A model to assess tractor operational efficiency from bench test data

Agricultural mechanization is required to sustain food production with high productivity, but fuel resource limitation has spurred both tractor manufacturers and users to take care of their fuel consumptions. Fuel consumptions of tractors are often assessed from tractor oriented testing procedures which are assumed to reflect a variety of agricultural operations. However, the resulting diagnostic, based on an annual average use, lacks consistency and users are asking for a more detailed fuel assessment according to real use. A novel approach is therefore proposed here which aims to be more representative of real field usage and to better express the related energy performance. The approach is designed in order to be suitable with automotive applications estimating fuel needs over some so-called driving cycles. The driving cycle, also named field working dynamics, is investigated by monitoring an experimental tractor throughout a whole year of field operations. Statistical analysis is applied to discriminate and characterize different tasks during the tractor use in field operation: displacement, poor idling, maneuvering and driving along the field. Time and mechanical energy needs are described for each subtask. Then, a parametric model is used to convert mechanical needs into a fuel demand. It is designed to predict operational efficiency as a function of agricultural parameters. The model is calibrated for a tractor by laboratory test procedures. For validation purposes, the model was applied to a plowing operation, in which the predicted efficiencies for fuel, time and field are compared to the actual efficiencies measured in the field. Lastly, the effect of operational parameters on efficiency is discussed through a sensitivity analysis that links fuel consumption and productivity. This analysis shows the main parameters that have to be defined to characterize agricultural work and convert an engine diagnostic into a user-oriented consumption. (C) 2014 ISTVS. Published by Elsevier Ltd. All rights reserved.