One of the promises of the smart grid is the opportunity to create self-healing grids that are able to perform near real-time isolation and restoration of power outages. And now the technology is at a point where it can be implemented virtually off-the-shelf, resulting in uptake by an increasing number of North American utilities.
“The United States holds an ‘undesirable’ award from the DOE and NERC – that of suffering more blackouts than any other developed nation, and moreover that number has been steadily increasing over the past decade,” Sacha Fontaine, head of Smart Grid at Theorem Geo Associates, told Engerati in an exclusive interview, explaining this growing uptake.
“Power quality and reliability is the number one factor in the JD Power ratings – and while interestingly, the more information customers are given on outages, the more accepting they are of longer outages – by reducing the duration of outages we are improving the reliability as well as providing business value.”
Self-healing team design
Fontaine, who is working with several North American utilities on self-healing grid implementation, explains there are just a few components required for a basic self-healing team between two substations. These are breakers at the substations, reclosers close to the middle of the feeders, either geographically or to the number of customers connected, and in between them a tie recloser (which is normally open).
Underlying these is a self-healing software which is monitoring the states and conditions at each of the points. If a fault is detected, it can be isolated to the smallest segment and if the conditions permit restoration will happen.
“Multiply this up to all the substations in a utility jurisdiction and one has a self-healing grid which is able to reconfigure in the case of a fault, minimising the number of customers who are affected,” he explains. “Typically restoration occurs within 30 to 60 seconds – usually before the grid operator is even aware there was a problem.”
Typical conditions monitored include absence of hot line tags, breaker lock out alarm, loss of three-phase voltage on the source side of the recloser and the load capacity.
The other underlying component is the communications network, e.g. cellular or mesh, depending on the utility’s requirements, and typically which is already in place for smart metering. In the US also, to meet NERC security requirements, security gateways are being installed at the connection points to avoid a breach of the grid and an upstream hack of the control centre.
In the most complex implementations currently, up to ten feeders have formed part of one self-healing team, offering multiple reconfiguration options.
A US utility self-healing team implementation
In one large US utility implementation, over 400 self-healing teams are being installed across almost the entirety of its jurisdiction – the first such large-scale implementation.
As of April 2016, 237 self-healing teams were in operation. In the year from May 2015, 468 self-healing operations had been performed and a total of more than 69.1 million customer minutes of interruption had been saved. This corresponds to 1,930 customer interruptions per team.
There also has been a 13% improvement in SAIFI (System Average Interruption Frequency Index). Fontaine says this is typical for such installations, and depends on factors such as the state of the network and the number of self-healing teams installed. For example, the biggest gains will be with self-healing teams installed in the most densely populated areas.
“The main business driver for this company, and indeed virtually all utilities, is to improve the SAIFI, which results from reductions in both the number of interruptions and their length,” he explains. “In addition there is the cost saving resulting from a reduced number of truck rolls, not having to send crews to fix problems or to patrol the feeders. Now we know which segment of the feeder has a problem and can target a crew directly there.”
For example, in the large US utility case, almost US$22,000 was saved in eliminated truck rolls in the the first quarter of 2016.
Self-healing teams with distributed generation
Current self-healing teams can be installed with off-the-shelf components and some basic software configuration.
Fontaine says the next phase of development will occur when entire networks are covered under control of a distribution management system (DMS). Reconnections will occur on the basis of load flow calculations and capacity verification within the DMS, effectively optimizing the use of the equipment in the field.
A further factor is the rise of distributed solar energy, which is offering an additional power source that can be drawn on for an outage restoration.
“This will require improved modelling to take into account the available capacity before reconnecting,” says Fontaine. “We don’t know yet whether we will need data on all the systems connected to a feeder or whether an average will be sufficient and because of the intermittency of solar and wind there is also the potential for imbalances.”
He adds that a pilot is expected to start at a US utility later in the year.
“This is our next challenge and by the end of the year we expect to demonstrate self-healing teams taking into account solar installations.”
Sacha Fontaine is the head of the Utility IoT practice at Theorem Geo Associates, an engineering consulting firm focused on the utility industry. He has over 15 years of experience leading grid modernization and renewable energy efforts. He supports several of the largest power suppliers in North America including Duke Energy, Siemens, GE and Hydro-Quebec.