Wind and Water
Neither water nor wind, birth nor death can erase our good deeds.
-Buddha (slightly modified by me)
At the beginning of this year I posted "Large Wind, Small Wind and Future Wind". This was a review of most wind power technologies (link below). Recently there has been more interest in building offshore wind farms. The U.S. finally commissioned our first offshore project, but the Europeans are well ahead of us. Read on for more details.
2.What does Europe Know that We Don't?
The score is the U.S. 30, Europe 15,780 (megawatts), and I would say it's a blow-out (pun intended). The following are the basic reasons:
2.1.The Price of Utility-Scale Offshore Wind Farms
The cost to implement an onshore wind farm is not easily quantified, and depends on many factors, but below we will present some official government numbers. The cost of building an offshore wind farm is much more variable. To give you a hint, I was slightly involved with the first attempted U.S. offshore wind farm (Cape Wind) where the project partners spent several $million. Political pressure resulted in ceaseless delays until the project supporters finally bailed, killing the project. The cost per megawatt: several $million / 0 MW = infinity. Not a good result going forward.
Finally, in 2016 the U.S. completed its first offshore wind farm near Block Island, RI. Block Island is a small island 13 miles south of mainland Rhode Island. Until the completion of the wind farm and the associated 21 mile long transmission cable to the mainland, Block Island relied exclusively on diesel generation.
Building this project was as much of a battle as Cape Wind. The developers, including Deepwater Wind and GE, were forewarned by Cape Wind and lined up very strong political support. This finally won the day in spite of strong opposition.
The price of the Block Island Wind Project was 300 $Million for a 30 MW project, but future projects should be less expensive. Consider the following facts:
- Onshore wind projects are much less expensive than offshore projects. Although the offshore costs will come down over time, the above number for Block Island is $10,000 per kW. Compare this with the U.S. DOE 2016 average cost of onshore projects which was $1,590/kW.
- Block Island is currently the best-case design because the ocean in this area is shallow enough for fixed-bottom (bottom-supported) turbines. If the depth is over 200 feet (60 m), fixed-bottom designs will not work. More about ocean depths and alternatives to fixed-bottom turbines in later sections.
- The reason there is a battle to site these wind farms has to do with views. Shore-line residents do not want their ocean-views spoiled by wind turbines. A DOE laboratory did a visual impact study for large wind turbines that indicated that they were a main focus of attention up to 12 miles away.
3.Potential of Offshore Wind
When planning an offshore project a major milestone is negotiating a reasonable power purchase agreement (PPA) with one or more utility-entities. By "reasonable" I mean a PPA that will guarantee the developers a profit on the project. We will go into more details about this later, but a major consideration is how rapidly nearby state governments want to increase renewables in their generation mix. Other factors are barriers to onshore wind farms (like lack of suitable sites).
The fuel for any wind turbine is wind. So an examination of the potential would start by looking at the wind-speed in off-shore areas. The figure below is wind speed at 90 m above the surface for U.S. potential offshore wind areas.
Next consider four facts:
- The average wind speed should be at least in the range of 18 to 20 miles per hour (8 to 9 m/s) at a height of 280 ft. (90 m) for a site to be considered.
- The energy in wind increases with the third power (cube) of the wind-speed, so an average speed over 22 mph (10 m/s) starts to look really good.
- In general, the wind offshore is much more consistent and has a higher speed than onshore.
- Almost all of the European offshore wind farms use fixed-bottom turbines.
The figure below shows the ocean depth for U.S. offshore areas (referenced source includes both wind speed (above) and depth maps).
If you will remember, any depth over 200 feet (60 m) will not work for fixed-bottom turbines. We will go into more details on what will work in deep-water areas a bit later.
Occasionally I get lucky. When looking for another document, I came across:
An Assessment of the Economic Potential of Offshore Wind in the United States from 2015 to 2030
Published by the National Renewable Energy Laboratory (NREL) last March, it used a reasonable technique to evaluate the title potential. The outputs of this study were clear definitions of the viability of all U.S. offshore areas. Although they were very thorough, they also added many caveats regarding their results. Some of these (and other technologies they didn't consider), may provide enough wiggle-room to extend their conclusions (and my paper), and allow doubtful areas to be reconsidered.
