This project has been developed in the framework of the MSc in Energy Systems and the Environment under the Sustainable Engineering program in the University of Strathclyde, Glasgow.

The aim of the project is to study the variation in correlation parameters for the Measure-Correlate-Predict (MCP) method between the wind speed data of pseudo wind farm site and meteorological site. We are also exploring behaviour of power curve for real wind farm in UK. We present our work in the form of website and presentation.

Presently wind power, which grew by 29 percent worldwide in 2008 to reach an estimated generation of 260 TWh per annum, has 1.5% contribution in global electricity consumption. This technology has edge over other conventional power generation methods as fuel for wind turbine is free of cost and inexhaustible. Wind speed prediction is not only useful for wind farm but for military and civilian air traffic control, ship navigation purpose also. It has significant impact on the energy yield from typical wind farm. It is of prime importance to wind farm operators to have confidence on long term wind speed prediction and hence in expected energy yield from a wind farm. Most commonly used techniques for estimating the long-term wind resource at a potential wind farm site are the Wind Atlas methodology and the Measure-Correlate-Predict (MCP) methodology.

In the project, we will concentrate on MCP methodology as proposed by Derrick (1992). He suggests that to predict the wind speed of proposed wind farm, one should have one concurrent year wind data for both proposed wind farm and nearest meteorological site along with 10 years of historical wind data from the same meteorological station. Derrick's method is based linear regression algorithm. Currently some other algorithms are also available such as matrix method, vector regression method, a model based on the ratio of the standard deviations of the two data sets etc.

Nature of wind is typically stochastic. So prediction of wind speed for any site has uncertainty and hence in energy yield also. One can refer our presentation for main sources of uncertainty associated with wind energy yield prediction. We found it is interesting to research in the variation in correlation parameters with time. We limited our work on MCP method to study the variation of correlation parameters affecting energy yield. Key findings of the project can be found here.