View in EDS HTML Full Text PDF Full Text

Machine-Learning-Assisted Viscoelastic Characterization of PC/ABS Blends via Multi-Frequency Dynamic Mechanical Analysis.

Saved in:
Bibliographic Details
Title: Machine-Learning-Assisted Viscoelastic Characterization of PC/ABS Blends via Multi-Frequency Dynamic Mechanical Analysis.
Authors: Sun, Yancai1,2,3,4 (AUTHOR), Deng, Wenzhong2,3 (AUTHOR), Wang, Haoran3,5 (AUTHOR), Jian, Ranran1,4 (AUTHOR), Bai, Wenjuan1,5 (AUTHOR), Chu, Dianming1,4,6 (AUTHOR), Hou, Peiwu1 (AUTHOR), He, Yan1,2,6 (AUTHOR) heyan@qust.edu.cn
Source: Polymers (20734360). Mar2026, Vol. 18 Issue 5, p599. 24p.
Subjects: Dynamic mechanical analysis, Viscoelasticity, Acrylonitrile butadiene styrene resins, Materials analysis, Machine learning, Rheology
Abstract: This study combines multi-frequency dynamic mechanical analysis (DMA) with machine learning (ML) to characterize and predict the viscoelastic properties of a commercial polycarbonate/acrylonitrile–butadiene–styrene (PC/ABS) blend. DMA temperature sweeps at four frequencies (1–10 Hz) in single cantilever mode yielded a glass transition range of 115.8– 123.2 °C ( E ″ peak), frequency sensitivity of 7.18 °C/ decade , and an apparent activation energy of 335 ± 85   k J   mol − 1 . Time–temperature superposition master curves were parameterized with a six-term Prony series ( R 2 = 0.998 ). Four data-driven models (RF, XGB, SVR, MLP) and a physics-informed NeuralWLF model were evaluated through a hierarchical validation framework. Temperature-blocked CV ranked MLP ( R 2 ¯ = 0.989 ) above RF (0.950) for interpolation; LOFO validation revealed that NeuralWLF achieved the best cross-frequency generalization ( R 2 > 0.92 for all targets) with interpretable WLF parameters ( C 1 ≈ 12.2 , C 2 ≈ 51.7   ° C ). A systematic block size sweep (5–30 ° C ) revealed a validation inflation effect in which MLP tan δ R 2 dropped from 0.986 to 0.592 as the gap-to-FWHM ratio increased from 0.5 to 3.1, establishing the gap/FWHM ratio as a quantitative validation stringency criterion. A physics–data crossover was identified at gap/FWHM ≈ 2 : beyond this threshold, NeuralWLF outperformed all data-driven models in tan δ prediction by up to + 0.300 in R 2 , while curriculum learning (freezing the WLF layer for 300 epochs) further improved the most stringent 30 ° C validation from R 2 = 0.660 to 0.731. The integrated framework demonstrates that honest evaluation of DMA–ML models requires validation gaps exceeding the characteristic feature width and introduces a quantifiable physics-data crossover criterion for selecting between data-driven and physics-informed architectures. [ABSTRACT FROM AUTHOR]
Copyright of Polymers (20734360) is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: egs
DbLabel: Engineering Source
An: 192641900
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Machine-Learning-Assisted Viscoelastic Characterization of PC/ABS Blends via Multi-Frequency Dynamic Mechanical Analysis.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Sun%2C+Yancai%22">Sun, Yancai</searchLink><relatesTo>1,2,3,4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Deng%2C+Wenzhong%22">Deng, Wenzhong</searchLink><relatesTo>2,3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Haoran%22">Wang, Haoran</searchLink><relatesTo>3,5</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Jian%2C+Ranran%22">Jian, Ranran</searchLink><relatesTo>1,4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Bai%2C+Wenjuan%22">Bai, Wenjuan</searchLink><relatesTo>1,5</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Chu%2C+Dianming%22">Chu, Dianming</searchLink><relatesTo>1,4,6</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Hou%2C+Peiwu%22">Hou, Peiwu</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22He%2C+Yan%22">He, Yan</searchLink><relatesTo>1,2,6</relatesTo> (AUTHOR)<i> heyan@qust.edu.cn</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Polymers+%2820734360%29%22">Polymers (20734360)</searchLink>. Mar2026, Vol. 18 Issue 5, p599. 