This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.

10,140 Members

Post

Energy Storage is Critical Issue at Utility Industry's Conference in New Orleans

energy storage and utilities

Earlier this week, the Edison Electric Institute (EEI) held their Annual Convention, which brought together leaders from the utility industry for three days in New Orleans. With energy executives from around the world in attendance, attention was squarely focused on the dynamic role of utilities and the evolving electrical grid, and how new assets like energy storage and renewable energy are increasingly augmenting grid operations.

On the 10th anniversary of Hurricane Katrina which devastated the Gulf Coast, the conference highlighted the immense importance of the utility companies in ensuring resiliency and reliability as increasingly intense storms challenge our aging infrastructure.

Energy storage was front and center, starting Monday morning with an overview of the Department of Energy’s Quadrennial Energy Review (QER) which Energy Secretary Moniz described as critical investments given the monumental change the U.S. grid is about to undergo as a result of the EPA’s proposed Clean Power Plan.  Energy storage throughout our energy infrastructure is highlighted as one of the key components of the QER, and Moniz indicated he was confident that Congress and the Administration would be supportive of prioritizing these critical recommendations.

The Secretary’s introductory remarks were followed by a fireside chat between Southern California Edison CEO Ted Craver with Elon Musk and JB Straubel from Tesla discussing their entrance into the energy storage industry earlier this year and their vision of the impact electric vehicles and storage technology will have. Peppered with amusing and intriguing anecdotes on his experience driving a culture of innovation in a rapidly growing company, Musk reflected on the unprecedented consumer demand that Tesla saw with the introduction of their new battery storage systems, which drew widespread media attention in May for its simple design and affordability.

Energy Storage was identified as a ‘critical issue’ by event organizers, and later on Monday I moderated a panel with leading energy and technology companies that drew a standing room only crowd and continued the dialogue about the drivers and value proposition of energy storage.  Panelists represented an investor-owned utility, a leading energy storage project developer, and three storage technology companies. The conversation featured expert insights into markets, policy and technology, and the many applications of storage in the energy economy.

With a total of 2 million MWh of energy storage service delivered and more than 1,000 MW’s of planned capacity, AES President of U.S. Strategic Business Unit Ken Zagzebski opened the session highlighting his company’s near decade of experience developing, installing and operating some of the largest energy storage systems in the world. Zagzebski focused his remarks on the system-wide benefits of his company’s energy storage installations, and the ability of storage systems to offset the need for new, often underutilized, ‘peaker’ capacity.

Questions quickly brought the conversation to the topic of resiliency and reliability, and Don Clevenger – SVP of Strategic Planning at Oncor Electric Delivery – discussed why his company has supported the need for more than 5 gigawatts of energy storage to be installed on the Texas electric grid.  Clevenger highlighted values ranging from ancillary services like frequency response to deferral of expensive transmission and distribution investments as reasons that Oncor should invest in energy storage. 

Clevenger noted in particular the benefits of increased system reliability and reduced outages, all while lowering consumers’ electric bills, as the key drivers of Oncor’s support of expanded deployment of distributed energy storage systems.

While all panelists agreed that the storage industry is going to continue its rapid growth, each highlighted current challenges built into regulatory structures and market designs.

With utilities around the world working with EOS Energy Storage, Chairman Steve Hellman highlighted a common thread in global energy markets – the difficulty for companies to capture these multiple value streams and monetize them.  Hellman noted that while energy storage provides a number of different valuable services to the grid, current regulations and markets generally make it impossible to monetize all those opportunities. 

The session closed with participants’ outlooks on the future of energy storage and a discussion of what we will be discussing at the next EEI conference.  Panelists agreed that in a year’s time there will be more experience to draw from, more data to analyze and more progress in creating competitive markets, and that the challenges that exist are far outweighed by the risk of inaction.

Ten years after Hurricane Katrina nearly crippled Gulf Coast states and hundreds of utilities banded together for weeks to bring the area out of darkness, at EEI’s annual event creating a more resilient and responsive energy grid is still a critical priority for the utility industry.

Photo Credit: Energy Storage/shutterstock

Matt Roberts's picture

Thank Matt for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.

Discussions

Bob Meinetz's picture
Bob Meinetz on June 12, 2015

Matt, storage is perceived as an environmental wundertechnologie because it can make wind and solar energy viable in “off hours”.

This perception is incorrect. In nearly all cases where grid-scale storage is being applied today, it permits utilities to deliver off-peak fossil generation at peak times, reducing the need to construct expensive new peaker plants. However, 5-10% of all generation stored is lost to inefficiencies during charging/discharging. Thus stored energy is 5-10% dirtier than the same energy delivered directly to the grid.

Occasionally, storage permits excess renewable energy to be used at a later time, displacing fossil energy. Though in those instances it undeniably creates a cleaner power mix, excess renewable energy is an exceedingly rare occurrence. In California, about 2% of renewable energy is deemed excess and dumped to maintain grid stability. Wind and solar make up 8% of the state’s total generation – so even if excess clean energy was stored it would displace less than two-tenths of one percent of fossil generation. Those savings are dwarfed by losses from storing fossil energy.

Storage saves money for utilities and their customers, but those savings come at the expense of the environment.

Matt Roberts's picture
Matt Roberts on June 13, 2015

Hi Bob – I think that might be an oversimplification of the benefits of a fast-responding, accurate asset on the grid. The underlying principle of why storage is beneficial is that it enables broader system optimization, not the round trip efficiency of any one charge.

