RELATIONSHIP BETWEEN THE 3D FOOTPRINT OF AN AGRICULTURAL TIRE AND DRAWBAR PULL USING AN ARTIFICIAL NEURAL NETWORK.

Improving the traction of an agricultural tractor on the field increases its working efficiency and capacity. Heavy work, like plowing, entails high levels of tire slip, which is directly related to power loss when the transmission of drawbar pull is required. Accordingly, it is possible to hypothesize that a tire with a higher traction capability could increase the working efficiency of the machine. The natural evolution for measuring the geometrical parameters of tires has led to the consideration of three-dimensional (3D) footprints since the distribution of the vertical stresses at the soil-tire interface may be highly non-uniform. In this study, the data acquired from 3D footprints of 10 sets of tires underwent processing along with the data from drawbar tests carried out with the same tires on soil terrain at different slip ratios. Subsequently, artificial intelligence multivariate methods based on artificial neural networks allowed traction prediction and verified the importance that the acquired geometrical parameters have on the measured drawbar pull. The study confirmed the correlation of the geometrical parameters of the 3D tire footprint with the drawbar pull and the results of the artificial intelligence modelling underlined the impact of these acquisitions. However, further work that considers various lug geometries is required to extend the generalizability of the studied methodology. [ABSTRACT FROM AUTHOR]