Renewable energy and intermittency

1,319 views

Published on

Wind and solar energy—two of the most popular
sources of renewable energy—are sometimes touted
as the answer to the world’s energy challenges.
Some advocates of these energy sources want us
to believe they can solve a plethora of problems,
ranging from avoiding the disastrous 2010 oil leak
in the Gulf region to materially reducing global
climate change. Wind and solar energy are also
routinely promoted with the promise of green jobs,
which will lead to a green technology revolution
while improving the environment—and making
us “energy independent” to boot. But how well do
wind and solar energy solutions actually perform on
these promises? Let’s take a rational loo

Published in: Technology, Business
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,319
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
52
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Renewable energy and intermittency

  1. 1. A RATIONAL LOOK ATRENEWABLE ENERGYAND THE IMPLICATIONS OF INTERMITTENT POWERBy Kimball Rasmussen | President and CEO, Deseret Power | November 2010, Edition 2.0 New Section — Natural Gas vs. Wind, p 14
  2. 2. TABLE OF CONTENTSForward................................................................................................................................................................. 2Wind Energy......................................................................................................................................................... 2 Fundamental Issue: Intermittency............................................................................................................ 3 Name-plate Rating versus Actual Energy Delivery............................................................................... 3 Wind is Weak at Peak. ................................................................................................................................ 4 . Texas ............................................................................................................................................................ 4 California...................................................................................................................................................... 4 The Pacific Northwest................................................................................................................................ 5 The Western United States........................................................................................................................ 5 Enter the “Twilight Zone”—A Control Area Nightmare......................................................................... 6 The Shadow Grid—The Fossil Fuel Stand-In for No Show Wind............................................................ 8 The Los Angeles Department of Water and Power (LADWP).............................................................. 9 Increase in Carbon Dioxide from Wind Power—It is Possible............................................................... 9 Got Transmission? Another Missing Cost Element.............................................................................. 10 . The Electric Continental Divide................................................................................................................ 11 Wind Energy Storage—Not Ready for Prime Time................................................................................ 12 Wind Turbines can Consume Electricity................................................................................................. 13 Value of Power—Demand versus Energy................................................................................................ 14 NEW IN 2.0 VERSION: Wind Power Compared to Natural Gas Power............................................... 14Solar Energy....................................................................................................................................................... 16 Not All Sunshine is Equal.......................................................................................................................... 16 How Expensive is PV Solar?..................................................................................................................... 17 Large PV Solar. .......................................................................................................................................... 17 . Concentrated Thermal Solar.................................................................................................................... 17 Solar Demand Versus System Peak........................................................................................................ 18 The Value of Solar Power—Demand Versus Energy............................................................................. 19 The Solar Synopsis.................................................................................................................................... 19 .The Renewable Portfolio Standard (RPS) or How 20 PercentCan Easily Become 100 Percent of a Utility’s Plant Investment. ....................................................... 19 .Summary............................................................................................................................................................. 21 Eyes Wide Open......................................................................................................................................... 22Special Thanks.................................................................................................................................................. 22
  3. 3. A RATIONAL LOOK AT RENEWABLE ENERGYand the Implications of Intermittent PowerBy Kimball Rasmussen | President and CEO, Deseret Power | November 2010, Edition 2.0There seems to be a common misperception, that produce breakthroughs, which could mitigate somewind turbines provide an inexhaustible supply of of the impediments to high penetrations of windcheap energy—after all, the fuel is free, isn’t it? and solar power to the electric grid, but as yet theseWhy don’t we simply build more and more impediments remain significant, and no economicalrenewable energy and achieve low-cost energy mitigation is on the immediate horizon.independence, while at the same time creatingmillions of new jobs to fuel a green economic FORWARDrecovery? What’s not to like about that? Wind and solar energy—two of the most popularUnfortunately for all of us, the real world poses sources of renewable energy—are sometimes toutedlimits on renewable energy technology, and with as the answer to the world’s energy challenges.those limits come costs—relatively high costs Some advocates of these energy sources want us(financial and otherwise), as will be shown in to believe they can solve a plethora of problems,this paper—that must be paid to integrate even a ranging from avoiding the disastrous 2010 oil leakmodest amount of renewable energy into the power in the Gulf region to materially reducing globalsupply portfolio. climate change. Wind and solar energy are also routinely promoted with the promise of green jobs,This paper will explore wind and solar energy in which will lead to a green technology revolutionterms of their environmental, operational and while improving the environment—and makingeconomic attributes. We will then place these in us “energy independent” to boot. But how well docontext to form a rational look at renewable energy wind and solar energy solutions actually perform onand the implications of intermittent power—the these promises? Let’s take a rational look.not-so-obvious operational challenges that have tobe addressed when large quantities of intermittent Wind Energyenergy must be accommodated on the electricity grid. Wind energy is becoming a significant consideration in the planning and development of the modernThis analysis is based on the current state of wind electric system. In the past decade, [the] Unitedand solar technologies. Further research and States (U.S.) wind energy output grew [as a sector]development efforts of these technologies may 10 times faster than [the combination of ] all other2 | A RATIONAL LOOK AT RENEWABLE ENERGY