Modulating large, slower-responding generating assets (like a natural gas plant) to perform ancillary services is inherently quite inefficient. These either need to be on and idling (burning fuel but not delivering electricity, or even running efficiently), or modulated at the turbine and jumping between different levels of efficiency. Also, taking minutes to respond to a grid signal means that the plant can be following a signal in the wrong direction, wasting energy and needing further correction.

By using storage systems, we can address two big inefficiencies in the system that will help reduce wasted energy. With storage allowing grid operators to more efficiently use thermal plants and deploy the ESS for grid services, we can have those fossil plants at peak efficiency more of the time. And, by avoiding building new combined cycle peaking plants and instead using storage to address those handful of hours each year, we can leverage clean energy or optimized fossil energy to address those rarer seasonal needs.

It does, as you noted, save utilities and consumers money, and part of that savings can be seen in reduced demand for frequency response on the PJM Interconnection following the introduction of pay-for-performance, as one example. Less energy bought for FR service because of the speed and accuracy of storage means less fossil fuel burned, and the energy that is burned is done so more efficiently.

Storage is a multifaceted asset on the grid, and the value exceeds any one application – and while it does lose power in its round trip, it enables us to use energy more efficiently, helps level out variable renewables and maximize their output, and defers unnecessary investments in new transmission equipment and power plants. 

The grid is a large interconnected system and it is aging quickly and not very nimble, and looking at any one node or one energy transaction does not show it’s wider impact and benefits. Storage is among the solutions (including efficiency, DR and renewables) we need to deploy to to enable the next generation, efficient electric grid.

Bruce McFarling's picture
Bruce McFarling on June 13, 2015

Matt, storage is perceived as an environmental wundertechnologie because it can make wind and solar energy viable in “off hours”.

This perception is incorrect. In nearly all cases where grid-scale storage is being applied today, it permits …”

Note that there is not actually a contradiction here … what a capability can support does not have to be the same as what a capability is being used for today. Technologies that have an ability to make an economic case under status quo conditions while also helping create a grid that can more easily integrate larger shares of variable renewables are excellent candidates for paving the way for growing shares of variable renewables.

In nearly all cases where grid-scale storage is being applied today, it permits utilities to deliver off-peak fossil generation at peak times, reducing the need to construct expensive new peaker plants.

However, 5-10% of all generation stored is lost to inefficiencies during charging/discharging. Thus stored energy is 5-10% dirtier than the same energy delivered directly to the grid.”

Except capacity installed as pure peaker capacity does not necesarily have the same energy efficiency as baseload or load-following plants … replacing the output of a single cycle NG turbine with an efficiency of 35%-42% with the output of a 55% efficient advanced NGCC generator at a cost of 5%-10% storage losses is a shift from 35%-42% efficiency to 49%-52% efficiency.

This general discussion conflates a range of target storage frequencies. For integrating variable renewables, some of the higher frequency storage is not about conventional peak demand as much as about peak ramping demand, where generation that is load-following generation can use the predictability of much of the variation in variable renewables to be ramped up in advance of an coming steap upward ramp in net-load.

At 90% storage efficiency and equivalent charge/discharge rates and with a single load-following source that can increase output in advance of the increase in load, the trough to peak increase when going from charging though storage to discharging is 190% of the charge rate. Mix two schedulable sources with different lead start-up times, and the start-up times can be scheduled for an even larger swing relative to charge rate.

 

Joe Deely's picture
Joe Deely on June 13, 2015

Bob,

You seem to always be looking in the rear-view mirror. You need to look ahead a little more so you don’t crash so often. 

  •  “In California, about 2% of renewable energy is deemed excess and dumped to maintain grid stability. Wind and solar make up 8% of the state’s total generation – so even if excess clean energy was stored it would displace less than two-tenths of one percent of fossil generation”

As per usual you are wrong with the CA data. Using EIA CA generation data for 2014 you can see that CA had 197,705 GWh of electricity generation and 23,667 GWh was from wind(13,776) and solar(9,891) – 12.0%. In 2015 wind and solar should be over 15% of CA in-state production.

Also, if we look out a few years we will start seeing many days where there will be “excess” solar. As an example look at this recent day(May 24th) on CAISO charts below  If you look at the Average Hourly graph you will see that thermal (in state NG generation) has been squeezed down to 3GW of production. Take into account the 5GW of solar that will be added onto CAISO over the next few years and you can see that not ony will the in-state thermal generation be squeezed down to nothing but the imports will be squeezed as well.  One last point on this – this year  is very low for Hydro. So throw in another 1-2GW of Hydro for a normal year. 

FInally, note Bruce’s point on efficiencly of Combined Cycle versus simple peakers. Storage will most likely eliminate simple peakers.  Storage will enable energy from fossil fuels to be cleaner rather than dirtier as you say. 


http://content.caiso.com/green/renewrpt/20150524_DailyRenewablesWatch.pdf


 

Bob Meinetz's picture
Bob Meinetz on June 13, 2015

Joe, again you employ the same tired fallacies on which renewables advocacy depends, namely “proportion perception deficit” and “trust me, the future will prove me right”.

2% of 12% equates to less than one-fourth of one percent of California’s grid mix which storage, applied only to available renewable energy, would help replace – if it were available. What do you find more encouraging, an infinitesimal or a slightly-less-than-infinitesimal measure of pixie dust?

As I’ve said before, I’m not the least bit interested in what you think “should be” California’s 2015 generation, or if or what you see when you “look out a few years”. I’ve been hearing heady projections since 1980 about wind and solar which have consistently fallen flat on their face. What would you think, if you were in my position?

