Tucker Decomposition-Based Feature Selection and SSA-Optimized Multi-Kernel SVM for Transformer Fault Diagnosis

Luping Wang, Xiaolong Liu, Zhihui Lai · 2025

Read Tucker Decomposition-Based Feature Selection and SSA-Optimized Multi-Kernel SVM for Transformer Fault Diagnosis by Luping Wang, Xiaolong Liu, Zhihui Lai (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12736887.txt.

Description

Luping Wang and co-authors build a three-stage pipeline to name the fault inside a sick power transformer from the gases in its oil, in this 2025 paper. Dissolved gas analysis is the standard diagnostic, but five gas readings carry tangled, nonlinear clues that simple ratios mishandle. Their framework engineers its way out.

First, the five measurements are expanded into twelve features, adding standard ratio indicators and statistical summaries. A Tucker decomposition then squeezes the set to seven features while keeping 95 percent of the discriminating signal, scoring each feature by both class separation and structural weight. Finally, a multi-kernel support vector machine, mixing three kernel shapes, classifies seven fault types. Its weights and settings are tuned by a sparrow search algorithm, which balances scouts and followers while hunting the best configuration.

Catalog record

CREATORLuping Wang, Xiaolong Liu, Zhihui Lai
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE43 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC12736887

Browse the full library · About the Tower of Babel