Interpretable Machine Learning Analysis of Design Factors in Hydrogel Supercapacitors

Understanding the relationships between design factors is crucial for the development of hydrogel supercapacitors, yet the relative importance and interdependencies of material properties and operating conditions remain unclear. This study employs interpretable machine learning to analyze the design...

Full description

Saved in:
Bibliographic Details
Main Authors: Liying Xu, Siqi Liu, Dandan Hu, Junhao Liu, Yuze Zhang, Ziqiang Li, Zichuan Su, Daxin Liang
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Gels
Subjects:
Online Access:https://www.mdpi.com/2310-2861/11/6/464
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849433423832154112
author Liying Xu
Siqi Liu
Dandan Hu
Junhao Liu
Yuze Zhang
Ziqiang Li
Zichuan Su
Daxin Liang
author_facet Liying Xu
Siqi Liu
Dandan Hu
Junhao Liu
Yuze Zhang
Ziqiang Li
Zichuan Su
Daxin Liang
author_sort Liying Xu
collection DOAJ
description Understanding the relationships between design factors is crucial for the development of hydrogel supercapacitors, yet the relative importance and interdependencies of material properties and operating conditions remain unclear. This study employs interpretable machine learning to analyze the design factors that affect hydrogel supercapacitor performance, using 232 experimental samples from 41 recent studies. SHAP analysis was implemented to quantify parameter importance and reveal feature interactions among 16 key design parameters, including polymer types, electrolyte formulations, and operating conditions. Results show that synthetic vinyl polymers most strongly influence specific capacitance, while conductive polymers predominantly affect cycle stability. Ionic conductivity emerged as the most impactful parameter despite moderate feature importance, indicating complex nonlinear relationships. Critical interdependencies between polymer concentration and electrolyte formulation suggest that optimal design requires coordinated parameter selection rather than independent optimization. This interpretable framework provides quantitative insights into design factor hierarchies and parameter interdependencies, offering evidence-based guidelines for rational material selection in hydrogel supercapacitor development.
format Article
id doaj-art-458e6c7fe4e34c19960f282fdb76394c
institution Kabale University
issn 2310-2861
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Gels
spelling doaj-art-458e6c7fe4e34c19960f282fdb76394c2025-08-20T03:27:02ZengMDPI AGGels2310-28612025-06-0111646410.3390/gels11060464Interpretable Machine Learning Analysis of Design Factors in Hydrogel SupercapacitorsLiying Xu0Siqi Liu1Dandan Hu2Junhao Liu3Yuze Zhang4Ziqiang Li5Zichuan Su6Daxin Liang7School of Food Engineering, Harbin University, Harbin 150086, ChinaKey Laboratory of Bio-Based Material Science and Technology (Ministry of Education), Northeast Forestry University, Harbin 150040, ChinaKey Laboratory of Bio-Based Material Science and Technology (Ministry of Education), Northeast Forestry University, Harbin 150040, ChinaKey Laboratory of Bio-Based Material Science and Technology (Ministry of Education), Northeast Forestry University, Harbin 150040, ChinaKey Laboratory of Bio-Based Material Science and Technology (Ministry of Education), Northeast Forestry University, Harbin 150040, ChinaKey Laboratory of Bio-Based Material Science and Technology (Ministry of Education), Northeast Forestry University, Harbin 150040, ChinaCollege of Chemistry and Chemical Engineering, China University of Petroleum (East China), Qingdao 266580, ChinaKey Laboratory of Bio-Based Material Science and Technology (Ministry of Education), Northeast Forestry University, Harbin 150040, ChinaUnderstanding the relationships between design factors is crucial for the development of hydrogel supercapacitors, yet the relative importance and interdependencies of material properties and operating conditions remain unclear. This study employs interpretable machine learning to analyze the design factors that affect hydrogel supercapacitor performance, using 232 experimental samples from 41 recent studies. SHAP analysis was implemented to quantify parameter importance and reveal feature interactions among 16 key design parameters, including polymer types, electrolyte formulations, and operating conditions. Results show that synthetic vinyl polymers most strongly influence specific capacitance, while conductive polymers predominantly affect cycle stability. Ionic conductivity emerged as the most impactful parameter despite moderate feature importance, indicating complex nonlinear relationships. Critical interdependencies between polymer concentration and electrolyte formulation suggest that optimal design requires coordinated parameter selection rather than independent optimization. This interpretable framework provides quantitative insights into design factor hierarchies and parameter interdependencies, offering evidence-based guidelines for rational material selection in hydrogel supercapacitor development.https://www.mdpi.com/2310-2861/11/6/464hydrogel supercapacitormachine learninginterpretable prediction
spellingShingle Liying Xu
Siqi Liu
Dandan Hu
Junhao Liu
Yuze Zhang
Ziqiang Li
Zichuan Su
Daxin Liang
Interpretable Machine Learning Analysis of Design Factors in Hydrogel Supercapacitors
Gels
hydrogel supercapacitor
machine learning
interpretable prediction
title Interpretable Machine Learning Analysis of Design Factors in Hydrogel Supercapacitors
title_full Interpretable Machine Learning Analysis of Design Factors in Hydrogel Supercapacitors
title_fullStr Interpretable Machine Learning Analysis of Design Factors in Hydrogel Supercapacitors
title_full_unstemmed Interpretable Machine Learning Analysis of Design Factors in Hydrogel Supercapacitors
title_short Interpretable Machine Learning Analysis of Design Factors in Hydrogel Supercapacitors
title_sort interpretable machine learning analysis of design factors in hydrogel supercapacitors
topic hydrogel supercapacitor
machine learning
interpretable prediction
url https://www.mdpi.com/2310-2861/11/6/464
work_keys_str_mv AT liyingxu interpretablemachinelearninganalysisofdesignfactorsinhydrogelsupercapacitors
AT siqiliu interpretablemachinelearninganalysisofdesignfactorsinhydrogelsupercapacitors
AT dandanhu interpretablemachinelearninganalysisofdesignfactorsinhydrogelsupercapacitors
AT junhaoliu interpretablemachinelearninganalysisofdesignfactorsinhydrogelsupercapacitors
AT yuzezhang interpretablemachinelearninganalysisofdesignfactorsinhydrogelsupercapacitors
AT ziqiangli interpretablemachinelearninganalysisofdesignfactorsinhydrogelsupercapacitors
AT zichuansu interpretablemachinelearninganalysisofdesignfactorsinhydrogelsupercapacitors
AT daxinliang interpretablemachinelearninganalysisofdesignfactorsinhydrogelsupercapacitors