Integrating wind power
When trying to couple a wind farm to a transmission system, one of the first technical problems encountered is that wind’s load profile is usually out of sync with what the system needs. Though the typical utility’s summer weekday peak demand occurs between 1 p.m. and 5 p.m., a wind farm’s production typically peaks in the late evening and early morning. This situation leads both to underproduction during peak demand periods (which utilities must address by dispatching peaking plants) and overproduction at other times.
There are likely to be a few hours during the year when a wind farm sends a utility more power than it needs. In such cases, the utility may need to reduce thermal generation or use the excess power to “recharge” the reservoirs at its pumped-storage plants (if it has any). Otherwise, the utility may have to sell the excess power on the spot market at a loss. Because utilities cannot depend on wind power being available during peak-demand periods, they typically have had to bring on-line gas turbine–based peaking plants at those times. Now, however, some may have another option for filling the gap: dispatching both wind farms and photovoltaic power plants. Such a combination may prove a less expensive way to keep the lights on.
The average production of a typical wind farm has a very different profile than its hourly generation. An examination of hourly generation makes clear that wind production is much more erratic than one might expect. Though a wind turbine typically runs 60% to 80% of the time, it typically runs at 100% capacity only 10% of the time. This intermittent characteristic of wind generation can weaken the stability of a grid. A good rule of thumb is that if the rated capacity of a wind farm exceeds 2% of the “fault duty” at its point of grid connection, measures should be taken to minimize the impact of dispatching the farm on the grid’s power quality. In such cases, dynamic compensation is usually the preferred measure.
Historically, utilities have been able to integrate wind power up to 20% of their portfolio’s total generating capacity. More than that can lead to stability problems on the transmission lines. The 20% figure is just a rough estimate, however, because utilities with larger hydro capacity can accommodate much more. For example, hydro-rich Denmark successfully integrates 60% wind into its grid.
Furthermore, due to wind’s intermittent nature, utilities must reserve the full capacity of a wind farm on the transmission system, although they can only reasonably expect it to have a capacity factor of 30% to 40%. This situation may lead to lost revenues from transmission capacity sales. How much wind energy a particular utility can handle depends on a number of operational factors, including: the amount of hydro generation, fast-starting generation, and spinning reserve available; the minimum loading and loading rates of fossil-fuel plants on the system; and the system’s load shape.
Low rider
An unresolved issue in the area of wind integration is how to meet FERC’s new Order 661A, which requires wind farms to remain on-line during system disturbances in which system voltages drop to zero volts. This requirement is much more stringent than that of Order 661, which required a low-voltage ride-through of 0.625 seconds. Stated another way, Order 661 required a wind farm to demonstrate its ability to remain on-line for 0.625 seconds after the system voltage dropped to 15% of normal due to a system disturbance. This requirement ensured that the wind farm would be available to support the system when there was a normally cleared fault on a single element, which typically took four to eight cycles.
Order 661A recognizes that a fault occurring near a wind farm could cause the voltage at the point of interconnection to fall to zero during clearing, and requires wind farms to remain on-line during that time. Because clearing a fault normally takes four to eight cycles, wind plants now are required to remain on-line for nine cycles, or 0.150 seconds, at zero volts.
Though SVCs and inverter-based VAR support schemes could be developed to meet the low-voltage ride-through requirement of Order 661, the zero-voltage ride-through requirement of Order 661A is problematic for those wind turbines that do not use inverters in each nacelle. The solution may require the use of energy storage devices such as ultra-capacitors.
—Robert D. Castro (robert.castro@alumni.usc.edu) teaches graduate-level power classes at the University of Southern California and negotiates wind generation contracts for a local utility.
Email
Comments
Print
Reuse
