حمید جلیل نژاد

In this research, field experiments were carried out in two types of soil (sandy clay loam and clay) to model the traction performance of the tractor considering drawbar power, rolling resistance, and traction efficiency, using deep learning and convolutional neural network while having some parameters such as soil type and conditions, tool parameters, and operation parameters. The tests were conducted within each soil texture in the form of factorial tests based on the randomized complete block design (RCBD) in triplicates. The tests were done in various moisture levels (8-17% for dry soils and 18-40% for moist soils), tractor forward speed (1.2, 1.6, 1.8, and 2.2 km h-1), working depth (30 and 50 cm), the number of pass (2 and 6 times), and tire inflation pressure (20 and 25 psi). The cone index, dynamic load, draft force, and moisture content were measured in each tests. The networks designed to model the drawbar power, rolling resistance, and traction efficiency were of convolutional neural network type. Various algorithms such as Sgdm, Adam, and Rmsprop were utilized to train the network. The results showed that the neural network developed by Sgdm algorithm outperformed the others. Therefore, this algorithm was utilized for the modeling process. Statistical criteria such as R2 and MSE were also employed to evaluate the performance of the network. For the drawbar power, the 8-499-499-1 architecture showed the best performance with R2=0.9953 and MSE=0.0016. Concerning the rolling resistance, the best performance was observed in 8-301-305-1 architecture with R2=0.9903 and MSE=0.0039. The best performance for the traction efficiency was obtained by 8-371-371-1 architecture with R2=0.9888 and MSE=0.003. The results showed that these networks can be used to model parameters by removing convolution layers and reducing dimensions.

Draft of different tillage tools is an important parameter for performance measurement, evaluation of tillage tools and also for determining the amount of required energy. Prediction of this parameter could be beneficial in many farm management practices, prediction of energy requirements and selecting appropriate tractor. In this study, field experiments were carried out at two soil types, namely, clay loam and loam clay, for predicting draft of a vertical narrow tillage tool, using artificial neural network and also, for comparison of developed model accuracy with that of regression models. Some parameters such as soil types, soil conditions, tools parameters and operational parameters were selected as inputs to artificial neural network. Within each type of soil, experiments were conducted in the form of factorial experiment based on randomized complete block design (RCDB) with three replications. Different levels of soil moisture content (factor A) 5-16 percent for dry soil and 17-38 percent for wet soil, tractor speed (factor B) at four levels of 1, 1.5, 1.8 and 3 km/hr, working depth (factor C) at four levels of 10, 20, 30 and 40cm and blade width (factor D) in four levels of 2.5, 3, 3.5 and 4cm were selected. Back propagation neural networks with three different training algorithms (gradient descending algorithm with momentum, descending scaled gradient and Levenberg-Marquardt) were adopted to predict the draft. Back propagation neural networks with Levenberg-Marquardt training algorithm presented better accuracy in simulation (95.05%) and correlation coefficient (R2 ) of 0.9935 as compared to others. The obtained data from neural network model were compared to ASAE and Ashrafizadeh (2006) models; the result showed that the predicted data by artificial neural network were very close to real data obtained from field experiments and the regression models did not have much proficiency for predicting draft at the studied area.