Air classification efficiency evaluation of landfilled municipal solid waste using experiments and a probabilistic method

Abstract

The air classification efficiency of landfilled Municipal Solid Waste (MSW) is critical for resource recovery but remains challenged by the heterogeneity and non-spherical morphology of waste. In this study, a probabilistic framework that integrates spheroid modeling with the Monte Carlo procedure to predict and optimize the separation efficiency was developed, and the results were compared using orthogonal experiments. The morphological distributions (elongation, flatness, and size) of 381 landfill samples and the density distributions of 184 landfill samples were statistically characterized. For the numerical model, spheroidal particles were generated by randomly sampling from each parameter’s distribution. Then, a numerical model that incorporates non-spherical drag coefficients was developed. The model achieved a Root Mean Square Error (RMSE) of < 0.13 in predicting separation indicators, compared to the experimental results. The experimental results demonstrated that, under the same airflow velocity conditions, the recovery of light substances (R_L) in landfilled MSW was lower than that of fresh MSW, partially due to the increase in density resulting from the degradation of organic matter. The numerical model revealed that the separation efficiency (E) exhibited velocity-dependent unimodal trends. The model further identified the optimal performance of the effective separation interval for an airflow direction of 15° under 21.40 m/s (50 Hz), and the interval length was 21 % and 15 % longer compared to 0° and 30°, respectively, under equivalent velocities. The results of this work provide a reference for optimizing the air classification apparatus of landfilled MSW, and a basic method for use in more thorough simulation studies.