Artificial Neural Networks for Predicting the Rice Yield in Phimai District of Thailand
Saisunee Jabjone · International Journal of Electrical Energy · 2013
Ab stract—This study aimed to find the model for predicting rice yield in Phimai district, Thailand. A classic multilayer feed-forward neural network with back-propagation algorithm was used throughout this experiment. Data from 2002 to 2007 were used as the training data for predict the rice yield between 2008 and 2012. The input data from six meteorological factors; rainfall, water distribution, evapotranspiration, temperature, humidity and wind speed were used. Evapotranspiration (ET) was found by using Pennman-Montieth equation. The result showed that ANN (8, 19, and 17) provided the lowest value of RMSE (10.57) and MAPE (2.3). The rice yield predicting of ANN (8, 19, 17) and actual data have linear relationship (R 2 =0.99). This