Pigging data (also known as In Line Inspection) provides information on the corrosion defect geometry, such as depth and length, orientation, location and the corrosion type. Every few years pipeline are inspected and such data is gathered.
I need data for the same pipeline over a number of years. I am going to use the data to train a neural network, which will be used to manage pipelines integrity.
I am wondering if someone would kindly provide me data for several years or even better for several pipelines over several years.
- It is helpful to review API 570 and review corrosion rate determination sub-section to dertermine the corrosion rate for a specified peroid and compare with the proposed three options in the standard.
Hi Mohamed,
Thank you for your attention.
Perhaps I was not quite clear. I am looking for actual field data measured over several years. This is to validate a method that I have developed for the prediction of corrosion evolution of pipeline. The intention is to see if the method is any good.
There are papers on the training of neural network, (as well as other regression and probabilistic methods), for corrosion progression. Authors of these papers did not included their data. Thus, the reported results in these papers cannot be replicated.
Codified recommendations are suitable for the design, not for the validation of a theory.
regards
Sirous
- Badges
- Science topic