24p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Dynamic+mechanical+analysis%22">Dynamic mechanical analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Viscoelasticity%22">Viscoelasticity</searchLink><br /><searchLink fieldCode="DE" term="%22Acrylonitrile+butadiene+styrene+resins%22">Acrylonitrile butadiene styrene resins</searchLink><br /><searchLink fieldCode="DE" term="%22Materials+analysis%22">Materials analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Rheology%22">Rheology</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: This study combines multi-frequency dynamic mechanical analysis (DMA) with machine learning (ML) to characterize and predict the viscoelastic properties of a commercial polycarbonate/acrylonitrile–butadiene–styrene (PC/ABS) blend. DMA temperature sweeps at four frequencies (1–10 Hz) in single cantilever mode yielded a glass transition range of 115.8– 123.2 °C ( E ″ peak), frequency sensitivity of 7.18 °C/ decade , and an apparent activation energy of 335 ± 85   k J   mol − 1 . Time–temperature superposition master curves were parameterized with a six-term Prony series ( R 2 = 0.998 ). Four data-driven models (RF, XGB, SVR, MLP) and a physics-informed NeuralWLF model were evaluated through a hierarchical validation framework. Temperature-blocked CV ranked MLP ( R 2 ¯ = 0.989 ) above RF (0.950) for interpolation; LOFO validation revealed that NeuralWLF achieved the best cross-frequency generalization ( R 2 > 0.92 for all targets) with interpretable WLF parameters ( C 1 ≈ 12.2 , C 2 ≈ 51.7   ° C ). A systematic block size sweep (5–30 ° C ) revealed a validation inflation effect in which MLP tan δ R 2 dropped from 0.986 to 0.592 as the gap-to-FWHM ratio increased from 0.5 to 3.1, establishing the gap/FWHM ratio as a quantitative validation stringency criterion. A physics–data crossover was identified at gap/FWHM ≈ 2 : beyond this threshold, NeuralWLF outperformed all data-driven models in tan δ prediction by up to + 0.300 in R 2 , while curriculum learning (freezing the WLF layer for 300 epochs) further improved the most stringent 30 ° C validation from R 2 = 0.660 to 0.731. The integrated framework demonstrates that honest evaluation of DMA–ML models requires validation gaps exceeding the characteristic feature width and introduces a quantifiable physics-data crossover criterion for selecting between data-driven and physics-informed architectures. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Polymers (20734360) is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=192641900
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.3390/polym18050599
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 24
        StartPage: 599
    Subjects:
      – SubjectFull: Dynamic mechanical analysis
        Type: general
      – SubjectFull: Viscoelasticity
        Type: general
      – SubjectFull: Acrylonitrile butadiene styrene resins
        Type: general
      – SubjectFull: Materials analysis
        Type: general
      – SubjectFull: Machine learning
        Type: general
      – SubjectFull: Rheology
        Type: general
    Titles:
      – TitleFull: Machine-Learning-Assisted Viscoelastic Characterization of PC/ABS Blends via Multi-Frequency Dynamic Mechanical Analysis.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Sun, Yancai
      – PersonEntity:
          Name:
            NameFull: Deng, Wenzhong
      – PersonEntity:
          Name:
            NameFull: Wang, Haoran
      – PersonEntity:
          Name:
            NameFull: Jian, Ranran
      – PersonEntity:
          Name:
            NameFull: Bai, Wenjuan
      – PersonEntity:
          Name:
            NameFull: Chu, Dianming
      – PersonEntity:
          Name:
            NameFull: Hou, Peiwu
      – PersonEntity:
          Name:
            NameFull: He, Yan
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 03
              Text: Mar2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 20734360
          Numbering:
            – Type: volume
              Value: 18
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: Polymers (20734360)
              Type: main
ResultId 1