Modeling of energy consumption and GHG (greenhouse gas) emissions in sugarcane production in ratoon farms using artificial neural networks (A case study in Ahvaz)

This study was carried out in Khouzestan province of Iran. Data were collected from 113 ratoon farms in Debel Khazai Agro-Industry with face to face questionnaire method. The objective of this study was to predict sugarcane production yield and GHG (greenhouse gas) emissions on the basis of energy inputs. Accordingly¡ several ANN (artificial neural network) models were developed and the prediction accuracy of them was evaluated using the quality parameters. The results illustrated that average total input and output energy of sugarcane production were 145117.7978 and 87096.42 MJ.ha-1¡ respectively. Electricity¡ chemical fertilizers and water for irrigation were the most influential factors in energy consumption. The ANN model with 6-7-19-1 and 5-5-1 structure were the best for predicting the sugarcane yield and GHG emissions¡ respectively. The coefficients of determination (R2) of the best topology were 0.96 and 0.99 for sugarcane yield and GHG emissions¡ respectively. The values of RMSE for sugarcane production and GHG emission were found to be 17763522 MJ.ha-1 and 528¡ respectively.