This study looked at two metrics in evaluating the viability of offshore wind projects: the levelized cost of energy (LCOE) and levelized avoided cost of energy (LACE). These were calculated as defined below for more than more than 7,000 potential U.S. offshore wind sites. Note that much of this work is documented in an earlier paper, referenced below. LCOE is the total cost of generating a unit of electric energy, commonly expressed in dollars- per-megawatt-hour (MWh), over the expected lifetime of a generating plant. LACE is a metric to approximate the electric system value of a generation technology over its expected lifetime and is commonly also expressed in dollars per MWh. Although calculating LACE is a very complex process, the referenced study used a simplified model. The higher LACE is, and the lower the LCOE is at a given site, the more likely it is for this site to be economically viable, all else being equal.
LCOE = [ (FCR x CapEx) + OpEx ] x 1,000 / AEPnet
LCOE = levelized cost of energy ($/MWh)
FCR = fixed charge rate (%, any type of fixed expense that recurs on a regular basis)
CapEx = capital expenditures ($/kW)
AEPnet = net annual energy production (kWh/kW/yr)
OpEx = annual operational expenditures ($/kW/yr)
LACE = [ (MP x AEPnet)+( CP x CC ) ] x 1,000 / AEPnet
MP = Average marginal generation price ($/kWh)
AEPnet = Net annual energy production (kWh/kW/yr)
CP = Capacity payment ($/kW/yr)
CC = Capacity credit (%, capacity credit is defined as the average discount of capacity payment, considering a generator's intrinsic-variability and availability)
Note that the capacity credit for fully dispatchable generation approaches 100%. The typical capacity credit for wind is in the 10% to 15% range.
The difference between LCOE and LACE at a given location (“net value”) can indicate the economic potential of a new offshore wind project at a high geospatial resolution. This is shown below for each location, i:
Net value ($/MWh)i = LACEi – LCOEi
The net value for each potential location was mapped, showing spatial distribution net value of offshore wind for five U.S. coastal regions. The value for the Atlantic Coast, Pacific Coast, Gulf of Mexico and the Great Lakes for each of the focus years (2015, 2022, and 2027) are presented below.
The only areas defined as having economic potential for off-shore wind by this analysis, and the potential capacity is given in the table below.
The above summary excludes many of the details of this analysis. If you wish to delve into these, go to footnote 5, click on the link, and start reading.
As most good analyses do, this document described the limitations of its work. The following are the most significant of these.
- To achieve the modeled cost reductions in the U.S., a key assumption is that there will be continued investments in technology innovation, developments, and the market visibility of a robust domestic supply chain.
- This analysis is limited by available data and a set of simplifying assumptions. It does not consider competition among technologies, dynamic feedback from increasing renewable deployment on wholesale electricity prices, or the alleviation of electricity system constraints (e.g., transmission constraints) over time.
- Policy-related factors that may influence the “net value” of a renewable energy project were not considered explicitly. For instance, renewable energy support mechanisms, energy sector and environmental regulations (e.g., carbon pricing, loan guarantee programs) may increase the “net value” and economic potential. Conversely, regulatory uncertainty and market barriers may decrease the “net value” and economic potential.
- No estimates were made to account for disruptive technologies (my insertion).
Some of the above limitations will be used in the next section to explore how some offshore projects can be moved from improbable to probable.
We live in a world where the disruptive is the norm. Climate change has started to accelerate, and governments have noticed. Historic storms, droughts, wildfires, flooding, etc. have demonstrated the future disruptions we might expect.
Many have criticized California (et al) for leaping forward to 50% renewables by 2030. If you include nuclear and large hydro (California doesn't), California might meet this goal by 2020, and 70% by 2030. California really wants to have net-zero fossil-generation ASAP. This means that we will place a premium on any renewable, and fully support sources like offshore wind.
4.1.Current State of the Art
Floating wind turbines are required for deep water (>200 ft.), and there is currently one wind farm that uses floating turbines: Statoil's Hywind Scotland Project (initial operation in 2017). This project is described in my prior paper (linked in Section 1 of this paper).
Three designs for floating turbines are shown below. These are (from left to right): Spar-buoy, Semi-submersible and Tension leg platform. Hywind Project used Spar-buoys.
