ORNL microgrid could standardize small, self-sustaining electric grids

November 5, 2014 by Katie Elyce Jones, Oak Ridge National Laboratory

When Department of Energy and Oak Ridge National Laboratory researcher Yan Xu talks about "islanding," or isolating, from the grid, she's discussing a fundamental benefit of microgrids—small systems powered by renewables and energy storage devices. The benefit is that microgrids can disconnect from larger utility grids and continue to provide power locally.

"If the is always connected to the main grid, what's the point?" Xu said. "If something goes wrong with the main grid, like a dramatic drop in voltage, for example, you may want to disconnect."

Microgrids are designed to not only continue power to local units such as neighborhoods, hospitals or industrial parks but also improve and reduce cost when connected to the main grid. Researchers predict an future more like a marketplace in which utility customers with access to solar panels, battery packs, plug-in vehicles and other sources of distributed energy can compare energy prices, switch on the best deals and even sell back unused power to utility companies.

However, before interested consumers can plug into their own energy islands, researchers at facilities such as ORNL's Distributed Energy Control and Communication (DECC) lab need to develop tools for controlling a reliable, safe and efficient microgrid.

To simulate real scenarios where energy would be used on a microgrid, DECC houses a functional microgrid with a total generation capacity of approximately 250 kilowatts (kW) that seamlessly switches on and off the main grid.

This grid includes an energy storage system that generates 25kW of power and uses 50kW•hours of energy built from second-use electric vehicle batteries, a 50kW- and a 13.5 kW-solar system and two smart inverters that serve as the grid interfaces for the distributed energy emulators. Programmable load banks that mimic equipment consuming energy on the grid can provide sudden large load changes and second-by-second energy profiles.

"A microgrid should run an automated optimization frequently, about every five to 10 minutes," Xu said.

To optimize grid operations, microgrid generators, power flow controllers, switches and loads must be outfitted with sensors and communication links that can provide real-time information to a central communications control.

"Microgrids are not widely deployed yet. Today, functional microgrids are in the R&D phase, and their communications are not standardized," Xu said. "We want to standardize microgrid communications and systems so they are compatible with the main grid and each other."

Now two years into the inception of ORNL's microgrid project—"Complete System-Level Efficient and Interoperable Solution for Microgrid Integrated Controls," or CSEISMIC—the microgrid test bed at DECC is functional and employs an algorithm developed at ORNL that directs automatic transition on and off ORNL's main grid.

Xu said the next year will focus on getting the energy management system (EMS) running. The EMS will drive optimization by allowing microgrid components to fluctuate operation based on parameters such as demand and cost.

"The EMS may, for instance, tell the PVs [solar cells] how much power to generate for the next five to 10 minutes based on the time of day and energy demand," Xu said.

The CSEISMIC team has long-term goals of partnering with industries to conduct field demonstrations of standardized grid prototypes.

"As soon as microgrids are standardized and easy to integrate into the main grid," Xu said, "we'll start seeing them in areas with a high penetration of renewables and high energy prices."

Explore further: Building a better telecom system

Related Stories

Building a better telecom system

July 22, 2008

Hurricane Katrina helped University of Texas professor, Alexis Kwasinski, formulate a new plan for the U.S. telecom system: a de-centralized power architecture that would have kept the lights and phones on in New Orleans. ...

Microgrids: So much more than backup energy

August 27, 2012

(Phys.org)—Most Americans don't have to think much about energy reliability. We plug in a computer and it powers up; we flip a switch and the lights come on.

Battery system will be able to light 2,500 homes

September 30, 2014

One of the largest, most environmentally-friendly, battery-based energy storage systems in the nation will be installed at the University of California, San Diego the campus announced today (Sept. 29).

Climate change shifts focus for energy system

May 30, 2014

The U.S. National Climate Assessment report states bluntly that streets in coastal cities are flooding more readily, that hotter and drier weather in the West means earlier starts to wildfire seasons, and that every region ...

Recommended for you

Technology near for real-time TV political fact checks

January 18, 2019

A Duke University team expects to have a product available for election year that will allow television networks to offer real-time fact checks onscreen when a politician makes a questionable claim during a speech or debate.

Privacy becomes a selling point at tech show

January 7, 2019

Apple is not among the exhibitors at the 2019 Consumer Electronics Show, but that didn't prevent the iPhone maker from sending a message to attendees on a large billboard.

China's Huawei unveils chip for global big data market

January 7, 2019

Huawei Technologies Ltd. showed off a new processor chip for data centers and cloud computing Monday, expanding into new and growing markets despite Western warnings the company might be a security risk.

20 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

Eikka
3.7 / 5 (6) Nov 05, 2014
"The EMS may, for instance, tell the PVs [solar cells] how much power to generate for the next five to 10 minutes based on the time of day and energy demand," Xu said


You don't "tell" a solar cell how much to produce. It gives you as much as it can at all times, and if you don't accept the energy it makes there and then, you're wasting money and resources.

An energy source like solar power becomes dispatchable only when you have a sufficient energy buffer. If the capacity factor of the PV system is 0.14 over the course of a full day, you will need something on the order of 3x its nominal capacity in kWh to actually smooth it out to the average output, so you could then serve such an average demand. For example, a 1 kW panel would require 3 kWh of batteries, which get topped up during the mid-day peak and drained during the rest of the day.



Eikka
4.2 / 5 (5) Nov 05, 2014
A couple years back I also ran a computer simulation about wind power on a national scale with an average capacity factor of 0.25 to see how much batteries you'd need to smooth its output.

The assumption was that the power limiters apply at 12 m/s and cut off happens at 20 m/s, and the wind speeds are Rayleigh-distributed around 6.25 m/s mean wind speed. The maximum power variation was then smoothed out to be no more than 4% of the nominal per hour to represent wide geographic spread.

I then assumed that all the power was put into a big battery which is being continuously drained at the same rate as the capacity factor, or 25% of the nominal peak output.

The battery in the simulation started to fill up in a random-walk fashion and peaked at roughly 150 MWh for every 1 MW of wind turbines in the system, staying mostly below 100 MWh for 2/3rd of the time over a simulation period of one year.

That illustrates what massive amounts of batteries are needed to smooth the output.
Eikka
4.2 / 5 (5) Nov 05, 2014
Now, the weakness of the abovementioned simulation is that it assumes all the turbines wind up and down at the same time, so it's not truly representative of a large geographical spread and that inflates the results as they'd apply on actual national scale.

But the assumptions work better for microgrids where the turbine system is constrained to a small geographical area, like one county. Such a microgrid utilizing wind turbines, if they were to disconnect from the grid at large, would need on the order of 4 days worth of battery reserve capacity and preferably 6 days worth, so they wouldn't have to dump the power all the time and then turn on other generators when it isn't available.

And then there's seasonal energy storage as well, because solar/wind output varies with the seasons, but that's a whole other can of worms.

Eikka
4.2 / 5 (5) Nov 05, 2014
Actually, that's 16-25 days of reserve capacity required, because I accidentally calculated it over the nominal power instead of the average power output, and it's the average output you're trying to meet the demand with.

The reason isn't that you'd have to subsist for 25 days in a row entirely without wind on battery power alone - the probability of that is exceedingly small - but that you often have to suffer weeks and weeks of reduced output because it just isn't very windy, and conversely you may have weeks and weeks of higher than normal winds that you need to back up somewhere.

gkam
2.3 / 5 (6) Nov 05, 2014
I guess the rest of the world did not read the notes from Eikka. Wind provided the most added capacity last year, and even over 50% of England's power in a high storm.
gkam
2.3 / 5 (6) Nov 05, 2014

Why do the simple folk assume our grids will be supported by only one source of power? Do they have ANY idea the availability and benefits of alternatives, such as geothermal? Look up US geothermal reserves.

Then, we have solar, and hydro, and smaller sources. All, including efficient thermal plants will be used, some fueled by natural gas.

Eikka
3.7 / 5 (6) Nov 05, 2014
I guess the rest of the world did not read the notes from Eikka. Wind provided the most added capacity last year, and even over 50% of England's power in a high storm.


That's a red herring. Added capacity doesn't mean it's actually good for anything. It just means someone is getting paid for erecting wind turbines.

Why do the simple folk assume our grids will be supported by only one source of power? Do they have ANY idea the availability and benefits of alternatives, such as geothermal? Look up US geothermal reserves.


I have, and they're minimal where the population is maximal.

