Grids are increasingly looking to storage as a critical asset for balancing load and renewables such as solar and wind.

Largely motivated by the challenges of high penetration of renewables, California‘s AB 2514 has set a target procurement of 1.3 gigawatts of storage by 2020 for California utilities. And recently, the California renewable energy portfolio standard added 50 percent by 2030 to the previous goal of 33 percent by 2020. This new goal will require many more gigawatts of multi-hour storage to shift midday over-generation of solar to the evening.

Given the emerging dependency of California and other grids on storage, it is essential for reliability, resiliency, environmental and economic reasons that the storage deployed be commercially viable. Some utilities have suggested that when they procure storage services from storage project owners or buy storage assets, the utilities should take the performance claims of the vendors at face value, since the contracts will penalize the vendors for non-performance. However, failures in large-scale storage deployments will reduce grid reliability. This will damage the storage and renewables industries, the utilities and their customers. It is important for all these reasons that utilities only procure storage assets and storage services that are commercially viable.

Commercially viable storage can be defined as storage that has very high probability of meeting the lifetime, functions, performance and other commitments of the suppliers of the storage asset or storage service.

Four factors for commercial viability

A full assessment of commercial viability of a particular storage alternative should at least consider the following four factors. First, what is the total number of projects and megawatt-hours deployed, anywhere in the world, using that technology, and what are its performance results? Second, is the claimed storage lifetime actually demonstrated, as evidenced by the actual years of operation for other projects using the same storage? Third, are these other storage projects providing similar grid functions to the current storage acquisition? If the other projects are not performing similar functions, their demonstrated performance and lifetime may need to be significantly discounted, or even eliminated from consideration. Fourth, do the storage vendors have the financial strength to meet their commitments, including service contracts, warranties and performance guarantees, even in the event of difficulties, setbacks or unexpectedly high costs?

The assessment of commercial viability should be of concern to project developers in selecting their storage technologies, to utilities (irrespective of whether the utilities are contracting for storage services from IPPs or buying storage directly), and to regulators in their role of approving the utility’s proposed contracts. (In this article, we use "independent power producer" or IPP as a generic term to refer to any non-utility entity providing storage services.)

Ensuring responsible choices

Responsible choices by these parties can be enforced by clear definition of their responsibility to make the case for commercial viability and to shoulder the risks if the commercial viability does not meet expectations.

Specifically, when the storage technology is selected by an IPP to provide storage services to a utility, the burden should rest on the IPP to prove to the utility that the storage is commercially viable.

Similarly, whether a utility is contracting for storage services or buying a storage asset, when the utility takes the proposed contract to the PUC for approval, the burden of proof should lie with the utility to make the case for commercial viability of the storage.

After the PUC approves a storage contract, the financial risk should remain with the IPP, and if the IPP fails to absorb the risk (e.g., due to the IPP’s insolvency), the risk should lie with the utility’s shareholders. If the utility owns the storage, the risk should lie with the utility shareholders. An IPP or utility may have performance guarantees from storage vendors to backstop the utility's and IPP’s risk, but if the storage vendor becomes insolvent (as has happened with a number of small, inexperienced storage vendors over the last few years), the risk should land on the IPP or utility.

Normally, projects will be partitioned into LLC structures, in part to insulate the IPP or utility from the risk of failure. However, it is up to the PUC to ensure there is sufficient financial depth of at least one party to make the ratepayer whole in the event that the storage fails to meet expectations. (This is one of the reasons for the fourth criteria given previously for commercial viability.) The only circumstances under which the risk should end up on the ratepayer is if the storage vendors, IPP (if any), and utility all become insolvent. By enforcing these burden-of-proof and risk rules, the process ensures each party is motivated to make responsible choices.

Are we creating a hurdle that is too high? Can any storage meet these criteria? Some absolutely can. Large pumped storage of gigawatt/multi-hour size has been extensively deployed -- over 150 gigawatts are in operation, and the technology has been proven to be commercially viable. However, smaller pumped storage (a few hundred megawatts and below) has far fewer deployments and may not meet these risk criteria.

With respect to batteries, sodium sulfur (NaS) storage systems provide a 15-year life and 4,500 cycles for six hours of discharge. NaS storage systems have been deployed for nearly two decades and there are now hundreds of projects in operation around the world. We would suggest that NaS storage systems are commercially viable.

Lithium-ion is on the borderline for these criteria. There are a number of projects deployed, but none that have yet demonstrated the promised lifetime for daily deep cycling for multi-hour shifting of over-generation of solar and wind.
Lead acid has been extensively deployed for uninterruptible power supply applications and is proven for this use. But for grid-scale applications, characterized by more frequent and deeper cycling, lead acid has far fewer demonstrations, and thus it is questionable whether it would meet the test of commercial viability for this application.

Nurturing immature technologies

Clearly, new technologies should be nurtured to see if they can develop into proven commercial viability. However, this should be done through non-mission-critical usage, so that if they fail (as will typically happen with any young technology), stakeholders can evaluate what went wrong, step back, take corrective action, and then try again. A few megawatts each for a number of storage vendors would be sufficient to give new storage technologies a valuable real-world proving ground, while at the same time not creating an aggregate megawatt risk that would threaten grid reliability.

Our company has long advocated that a storage testbed be established on the grid that allows emerging storage technology vendors to easily connect and operate as grid resources, in order to help them attain real-world experience. We call this concept the California Renewables Energy Storage Park. The connections at CRESP should be plug-and-play, up to approximately 5 megawatts per pad, with all grid interconnections and communications ready to go -- think of it as being like an RV campground hookup, but at a 5-megawatt-per-pad scale. The 5-megawatt size is small enough that failure of storage at multiple pads at the same time would not affect California's grid reliability.

The storage devices connected at CRESP should be allowed to operate as full grid resources, buying and selling into the wholesale market. Multi-year tenancy should be readily available -- it will be needed to get the bugs out of the new technologies.


David MacMillan and Ed Cazalet are the founders of MegaWatt Storage Farms.