Development of feed factor prediction models for loss-in-weight powder feeders

Feed factor profile prediction model for two-component mixed powder in the twin-screw feeder

“…In discussing absolute rather than relative errors, Shier et al (2022) stated that the goal was ideally to achieve prediction errors of feed factors within 0.1 g. Tahir et al (2020) considered prediction errors of feed factors of 0.25–0.43 g high and successfully reduced the predictive errors to 0.10–0.17 g by constructing models for clusters of powders with similar material properties. However, the absolute values of feed factors in the present study are significantly larger than those of other studies, and the acceptable absolute errors become larger.…”

“…The feed factor [g] was calculated using the flow rate [g/min] and the screw rotation speed [rpm] ( Tahir et al, 2020 ). is the instantaneous flow rate and calculated every second.…”

“… Chen and Ierapetritou (2020) ; Shier et al (2022) developed gray-box models based on Heckel's equation ( Heckel, 1961 ) based physical model proposed by ( Wang et al, 2017b ). The statistical models were developed to link the feeding performance with material properties, process parameters, and feeder configurations ( Jia et al, 2009 ; Pordesimo et al, 2009 ; Wang et al, 2017a ; Santos et al, 2018 ; Escotet-Espinoza et al, 2018 ; Wang et al, 2019 ; Bostijn et al, 2019 ; Stauffer et al, 2019 ; Yadav et al, 2019 ; Tahir et al, 2020 ; Bekaert et al, 2021 ).…”

“…In other studies working on the prediction of FFP, many experiments using various kinds of powder materials were conducted for model development. For example, Tahir et al (2020) conducted 224 experiments. They also assume large databases of material properties.…”

Classification of granular materials via flowability-based clustering with application to bulk feeding

“…Moreover, these statistical tools have been used to predict screw feeder performance at powder feeding rates between 0.1 kg h - 1 and 99.6 kg h - 1 [17--22]. Particularly, PCA has been reported to enhance the \sansnin \sansfiv visualisation of pharmaceutical powder clusters with similar flowability [21].…”