The Cost of Wind Power Variability

The cost of wind power variability

Warren Katzenstein, Carnegie Mellon University, Phone +1 412-390-6550, E-mail: ay Apt, Carnegie Mellon University , Phone +1 412-268-3003, E-mail:

Overview

In this paper we develop a cost metric to value the variability of individual wind plants and then show its use in valuing reductions in wind power variability. System operators have recognized integrating wind power into their systems increases their operational costs because they need to “secure additional operating flexibility on several time scales to balance fluctuations and uncertainties in wind output” (Northwest Power and Conservation Council, 2007). There is interest in using storage technologies or fast-ramping fossil fuel generators, deemed flexible resources, to mitigate, or smooth, wind power variability and decrease the costs of integrating wind power into electrical systems. Previous research and wind integration studies performed by Independent System Operators (ISOs) and Regional Transmission Operators (RTOs) have estimated the cost of integrating wind power ranges from $0.5 to 9.5 per MWh for wind penetration levels ranging from 3.5 to 33% (). The methods used to estimate the integration costs of wind energy are not suitable to evaluate reductions in wind power variability for individual wind plants because they use complex priopriatary methods and typically estimate future wind penetetration scenarios.

Methods

A statement made by Keith and DeCarolis in their article “The Costs of Wind’s Variability: Is There a Threshold?” serves as the inspiration for the cost metric detailed below. Specifically, they state that it should be “possible…to assess the overall cost of wind’s intermittency” by “portioning the cost of wind’s variability between various markets…and market participants.” Here we present an unbiased method to partition wind energy between various markets (hourly and subhourly) and use the corresponding market prices to determine the cost of wind’s variability.

Equation 1 is the general formulation of the hourly value of wind energy from a system operator’s point of view described in figure 1. For each hour the value of wind energy is the value of its hourly energy plus its capacity value minus its variability costs. The yearly value of energy from a wind plant is the sum of its hourly values. Equation 1 can be modified to represent any electricity market in the United States. In the next section we modify equation 1 for ERCOT markets.

and

Where PH is the hourly price of energy

Pk is the subhourly price of energy

PC is the hourly price for capacity

PAC-UP is the subhourly price for up regulation capacity

PAC-DN is the subhourly price for down regulation capacity

qH is the amount of firm hourly energy bid

qC is the amount of hourly capacity bid

, the amount of subhourly energy per time period k

In order to create an unbiased metric, we find the qH that minimizes the variability costs based on the prices and wind power data.

Minimize:

Wherek = 1:n

Subject to , k = 1:n

k = 1:n

k = 1:n

We use 15-minute time sampled wind power data from 20 ERCOT wind plants in 2008 and 2009. In addition, we use 15-minute ERCOT balancing energy service (BES) price data and hourly load following and regulation capacity price data for years 2004 through 2009.

Results

Figure 1 - Wind plant variability costs decreases as the capacity factor of a wind plant increases.

Figure 2 - Variability costs decrease as more wind plants are interconnected. Only 8 wind plants need to be interconnected to achieve the majority of the reductions in variability costs.

Conclusions

We developed a cost metric capable of estimating the variability cost of individual wind plants and found it produces results similar to integration studies produced by the major electricity market operators in the United States. Variability costs decline as the capacity factor of wind plants increase. We find that a set of individual wind plants respond equally to price spikes in ancillary service markets and the relative ranking of wind plants based on variability costs is dependent on the wind power produced from the wind plants. If electricity systems recover the integration costs of wind energy from the wind plants in its system, then it should consider implementing a tariff based on the capacity factor of the wind plants. They should also ensure the amount charged is based on the prices for the current year as variability costs can vary significantly from one year to the next. Finally, they should also offer a reduced tariff for wind plants that install systems to mitigate their variability.

References

Northwest Power and Conservation Council (2007). The Northwest Wind Integration Action Plan. Available:

Ryan Wiser andMark Bolinger (2008). 2008 Wind Technologies Market Report. US DOE. Available:

Joseph F. DeCarolis and David W. Keith (2005). The Costs of Wind's Variability: Is There a Threshold?The Electricity Journal,18: 69-77.