Scholarlink

<p>Modeling single-screw extrusion processes is essential for process optimization, yet existing comprehensive models are either computationally expensive or limited to steady-state conditions, hindering their use for dynamic scenario evaluation and offline optimization studies. This study presents a dynamic 1D model (spatial discretization along the axial screw direction) that integrates mass, momentum, and energy balances across all key extrusion zones—solids conveying, melting, melt conveying, and die flow—in a fully coupled framework where all balance equations are solved simultaneously. The model was implemented for a conventional screw geometry with smooth feed section using upwind finite difference schemes, resulting in a differential–algebraic equation (DAE) system validated against experimental data from an industrial extruder processing polypropylene (PP). The model achieves high predictive accuracy with Mean Absolute Error (MAE) values of 7.45&#xa0;<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\times\)</EquationSource> <EquationSource Format="MATHML"><math> <mo>×</mo> </math></EquationSource> </InlineEquation>&#xa0;10<sup>−6</sup>&#xa0;m<sup>2</sup> for solid flow area, 0.51–8.96&#xa0;bar for pressure, and 1.66&#xa0;°C for melt temperature, while maintaining computational efficiency with simulation times of 625–875&#xa0;s (simulation-to-real-time ratios of 2.7:1 to 6.7:1). As far as we know, this is the first dynamic 1D model providing transient predictions across process zones in a unified framework, overcoming the steady-state limitations of previous comprehensive models. This balance between accuracy and computational speed enables offline process optimization, parametric studies, and sensitivity analysis for process engineers, production engineers, and researchers. Additionally, its potential as a generator of synthetic data for the training of machine learning models positions it as a tool for developing soft sensors and advanced control strategies with reduced sensor requirements.</p> Graphical abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Dynamic modeling and simulation of mass, momentum, and energy transfer in a single-screw plastic extrusion process

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月19日

查看全文
  • Fabian Luna,
  • Juan C. Maya,
  • Farid Chejne

摘要

Modeling single-screw extrusion processes is essential for process optimization, yet existing comprehensive models are either computationally expensive or limited to steady-state conditions, hindering their use for dynamic scenario evaluation and offline optimization studies. This study presents a dynamic 1D model (spatial discretization along the axial screw direction) that integrates mass, momentum, and energy balances across all key extrusion zones—solids conveying, melting, melt conveying, and die flow—in a fully coupled framework where all balance equations are solved simultaneously. The model was implemented for a conventional screw geometry with smooth feed section using upwind finite difference schemes, resulting in a differential–algebraic equation (DAE) system validated against experimental data from an industrial extruder processing polypropylene (PP). The model achieves high predictive accuracy with Mean Absolute Error (MAE) values of 7.45  \(\times\) ×  10−6 m2 for solid flow area, 0.51–8.96 bar for pressure, and 1.66 °C for melt temperature, while maintaining computational efficiency with simulation times of 625–875 s (simulation-to-real-time ratios of 2.7:1 to 6.7:1). As far as we know, this is the first dynamic 1D model providing transient predictions across process zones in a unified framework, overcoming the steady-state limitations of previous comprehensive models. This balance between accuracy and computational speed enables offline process optimization, parametric studies, and sensitivity analysis for process engineers, production engineers, and researchers. Additionally, its potential as a generator of synthetic data for the training of machine learning models positions it as a tool for developing soft sensors and advanced control strategies with reduced sensor requirements.

Graphical abstract