  4. 4. sources of electric energy.1 The growth in wind Name-plate Rating versus Actualturbines is remarkable, given that the U.S. wind Energy Deliveryindustry installed more than 60 percent of all For the sake of this discussion it’s important toexisting turbines in just the past four years. know that all power producing equipment comesWe are truly in a wind boom. This is attributable with an output rating stating how much powerto a number of factors, including the fact that wind the facility will produce. This is referred to asprojects are much quicker to design, permit and name-plate capacity and it is expressed in kilowattsconstruct than traditional coal or nuclear plants, (kW) or megawatts (MW).2 For large utility gradeand wind energy tends to be one of the least generators the customary expectation is that onceexpensive renewable energy options. As a result, installed, they will deliver the name-plate output,the U.S. now leads the world in total wind energy on demand, when supplied with sufficient fuel.connected to the grid, surpassing nations such as Additionally, they will operate, if required, aroundGermany and Denmark. Where we go from here the clock. In the case of wind energy installationsdepends on how much we choose to subsidize this this is simply not the case. The output over time isalternative, as well as on some specific attributes of only a small fraction of name-plate rating because ofwind and the systems required to deliver it. the intermittency of the fuel resource (moving air). The ratio of actual output divided by maximumFundamental Issue: Intermittency potential output is defined as capacity factor.Despite robust wind development in the U.S., The entire sector of U.S. wind energy is currentlywind faces a nearly insurmountable issue: operating at a capacity factor of only 30 percent.3intermittency. Simply put, the intermittent natureof wind makes it difficult to harness effectively on a It is troubling that we see some astonishinglypower grid that is finely tuned to deliver electricity simplistic reports in the media which assert thearound the clock, matching demand second-by- number of homes that a given “wind farm” willsecond. The down side of this intermittency is allegedly supply. When reliability, expressed asclearly evident in the actual performance data of capacity factor, is taken into account, thewind turbines already installed. Wind performs serviceability of wind is much lower than advertised.poorly across all traditional utility metrics for Our modern understanding of supplying electricitygenerating resources. For reliability, stability, to a home is that a fully sufficient amount will beforecast ability, proximity to load centers, dispatch made available 24/7/365. In reality, it would be rareability and economics, wind power is a poor choice that any wind development could actually supplyfor large-scale power production. fully sufficient electricity to even a single home 24/7/365. Such misleading claims by developers may contribute to the sentiment that renewable energy can easily replace fossil fuels—it cannot.1 U .S. DOE Energy Information Administration, Net Generation by Energy Source, (2010), http://www.eia.gov/electricity/monthly/index.cfm. Wind data shown in [on] Table 1.1.A. Net Generation by Other Renewables: Total (All Sectors), 2000 through July 2010. Based on a comparison of 2010 versus 2000, wind energy output expanded by 1600 percent, natural gas electric generation grew 63 percent, nuclear grew seven percent, coal declined six percent and hydroelectric declined seven percent. The entire electric mix from all sources grew a total of eight percent.2 One megawatt is equal to one thousand kilowatts.3 U .S. DOE Energy Information Administration, Electric Power Monthly, Table 1.1.A, Net Generation by Other Renewables, www.eia.gov/electricity/ monthly/index.cfm. Annual capacity factor is calculated by dividing 2010 wind generation by the monthly average of installed capacity. A RATIONAL LOOK AT RENEWABLE ENERGY | 3
  5. 5. Wind is Weak at Peak CALIFORNIAThe intermittent and unpredictable nature of The State of California ranks third in the U.S.wind is further compounded by the fact that the for total installed wind energy (behind Texaswind tends to be weak during electrical peak and Iowa). California is also the third largestload conditions. Wind blows most consistently state geographically (behind Alaska and Texas).and creates the best generation opportunities during According to the California Independent Systemoff-peak hours, cooler days and evening hours. Operator (CAISO),Unfortunately, this is directly opposite the electric “ alifornia is a national leader in the Ccustomer demand profile. This disparity is a natural development of renewable resources. consequence of the climate forces that determine Because California has large quantitieswind: daily and seasonal temperature differentials. of renewable resources already on-line, a significant amount of historical data isFor example, on the hottest days of the summer available to accurately model and forecastthe wind tends to be low or non-existent when future performance of the various types ofair conditioning demands are at their peak. renewable resources.”Then when it gets windy, the temperatures willnaturally moderate and air conditioning loads drop “ ind generation presents . . . significant Woff just in time for the wind energy to pick up. operational challenges. Wind generationSo wind supply and user demand are out of sync. energy production is extremely variable,The same phenomena can be demonstrated to occur and in California, it often produces itsduring winter peak conditions. The very coldest highest energy output when the demanddays are also the days when the wind is not blowing. for power is at a low point.” 5For this reason, utility-scale balancing regionssimply do not plan for significant contribution of CAISO’s graph demonstrates its summer windwind at peak demand periods. This can be amply generation and average variation by hour:demonstrated in real-world, large scale examplesfrom Texas, California, the Pacific Northwest 2008 SUMMER LOADS RESOURCESregion, and the entire western United States. OPERATIONS PREPAREDNESS ASSESSMENT 1200TEXAS 1000Texas is home to the largest collection of wind 800 MWgeneration facilities in the nation. More than one 600out of every four wind turbines in America is 400found in Texas. The Electric Reliability Council of 200Texas (ERCOT) only plans for 8.7 percent of windname-plate rating as the “dependable contribution 0 7/17/06 0:00 7/17/06 12:00 7/18/06 0:00 7/18/06 12:00 7/19/06 0:00 7/19/06 12:00 7/20/06 0:00 7/20/06 12:00 7/21/06 0:00 7/21/06 12:00 7/22/06 0:00 7/22/06 12:00 7/23/06 0:00 7/23/06 12:00 7/24/06 0:00 7/24/06 12:00 7/25/06 0:00 7/25/06 12:00 7/26/06 0:00to peak requirements” (also known as CapacityValue), in accordance with ERCOT’s stakeholder-adopted methodology.4 This means that more than Average hourly wind generation Average output at peak91 percent of Texas wind turbines are expected to beoff-line when it matters most—at peak load periods.4 Kent Saathoff, ERCOT Expects Adequate Power Supplies for Summer, ERCOT, May 12, 2010, www.ercot.com/news/press_releases/2010/nr-05-12-10.5 2008 Summer Loads and Resources Operations Preparedness Assessment, Figure 5, Grid Assets, California ISO Version 1.0, April 28, 2008.4 | A RATIONAL LOOK AT RENEWABLE ENERGY
