The most important chart in energy


There’s a chart you’ve never heard of that perfectly represents how inefficient the power grid is and how much we need more distributed energy solutions. Agencies responsible for the reliability of the grid, including the California System Independent System Operator (CAISO), use what are called Load Duration Curves, or LDCs, to plan investments and monitor efficiency. The area under the LDC (see image below) represents the energy demanded by the system and the curve illustrates the relationship between energy use and generating capacity needs.

Screen Shot 2013-02-05 at 6.09.08 PM

In order to meet California’s peak load in 2011, CAISO had to secure the equivalent energy of 20 nuclear power plants. This level of energy represents what is needed to serve peak capacity rather than average capacity. On most days, California’s electricity demand ranges between about 23,000 MW and 36,000 MW. In 2011, California demand exceeded 40,000 MW for only 0.8 percent of the year, or about 70 hours total (for reference, California hit an all-time for peak energy usage, reaching a total load of 50,270 MW in July, 2006).

Without transforming the power system to include energy storage and other distributed energy technologies, we have to build and maintain enough power plants to meet peak demand in real-time. Utilities will have to build additional generation assets, leading to increased generation investment while using this capacity for very few hours of the year.

stem-PowerStoreWe clearly have a huge opportunity to improve our current electrical power infrastructure since most of our generation capacity is only required a brief percentage of the year.  As the LDC illustrates, satisfying our peak load creates a massive discrepancy between our old generation capacity and asset utilization.  According to the Energy Information Administration, our national capacity factor, or the amount we actually generate vs. the amount we are capable of generating, is at or less than 40 percent. Most hours of the year, a massive fleet of generators sits cold and idle, not providing power to customers or earning revenue for their owners. Even idle, these plants require upgrades and maintenance to stay current.

With the problem of low asset utilization, the more optimal solution may lie on a micro- and not a macro- scale.  Many influential organizations, as varied as the Rocky Mountain Institute and the United States Military, are realizing the promise of smaller scale distributed energy generation and electricity storage as a way to combat the exact discrepancies that the LDC highlights.

This move towards supplementing our power with decentralized capacity has the added benefit of circumventing many of the negative economic and environmental impacts associated with our aging grid. Relying on centralized power systems alone to meet peak load is costly. Power generated from peaking-units (which currently make up around 14 percent of America’s 2,600 total power plants), can cost upwards of $100 more than the typical megawatt-hour.

StemAdditionally, the cost of transmitting energy across an aging grid is increasing and inefficient as well (according to the Department of Energy, around 75 percent of transmission lines and transformers are older than 25 years or older and roughly 60 percent of circuits have been operating for over 30 years, and transmission losses can range from 7-10 percent).  William Pentland of ClearEdge Power points out that “the typical electric utility customer in New York City is charged more for the delivery of an electron than the generation of electron”.

Recent advancements in distributed energy storage technology — including the adoption of lithium-ion batteries similar to those used in electric vehicles as onsite storage — are allowing business owners to dispatch power from the grid at those times that it is most cost-effective to do so.  (Many large automakers, including GM, Chevy and Nissan, are catching on to this potential by developing grid storage units with their used EV batteries.) At Stem we use energy storage systems to provide on-site electricity to customers at times when energy from the grid is more expensive, such as during hot summer days.Stem

It is important to note that even if the cost of energy drops, the LDC problem persists. While many in the energy community are excited by the shale gas revolution, without storage, generators will still need to be sized to meet peak demand and therefore will continue to operate at 40 percent or less capacity factor. Electric utilities and their rate base will still have to pay to build assets that sit idle for most of the year regardless of the mix of energy sources in the future.

Due to their ability to mitigate peak energy usage where it occurs, distributed energy storage systems have the potential to change the way we draw power at peak times. This could, in effect, transform the LDC into a much more efficient rectangle, flattening the load and avoiding generation overbuild.  Distributed energy storage that is implemented as the smart solution will prove that grid stability and grid efficiency no longer have to be treated as mutually exclusive.

Salim Khan is the CEO of Stem, an energy technology company that enables businesses to control their electricity expenses and helps the electrical grid to be more efficient in managing peak usage.



> Pour Vincent Bolloré,
> Quand vont’ils se réveiller: Créons des emplois: les voitures de 5 ans
> et plus transformons-les en voitures électriques. Pour l’autonomie
> développons une nouvelle technologie qui va rendre les voitures autonomes
> de même que les maisons autonomes voir reportage:
> BONJOUR a toutes et tous dans ( NEXUS) DU MOIS DE SEPTEMBRE voir le
> reportage d’un chargeur autonome qui puise l’énergie du vide pour charger
> les batteries voir aussi un autre reportage
> Clément Gauthier Abitibi Canada
> 819-735-4331


