Simulation study on the screening effect of flat pendulum composite vibration based on DEM

In order to improve the classification effect of particle groups, the stratification effect of raw coal particle groups in the vibration screening process, and the utilization efficiency of raw coal materials, a composite vibration form of pendulum and linear vibration was proposed on the basis of a traditional linear vibrating screen. An orthogonal experiment was designed based on on-site parameters using conventional linear vibrating screens and composite vibrating screens as experimental objects. Swing frequency, swing angle, vibration frequency, amplitude, vibration direction angle, and amplitude difference were among the parameter variables. The new composite force field vibrating screen was found to have substantial advantages over existing vibrating screens in terms of both screening efficiency and layering rate. To investigate the complex components’ combined impacts on the particle groups during the screening process, the six screening machine parameters were used as independent variables. The screening outcomes were then subsequently subjected to a BP neural network model. Regression coefficients R values were 0.99511, 0.96954, 0.94360, and 0.92158 for the training sample, prediction sample, test sample, and prediction outcomes, respectively. This demonstrated that the BP neural network model was highly effective for optimizing vibrating screen parameter design.