Bob Meinetz's picture
Bob Meinetz on June 13, 2015

Matt, to be entirely accurate storage does not allow us to use energy more efficiently, as you state (energy is always used most efficiently applied directly to the application for which we need it). Storage may permit us to generate energy more efficiently – but do you have any evidence to show gains in efficiency outweigh the resistance losses it incurs?

Bob Meinetz's picture
Bob Meinetz on June 13, 2015

Bruce, because load is distributed among millions of users, grids never experience a “steap upward ramp in net-load”, so the value of storage’s ability to ramp up quickly to a grid is an invented one (see CAISO’s current load here, which is ramping up right now at 1GW/hour – well within CAISO’s distributed utility generation capacity).

Storage does help smooth the output of variable renewable generation, which shows significant and unpredictable ramps up and down. But what variable renewables need is the ability to time shift gigawatthours of generation to times during a day or week which don’t coincide with its generation at all. Renewables advocacy deliberately blurs the distinction between these two applications of storage, one which is feasible, one which is not remotely so. One does not easily “pave the way” for the other except in fantasy.

Matt Roberts's picture
Matt Roberts on June 13, 2015

Hi Bob- using a system that can accurately follow a grid signal with the exact amount of energy you need in an instant is more efficient.  See PJM’s reduction in frequency response as just one example, as I noted above. With a few dozen systems installed on their grid, there are already major reductions and more on the way. 

Matt Roberts's picture
Matt Roberts on June 13, 2015

Bob, ramping is an important issue and is already having negative impacts on the grid in places like HI, CA, and others.  When large amounts of solar drop off each day, inconveniently at the same time as the system peak, we need a lot of generation very quickly.  Storage and demand response are critical solutions to this challenge, and with more renewables on the way, it is just going to become a bigger hurdle. Do am image search for the ‘duck curve’ in California , which while just one example, it is indicative of what many grid operators see on the horizon for their systems. 

Bob Meinetz's picture
Bob Meinetz on June 13, 2015

Matt, I don’t doubt storage is useful for frequency regulation purposes. These systems are tiny (3MW) compared to the output of peaker plants (300MW or more) and thus are incapable of avoiding any significant quantity of peaker generation.

Matt Roberts's picture
Matt Roberts on June 13, 2015

Once again Bob, I hate to do this, but you are not using any data and just guessing at numbers.  Storage systems come in all shapes and sizes – from a handful of kilowatts to hundreds of megawatts.

To replace a peaker, you are not competing with its nameplate capacity – mostly because they are primarily only used about 10-15% capacity factor – you are replacing the application it was performing. By those metrics you could offset the need for a 300 MW facility with a 30-40MW of storage, and avoid the brownfield site, inefficient operation, and wasted investment that goes along with it.

I hope you have found this conversation informative and that you will take the time to learn more about storage systems and their applications. While storage is not the silver bullet it is some times held up as in the media that will solve all our problems, it is a flexible resource that can do many things very quickly and efficiently – and it’s value is dynamic, and evolves with the grid.

It’s more of a Swiss Army knife, and when you know what all the doodads on your handy tool can do, you will find it very useful for handling many, but not all, jobs.

Joe Deely's picture
Joe Deely on June 13, 2015

Bob,

1) You might want to read up on what “marginal” means. In the scenario I showed the graphs for – once we add an additional 3-6GW of solar capacity  in CA the “excess” solar moves rapidly from the 2% you mentioned towards 100%. So potentially, each additonal GWh of solar generation added from that point on could result in a GWH moved to storage and a GWh of avoided fossil generation later in the day. It really is simple math.

2) Your comment: 

“I’ve been hearing heady projections since 1980 about wind and solar which have consistently fallen flat on their face.What would you think, if you were in my position?”

Again with that rear-view mirror. See data below. The trends are pretty simple. You can also see what is being built now and the capacity increases trhrough 2017 are already mapped out. No major prognostication here.

Solar in CA from EIA

  • 2001- 2009     0.3% of generation
  • 2010-2011      0.4% of generation
  • 2012              0.7% of generation
  • 2013              1.9% of generation
  • 2014              5.0% of generation

2014Q1 – Solar @3.5%    2015Q1 – Solar @7.0%


3) By the way Bob, here is some more heady reading for you from a recent CAISO proposal. – 

  • “Distributed energy resources – i.e., resources on the customer side or the distribution grid side of the electric system, such as rooftop solar, energy storage, plug-in electric vehicles, and demand response – are growing and will represent an increasingly important part of the future resource mix. Integrating these new, distributed resources will help lower carbon emissions and can offer operational benefits. The ISO is therefore working to facilitate their participation in wholesale markets, consistent with reliable system operations. “

You will be hearing a lot more about DERPs in CA over the next few years. 

If we had this in place 2-3 years ago there would be no discussion on the need for a new NG plant in Carlsbad.

 

Nathan – I think you in particular will like this concept.

Bob Meinetz's picture
Bob Meinetz on June 13, 2015

Matt, if by “system peak” you mean peak load, yesterday on the island of Hawaii that occurred at 7:30PM when solar was generating no energy at all.

Storage could, in theory, shift the 30MW of maximum solar generation on the island forward 6-1/2 hours until peak load. There are two problems with that scenario:

  • It would require roughly 165MWh of storage, $41 million at AES’s current price, and would be worse from an environmental perspective (it’s doing more good replacing Hawaiian coal generation earlier in the day without resistance losses).
  • Almost of all the island’s solar is distributed, so when you store Hawaii’s grid mix you’re storing even more coal generation and making it 5-10% dirtier.

