1Department of Agricultural Machinery Engineering, Faculty of Agricultural Engineering and Technology, University of Tehran, Karaj, Iran
2Iranian Agricultural Engineering Research Institute, Karaj, Iran
Seed corn is one of the most sensitive crops during harvesting operation. Due to its high economic value and in order to decrease the amount of losses, harvesting operation of this crop should be done precisely. There are several factors which can affect grain combine harvesting losses. In order to find and predict the relationship between cylinder and travel speed of grain combine and seed corn harvesting losses, various Artificial Neural Networks (ANNs) were developed. For this purpose, data were collected from seed corn farms during harvesting operation in different levels of grain combine cylinder and travel speed. The developed ANN was a multilayer perceptron (MLP) with two neurons in the input layer, one and two hidden layers of various numbers of neurons and one neuron in the output layer. Results showed that the ANN model with 2-7-10-1 topology can predict the harvesting losses value well. For the optimal model, the coefficient of determination (R2), the mean absolute error (MAE) and the root mean square error (RMSE) was calculated as 0.93, 15.48 and 17.00, respectively. Finally, sensitivity analysis revealed that cylinder speed is the most significant parameter in seed corn harvesting losses.
Artificial neural networks, Grain combine, Seed corn, Harvesting losses, Travel speed, Cylinder speed
