08.11.2014 Views

Probabilistic Performance Analysis of Fault Diagnosis Schemes

Probabilistic Performance Analysis of Fault Diagnosis Schemes

Probabilistic Performance Analysis of Fault Diagnosis Schemes

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Fix a parameter sequence θ = ϑ and let the residual generator F be partitioned as F = [ F 1 F 2<br />

]<br />

.<br />

Divide the residual into its non-random and random parts, as follows:<br />

where<br />

r = r unc + r rnd ,<br />

r unc = F 2 G 21,θ β + F 2 G 23,θ f (θ) + (F 2 G 24,θ + F 1 )u<br />

r rnd = F 2 G 22 v.<br />

Since v is zero-mean by assumption, the conditional mean <strong>of</strong> the residual at time k is<br />

and the conditional variance at time k is<br />

ˆr k = E(r k | θ 0:k = ϑ 0:k ) = r unc<br />

k<br />

,<br />

Σ k = E ( (r k − ˆr k ) 2) ( (r ) )<br />

rnd 2<br />

= E .<br />

Note that, as desired, the sequence {Σ k } does not depend on β or θ.<br />

k<br />

Maximizing the Probability <strong>of</strong> False Alarm<br />

Assume that no faults have occurred (ϑ = 0). Recall that the worst-case probability <strong>of</strong> false<br />

alarm is<br />

Pf ⋆ = 1 − min min P( )<br />

|r k | < ε k | θ 0:k = 0 0:k .<br />

∆∈P ∆ 0≤k≤N<br />

As explained in Section 5.2.2, the crux <strong>of</strong> computing P ⋆ f<br />

ˆr ⋆ k = max<br />

∣ ∣r unc∣ ∆∈P<br />

k ,<br />

∆<br />

is solving<br />

for k = 0,1,..., N . There are two cases to consider: P ∆ = ∆ 2,lti (γ) and P ∆ = ∆ 2,ltv (γ).<br />

Case 1. Suppose that ∆ belongs to the set ∆ 2,lti (γ) and assume that G 11,0 is an lti operator<br />

with ‖G 11,0 ‖ i 2 < 1 γ . Then, for k = 0,1,..., N, applying Theorem 5.6 yields the following<br />

optimization:<br />

ˆr ⋆ k = maximize<br />

β<br />

∣ r<br />

unc∣<br />

k<br />

subject to r unc = F 2 G 21,0 β + (F 2 G 24,0 + F 1 )u<br />

z = G 13,0 f (0) +G 14,0 u<br />

J (β) ≽ 0,<br />

96

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

Saved successfully!

Ooh no, something went wrong!