Prognostication of energy indices of tractor-implement utilizing soft computing techniques

Abstract

Energy indices (energy requirement for tillage implement (ERTI) and tractor overall energy efficiency (TOEE)) of tractor-implement during tillage operations were aimed to be investigated in this study. To generate a new comprehensive model, the effects of forward speed at three levels (2, 4 and 6 km/h) and plowing depth at three levels (10, 20 and 30 cm) on energy indices were experimentally evaluated. Two soft computing techniques, artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS), were employed to prognosticate energy indices. Comparison between the best developed structure of each soft computing technique demonstrated that one comprehensive ANN model was preferred than two individual ANFIS models. According to the ANN prognostication results, simultaneous increase of forward speed from 2 to 6 km/h along with plowing depth increment from 10 to 30 cm led to nonlinear increment of the ERTI and TOEE from 33.87 to 122.66 MJ/ha and 4.65 to 17.85%, respectively. Moreover, interaction of forward speed and plowing depth on energy indices was congruent. Development of comprehensive ANN model now makes it possible to answer fundamental questions in domain of the effect of plowing depth and forward speed on energy indices of tractor-implement that were previously intractable. Hence, to properly manage energy indices and reduce energy dissipation of tractor-implement, application of the new developed ANN model is strongly recommended. Keywords: Artificial neural network, Energy requirement for tillage implement, Tractor overall energy efficiency, Adaptive neuro-fuzzy inference system