Let’s imagine we are now living with a smarter, cleaner grid.

In this network utopia, much progress has been made electrifying transport and heating, with ‘smart systems’ allowing much of the national fleet of electric vehicles (EVs) and heat pumps to adjust their demand patterns according to the capacity available on the network.

Fast-forward to 6pm on a cold winter’s day, and the network has become fully loaded. In response, the distribution system operator’s (DSO) plan is to suppress EV charging and use up all available thermal storage to limit the demand from heat pumps.

Thanks to the digitalised energy system, the network can communicate with the home energy management systems (HEMS) installed in thousands of homes to make this happen without impacting consumers.

But what would happen if these communication channels failed, perhaps caused by a temporary internet outage? Or what if the DSO systems were compromised accidentally or deliberately?

If this scenario is not planned for thoroughly, there is a risk that the network will become overloaded, leading to blackouts potentially damaging equipment that must be repaired or replaced before supplies can be restored.

A possible solution would be to design a HEMS such that, in the event of a loss of communications, it disconnected from the power system, cutting the power supply in the process. This would prevent the worst potential consequences, but at the expense of unnecessary power outages for customers.

Alternatively, we could design in a “fall-back” mode. At its simplest, this could restrict each household’s demand so that it was impossible for the network to be overloaded. This would only apply in the rare event of a comms failure in order to prevent the network being overloaded and the risk causing blackouts and/or damage to the system.

The team at Energy Systems Catapult have been considering how we could implement a fall-back mode that would be effective, cheap, robust and minimise the disruption to customers.  We’ve outlined one possible approach not as a proposed solution, but to explore the issues involved and, hopefully, stimulate other ideas.

How it could work

In the absence of any communications, how could a HEMS estimate the loading on the network? The only data points available are parameters that can be measured locally, such as load current, frequency and local voltage.

Load current gives little information about what is happening on the wider network and could only be used to limit each household’s demand. To protect the network this would have to be set at a low value, causing great inconvenience to customers. Frequency gives a view of the national load/supply balance, but no information about the loading of the network.

Local voltage, as a single data point, also has limited value. However, if the HEMS stores a record of the interventions received from the smart grid and the local voltage when the message was received, it can use AI to learn the pattern linking the two. In the absence of smart grid communications, it would then be able to measure the local voltage and take actions based on its previous experience.

Clearly, this process can only provide an approximation of what would happen with full communications. However, if it was deliberately designed to err on the side of caution, it would allow the network to continue to operate without risking damaging overloads.

While improved network monitoring and demand response will boost the energy that a network can deliver, thought must be given to what happens if the control system fails due to a fault or malicious action.

This approach is by no means the only one and would need to be tested in real homes to ensure it would be acceptable to consumers. However, based on the Catapult’s energy system expertise and the work we have delivered for the Future Power Systems Architecture project, we believe that it can successfully control the system under such circumstances, providing a cheap, robust way to balance the protection of the network against the impact on customers.

Phil Lawton is a power systems engineer at Energy Systems Catapult. This article was originally published in Network.