Wind speed prediction using statistical regression and neural network

Abstract

Prediction of wind speed in the atmospheric boundary layer is important for wind energy assessment, satellite launching and aviation, etc. There are a few techniques available for wind speed prediction, which require a minimum number of input parameters. Four different statistical techniques, viz., curve fitting, Auto Regressive Integrated Moving Average Model (ARIMA), extrapolation with periodic function and Artificial Neural Networks (ANN) are employed to predict wind speed. These methods require wind speeds of previous hours as input. It has been found that wind speed can be predicted with a reasonable degree of accuracy using two methods, viz., extrapolation using periodic curve fitting and ANN and the other two methods are not very useful.

RETRACTED: Artificial neural networks applications in wind energy systems: a review

“…Kulkarni et al [43] used four different statistical techniques, curve fitting, ARIMA, extrapolation with periodic function and ANN, to predict wind speed. In this study, wind speeds of previous hours were used as input.…”

“…vii. MLFFN with statistical data weighting pre-processing reduces the number of training data [43]. This concept may be developed for WECS applications.…”

Short term wind speed prediction using artificial neural networks

“…There are a number of traditional methods for wind speed prediction; most of them involve the statistical analysis of wind speed data from the past and the development of an empirical model using mathematical methods, such as least squares curve fitting or non-linear regression ( [11]). These methods are simple but generally result in low accuracy.…”

Estimation and Forecast of Wind Power Generation by FTDNN and NARX-net based models for Energy Management Purpose in Smart Grids

“…With reference to wind-farm power generation, modelling wind speed by means of time series forecasting techniques is going to be considered more and more promising [1], [7]. The high complexity and the intermittent and non-linear behaviour of the wind speed makes necessary the use of autoregressive, fuzzy and neural techniques and forecast model based on hybrid algorithms [8]- [10]. In general, Artificial Neural Networks (ANNs) and hybrid algorithms provide to be more effective than other classical autoregressive predictors for both wind speed and solar radiation [1], [7], [11], [12].…”