The closure of the San Onofre Nuclear Generating Station in 2013 left a stranded coastal asset. However, the existing transmission infrastructure, site control, and geographic topology make it an ideal site for a large-scale application of nascent seawater pumped storage technology.
Southern California Edison’s (SCE’s) San Onofre Nuclear Generating Station was retired in 2013 due primarily to premature wear in replacement steam generator tubing. The shutdown stranded approximately 2,500 MW of substation and transmission capacity from San Clemente into the California Independent System Operator (CAISO) grid. This infrastructure could be redeployed to enable a seawater pumped storage (SPS) facility, which could help facilitate California’s transition to 100% renewable energy.
The San Onofre site currently houses a decommissioned nuclear plant, along with its spent nuclear fuel. Turning a nuclear facility into an environmentally friendly energy facility that could store excess renewable energy during the day and power up Southern California at dusk is an exciting prospect—potentially eliminating “Duck Curve” issues for much of the region.
Designing a Seawater Pumped Storage Facility
The selection of SPS technology as an energy storage option is suggested for a number of reasons, but it’s primarily driven by the need to integrate renewable energy resources to meet ever-growing renewable portfolio standard (RPS) mandates. Pumped storage provides a large impact that other existing storage technologies, including battery storage, cannot yet match. However, the scarcity of freshwater resources in California makes the adoption of a typical pumped storage facility fed by freshwater untenable.
The San Onofre site offers a number of attractive characteristics for a possible SPS facility. The topography of the area is ideal for a pumped storage project, having sufficient land nearby with the requisite elevation differential for an upper reservoir and a mild slope to the coastline for penstocks or tunnels. The slope is in-line with typical penstock best practices, which recommend penstock lengths be less than five times the penstock elevation differential between its endpoints. The site’s proximity to the ocean makes possible a pumped storage facility that does not impact local freshwater supplies. The site also has good access from land or sea, and the San Onofre plant infrastructure is sufficient to support a 2,000-MW SPS facility.
An SPS facility would entail filling a 40,000 acre-feet reservoir, located across from the Interstate 5 freeway, where a natural 800 to 1,300-foot embankment would require only minimal disturbance to achieve the requisite head differential with respect to sea level. Reversible turbines could then pump water from ocean inlets to the reservoir during periods of low power demand, sufficient to provide eight 250-MW, 18-kV reversible pump-generators. This would provide a nominal SPS facility rating of 2,000 MW, at an 80% pump-to-generator operating efficiency. With the available reservoir size, head, and expected flow rate, it is recommended that the SPS facility use Francis pump-turbines, which should be sufficient to provide 10 hours of full capacity.
1. This image shows the existing Castaic six-unit pumped storage plant, which uses fresh water and a surge chamber. The facility also includes a smaller seventh unit for pump starting. Courtesy: Los Angeles Department of Water and Power
Operating all eight generators at full load would result in significant friction losses and would result in a net SPS facility rating of 1,800 MW. Ideally, one of the units could be configured as a pump-starting unit for the other generators, increasing SPS facility reliability. A surge chamber, measuring approximately 120 inches in diameter and 450 feet tall, could also increase reliability by relieving excess pressure in the penstocks. Figure 1 illustrates a similar 1,500-MW configuration used by the Los Angeles Department of Water and Power (LADWP) in its Castaic Pumped Storage Power Plant.
To maximize operational capability and minimize capital expenditures, four of the turbines could be traditional fixed-speed turbines, with two variable-speed capable units having a ramp rate of about 20 MW per second, fully synchronized. Leveraging the existing infrastructure would allow the SPS facility to be available as early as 2027, costing an estimated $8 billion to complete. Grid interconnections could use existing infrastructure, which has close proximity to large load centers and market hubs.
Seawater Pumped Storage
SPS has the advantage of only requiring one reservoir, as the water would be drawn directly from the Pacific Ocean. The SPS facility is estimated to cost about 20% more than comparable freshwater facilities because of increased costs for corrosion prevention and groundwater contamination prevention. Because corrosion is the biggest concern differentiating fresh and saltwater pumped storage, some of the main concerns and mitigation solutions will be addressed here.
Much of the corrosion control efforts are derived from lessons learned at the Yanbaru project, the world’s first SPS facility. The Yanbaru plant, a 30-MW facility, was installed in Kunigami Village in Okinawa, Japan, in March 1999, and it operated successfully before being decommissioned in 2016. The results of the Yanbaru project demonstrated the viability of SPS schemes and spurred interest in the technology.
