The prediction of wind speed is imperative nowadays due to the increased and effective generation of wind power. Wind power is the clean, free and conservative renewable energy. It is necessary to predict the wind speed, to implement wind power generation. This paper proposes a new model, named WTGWOBPNN, by integrating Wavelet Transform (WT), Back Propagation Neural Network (BPNN) and Grey Wolf Optimization (GWO). The wavelet transform is adopted to decompose the original time series data (wind speed) into approximation and detailed band. GWO – BPNN is applied to predict the wind speed. GWO is used to optimize the parameters of back propagation neural network and to improve the convergence state. This work uses wind power data of six months with 25, 086 data points to test and verify the performance of the proposed model. The proposed work, WTGWOBPNN, predicts the wind speed using a threestep procedure and provides better results. Mean Absolute Error (MAE), Mean Squared Error (MSE), Mean absolute percentage error (MAPE) and Root mean squared error (RMSE) are calculated to validate the performance of the proposed model. Experimental results demonstrate that the proposed model has better performance when compared to other methods in the literature.
The rapid growth of the world economy, the renewable energy sources such as solar, tidal, wind and geothermal energy has significantly shown its importance around the globe. Wind power is one of the most powerful ways to generate electricity from renewable sources. The amount of wind around the wind farm must be estimated to forecast the wind power. Wind speed is an important factor that affects the wind power. The other factors include location of wind farm and weather. There are two types of wind speed predictions, namely, short term and long term based on the time scale of the prediction. The farmer is more reliable than the later [
Statistical and physical methods are also named as direct and indirect methods respectively. In direct method, there is a linear relationship among input time series data and the predicted results. Some of the direct method for prediction of wind speed are presented in [
In Zhou et al. [
WT is one of the prevailing tools to analyze variety of resolution images and time series data. During wavelet analysis the input signal is decomposed into shifted wavelets and scaled wavelets. Spare representation and nonredundant computation are the advantages of DWT than the Fourier Transform (FT), and Gabor transform while extracting features. DWT has been successfully used for feature extraction and achieve better results when compared to other methods [
BPNN is a variety of ANN, supervised learning, multilayer feedforward network. The layers in BPNN are input layer, multiple hidden layers and output layer [
The GWO is one of the metaheuristic algorithms based on the grey wolf’s hunting style [
In GWO
Where A and C are the coefficient vectors. The position of victim is Xp and the position of grey wolf is X. Both Xp and X are vectors. Vector
This section presents details of some of the prediction models of wind speed available in the literature. Jianguo Zhou et al. have proposed a wind power forecasting named ESMDPSOELM [
Han et al. [
Fredy H. Martínez S et al. [
Jinli Dou et al. [
Younes et al. [
In order to predict wind power, a hybrid deep learning neural network system has proposed in Liu et al. [
S. No.  Methodology  Dataset Count/Year/Location  Performance Parameters  Used Models  Results 

1.  ESMDPSOELM [ 
2880/April 2016/China  MAPE, MAE, RMSE  BPNN, ELM, PSOELN, EMD BPNN, EMD ELM, EMD PSOELM, ESMD BPNN, ESMD ELM,  MAPE is 4.76, MAE is 2.23 and RMSE is 2.70 
2.  ATLDNN [ 
1317 months of 5 wind form data  MeanAbsoluteError, RootMeanSquaredError, and StandardDeviationError  ARIMA, 
MAE is 0.0637, 
3.  Machine Learning [ 
2013 to 2015/ University of Waterloo weather station  RMSE and MAE  BPN, RBF, NARX, 
RMSE is 0.5814 
4.  IHGWOSCA [ 
7200/2017/Sotavento Galicia (SG) wind farm and Inner Mongolia (IM) wind farm  RMSE, MAE and MAPE  GSSVM, GSBP, GSELM, OriginalIHGWOSCAPSRELM, OriginalSSAIHGWOSCAPSRELM, OVMDSSACCPSRELM, EMDIHGWOSCAPSRELM, OVMDIHGWOSCAPSRELM, EMDSSAIHGWOSCAPSRELM, OVMDSSAIHGWOSCAPSRELM  RMSE is 0.0675 
5.  Long ShortTerm Memory (LSTM) [ 
10 Years/ Guajira (Colombia)  Root Mean Squared Error (RMSE)  –  RMSE is 4.223 km/h 
6.  ESELM [ 
January 2013 to 31st December 2014/five airports in U.K.  Average accuracy, MAPE, VAPE, R^{2}  EMDSAE, EMDELM, and SAEELM  Average accuracy is 93.73 
7.  CNN [ 
Root mean square error, Average absolute error  FNN, CNN 
CNN 

