Comparing nuclear power plants and wind farms resilience to hurricanes

Hurricane Matthew affected the continental US last week, the first since 2005. It was a category 5 hurricane that caused more than 1000 deaths, mostly in Haiti, and about 7 billion dollars damage as a preliminary estimate.

As Matthew quickly moved toward Florida and the Carolinas, rigorous procedures to ensure safe operations of nuclear power plants in the affected areas were implemented.

Four NPP are located within the affected area: St. Lucie (FL), Robinson (SC), Harris and Brunswick (NC), for an overall capacity of about 5600 MW.

Previous hurricanes have shown that NPP are robust facilities able to withstand strong hurricane winds and storm surge [1], nevertheless the “unusual event” status – the lowest NRC’s emergency condition- was declared for all the plants [2].

The plant personnel made sure that all the equipment potentially affected by heavy wind and rain were secured and a “walkdown” inspection through the plants response to disaster condition was initiated, including assessing the availability of emergency diesel power generators for at least a week.

When Matthew made landfall in South Carolina, early on Saturday, Robinson plant safely shut down due to loss of power and flooding of transformers. Harris suffered power outage too, but was already shut down for scheduled refueling. Brunswick was instead fully operational but was required to modulate down to 50% capacity on Sunday in order to respond to reduced capacity of the grid. By Wednesday all nuclear plants exited their “unusual event” status and, after routine safety inspections, ramped back to full capacity.

Wind power is the fastest growing renewable source in the US. According to DOE scenarios of 20% wind power in the US by 2030 (2008, [3]), offshore wind should contribute with 54 GW. Most of this power should come from shallow to intermediate depth farms along the Atlantic coast, that has a potential capacity of 920 GW and the Gulf region, with a potential capacity of 460 GW [4]. Incidentally, those regions are the preferred hurricanes corridors! Regardless the accuracy of these estimates and the feasibility of the envisioned goals, how would the wind farms stand a hurricane?

In August 2003, the typhoon Dujuan hit the southern part of China and caused severe damage to a wind farm located in the coastal area of the Guangdong province. The wind turbines were designed to survive a maximum gust of 70 m/s, but a maximum gust simultaneously with significant yaw error and rotor standstill had not been considered. The actual maximum gust did not exceed the design maximum gust of 70 m/s. Several wind vanes were damaged during the cyclone’s passage [5].

Few days later, typhoon Maemi almost flattened a wind farm on Miyakojima Island (Japan) [6].

While blades are relatively easy to replace, tower buckling is a severe damage that can require months to years for restoration [7].

At present there are no wind farms offshore the US East coast and in the Gulf of Mexico, but several are planned.

Thus a recent paper [7] estimated the resilience of offshore wind farms to storm conditions. Wind turbines are designed to operate with winds up to 25 m/s, over this threshold they shut down for safety reasons. The turbines currently on the market (Class 1) may (ideally, as Dujuan typhoon taught us) stand winds up to 70 m/s, but hurricane winds often exceed 80 m/s. Although the design of hurricane resilient turbines would be possible (Class S), this option comes with compromises on the productivity (i.e. they need stronger cut in wind to operate) besides higher costs [8].

Rose et al. (2012) [7] model both the risk from a single hurricane and the cumulative risk over the lifespan of a wind farm, through 4 sites offshore the Gulf and the Atlantic coast where farms are planned. Considering a farm size of 50 turbines, a considerable number of them is expected to be buckled down over a 20 years period by passing hurricanes: 16 out of 50 in Galveston County (TX) and 8 out of 50 in Dare County (NC), while numbers decrease in NJ and MA, where usually hurricanes loose strength.

Stronger hurricanes, category 4 and 5, cause more damage although occur less frequently. Overall, the damage occurrence over a turbine lifespan is dominated by one or two hurricanes. Most of the offshore wind potential is concentrated offshore Texas, Louisiana and North Carolina. The same for hurricane occurrence, at least one every 4 years, thus making the risk of significant loss of the capital investment relevant over a 20 years long period.

Yawing wind turbines, i.e. those that can oscillate and move accommodating fast wind direction changes, have better chances of survival. This would come to the expense of providing them a power back up, worth $ 30.000-40.000 each. The overall additional costs to improve hurricane resilience are estimated in 20-30% for onshore turbines, something less for offshore.

Another concern of a massive penetration of wind farms in hurricane prone areas would be how to assure the stability of the grid, thus of the power supply, through the shutdown period or the even longer time span required to restore severely damaged turbines. Again, as the recent case of Southern Australia showed [9], a base load of non intermittent and programmable is required. Thus here we are again talking about conventional thermoelectric power, you would think. Indeed “new energy” is often synonym of new unreliable installations with expensive back-up powered by fossil fuels, at best by natural gas. No thanks: our mind always run back to nuclear power!

 

Sources

[1] https://public-blog.nrc-gateway.gov/2016/10/06/hurricane-matthew-and-the-nrc/

[2] http://www.nei.org/News-Media/News/News-Archives/Southeast-s-Nuclear-Plants-Easily-Weather-Hurrican

[3]Shwartz M, Heimiller D, Haymes S, Musial W (2010) Assessment of Offshore Wind Energy Resources for the United States. (National Renewable Energy Laboratory, Gold- en, CO).

[4] Lindenberg S, Smith B, O’Dell K, DeMeo E, Ram B (2008) 20% Wind Energy by 2020: Increasing Wind Energy’s Contribution to US. Electricity Supply (National Renewable Energy Laboratory, Golden, CO).

[5] Clausen N, et al. (2007) Wind farms in regions exposed to tropical cyclones. (Germanischer Lloyd WindEnergie GmbH, Hamburg) European Wind Energy Conference and Exhibition.

[6]Takahara K, et al. (2004) Damages of wind turbine on Miyakojima Island by Typhoon Maemi in 2003.

[7] Rose, S., Jaramillo, P., Small, M. J., Grossmann, I., & Apt, J. (2012). Quantifying the hurricane risk to offshore wind turbines. Proceedings of the National Academy of Sciences, 109(9), 3247-3252.

[8] Musial, W. (2011). Large-Scale Offshore Wind Power in the United States: Assessment of Opportunities and Barriers. DIANE Publishing

[9] http://www.news.com.au/technology/environment/preliminary-report-reveals-cause-of-south-australia-blackout/news-story/92606772798e23e1ceec8c53f4256900