Ideally – if you had 160MW of utility solar generation, at a cost of ~$1 billion, you could shift roughly 1.32GWh of daily solar generation forward 6-1/2 hours and replace about half of Hawaii’s coal generation. Storage would cost $330 million, and for those cloudy days you’d still have to maintain all of Hawaii’s coal plants, because the poor island wouldn’t be able to afford any CCGT after spending all that money on solar and storage.

That’s only one island in the state, and from a solar perspective, one of the best possible scenarios.

Matt Roberts's picture
Matt Roberts on June 13, 2015

You are conflating multiple concepts into one conversation Bob, it’s difficult to address your questions when you are jumping around so much.  We were just discussing ramping, now we are back to peak load – one is about binary on/off while the other is about the delta and dynamic change.

I would be happy to help you find some resources so we could have a productive conversation. Your last comment is still binary thinking and doesn’t reflect dynamic change. Time shifting solar and arbitrage is not a primary driver for storage, it’s not about 24/7 solar power.  

Bob Meinetz's picture
Bob Meinetz on June 13, 2015

Matt, I think you’re the one who’s guessing at numbers. Here’s PJM’s total energy storage commitment:

Battery storage – A one-megawatt array of lithium-ion batteries provided regulation service inhe PJM market for several years. The battery facility, housed in a trailer on the PJM campus, was owned by AES Energy Storage, a subsidiary of The AES Corp. AES now has a two-MW battery facility on the PJM campus. A much larger battery facility, 64-MW AES Laurel Mountain in West Virginia, went into operation in 2011 in conjunction with a 98-MW wind farm. The battery facility helps PJM quickly balance variations in load to regulate frequency as an alternative to adjusting the output of fossil-fuel generators; it is capable of changing its output in less than one second. In response to PJM requests to balance the grid, the battery unit can supply power into the grid by discharging its batteries or store excess electricity from the grid to charge its batteries.

No “hundreds of megawatts” anywhere to be found, and the 64-MW facility would be unnecessary if it weren’t for the intermittency of wind.

It would be helpful if you provided some links in your article so readers didn’t have to hunt for data themselves. Though I’m sure storage is useful for regulating voltage and the output of renewable energy facilities, in terms of replacing peakers, In my opinion, it’s wildly oversold.

Bob Meinetz's picture
Bob Meinetz on June 13, 2015

Joe, rearview mirrors are, in general, more dependable than crystal balls. With “once we add an additional 3-6GW” and “you will be hearing a lot more about DERPs” you’re wildly hypothesizing again.

I think the environment deserves a bit more consideration than fortune-telling. Are you just making this stuff up?

 

Bob Meinetz's picture
Bob Meinetz on June 13, 2015

Fair enough, Matt – here’s one concept. You write:

When large amounts of solar drop off each day, inconveniently at the same time as the system peak, we need a lot of generation very quickly.

As you can see, Hawaiian solar is generating no energy whatsoever at system peak. Can you clarify?

Nathan Wilson's picture
Nathan Wilson on June 13, 2015

Good points Bruce.  I would also add that the storage projects that utilities are installing today are much different than 30 years ago.  Back then, we built pumped hydro storage with 8 hours of endurance for actual load shifting of demand peaks (because fossil gas was expensive and combustion turbines were inefficient).  This is talked about a lot today, but what utilties are actually buying today is more often battery systems with 15-60 minutes of storage; which is enough to allow a slow-throttling steam or combined cycle plant to serve peak demands and provide spinning reserves.

So thus far, fossil fuel is in no danger of being replaced by renewables with storage; renewables with fossil balancing is much cheaper.

Nathan Wilson's picture
Nathan Wilson on June 14, 2015

Bob, I think you’re over-stating the role of coal in Hawaiian electrical system.  As you’ve noted, Hawaii is about the best location in the US for renewables; not because their resources are so great, but because most of their electricity comes from very expensive oil, and the total grid demand is too small for conventional nuclear.

As noted in this NREL report, on the island of Oahu (the most populous island, which contains the city of Honalulu) has a total firm generation of just 1,737 MWatts, only 11% of which is from coal.

The AWEA reports that as of 2014, Hawaii had 206 MW of windpower and Hawaii got 5.9% of its electricity from wind.

It strikes me as very likely that over the next decade, it will be possible for Hawaii to economically replace half or so of their oil-fired electricity with solar and wind, but they’ll need a lot of storage for grid stabilization and load shifting at that high penetration (the NREL report used storage for stabilization only, and only reach around 23% penetration!).  Geothermal would probably be cheaper for them, as batteries would not be required, and they may end up building undersea cables to connect the islands anyway, but there is strong opposition to geothermal there.  However, they are also considering building a Liquid Natural Gas terminal, which would likely undercut the cost of batteries and put a cap on renewable penetration.

So regardless of how we feel about renewables, solar+batteries can at least buy more time (for SMR nuclear and nuclear and/or geothermal public acceptance to develope further).  Once the LNG terminal goes in, it will likely dominate for the next half century.

Bruce McFarling's picture
Bruce McFarling on June 14, 2015

“Bruce, because load is distributed among millions of users, grids never experience a “steap upward ramp in net-load”, so the value of storage’s ability to ramp up quickly to a grid is an invented one (see CAISO’s current load here, which is ramping up right now at 1GW/hour – well within CAISO’s distributed utility generation capacity).”

I’m surprised that you are unaware of the CAISO “duck chart” diagram showing typical early spring net-load (load net of supply of variable renewables) with increasingly steap ramping at solar penetrations predicted for 2015-2020.

