Search

[Display omitted] • A stepwise hybrid model is proposed for hourly solar irradiance forecasting. • The correction errors are used by a step-by-step approach. • The VMD method is proposed to reduce non-stationarity of irradiance series. • The CNN-LSTM-MLP and DBN are combined. • The feature of high and low frequencies are predicted separately. Accurate and reliable solar irradiance forecasting is critical for distribution planning and modern smart grid management and dispatch. However, due to the time series of solar irradiance with the nonlinearity and nonstationarity, some researches up to now are still unsatisfactory in terms of prediction accuracy and model generalization ability. Therefore, to improve the comprehensive performance of the model, a novel forecasting framework of hourly stepwise forecasting for solar irradiance using integrated hybrid models CNN-LSTM-MLP combined with error correction and variational mode decomposition (VMD) was proposed. In this paper, three different datasets were pre-processed and then the step-by-step predictions of seven different methods were completed based on the proposed procedures. Finally, comprehensive analysis results of the model indicated that the prediction scheme proposed in this paper made full use of, step-by-step prediction, VMD method, integrated hybrid model, and error correction four major advantages greatly improved the anti-interference ability of the model, so that the average performance metrics RMSE, nRMSE, MAE, and R2 of the three datasets reach to 12.53 W/m2, 6.04 %, 7.65 W/m2, and 99.79 %, respectively. Moreover, the optimal promoting percentages of the R2(P R 2 )indicator compared to the persistence model (Per) is by 20.17 %. It is found that the model is superior to a large number of traditional alternative approaches in terms of accuracy and robustness, which may provide a reference for comprehensive performance optimization of the model in the future. [ABSTRACT FROM AUTHOR]

Accurately forecasting wind speed is a critical mission for the exploitation and utilization of wind power. To improve the prediction accuracy, the nonlinearity and nonstationarity embedded in wind speed time series should be reduced. Because the subseries has less nonlinearity and nonstationarity after decomposition, the decomposition-based forecasting methods are widely adopted to provide the higher predictive accuracy. However, latest studies showed the real time decomposition-based forecasting methods could be worse than the single forecasting models. The aim of this study is to improve the performance of the real time decomposition-based forecasting method after the factors attributed to its unsatisfactory performance are uncovered. In this paper, the feature selection and error correction are adopted in the real time decomposition-based forecasting method to enhance the prediction accuracy. In the proposed method, the raw wind speed time series is decomposed into a number of different subseries by ensemble empirical mode decomposition; then two feature selection methods including kernel density estimation-based Kullback-Leibler divergence and energy measure are used to reduce the disturbance of illusive components; further the least squares support vector machine is adopted to establish the one-step ahead forecasting models for the remaining subseries; finally, the hybrid of least squares support vector machine and generalized auto-regressive conditionally heteroscedastic model is introduced to correct resulting error component if its inherent correlation and heteroscedasticity cannot be neglected. Based on two sets of measured data, the results of this study show that: (1) the real time decomposition-based method may be ineffective in practice; (2) both the feature selection and error correction can improve forecasting performance in comparison with the real time decomposition-based method; (3) compared with other involved methods, the proposed hybrid method has the satisfactory performance in both accuracy and stability. [ABSTRACT FROM AUTHOR]

• Propose a novel short-term wind speed prediction model base on deep learning. • Employ the double decomposition method to process wind speed data. • Modify the forecasting errors in the error correction strategy. • Verify the validation of the proposed model by four real cases. As wind power accounts for an increasing proportion of the electricity market, the wind speed prediction plays a vital role in the stable operation of the power grid. However, owing to the stochastic nature of wind speed, predicting wind speeds accurately is difficult. Aims at this challenge, a new short-term wind speed prediction model based on double decomposition, error correction strategy and deep learning algorithm is proposed. The complete ensemble empirical mode decomposition with adaptive noise and variational mode decomposition are applied to decompose the original wind speed series and error series, respectively. The deep learning algorithm based on long short term memory neural network, is utilized to detect the long-term and short-term memory characteristics and build the suitable prediction model for each sub-series. In the four real forecasting cases, nine models were built to compare the performance of the proposed model. The experimental results show that the proposed model performs better than all other considered models without double decomposition, and the variational mode decomposition for error series can improve the effect of error correction strategy. [ABSTRACT FROM AUTHOR]

• The proposed model can extract unpredictable components from raw wind speed. • Two models are built on predictable and unpredictable components respectively. • An adaptive multiple error corrections method is developed. • The bivariate Dirichlet process mixture model can produce heteroscedastic results. Wind power is promising renewable energy. Wind speed forecasting is essential for wind energy integration and application. The wind speed series can be divided into two components, including predictable and unpredictable components. In this study, a novel two-stage forecasting model is proposed for dealing with those two components separately. In the first stage, a novel model consisting of wavelet packet decomposition, convolutional neural network and adaptive multiple error corrections is proposed to forecast the predictable components. The adaptive multiple error corrections method can eliminate the predictable components in forecasting residuals of the wavelet convolutional neural network model, which is unique. In the second stage, bivariate Dirichlet mixture model is proposed to model heteroscedasticity of the unpredictable residuals with the non-parametric distribution. Several important hyper-parameters are selected by blocked cross validation. Three actual wind speed series are utilized to verify the effectiveness of the proposed model. The results show that the proposed model outperforms the benchmark models. [ABSTRACT FROM AUTHOR]