8.  Machine Learning 
8 years (2005–2011)/ Kersey site, Colorado, USA  correlation 
MLP, ANN, Genetic Programming, Persistence method  ANN 
9.  SSACNNGRUSVR [ 
2688 each /April 2015 to May 2015, May 2016 to June 2016, September 2014 /China  MAPE, MAE, RMSE  ARIMA model, PM 
MAPE is 0.97 
10.  BPNN [ 
Iran  MAPE, MAE, RMSE, WSHE, MAWEEE  RBFNN, ANFIS  WSHE is 2.6% 
11.  Deep Belief Network (DBN) [ 
three months 
NNSHL, NNMHL, SVR  DBN has less error in prediction  
12.  EWTLSTMElman [ 
700 samples  MAPE, MAE, RMSE  ARIMA, GRNN, Elman, EWTElman, BP, LSTM, EWTBP, WPDLSTMElman, EWDLSTMElman  EWTLSTMElman model depicts best performance 
13.  ANN [ 
2 March and 2 April 2016 / Eskisehir  RMSE, MAE  –  RMSE is 0.6508 
14.  ANN [ 
2003 and 2004 / Faro, Portugal  MSE  –  MSE is 1.288 
15.  DNNMRT [ 
five wind farms / Europe  RMSE, MAE, SDE  ARIMA, SVR  DNNMRT performs better 
The proposed model WTGWOBPNN is used to predict the wind speed. The WT algorithm is used for decomposing the input signal ie., time series data. GWOBPNN is used for predicting the wind speed based on the characteristics of the time series data. In this model GWO is used for tuning the weights during the BPNN training. The wind speed data is taken from the publicly available data sources for six months.
Sample  Mean  Median  Minimum  Maximum  Std.  Skewness  Kurtosis 

25086  13.48  12.9  0  58.4  6.54  0.724  3.914 
Usually, MAPE – Mean Absolute Percentage Error, MAE – Mean Absolute Error, RMSE – Root Mean Square Error and MSE – Mean Square Error are the performance measures for any predictive model [
Evaluation parameters  Formula 

MSE  
MAE  
RMSE  
MAPE  
AActual value, PPredicted value 
This section deals with the experimental results obtained using the proposed system WTGWOBPNN. 25,086 data points are chosen as input for prediction. Seventy percentage of input data points are considered for training and remaining thirty percentage for testing.
The error for training and testing data is shown in
Performance Metrics  1Step  2Step  3Step 

MSE  2.69  4.95  6.62 
RMSE  1.64  2.22  2.57 
MAE  1.13  1.54  1.80 
MAPE (%)  11.68  16.28  19.03 
Wind speed is one of the sources of renewable energy. Wind speed prediction is required to predict wind power. Due to the vague environment, the wind speed prediction is more difficult process. This paper has presented a wind speed prediction model employing a hybrid method named as WTGWOBPNN. The proposed method uses six months wind speed data of about 25,086 samples for prediction. The performance of the proposed model is evaluated and compared with the existing models in terms of MSE, MAE, RMSE and MAPE. It is observed that the proposed model poses low MAE when compared to all other models such as Elman, ARIMA, WPDELM, GRNN, EWTElman and BPNN. Additionally, the value of MSE and RMSE is better than Elman, ARIMA and BPNN. Stepwise performance comparison of WTGWOBPNN proved that the proposed method is used for prediction of wind speed during climatic changes. In future research, deep neural network methodologies can be applied with more sample data to improve the prediction of wind speed.
The authors acknowledge the support and encouragement of the management and Principal of Sri Ramakrishna Engineering College. We would also like to thank the referees and the editors for their helpful comments and suggestions.