In late 2014, a virtually unknown company called Alevo announced it was entering the energy storage market with a new inorganic, sulfur-based lithium ion battery technology that it had acquired from the bankrupt German company fortu PowerCell. Alevo entered the U.S. with a big splash, investing over $68 million in the 3.5 million-square-foot former Philip Morris tobacco factory in Victory, North Carolina, outside Charlotte. It also announced that it would hire up to 2,500 workers over three years, with a potential maximum workforce of 6,000 capable of turning out thousands of megawatts of electricity storage products annually. In other words, Tesla would not be the only storage company with a gigafactory.
Unlike many of the major storage manufacturers of the world – the Panasonics and the LG Chems – Alevo wasn’t simply going to sell its product into a competitive market. Instead, the plan was to becom a developer as well, deploying its own technology. This was in contrast with most of the other energy storage developers (RES and AES AES +0.68% are technology agnostic and deploy different storage technologies from multiple companies – AES, for example, tests and certifies various technologies at its own lab). By contrast, Alevo’s stated goal was to package its cells into two-megawatt, one megawatt-hour GridBanks and offer them turnkey into the marketplace.
Things were fairly quiet until Alevo’s announcement in February of last year, that it had signed a ‘joint operational agreement’ with Customized Energy Solutions to provide 200 megawatts (MW) of frequency regulation services (fast responding near-instantaneous release or absorption of energy to stabilize voltage on the power grid). While Customized Energy Solutions has been active in multiple power markets in the U.S. for well over a decade, providing a broad suite of services to many players, the announcement was seen by some as aspirational in that it did not identify specific deployments. After that announcement, Alevo went rather quiet again.
Until March of this year. when the company announced its first two major deployments: an 8 MW, 4 MWh storage system at the municipal utility in Lewes, Delaware, and a 10 MW, 3 MWh project in Georgetown, Texas, to be jointly owned and operated by Ormat Technologies (with the majority ownership going to Ormat – a company known to many for its expertise in geothermal power).
According to Alevo’s press release, the Lewes project is “the first in a series of major commercial deployments the company has scheduled for 2016.” The Lewes project involves sale of ancillary services into the PJM (Mid-Atlantic power pool) while improving power quality and reliability for the local power system (which has both wind and solar on its grid) and reducing customer exposure to peak demand charges set during periods of highest consumption of power on the grid. The system will be dispatched by Customized Energy Solutions.
The Georgetown project – expected to last 20 years – will be located at a substation owned by the local municipality. The utility will receive fast response regulation service (this is the first commercial project to offer this service in Texas) that will help it to better manage its renewable resources.
Alevo was the conference host for the recent Energy Storage Association Conference and Expo in Charlotte, so I was able to sit down with some of its senior officers (Scott Schotter – Chief Marketing and Sustainability Officer, Chris Christiansen –President, and Jeff Gates – VP of Sales) and learn more about their recent activities and designs on the market.
The team commented that its 2009 licensing arrangement with fortu for its core inorganic lithium technology was when the company really got started. Christiansen observed that the stability of the technology was a key attribute that attracted them.
“It’s impossible for the battery to burn. Every lithium ion battery has a connecting salt that has a solvent like paint thinner. But ours has no methane, propane, no chain of hydrogen or carbon atoms.”
The original business model concept was that this technology could be developed for heavy trucks or buses that would be connected to the grid.
“Then we thought, why a bus or a truck? We should just be connected to the grid. Then FERC 755 (the Federal Energy Regulatory Commission order that supported the ability of fast-acting frequency regulation resources such as batteries and flywheels to interact commercially with power grids) came along, and we could see something evolving. But frequency regulation alone was not sufficient to make money.”
At this point, the company made a move that they see as being a critical differentiator: they invested in high-performance computing and analytics capabilities. The company needed to understand all of the different value streams and the drivers of costs on the grid so they could make sense of the entire economic landscape and value drivers for storage.
They looked at utility duration curves — essentially a visual curve depicting the frequency at which various energy prices occur. They evaluated grid operations and maintenance costs, as well as heat rate curves (the efficiency of converting raw fuel into electrons). They created powerful analytics capabilities to understand all of the value streams storage could possibly provide to the grid.
The computer they bought was a monster. According to Schotter, “we thought we needed to do that to be able to do the work to analyze the grid.” So they installed a machine with over 5,000 cores – for parallel processing – with up to 21 terabytes of memory. When fully built out, the machine will be capable of three petaflops (a petaflop is one quadrillion floating operating points per second) – for comparison’s sake, this machine will rank among the 25 most powerful computers currently in the world. According to the team, in addition to the overall power, the SMP (symmetrical multiprocessing systems) architecture allows the system to share all resources (processors, core and memory) and provides the benefits of a large machine at a fraction of the cost of clustered supercomputers).
