Horizontal Extrapolation of Wind Speed Distribution Using Neural Network for Wind Resource Assessment

Abstract

Abstract:To evaluate the wind potential on a site for future wind energy project, an accurate representation of the wind speed distribution is required. However, due to the lack of observations, wind engineers are conducted to use some statistical tools to estimate the characteristics of wind by the measurements from a nearby reference or data obtained from a short period. In this work, we aim at applying an information processing paradigm that is inspired by biological neurons, formal neurons, for the assessment of wind speed distribution. Two different learning algorithms are used so as to generate Artificial Neural Network with one hidden layer. Results prove that learning by means of Bayesian regularization, in comparison with Levenberg-Marquardt learning algorithm, gives the best performance. In addition, the proposed network allows significant results in horizontal wind extrapolation.

Assessing the Wind Power Potential in Naama, Algeria to Complement Solar Energy through Integrated Modeling of the Wind Resource and Turbine Wind Performance

“…Gaps up to 2 h were addressed through linear interpolation. The resulting refined decade-long Naama dataset underpinned subsequent wind distribution modeling [25].…”

“…Additionally, quantile-quantile plots compared observed data against standard theoretical distributions to assess congruence. Preliminary examination revealed a pronounced Weibull distribution marked by a unimodal shape and positive skew, thereby indicating higher occurrence of lower wind speeds [23,25]. Data were then aggregated hourly and monthly to establish a foundation for more complex distribution modeling.…”

“…Furthermore, the dataset was meticulously scrutinized for sensor malfunctions and icing occurrences, with anomalies rectified through manual inspection and crossheight correlation. Linear interpolation was utilized to bridge gaps in data continuity, which spanned up to two hours [25]. Subsequent to these refinements, the resulting decade-long dataset from Naama underpinned subsequent wind distribution modeling efforts [23].…”