Floating turbines should be roughly the same costs as fixed-bottom turbines by the late 2020s. The higher capital expense of the platform, moorings, and anchors required by floating turbines will be offset by lower installation costs and lower operating expense driven by cheaper repair costs for major components."
4.2.Future Disruptive Technologies
The two largest renewable generation technologies are photovoltaic (PV) and wind. Both of these are highly intermittent, although the dynamics are somewhat different. Many future large solar projects will be paired with battery energy storage systems (BESS) to mitigate their intermittency. This process has also started for large wind farms (see section 4.1.2 in my prior paper, linked in section 1).
As we mentioned above, Statoil's Hywind Scotland Project, which started operation last year, is the first floating wind farm. By the way, on Tuesday of this week (5/15) as I'm writing this, Statoil ASA (ASA = Assembly of Statoil) changed its name to Equinor ASA. Go through the link below for more information. Hereafter we refer to this firm as Equinor.
Equinor already has plans to add a 1 MW BESS to the Hywind Project next year.
Adding a BESS to a wind farm allows the combined project to be (at least) partially dispatchable, greatly increasing the LACE through an increase in the capacity credit (see section 3.2). The capacity credit for a normal wind farm is 10-15%. For fully dispatchable generation this approaches 100%. Adding storage to wind dramatically changes its economics.
In addition, storage excels at fast frequency regulation (an ancillary service), and a premium may be offered for this.
The other modern technology is HVDC under-water transmission lines. In a dress rehearsal for the advanced DC cables that will be required to support offshore wind farms, a unique project was implemented a few years ago to solve a very large problem that San Francisco had. The problem was that they did not have enough power to supply the city without keeping a couple of very old peaking plants in the city operating. Running a transmission line up the San Francisco Peninsula would be extremely expensive and would require many years to complete. This area has some of the most expensive property in the U.S., so a transmission line would also be very unpopular, as was anything that involved replacing old peaker plants with new ones.
Around 2000 the California ISO organized a team called the San Francisco Stakeholder Study Group. After reviewing all possible alternatives they decided on the Trans Bay Cable (TBC) Project. Pattern Energy completed the TBC project in 2010.
TBC involved a 53 mile long high voltage DC transmission cable running from the PG&E Pittsburg (CA) Substation to their Potrero Substation in San Francisco (see the figures below).  This cable could transmit up to 400 MW of power, and provide as much as 40% of San Francisco's demand.
The project used a new generation of voltage-sourced converter technology, with insulated gate bipolar transistor (IGBT) as the switching device. This design (Siemens HVDC PLUS) has several advantages over other converter technologies:
- A much smaller footprint (a third to a quarter compared to other technologies),this is particularly important in with the cost of land in San Francisco)
- Independently controls active and reactive power
- No AC filtration is required
- Black-start capable
In the next section I will spin a scenario for cost-effectively deploying offshore wind in the Pacific north of San Francisco.
5.Scenario for Making the Improbable Probable
First some caveats: I am in no way qualified to even perform the basic engineering required to conceptualize this scenario. Although I have managed some major projects, they were not using these technologies, but rather ones where I am somewhat qualified. My hope is that someone will be sufficiently intrigued by the following and put together a team that can access whether this is a reasonable idea. If so, when?
When I started thinking about this paper, whether Northern California could reasonably host off-shore wind turbines was at the front of my mind. I knew there were major challenges, including:
- The coastal waters are not just too deep for fixed-bottom turbines, they are really deep (my guess is over 1,000 m deep where I would site turbines).
- There are major sea-lanes about 15-miles off the coast. This would push any off-shore wind farms out to at least 20 miles.
- Going north from San Francisco, there are no major (or even medium-sized) cities near the coast. Santa Rosa (pop 175,000) is the closest, and it's about 15 miles away from the coast. Also the population density is very low near the coast. Both of these (and other factors) mean that there are no high capacity transmission lines or substations near the coast.
And then there are the economics as projected in section 3.2. I believe that storage and modern HVDC underwater cables, as outlined in section 4.2, can shift the economics to make them acceptable.
And then there is the good news:
- Northern Californians readily accept the sight of wind turbines. I live in a city that is only a few miles from the Altamont Pass – a major site for large (1.8 to 2.4 MW) turbines. Many residences can see these from their property, and I have never heard of a complaint about these.
