Indirect Estimation of the Volumetric Throughput Performance in the Shredding of Solid Waste

The volume or mass throughput of a mechanical treatment plant for commercial waste represents a key performance parameter. This measurement parameter is often unavailable, as the sensor technology required is often expensive or does not provide accurate data. The first process stage is usually a shredding machine, converting the waste into a transportable and separable fraction size. Here, a methodical approach is pursued which enables an indirect estimation of the volume throughput capacity based on further machine parameters, such as the drum speed and the drum torque. Based on 32 test data sets, two models were developed to approximate the volume throughput rate. The two models developed are the regression model and the displacement model. Furthermore, two reference models were defined to evaluate the accuracy of the two approaches developed: the so-called mean value model and the ANOVA model. When looking at the 80th percentile of the sign-adjusted relative deviation, the results show that the regression model, with ±40%, followed by the displacement model, with ±42%, enable significantly more accurate estimates of the volumetric throughput performance than the two reference models, with ±63% and ±71%, respectively.