see Northern Powergrid delegate some of the emerging DSO functions to four local communities around the country, allowing them to take responsibility for managing the operation of their local electricity network, albeit within limits set by the company.
Miller says this local energy system balancing will enable the communities to get the most out of both their existing assets and the network and create space for the connection of more low-carbon technologies such as solar panels, heat pumps and electric vehicle chargers.
A key challenge is resolving issues around the interoperability of their assets – “getting different people’s equipment to talk to each other.” Recounting his own personal experiences, Miller says this can sometimes be problematic even within a single home.
Glendinning makes a comparison with the Smart Energy Network Demonstrator on the campus of Keele University, which was “done by one mastermind brain” using the same equipment and communication protocols, and all behind the meter: “That’s unachievable for a community.”
He says the Community DSO project will seek to create a similar system but where “the single point of contact, the brain if you like, could just be a housing association or a parish council”.
“It’s really about trialling the technical solutions, which we think are available, but need to be tested to make sure they work in a plug-and-play situation,” he adds.
The project will develop and test the idea of organising local energy systems into a hierarchical “nested cell” structure mirroring the voltage levels of the power grid.
At the lowest level, a cell could comprise a single low-voltage feeder, perhaps serving a hundred customers or fewer. Their assets would be integrated together to form a smart local energy system, with their actions being optimised within the constraints imposed by the distribution network operator (DNO) for that particular feeder.
At the next level up, a number of feeders below a transformer could be combined together to form another cell. The actions of the lower levels cells would likewise be optimised within the constraints set by the DNO for that transformer. Transactions would only take place within cells. This structure could be replicated for each voltage level of the power grid.
Miller said under this model assets will only need to be interoperable with other assets within their immediate cell. Different software platforms could be used for different cells, meaning they could be selected to meet the specific needs of that cell.
“National Grid control has a really simple system,” he explains. They’ve got maybe 1,000, 2,000 ends, depending how you count them.
“We have 3.9 million ends on our system and if you go across the country, that’s 30 million ends on distribution systems. Imagining a platform that can look after 30 million ends, or even 3.9 million ends in real time, doesn’t bear thinking about.
“If I have one mega platform to look after everybody that is the IT project to end all IT projects.”
Glendinning says this approach is already used in industrial settings, where there are controllers for each individual process that then send information back to a central brain for the whole facility.
Miller says this approach is easily scalable: “This you can roll out on one street at a time because I get benefits back when its one street’s worth. I get benefits when I link those streets together. I get benefits when I link towns together.
“You can build that up piece-by-piece, one house at a time, one street at a time, one town at a time.”
Northern Powergrid’s submission to Ofgem says this cellular approach will empower communities to “reflect their needs and ambitions, with less centralised DNO decision making, while still preventing unintended consequences for the wider system.”
The four-year project will take place in three phases. In the first, Northern Powergrid will design and develop the architecture and solutions for the delivery of cellular community DSO functions, including analysis and control software and hardware requirements for network monitoring and control.
“Novel predictive and decision-making algorithms that incorporate machine-learning and data science are likely to be very important for community DSO operation, as there is only partial visibility and controllability of low-voltage networks,” the document explains. “This work package will include simulation and trialling of the developed approaches within a virtual modelled environment.”
In the second phase, the DNO will select four communities from across its region, some urban and some rural, to take part in the project. Although they haven’t “signed anybody up on the dotted line,” Miller says they have already identified some potential communities.
“This won’t be a hard sell,” adds Glendinning. “There’s plenty of communities out there that want to do this type of thing and find some networks a bit of an issue to deal with.”
Northern Powergrid will conduct field trials over a period up to three years using an iterative approach of trial, review and modify. This will include multi-cell management of a section of network using adjacent cells and the simulated management of cells across a broader network area.
In the third stage, Northern Powergrid will analyse the findings and identify the necessary regulatory changes and commercial arrangements to roll out the model as business-as-usual.
“There are some technical challenges around data and control of the equipment but the biggest challenge is a regulatory and commercial one,” says Glendinning.
“I think the commercial and regulatory issues that will come out of the back end of this are probably where the biggest prize is”.