Field data collection, evaluation and use for corrosivity prediction and validation of models. Part II: Evaluation of field data and comparison of prediction models
Nyborg, R.
Corrosion 2006., San Diego, Calif., 2006-03-12--03-16. Paper 06118
- Publ. year
- 2006
- Publ. type
- paper
- Abstract
- A corrosion field database has been developed by collecting corrosion field data gathered by several operators. This has proven to be a difficult exercise, as the amount of reliable corrosion field data is scarce. The amount of available data varied considerably from case to case. For many of the cases where corrosion problems were encountered it was difficult to trace back to obtain all the relevant information from earlier stages of the history of the field. The different corrosion prediction models were evaluated against field data by running different corrosion prediction models for a set of cases taken from the corrosion field database. It varied considerably from case to case which models were most successful in their prediction, and it was not possible to declare one or two models as better than the others. An evaluation of any model against field data can be strongly dependent on the selection of field data used for the evaluation or validation and the validity and accuracy of the field data. It is important to understand the limitations and uncertainties in the corrosion prediction models. ©NACE International. All rights reserved. Paper reproduced with permission from CORROSION/2006 Annual Conference and Exhibition, San Diego, California. <A HREF="http://www.nace.org" target=_blank>www.nace.org</A>
Corr06118.pdf
—
PDF document,
300 kB (307375 bytes)
