Skip to main navigation Skip to search Skip to main content

A Parameter Identification Method Based on Data and Physical Information

  • Cuiyu Liu*
  • , Ziwen Xiao
  • , Gang Xiang
  • , Zhiming Yang
  • , Yang Yu
  • *Corresponding author for this work
  • School of Electronics and Information Engineering, Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Data-driven parameter identification methods are prone to deviating from physical meaning and falling into local optimal solutions. To address this issue, this paper proposes a novel parameter identification approach that integrates both data and physical information. The contributions of this work include achieving accurate parameter identification by integrating physical models with machine learning, requiring less data, and eliminating the need for external signal injection. The method is applied to power electronic converters, and simulation results demonstrate its effectiveness and superiority over traditional methods in terms of accuracy and robustness.

Original languageEnglish
Title of host publicationIntelligent Manufacturing and Cloud Computing - Proceedings of the 2nd International Conference, ICIMCC 2025
EditorsIsabel S. Jesus, Ke Wang
PublisherIOS Press BV
Pages71-78
Number of pages8
ISBN (Electronic)9781643686561
DOIs
StatePublished - 13 Mar 2026
Externally publishedYes
Event2nd Annual International Conference on Intelligent Manufacturing and Cloud Computing, ICIMCC 2025 - Wuhan, China
Duration: 12 Dec 202514 Dec 2025

Publication series

NameAdvances in Transdisciplinary Engineering
Volume91
ISSN (Print)2352-751X
ISSN (Electronic)2352-7528

Conference

Conference2nd Annual International Conference on Intelligent Manufacturing and Cloud Computing, ICIMCC 2025
Country/TerritoryChina
CityWuhan
Period12/12/2514/12/25

Keywords

  • Parameter identification
  • physics-informed machine learning
  • power electronic converters

Fingerprint

Dive into the research topics of 'A Parameter Identification Method Based on Data and Physical Information'. Together they form a unique fingerprint.

Cite this