Affordable, self-healing power grids are closer than you think
- Reference: 1706794208
- News link: https://www.theregister.co.uk/2024/02/01/selfhealing_microgrid_sandia/
- Source link:
When it comes to the power grids that distribute electricity to homes and businesses, however, a lot less has changed. Sure, there've been tweaks added here, and there and new forms of electricity generation have been introduced, but by and large the design is the same.
Our current electrical paradigm isn't sustainable. In just 142 short years power generation from [1]burning fossil fuels has [2]changed the world's climate, necessitating yet another wave of electrification – this time from clean, renewable energy sources including solar and wind. With that new energy paradigm comes the need for [3]a new grid , and with it a host of challenges to overcome.
[4]
Sometime soon large, regional power grids supplied by a few solitary fuel-burning giants will hopefully be gone. In their place will be interconnected microgrids fueled by smaller distributed power generation plants, such as wind and solar farms, Dr Michael Ropp, an electrical engineer at the Sandia National Lab over in America, told The Register this week.
[5]
[6]
Rethinking the grid isn't simple. Grids are mostly designed with single one-way power lines feeding AC current from power plants to customers. Renewable energy sources like solar and wind typically produce direct current electricity, requiring an inverter to turn it into alternating current. All those distributed inverters spread over a whole bunch of small grids mean it's much easier for a grid to end up in a loop, as power flows in different directions among small, interconnected systems.
Keeping a bunch of microgrids playing nice with each other – and not destabilizing due to the creation of unintentional closed loops – will be tricky, if not impossible, without a bunch of new tech. The US power system as it stands isn't designed for such decentralization.
[7]
That's where Ropp and his fellow engineers at Sandia and its partner facilities come in. They've been working on methods to create the ideal self-healing power grid, and they think they've found a far more reliable way to do it than has been tried to date. This technology could be deployed anywhere, really, in theory and depending on the circumstances.
The modern self-healing grid: Not sci-fi, but not cheap
There's no need to wait for a future of electrical lines filled with self-replicating nanites for a self-healing grid to become a thing – it's not even a new concept.
Development of such power-shifting systems has been a stated priority of Uncle Sam since the [8]codification of the US Energy Storage Competitiveness Act of 2007, which was designed to spur development of a number of electrical innovations – self-healing grids among them.
The US code defines a self-healing grid as one "capable of automatically anticipating and responding to power system disturbances, while optimizing the performance and service of the grid to customers." Such technology has even been [9]deployed by power providers like Charlotte, North Carolina-based Duke Energy in several states.
Duke's system is typical of existing self-healing grids. It involves "remote sensors and monitoring, as well as advanced communication systems that deliver real-time information from thousands of points along the grid … to make real-time decisions to keep power reliable," according to its website.
[10]
Self-healing grid technology, said Duke, can reduce the number of customers affected by an outage, decrease the time necessary to locate a problem, speed up power restoration and reduce downtime due to natural disasters and other events.
[11]US cities are going to struggle to green up their act by 2050
[12]Microsoft hires energy mavericks in quest for nuclear-powered datacenters
[13]Google goes geothermal to power some bitbarns
[14]Ireland to develop datacenter powered by fuel cells
Of course, those advancements aren't without their own impediments. Such self-healing grid technology is expensive and – like the current grid – centralized. So a failure could knock the entire thing offline.
Networks of fiber optic cables, monitoring equipment, and lots of other costly hardware is necessary to make self-healing grids like Duke's possible. Using traditional telecommunications to monitor the grid also means there's a potential for cyber attacks, and scaling such systems is a further problem.
"In a major problem situation of any type, you may lose those communications," Ropp told The Register . "And in some cases, those comms are expensive."
With those drawbacks in mind it would be hard to justify wide-scale deployment of such self-healing technologies to modernize the grid – especially given so many clean energy projects are [15]already behind schedule and threaten to [16]derail clean energy goals.
Ropp and his fellow researchers want to solve this problem by looking at ways to prevent the formation of closed loops without needing the total situational awareness provided by such self-healing designs.
"We're trying to figure out how to avoid creating a loop if the only information I can see is the information right where I am," explained Ropp, emphasizing that a key goal is avoiding reliance on a system of expensive communications equipment.
The future-future grid is already ready
Ropp and his Sandia-led team of researchers, with collaboration from boffins at New Mexico State University, have developed a method of detecting potential disruptions between microgrids using nothing but software algorithms. Better yet, the system doesn't require any new hardware, and could be readily deployed as relays – the microprocessor controls for grid switches that reconfigure electrical systems in various ways.
"New software on existing hardware was our focus on this project," Ropp said. "Almost all of what we're doing is deployable on existing commercial hardware."
[17]
Sandia National Lab's Dr Michael Ropp, who led development of an algorithm that could make future electrical grids self-healing without the need for new hardware ... Click to enlarge.
As described in a pair of papers published in [18]2022 and [19]2023 , Ropp and his team have sussed out a system that works at each relay switch – without any knowledge of the rest of a larger system of microgrids between which relays would form bridges.
