A new hybrid approach for the prediction of corrosion due to DC stray currents emitted by electrical traction and nonlinear dynamic process monitoring

• K. Khattab^1,2, E.B. Azzag³, K. Allali⁴ and B. Hamaidi²

¹ Badji Mokhtar University, Faculty of Technology, Department of Electromechanics, 23000, Annaba, Algeria
² Badji Mokhtar University, Electromechanical Engineering Laboratory, 23000 Annaba, Algeria
³ Badji Mokhtar University, Faculty of Technology, Department of Electrical Engineering, 23000, Annaba, Algeria
⁴ Mentouri Brothers University, Faculty of Science and Technology, Transport Engineering Department, 25000, Constantine-1, Algeria

Abstract: The purpose of this paper is to present the results of an analysis of buried networks’ susceptibility to stray currents and to define the methods and tools for tracking their damaging effects on potential victims as a result of the construction of tramways near gas networks. With the development of urban transport by electric tramways located near to buried metal pipes, corrosion defects due to stray currents have become a significant threat to these metal pipes, especially in Algeria. The ANN method has been chosen to solve the non-linear problems arising from leakage currents. This paper’s originality lies in the use of advanced machine learning techniques, including ANN, the least squares method, and stepwise with interaction filtered by Lasso regression, to investigate the effects of DC stray current for the first time in Algeria. Finally, a sensitivity analysis using a Monte Carlo simulation algorithm was performed for validation of the method. This study was done using real data collected near the tramway in the city of Constantine (East Algeria).

Keywords: gas pipelines, stray current corrosion defect, potential, cathodic protection, artificial neural networks

Int. J. Corros. Scale Inhib., 2023, 12, no. 1, 346-365
doi: 10.17675/2305-6894-2023-12-1-20

Download PDF (Total downloads: 70)

Back to this issue content: 2023, Vol. 12, Issue 1 (pp. 1-365)