As that summary fact sheet explains, “Overgeneration happens when more electricity is supplied than is
needed to satisfy real-time electricity requirements. The ISO experiences overgeneration in two main operating
conditions. The first occurs as the ISO prepares to meet the upcoming upward ramps that occur in the
morning and in the late afternoon. The existing fleet includes many long-start resources that need time to come on line before they can support upcoming ramps. Therefore, they must produce at some minimum power output levels in times when this electricity is not needed. The second occurs when output from any non-dispatchable/must-take resource further increases supply in times of low electricity need, typically in the nighttime hours. Historically, this condition was most likely to occur in the early morning hours when low demand combines with electricity and generation brought on line to prepare for the morning ramp. The duck curve in Figure 2 shows that overgeneration is expected to occur during the middle of the day as well.”

Its also surprising that you point to a total load chart in the summertime to response to an issue that CAISO has presented as a net-load issue that is less serious in the summertime than in the spring.

“Storage does help smooth the output of variable renewable generation, which shows significant and unpredictable ramps up and down.”

Storage increases the energy efficiency of smoothing variability in net-load whether it is predicted or unpredicted.

“But what variable renewables need is the ability to time shift gigawatthours of generation to times during a day or week which don’t coincide with its generation at all.”

In the present stage of the roll-out of renewables, no, variable renewables does not need that to increase its share of power generation from ~5% to 20%-25%. But even before those levels of penetration, seasonal periods of rapid ramps of net-load have been projected to lead to energy losses with the current generation fleet. Storage reduces those energy losses in two ways … first, by time shifting the early over-generation into the later part of the ramp, and second, by allowing a later scheduling of some of the slower responding generation.

How much “variable renewables” needs to shift to expand beyond that, as part of a high renewables generation share, depends upon the characteristics of the net-load. That discussion has a substantial speculative component, since it is a decade or more in the future, but trying to shift all of it with storage, without any dispatchable generation and by ruling out an economically optimal degree of excess variable renewables supply is not be a cost-optimizing system in any of the modeling exercises that I have seen, even if the dispatchable generation is constrained to dispatchable renewables.

Bob Meinetz's picture
Bob Meinetz on June 14, 2015

Nathan, thanks for that clarification. I agree that Hawaii is a unique example of a grid suited for replacing fossil energy with renewable generation, however it’s conveniently used as an example for California, or Texas, or Vermont. Using a tropical resort island as a template for industrial economies would be disastrous – although there are some similarities between island superstitions and objections of the antinuclear movement.

Bruce McFarling's picture
Bruce McFarling on June 14, 2015

“So thus far, fossil fuel is in no danger of being replaced by renewables with storage; renewables with fossil balancing is much cheaper.”

Not necessarily economically cheaper, though of course commercial decisions are not based on fuill economic costs and benefits, but rather on commercial costs and benefits, and it certainly is commercially cheaper at present, when including being allowed to use a scarce CO2 cycle sequestering resource for little or no pay.

That kind of 15-60 minute storage seems highly likely to remain an important capability no matter what the optimal balance of storage, dispatchable supplies, dispatchable demands and surplus generation capacity of variable reneweables may be. The quickest responding among dispatchable renewables, reservoir hydro, has constrained total energy resource. The efficiency benefits of that kind of storage with dispatchable supplies is the source of the current business case, and would be retained by dispatchable renewables, or nuclear operated in a partial load-following manner. The largest potential for dispatchable demand is in decentralized dispatchable demands, which may well require 15-60minute lead times for full response. And while surplus variable renewables generation capacity reduces total net load, it seems like it would more shift the timing of the need for storage at the 15-60 minute time scale than reduce the amount of storage required at that time scale.

Bruce McFarling's picture
Bruce McFarling on June 14, 2015

“Joe, rearview mirrors are, in general, more dependable than crystal balls.”

Neither are a dependable guide to future road conditions … assuming you have a choice between one or the other and no other options is a false dichotomy.

Bruce McFarling's picture
Bruce McFarling on June 14, 2015

What is there to clarify? As of today, the chart for the day before yesterday show precisely the effect he referred to, with the decline from 3:30pm to 6pm coinciding with an increase toward the peak, and so as you can see, the steepest ramp in net-load is between ~3:45-4:30pm.

Or that in this context, “each day” is a claim regarding average daily loads and not a claim that the system peak occurs at the same time each and every day? But that clarification would be unecessary, as surely only the most strained effort at “gotcha” commenting would try to pretend to have read it as intending to say the latter.

Joe Deely's picture
Joe Deely on June 14, 2015

Your comment: “With “once we add an additional 3-6GW” and “you will be hearing a lot more about DERPs” you’re wildly hypothesizing again.”

Wildly hypothesizing?? 

CA now has 9GB of solar installed.  3.5GB was installed in 2014 and 700MW in Q1 2015. Developers will be scrambling to beat the expiration of ITC at end of 2016.   So 3-6GW sounds like a really tough stretch. Right?

http://www.seia.org/research-resources/solar-market-insight-report-2015-q1

No crystal ball needed here.

Bob Meinetz's picture
Bob Meinetz on June 14, 2015

Bruce, I’m well-aware of the “duck chart”. What you describe is not a function of load but of renewable energy’s inability to meet it. We can either fret about fixing problems of variability within a renewables-saturated grid, or attribute those problems where they accurately belong – to the sources of energy themselves – and reconsider whether there aren’t better and more economical ways of meeting our clean energy obligations.

Joe Deely's picture
Joe Deely on June 14, 2015

Interesting Bob. I do what I assume is the same search as you ” pjm energy storage” and I get the same historical data that you reference in your link – the PJM Elelectricty Storage page. Pretty much says what you have quoted and its dated at the bottom 03/31/2015 – a date in the recent past.

