An Improved Switching Method-Based Diagnostic Strategy for IGBT Open-Circuit Faults in Hybrid Modular Multilevel Converters

  • Hong Wu
  • , Yue Wang
  • , Yi Liu
  • , Yonghui Liu
  • , Yufei Li

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

The hybrid modular multilevel converter (HMMC) has attracted extensive attention due to its dc fault ride-through capability. However, the research on its insulated gate bipolar transistor (IGBT) open-circuit fault (OCF) diagnosis is still barely seen, which becomes a research gap that may jeopardize the reliability of the HMMC system. To address this issue, this article first proposes an improved switching method, which can ensure all six types of IGBT OCFs unveil their fault characteristics thoroughly by using an extra bypassed switching sequence. Furthermore, the improved switching method remains the same switching loss as the existing switching method. Afterward, a diagnostic method is brought forward to identify the faulty SM by its fixed specific sequence in a period of time while the switching sequences of other SMs with the same type are dynamic. On this basis, an improved switching method-based diagnostic strategy that consists of various types of diagnostic methods is proposed, which can diagnose all six types of IGBT OCFs in the HMMC within 20 ms and have immunity to dc faults and load change. Experimental results in a hardware-in-the-loop platform verify the effectiveness of the proposed switching method and diagnostic strategy.

Original languageEnglish
Pages (from-to)3578-3599
Number of pages22
JournalIEEE Transactions on Power Electronics
Volume40
Issue number2
DOIs
StatePublished - 2025

Keywords

  • Diagnostic strategy
  • hybrid modular multilevel converters
  • open-circuit fault
  • switching loss
  • switching method

Fingerprint

Dive into the research topics of 'An Improved Switching Method-Based Diagnostic Strategy for IGBT Open-Circuit Faults in Hybrid Modular Multilevel Converters'. Together they form a unique fingerprint.

Cite this