  6. 6. The wind capacity available at California peak On Tuesday December 16, 2008, the BPA system demand times is about 200 MW. The name-plate reached its peak for the entire year, with a demand capacity of California-based wind generators is of 10,762 MW. At the time of peak demand, about 2,600 MW. Hence, the wind power available the output of the entire fleet of wind resources, at peak is less than 10 percent, which is very with a name-plate value of 1,599 MW, was only similar to the Texas experience. In other words, 116 MW, or about seven percent of the name- about 90 percent of California wind turbines plate potential. This is very similar to the Texas and cannot be depended on to be producing at peak California wind experience, only in this case about load conditions. 93 percent are not producing at the winter peak. Note that Texas and California are both summer- THE WESTERN UNITED STATES peaking systems. Let us consider a vast winter- Now let us consider an even broader region—all peaking region—the Pacific Northwest—to see eleven western states, from Montana to New Mexico, how wind energy performs in that situation. from Washington to California, and everything in between. This vast area is served as a single THE PACIFIC NORTHWEST “reliability” region known as The Western Electricity Oregon and Washington rank fourth and fifth Coordinating Council (WECC). During the heat in the U.S. for total installed wind energy. wave of July 2006, the WECC system reached its The prominent Federal Power provider in the peak on Monday, July 24, 2006. The hottest day region—the Bonneville Power Administration was actually July 23, 2006, but this was a Sunday (BPA)—is a winter-peaking system with about so total loads did not peak until Monday. On the 10,000 MW of load. hottest day, the capacity factors for wind resources through most of WECC were well under five BONNEVILLE POWER ADMINISTRATION6 percent, and on the peak day, which was a slightly ACTUALS - WEEK DECEMBER 14, 2008 cooler day, the wind capacity factors were less than 12,000 10 percent.7 Again, this is very similar to Texas, 2008 BPA Peak Load California and the Pacific Northwest. 10,000 These real-world lessons illustrate a graveMegawatts 8,000 shortcoming of wind. Approximately 90 percent of 6,000 wind turbines can be expected to NOT PRODUCE 4,000 Load power at peak load periods, even when distributed Wind Nameplate over broad geographic areas. 2,000 Incidentally, I recently had a conversation with a 0 Sun Mon Tue Wed Thur Fri Sat trustee of a large mid-western utility that is home to 450 MW of wind generation. He asked me to guess Data: Bonneville Power Administration how much of the 450 MW of wind was actually producing during their system peak. I responded, “Probably between 30 and 40 MW.” He gasped, “How did you know? That is exactly what we are seeing!” Yes, wind is weak at peak. 6 Bonneville Power Administration, http://transmission.bpa.gov/Business/Operations/Wind/default.aspx. 7 WECC, Wind Capacity Issues, Working Draft, March 17, 2010. A RATIONAL LOOK AT RENEWABLE ENERGY | 5
  7. 7. ENTER THE “TWILIGHT zONE”— ramped up by 1,200 MW in only one hour, A CONTROL AREA8 NIGHTmARE and then down 800 MW in only 20 minutes. The demonstrated low performance of wind energy Such rapid changes cause extreme stress to a control during peak load conditions is only one side of area and in many cases result in market price the coin. The other side occurs during off-peak distortions and environmental degradation. periods when unscheduled, unanticipated wind energy comes booming onto the system ready to CONTROL AREA/TWILIGHT zONE IN ACTION:10 serve loads that are nowhere to be found. BONNEVILLE POWER ADMINISTRATION ACTUAL — APRIL 27, 2010 1,800 This can easily happen because of the physics of 1,600 60 Min wind energy: the power output of a wind turbine 1,400 -800 accelerates at a much faster rate than the simple Megawatts W 1,200 0M MW change in wind speed. For instance, if the wind 0 1,000 +1,2 speed changes from 10 to 20 mph (a doubling of 800 the wind speed) the associated power output will 600 20 Min change by a factor of eight.9 400 200 0 An actual case with the BPA brings the control area 4:00 5:00 6:00 2:00 3:00 4:30 5:30 6:30 2:30 3:30 7:00 7:30 problem into perspective. On April 27, 2010 about 3:00 a.m., wind generation on the BPA system Time of Day BONNEvILLE POWER ADmINISTRATION 2009/10 TOTAL SYSTEm LOAD AND WIND CONTRIBUTION 12,000 SYSTEM ANNUAL TOTAL SYSTEM LOAD PEAK LOAD 10,000 WIND NAME-PLATE WIND ACTUAL 8,000MEGAWATTS 6,000 4,000 Wind contribution 2% of name-plate 2,000 0 NOVEMBER 2009 8 Control Area - A power generation regulation region that maintains and balances its power load and power interchanges with other control areas. See also, Control Area Concepts and Obligations, North American Electric Reliability Council, 1992. 9 Note that the physics of wind energy is such that the change in power of a wind turbine is proportional to the cube of the change in wind speed. This means that if the wind speed cuts in half, the power output will cut to one-eighth. See also, Wind Systems Power Calculation, http://wind- power.generatorguide.net/wind-speed-power.html. 10 Bonneville Power Administration, http://transmission.bpa.gov/Business/Operations/Wind/default.aspx or ibid. 6 |.A.RATIONAL.LOOK.AT.RENEWABLE.ENERGY
  8. 8. Such erratic changes in generation run directly base-load, coal-fired generators but they have been counter to the needs of utility operators who select reduced to minimum-load status. The nuclear plants from a pool of different traditional generators to are running because they remain in “must-run” provide the right amount of power at the instant condition for safety and economic reasons. The wind it’s required. In a normal day they blend the turbines are cruising along at a modest output. outputs of traditional power plants that include coal, nuclear, natural gas, and in some regions Now assume that a sudden, unanticipated, change hydroelectric to work in concert to minimize in the weather brings with it a rapid ramping of operating costs while maintaining reliability. wind energy output. This can result in a large block of several thousand MW of unplanned energy that, Now we have the advent of wind. The use of wind when combined with the operating status just energy creates an unprecedented challenge, which described, can easily swamp out the total load can easily launch utility power systems into an off- requirements of the utility—meaning there’s literally peak condition, something that can be described as no place for the energy to go. the “twilight zone.” Now the utility is forced to make quick and drastic Consider an event that occurs during off-peak or decisions to balance loads and resources. I call this twilight hours. The various utilities are operating the twilight zone—a control area no-man’s land. with all of the peaking plants off line and many of One option might be to enact the costly decision to the intermediate resources off line. Still running are shut down a base-load resource, such as a nuclear WIND PERFORMANCE SCORECARD Total hours in peak period 2208 or 3 months Percent of period when output was below 5% of name-plate 45% Total hours in period below 5% 982.6 or 5.9 weeks Consecutive hours of generation below 5% 103.9 or 4.3 days Capacity Factor during peak period 16%DECEMBER 2009 JANUARY 2010 A RATIONAL LOOK AT RENEWABLE ENERGY | 7
