Fundamental Issue: Intermittency
Despite robust wind development in the U.S., wind faces a nearly insurmountable issue: intermittency. Simply put, the intermittent nature of wind makes it difficult to harness effectively on a power grid that is finely tuned to deliver electricity around the clock. The down side of this intermittency is clearly evident in the actual performance data of wind turbines already installed. Wind performs poorly across all traditional utility metrics for generating resources. For reliability, stability, forecast ability, proximity to load centers, and economics, wind power is a poor choice for large-scale power production.
Nameplate Rating versus Actual Energy Delivery
For large utility grade generators the customary expectation is that once installed, they will deliver the name-plate output when supplied with sufficient fuel. Additionally, they will operate, if required, around the clock. In the case of wind energy installations this is simply not the case. The output over time is only a small fraction of name-plate rating because of the intermittency of the fuel resource. The ratio of actual output divided by maximum potential output is defined as capacity factor. The entire sector of U.S. wind energy is currently operating at a capacity factor of only 25 percent.
Wind Is Weak at Peak
The intermittent and unpredictable nature of wind is further compounded by the fact that the wind tends to be weak during electrical peak load conditions. Wind blows most consistently and creates the best generation opportunities during off-peak hours, cooler days and evening hours; directly opposite the electric customer usage profile. This is a natural consequence of the climate forces that determine wind: daily and seasonal temperature differentials. On the hottest days of the summer the wind tends to be low or non-existent when air conditioning demands are at their peak. Then when it gets windy, the temperatures will naturally moderate and air conditioning loads drop off just in time for the wind energy to pick up. Therefore, during the summer months, wind generation is low during high demand times, and can be shown to reach maximum generation when power demands are down. The same phenomena can be demonstrated to occur during winter peak conditions. The very coldest days are also the days when the wind is not blowing. For this reason, utility-scale balancing regions simply do not plan for significant contribution of wind at peak demand periods.
Texas is home to the largest collection of wind generation facilities in the nation. More than one out of every four wind turbines in America is found in Texas. The Electric Reliability Council of Texas (ERCOT) only plans for 8.7 percent of wind name-plate rating as the “dependable contribution to peak requirements,” in accordance with ERCOT’s stakeholder-adopted methodology. This means that more than 91 percent of Texas wind turbines are expected to be off-line when it matters most—at peak load periods.
The State of California ranks third in the U.S. for total installed wind energy (behind Texas and Iowa). California is also the third largest state geographically (behind Alaska and Texas). … The wind capacity available at California peak demand times is about 200 MW. The name-plate capacity of California-based wind generators is about 2,600 MW. Hence, the wind power available at peak is less than 10 percent, which is very similar to the Texas experience. In other words, about 90 percent of California wind turbines are idle at peak load conditions.
Oregon and Washington rank fourth and fifth in the U.S. for total installed wind energy. The prominent Federal Power provider in the region—the Bonneville Power Administration (BPA)—is a winter-peaking system with about 10,000 MW of load. On Tuesday December 16, 2008, the BPA system reached its peak for the entire year, with a demand of 10,762 MW. At the time of peak demand, the output of the entire fleet of wind resources, with a name-plate value of 1,599 MW, was only 116 MW, or about seven percent of the name-plate potential. This is very similar to the Texas and California wind experience, only in this case about 93 percent are not producing at the winter peak.
Western United States
Now let us consider an even broader region— all eleven western states, from Montana to New Mexico, from Washington to California, and everything in between. This vast area is served as a single “reliability” region known as The Western Electricity Coordinating Council (WECC). During the heat wave of July 2006, the WECC system reached its peak on Monday, July 24, 2006. The hottest day was actually July 23, 2006, but this was a Sunday so total loads did not peak until Monday. On the hottest day, the capacity factors for wind resources through most of WECC were well under five percent, and on the peak day, which was a slightly cooler day, the wind capacity factors were less than ten percent.
These real-world lessons illustrate the grave shortcomings of wind. Approximately 90 percent of wind turbines can be expected to NOT PRODUCE power at peak load periods, even when distributed over broad geographic areas.
