Texas has, by far, the highest penetration of wind among the three main US interconnections (Eastern, Western, and Texas), and the Electric Reliability Council of Texas (ERCOT) has met the challenges of wind integration. ERCOT is set to get even more wind power, which will present even greater challenges, because of the relationship between wind production and electricity demand.
Several of my students, most recently Dr. Duehee Lee, have investigated the statistics of wind production to understand this relationship. Recent joint work with Lee (reported at Wind Farms in Dallas in May, and more recently at the Austin IEEE Power and Energy Society chapter meeting and the Berlin Conference on Energy and Electricity Economics) analyzed the relationship between wind and load variation at various timescales. We want to understand how the variability of wind at various timescales will affect the operation of ERCOT, and what resources will therefore be necessary to compensate for this variation, whether it is agile generation or battery storage.
West Texas inland wind presents a particular problem, because it tends to be negatively correlated with load, while near-coastal wind is weakly correlated and therefore a better match to load. This means that increased amounts of inland wind will not offset the need for other generation capacity to cover the peaks in demand. Moreover, this generation capacity will also need to be increasingly flexible to accommodate wind production that occurs off-peak.
Furthermore, while aggregating large numbers of farms tends to smooth intra-hour fluctuations of wind, in contrast, the longer term fluctuations of wind, and particularly the negative correlation of inland wind and load will not be “solved” by aggregation of West Texas wind.
Coastal Texas wind has better correlation with load from this perspective, but it may be more difficult to build much more coastal wind due to environmental and other considerations. An open question is how much more wind can be accommodated in the ERCOT market before we need to build large-scale storage.