Variable electrical generation (VG) sources such as wind farms are an increasing percentage of total electrical generation in the Bonneville Power Administration (BPA) balancing area and are starting to impact the ability of the regional balancing authority to control the electric grid. Wind farms are not dispatchable and challenge historical electric grid control methods. Successful integration of VG at high penetration levels of wind needs to address increased overall system variability and the rapid power ramp rates caused by wind. One of the new control paradigms needed is accurate wind speed prediction which directly relates to wind farm power output. With an accurate wind speed forecast other generation sources can be dispatched as needed to ensure grid stability. This work uses BPA metrology station (MS) data to make predictions for short-term wind speed where short-term is defined as a one hour prediction horizon. It is shown that, using the available metrology station data and several different prediction methodologies, only small improvements in short-term wind speed prediction can be achieved with the available data for the algorithms analyzed. / Graduation date: 2013
Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/30474 |
Date | 15 June 2012 |
Creators | Davidson, James D. (James Douglas) |
Contributors | Brekken, Ted K. A. |
Source Sets | Oregon State University |
Language | en_US |
Detected Language | English |
Type | Thesis/Dissertation |
Relation | Lower Columbia Explorer |
Page generated in 0.0427 seconds