However, I take two minutes to look a little further at other links that come up in search results  and I find the following. Just curious – did you miss these? Looking in that rear-view mirror again? 

1) http://www.solarserver.com/solar-magazine/solar-news/current/2015/kw21/i...

  • “On May 14th, 2015, clean energy company Invenergy LLC (Chicago, Illinois, U.S.) announced the start of commercial operations of its 31.5 MW Grand Ridge energy storage project in La Salle County, Illinois.”

2) from same article

  • “Invenergy also will bring online this year the 31.5 MW Beech Ridge energy storage project in West Virginia, the company announced. Both Beech Ridge energy storage and Grand Ridge energy storage are utilizing BYD America’s Containerized Energy Storage System.”

3)  http://www.utilitydive.com/news/nec-energy-solutions-plans-60mw-of-stora...

  • “The PJM Interconnection will add a 60 MW networked energy storage battery system built by NEC Energy Solutions (NEC), in partnership with Amergin Energy and Akula Energy Ventures. The system, designed for fast-response frequency regulation, is expected to be operational by mid-2016.”
  • “The PJM Interconnection — the largest US wholesale power market — has 100MW of storage resources on line and another 500MW in the interconnection queue, according to PJM senior analyst Scott Baker.”

Seems like PJM has a somewhat higher “commitment” versus what you have laid out.

Bob Meinetz's picture
Bob Meinetz on June 14, 2015

Bruce, perhaps inadvertently you’ve clarified the situation, and that is by highlighting the renewables- and storage-centric concept of “net load”.

You’ll notice corresponding with that sharp drop in “net load” is a sharp increase in solar generation. That happens frequently in Hawaii, when clouds uncover the sun in densely-populated areas, and thousands of PV panels become suddenly useful. In many areas of the world, we can get rid of this annoying phenomenon by ditching those panels in the nearest toxic waste area; in Hawaii, it makes both environmental and economic sense to figure out a way to accommodate them.

“Net load” is a prime example of renewables doublespeak which attempts to shift the shortcomings of inadequate and unruly sources of energy to a system not designed to accept them. Another is your own strained interpretation of “each day, inconveniently at the same time as the system peak” as a reference to “average daily loads”. Whatever.

Bruce McFarling's picture
Bruce McFarling on June 14, 2015

In many areas of the world, we can get rid of this annoying phenomenon by ditching those panels in the nearest toxic waste area; in Hawaii, it makes both environmental and economic sense to figure out a way to accommodate them.”

The reason it makes “economic” sense in Hawaii is that the commercial cost of fossil fueled energy in Hawaii is much closer to the economic cost of fossil fueled energy in the Continental US. Of course, we don’t charge the economic cost of fossil fueled energy on either Hawaii or the mainland, preferring to give fossil fueled power an in-kind subsidy by allowing it to consume a scarce resource for little or nothing … so where the commercial benefit is present in Hawaii, the economic benefit is even greater.

“Net load” is a prime example of renewables doublespeak which attempts to shift the shortcomings of inadequate and unruly sources of energy to a system not designed to accept them.”

We agree that the present system is not designed to accomodate a large supply of variable renewables … we disagree that the first priority of electricity supply is to be convenient to a status quo electricity system that has been developed primarily to accommodate the needs of electricity sources that we can no longer afford to use.

Another is your own strained interpretation of “each day, inconveniently at the same time as the system peak” as a reference to “average daily loads”.”

It seems clear that its average daily loads he is talking about … 

Bruce McFarling's picture
Bruce McFarling on June 15, 2015

“… and reconsider whether there aren’t better and more economical ways of meeting our clean energy obligations.”

By which you mean nuclear.

But the best ways to meet our clean energy needs are ways that we can actually get deployed: the merits of low carbon / no carbon electricity sources that we cannot get deployed are purely hypothetical, except if they are used as a pretext to interfere with the roll-out of low carbon / no carbon electricity sources that we can get deployed. At present, there in the US, variable renewables are succeeding in getting deployed, while new nuclear is struggling.

Here in China, both are being rolled out as fast as practicable, along with new reservoir hydro, but in the face of energy demand growing so rapidly that it is only slowing the growth of coal consumption.

But it would be a mistake to treat storage as “renewables plus storage versus nuclear”, since a high nuclear penetration system has its own version of net-load and its own increase in storage requirements versus a fossil fueled power. Recall how the load-variable nuclear output of France is managed: freshly fueled nuclear power plants are equipped with light control rods that allow the output to be modulated up and down, on a schedule. As the plant runs through its fuel cycle, the effectiveness of output modulation declines, until power plants that have passed a threshold level of fuel use are operated as baseload plants. So the nuclear output varies through the day, but it is not load-following in the traditional sense. France’s hydro capacity, extended by cross-haul with Swiss hydro and German fossil generators, is used to close the load net of nuclear output, supplemented by France’s fossil fueled generation that is largely used on a seasonal basis.

In a grid with a mix of solar PV, wind, that kind of “scheduled cyclical output” nuclear and dispatchable renewables, the nuclear and variable renewables would be cheaper than the dispatchable renewables except for reservoir hydro, which would be supply constrained, so there is still an economic benefit to having one to three hours of storage / dispatchable demands to fit the slower moving supply ramp of the nuclear to the needs of the system as a whole.

And taking out the variable renewables doesn’t really change that, since then the load net of nuclear output is carried entirely by dispatchable renewables{+}, including substantial energy lost to spinning reserves, unless there is storage / dispatchable demands available to shift nuclear output from lower load net of nuclear period to higher load net of nuclear periods.

