Exploring the potential of transfer learning in extrapolating accelerated corrosion test data for long-term atmospheric corrosion forecasting
Source abstract
Abstract
This study utilizes transfer learning (TL) to enhance long-term atmospheric corrosion predictions. Using a Fe/Cu galvanic-type sensor, we gathered data in a controlled SAE J2334 salt spray setting and transferred this to an uncontrolled outdoor environment. Among TL methods tested, freezing the initial layer and fine-tuning others at a lower rate was most effective. The approach excelled at forecasting outdoor corrosion behaviour using a limited dataset. This approach could provide a solution to extrapolate results from controlled corrosion tests to unpredictable outdoor conditions and addressing data scarcity in machine learning modelling in the context of atmospheric corrosion.
Source: openaire · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Non-Destructive Testing Techniques · Infrastructure Maintenance and Monitoring · Water Systems and Optimization
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (39)
Show source-provided references
1. Atmospheric corrosivity modeling — a review10.1016/S0261-3069(01)00051-6 ↗
2. Leygraf · 2016
3. Atmospheric corrosion prediction: a review10.1515/corrrev-2019-0100 ↗
4. The classification system of ISO 9223 standard and the dose–response functions assessing the corrosivity of outdoor atmospheres10.1023/B:PROM.0000049517.14101.68 ↗
5. Atmospheric corrosion modeling10.1515/corrrev-2014-0023 ↗
6. Effect of environmental conditions on corrosion rates10.1061/(ASCE)0899-1561(2007)19:2(121) ↗
7. Corrosion monitoring in atmospheric conditions: a review10.3390/met12020171 ↗
8. Investigation of corrosion behaviors on an fe/cu-type acm sensor under various environments10.3390/met10070905 ↗
9. Corrosion monitoring and materials selection for automotive environments by using Atmospheric Corrosion Monitor (ACM) sensor10.1016/j.corsci.2014.02.020 ↗
10. A method for estimating time-dependent corrosion depth of carbon and weathering steel using an atmospheric corrosion monitor sensor10.3390/s19061416 ↗
11. Corrosion assessment of carbon steel in Thailand by atmospheric corrosion monitoring (ACM) sensors
12. Electrochemical monitoring of atmospheric corrosion phenomena10.1016/0010-938X(76)90052-4 ↗
13. Choosing an accelerated corrosion test10.1016/S0026-0576(00)83115-3 ↗
14. Accelerated corrosion testing10.1016/S0026-0576(10)80256-9 ↗
15. Comparing modeled and experimental accelerated corrosion tests on steel10.1149/2.0951709jes ↗
16. Extrapolating short-term corrosion test results to field exposures in different environments10.1016/j.corsci.2021.109455 ↗
17. Reviewing machine learning of corrosion prediction in a data-oriented perspective
18. Towards understanding and prediction of atmospheric corrosion of an Fe/Cu corrosion sensor via machine learning10.1016/j.corsci.2020.108697 ↗
19. L. Mansilla R. Echeveste D.H. Milone E. Ferrante Domain Gen. via Gradient Surg. 2021.
20. S.J. Pan Q. Yang A survey on transfer learning IEEE Trans. Knowl. Data Eng. 22 2010 1345 1359.10.1109/TKDE.2009.191 ↗
21. Transfer learning with time series data: a systematic mapping study10.1109/ACCESS.2021.3134628 ↗
22. A comprehensive survey on transfer learning10.1109/JPROC.2020.3004555 ↗
23. Aircraft fuselage corrosion detection using artificial intelligence10.3390/s21124026 ↗
24. Inspection of surface defects on stay cables using a robot and transfer learning10.1016/j.autcon.2020.103382 ↗
25. CorrDetector: a framework for structural corrosion detection from drone images using ensemble deep learning10.1016/j.eswa.2021.116461 ↗
26. A transfer-learning approach for corrosion prediction in pipeline infrastructures10.1007/s10489-021-02771-y ↗
27. Update on the developments of the SAE J2334 laboratory cyclic corrosion test
28. Understanding environmental impacts on initial atmospheric corrosion based on corrosion monitoring sensors10.1016/j.jmst.2020.01.023 ↗
29. Investigation of corrosion behaviors on an fe/cu-type acm sensor under various environments10.3390/met10070905 ↗
30. Long-term corrosion monitoring of carbon steels and environmental correlation analysis via the random forest method10.1038/s41529-021-00211-3 ↗
31. An integrated modeling approach for atmospheric corrosion in presence of a varying electrolyte film10.1016/j.electacta.2015.11.010 ↗
32. Influence of environmental factors on atmospheric corrosion in dynamic environment10.1016/j.corsci.2018.03.042 ↗
33. R. Mushtaq Test. TIME Ser. DATA STATIONARITY 2011.〈http://ssrn.com/abstract=1911068Electroniccopyavailableat:https://ssrn.com/abstract=1911068〉.
34. The asymptotic size and power of the augmented Dickey–Fuller test for a unit root10.1080/00927872.2016.1178887 ↗
35. Long-term atmospheric corrosion of mild steel10.1016/j.corsci.2010.10.007 ↗
36. Corrosion behavior of weathering steel in marine atmosphere10.1016/S0254-0584(02)00110-4 ↗
37. Goodfellow, I.J., Mirza, M., Xiao, D., Courville, A. Bengio, Y. An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks. Preprint at (2015).
38. An overview of overfitting and its solutions
39. First steps toward understanding the extrapolation of nonlinear models to unseen domains