Gates previously worked for Duke Energy DUK +1.46%, where he was involved in the NoTrees wind and storage integration project in Texas. The company understood the advantages of storage in general but nobody could effectively quantify and support the various value propositions.
“Being on the buy side from Duke, I had seen all the different batteries…We (at Duke) understood the technology and services, but we still could not make the value proposition to regulators in North Carolina as to why storage should be included in the RFP process. If Duke couldn’t figure it out, who could? The analytics guys at Alevo were the first to solve the problem and calculate the benefits in dollars to the ratepayer.”
Schotter indicated that Alevo undertook a number of grid modeling exercises as part of this value calculation. One was a simulation of the South African grid, where two gigawatts (two thousand megawatts) of storage were theoretically deployed on Eskom’s system. Schotter indicated that the results were quite promising.
“By reducing fuel costs, spinning reserves, and operations and maintenance, we calculated savings at just under one billion dollars.”
Alevo also simulated the Western Electric Coordinating Council (the entire U.S. western power grid), an exercise that originally took weeks of computing time. In its current configuration, Alevo’s supercomputer can do this within hours, and when built out to full capacity, the exercise is expected to take just minutes.
Why is this so critical? Among other things, it will help the company to figure out exactly where storage can provide the greatest value. To better illustrate the opportunity, an example may prove instructive:
Assume two 500 MW power plants. One is in operation and the other is awaiting additional demand on the system. Also assume that demand is ramping up, so that at some point it just barely exceeds 500 MW. Without storage, one would have to fire up the second 500 MW facility, while ramping down the first plant significantly, so each would run at well less than full load and at resulting lower efficiencies that are costly to everybody. Storage can help avoid some of that issue of ‘lumpiness’ and allow one to operate the grid at higher efficiencies.
Gates puts it this way:
“My analogy is mileage in a car. You get 18 mpg city and 30 highway. Why not just drive on the highway? Put the plant at 55 with cruise control and let the batteries do the hard work.”
In other words, let the power plants run with less ramping up and down, and let batteries accommodate to – and meet the fluctuations of – the power grid. With the right amount of storage added and subtracted at the right times, you can more optimally run generators. Alevo’s supercomputer is critical in helping the company determine where, when, and what resources to add for maximum economic efficiency. In fact, these insights are so valuable that Alevo now sells the modeling capabilities to other entities who are not competitors.
As far as the batteries themselves, they are built for heavy-duty use and the highest amount of cycling capability in the industry – tested in their lab in Germany at 55,000 cycles, and going from from 100% to 0% charge on every cycle. In the tests, the batteries were fully discharged and recharged every 30 minutes for the first 16,000 cycles, and then every 20 minutes for the remainder. That compares very favorably with your typical lithium ion battery that is designed for approximately 5,000 cycles and 80% discharge. This ability to release power deeply and quickly is important, since much of the value of storage depends on how much energy can be released or absorbed over very short timeframes.
And while the batteries have not yet been independently tested by a third party (55,000 cycles of testing takes about three years), Christiansen stated that the company logged all of the information and will make it available to third parties as needed. Alevo is contracting consulting firm DNV GL to do extensive testing on our battery and its ABT facility, an engagement that will begin shortly.
Meanwhile, Alevo continues to make improvements in the battery chemistry, with the specific goals of increasing the energy density and decreasing ‘fade’ (the heat given up as the battery ages – a common experience for many smart phone users).
Alevo is currently looking at financing projects through third-party leasing and power purchase agreements. Christiansen indicates that third party financial support will be critical in helping Alevo to meet its long-term goals.
“We want to deploy hundreds of gigawatts (a gigawatt is a thousand megawatts), so we need to set up an adequate financial structure. We definitely believe the market will materialize.”
Assuming the projects and financing materialize, Alevo is ready to move quickly. The company currently has one production line that is fairly well automated and capable of producing 600 MW and 300 MWh annually for both the utility-scale projects as well as smaller commercial and industrial customer installations.
So the building blocks appear to be in place. The big question now is whether or not there is any technology risk to their relatively new battery chemistry, and whether the market wants what Alevo has in sufficient quantities to make this a profitable business. The company thinks it has what it takes to win in this new and growing industry and that it can compete with the likes of LG Chem , and Tesla.