  9. 9. or coal unit, and then subsequently face a high cost Do you think the twilight zone problem“re-start” with its attendant unusual wear and tear is insignificant? Is this just a remote hypothetical?on the affected units. In the case of a coal-fired unit, Think again. Many utilities have found themselvesemissions will increase as the unit and its pollution in precisely this situation. For this reason somecontrol equipment ramp up during the few hours system operators are now requiring wind turbinesafter startup. Another significant problem with this to be equipped with a “cut out” switch thatchoice is illustrated in the BPA example above. disconnects the wind farm from the grid byIf a base-load source like coal is shut down, and remote control. But it doesn’t end there as the windthen the wind supply suddenly drops, it will not be developers still want to get paid for the energy theypossible to re-energize the base-load facility quickly might have been able to deliver. When they areenough to compensate for the wind loss. This could denied this payment, they sue.13 No matter whatresult in rolling blackouts—which puts us out of the the result of such a suit, there is an obvious waste oftwilight zone and into an electric abyss. money and energy.Another twilight zone choice is to try to sell the THE SHADOW GRID—THE FOSSIL FUEL“hot potato” energy to a neighboring utility or to STAND-IN FOR NO SHOW WINDanother control area authority. What if the neighbor Wind’s unpredictable nature tends to provide energyalready is operating at (or close to) optimum that does not match user demand. As noted in thebalance—a likely scenario. The choices are quite examples of ERCOT, California and the Pacificlimited and whatever option is chosen there will Northwest, wind volatility makes it unsuitable forbe a substantial losses incurred by the wind energy electricity planners to rely on wind energy to meetutility (i.e. ratepayer). peak demand needs. The same applies to base-load and load-following demands. In order to mitigateIn the first case the electricity is sold at a these negative effects, the grid operators andsubstantially discounted price (e.g. what costs the planners must construct a shadow grid, typicallyutility $135/MWh is sold for $20/MWh).11 consisting of fossil-fueled power plants (particularly fast responding gas-peaking units). This shadow gridIn the second case the market price for electricity stands as reserve generation for those times whencan plunge so low that the price actually wind has unexpected supply variations, which aregoes negative. The host utility might actually have inherently a daily matter. This augmentation needsto pay a neighboring utility to accept the surplus to be distinguished from the former backup ofschedule and allow delivery onto its system. conventional sources built into the grid, which wasThis absurd result is a reality in a system that has a primarily designed to deal with demand variations.high percentage of wind generation installed and Expressed another way, conventional sources have acan be very costly to the host utility.12 Capacity Value of well over 90 percent, while wind energy’s Capacity Value is under 10 percent. This is a profound difference.11 FPComment, Ontario’s Power Trip: Power Dumping, http://opinion.financialpost.com/2011/07/20/ontarios-power-trip-power-dumping/.12 K nowledge Problem, Frequent Negative Power Prices in the West Region of ERCOT Result from Wasteful Renewable Power Subsidies, http://knowledgep- roblem.com/2008/11/20/frequent_negati/.13 U nited States of America Before the Federal Energy Regulatory Commission, Complaint and Petition for Order Under Federal Power Act Section 211A Against Bonneville Power Administration Requesting Fast Track Processing, http://www.bpa.gov/corporate/agencytopics/columbiariverhighwatermgmnt/ otherdocs/2011-06-13_complaintagainstbpa.pdf.8 | A RATIONAL LOOK AT RENEWABLE ENERGY
  10. 10. Typically, we build new fossil-fueled The need to develop a shadow grid has also resultedpeaking power plants (usually natural in the actual filing of new tariffs to charge forgas) to augment the wind resources that the cost of such a grid. Puget Sound Electric haswere intended to eliminate fossil-fueled recently filed a tariff with a proposed charge ofresources in the first place. $2.70 per kilowatt-month to offset the carrying cost of a shadow grid of gas turbines that are required toThis duplication of costs is forced onto consumers, stabilize the volatility of wind.14 This can result inwho must pay for both the wind turbine and the an energy charge of one- to two-cents per kWh—back-up generator. or an additional 10 to 20 percent (or more) tacked onto the already high cost of a wind turbine inThe Los Angeles Department of Water order to integrate it operationally into the grid.and Power (LADWP) recognizes the need toback up wind with gas in order to maintain capacity Increase in Carbon Dioxideand reliability. Consider the following statement from Wind Power—It is Possiblefrom the LADWP’s executive summary of its 2010 In addition to the obvious investment and operatingDraft Integrated Resource Plan: cost of the shadow grid, there is another unintended consequence of this fossil-fueled backstop system: “There is ongoing debate regarding the CO2 [carbon] emissions. As discussed above, level of on-peak reliability of renewable a significant penetration of wind turbines into resources. However, the renewable an electric grid can cause base and intermediate resources were added mainly to satisfy resources to be fired up and energized onto the grid Renewable Portfolio Standard (RPS) or dispatched at levels where design efficiencies are target requirements, while natural gas very poor. This results in increased CO2 emissions resources were incorporated to ensure from what might otherwise be expected. system reliability.” Think of it like this: suppose that you were goingIn other words, the LADWP overtly recognizes from one part of a city to another, where you arethat the wind projects on the system are only required to maintain an average speed of 30 mph.meeting the legislatively mandated RPS as they In the base case you take an unobstructed routeprovide intermittent energy. But to actually operate where you can set the car on cruise control ata reliable system, with capacity and energy, 30 mph. Now consider if another equally long roadLADWP must install additional natural gas you took had stoplights and periodically congestedgeneration resources. In spite of the purported traffic—which would result in unpredictable stopsenvironmental objective of wind energy, the shadow or slowdowns. To maintain the same 30 mph yougrid of gas generation will result in consequential would have to rapidly accelerate to 50 mph+ forair emissions, including carbon dioxide. To meet certain intervals. Clearly there would be muchthe fast-acting needs of their wind partner, many higher fuel used in the second case. This is moresuch generators are “simple cycle” peaking units, or less what happens to an electric system thatwhich tend to be less efficient and have the highest attempts to accommodate a high percentage of windemissions among gas-fired generators. resource into the grid.14 E nergy News Data, California Energy Markets, July 16, 2010, No. 1087, 11-12. A RATIONAL LOOK AT RENEWABLE ENERGY | 9
