Wind potential and economics

Victor Niemeyer, the technical executive for EPRI’s climate change program, offered an objective view of wind drivers and economics. It was wind from 30,000 ft, quite unlike most other presentations at the workshop, which focused on specific ways to accommodate the variable nature of wind and to move the electricity it produces from point A to point B.

He said that if the federal government were to write into law a national policy to curb CO2 emissions below current levels, the legislation would likely initiate a competition to replace existing coal. Renewables, nuclear, and carbon capture and storage (CCS) would be among the high-profile solutions. Passage of climate legislation, Niemeyer continued, would initiate a “voyage of discovery” leading to an energy paradigm for the future that is much different than exists today.

Wind resource potential is huge, he said, exceeding half of the nation’s current electric needs at $90-$100/MWh, and possibly exceeding current generation from coal (Fig 1). Unsubsidized wind, the economist added, is competitive with $8/million Btu gas on a pure energy basis.

Niemeyer qualified his remarks by saying wind potential is adversely impacted by:

• Long periods of calm with no output over large geographic areas—swaths of real estate that may extend over several states.
• Rapid changes in wind generation over relatively short periods of time—an hour or less in some cases.
• “Anti-correlation” of wind with load.
• The cost/ability to permit new transmission needed to move wind energy to consumers.

The economist added that while he was not expecting climate legislation anytime soon, because it could happen might discourage power generators from considering coal as a viable option for future capacity. One of his slides plotted the CO2 emissions reductions required to satisfy several pieces of legislation introduced in the last several years. A couple of the laws proposed would require that CO2 emissions be halved from 2005 levels within 25 years, and cut to 80% below historic levels by 2050.

Next came a few numbers to illustrate the magnitude of the challenge associated with curtailing CO2 emissions—and the cost. The electric sector, Niemeyer said, produced 39% of the country’s CO2 (and one third of its total greenhouse-gas emissions) in 2006; 83% of the industry’s CO2 was emitted from coal-fired plants burning fuel at an average cost of $2.50/million Btu.

He believes that any cap-and-trade legislation seeking to cut emissions well below current levels would include incentives for new generation to back out coal; also, the national CO2 “price” would be whatever it takes to displace existing coal.

Wind potential. EPRI engaged AWS Truepower LLC, Albany, NY, to get a comprehensive assessment of wind resource potential. It identified more than 5000 viable utility-scale (100 MW minimum) wind-farm sites nationwide based on actual hourly meteorology from 1997-2008, assuming installation of 1.5-MW turbines.

Variables such as distance to the grid, terrain/wake effects, and exclusion areas were factored into the analysis. Fig 2 shows were the wind farms (capacity factors greater than 35%) would most likely be located.

An example analysis conducted for the North West Central (NWC) region is instructive, illustrating key points regarding the behavior of wind. About half the nation’s wind potential exists in this seven-state area (Minnesota, North and South Dakota, Nebraska, Iowa, Missouri, and Kansas). However, regional demand would be largely satisfied with less than 10% of that amount were it built. “Use it or move it,” Niemeyer said. Easier said than done: The regional grid can’t handle anywhere near the amount of wind power that could be produced.

More wind than load produces a local surplus that must be “spilled” if it can’t be exported. If you must spill, capacity factor of your wind turbines decreases and your pro forma takes a hit. Given adequate wind resources, the challenge is to match market needs.

The EPRI study was based on the following facts and assumptions: (1) actual state hourly load data for 2007 from Energy Velocity LLC, Boulder, Colo; (2) correlation of energy consumption with meteorological data to quantify the impact new wind generation would have on regional demand; (3) an additional 50 GW of new wind capacity installed in the region at qualified sites offering the highest capacity factors.

Figs 3-5 offer grid operators and power generators unfamiliar with wind’s idiosyncrasies a short course on its variability and electric system impacts. The NWC time series from Feb 28, 2007 through March 7 (Fig 3) reflects the week of highest wind output for the 2007 simulation. Keep in mind that this simulation includes the 50-GW addition to the wind capacity existing in the region.

The wind curve illustrates the variability in electric production experienced. Dispatchable capacity early in the week is close to 50 GW as a cold front moves through the region, and less than 5 GW at the end of the week after the front has passed. “Wind comes and goes,” Niemeyer said, putting up the next slide in the series.

Fig 4, for May 5-12, 2007, illustrates a prolonged period of low wind; a dead calm is experienced the afternoon of May 8. The EPRI executive said many people believe that although wind might not be blowing at any given point in the region, it is blowing elsewhere. That’s not necessarily true, he continued, showing a national weather chart for May 8 that revealed a stall extending over more than just NWC.

The time series from Aug 9-16, 2007 illustrates a typical summer pattern in NWC (Fig 5). Note that the wind pattern is more consistent day to day in summer than it is in winter (refer back to Fig 3). More importantly, the chart clearly shows the anti-correlation of wind with load—that is, wind production is highest when demand is lowest. Looking ahead, conventional assets backing-up wind will require the ability to ramp up and down quickly to maintain the continuous balance between load and generation needed for a reliable power system.

Niemeyer summed up his thoughts with these three points:

• Adding transmission enables greater utilization of wind resources.
• The cost of wind delivered to load is much higher than the simple cost of generation.
• Anti-correlation with load and the need for new interregional transmission capacity limit the amount of existing coal-based generation that wind can displace in a decarbonized electric future.

How much will wind cost? Niemeyer answered this question based on AWS TruePower’s national wind energy supply curves and EPRI’s estimates of generation and transmission asset costs. The exercise was to estimate the cost of producing and delivering 1000 TWh, or about 50% of the energy supplied by coal-fired powerplants (Fig 6).

The answer: It would take about 175,000 1.5-MW wind turbines to accomplish the goal at a total installed cost of about $650 billion. Delivery of the power produced would require about 13,000 line-miles of extra-high-voltage (800 kV dc) transmission lines at a cost of approximately $50 billion. Niemeyer mentioned that the biggest question is not about the cost of transmission, but rather if you can construct it at all.

Referring to the chart, he said that delivered cost jumps up right away (point A), because you must integrate the new capacity with the existing grid. It is about $450/kW for induced transmission associated with backing up wind, Niemeyer continued. If you add a couple of wind turbines, there is no associated grid cost impact because you can squeeze them onto the existing infrastructure.

But the installation of wind farms rated in the hundreds of megawatts require upgrades to transmission assets. And the further wind resources are located away from the load, the higher the voltage must be to reduce line losses. The curves of delivered cost are steep, he added, because wind does not line up with load (anti-correlation) and capacity factors for wind turbines decrease as you add more and more wind.

“The bottom line is that the country has a vast potential wind resource, but there are fundamental forces that limit how much we can use,” Niemeyer said in his summary remarks. The biggest is that wind output just does not line up that well with loads. The anti-correlation effect clobbers the economics of wind once you start to generate more than 10% to 20% of total electricity demand—depending on whether or not a large amount of interregional high-capacity transmission can be built.

The climate-change expert closed with this thought: Wind can play a strong role in a low-carbon electric future, but it will not be a dominant role.”