Construction of a Regression Model for Assessing the Efficiency of Separation of Lightweight Seeds on Vibratory Machines Involving Measures to Reduce the Harmful Influence of the Aerodynamic Factor

Abstract

When separating (cleaning) lightweight seed mixtures with the help of vibratory machines, there is an issue related to the harmful effect of air movement in the gaps between parallel working surfaces of vibratory machine units. This factor is particularly harmful to seed material, which is sensitive to air movement (some medicinal and vegetable crops). To address this issue, the design of vibratory machines is changed while their operational regimes are configured accordingly. This requires many full-scale experiments and (or) time-consuming personal computerbased simulation of the working processes of the vibrational motion of such seed mixtures. This paper proposes several regression models that make it possible to replace time-consuming numerical modeling with simple analytical expressions (regression equations). These equations are used for a quantitative assessment of the degree of influence of aerodynamics on the kinematic parameters of the vibrational motion of particles of seed mixtures. The assessment is derived depending on the geometric characteristics of the aerodynamic screen, the design of the unit, and the amplitude of oscillations of the working surfaces of a vibratory machine. The models take the form of equations of multiple linear regression of the second order, obtained on the basis of a series of numerical experiments. The processes of vibration movement of the seed material of parsnips, lettuce, and fragrant dill were investigated. The coefficient of determination equaled 0.956…0.967. The results reported here are useful for the construction of algorithms to optimize the design and adjust the operating modes of vibratory separators according to the criterion of minimizing the harmful effects of the aerodynamic factor.