Development of Solar Radiation Estimation Model for North Eastern Stations of India Using Artificial Neural Network
Estimation of solar radiation is a thrust area for the number of researchers since the 20th century for optimal utilization of solar energy. Due to the scarce availability of meteorological stations, estimation is carried out by mathematical and soft computingbased solar radiation estimation models. In this paper, artificial intelligence technique, in particular, artificial neural network-based solar radiation estimation model is developed for North Eastern part of India using latitude, longitude, altitude, months of a year, maximum temperature, minimum temperature, humidity and sunshine hour. These data are collected from CROPWAT-8.0 for ten stations of North-East part of India. These stations are mostly hilly and sunshine duration and temperature remain lesser than the other parts of India. Meanwhile, the humidity remains generally higher. The LM algorithm with a feed-forward backpropagation network is used in the present analysis. For simulation purposes, MATLAB R2016a is used. The obtained results are compared with the measured data and RMSE is calculated for each station and overall too. The RMSE in the developed model varies from 0.39842 to 0.17586 for North Eastern stations of India.