Wind power statistics in Britain, and their consequences for the integration of wind generation into electricity grids

Lowe, Robert John (1984). Wind power statistics in Britain, and their consequences for the integration of wind generation into electricity grids. PhD thesis The Open University.

DOI: https://doi.org/10.21954/ou.ro.0000de42

Abstract

This thesis describes work on the statistics of the temporal and spatial variation of wind power in the UK, and the consequences of these variations for the use of wind generation of electricity both by the Central Electricity Generating Board (CEGB) and more generally. The statistical work is based on the use of spectral analysis, and cross correlation analysis of wind power time series estimated from Meteorological Office records of hourly mean wind speeds at 30 UK recording sites over a period of 9 years. Estimates of wind power capacity credit and wind operating reserve requirements as functions of wind power penetration into the conventional grid, geographical separation of wind power plant and specific power of wind turbines are made. An important result from this work, is that cross correlation coefficients of changes in wind power for lead times of a few hours are dominated by diurnal fluctuations in wind speed. These in turn are highly variable in both magnitude and sign, depending on among other things measurement height. The dominance of diurnal fluctuations can be seen more clearly in the coherence function plots for wind power at pairs of sites. The most important consequence of the resultant uncertainty in cross correlation coefficients will be on estimates of wind operating reserve requirements, which to a first approximation are linearly dependent on these coefficients.