Artificial Neural Network Based Modeling of Tractor Performance at Different Field Conditions

Salim Almaliki, Reza Alimardani, Mahmoud Omid


Application of tractors in farming is undeniable as a power supply. Therefore, performance model for evolving parameters of tractors and implements are essential for farm machinery, operators and manufacturers alike. The objective of this study was to assess the predictive capability of several configurations of ANNs for performance evaluating of tractor in parameters of drawbar power, fuel consumption, rolling resistance and tractive efficiency. A conventional tillage system which included a moldboard plow with three furrows was used for collecting data from MF285 Massey Ferguson tractor. To predict performance parameters, ANN models with back-propagation algorithm were developed using a MATLAB software with different topologies and training algorithms. For drawbar power. The best result was obtained by the ANN with 6-7-1 topology and Bayesian regulation training algorithm with R2 of 0.995 and MSE of 0.00024. The ANN model with 6-7-1 structure and Levenberg-Marquardt training algorithm had the best performance with R2 of 0.969 and MSE of 0.13427 for TFC prediction. The 6-8-1 topology shows the best power for prediction of AFC with R2 and MSE of 0.885 and 0.01348, respectively. Also, the 6-10-1 structure yielded the best performance for prediction of SFC with R2 of 0.935 and MSE of 0.012756. The obtained result showed that the 6-7-1 structured ANN with Levenberg-Marquardt training algorithm represents a good prediction of TE with R2 equal to 0.989 and MSE of 0.001327. The obtained results confirmed that the neural network can be able to learn the relationships between the input variables and performance parameters of tractor, very well.


Artificial neural network, Tractive efficiency, Rolling resistance, Drawbar power, Fuel consumption.

Full Text: