Modeling environmental indicators for land leveling, using Artificial Neural Networks and Adaptive Neuron-Fuzzy Inference System

Abstract

Land leveling is one of the most important steps in soil preparation and cultivation. Although land leveling with machines requires considerable amount of energy, it delivers a suitable surface slope with minimal soil deterioration as well as damage to plants and other organisms in the soil. Notwithstanding, in recent years researchers have tried to reduce fossil fuel consumption and its deleterious side effects, using new techniques such as Artificial Neural Networks (ANNs) and Adaptive Neuron-Fuzzy Inference System (Fuzzy shell-clustering algorithm) models that will lead to a noticeable improvement in the environment. The present research investigates the effects of various soil properties such as Embankment Volume, Soil Compressibility Factor, Specific Gravity, Moisture Content, Slope, Sand Percent, and Soil Swelling Index in energy consumption. The study consists of 90 samples, collected from three different regions. The grid size has been set on 20 m * 20 m from a farmland in Karaj Province, Iran. The aim is to determine the best linear model, using ANNs and ANFIS model to predict environmental indicatorsand find the best model for land leveling in terms of its output (i.e. Labor Energy, Fuel energy, Total Machinery Cost, and Total Machinery Energy). Results show that ANFIS can successfully predict labor energy, fuel energy, total machinery cost, and total machinery energy. All ANFIS-based models have R² values above 0.995 and MSE values below 0.002 with higher accuracy in prediction, given their higher R² value and lower RMSE value.