Abstract
Accurate solar forecasting facilitates the integration of solar generation into the grid by reducing the integration and operational costs associated with solar intermittencies. A novel solar radiation forecasting method was proposed in this paper, which uses two kinds of adaptive single decomposition algorithm, namely, empirical mode decomposition (EMD) and local mean decomposition (LMD), to decompose the strong non-stationary solar radiation sequence into a set of simpler components. The least squares support vector machine (LSSVM) and the Volterra model were employed to build forecasting sub-models for high-frequency components and low-frequency components, respectively, and the sub-forecasting results of each component were superimposed to obtain the final forecast results. The historical solar radiation data collected on Golden (CO, USA), in 2014 were used to evaluate the accuracy of the proposed model and its comparison with that of the ARIMA, the persistent model. The comparison demonstrated that the superior performance of the proposed hybrid method.
Highlights
Solar radiation is the most important factor affecting photovoltaic power generation [1].Accurate solar forecasting facilitates the integration of solar generation into the grid by reducing the integration and operational costs associated with solar intermittencies [2]
In terms of mean absolute error (MAE), the persistence model was better than the least squares support vector machine (LSSVM) and the Volterra model for all the datasets, which may be attributed to the LSSVM and the Volterra models are based on the principle of least squares error minimization, while the persistence model is an empirical model
Models combined to develop forecasting models (namely empirical mode decomposition (EMD)-LSSVM-Volterra and local mean decomposition (LMD)-LSSVM-Volterra, Analysis has shown that LSSVM is suitable for the high-frequency components (i.e., IMF1 and respectively)
Summary
Solar radiation is the most important factor affecting photovoltaic power generation [1].Accurate solar forecasting facilitates the integration of solar generation into the grid by reducing the integration and operational costs associated with solar intermittencies [2]. Many modeling methods have been used to describe solar radiation sequences, including regression methods such as linear regression [4], autoregressive model (AR) [6], autoregressive moving average (ARMA) [7], multi-dimensional linear prediction filters [8], least absolute shrinkage and selection operator (LASSO) [9], and nonlinear model approximators such as artificial neural network (ANN) [10], adaptive neuro-fuzzy inference system [11], hidden Markov model [12], fuzzy logic [13] and Angstrom–Prescott equations [14].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