By looking at the frequency of voltages on either side of a relay and running the measurements through an algorithm, Ropp's software arrives at a correlation coefficient between the two sides that determines whether two microgrids should be disconnected to prevent a loop forming.
Each microgrid relay, equipped with the necessary code to make that determination, could act independently to prevent grid malfunction. [20]According to Sandia, those algorithms could be used to determine when a portion of a grid should be shut off to maintain power supplies to critical resources (like hospitals), and could reorganize to avoid damaged microgrids – much like the existing centralized systems in use by companies like Duke.
With the hardware necessary for such a system largely in place, this isn't a distant, far-term project – it could be in place in less than five years, it's claimed.
We've got the technology, we're ready to go with this
"We use a lot of existing functions that are already used in the power system – we just use them in new ways to try to detect new things," Ropp told us. "We came up with [the loop detector] ourselves to solve a specific problem, but the whole idea is that this is something that can be practically applied on power systems tomorrow. We've got the technology, we're ready to go with this."
Of course, testing will be needed to ensure the preliminary results demonstrated in the papers bear out in the real world. "We want to pound the living daylights out of it to make sure that it really does work," Ropp explained. "We're confident, but we haven't done larger scale testing yet."
The team is already setting up test facilities at Sandia, and has partnered with several utility companies around the US to ensure the concept works across different power system design philosophies. Ropp isn't sure where the tech may be deployed first, but suggested it could end up being tested in multiple locations, once validated in the lab.
As for whether we can make the transition from our old centralized electrical paradigm to a world of distributed generation and microgrids, Ropp has faith we can, with the matter all boiling down to how affordable we can make it.
"What we're trying to do is to create solutions that allow us to meet the challenge without breaking the bank," he declared, "and [the self-healing grid algorithm] is what that's all about." ®
Get our [21]Tech Resources
[1] https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions
[2] https://www.theregister.com/2023/09/27/climate_change_carbon_capture/
[3] https://www.theregister.com/2023/10/18/electrical_grid_investment_is_the/
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/publicsector&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2ZbvOPrKKzWZPVXzUFf@wyAAAAEM&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/publicsector&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44ZbvOPrKKzWZPVXzUFf@wyAAAAEM&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/publicsector&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33ZbvOPrKKzWZPVXzUFf@wyAAAAEM&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[7] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/publicsector&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44ZbvOPrKKzWZPVXzUFf@wyAAAAEM&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[8] https://www.law.cornell.edu/uscode/text/42/17231
[9] https://www.duke-energy.com/our-company/future/self-healing-technology
[10] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/publicsector&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33ZbvOPrKKzWZPVXzUFf@wyAAAAEM&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[11] https://www.theregister.com/2024/01/19/us_cities_renewables_research/
[12] https://www.theregister.com/2024/01/23/microsoft_nuclear_hires/
[13] https://www.theregister.com/2023/11/29/google_running_on_geothermal/
[14] https://www.theregister.com/2023/11/07/ireland_to_develop_datacenter_powered/
[15] https://www.theregister.com/2024/01/19/us_cities_renewables_research/
[16] https://www.theregister.com/2022/07/01/2050_carbon_emission_goals_need/
[17] https://regmedia.co.uk/2024/01/31/michael-ropp-sandia.jpg
[18] https://ieeexplore.ieee.org/document/10012232/authors#authors
[19] https://ieeexplore.ieee.org/document/10071548/authors
[20] https://newsreleases.sandia.gov/healing_grid/
[21] https://whitepapers.theregister.com/
Re: I beg to differ
Depending on where they are in the world they might be solar, they might be wind, but let's scupper this greeny folklore about smaller plants being the future. I'm surprised the article didn't throw in worlds like "pro-sumer", "wholegrain", "community cooperative", "beard oil", and the like.
I think to an extent, it depends on the market. The US is already to a degree a series of micro-grids, ie states like Texas or California doing their own thing and being semi-isolated from the US 'National Grid'. Which at times causes both them and the grid problems. Other countries like the UK aren't as easy, especially if extended to dependent neighbours like NI, or Ireland itself.
Re: I beg to differ
"The US is already to a degree a series of micro-grids, ie states like Texas or California doing their own thing"
Most US states including those two names hardly qualify as micro-grids. If California were a country they'd be the 5th-6th largest economy in the world, with a population as great as Canada (or Poland, or Morocco), and a land area greater than Germany. Texas has a much greater area, slightly smaller population, but similar comments apply - these aren't micro-grids, they are near enough similar status to a fully fledged nation, each state having their own energy regulation arrangements. Almost all adjacent nations have a few cross border links, meaning that power is both imported and exported - true of the US as a whole, its component states, and the countries of Europe and elsewhere.