  11. 11. The concept and conclusion is as valid as it populous areas far away from the big cities onis alarming: either coast. This mismatch between resource and population is one of the reasons that developersWind power does not produce all of the of wind energy are challenged to find and exploitclaimed benefits of reductions in fossil locations close to existing high-voltage transmissionfuel consumption and CO2 emissions lines that can carry electricity from wind turbineswhen the fuel consumption and related to big city distribution lines. As more of theseemissions of the shadow grid of gas- locations become occupied, adding more windfired resources are taken into account. generators can only happen in locations where new additional transmission corridors are cleared andThe actual benefits are much less. When real world constructed, to carry out the delivery process fromefficiency losses and additional emissions from the high wind zones to urban centers. Obviously thegas turbines are taken into account, the perceived lack of new transmission adds a significant hurdleenvironmental savings of wind energy are when considering wind development. Too oftengreatly diminished.15 this substantial component of cost gets overlooked in discussing the relative costs and benefits ofGot Transmission? wind energy.Another Missing Cost ElementNo matter where you live, or how windy you think The western continental United States is home toit is, some regions of the country see relatively little eight of the largest states, in terms of land areasustained wind at all. Below is a basic map of the (with Alaska and Texas completing the top 10).United States’ wind energy potential.16 Of these eight states, only California ranks in the top 10 in terms of population. The relatively sparseNote that the regions of maximum wind potential population in the vast, open areas calls for lengthy,(the areas of red, purple and blue) do not coincide and therefore costly, transmission lines. There canwith the areas of dense population. The wind speed also be significant line losses.and duration are generally the greatest in the least Wind Speed m’s 10.0 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 Wind resource data developed by AWS 4.5 Truewind, LLC for windNavigator® 4.0 4.0 Red, purple and blue areas: most economic potential Brown and possible green areas: most technical potential15 B enteck Energy, How Less Became More—Wind, Power and Unintended Consequences in the Colorado Energy Market, http://docs.wind-watch.org/ BENTEK-How-Less-Became-More.pdf.16 U.S. Department of Energy, Wind Powering America, http://www.windpoweringamerica.gov/wind_maps.asp.10 | A RATIONAL LOOK AT RENEWABLE ENERGY
  12. 12. The Electric Continental Divide The Western Interconnect boundary (WECC)The United States electric delivery system does not consists of 30 such “Control Areas” that includeoperate as a single grid, but rather as three separate most of Montana, Colorado, New Mexico andgrids as shown on the map below: all states to the west. The Eastern Interconnect includes everything east of this border, with aboutNORTH AMERICAN ELECTRIC RELIABILTY 120 individual Control Areas. And Texas? I guessCORPORATION (NERC) REGIONS you don’t mess with Texas! They do their own thing down there. These three grids operate independently from one another. Since the three large grids are not in synchronous operation with one another, they cannot be interconnected with one another through traditional “alternating current” (AC) transmission lines. The only possible means of interconnection is through “direct current” (DC) and this is very costly. Consequently there are only six DC ties connecting the Western Interconnect and the Eastern Interconnect in the United States and one additional DC tie in Canada. The capacityBecause electric energy is instantaneously generated of these ties is quite limited (due to cost).and consumed, the operation of these three grids The current interties, being already in service, haverequires a coordinated balancing of generation little excess capacity to move new renewable power.and consumption of power within each grid. Six or seven interties of several hundred MWControl Area Operators (CAOs) perform this will simply not get the job done especially whenfunction, as well as other important tasks, that allow compared to the million megawatt size of ourthe interconnected electric power systems and their electric system.components to operate together both reliablyand efficiently. There are approximately 150 Control The aforementioned transmission challenges createAreas in the nation. Most are run by the dominant a virtual wall between east and west and Texas.large investor-owned utility in a geographic area Unfortunately, the greatest natural source of winddefined by an interconnected transmission grid and energy is electrically trapped in the Midwest. It ispower plant system. The CAOs dispatch generators virtually impossible, or at least very cost prohibitive,from a central control center with computerized to consider transmitting this resource to the west.systems in such a way as to balance supply and It is also impractical and cost prohibitive to transmitdemand and maintain the transmission system this energy to the east coast population centers thatsafely and reliably. This also protects the sensitive are, in some cases, more than a thousand difficultelectronics to which we’ve grown accustomed by miles away.providing a consistent voltage, 24/7/365. A RATIONAL LOOK AT RENEWABLE ENERGY | 11
  13. 13. Just as precipitation will naturally drain within a there is much effort underway to develop newcontinental divide, in similar manner the nation’s storage technologies. An ideal storage mechanismwind energy resources are virtually constrained would be able to economically capture very largeto remain within the three “electric continental quantities of electricity, at a near instantaneous ratedivides” of the West, the East and Texas. of charge and discharge on demand. It would alsoWhat happens in the east stays in the east, be able to safely hold a large charge for long periodswhat happens in the west stays in the west and of time, with little or no losses. Unfortunately,what happens in Texas, stays in Texas. as of today, this dream set of criteria is a fantasy, with nothing on the horizon likely to be evenWind Energy Storage—Not Ready close. Unfortunately, the wave of new, wind andfor Prime time solar energy sources can only function properly inWhat if we could just store the wind energy when it a world that has optimized such near-ideal energyis produced (and not needed) and then call on it in storage opportunities.times of need? Consider this statement from the NorthAmerican Electric Reliability Corporation (NERC):17 It is true that devices have been invented to store bulk electric energy. “Unlike water or gas, electricity These tend to be miniscule in scale, cannot be stored. It must be generated and expensive to acquire and operate. as it is needed, and supply must be kept in balance with demand. Furthermore, One particular state-of-the-art storage device electricity follows the “path of least consists, essentially, of a high-speed flywheel resistance,” so it generally cannot (30,000 rpm) that is suspended (or levitated) above be routed in a specific direction. cleverly designed magnets, resulting in a storage This means generation and transmission that is almost frictionless. Such a device can begin operations in North America must charging (or discharging) in a fraction of a second— be monitored and controlled in real time, it is clearly able to respond to any sudden changes 24 hours a day, to ensure a consistent and in wind or solar output. It is recommended by the ample flow of electricity. This requires the developer of this technology that about 2.5 MWh cooperation and coordination of hundreds of storage capacity (at a cost of $1.4 million) will of electricity industry participants.” accommodate a wind turbine of about one MW name-plate rating. The one MW wind turbine,Storage of electricity would, indeed, answer many according to EIA data,18 would come with anof the operational concerns raised when it comes installed cost of about $2.4 million. Storage,to renewable energy. In the above statement saying in this case, therefore adds about 60 percentelectricity “cannot be stored” would be better to the installed cost of wind. Such a storagephrased as “cannot be practically and economically device is capable of offsetting any unanticipatedstored, at this time.” To fill this perceived need, effects of the wind system for a period of at least17 North American Electric Reliability Corporation, About NERC: Understanding the Grid, http://www.nerc.com/page.php?cid=1|15.18 U .S. DOE Energy Information Administration, Updated Capital Cost Estimates for Electricity Generation Plants, November 2010, http://www.eia.gov/ oiaf/beck_plantcosts/pdf/updatedplantcosts.pdf.12 | A RATIONAL LOOK AT RENEWABLE ENERGY