Enter the “Twilight Zone”—A Control Area Nightmare
The demonstrated low performance of wind energy during peak load conditions is only one side of the coin. The other side occurs during off-peak periods when unscheduled, unanticipated wind energy comes booming onto the system ready to serve loads that are nowhere to be found.
This can easily happen because of the physics of wind energy: the power output of a wind turbine accelerates at a much faster rate than the simple change in wind speed. For instance, if the wind speed changes from 10 to 20 mph (a doubling of the wind speed) the associated power output will change by a factor of eight.
An actual case with the BPA brings the control area problem into perspective. On April 27, 2010 about 3:00 a.m., wind generation on the BPA system ramped up by 1,200 MW in only one hour, and then down 800 MW in only 20 minutes. … Such erratic changes in generation run directly counter to the needs of utility operators who select from a pool of different traditional generators to provide the right amount of power at the instant it’s required. In a normal day they blend the outputs of traditional power plants that include coal, nuclear, natural gas, and in some regions hydroelectric to work in concert to minimize operating costs while maintaining reliability.
Consider an event that occurs during off-peak or twilight hours. The various utilities are operating with all of the peaking plants off line and many of the intermediate resources off line. Still running are base-load, coal-fired generators but they have been reduced to minimum-load status. The nuclear plants are running because they remain in “must-run” condition for safety and economic reasons. The wind turbines are cruising along at a modest output. Now assume that a sudden, unanticipated, change in the weather brings with it a rapid ramping of wind energy output. This can result in a large block of several thousand MW of unplanned energy that when combined with the operating status just described, that can easily swamp out the total load requirements of the utility—meaning there’s literally no place for the energy to go. Now the utility is forced to make quick and drastic decisions to balance loads and resources.
I call this the twilight zone—a control area no-man’s land. One option might be to enact the costly decision to shut down a base-load resource, such as a nuclear or coal unit, and then subsequently face a high cost “re-start” with its attendant unusual wear and tear on the affected units. In the case of a coal-fired unit, emissions will increase as the unit and its pollution control equipment ramp up during the few hours after startup. Another twilight zone choice is to try to sell the “hot potato” energy to a neighboring utility, or to another control area authority. What if the neighbor already is operating at optimum balance? … The host utility might actually have to pay a neighboring utility to accept the surplus schedule and allow delivery onto its system.
Many utilities have found themselves in precisely this situation. For this reason some system operators are now requiring wind turbines to be equipped with a “cut out” switch that disconnects the wind farm from the grid by remote control. This becomes an obvious waste of energy.
The Shadow Grid—The Fossil Fuel Stand-In for No-Show Wind
Wind’s unpredictable nature tends to provide energy that does not match consumer demand. As noted in the examples of ERCOT, California and the Pacific Northwest, wind volatility makes it unsuitable for electricity planners to rely on wind energy to meet peak demand needs. In order to mitigate these negative effects, the grid operators and planners must construct a shadow grid, typically consisting of fossil- fueled power plants (particularly gas-peaking units). This shadow grid stands as reserve generation for those times when wind resources are not delivering their potential capacity.
Effectively, we end up building new fossil-fueled peaking power plants (usually natural gas) to back up the wind resources that were intended to eliminate fossil-fueled resources in the first place.
Los Angeles Department of Water and Power
[T]he LADWP overtly recognizes that the wind projects on the system are only meeting the legislatively mandated RPS as they provide intermittent energy. But to actually operate a reliable system, with capacity and energy, LADWP must install natural gas generation resources. In spite of the obvious environmental objective of wind energy, the shadow grid of gas generation will result in air emissions, including carbon dioxide. Many such generators are “simple cycle” peaking units, which tend to be less efficient and have the highest emissions among gas-fired generators.
Increase in Carbon Dioxide from Wind Power—It Is Possible
In addition to the obvious investment and operating cost of the shadow grid, there is another unintended consequence of this fossil-fueled backstop system: carbon emissions. As discussed above, a significant penetration of wind turbines into an electric grid can cause base and intermediate resources to be fired up and energized onto the grid or dispatched at levels where design efficiencies are very poor. This results in unintended carbon emissions.
