21.04.2016 Views

Remaining Life of a Pipeline

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Maximun Likelihood parameters estimation<br />

To completely define eq. (iii), it is necessary to find the values <strong>of</strong> the parameters “ν” and “σ E ”. This is<br />

possible by applying the method <strong>of</strong> Maximum Likelihood Estimator, as follows:<br />

ln( likelihood)<br />

= Λ =<br />

n<br />

∑<br />

i=<br />

1<br />

⎡ 1 ⎡ Ei<br />

Ni<br />

ln⎢<br />

φ ⎢<br />

⎣σ<br />

EEi<br />

⎣<br />

− E0e<br />

σ<br />

E<br />

−υt<br />

⎤⎤<br />

⎥⎥<br />

⎦<br />

⎦<br />

(iv)<br />

The parameters estimate “υˆ ” and “ σˆ ”, will be found by solving for all the evidence equation (iv),<br />

so that:<br />

E<br />

∂Λ<br />

∂υ<br />

= 0<br />

∂Λ<br />

= 0<br />

∂σ E<br />

(v)<br />

(vi)<br />

Once the estimators “υˆ ” and “ σˆ ” have been found based on the evidence, the equation (iv) is<br />

E<br />

totally defined for any value <strong>of</strong> t i , and it represents the distribution function <strong>of</strong> the stress.<br />

pdf<br />

E(<br />

t))<br />

= f ( E,<br />

t)<br />

=<br />

ˆ σ<br />

1<br />

e<br />

2π<br />

1 ⎡<br />

− ⎢<br />

2 ⎢⎣<br />

( E(<br />

t )<br />

−E<br />

e<br />

0<br />

2<br />

ˆ σ E<br />

(<br />

E<br />

) ⎤<br />

⎥<br />

⎥⎦<br />

− ˆ υt<br />

2<br />

(vii)<br />

34

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!