Over the last six years, industry and government have spent billions demonstrating the technical merit of using manufactured energy storage systems for grid-scale applications.
Industry analysts forecast that annual revenues from sales of grid-scale energy storage systems will grow from about $200 million in 2014 to almost $20 billion by 2023.
Most investor perceptions of how grid-scale energy storage will evolve are 180 degrees out of synch with the technical and economic realities of energy storage.
While existing regulations don’t require renewable power producers to pay intermittency abatement costs – the cost of converting unstable current into clean current – it's only a matter of time.
Investors who want to profit from the emerging energy storage mega-trend must learn to distinguish between sensible business strategies and imagination-fueled speculation.
Conceptually, batteries and other manufactured energy storage devices are nothing more than expensive bottles that store electricity, a commodity most of us want to buy as cheaply as possible. While using costly bottles to store electricity for its "energy value" is rarely economic, using stored electricity for its instantaneous "power value" can be a compelling opportunity.
The value of a single energy storage cycle is measured by the "spread" between the cost of the electricity used to charge the system and the sum of the avoided costs and incremental revenues received when the stored energy is used. For example, if you have one megawatt-hour of stored electricity:
- It can be worth about $150 to individual consumers; or
- It can be used to abate the intermittency of up to 25 MW of PV solar; or
- It can be used to stabilize the frequency of up to 100 MW of conventional generation; or
- It can be used to carry a factory or server farm through a brief power interruption.
The hierarchy of value is undeniable. In grid-scale applications, the value of storage arises from the instant dispatchability of the stored power rather than the intrinsic value of the stored power.
The net annual benefit of an energy storage system is equal to the average per cycle spread multiplied by the total number of cycles per year, less operating and maintenance costs.
The total economic value of an energy storage system is the discounted present value of expected net annual benefits over the useful life of the system.
When it comes to the economics of energy storage, the two immutable cardinal rules are:
- Applications with large spreads and multiple value streams usually are more valuable than applications with small spreads and limited value streams; and
- Applications that cycle a system dozens or hundreds of times a day are usually more valuable than applications that only cycle a system once or twice per day.
According to the Energy Storage Association, representative present values of energy storage benefits range from highs of about $1,150 per kW for rapid cycling applications like frequency regulation and short-duration renewables integration to lows of $100 per kW for slow cycling applications like time shifting and price arbitrage. While investor imaginations usually flash on ideas like storing off-peak energy at night and using it the next day or storing solar power during the day and using it at night, both of these applications are very uneconomic in an industrialized society with a reliable grid that doesn't run on fuel oil.
While the economics of energy storage are complex, multifaceted and far from obvious, the technical attributes of energy storage systems are a real can of worms.
Every storage system has a theoretical capacity - the maximum amount of energy that can be stored and recovered during a charge-discharge cycle. It also has a smaller working capacity - the optimal amount of energy that should be stored and recovered during a charge-discharge cycle to maximize performance, economics, system life and safety. Stressing any storage system beyond its working capacity is always a bad idea because it shortens life and increases safety risks.
The most important concept in any discussion of grid-scale energy storage is round-trip cycle duration - how long does it take for a particular system to discharge and recover its working capacity. The reason I focus on round-trip cycle duration is that most storage technologies discharge energy faster than they recharge and since storage economics depend on the number of cycles a system can support in a 24-hour period, the most critical metric is the time required to complete a full charge-discharge cycle.
For low value applications like time shifting and price arbitrage, cycle durations of several hours are acceptable. For medium value applications like power quality and demand charge management, cycle durations of two to four hours are common. For high value applications like frequency regulation and renewables smoothing, cycle durations of a few seconds to 60 minutes are essential. These differences between the cycling requirements of various applications are the primary reason experts agree that a comprehensive energy storage strategy will require a multi-pronged approach that selects the best storage technology for a particular application.
There are no silver bullet solutions in stationary energy storage, only silver buckshot.
The following table presents a simple hierarchy of round-trip cycle durations for some of the principal energy storage technologies that are being developed or commercialized by public companies I follow. Since the goal of every stationary storage application is to optimize economic value, system life and safety, the table highlights the total time required to charge and discharge a system's working capacity, rather than its theoretical capacity. It also distinguishes between lithium-ion power batteries like lithium titanate and lithium iron phosphate, and the less costly lithium-ion energy batteries used in consumer goods and plug-in vehicles.
In general, the storage technologies with the shortest cycle durations are best suited to high-value power applications while the storage technologies with the longest durations are best suited to lower value energy applications.
Over the last month, the big energy storage news has been Hawaiian Electric's plans to install 200 MW of energy storage capacity on the Island of Oahu over the next three years. If you dig into the project specifications included in HECO's request for proposals, you'll learn that the primary uses of the stored electricity include:
- Minute-to-minute frequency regulation to maintain a stable 60Hz current throughout the HECO grid;
- Short duration buffering to avoid power outages from generator failures; and
- Short-duration intermittency abatement for PV solar.
It's a classic example of a grid-scale storage user trying to aggregate several value streams using the same operating asset. To satisfy HECO's requirements, the systems must:
- Provide megawatts of supplemental power within three cycles after a generator fault and full output response within 100 milliseconds;
- Provide 150% of nominal power output for 10 seconds and 100% of nominal power output for 30 minutes;
- Offer a round trip efficiency of 90% or more; and
- Offer a 15-year lifespan.
Based on the HECO requirements, I believe three companies in my table are commercializing suitable technologies for the HECO project: Altair Nanotechnologies (NASDAQ:ALTI), Axion Power (NASDAQ:AXPW) and China's BYD (OTCPK:BYDDF). The flywheel and ultracapacitor solutions are too fast and don't have enough energy for the work. Similarly, the battery technologies with long round-trip cycle durations don't have enough power for the work. While there's no way to know whether these three manufacturers are participating in the RFP process, I have to believe they're all likely bidders. Given the sheer size and scope of the HECO project I think it's highly unlikely that a single bidder will win a sole source contract.
Regardless of the outcome of HECO's current plans, I believe the fundamentals of the business transaction are very instructive for investors who want to position their portfolios for the coming grid-scale storage boom. While many companies like SolarCity (NASDAQ:SCTY) and Tesla Motors (NASDAQ:TSLA) are talking about small storage systems for owners of residential and small commercial PV solar, these proposed products are marginal economic propositions that are only feasible because of current tax and regulatory incentives. When the incentive regime changes to more fairly reflect the problems that arise when small power producers deliver unstable current to the grid, I believe the widely publicized market for residential and small commercial storage systems will wither and die. Given the technological and economic realities of energy storage, it's clear that the lion's share of the future revenue growth will come from large commercial power users and utilities that can aggregate value streams and derive more benefit than consumers and small businesses.
In 2011 Bill Gates took a ton of heat for characterizing rooftop solar as a "cute" solution that could never deal with the bigger issue of climate change and powering the developing world. While Mr. Gates hasn't discussed residential energy storage systems, his recent investments in grid-scale storage technologies that will only be suitable for large commercial power users and utilities leave me convinced that he would include residential energy storage in the cute class.
Disclosure: The author is long AXPW. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: The author served as a director of Axion Power International (AXPW) from 2005 through early 2007 and holds a substantial long position in its common stock.
Editor's Note: This article covers one or more stocks trading at less than $1 per share and/or with less than a $100 million market cap. Please be aware of the risks associated with these stocks.