.png){kind=icon}
.png){kind=icon}
Return
Stray current inversion method based on sparrow search algorithm with kernel extreme learning machine
.png){kind=icon}
PRE
AI
DOI：10.1007/s13042-026-02998-6.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
With the continuous advancement of engineering construction, stray current has become a typical factor causing pipeline corrosion. Therefore, it is particularly important to strengthen the research on stray current detection to enhance the technical capability of pipeline corrosion prevention. In this paper, a stray current inversion method based on Sparrow Search Algorithm with Kernel Extreme Learning Machine (SSA-KELM) is proposed, and realize the quantitative analysis of stray currents and defects in complex situations. Specifically speaking, after defining and extracting the features of the electrical signal, the KELM model is parameter optimized using the SSA algorithm. The results of single- and double-output inversion of the same test samples by SSA-KELM model, the basic ELM model and KELM model are compared to verify the superiority of SSA-KELM model in the inversion of stray currents and defects, and to realize the quantitative analysis of stray currents and defects in complex situations. The experimental results demonstrate that the $$R^2, MSE, MAPE$$ , and RPD of our model reach 0.93348, 5.4197, 13.976%, and 6.9021 for the single output of defects, respectively; for the dual output of defects and stray currents, the above four metrics for defects outputs are 0.89583, 29.8813, 40.757%, and 3.0416, respectively, and for stray currents 0.95794, 0.52179, 4.2657%, 6.8561.
Keywords:
Defect recognition
Metal pipe
Stray current
Quantitative testing
SSA-KELM
Journal
IF:
2.7
Papers:
3.1K
Citations:
5.6K