IMO, there will be a fairly low threshold on the total kWh that can be shifted by dispatchable demands on very short notice, and aggregating dispatchable demands will be on a rising cost both per MWh shifted and declining cost per hour advance notice available. Meanwhile, for long duty-cycle life batteries, the total capacity cost dominates. If that is correct, then that works to place the battery capacity at the rapid response, higher frequency of use side of the task and to place dispatchable demand on the time shifts that are predictable one to three hours in advance.

 {+ Or commercially viable CCS fossil fueled power, if what is presently as common as magical flying carpets actually does make an appearance, but that doesn’t imply a loss of market niche for storage relative to today … given the energy cost of CCS and the likelihood of CCS in practice being lossier than CCS in theory, we would ideally prefer that to be CCS added to the tail end of an advanced NGCC power plant.}

Clayton Handleman's picture
Clayton Handleman on June 14, 2015

“the present system is not designed to accomodate a large supply of variable renewables “

Exactly.  I think that the opportunity is to monetize power in near real time.  This will create a marketplace in which agile generation AND agile loads can negotiate optimal operating conditions.  Monetizing time of use can give loads the opportunity to shift in response to price signals.  This also will create fertile ground for innovation and wealth building as the industry shifts.

Of course, as you have pointed out frequently, another important piece is to monetize the damage from carbon and internalize that cost.  This is not to be done as a punishment but rather as an economic statement of what is so.  Since those borrowing from future generations will not be there to pay for the damages, it is imperative that they (us) pay up front.   

 

Bob Meinetz's picture
Bob Meinetz on June 15, 2015

Bruce, I think that we would agree that the first priority is drastically reducing anthropogenic carbon emissions, and that will only be possible by creating suitable replacements for fossil energy with no carbon emissions at all.

There’s nothing about existing transmission grids which specifically favors fossil generation. They undeniably favor centralized generation, but centralized generation has critical benefits, including safety standards and efficiencies of scale.

Perhaps most importantly, centralized generation offers public control over emissions – entirely appropriate, given emissions are a public liability. There seems to be a common misperception that distributed generation will be inherently cleaner, but we have absolutely no guarantee it will pan out that way. A situation where distributed fossil generation becomes cheaper than utility electricity could have disastrous consequences.


Distributed heat generation in London, 1952. Thousands died.

I harbor no special allegiance to our “status quo electricity system”, but I do have an allegiance to the $800 billion investment the U.S. transmission grid represents. Minus a feasible and affordable replacement, that’s what we’ve got to work with, and there’s no evidence such a replacement will be available soon enough to address climate change.

Clayton Handleman's picture
Clayton Handleman on June 15, 2015

“centralized generation has critical benefits, including safety standards and efficiencies of scale.”

DG has mature safety standards, from fire safety through Underwriters Laboratories through interconnection standards through a rigorous process that included experts from around the country from national labs, the renewable energy industry and utilities.  Wiring of DG has been in the National Electric Code artical 690 for decades.

Regarding scale, DG continues to benefit from economies of production

“I do have an allegiance to the $800 billion investment the U.S. transmission grid represents.”

Recommended changes to the transmission grid are generally in the form of upgrades and improvements, not a ground up rebuild.  Currently our transmission grid is vulnerable to a coordinated attack.  Adding redundancy would make it much more difficult for a small group to orchestrate crippling attacks.  In other words, much of the upgrading is already justified as a defense budget item.

Bob Meinetz's picture
Bob Meinetz on June 15, 2015

Clayton, “wiring of DG” is not covered under Article 690, which only applies to solar PV wiring and labeling. The assumption is apparently that distributed generation will be limited to PV.

For the price of a Tesla Power Wall and a natural gas generator, I can generate my own electricity right now for $.16/kWh. Should the price of storage be halved in coming years, as PV enthusiasts eagerly anticipate, we face a situation where generation of electricity – and resulting emissions – will be outside public purview.

“Regarding scale?” Nice try, but generating electricity efficiently is inversely correlated with economies of production of PV panels, which encourage breaking a task into pieces accomplished more efficiently as a whole. Similarly, you mistakenly equate distributed generation with “redundancy”. Millions of independent generators will not work in spontaneous cooperation to keep a grid functional, but instead exponentially complicate the task of maintaining its stability, and increase its vulnerability to attack.

Bob Meinetz's picture
Bob Meinetz on June 15, 2015

Bruce, agreed that nuclear’s load following constraints currently require a dispatchable source like CCGT as a complement to meet load.

But the cost of battery storage required to modulate a 2GW nuclear plant’s output would be prohibitive. Even though less efficient, wouldn’t it ultimately be more cost effective to store off-peak generation as hydrogen from electrolyzed water?

Bruce McFarling's picture
Bruce McFarling on June 15, 2015

In what possible world would it be either/or?

If you split off the charging cost, the storage cost and the discharge cost, the storage cost per kWh for an electrofuel would be much lower than the storage cost per kWh of battery storage, because the battery is a charge, store, discharge complex and holding power in storage over a longer term is losing charge and discharge opportunities. Meanwhile the storage for an electrofuel is a tank, external to the “charge” and “discharge” cycles and you can buy another tank far more cheaply than you can buy another battery with equivalent capacity.

If you look at the cost of generating the electrofuel as a pure baseload demand, there will be a trade-off of extra generating capacity that can be paid for by not running the electrofuel during peak net-load when electricity has its highest opportunity cost, but cost effective balancing with electrofuel generation will still likely involve using a smaller ceiling amount of energy for a longer percentage of hours in the year and then focusing a large capacity of generation when it has the highest marginal benefit.

