Using Genetic Fuzzy Algorithms to Model the Evolution of Climate Variables at San Jorge Gulf Area
MICAI '08 Proceedings of the 7th Mexican International Conference on Artificial Intelligence: Advances in Artificial Intelligence
Hi-index | 0.00 |
Uncertainty and variability in the wind resource create obstacles for the participation of wind power in forward markets, such as regional day ahead electricity markets. Studies performed in various states have developed methods to improve wind forecasting and so reduce the inherent uncertainty in a day ahead schedule for wind power generation. This paper addresses the issue of the variability in wind power generation by estimating the next ten-minute production level for a hypothetical wind farm, and then dispatching additional dedicated resources, such as responsive load or a gas turbine, in order to reduce the net variability of the generation in the next ten- minutes. Historical wind data from ISO-ne are used with an auto-regressive moving average model to develop the next ten-minute forecast. Preliminary results estimate the capacity required for the dedicated resources to maintain the wind output within a specified percentage of the submitted day ahead schedule.