Modeling and Optimization of Energy Inputs and Greenhouse Gas Emissions for Eggplant Production Using Artificial Neural Network and Multi-Objective Genetic Algorithm

Abstract

This paper studies the modeling and optimization of energy use and greenhouse gas emissions of eggplant production using artificial neural network and multi-objective genetic algorithm in Guilan province of Iran. Results showed that the highest share of energy consumption belongs to diesel fuel (49.24%); followed by nitrogen (33.30%). The results indicated that a total energy input of 13910.67 MJ ha^-1 was consumed for eggplant production. In ANN, the Levenberg-Marquardt Algorithm was examined to finding best topology for modeling and optimization of energy inputs an GHG emissions for eggplant production. The results of ANN indicated the best topology with 12-9-9-2 structure had the highest R², lowest RMSE and MAPE. Also, the multi-objective optimization was done by MOGA. In this research, 42 optimal was introduced by MOGA based minimum total GHG emissions and maximum yield of eggplant production, in the studied area. Also, the results revealed that the best generation with lowest energy use was consumed about 4597 MJ per hectare. The GHG emissions of best generation was calculated as about 127 kg CO_2eq.ha^-1. The potential of GHG reduction by MOGA was computed as 388.48 kg CO_2eq.ha^-1. Also, the highest reduction of GHG emissions belongs to diesel fuel with 65.05%