And if you have electrofuels used that way, you have the same benefit in having 15 minutes to an hour of battery storage on the grid that you have with NG and coal today … that flexibility allows avoiding the use of the fueled power as spinning reserve, making the (remember, more commercially expensive than today, because they become commercially viable because externalized CO2 emissions costs have been internalized) fueled power more efficient, and allows using more efficient use of combined cycle generation, since even with independent block combined cycle, starting multiple turbine blocks at the same time allows the second cycle to be more efficient than phasing the blocks in one at a time.

Electrofuels would be excellent for seasonal peaks. They are excellent for rough-tuning weekly fluctuations in net-load, whether that is weekly fluctuations in load net of a scheduled-variable output nuclear or weekly fluctuations in expected variable renewables harvest. Batteries are better suited for higher frequency charge/discharge cycles: they can bid to by a flexible reserve load when charging until charged and then a flexible reserve supply when discharging until discharged and then turn around and bid to be a flexible reserve load again.

The fixed cost of batteries weighs more heavily on storage capacity than on charging rate, the fixed cost of electrofuels would weigh more heavily on charging rate than on storage capacity. So while there are costs where one will tend to dominate the oher, its more likely that will mean one or the other tends to dominate in tasks in the middle.

If we could more effectively monetize storage, we’d have more batteries, and the electrofuels would become commercially viable at a lower carbon price as well.

But then, that gets into issues similar to PHS, where the traditional daily baseload-charge, peak discharge economics of PHS ~ which is why we were building PHS when we were building out nuclear power plants ~  does not actually deliver its cheapest peak capacity. In a setting where there are a number of days of multiple charge/discharge cycles and the storage requirement is 3 hours rather than 6 hours, and if reservoir costs are 50%+ of total fixed costs, you can easily have PHS generating capacity at half its current cost per kW … twice as much capacity for the same reservoir and the reservoir cycled 300 times a year rather than 200 times a year would cut the cost per kW generating capacity in half, though the cost per kWh stored would increase. But if we regulate the PHS as a generator, that limits its ability to sell grid services, which means that a large part of the benefit to the grid of increasing its generating capacity and allowing it to cycle between providing spinning load and spinning reserve cannot be effectively monetized. And if we regulate the PHS as a provider of transmission services, that limits its ability to effectively maximize the value of the sale of the energy that it stores.

 

Clayton Handleman's picture
Clayton Handleman on June 15, 2015

 

wiring of DG” is not covered under Article 690, which only applies to solar PV wiring and labeling.

Good sounds like we agree that solar DG is safe.  Which DG are you suggesting is unsafe and why?

 

Bruce McFarling's picture
Bruce McFarling on June 15, 2015

“There’s nothing about existing transmission grids which specifically favors fossil generation. They undeniably favor centralized generation, but centralized generation has critical benefits, including safety standards and efficiencies of scale.”

You’ll note that I said “system”. So, for example, the pricing system for wholesale power generation favors fossil fueled power (over both variable renewables and nuclear, as a matter of fact). And the weakness in our electricity transmission planning system in being able to take regional needs into effect has led to a bias in favor of trading off natural gas transmission for electricity transmission, siting new natural gas peaker plants close to demand, which is a bias in favor of fossil fueled power.

As far as the argument that the transmission system ought to be designed to be biased in favor of centralized generation because centralized generation has economies of scale, I don’t see the economic sense of it. The cost-efficient degree of concentration for those technologies that have economies of scale is the amount that there would be with, among other things, a transmission system that is not biased in favor of either centralization or decentralization. Introducing a bias one way or the other would then lead to cost sub-optimal over-centralization or over-decentralization.

Nathan Wilson's picture
Nathan Wilson on June 16, 2015

it would be a mistake to treat storage as ‘renewables plus storage versus nuclear’, since a high nuclear penetration system has its own version of net-load and its own increase in storage requirements

Bruce, it’s a bigger mistake to ignore the differing storage needs.  With reasonable amounts of accommodations like curtailment, without relying on storage, I would expect a nuclear+hydro electrical system to reach 70-90% non-fossil, whereas wind+water+sun is likely to reach 40-60% non-fossil.  Adding storage, demand-response, or CC&S will help either portfolio, so basically a renewable-rich portfolio likely emits 2-3 times the CO2 and air pollution as a nuclear-rich one, for similar levels of effort.

Also, the load-following performance of 1980s era French reactors should not be considered a limit to what is feasible.   For example because neither fast reactors nor molten-salt fueled reactors suffer from xenon transient, both should offer improved load following.  And both operate in suitable temperature ranges for the use of solar salt for thermal energy storage (which is likely cheaper than batteries).


the best ways to meet our clean energy needs are ways that we can actually get deployed … in the US, variable renewables are succeeding in getting deployed, while new nuclear is struggling.

A false choice; similarly if the car breaks down, we can either start walking or accept our new location.  Often, the best choice is unstate but obvious: fix the car.  The reason nuclear (which remains cost competitive with renewables when system costs are included) is struggling is almost entirely because the environmental groups have joined fossil fuel interests in opposing nuclear (in response to wildly exaggerated fears), rather than joining with the scientists who say we need it (to control pollution and environmental impact at the lowest cost).  The way to fix nuclear’s problem is speak the truth to the environment groups, not to ignore the deception at the root of the problem.

Joe Deely's picture
Joe Deely on June 17, 2015

Some competiton for Tesla… article here.

 

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »