Title: Short term solar energy forecasting using GNN integrated wavelet-based approach

Authors: Priyanka Chaudhary; M. Rizwan

Addresses: Department of Electrical Engineering, Delhi Technological University, 110042, Delhi, India ' Department of Electrical Engineering, Delhi Technological University, 110042, Delhi, India

Abstract: The power generation from solar energy is gaining more attention because of advancement in the solar photovoltaic technology including enhanced efficiency of solar cells by incorporating the good materials. In the present scenario the bidding of power is done on 15 minutes basis by many distribution companies. Keeping in mind aforesaid, 15 minutes ahead short term solar energy forecasting has been done and presented. As the power from solar energy is fluctuating and nonlinear in nature, the results obtained from mathematical models are not found satisfactory. Therefore, an intelligent approach based on wavelet transform and generalised neural network (GNN) is developed and applied for the short term solar energy forecasting problem. The results obtained from the proposed model are evaluated on the basis of statistical indicators like root mean square error (RMSE) and mean absolute error (MAE). It is concluded that the performance of the proposed model is found better as compared to GNN model.

Keywords: generalised neural network; GNN; wavelet transform; grid integrated solar PV systems; solar energy forecasting.

DOI: 10.1504/IJRET.2019.101729

International Journal of Renewable Energy Technology, 2019 Vol.10 No.3, pp.229 - 246

Available online: 02 Aug 2019 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article