Estimation of wind resources in the coast of Ceará, Brazil, using the linear regression theory

This work is concerned with the estimation of onshore wind resources in the coast of Ceará, Brazil by using the linear regression method. The main focus is to estimate the average wind speed at several altitudes from data collected at the surface. Two specific areas are investigated i.e. Paracuru and Camocim, which are located in the state of Ceará, Brazil. The same methodology is adopted in both cases, where the regions are initially characterized by obtaining the daily and monthly average wind speed profiles from raw data collected by a Platform for Data Collection (PDC) and an Anemometric Tower (AT). Data regarding the prevailing wind direction are also recorded. By using the logarithmic wind profile equation, it is possible to estimate the average wind speeds at altitudes of 20m, 40m, and 60m from data collected at 10m, as it is possible to determine correlation coefficients between the data and those collected by the AT. Linear regression model is used to estimate the average speed for new altitudes. This procedure is carried out during the calibration and model validation. In both periods, the linear regression model has shown good performance in terms of high level of agreement for the data series, estimated data and related correlation coefficients, and also low error values involving the aforementioned series.