Comparative study of artificial neural network (ann) and Adaptive Neuro-fuzzy Inference System (ANFIS) based models for solar radiation forecasting
| dc.contributor.advisor | Pravendra Kumar | |
| dc.contributor.author | Gaur, Indramani Pratap Singh | |
| dc.date.accessioned | 2019-01-10T05:56:20Z | |
| dc.date.available | 2019-01-10T05:56:20Z | |
| dc.date.issued | 2018-07 | |
| dc.description.abstract | Solar radiation is a electromagnetic radiation which we get from the Sun. Solar radiation is converted into solar energy by the uses of different techniques which include solar thermal energy, photovoltaics and solar heating. In this study an effort has been made for the development of artificial neural network (ANN) and Adaptive neuro-fuzzy inference system (ANFIS) models for daily solar radiation forecasting of Allahabad in Uttar Pradesh, India. The latitude, longitudes and elevation of study station are 25° 28' 22'' N, 81° 52' 42'' E and 98 m above mean sea level. The total geographical area of Allahabad district is 5482 Km2 according to census 2011. The daily meteorological data of 11 years (3 March, 2006 - 20 May, 2016) were collected from meteorological observatory located at Sam Higginbottom Institute of Agriculture, Technology and Science (SHIATS), Allahabad. The 70 percentage data (3 March, 2006 - 24 April, 2013) were used for model calibration and remaining 30 percentage data (25 April, 2013 - 20 May, 2016) were used for validation. The best combination of input variables was selected on basis the of Gamma statistic. In ANN model, back-propagation algorithm and log sigmoid activation function were used to train and test the models while in ANFIS models, triangular, trapezoidal, Gaussian and Generalized bell membership functions were used. The performance of the models were evaluated qualitatively by visual observations and quantitatively using various performances indices Viz. RMSE, correlation coefficient, MAPE, coefficient of efficiency and pooled average relative error. Finally, it can be concluded from the results that the performance of the ANFIS model (Gauss, 4) is better than the ANN model (11-20-1). The result showed that the current day solar radiation depends on the current day’s vapor pressure and sunshine hours, current day and one lag day of relative humidity and current day, one lag day and two lag day of temperature and wind speed. | en_US |
| dc.identifier.uri | http://krishikosh.egranth.ac.in/handle/1/5810089594 | |
| dc.keywords | neural network, fuzzy logic, solar radiation, forecasting | en_US |
| dc.language.iso | en | en_US |
| dc.pages | 88 | en_US |
| dc.publisher | G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand) | en_US |
| dc.research.problem | Neural Networks | en_US |
| dc.sub | Agricultural Engineering | en_US |
| dc.subject | null | en_US |
| dc.theme | Soil and Water Conservation | en_US |
| dc.these.type | M.Tech | en_US |
| dc.title | Comparative study of artificial neural network (ann) and Adaptive Neuro-fuzzy Inference System (ANFIS) based models for solar radiation forecasting | en_US |
| dc.type | Thesis | en_US |