Zeitschrift für Ökologie und Toxikologie

Abstrakt

Prediction of environmental indicators in land leveling using artificial intelligence

Isham Alzoub

Land leveling is one of the most important steps in soil
preparation and cultivation. Although land leveling with
machines require considerable amount of energy, it delivers
a suitable surface slope with minimal deterioration
of the soil and damage to plants and other organisms in
the soil. Notwithstanding, researchers during recent years
have tried to reduce fossil fuel consumption and its deleterious
side effects using new techniques such as; Artificial
Neural Network (ANN),Imperialist Competitive Algorithm
–ANN (ICA-ANN), and regression and Adaptive
Neuro-Fuzzy Inference System (ANFIS) andSensitivity
Analysis that will lead to a noticeable improvement in the
environment. In this research 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
were investigated. The study was consisted of 90 samples
were collected from 3 different regions. The grid size
was set 20 m in 20 m (20*20) from a farmland in Karaj
province of Iran. The aim of this work was to determine
best linear model Adaptive Neuro-Fuzzy Inference System
(ANFIS) and Sensitivity Analysis in order to predict
the energy consumption for land leveling. According to
the results of Sensitivity Analysis, only three parameters;
Density, Soil Compressibility Factor and, Embankment
Volume Index had significant effect on fuel consumption.
According to the results of regression, only three
parameters; Slope, Cut-Fill Volume (V) and, Soil Swelling
Index (SSI) had significant effect on energy consumption.
using adaptive neuro-fuzzy inference system for prediction
of labor energy, fuel energy, total machinery cost,
and total machinery energy can be successfully demonstrated.
In comparison with ANN, all ICA-ANN models
had higher accuracy in prediction according to their higher
R2 value and lower RMSE value. The performance of
the multivariate ICA-ANN and regression and artificial
neural network and Sensitivity analysis and Adaptive
neuro-fuzzy inference system (ANFIS) model was evaluated
by using statistical index (RMSE, R2 )). The values of
RMSE and R2 derived by ICA-ANN model were, to Labor
Energy (0.0146 and 0.9987), Fuel energy (0.0322 and
0.9975), Total Machinery Cost (0.0248 and 0.9963), Total
Machinery Energy (0.0161 and 0.9987) respectively, while
these parameters for multivariate regression model were,
to Labor Energy (0.1394 and 0.9008), Fuel energy (0.1514
and 0.8913), Total Machinery Cost (TMC) (0.1492 and
0.9128), Total Machinery Energy (0.1378 and 0.9103).Respectively,
while these parameters for ANN model were,
to Labor Energy (0.0159 and 0.9990), Fuel energy (0.0206
and 0.9983), Total Machinery Cost (0.0287 and 0.9966),
Total Machinery Energy (0.0157 and 0.9990) respectively,
while these parameters for Sensitivity analysis model were,
to Labor Energy (0.1899 and 0.8631), Fuel energy (0.8562
and 0.0206), Total Machinery Cost (0.1946 and 0.8581),
Total Machinery Energy (0.1892 and 0.8437) respectively,
respectively, while these parameters for ANFIS model
were, to Labor Energy (0.0159 and 0.9990), Fuel energy
(0.0206 and 0.9983), Total Machinery Cost (0.0287 and
0.9966), Total Machinery Energy (0.0157 and 0.9990)
respectively, Results showed that ICA_ANN with seven
neurons in hidden layer had better.