We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Non-Markovian model for the study of pitting corrosion in a water pipe system.
- Authors
Rosa, A. C. P.; Vaveliuk, P.; Moret, M. A.
- Abstract
The main studies on pitting consist in proposing Markovian stochastic models, based on the statistics of extreme values and focused on growing the depth of wells, especially the deepest one. We show that a non-Markovian model, described by a nonlinear Fokker-Planck (nFP) equation, properly depicts the time evolution of a distribution of depth values of pits that were experimentally obtained. The solution of this equation in a steady-state regime is a q-Gaussian distribution, i.e. a long-tail probability distribution that is the main characteristic of a nonextensive statistical mechanics. The proposed model, that is applied to data from four inspections conducted on a section of a line of regular water service in power water reactor (PWR) nuclear power plants, is in agreement with experimental results.
- Subjects
MARKOVIAN jump linear systems; PITTING corrosion; WATER pipelines; STOCHASTIC models; NUCLEAR power plants
- Publication
International Journal of Modern Physics C: Computational Physics & Physical Computation, 2015, Vol 26, Issue 10, p-1
- ISSN
0129-1831
- Publication type
Article
- DOI
10.1142/S0129183115501193