I think we should all first thank the author for writing this article and bringing the subject up, as stated by Mr. Gupta, outside of PUC/PSC hearing rooms. The article paints an interesting, and yet as stated by Mr. Hails, incomplete picture. However, let’s not get t0o critical too fast. The LDC presented is VERY generic and honestly, if you don’t know how that curve is created and what those horizontal lines mean (some are left off too) then the graph is essentially meaningless to you. But, the point he is making is not incorrect – that the establishment of “micro-grid” style DG is going to be a major answer to the future energy needs of the US and many other countries. When the grid was being developed here in the US, circumstances were very different than they are now, and today there are different concerns, different resource shortages, newer technologies, etc. I agree with Mr. Hails that the traditional baseload generation units cannot just be ignored while we “look” for other solutions. That problem needs to be confronted now – both the construction of new units, training of qualified personnel, etc. However, the argument that DG, storage, renewables, smart grid, etc…is not going to be the answer as stated by many very learned power engineers (my hand just went up too) is simply incorrect. Mr. Hails, the existing and antiquated paradigm that defines the US electrical power infrastructure is in desperate need of change. Let’s discuss the host of problems from enormous cost of emissions control that is now required on traditional fossil facilities, the unnaturally low cost of natural gas (highly unlikely this is sustainable far into the future), the utter disaster that is waiting to happen in the US transmission system and the old distribution networks across the entire US. There is no national HVDC transmission in place yet, mostly due to a complete lack of national policy/regulation to make it possible. To talk about blackouts, I believe that new modern devices, namely synchrophasors, (coupled with some other pieces of technology and newer software) are going to be the key to eliminating that problem – but only if we solve the generation-load problem. Management of the grid is crucial, but that’s like making sure you change the oil every 3000 miles in an old beat up Buick – it’s only going to do so much before you need a new engine. I’m bringing up a myriad of different issues basically to point out this major fact: there isn’t now, never was and will never be one single silver bullet answer. I think the closest thing we have to an all-encompassing answer is the CONCEPT embodied by micro-grids and DG with storage. This concept itself is seeking to drive a whole new way to approach the grid, electric power generation and storage in the US. This is what is needed. Yes, we need to solve the immediate problem with the aging plants and workforce, but only as a stop gap. We need to examine and commercialize the new concepts as presented in this article. The traditional LDC exists around the traditional electric power industry as we know it today. Why can it not change?
10 years ago never in a million years would we have thought our phones could do what they are doing today. I think this is the attitude we should be taking towards this problem.

D Christian

Here’s the chart that is truly most important to US residents:

The chart cited in this article is only a generic example, and is not itself important.
A similar specific chart of your use in your area of what’s important to you right now.

Pradeep Gupta

Mr. Hails very lucidly explained the rationale behind investments and operations of energy resources which are dictated by the necessity that electricity supply should match the consumers’ demand at each instant. But the story goes a bit further.
The “load duration curve”(LDC) is treated as a given fact generally as in this article, and the resource portfolio, is treated as the only correct solution. However, in fact, LDC is just the sum of individual patterns of electricity use by all of us; residences using electricity mostly in the evenings, offices during the day, no one using much during the wee hours of night etc. But if you look at the situation little differently, where the “load” side could also be modified, perhaps even at a cost lower than the cost of corresponding supply resource, we could end up with a lower costs overall.
This approach is called “Integrated Resource Planning”. The “load” side can be modified by using storage batteries, pre-heating or -cooling commercial buildings, all of us using more efficient appliances or the industries rescheduling their manufacturing shifts or utilizing local energy resources such as solar. But note that each of these measures come at a cost, which could be more or less than the matching “supply” side options.
I am delighted that these issues are being discussed here at the Patch, beyond the hearing rooms of Public Utilities Commission in San Francisco by only engineers and lawyers.
Pradeep Gupta
Council Member- South san Francisco

Jiminy StAck

V2G Vehicle To GRID can provide Distributed store at low cost and pays off for the owners of the Vehicles. Most vehicles site 20+ hours a day doing nothing.
Read to see how this is growing and in use today.

R. L. Hails Sr. P. E.

This is not the whole story. On the grid, the base loaded units generate juice at the cheapest cost. Then as the load increases, less efficient units, and units at a distance are called on to produce. As the load peaks at maximum load, everything is utilized with a thin margin of excess capacity. If the load gets too big, the system will tear itself apart if load shedding, black outs, are not used.

Loads which can tolerate interruption of service pay less, for less reliable juice. It is basic electrical engineering, and utility policy, that the grid is designed to provide ubiquitous, cheap, continuous, safe juice. If everyone has their whole distributed (independent) generator, or storage scheme, these can compete with the expensive last stage generators. But not the base loaded units. America has allowed the base loaded units to decay, most are far beyond retirement age. The men who know their ancient machines are dying out.

Conservation, and new base loaded units are desperately needed, now.


A lot of misleading stuff here. Do you think utilities and their regulators are trying to build the most inefficient system they can? (California excepted). This ill-defined storage is the only thing more expensive than solar cells. you have lots of stuff that only gets used a few hours a year. Is that a problem in and of itself?


Just kind of humorous.. India’s electric generation and distribution system might be the worst in the world.

paul martin

Still staring at this chart and am not convinced that it is that crucial to me personally .. I want to see the one that relates to all the battery problems that seem to be occurring. Boeing might win a few friends if it updated Peukert’s Law …. maybe they did already


Distributed energy storage is only the first step towards better energy management. Without distributed energy generation and optimal utilization of energy, the former creates a very small difference. Creating cost-effective distributed energy storage is one of the biggest problems facing our society today, along with reckless consumption of energy.

Comments are closed.