#1 cleantech news, reviews, & analysis site in the world. Subscribe today. The future is now.


Published on December 22nd, 2015 | by Jake Richardson


AES Reveals Another Energy Storage Alternative To Peaking Power Plants

December 22nd, 2015 by  

Energy storage leader AES recently revealed its new energy storage platform for commercial deployments. AES installed one of its systems in Maryland, and it is the largest grid-scale battery in that state. Brian Perusse, VP of International Market Development for AES, answered some questions about the platform for CleanTechnica.

In what ways is the Advancion 4 system a better alternative to peaking power plants?

Advancion 4 saves customers money, reduces emissions, provides more flexibility, and is more reliable than a peaking power plant. Advancion arrays are continuously and instantly available without burning a fuel in “standby mode” and can deliver cheaper energy than a peaking power plant. The arrays serve as both generation and load, marking a turning point in grid flexibility and renewable integration, by giving power producers more than twice the flexibility compared to a gas peaker plant on the same interconnection. Advancion can deliver power significantly faster, more reliably and with more accuracy than a gas peaker and thus helps improve both the reliability and efficiency of the whole power grid.

What is the capacity in megawatts and megawatt-hours?

Advancion 4 has a unique scalable design that allows for standard configurations from 100kW to over 1,000 MW, and from 15-minutes of duration to over four hours without any reengineering. The largest AES Energy Storage project to-date is a 100 MW Advancion Array, that will deliver 400 MWh of energy to Southern California Edison.

What is the cost of a system?

System costs vary depending on the size, duration, and online date of the project.

Does it work well with renewable energy like solar power?

Yes, Advancion provides flexibility, peak capability, and grid-stabilizing capabilities that are critical for integrating renewable energy, such as solar and wind power, into our electric grid. Even better, Advancion does not need to be collocated with renewable generation; it can be sited at critical points on the grid, and still deliver the same benefits to renewable energy.

What makes the Advancion system suitable for commercial operations like utilities?

Advancion 4 is among the most proven energy storage platforms available, resulting from AES’ more than eight years of commercial experience operating grid-connected energy storage. AES developed, owns and operates the largest fleet of grid connected energy storage, with nearly 3 million megawatt-hours of service delivered. Advancion 4 delivers the highest reliability, has the lowest total cost of ownership, and is available to utilities, developers and power system operators to own directly or through a managed service contract.

Do you anticipate installing any that are 100 MW or larger?

Yes. AES has been awarded a 20-year Power Purchase Agreement (PPA) by Southern California Edison (SCE), to provide 100 MW of interconnected battery-based energy storage. Advancion 4 is also being evaluated for other projects larger than 100MW in several markets and countries.

Is the Advancion system containerized, or is it located within some other structure?

The Advancion solution can be deployed either in containers, or for larger scale projects, in a building.

What type of batteries are used?

As part of the Advancion solution, AES has developed relationships with the world’s leading battery manufacturers to supply the most advanced and highest quality batteries on the market. The current preferred technology is Lithium Ion, but as battery technology evolves, Advancion’s certification process will allow for these new technologies to be incorporated – effectively future proofing Advancion.

Who is the battery manufacturer?

Advancion has relationships with the world’s leading battery manufacturers, supplying the most advanced and highest quality batteries on the market.

What does it mean that the Warrior Run Advancion Energy Storage Array has 10 MW of capacity that is equivalent to 20 MW of flexible resource?

Batteries can either charge or discharge, both of which are valuable to the grid operator for stabilization of the grid. The grid operator can use batteries to absorb up to 10 MW of excess power from the grid, or supply up to 10 MW of power to the grid. Both of these services are considered a valuable resource to the grid, and this -10 MW to +10 MW range is considered a 20 MW resource.

Once the system is installed, who monitors its performance, and is that done with software and real-time analytics?

The Advancion system has sophisticated software that can, in real-time, monitor the health and capability of the battery system. For example, the Warrior Run installation regularly measures and records nearly 80,000 separate data points, representing more than two terabytes of data per year. This data is evaluated and used to make real-time and longer term decisions over the life of the system.

What kind of maintenance is required?

AES operates the largest fleet of battery based energy storage projects in the world and has a proven track record of safely operating and maintaining these systems. The systems require scheduled maintenance such as changing air conditioner filters to replacing parts as needed. The Advancion architecture allows for segmented battery maintenance. This feature maximizes system availability even while maintenance activities are being performed.

Image Credit: AES

Tags: , ,

About the Author

Hello, I have been writing online for some time, and enjoy the outdoors. If you like, you can follow me on Google Plus.

  • eric mair

    I don’t see any mention of when these batteries will need replacing?
    It’s not sufficient to compare cost/kW or even cost /kWh. You need an LCOE number and a life expectancy in order to compare apples with apples. (remember to include decommissioning costs)
    I don’t think batteries are the answer at this level – too costly and to short a life. There is no real answer yet to pumped hydro except for the permitting issues, oh, and the fact that you need a hill or a mountain close by.

  • Pat Campbell

    This sounds more like a company brochure than an article by a professional journalist. Could use some vetting please.

    • Bob_Wallace

      You probably would have benefitted by reading the other comments prior to commenting. You’re late to the dance….

  • Bristolboy

    To me this reads very much like an advertorial – I assume Cleantechnica were well paid?

    • bink

      what is your problem. peakers are expensive

      • Bristolboy

        I’m not peaker plants are expensive and that energy storage will one day (if not already) be able to compete without subsidy. My query is in relation to publishing a complete article which reads like a poorly disguised press release. There is no “new” information or anything which is unique – note the avoidance of reference go which battery supplier, costs etc. However, what there is a lot of is references to how great AES and the system is.

        • Jake often does interviews like this. Typically, the answers are more on point, more candid, and less sales-y than some of the ones here. If AES/Brian Perusse chooses to go this route, though, that’s its/his call. I wouldn’t advise such an approach, and don’t think it is beneficial to the company, but there are differences of opinion on the matter.

          However, I think if an article reads like an advertorial to you, I should take a closer look at these interviews before they go live and ask Jake to work on the company to provide a better interview… or just delete some of the less useful sections. Like the following one?

          “Who is the battery manufacturer?

          “Advancion has relationships with the world’s leading battery manufacturers, supplying the most advanced and highest quality batteries on the market.”

          Yeah… that doesn’t tell us much, avoids the question, and would just piss me off as a potential customer reading this article.

          • ROBwithaB

            The question still remains:

            Did money change hands?
            However negligible, if there was any sort of contribution from the company being featured (even if they simply bought some advertising space) I think your readers deserve to know.
            We’re all grown adults here. We understand that you need to make money to keep the site going. That’s all good.
            But there is an ethical line in the sand. And there’s a very easy way to stay on the right side of it. Just declare the interest somewhere at the bottom (actually, preferably at the top) of the story. “AES is an advertising partner of Cleantechnica” or “AES covered the costs for publishing this story” or “Brian occasionally buys Jake a beer” or whatevs.

            Maybe even “AES does not advertise with us (yet). But we’re hoping for some of that sweet goodness, and our normal critical thinking may be somewhat impaired as a result.”

            I gotta agree with Bristolboy on this one. It really did sound like an advertorial. A real love-fest. If it wasn’t paid, then maybe you should be asking some tougher questions of your interviewees. Or maybe doing some background research to provide a bit of balance.

            Making sure that your readers always know exactly which stories are sponsored vastly increases the value of ALL the stories you publish. And, ironically, should allow you to attract more advertising.

        • But I’ll note that this is almost always the kind of response we get on questions about cost:

          “System costs vary depending on the size, duration, and online date of the project.”

          If we don’t include our questions about cost, readers will surely complain. If this is all we can get from a company, this is all we can get. Unfortunately, energy storage companies aren’t keen on providing much cost detail to us.

          • ROBwithaB

            Sorry, Zach.

            I’m not really buying it. That’s a bit of a cop-put. You could always say something like:
            “Hmmm. Okay. Sorry, you seem to have dodged the question there. We understand you don’t want your customers or competitors to know exactly what you’re charging on this particular deal, but…. you said that your system can save customers money, but you haven’t given any numbers to back that up.
            Look, our readers are pretty up-to-date with industry trends. A lot of them are actually electrical engineers and the like, some of them in purchasing positions with utilities and municipalities. They’re a demanding bunch, to be honest. And we get huge flak in the comments if don’t give them any hard data to work with.
            So yeah… Unfortunately, we’re not really going to be able to publish this story unless we give them some sort of indications on cost. If YOU don’t give it to us, we’ll have to find that information elsewhere. But it seems better to get if from the horse’s mouth. So….Maybe just some ball park figures? A range for cost estimates, high to low? Something? Maybe some of the LCOE pricing was public information?
            Or maybe we should get back to you in a year or two when you actually know what you’re doing?”

            You know. A little bit more hardball.
            Less bromance. More hard-to-get.
            A “contact” in the industry isn’t really worth anything if they keep serving you soft porridge. Cultivate that relationship all you want, and it’s never going to be worth anything, except to them, when they decide they need you.

            P.S. You’re welcome to save my speech and read it out next time. No copyright fees required 🙂

      • evfan

        bink, I agree peakers are expensive, but so are batteries. See my posting above. Do you have any details?

    • Bob_Wallace

      Do you see a statement to the effect that this is a pay to publish article?
      If not, no money exchanged hands.

      Sometimes articles read like press releases because that’s all the author has to go on. Your choice, read the news or not.

      • Bristolboy

        No evidence, just a suspicion I had as normally stories on Cleantechnica are very technical and less salesman like than compared to this.

        I’ve also recently seen “Sponsored” stories on the Guardian which involve AES amongst others – and assumed this was an extension of such marketing exercises.

        • Ha, wish AES sponsored this one! Maybe we need to pitch them. 😀

          These interviews Jake does can run on the sales-y side. Depends on how the person/company decides to answer. I think it is something for us to address going forward.

      • ROBwithaB

        Oops. I suppose I should have read through the entire comment exchange first…
        Is that an absolute hard and fast rule of the site? If any monies or favours change hands, then the readers will be clearly informed?
        Even advertising not directly related to this particular story? Or “favoured” releases of information? Or tickets to a convention or sporting matches, or meals or drinks at the pub?
        Because that’s absolutely the way it SHOULD work.

        But I see Zach already responded directly to Bristolboy below here. And is clearly already giving some thought to the “sales-y” tone of some of these pieces.
        (And the admitted wish to have AES as a sponsor might indeed blur the lines a little.)

        But there’s no need to get defensive, Uncle Bob. It’s a perfectly normal conversation to be having, and the fact that it is occurring openly here is a GOOD thing. It means that the readers have higher expectations of Cleantechnica than for many other “news” sources that have already sold their souls.

    • Nope. No payment at all, of any kind.

      And just to be clear, we always clearly label an article if it is an advertorial.

      • Bristolboy

        Thanks for the clarification!

        • No problem. It’s not the first time in the past month someone has presumed a normal article was an advertorial. I’m not entirely clear if that means we need to change how we are reporting on some things or not.

          We definitely like to feature good cleantech companies, but maybe we’re not throwing in enough of a challenge to them at the same time…

          Stuff to think about.

  • evfan

    I’d be the first to cheer if battery storage could replace peakers, but I have my doubts if this makes economic sense. Does anybody know if some real numbers are available anywhere?
    I say this because battery systems are great, but expensive. In order to get good ROI on a battery system, you try and cycle them once or twice daily. Peakers on the other hand are cheap but inefficient generators that sit idle until the price of electricity rises high enough for them to make money.

    • Martin

      As for the ROI, what is the cost of a peaker plant and time to build as for the cost of a battery system and time to install?
      Both are money tied up $’s, and if the battery system is containerized it could be moved, try that with a peaker plant.

      • evfan

        Martin, sorry I do not understand. It seems like you think peaker plants should be portable. That is news to me. Do you care to elaborate, please?

        • Martin

          No my point was a container version could be moved if the need arises while a fixed version can not be.
          Because some times the need of the grid change, loss of industry, people etc.
          But as well the cost of both system money tied up, I think the cost per MWh are less for the battery system than the peaker plant.

          • bink

            Martin, evian is like most on here he does not understand how the grid works let alone the benefits of energy storage. Circuits get overloaded and instead of installing an expensive transformer you drop an energy storage system to defer upgrade for several years then you move it somewhere else where there are congestion and load problems. Also peakers are not cheap when they sit idle for on average 95% of the year. Batteries provide many grid benefits that peakers dont and they respond faster saving energy. With a conventional peaker it takes 50-100MW’s of power to bring 100MW into balance this is largely a reflection of response and ramp time. A battery will respond in milliseconds and ramp 1000 a min
            whereas a peaker response time is 10-12 mins ramping at 20MW a min. so to bring 100MW into balance 5 mins.. 10=5 + 15 -17mins just to balance

          • I love your detailed responses. I just wish you would tell us at some point what company you work for. 😀

            But happy to see you commenting again.

          • karrlsam

            A point made in the article is that batteries can take power off the grid. Quickly. When a transformer blows. ( storm ), or a large factory shuts down , ETC.
            This prevents voltage spikes and conditions the grid all year long.
            Peakers don’t.

          • Bob_Wallace

            That role was largely covered by running large thermal plants a few percent below their optimal output level. Coal plants apparently complained that they were being required to furnish free backup for reactors which were allowed to run full on.

          • ROBwithaB

            I learn a lot by reading your comments (and then googling the stuff I don’t understand).

          • evfan

            Martin, I am not sure you understand what a peaker is. I do not have all the numbers, but I do know that one extra MWh for a peaker costs less than $300, it is just some extra fuel to pay for. You cannot buy many batteries for that.

    • Bob_Wallace

      Gas peakers run 17 cents to 22 cents per kWh.

      The EOS Energy zinc-air battery sells for $160/kWh. If cycled daily that would cost 2.5 cents per kWh.

      If you have a situation where peakers are being run frequently it’s likely that storage is a better deal. Peakers are going to continue to be the least expensive option for power which is rarely demanded.

      • evfan

        Bob, thanks for the information. Now batteries just store energy, so if the battery is cycled daily, I presume you are charging it at night when it is cheap, and selling the power later afternoon. I am curious how you get the 2.5c per kWh. How many cycles in zinc-air battery? And what do you assume for the power you buy? And what ancillary costs are there?

        • Bob_Wallace

          I used 7% financing for 20 years. That might be too long for the EOS zinc-air which has a 15 year calendar life. I just plugged the cost into a spreadsheet I already had set up.

          EOS says 5,000 cycles. That would be 1x a day for 13.7 years.

          EOS delivers their battery in a 40′ shipping container. There would be some cost in wiring it into the grid.

          Once a day would likely be low for the first time-shifting batteries we add to the web. They could cycle twice some days, wind at night and solar during the day.

          I didn’t include cost of electricity in that rough estimate. Just the cost of storage.

      • juxx0r
    • eveee

      Battery storage systems are cheaper than gas peakers.

      Brattle Group did a study. Storage replaces gas peakers at $350/kwhr.

      Tesla PowerBlock goes for $250/kwhr. 620 million in orders in 7 days. The revolution has started.


      • evfan

        eveee, thanks for the link, but that article does not cover all cases. I know batteried help a lot with power quality and stability, but they do not eliminate regular peakers that can run for hours or days at a time in case of an emergency, such as weather, equipment failure, etc. Do you have any numbers on that?

        • Bob_Wallace

          Peakers run around 20 cents/kWh. If they had to fill in for a long term outage (like the two SONGS reactors closing without prior notice) the cost would come down some as their capital costs were spread over more MWh.

          • evfan

            Bob, you are touching on another topic. One reason why utilities are fighting rooftop solar is that solar reduces peaker MWh, but not peaker MW, so the peakers run less and cost more per hour. And now batteries are adding another variable they have to manage.

          • Bob_Wallace

            Yeah. Utilities are undergoing a lot of disruption. No one likes have their job made harder and their income threatened. Some utilities will fight the change more than others.

          • eveee

            Utilities don’t seem to mind batteries. Judging from their orders of over 620 million worth of Tesla storage in 7 days, it looks like they have warmed up to them nicely.

            They realize they are cheaper than peakers at those prices. Utilities problem is that they have existing investments in facilities that may become “stranded assets” and they don’t know how to deal with that. Thats the price of change. The fight is over who pays.

            But none of that has to do with batteries or storage and how utilities are embracing batteries to replace gas peakers.



        • eveee

          No. Read the Brattle Group Study. Batteries replace gas peakers. Period.

          You are talking as if its a theory. Utilities have already replaced gas peakers with lithium and other storage.

          This is industry stuff.

          “Southern California Edison has contracts for more than 275 megawatts of storage capacity from suppliers that include AES Corp. and Advanced Microgrid Solutions LLC. It plans to solicit as much as 120 megawatts of additional capacity in 2016.”


          All cases? Are you a power engineer? All power systems have a variety of units of various types in use and in reserve dispatched on a daily basis in response to demand. One unit does not one for one replace another.

          AES has been doing this for some time. They detail 95% availability at their Laurel Mountain facility. A multitude of energy storage units are used to provide continuous service, just like all other sources.


          • evfan

            Sorry, I disagree, peakers will remain for some time to come. To quote from Bob above
            “Peakers are going to continue to be the least expensive option for power which is rarely demanded.”

          • eveee

            OK. Argue with a utility industry exec, then. NextEra.

            Robo said, “Post-2020, there may never be another peaker built in the United States — very likely you’ll be just building energy storage instead.”


          • JamesWimberley

            Seconded. A reliable grid needs a large reserve that is rarely used; many old gas peaked plants will sit idle almost all the time, like that emergency generator in your uncle’s garage.

          • Bob_Wallace

            When SONGS went down SDG&E pulled a gas generator out of storage that hadn’t been used in years and hooked it back up.

            As we build up renewables we may see both gas and coal plants mothballed on a “just in case” basis.

          • Tim

            When “many old gas peaker plants will sit idle almost all the time,” they don’t make much CO2. So let ’em sit there – long as no new coal or gas baseload plants go up, existing coal plants get torn down, and existing gas plants become peakers.

          • Bob_Wallace

            You’re not picking up on my point. Peakers are likely to stay around for a long time but their role as frequent providers is going away. As the price of batteries drop peakers will be run less and less.

            “power which is rarely demanded”

            Peakers are not going to be “peakers” but rather “deep backup”.

          • evfan

            Bob … actually, I get your point. It is true that peakers will be needed less frequently, but they will still be needed from time to time. Batteries are ideal to react quickly and stabilize the grid for a few hours, but not much longer. The premise of the article is that Battery Storage is an alternative to peakers, which I do not believe. I think they augment peakers, and help utilities be much more efficient, reduce the amount of spinning reserves, etc

          • eveee

            Why wouldn’t batteries ( or flow ) replace gas peakers? What would limit them?

          • evfan

            Cost is the main reason. Batteries to store 1MWh cost $160000, plus the cost of electricty. 1MWh of fuel for a peaker costs $300, total.

          • Bob_Wallace

            That $160,000 worth of batteries (assume EOS zincs) can be cycled 5,000 times. They’re going to hold 1 MWh 5,000 times during their lifetime.
            That brings the cost per MWh down to $32/kWh plus electricity (5 cents?) + interest (higher than the gas peaker).

          • evfan

            “They’re going to hold 1 MWh 5,000 times during their lifetime”.

            This is where things get tricky depending on battery size: A small battery will get its full 5000 cycles. A bigger battery may only cycle partially, not full depth.
            If we look at the late afternoon peak on the duck curve, it is not the same size every day. Let’s consider the 2 extreme scenarios.

            1) If you buy batteries to smooth out the peak on the lowest peak day, then the peaker still runs every day, but just for fewer hours.

            2) if you buy batteries to smooth out the peak on the hottest summer day, then for most days of the year, you will not need to use the entire battery, which means that some of your MWhours (that cost $160000 each) is not working that day. And that drives up the average cost of the battery storage.
            These were extreme cases, in reality all installations will be somewhere in between.

          • Bob_Wallace

            I’m betting that whatever the cycling is like the cost is going to work out to less than $160,000/MWh…. ;o)

          • evfan


          • eveee

            Ouch. That math. You have to pay for the turbine, not just the fuel. Lets skip directly to the experts, Lazards:

            “For applications in front of the meter, the paper compares energy storage facilities to Lazard’s $165-$218/MWh levelized cost for a natural gas peaker plant, giving a “rough proxy” for how storage competes against conventional generation, Tappin said.”

