|Title : A Review of MCP Techniques
Authors : Mike Anderson, RES
This report reviews a range of methods used to estimate the long term wind speed at potential wind
farm sites, where only a limited amount of on-site data are available.
The methods which are reviewed here comprise a family of generic ‘measure-correlate-predict’
(MCP) methodologies. These methods proceed by ‘measuring’ the winds at a target site,
‘correlating’ them with winds from a nearby reference site, and then by applying these correlations
to historical data from the reference site, to ‘predict’ the long term wind resource of the target site.
OUTLINE OF MCP METHODOLOGY
The general methodology of the MCP process proceeds as follows:
i) collect wind data at the predictor site for an extended period;
ii) identify a reference site, for which high quality, long term records exist, in the vicinity of
the predictor site, and which has a similar exposure - this is hereafter referred to as the
iii) obtain wind data from the reference site for the same time period as for the predictor site -
this period is hereafter referred to as the ‘concurrent period’.
iv) establish a relationship between the data from the reference and predictor sites for the
v) obtain wind data from the reference site for a historic period of 10 to 20 years duration -
this period is hereafter referred to as the ‘historic’ period;
vi) apply the relationship determined from (iv) above to the historic data from the reference
site to ‘predict’ what the winds would have been at the predictor site over that period.
that this is a prediction of the winds that would have been observed had measurements been
made at the predictor site for the same period as the historic data, rather than a prediction of
winds that will be observed in future.
Dr. Mike Anderson,
RES group, UK
Download as PDF