For years the UK has relied upon cheap excess French nuclear power to support its grid, and this has encouraged the dim thinking of Ofgem a whole load of new undersea connectors in the mistaken belief that as UK power demand rises and our conventional and nuclear generating fleet wither, other countries will magically have a surplus of cheap power when Britain needs it. With French nuclear capacity set to decline as the old stations close faster than new ones are being built, the position of France as a net exporter is going to fade away, and with the electrification of heating and transport countries will find that expecting somebody else to provide the raw generation capacity is a plan doomed to failure. Cross border or state line transfers will continue, and even increase in importance, but they're not going to be a magic bullet, and they don't address the problem that both peak demand and shortfalls in renewables between adjacent countries tends to be at similar times. Microgrids aren't going to be the answer, the answer is proper, rational strategic planning, large scale generation assets, and a focus on balancing worst cases against costs, plus putting affordability far closer to the heart of energy policy than it has been in the past fifty years.
More for interest than because it supports any particular position, there's a beautiful infographic here on European energy transfers:
https://ig.ft.com/electricity-sharing/
Well, duh.. Time for DC?
Renewable energy sources like solar and wind typically produce direct current electricity, requiring an inverter to turn it into alternating current...
...Keeping a bunch of microgrids playing nice with each other – and not destabilizing due to the creation of unintentional closed loops – will be tricky, if not impossible, without a bunch of new tech.
Not to mention 'renewables' like solar and wind being variable, intermittent, and the cause of the problem. So the solution is to throw even more money at the problem rather than recognising the root cause of the problem in the first place. With 'microgrids', one problem should be very obvious. So a grid sees a frequency or voltage drop and trips the relays to isolate. Yey! The microgrid is now protected. Or in other words.. Isolated. Where will the microgrid be getting fed power from?
This may make more sense with widespread SMR deployment where a microgrid has enough SMR-generated power to keep the lights on when the relays trip out. But I'm taking a wild guess here and thinking the person interviewed just happens to have a bunch of new tech to flog.
But if there's going to be massive grid re-engineering, maybe we should just bite the bullet and go DC instead. It'll be massively expensive, but that's the 'Net Zero' way.
Re: Well, duh.. Time for DC?
Interconnect between "the continent" and the UK are already DC, aren't they?
Titanic
“Self healing power grids are closer than you think.”
So are unsinkable ships…
Self-healing grids in UK
The GB grid is, in fact, quite far advanced when it comes to self-healing grid, albeit using the 'traditional' methods attributed to Duke in this article. For over 10 years, the UK DNO's have been rolling out technology called "Adaptive Power Restoration Scheme", a self-healing grid technology developed jointly between General Electric and UK DNO's to algorithmically restore supply in under 1 minute in case of an outage on the HV network. The maximum number of customers are restored without control field staff involvement. Many DNO's have now rolled this technology out to > 90% of their network. It does require investment in remote monitoring and control, but comms links are generally reliable. The UK regulator has been helpful in this regard by incentivising quality of service through rewards and penalties.
In a global context, this type of application is generally called Fault Location, Isolation and Service Restoration (FLISR). UK DNO's are ahead of most companies globally in use of this type of technology.
But looking at field-technologies without the need for central control may have some advantages, though concerns will remain around the safety of the approach in underground networks where engineers may be on-site and there will be risks where there are automatic actions while they are potentially working on the network. A comparative project has recently been progressing to live trial with Scottish Power and DRAX, looking at how similar technology can establish and sustain microgrids in the case of major power outages (https://www.smart-energy.com/industry-sectors/energy-grid-management/synergy-project-demonstrates-renewable-energys-black-start-potential/).
It's world-leading tech, and developed by an amazing group of engineers in Blighty (Edinburgh) in close collaboration with the UK network companies :-)
Intermittent generation
With intermittent generation without some kind of buffer (and no batteries is not the answer) budling a functional resilient grid is hard.
I beg to differ
"interconnected microgrids fueled by smaller distributed power generation plants, such as wind and solar"
That's fairytale thinking. To manage a grid, keep inertia, and have the resilience and excess capacity to meet modern demand you need huge plants. To meet consumer and business needs, not to mention decarbonisation of heating, cooling and transport, you need lots of huge plants. Depending on where they are in the world they might be solar, they might be wind, but let's scupper this greeny folklore about smaller plants being the future. I'm surprised the article didn't throw in worlds like "pro-sumer", "wholegrain", "community cooperative", "beard oil", and the like.
The biggest solar plants are in the 1-2 GW range, not a collection of unmanaged panels scattered across the roofs of pensioners bungalows. China hold the title of the world's largest solar farm - a monster outlier at 15.7 GW across nearly 350 square kilometres. The same is true of wind farms, where big is beautiful. Hornsea 1 is 1.2 GW, and Hornsea 2 is 1.4 GW. The latest wind turbines aren't for crappy little micro-grid setups of one or two local units, they have individual outputs in the range of 14 MW and rotor diameters of around 240 metres, with a total height in excess of a third of a kilometre. You don't scatter a few of these around and call it an "affordable self healing grid", you have one of a small number of globo corp build them, another globo corp project manages the construction a farm of hundreds of these, yet another finances them, and another still operates them. And this isn't "self healing" in any way at all. Solar won't offset low wind conditions as you build the solar output into your ops profile. Like wise high wind won't necessarily correlate with any demand spikes, or loss of output from other units or grid linkages.