Currently the only systems that are truly scalable to the whole grid scale are wind and solar, which happen to be antagonistic towards each other because their production overlaps and neither is dispatchable. In essence, you can mostly choose just one and discard the other, so the options ARE rather limited.
Eikka
4.2 / 5 (5) Nov 05, 2014
and even over 50% of England's power in a high storm.


That's btw. exactly the reason why it is problematic.

If England had twice the wind turbines, it would produce 100% of the power in a high storm. Trouble is, it would still produce only 14% on an average day.

At the amount of capacity where you would technically reach half of your demand, you're getting production spikes that equal four times the electricity demand, and there's absolutely nowhere you can put the power. Nobody's going to buy random power spikes the magnitude of a country because the grids just can't handle it.

You simply have to waste it. That in turn detracts from your production average and so adding more capacity simply means you're wasting more and more money for smaller and smaller gains in energy production.

retrosurf
4.2 / 5 (5) Nov 06, 2014
"You simply have to waste it. That in turn detracts from your production average and so adding more capacity simply means you're wasting more and more money for smaller and smaller gains in energy production."

And there's nothing wrong with that. The economics of wasting (or rather, not utilizing) power from wind and sun are very different from the economics of wasting power from conventional fossil-fueled generation. There are no fuel costs to solar generation, which are approximately 80% of power production costs for coal and natural gas, and approximately 35% for nuclear power.

Dispatchable solar power can (and will be) provided with Smart Grid, without batteries. The grid-ties from solar arrays will be instructed to "sell" as much or as little of their capacity as is needed to smooth output. Overcapacity will provide the energy buffer, not batteries.
Lord_jag
1 / 5 (2) Nov 06, 2014
You have to waste it. So what?

So... your solar farm produced 200KW and you only used 50KW. So What? That's 50KW for several hours that didn't have to come from another source.

Wind turbines making more power than you need? Don't use it. So what?

Extra production capacity in renewables isn't a problem. A solar panel is quite happy if you don't draw power from it. Nothing bad will happen.

A nuclear reactor, however, is a lot less happy if you decide not to draw your Megawatt of power from it. In fact, not taking the power it made might be enough to make it want to kill us all.
Lord_jag
1 / 5 (1) Nov 06, 2014
And... you know what? If a huge solar farm was constantly producing energy not being used I'm sure someone would come up with a way to use FREE, yes FREE electricity on site.

"Oh you paid to install 150KW of production that's not being used right now? Well let me see if there's *Something* I can do with this free energy that I can make money with"

I can't even imagine the possibilities of what I could do with it.

You could even balance the load by using the smart meters for the reason they were sold in the first place *GASP* you could reduce the price of energy when the supply was excessive and increase it when it wasn't. Then people all over the grid could set appliances to use more power when it was cheaper.

Just imagine if at 3am the energy was free to you instead of paying a neighboring state to take it from you.
Modernmystic
not rated yet Nov 06, 2014
The problem isn't waste the problem is dispatchability. You effectively need 70% at the moment, which means when you're building a wind farm what you're really doing is building 30% wind farm and 70% coal, nuclear, or other grid friendly power plant.
Eikka
5 / 5 (3) Nov 10, 2014
And there's nothing wrong with that.

You have to waste it. So what?


The economics of a wind turbine are not so different from the economics of a conventional powerplant operating as baseload. The "fuel" is exceedingly cheap, but the plant itself is not, and the plant has a limited operating life and ongoing financial and maintenance costs, land rents etc.

That means, for X amount of dollar paid over the lifetime of a wind turbine, you get Y - N amount of energy where Y is the amount you could make and N is the amount you waste, so the price of the energy is X : (Y - N).

The cost of energy increases.

Wasting the energy also means lowering your energy yield per square meter, so you have to allocate more land for your turbines and solar panels, which has a marginal cost effect of having to put the turbines in worse locations which again increases cost and decreases output.
Eikka
5 / 5 (3) Nov 10, 2014
If I check back on the simulation model I used, I can see that you can increase the capacity factor of wind power to 58% by wasting 52% of your potential energy gains, and to 75% availability by wasting 78% of the energy. This would be done essentially by applying the brakes at lower wind speeds all the way down to 5 m/s, so you can pretend that the turbine has a lower nominal power. These scenarios represent an increase in cost of wind energy by a factor of 2 to 4, which translates to building 2-4 times as many wind turbines to gain the same amount of energy.

