Title: Adaptive neuro-fuzzy approach for prediction of global solar radiation for 25 cities falling under seven Köppen climatic zones

Authors: Vinay Anand Tikkiwal; Sajai Vir Singh; Dinesh Bisht; Hari Om Gupta

Addresses: Department of Electronics and Communication Engineering, Jaypee Institute of Information Technology, Noida-201304, India ' Department of Electronics and Communication Engineering, Jaypee Institute of Information Technology, Noida-201304, India ' Department of Mathematics, Jaypee Institute of Information Technology, Noida-201304, India ' Department of Electronics and Communication Engineering, Jaypee Institute of Information Technology, Noida-201304, India

Abstract: Estimation of solar energy is essential for the identification of suitable locations for solar-based energy systems and their optimal sizing. In this work, capability of adaptive neuro-fuzzy inference system (ANFIS) for the estimation of global solar radiation has been investigated for 25 different locations in India. Using geographical parameters like latitude, longitude and altitude, along with meteorological parameters such as sunshine duration, temperature as inputs, five ANFIS models have been developed. The relevant significance of input parameters with respect to global solar radiation has been evaluated using the regression method. The accuracy of prediction for the models has been evaluated using standard statistical indicators including mean absolute percentage error (MAPE), mean bias error (MBE), root mean square error (RMSE) and t-statistic. The models have been ranked using the rank score method. For the best ANFIS model, MAPE = 9.7%, RMSE = 0.80 and MBE = 6.6E-06 were obtained. The results indicate that the developed models have good prediction accuracy for cities having starkly different climates.

Keywords: global solar radiation; neuro-fuzzy computing; ANFIS; prediction; solar energy; renewable energy; artificial intelligence.

DOI: 10.1504/IJCAET.2021.118469

International Journal of Computer Aided Engineering and Technology, 2021 Vol.15 No.4, pp.501 - 515

Received: 13 Nov 2018
Accepted: 28 Jan 2019
Published online: 27 Oct 2021 *

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