Desenvolvimento e validação de um dispositivo para monitoramento da qualidade da massa de grãos de soja no transporte rodoviário

Abstract

The objective of this research was to develop and validate a hardware device with temperature, relative humidity and CO2 sensors for indirect monitoring of grain mass quality. All experimental steps were carried out at the Post-Harvest and Electrical Engineering Laboratory of the Federal University of Santa Maria, Campus Cachoeira do Sul. Chapter I comprised the construction / development and validation stages of the equipment. In the first part, a probe for conditioning the sensors was tested, in which the hole diameter and perforation height were changed. In the second part, mathematical modeling was used, evaluating different equilibrium moisture equations. In the last part of Chapter I, we sought to validate the harvest in soybeans with different qualities. With the results, it was observed that a probe with a bore diameter of 6.5 mm and a perforation height of 235 mm had a faster response for stabilizing the readings. The quantification of CO2 made it possible to carry out correlations with the quality of grain mass. Water contents of 10 and 13% had lower levels of carbon dioxide, while grains with water contents of 25% achieved a CO2 concentration of 5000 ppm. In evaluating the moisture balance of the intergranular air in the grain mass, it was observed that the Sigma Copace equation was the one that best fit the results. The quality results obtained by the germination and electrical conductivity tests in the grains were correlated with the concentration of carbon dioxide obtained in the grain mass. Grains with germination percentage of 78% and electrical conductivity of 126.75 μS cm-¹ had low CO2 values. As for grains with low germination values (6%) and high electrical conductivity values of 426.54 μS.cm-1, the concentration of carbon dioxide was 3500 ppm. In Chapter II, transport conditions were simulated, using grains with water contents of 11, 14 and 18%. Grains were monitored at three heights in three layers in the conveyor system. The results obtained have direct application to producers and grain storage and processing industries, in the monitoring and prediction of grain quality, as well as in storage and transport logistics. The water content associated with time were the factors with the greatest influence on the quality of transported grains. Displacement time is the determining factor for controlling the quality of grains during transport. Soybeans harvested from crops with a water content between 14 and 18% must not exceed 120 minutes of transport time to maintain quality. It is recommended that the transport time of grains shipped from storage units to processing industries, with a water content between 11 and 14%, does not exceed 840 minutes. The monitoring of indirect quality measurement variables associated with the application of Machine Learning models satisfactorily predicted the physical quality of the grain mass, along the transport time, for the different conditions tested.