Statistical modelling of wind speed measurement with accuracy tests models

In harnessing wind energy, the development of modelling wind speed plays a vital role in helping to obtain the actual wind speed data for the benefit of power plant planning in the future. More importantly, the accuracy of the wind distribution model needs to be carried out. In this paper, six selected locations in Nigeria, Jos, Katsina, Kano, Kaduna, Bauchi, and Maiduguri, were considered for investigation. Real-time wind speed data were obtained from the Nigerian Meteorological Agency for these locations. The paper examines different statistical models in the literature. The Weibull distribution model was used to carry out the wind speed distribution modelling. Also, the accuracy of the predicted wind distribution was validated using three different methods, namely, root mean square error, mean absolute error, and chi-square test. For improved results of the test, the test was done at three different height positions of 10m, 30m, and 60m respectively. The results obtained showed three different outcomes from the different heights that were considered for a whole month for the six selected locations with little error margin.