Deterministic prediction of surface wind speed variations

Abstract

Abstract. Accurate prediction of wind speed is an important aspect of various tasks related to wind energy management such as wind turbine predictive control and wind power scheduling. The most typical characteristic of wind speed data is its persistent temporal variations. Most of the techniques reported in the literature for prediction of wind speed and power are based on statistical methods or probabilistic distribution of wind speed data. In this paper we demonstrate that deterministic forecasting methods can make accurate short-term predictions of wind speed using past data, at locations where the wind dynamics exhibit chaotic behaviour. The predictions are remarkably accurate up to 1 h with a normalised RMSE (root mean square error) of less than 0.02 and reasonably accurate up to 3 h with an error of less than 0.06. Repeated application of these methods at 234 different geographical locations for predicting wind speeds at 30-day intervals for 3 years reveals that the accuracy of prediction is more or less the same across all locations and time periods.Comparison of the results with f-ARIMA model predictions shows that the deterministic models with suitable parameters are capable of returning improved prediction accuracy and capturing the dynamical variations of the actual time series more faithfully. These methods are simple and computationally efficient and require only records of past data for making short-term wind speed forecasts within practically tolerable margin of errors.

The effect of shear web on the dynamic response of a wind turbine blade subjected to actual dynamic load

“…But in reality, wind is highly unsteady and to compute the actual load, an unsteady BEM is required. 11,31 Since the wind velocity is a function of both space and time, it is important to know the position of the blade at any instant with respect to a fixed inertial coordinate system. This inertial coordinate system is placed at the bottom of the tower.…”

“…whereṼ 1 is the wind velocity with respect to the ground andṼ 0 is the wind velocity with respect to the rotor,ã 34 ;ã 23 ;ã 12 are the transformation matrices to transform the velocity from ground to rotor frame of reference. 11,31 Figure 5 shows the different coordinate systems used with respect to which the velocities are referred. The relative velocity with respect to the rotor can be found as (Figure 6)…”

“…The present analysis is performed using a specific deterministic wind distribution captured with an anemometer at a height of 80 m and a location of 34:98420 N latitude and 104:03971 W longitude. 31 The variation of the wind velocity with height is neglected. Figure 8 shows the variation of the wind velocity with time for a period of 2000 s. The response of the NPS100 subjected to this wind velocity variation is analyzed in the present work and the effect of the position of shear web is investigated.…”

Cross-location wind speed forecasting for wind energy applications using machine learning based models

“…However, it has recently been demonstrated that the underlying dynamics of apparent random-like fluctuations of wind speed measurements is deterministic, low-dimensional and chaotic. 4,5,6 Hybrid models have been developed recently by combining different methods such as physical, statistical and machine learning methods to enhance prediction accuracy. 7,8 Models based on artificial neural network and other data learning techniques have also been gaining increased attention in literature in the recent past.…”

Empirical mode decomposition and chaos based prediction model for wind speed oscillations

“…Apart from these traditional methods, it has been recently reported that apparent random oscillation of wind speed data is chaotic, suggesting non-linear deterministic prediction schemes can be used for accurate short-term predictions [13]. In our previous work we have analysed the scope of deterministic methods to predict wind speed variations, and demonstrated that these methods are capable of providing reasonably accurate predictions up to a few hours [14].…”