  14. 14. 2.5 hours, or longer. This gives the utility Wind Turbines can Consume Electricitysystem more opportunity to operate with more One of the little known ironies about utility scalepredictability; it also mates some capacity with wind turbines is that they require an externalthe wind energy. The storage does not, however, source of grid-provided electricity in order to runmitigate wind’s nationwide annual average capacity numerous functions.19 For example, particularlyfactor of only 30 percent, or provide for reliability in cold climates, where much of the best windor dispatch ability on a par with our conventional resources can be found, these units must be heatedsources. to maintain proper viscosity in lubricating fluids and to protect vital components from damage.Based on the above estimates, storage would Even more of a draw is that blades need to beadd about 4.7 cents per kilowatt-hour to the periodically de-iced. When it’s cold in Wyomingcost of wind. It is doubtful that such a system is and up into the Dakota badlands where the calmeconomically justified, but it is quite interesting night air drops to below-zero, it will be the fossil-nonetheless. Such exotic storage systems are based fuel from gas and coal-fired power plants intypically reserved for the most rare of applications— the region that are called upon to warm the massiveremote islands, arctic outposts, etc., or for research wind turbines towering hundreds of feet above theand development pilot projects. None of these windswept plains. These considerations need to betechnologies currently exist with sufficient supply factored into not only wind economics but also theand at a low enough cost to make a meaningful actual amount of CO2 saved.difference to the bulk power system. As we gather more and more real-world data inNot every region or location is suitable for the most the production of wind energy, it is apparent thispromising storage opportunities. It should also resource has a long way to go before becomingbe noted that storage technologies always come a viable contributor to the world’s energy needs.at not only a capital cost to develop and acquire While there may be a worthy role for subsidiesthe storage mechanism, but also an operating cost and taxpayer support of wind, the inescapable factor storage penalty (essentially the execution of is that wind is unlikely to ever be more thanthermodynamic laws). There is always some amount high-cost/low-benefit supplemental resource,of energy loss associated with storage. The flywheel which must be backed up by natural gas plantssystem previously described claims a storage penalty and/or energy storage technology.of about five percent, including transformation,while hydroelectric pumped storage (the onlylarge-scale storage mechanism built to date) requiresabout 30 percent more energy to fill the storagepond than can be extracted upon retrieval.The energy output of storage is always net negative.19 AWEO, Energy Consumption in Wind Facilities, http://www.aweo.org/windconsumption.html. A RATIONAL LOOK AT RENEWABLE ENERGY | 13
  15. 15. Value of Power— certainly also add in the cost of frequently callingDemand versus Energy a taxi when your car stalls half way down the roadCommercial and industrial electric power to town (or on the middle of your cross-countryis typically priced and valued based on two vacation), the cost of the tow truck to return yourcomponents: demand and energy. Demand, or vehicle, the cost of your lost time, the cost of lostcapacity, is the ability to supply electricity at the business, etc. It doesn’t take much thought to seevery instant it is needed. Energy refers to the that a focus on the fuel economy (as an accurateamount of electricity that is actually delivered and indicator of overall cost) is an illusion, as the realused over the course of time (such as a monthly cost of such a vehicle is considerably more than thebilling period, typically in kWh). Depending on the conventional options. And who wants a car thatutility rate structure, the demand charge component only runs on its own schedule?can be as large as or larger than the energycharge component. This reflects the fact that the Some wind proponents say that the reliability willutility must purchase, construct and have available be equivalent to our conventional sources if we joinall of the generating and transmission resources together several wind facilities, spread over a verynecessary to meet peak demand needs, as well as the wide area. In our analogy that would be like buyingcumulative energy needs measured over a period of multiple undependable cars hoping one of thema month or a year. would likely work at any time. Clearly the cost (e.g. multiple insurance policies, etc.) and nuisanceSatisfying demand is closely associated with the would far exceed the expense of one reliable vehicle.notion of dependability or reliability. Empirical evidence doesn’t support this claim either. See the earlier discussion, “Wind is Weak at Peak.”Consumers want the power to be therethe very instant that it is demanded. This “value of power”concept is very relevant to a discussion about renewable energy. A claim mightHaving electricity intermittently available, at be made, for instance, that a certain wind turbineunpredictable times and quantities, is not acceptable can produce power at a cost in terms of cents perin today’s modern electric system. kilowatt hour (kWh) that approximates the cost of coal or gas. But energy cost is only half the story.A practical example will help illustrate this point. The actual value of such power is properly assessedWhen it comes to our automobiles, we have a by considering both the demand and energytendency to demand cars be reliable and to meet provided by any given resource, and theour wants and needs at our beck and call. related dependability.Consider a choice between two automobiles:one gets 50 miles per gallon, but only runs WIND POWER COMPARED TOintermittently about 25 percent of the time; NATURAL GAS POWERthe other car gets about 20 miles per gallon, Suppose we are weighing the pros and cons of abut it runs all of the time. How would you value decision to add a new resource. The two finalistseach of these cars? are (1) a 100 MW wind farm, and (2) a 100 MW combined-cycle gas plant. Our goal is to offsetIf the first car had low fuel cost, but also low emissions from an existing 100 MW coal plantreliability, how much would you pay for such a car? that is nearing retirement.Note that when considering the cost, you would14 | A RATIONAL LOOK AT RENEWABLE ENERGY
  16. 16. Our target coal plant is currently operated a little Since our target coal unit operates with 67 averageless than 67 percent of the time, primarily due to MW, the maximum replacement (albeit highlythe fact that the coal plant is dispatched, or operated unlikely or even impossible) that we can get from theto conform to load requirements. Our plant runs wind farm is 30 of the 67 average MW, or aabout the same as the entire fleet of U.S. coal plants. 45 percent displacement. We are unavoidably left withIt normally runs at very high availability rates—in at least 37 average MW of coal that must operateexcess of 90 percent—but is scheduled to match when the wind is not blowing. But, in fairness,consumer load requirements. Our prospective gas 45 percent coal displacement is an extreme upperplant can easily take the place of the retiring coal unit limit of what wind can do. In actuality, the wind farmas it performs with similar characteristics. It can be is likely to produce some of its energy (probably aboutsited flexibly near load. It depends on a combustion 30 percent of the time) during periods when the coalresource and therefore can be operated to match the plant is not operated at all. In addition, wind tendsconsumer load requirements at any time. to be strongest during off-peak periods—the same periods when the coal plant is most likely to be scaledOur prospective wind resource on the other hand, back to minimum load. And, if the wind energycan only be scheduled according to the operator’s best comes on during these periods, the coal plant cannotestimate of forward-looking weather patterns. The be further reduced, unless it is taken completely offprimary challenge of wind is the intermittency of the line (which would subsequently incur restart costssupply. Unless it is backed up with another peaking and ramping challenges). Therefore, the wind willresource (usually gas turbines), we cannot use the most likely displace some other resource, such as gaswind by itself as a complete coal replacement. But, the or possibly hydro. Obviously, the carbon reductionsobvious allure of wind is the promise of low emission from wind will be much lower, or even nonexistent,electricity (e.g. carbon dioxide, sulfur dioxide, nitrous in this circumstance. So, the range of coaloxides, particulates, etc.). displacement that can be accomplished by wind is somewhere in the range of 20 percent to 45 percent,So, how well does wind energy perform? with 30 percent as a reasonable central estimate.The national average output efficiencyfor wind is slightly less than 30 percent. Now, let’s consider the gas alternative.This means, for instance, that a 100 MW wind farm Combustion of natural gas produces aboutwill average only 30 MW of output. This might come in 45 percent less carbon dioxide emissions thanthe form of near 100 percent output for 30 percent of combustion of coal. Further, gas units are morethe time, and 0 percent output during nearly 70 percent efficient than their coal counterparts, and when thisof the time, with numerous iterations in between. But, efficiency is accounted for a combined-cycle gasover time, the average will be about 30 MW. Thus, the plant can be expected to produce about 60 percentability of wind to displace our coal is energy limited less carbon dioxide than coal.20because of the intermittent actual output.20 According to the Energy Information Administration, as a national average, coal-fired power plants emit 210.6 lbs. CO2 per MMBTU. Gas, on the other hand, emits 117 lbs. CO2 per MMBTU. So, comparing these two fossil fuels, gas emissions are about 55 percent of the coal counterpart. This means that there would be a 45 percent savings—but this is only part of the story. These emission rates are further adjusted by equipment efficiency. The term used in industry for this efficiency is heat rate, where a lower heat rate constitutes a more efficient machine (the machine produces more electrical energy per unit of heat or BTU input). The national average heat rate for coal is 10,355 BTU per kWh, while the national average heat rate for gas is 7,620 BTU per kWh. When we combine heat rate with the emission attributes, we find that natural gas-fired emissions are about 59 percent lower than coal-fired emissions per unit of energy, as shown in this formula: [ (210.6 X 10,355) – (117 X 7,620) ] / (210.6 X 10,355) = 59 percent. When we employ the same formula for an efficient combined-cycle unit with a heat rate of 7,000 BTU/kWh, can demonstrate CO2 reductions of 64 percent. A RATIONAL LOOK AT RENEWABLE ENERGY | 15
  17. 17. Because gas can be run essentially all of the time, profile as a viable replacement for coal. If theand can easily ramp up or down to match consumer national policy is to reduce emissions fromloads, it is a natural operational substitute for coal. coal-fired units, policy makers should abandonTherefore, if we were to install a new 100 MW so-called “renewable portfolio standards” thatcombined-cycle, gas-fired plant (instead of the 100 mandate wind and solar, and consider policiesMW wind farm in our example), this would result to promote the use of natural gas as thein a net carbon dioxide emissions reduction of preferred alternative.about 60 percent, compared to coal. Solar EnergyConsider the significance of this analysis. There is significant national and internationalWith the same installed capacity, a combined-cycle interest in the development of new solar energy.gas turbine can provide net reductions in carbon While this is obviously an energy source that hasdioxide that are greater than wind—approximately more applicability to regions with high levels ofdouble the benefit. In addition, wind is significantly sunshine, it is a promising technology formore costly when compared to gas. A new wind many reasons. The single greatest challenge to solarfarm can be expected to have an installed cost that power is the immutable fact that the sun is onlyis about double the price of a new combined-cycle available, at best, half of the time, no matter howgas turbine.21 ideal other conditions may be.Emissions Reductions using Natural Gas to Replace Coal A well-designed and situated solar Nitrogen Oxides -85.20% project will typically provide available Particulates -99.80% energy about 20 percent of the time. Sulfur Dioxide -100% At this low availability, solar energy Mercury -100% can never be more than a supplement (Assumes gas heat rate of 7,620 BTU/kWh and coal heat rate of 10,355 BTU/kWh) to a larger portfolio of power generating resources.In the above example, we were focused solely oncarbon dioxide. How does the example pan out And like wind, solar energy begs for supplementalin terms of other pollutants? Gas performs even storage in order to provide a degree of reliabilitystronger when considering other forms of emissions. to the grid.Gas yields an 85 percent reduction in nitrogenoxides versus coal. And, gas-fired power plants have Not All Sunshine is Equalvirtually no particulate emissions, and no sulfur Photovoltaic cells, or PV solar, are by far the mostdioxide or mercury, compared to coal.22 So, in these common application for electric generation fromregards, it is impossible for wind to outperform gas. solar energy. Although there are other forms of solar renewable projects, given the availability andConclusion: gas outperforms wind in all popularity of PV, we will focus on it first.emission-reduction categories when balanced PV panels are made from materials such asagainst average wind performance across the crystalline silicon and cadmium telluride, whichU.S. Further, gas can run with the same reliable convert photons from the sun’s rays into electric21 Calculations based on JEDI model found in U.S. Department of Energy, Wind Powering America, www.windpoweringamerica.gov/economics_jedi.asp and www.20percentwind.org.22 Energy Information Administration—EIA, Voluntary Reporting of Greenhouse Gases Program Fuel Emission Coefficients, 2012.16 | A RATIONAL LOOK AT RENEWABLE ENERGY
