Optimisation and modelling of draft and rupture width using response surface methodology and artificial neural network for tillage tools

Abstract

Context: Soil–tool interaction modelling and optimisation reduce manufacturing costs and energy requirements for precision tillage equipment design. Diverse tillage tools have been designed to reduce draft requirements and desirable soil disturbance, but this is not fully understood. Aims: The current study investigated the effects of tool width, cone index, depth, and forward speed on draft with corresponding rupture width in order to develop response surface methodology (RSM) and artificial neural network (ANN) models and compared them to other models in order to predict draft and rupture width. Methods: Experiments were carried out in a soil bin with a vertisol, and rupture width was measured using an image processing technique. Key results: Using RSM, the optimum values for minimum draft with maximum rupture width within a range of independent variables were found to be 100 mm tool width, 600 kPa cone index, 141.63 mm tillage depth, and 3 km/h forward speed. For predicting the draft, the coefficients of determination (R2) for ANN and RSM models were 0.997 and 0.987, respectively; for rupture width prediction, R2 were 0.921 and 0.976. Conclusions: Developed ANN and RSM models of draft and rupture width were better than other analytical or numerical models, and both models’ predictions were in good agreement with experiment values within the range of ±5% uncertainty. Implications: The developed models can be used to predict the draft and soil disturbance requirements of tillage tools and design precision tillage tools.