- Northern California frequently has either fog along the coast or a fog bank a few miles off shore. This would obscure (or almost obscure) large (10 MW) turbines 20 miles off-shore a large percentage of the time.
- Although we have Pacific storms, as far as I know, the Northern California Coast has never had a Hurricane, and storms are rare from June through September.
Look at the wind-speed map in section 3.1. Note that the orange area (9.0 to 9.5 m/s) starts about 70 miles north of the Golden Gate (entry to San Francisco Bay). The red area (9.5 to 10 m/s) starts about 150 miles north of the Golden Gate. A single high-capacity HVDC cable servicing multiple adjacent wind farms in this area will spread the cost of this cable (and supporting infrastructure) and help the economics.
I am proposing the following:
Start building a medium-sized wind farm (200 MW) about 90 miles north of the Golden Gate and about 20 miles off-shore (see the map below, the red "X"). The turbines can be assembled in San Francisco Bay, and towed into place. A creative solution may be needed to get them under the Golden Gate Bridge, also the bay may be too shallow for the full (final) depth of a spar-buoy design.
A floating platform should be placed near the southernmost turbine position (near the red "X". This platform will contain the sea-side HVDC converter station. I would guess that this will be a DC/DC converter design to convert the HVDC used for the transmission cable to MVDC used to collect energy from the turbines.
A HVDC transmission cable will be run south from the platform, through the Golden Gate, and into the east side of San Francisco as shown in the map above (red line). Oversize the transmission cable from the (first) wind farm to San Francisco to flow up to 1 GW.
Upgrade the Trans Bay Cable to provide bi-directional power flow.
The battery energy storage system could be put on the same platform as the convertor, a separate platform or integrated into the turbines. A spar buoy design turbine uses extensive ballasting below the waterline for stability, and batteries could provide at least part of this weight.
Daisy-chain additional wind farms further north from the first wind-farm (and into higher wind (red) area). The cable to the second wind farm would be connected into the original HVDC cable, the third cable into the second cable, and so on. This would result in a total of approximately five 200 MW wind farms.
The above could help completely displace all fossil generation in Northern California by 2040.
 Tatiana Schlossberg, New York Times, "America’s First Offshore Wind Farm Spins to Life", Dec. 14, 2016, https://www.nytimes.com/2016/12/14/science/wind-power-block-island.html
 U.S. Department of Energy, “2016 Wind Technologies Market Report”. August 2016. https://energy.gov/eere/wind/downloads/2016-wind-technologies-market-report
 U.S. Department of Energy, Argonne National Laboratory, " Wind Turbine Visibility and Visual Impact Threshold Distances in Western Landscapes", http://visualimpact.anl.gov/windvitd/docs/windvitd.pdf
 Marc Schwartz, Donna Heimiller, Steve Haymes, and Walt Musial, NREL, "Assessment of Offshore Wind Energy Resources for the United States", June 2010, https://www.energy.gov/sites/prod/files/2013/12/f5/45889.pdf
 Beiter, P., W. Musial, A. Smith, L. Kilcher, R. Damiani, M. Maness, S. Sirnivas, T. Stehly, V. Gevorgian, M. Mooney, and G. Scott. 2016. A Spatial-Economic Cost Reduction Pathway Analysis for U.S. Offshore Wind Energy Development from 2015–2030 (Technical Report). NREL/TP-6A20-66579. National Renewable Energy Laboratory (NREL), Golden, CO (US). https://www.nrel.gov/docs/fy16osti/66579.pdf .
 Rhodri James and Marc Costa Ros, Carbon Trust, "Floating Offshore Wind: Market and Technology Review", June 2015, https://www.carbontrust.com/media/670664/floating-offshore-wind-market-technology-review.pdf
 Introduction to The Trans Bay Cable Project, http://www.ewh.ieee.org/r6/san_francisco/pes/pes_pdf/TransBayCable2014.pdf
 Justin Gerdes, Living Energy Issue July 2011 "Siemens Debuts HVDC PLUS with San Francisco’s Trans Bay Cable", https://www.energy.siemens.com/hq/pool/hq/energy-topics/living-energy/issue-5/LivingEnergy_05_hvdc.pdf