            “The analysis shows that for grid-scale applications, the lithium-ion battery’s $211-$275/MWh can compete in some cases with the cost of natural gas peaker plants, and the LCOS of $276/MWh for flywheels isn’t far off.

            “Both technologies, co-located with renewables installations, have won contracts in competitive bidding to sell frequency regulation services to the PJM Interconnection
            November 24, 2015
            It appears there is a switch to storage and the reason is cost.

          • evfan

            You said it best yourself “Peakers wont go away, because existing ones will soldier on. But new ones will not be favored as replacements.”

            W.r.t. the existing peakers and mothballed older plant, there is not need to acquire a turbine, but it costs more than just fuel to run them.

            Right now, I think battery systems augment peakers, and help utilities be much more efficient, reduce the amount of spinning reserves, etc.
            It is a really tough time to manage power utilities. Typically they have to make acquisition decisions for equipment that lasts decades. Right now it is very hard to predict what the future needs will be. The old rules of thumb no longer apply.

          • eveee

            Yes. Thats right. Its tough for utility planning right now. The technologies are changing, and the regulatory and economic landscape is quite unsettled. Demand management is a much cheaper way of getting flexibility. US existing transmission lines have been neglected and some new ones are coming online, but its slow.

            The figures show storage making inroads, but its just started. New gas peakers are not gone yet. My guess is they will be used for decades. If the maintenance is kept up, they could be used infrequently for a long time.

            But I agree with you. I don’t see batteries alone taking over that role. Flow and other technologies like sodium do a better job for longer term or larger bulk storage. What really makes batteries exciting is that they do so much for T and D, and stability as well.

            One of the biggest problems is that the regulatory frameworks have been set up to monetize existing technology. The regulatory framework has to catch up to new technology. Its hard to monetize T and D savings properly for one. I am particularly concerned with the lack of proper price signals for demand management. IMO, it should be encouraged much more.

          • Bob_Wallace

            What many of us believe is that battery prices drop “peakers”, dispatchable generators, will be called on less as it will be cheaper to store and use electricity rather than to fire up a generator that uses fuel.

            At some point it will probably no longer be appropriate to call these dispatchable generators peakers but to call them deep backup or some other term. They will no longer be called on to supply power to demand peaks under normal circumstances.

          • evfan

            I think this is a fascinating concept that batteries might replace gas peakers, and that is why I hope somebody will publish actual numbers.

          • neroden

            Peakers are *dead*. “Deep backup generators” will still be around. But for a deep backup generator it doesn’t actually make sense to use the inefficient simple-cycle technology currently used for peakers (which is optimized for fast startup and shutdown); use one of the mothballed gas plants with higher efficiency (of which there will be quite a lot).

    • Matt

      The electric grid is more complex than that. There are plants running in stand by mode everyday, all day long. If not then when a coal plant fails the whole grid would go down. Since you have to match power to load, this use can be greatly reduced with batteries. Even if you only use it so that you always run the gas plant at best efficiency but for a shorter time. Then there are plants that only run a small amount each year. So there is a lot of space on the grid for fast response, before you get to the point of filling that space of plants that run 2 weeks a year.

    • Ronald Brakels

      Battery storage (or other forms of energy storage) can potentially eliminate the need for peak generation in the following way:

      Grid operators predict demand for the next day and say something like, “At about 6:00 pm peak generators will need to come online – or they would if we didn’t have energy storage.” Then instead of switching off the load following combined cycle natural gas plant that normally shuts down once demand falls late in the evening, they leave it operating and use the electricity generated to charge the energy storage. Then at around 6:00 am the next day they use the stored energy to meet demand instead of switching on peak generators.

      This has the advantage of not having to pay for peak generators that spend most of their time doing nothing and may only operate for 40 hours a year. And it has the advantage that natural gas is burned more efficiently, resulting in less emissions. Of course the disadvantage is that a large amount of energy storage will be required.

      Places with sufficient hydroelectric capacity use it in this way. We try to use our hydro power and pumped storage this way in Australia, but it’s a bit dry here and the grid is very “peaky” so we still have plenty of natural gas, diesel, and kerosene peak generation capacity. Currently we may have more than we need thanks to the growth of rooftop solar.

  • omar

    Good step to later replace the hall station by solar and wind

Back to Top ↑