Most large turbines won't even start until 6 m/s, which means you have to go back to tiny cost-inefficient turbines like the ones in Altamont Pass.

The effect is quite dramatic, and the moment you reach the amount of wind power that you can comfortably fit in the grid without spilling over, building more turbines will see a massive cost increase for neglible gains.
Eikka
5 / 5 (2) Nov 10, 2014
And... you know what? If a huge solar farm was constantly producing energy not being used I'm sure someone would come up with a way to use FREE, yes FREE electricity on site.


I'm sure someone could use cheap spill-over energy from wind turbines and solar panels, but the producer isn't getting their investment's worth by selling it because the demand for such power would only exist if the price was lower than what they could already buy from the grid.

And the ultimate trouble is that when speaking of powering entire counties, states or countries, the "surplus" is on the order of dozens of gigawatts, even hundreds, and there's no transmission grid that can transmit it, nor are there any consumers that can kickstart a factory or some process that would consume as much while remaining idle for over 90% of the time.

Maybe you could heat a big lake with it and go for a swim in the winter.
Eikka
5 / 5 (2) Nov 10, 2014
Just imagine if at 3am the energy was free to you instead of paying a neighboring state to take it from you.


There is never such a thing as free energy.

If you don't pay for it directly, you pay for it somewhere else because someone has to pay the upkeep of the turbines and panels, and that someone will pass the expense on to someone else, and eventually, you.

Currently you're paying for it in taxes, because public subsidies are used to make up the difference between what the market pays, and what they have to pay to keep the system running.

Plus a nice profit to the owners of course - to be invested in "Research & Development", which is a pseudonym for "how can we make even more money out of this?".
Lord_jag
1 / 5 (1) Nov 13, 2014
Don't talk to me about subsidies. Today the report came out about how much tax dollars are spent subsidizing GHG production. 2 trillion dollars.

I still see you're still waving the red herrings of energy storage and energy transmission. Those are truly red herrings.

All power is transmitted right now. Decentralizing the power production to the area's it's needed would need less transmission not more.
Lord_jag
1 / 5 (1) Nov 13, 2014

And the ultimate trouble is that when speaking of powering entire counties, states or countries, the "surplus" is on the order of dozens of gigawatts, even hundreds, and there's no transmission grid that can transmit it, nor are there any consumers that can kickstart a factory or some process that would consume as much while remaining idle for over 90% of the time.

Maybe you could heat a big lake with it and go for a swim in the winter.


You apparently didn't see or read the part about "using smart meters for the reason they were sold to us"

If we were sold energy for even close to a reasonable markup from the current market price, the surplus would magically disappear. We have, right now many many different devices in our homes and businesses that are smart enough to decide when to use energy: Water heaters, air heaters, air conditioners, car chargers, washing machines and driers.

All these devices could be set up to use energy when it's cheap.
Lord_jag
1 / 5 (1) Nov 13, 2014

That means, for X amount of dollar paid over the lifetime of a wind turbine, you get Y - N amount of energy where Y is the amount you could make and N is the amount you waste, so the price of the energy is X : (Y - N).

The cost of energy increases.

Wasting the energy also means lowering your energy yield per square meter, so you have to allocate more land for your turbines and solar panels, which has a marginal cost effect of having to put the turbines in worse locations which again increases cost and decreases output.


Again, so what? This is the basis of a feedback mechanism. Stop subsidizing oil and let the market figure out what's cheaper.

Solar is just getting started. Right now a 10KW rooftop installation is less than $10K. When it's as cheap as an iPhone people won't care if it's not used to it's maximum capacity any more than they care if every clock cycle on their computer or phone is calculating something important.
Lord_jag
1 / 5 (1) Nov 13, 2014
Of course... You would NEVER question any non-renewable in this way.

When the cost of billions of dollars goes to setting up a natural gas burning electric plant, no one ever noticed that it will often be shut down and not producing electricity.

How about if you take one second and consider the same issues with the glass houses our current energy production technology lives in before you go throwing rocks at the brick houses of renewable? You think you're hurting it, but all you're doing is increasing awareness of the closed mindedness of oil industry.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.