Predicting the non-linear conveying behavior in single-screw extrusion : a comparison of various data-based modeling approaches used with cfd simulations / W. Roland in INTERNATIONAL POLYMER PROCESSING, Vol. 36, N° 5 (2021)  

[article]  inINTERNATIONAL POLYMER PROCESSING > Vol. 36, N° 5 (2021) . - p. 529-544 Titre : Predicting the non-linear conveying behavior in single-screw extrusion : a comparison of various data-based modeling approaches used with cfd simulations Type de document : texte imprimé Auteurs : W. Roland, Auteur ; C. Marschik, Auteur ; M. Kommenda, Auteur ; A. Haghofer, Auteur ; S. Dorl, Auteur ; S. Winkler, Auteur Année de publication : 2021 Article en page(s) : p. 529-544 Note générale : Bibliogr. Langues : Anglais (eng) Catégories : Extrudeuse monovis Matières plastiques -- Extrusion:Polymères -- Extrusion Prévision, Théorie de la Rhéologie Simulation par ordinateur Index. décimale : 668.4 Plastiques, vinyles Résumé : The traditional approach to modeling the polymer melt flow in single-screw extruders is based on analytical and numerical analyses. Due to increasing computational power, data-driven modeling has grown significantly in popularity in recent years. In this study, we compared and evaluated databased modeling approaches (i. e., gradient-boosted trees, artificial neural networks, and symbolic regression models based on genetic programming) in terms of their ability to predict – within a hybrid modeling framework – the three-dimensional non-linear throughput-pressure relationship of metering channels in single-screw extruders. By applying the theory of similarity to the governing flow equations, we identified the characteristic dimensionless influencing parameters, which we then varied to create a large dataset covering a wide range of possible applications. For each single design point we conducted numerical simulations and obtained the dimensionless flow rate. The large dataset was divided into three independent sets for training, interpolation, and extrapolation, the first being used to generate and the remaining two to evaluate the models. Further, we added two features derived from expert knowledge to the models and analyzed their influence on predictive power. In addition to prediction accuracy and interpolation and extrapolation capabilities, we evaluated model complexity, interpretability, and time required to learn the models. This study provides a rigorous analysis of various data-based modeling approaches applied to simulation data in extrusion. Note de contenu : - MELT-CONVEYING SIMULATIONS : Governing equations - Dimensionless representation - Traditional melt-conveying model - Parametric design study - Numerical simulations DATA-BASED MODELING : Data preparation – Modeling - Modeling results - Table 1 : Range of values for n, t/Db, and h/w of the new (merged) dataset - Table 2 : Range of Pp,z depending on the power-law exponent of the merged dataset. The numbers of equidistant steps are 61 and 31, respectively, for the "original" and the “additional" data spaces - Table 3 : Number of samples per dataset after data splitting - Table 4 : Value ranges of each feature for the three different datasets - Table 5 : Model performance for the regression task of predicting ΠV from the influencing parameters - Table 6 : Model performance for the regression task of predicting ΠV. Comparing the influence of the additional features F1 and F2 on the prediction error - Table 7 : Performance of the traditional Newtonian melt-conveying models for predicting ΠV - Table 8 : Model constants for the best performing symbolic regression model DOI : https://doi.org/10.1515/ipp-2020-4094 En ligne : https://drive.google.com/file/d/12pNqnSQWNmTVTc1gCK_cszVmf9dWA1eR/view?usp=shari [...] Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=36520 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
23735	-	Périodique	Bibliothèque principale	Documentaires	Disponible

Using symbolic regression models to predict the pressure loss of non-newtonian polymer-melt flows through melt-filtration systems with woven screens / S. Pachner in INTERNATIONAL POLYMER PROCESSING, Vol. 36, N° 4 (2021)  

[article]  inINTERNATIONAL POLYMER PROCESSING > Vol. 36, N° 4 (2021) . - p. 435-450 Titre : Using symbolic regression models to predict the pressure loss of non-newtonian polymer-melt flows through melt-filtration systems with woven screens Type de document : texte imprimé Auteurs : S. Pachner, Auteur ; W. Roland, Auteur ; M. Aigner, Auteur ; C. Marschik, Auteur ; U. Stritzinger, Auteur ; J. Miethlinger, Auteur Année de publication : 2021 Article en page(s) : p. 435-450 Note générale : Appendice - Nomenclature - Bibliogr. Langues : Anglais (eng) Catégories : Armures (tissage) L'armure, dans la terminologie du tissage de textile, est le mode d'entrecroisement des fils de chaîne et des fils de trame. Il existe trois armures principales : toile, sergé et satin. À partir de ces trois armures sont fabriqués de nombreux tissus avec aussi des combinaisons des armures principales. Equations Etat fondu (matériaux) Filtration Fluides non newtoniens Polymères -- Recyclage Pression Rhéologie Simulation par ordinateur Index. décimale : 668.4 Plastiques, vinyles Résumé : When selecting a melt-filtration system, the initial pressure drop is a critical parameter. We used heuristic optimization algorithms to develop general analytical equations for estimating the dimensionless pressure loss of square and Dutch woven screens in polymer processing and recycling. We present a mathematical description – without the need for further numerical methods – of the dimensionless pressure loss of non-Newtonian polymer melt-flows through woven screens. Applying the theory of similarity, we first simplified, and then transformed into dimensionless form, the governing equations. By varying the characteristic independent dimensionless influencing parameters, we created a comprehensive parameter set. For each design point, the nonlinear governing equations were solved numerically. We subsequently applied symbolic regression based on genetic programming to develop models for the dimensionless pressure drop. Finally, we validated our models against experiments using both virgin and slightly contaminated in-house and post-industrial recycling materials. Our regression models predict the experimental data accurately, yielding a mean relative error of MRE = 13.7%. Our modeling approach, the accuracy of which we have proven, allows fast and stable prediction of the initial pressure drop of polymer-melt flows through square woven and Dutch weave screens, rendering further numerical simulations unnecessary. Note de contenu : - Non-newtonian flow analysis - Geometry - Governing equations - Theory of similarity - Numerical solution : Numerical simulation - Numerical results and discussion - Regression analysis : Symbolic regression modeling - Symbolic regression results and discussion - Experimental validation : - Table 1 : Performance of various filter media - Table 2 : Range of values of PD and n defined by minimum, maximum and step size - Table 3 : Range of values of A, D and n, defined by minimum, maximum and step size - Table 4 : Target and input variables for symbolic regression modeling - Table 5 : Mathematical building blocks provided to the heuristic optimization algorithm - Table 6 : Results of the error analysis assessing the accuracy of the models developed - Table 7 : Filtration parameters of the square-woven (SW) and Dutch-weave (DW) metal screens used - Table 8 : Results of the error analysis of the approximation models for square-weave and Dutch-weave screens DOI : https://doi.org/10.1515/ipp-2020-4019 En ligne : https://drive.google.com/file/d/13QLZw5Ek03dOo1EksPxH18yFIx-s1Fr9/view?usp=shari [...] Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=36717 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
23734	-	Périodique	Bibliothèque principale	Documentaires	Disponible

---

1 (1 - 2 / 2)