ABSTRACT This study aimed to develop artificial neural networks for the estimation of tractor fuel consumption during soil preparation, according to the adopted system. The multilayer perceptron network was chosen. As input data: the soil mechanical penetration resistance, the mobilized area by implements, the working gear and the tractor engine speed. The number of layers and neurons varied to form different architectures. The adjustment was verified based on various statistical criteria. The values estimated by the networks did not differ significantly from those obtained experimentally. The conclusion was that the networks showed adequate reliability and accuracy to predicting the fuel consumption in each tillage system, in function of the input data and this can be a useful tool for planning and management of agricultural operations.

No presente estudo, foram elaborados modelos empíricos para determinar a força de tração demandada por arados de discos, escarificadores e semeadoras adubadoras, em função da resistência mecânica do solo à penetração. A média da referida resistência, determinou-se até as profundidades de 25; 35 e 15 cm, em correspondência com as regulagens do arado de discos, escarificador e semeadora adubadora. O ajuste foi verificado de acordo com o coeficiente de determinação, gráficos de dispersão, análise residual e teste t (Student). Com base nessas análises, confirmou-se a normalidade dos resíduos, e foram estabelecidos intervalos de confiança com 95% de probabilidade. Os valores estimados e obtidos experimentalmente não discreparam significativamente. Concluiu-se que as funções exponenciais associadas apresentaram adequada precisão e confiabilidade para predizer a força de tração, considerando-se a resistência mecânica do solo à penetração e que os modelos podem ser uma ferramenta útil para o planejamento e gestão de operações agrícolas mecanizadas em solos com textura similar às desta pesquisa.

In this study, empirical models were developed to determine the tensile force demanded by disc plows, harrows and planters fertilizers, as function of soil mechanical resistance to penetration. The mean of the referred resistance was determined until 25, 35 and 15 cm depths, according to the regulations for the disc plow, springs harrow and fertilizer seeder. The adjustment was verified based on the coefficient of determination, dispersion graphics, residual analysis and t test (Student). It was confirm the normality of the residuals and to establish intervals of confidence with 95% of probability. The estimated and experimentally obtained values did not differ significantly. It was concluded that the associated exponential functions presented appropriate precision and reliability to predict the traction force being considered the soil mechanical resistance to penetration and that the models can be a useful tool for planning and management of mechanized farming operations in soils with texture similar to those of this research.