  18. 18. energy. To make use of the energy produced by the bottom line energy cost to the consumerthese cells, an inverter is attached to a PV array would be in the range of 15 to 20 cents per kWh.to create alternating electric current. Some PV To be competitive, solar would need to cut evenpanels are small, roof top applications, and a few further, probably another 50 to 70 percent beloware larger, utility scale facilities. PV solar panels even these levels.have no moving parts. Hence the operations andmaintenance consists largely of a careful cleaning Large PV Solarfrom time to time with glass cleaner. But even a Three years ago the much-publicized PV solarvery large PV solar project will have a fairly facility at Nellis Air Force Base was the largest suchmodest output. facility in North America, and the third largest in the world. It sits on 140-acres and produces aboutThe entire United States’ output of PV 30,000,000 kWh per year. Yet this amount ofsolar for the year 2009 was 807,988 production is only equivalent to one day’s outputMWh, about one-tenth of one percent of a 1,200 MW coal-fired plant. If we were toof the U.S. nuclear output. attempt to replace the entire fleet of coal-fired electrical generation in the United States withHow expensive is PV Solar? large PV solar projects, we would have to installApart from the day/night cycle of solar power, a Nellis-sized facility each month for each of thewhich can’t be avoided, another disadvantage next 5,000 years.24 Indeed we are a long way fromof PV solar is its high cost. California provides accomplishing much with PV solar energy.robust rebates and incentives under its California With growth in the solar industry, there are nowSolar Initiative and has produced some valuable three other PV solar facilities in the United Statesbenchmarks for the cost of solar power. that are larger than the Nellis facility, and 40According to a study produced for the California larger PV facilities in the world.Public Utilities Commission in 2009, the priceof installed PV under the California program Given the inefficiencies of scale associated with PVaveraged $7,090 per kW for large industrial solar, it is not realistic to envision the entire electriccustomer installations, and $8,490 per kW for system consisting solely of such distributed units.residential installations.23 Assuming a 20 percent Homes cannot run entirely from PV solar panelscapacity factor, a cost of capital of six percent without some form of backup or battery storage.and a life of 25 years, the cost per kWh of these Even large arrays on commercial buildings areinstallations would run from 32 to 38 cents per kWh. almost always tied into the electric grid because of the various shortcomings in PV systems, and largeThis example helps to explain why solar energy scale utility systems require enormous tracts of landis only a miniscule resource in the United States. while providing only modest energy output.Still, solar is a growth industry and significantimprovements in both design and cost Concentrated Thermal Solarare forthcoming. Indeed there are anecdotal PV technology directly converts solar energy intoevidences of less costly solar installations—as little electrical energy through panels. Concentratedas $4,000 per kW—but even at that installed cost, thermal solar, on the other hand, uses parabolic23 Energy News Data, California Energy Markets, July 2, 2010.24 ased on the United State’s coal-fired electrical generation of 2 billion MWh per year compared to Nellis’ advertised annual output of 30,000 MWh B per year. (2 billion / 30,000) / 12 months = 5,555 years.) A RATIONAL LOOK AT RENEWABLE ENERGY | 17
  19. 19. mirrors, or similar technology, to focus solar energy very predictable, barring the unpredictable effectsinto heating a fluid that then goes through a heat of intermittent cloud cover. If solar panels aretransfer process that is not unlike a traditional spread over a wide enough area, some of the cloudgas- or coal-fired steam electric turbine. In fact, cover effect can be mitigated through diversity.many concentrated solar facilities will have natural However, even in the best case the peak solar outputgas-based generation as a backup or supplement. tends to occur prior to the time of peak load for theConcentrated solar installations tend to cost around utility shown in blue.two-thirds, or less, compared to the cost of a St. George City peak loadPV installation. This is a significant step in the right vs. SunSmart’s output:direction, but still very expensive power compared 100%to traditional base load resources. 90% RATIO: HOURLY TO PEAK 80% 70%Nevada Solar One boasts one of the newest and 60% Solar output falls off daily just as loadlargest concentrated solar facilities in the 50% 40% in St. George peaksUnited States. This project delivers 64 MW of 30% 20%capacity and approximately 134,000 MWh of 10%energy per year. Gilbert Cohen, vice president of 0% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Engineering and Operations for Solargenix, said the HOUR OF PEAK, JULY 17, 2009 Dixie’s St. George Loadproject installation costs are somewhere in the range SunSmart Outputof $220 to $250 million. At that price, the power ismore expensive than most wind power projects, but Based on the graphic, about 30 to 60 percent ofless expensive than typical PV projects. Energy from the solar peak was useful during the peak hour.Nevada Solar One currently costs about 13 cents The fit between solar-produced energy and theper kWh.25 The developers of Nevada Solar One demand curve for electricity usage is not as close asbelieve that a target of seven cents per kWh will be one might expect, and certainly not as good as oneachievable in the future. At that price, concentrated would hope. This can be significantly mitigated andsolar would be fairly competitive as a viable, utility improved if the solar project is combined with angrade source of power. energy storage facility, but that would add to the cost. A similar result can be demonstrated across a muchSolar Demand versus System Peak larger system. Last year, CAISO reported peakA desirable attribute of solar energy is that it is demand of 45,994 MW, which occurred at 3:00 p.m.produced during hours that roughly coincide with on September 3, 2009. In that hour, even thoughutility system peak loads. The coincidence is not California had installed PV capacity of nearlyperfect, but much better than wind. The chart in 250 MW, that was operational and online on thethe following column is an actual output profile of state’s electric grid, only about 144 MW of solarthe 100 kW SunSmart project in St. George, Utah. energy was being generated to help serve the peak demand, or around 58 percent of the amount thatThe solar output shown in yellow tends to ramp the installed solar units were capable of producing.up around 10 a.m. and then ramps down in By contrast, as of noon that same day, the PVthe afternoon. The shape of the output curve is solar units reached their maximum capacity factor at about 72 percent, which is the typical peak25 Jesse Broehl, Renewable Energy World.Com, A New Chapter Begins for Concentrated Solar Power, (2006), www.renewableenergyworld.com/rea/news/ article/2006/02/a-new-chapter-begins-for-concentrated-solar-power-43336.18 | A RATIONAL LOOK AT RENEWABLE ENERGY

×