Remaining Life of a Pipeline
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
6.2. About the Statistical and Reliability Analysis<br />
As it was mentioned in section five (5), the “advanced first order-second moment method” allows us<br />
to estimate its mean value and its standard deviation <strong>of</strong> the remaining strength (Pf), but not its<br />
“distribution function” or pdf (Pf). Therefore the level <strong>of</strong> confidence <strong>of</strong> this estimate can not be<br />
obtained.<br />
To solve this problem, a Montecarlo Approach is proposed. This approach can be really useful if a<br />
precise estimation <strong>of</strong> the confidence interval is required. These calculation where developed using<br />
“MathCad 5”.<br />
Probabilistic estimation <strong>of</strong> remaining life <strong>of</strong> a pipeline<br />
in the precence <strong>of</strong> active corrosion defects<br />
Terms definitions:<br />
Sy = yield strength <strong>of</strong> pipe material<br />
T = Wall Thickness<br />
Do = Measured depth <strong>of</strong> the defect at t=to<br />
Lo = Measured length <strong>of</strong> the defect at t=to<br />
D = Pipe diameter<br />
Rd = Steady-state corrosion rate in the direction <strong>of</strong> depth<br />
Rl = Steady-state corrosion rate in the direction <strong>of</strong> length<br />
M = Folias factor = 1 0.6275. L2<br />
DT . 0.003375.<br />
L 4<br />
D 2 . T 2<br />
L = length <strong>of</strong> the defect at any time (t) = Lo+Rl(t-to)<br />
d = depth <strong>of</strong> the defect at any time (t) = Do=Rd(t-to)<br />
19
Change language