2. This concept drawing shows a proposed pumped storage facility using seawater, which is under consideration in Australia. Courtesy: EnergyAustralia
Arup and EnergyAustralia are currently pursuing the Cultana Pumped Hydroelectric Energy Storage Project, a 225-MW pumped storage facility using seawater, which could potentially generate 1,800 MWh of energy daily. The Cultana project’s planned installation is near the northwestern tip of the Spencer Gulf in South Australia. It has been estimated construction could take three years to complete. Arup is currently undertaking a front-end engineering design study to progress the Cultana project to a final investment decision. Figure 2 demonstrates a conceptual drawing of the proposed Cultana project, which serves to illustrate the operation of the SPS facility as well.
One of the primary foci in the construction of the Yanbaru project was corrosion-mitigation based. Conventional anti-corrosion efforts remain problematic because pump-turbines have a number of major parts embedded and many other parts have limited spacing. All the materials exposed to water passage were made as smooth as practicable to minimize the initiation of crevice corrosion. The bearing housing insulation and seals were doubled to prevent seawater entry. Wherever possible, bolts were sealed by rubber gaskets to isolate seawater incursion. The main shaft was provided with a slip ring to provide cathodic current. For those areas where cathodic current was impractical to prevent corrosion, a sacrifice electrode system was used.
Some modifications to the Yanbaru project that are recommended for the SPS facility include using stainless steel for all flow portions and the application of cathodic protection, with adjustable corrosion preventive current. To reduce cost, a lower-grade steel may be substituted, if coated with a thick film of mixed phenol and epoxy resins for corrosion prevention.
In addition to the Cultana project in Australia, a number of other seawater pumped storage projects are currently being considered around the world. A 300-MW project is being considered in Lanai, Hawaii, as is a 300-MW project in the Atacama Desert in Chile. There are also projects being considered in the Middle East, Europe, and Mexico.
Benefits of the SPS Facility
The SPS facility (Figure 3) offers a number of benefits to the ratepayers of Southern California. The SPS facility would use the existing electrical utility infrastructure, substation, and transmission line, bringing a significant revenue stream to owners of those stranded assets. It would also make use of valuable beachfront property, the use of which may be viewed with uncertainty due to its historical proximity to a nuclear generating plant.
3. This concept drawing shows a recommended seawater pumped storage facility reservoir design, located southeast of the San Onofre Generating Station. As proposed, the reservoir would reside on federal land. Source: Robert D. Castro and Daniel Scorza
The grid-scale bulk storage of excess renewable power would allow for the requisite flexibility in integrating renewable energy into electrical grids built primarily for dispatchable generation. The SPS facility would supply flexible deployment capability to provide a much-needed, carbon-free, baseload capability for the next 50 years.
Preliminary analyses indicate that the SPS facility may reduce annual system production costs by $20 million to $70 million per year, depending on RPS requirements. As higher RPS levels are instituted, the decrease in curtailment due to the implementation of the SPS facility may be as high as 4%.
Other ancillary benefits include frequency regulation and improved voltage stability. The SPS facility would reduce resource cycling and renewable curtailment. Estimated economic benefits to the state include an additional $5 billion in statewide gross domestic product and 50,000 additional job-years of employment over a 25-year period.
As the proposed location is along the California coastline, the project would have to comply with the more-stringent California Environmental Quality Act (CEQA) requirements in addition to National Environmental Policy Act (NEPA) provisions. One of the most-concerning aspects is the applicability of the U.S. Environmental Protection Agency’s (EPA’s) Clean Water Act Rule 316(b), which is a regulation intended to reduce the injury and death of fish and other aquatic life caused by cooling water intake structures of power plants and factories. While the SPS facility would not have cooling water intake structures, and would therefore seem to be exempt from Rule 316(b), there do not appear to be other rules in place to specifically address the seawater intake for nascent SPS projects.
The intent of Rule 316(b) is to reduce impingement and entrainment of aquatic organisms becoming trapped on intake water screens, causing mortality. A possible way to meet the intent of Rule 316(b) would be to use “fish-friendly” modified traveling water screens in the intake structure. This approach could comply with the best-available technology for reducing impingement/entrainment in SPS facility operations. The Electric Power Research Institute designed and fabricated a set of five traveling screens that demonstrated over a one-year period in 2013 that such a system returned 92% of the fish and 89% of the debris back to the source. This 92% return rate would seem to comply with the Rule 316(b) intent, which targeted a minimum 90% return rate.
In lieu of the penstock approach illustrated in Figure 1, three tunnels could be employed to transport water and minimize aesthetic and above-ground impacts. The San Onofre plant used seawater for cooling, so some of the existing seawater inlets could be modified for use in an SPS facility. Preliminary studies indicate that SPS operation would result in less than 1,000 acre-feet of evaporated water loss per year. The prevention of seawater intrusion into the water table was largely addressed in the Yanbaru project by lining the structure in a detailed fashion, using multiple layers of rubber sheets, supplemented with filling concrete and drainage piping.
Experience in once-through ocean cooling of generators has found that barnacles, and barnacle-type organisms, are apt to adhere to areas where water flow is less than 5 meters/sec. Those areas should be coated with water-repelling paint to discourage barnacle attachment.
The pump seawater intake should be at least a half mile from the shoreline, at a depth no higher than 70 feet below the surface. This minimum depth ensures that surface wave impact is negligible on intake performance.
Earthquakes are always a concern for California infrastructure, but in this instance, the mitigation efforts undertaken to protect the San Onofre plant can be leveraged to mitigate concerns for the development of an SPS facility. The closest tectonic fault is located less than a mile away, the Christianitos fault, but it is considered inactive. The closest active fault is about five miles offshore, so protective measures have been put in place to protect against tsunamis. One such existing protective measure at San Onofre is the installation of a 25-foot seawall that could be used to protect an SPS facility. The Yanbaru project used tetra-pods to provide the tailrace protection from the waves.
New technologies are typically more risky than proven technologies, and, while pumped storage has been around for hundreds of years, SPS has not yet been implemented on a large scale. Prudent project management dictates that measures should be taken to minimize financial exposure. Primacy among these efforts would be to spread the development risk to multiple off-takers.
LADWP has implemented a few different models toward this implementation, including having a number of utilities finance the project (similar to the 1,900-MW Intermountain Generating Station); securing site control and having a developer build on the land (similar to the 250-MW Beacon Solar Facility); or having a developer engineer, procure, and construct the facility and entering into a power purchase agreement (PPA) to secure the generation output. Given that developers are typically leaner than utilities when constructing generation facilities, the PPA route may be the quickest way to get the SPS facility on-line, and, provided that adequate controls, specifications, guarantees, representations, warranties, oversight, and testing are in place, may result in the best long-term value to utility ratepayers.
A reliability assessment would be needed to further vet the reliability and benefits of an SPS facility. The assessment should include a feasibility study, a system impact study, a facility study, stability analysis, and a path transfer limit study. Most of these could be achieved as part of a large generator interconnection agreement with SCE. These studies should be consistent with North American Electric Reliability Corp. reliability standards and Western Electricity Coordinating Council path rating processes. A remedial action scheme for relay operation and subsequent generation islanding would also need to be considered, if the project moves forward, to ensure full integration for the off-takers.
The interconnecting balancing authority, CAISO, uses nodal pricing to account for systemwide transmission congestion. Accordingly, production cost modeling incorporating locational marginal pricing needs to be performed on various scenarios of projected power flows to determined annual generation costs. Projections valuing storage benefits, such as energy arbitrage, curtailment avoidance, and integration costs, need to be assessed, with ancillary services like frequency response, reserve, and ramping benefits included, as these benefits supply up to 50% of the projected value.
A more-thorough vetting could be accomplished with a stochastic benefit-cost sensitivity analysis considering the probabilistic impact of key variables. A sensitivity analysis incorporating various risks and scenarios of policy and market conditions based on the latest regional integrated resource plans, transmission plans, and regulatory/public policy environment would be essential when considering capital expenditures of this magnitude.
One contemporary example of risk to be considered is the potential developer or off-taker exposure from transmission through high wildfire areas and California’s inverse condemnation laws. Such liabilities have already forced California-based Pacific Gas and Electric into Chapter 11 bankruptcy.
Land acquisition and site control should be investigated and potential impediments identified. Environmental studies, including CEQA, are typically critical-path items that need to be initiated early to avoid schedule slippage. Engagement with the community and public relations efforts on repurposing the nuclear site into a more benign generation facility should also begin to identify the level of public acceptance/opposition.
Micrositing for the pumping station and conveyance route should be considered, along with general plant arrangement drawings. This would help determine suitable tunnel/penstock and pump house excavation technologies.
Geotechnical engineering should commence with investigations into tunnel construction, intake structures, and underground pump cavern. These would require a site geology review, a powerhouse seismic/tectonic study, and initial borings along the conveyance route for preliminary design requirements.
Identifying specific off-takers could help define how the SPS facility could best be designed and operated to maximize the benefits for the particular utilities using the facility. Concerns like resource adequacy, curtailment impacts, and energy reserve requirements could be addressed in this fashion. These would help determine a preliminary schedule and budget for planning, engineering, procurement, construction, and commissioning. ■
—Robert D. Castro teaches graduate level power courses at the University of Southern California, and Daniel Scorza is a former power engineering manager with LADWP.