Think of it like this: Suppose that you were to go on a road trip where you are required to maintain an average speed of 60 mph. In the base case you do this by setting the car on cruise control. Now imagine an outside influence that requires you to suddenly stop, and then rapidly accelerate to 120 mph, and to do so at unpredictable intervals, all the while you are required to average 60 mph. Can you imagine the fuel economy differences between these two cases? This is more or less what happens to an electric system that attempts to accommodate a high percentage of wind resource into the grid.
Got Transmission? The Missing Cost Element
[T]he regions of maximum wind potential (the areas of red, purple and blue) do not coincide with the areas of dense population. The wind speed and duration are generally the greatest in the least populous areas far away from the big cities on either coast.
Technical Potential versus Economic Potential
This means that, in terms of their operating characteristics, and even for the best wind resources, the grid must be designed and operate as if 60 to 75 percent of the time a typical wind turbine produces very little or nothing at all.
The Electric Continental Divide
In addition to the obvious transmission challenges of renewable energy, there is a virtual wall between east and west. Unfortunately, the greatest “economic potential” of wind energy is electrically trapped in the Midwest. It is virtually impossible, or at least very cost prohibitive, to consider transmitting this resource to the west. It is also impractical and cost prohibitive to transmit this energy to the east coast population centers that are, in some cases, more than a thousand difficult miles away.
Wind Energy Storage—Not Ready for Primetime
Storage of electricity would, indeed, answer many of the operational concerns raised when it comes to renewable energy. The notion that electricity cannot be stored is not entirely accurate, and in fact, there is much effort underway to develop new storage technologies. An ideal storage mechanism would be able to capture unlimited quantities of electricity, at a near infinite rate of charge and discharge on demand. It would be able to hold a charge for long periods of time and would be free, or at least very inexpensive to install and operate, with little or no losses. Unfortunately, as of today, this dream set of criteria is a fantasy, although there is an obvious need for energy storage technology. An effective wave of new, renewable energy can only function properly in a world that is ripe with near-ideal energy storage opportunities.
It is true that devices have been invented to store bulk electric energy. These are all minuscule in scale, and expensive to acquire and operate. … It should also be noted that storage technologies always come at a cost—both a capital cost to develop and acquire the storage mechanism, as well as an operating cost or storage penalty (essentially the execution of thermodynamic laws). There is always some amount of energy loss associated with storage. The flywheel system previously described claims a storage penalty of about five percent, including transformation, while hydroelectric pumped storage requires about 30 percent more energy to fill the storage pond than can be extracted upon retrieval. The energy output of storage is always net negative.
Wind Turbines Can Consume Electricity
One of the little known ironies about utility scale wind turbines is that they require an external source of grid-provided electricity in order to run properly. Particularly in cold climates, where much of the best wind resources can be found, these units must be heated to maintain proper viscosity in lubricating fluids and to protect vital components from damage. When it’s cold in Wyoming and up into the Dakota badlands where the calm night air drops to below zero, it will be the fossil-based fuel from gas and coal-fired power plants in the region that are called upon to warm the massive wind turbines towering hundreds of feet above the windswept plains.
The Hard Realities of Renewable Pricing—Value of Power—Demand versus Energy
Having electricity intermittently available, at unpredictable times and quantities, is not acceptable in today’s electric system. A practical example will help illustrate this point. When it comes to our automobiles, we have a tendency to demand cars be reliable and to meet our wants and needs at our beck and call. Consider a choice between two automobiles: one gets 50 miles per gallon, but only runs intermittently about 25 percent of the time; the other car gets about 20 miles per gallon, but it runs all of the time. How would you value each of these cars? If the first car had low fuel cost, but no reliability, how much would you pay for such a car, and are you prepared to call a taxi when your car stalls half way down the road? If the value of a car is based, shall we say, half on fuel economy and half on reliability, then the market value of the intermittent car will be intrinsically lower because it fails to meet the primary purpose of reliable transportation. Who wants a car that rarely runs?
This concept is very relevant to a discussion about renewable energy. A claim might be made, for instance, that a certain wind turbine can produce power at a cost of 8 cents per kilowatt hour (kWh). But cost is only half the story. The actual value of such power is properly assessed by considering both the demand and energy provided by any given resource.
Download original document: “A Rational Look at Renewable Energy ”