Engineering Risk Benefit Analysis: Rpra 3. Probability Distributions in RPRA
Engineering Risk Benefit Analysis: Rpra 3. Probability Distributions in RPRA
Engineering Risk Benefit Analysis: Rpra 3. Probability Distributions in RPRA
1.155, 2.943, 3.577, 6.938, 10.816, 13.621, 16.862, 22.82, ESD.72, ESD.721
RPRA 3.
Spring 2007
RPRA 3. Probability Distributions in RPRA 1
Overview
We need models for: The probability that a component will start (fail) on demand. The probability that a component will run for a period of time given a successful start. The impact of repair on these probabilities. The frequency of initiating events.
Failure to start
N k N k Pr[ X = k ] = q (1 q ) k
This is the probability mass function of the Binomial Distribution. It is the probability of exactly k failures in N demands. The binomial coefficient is:
N N! k k! (N k )!
RPRA 3. Probability Distributions in RPRA 5
N k N k =1 q (1 q ) k =0 k
m N k
CDF
Notes: 1. We have assumed nominally identical and independent components in both cases. 2. If the components are not nominally identical or independent, the structure function approach still works, but the binomial distribution is not applicable. Why?
RPRA 3. Probability Distributions in RPRA 7
Pr[k ] = e
k! 1*2**(k-1)*k
( t )
k!
0! = 1
m = t
= t
8
t = 10-2*100 = 1 Pr[1 repl.] = e-t = e-1 = 0.37 = Pr[no replacement] Expected number of replacements: 1
2 1 1 e = 0.185 Pr[ 2repl] = e 1 = 2! 2
10
The distinction between h(t) and f(t) : f(t)dt: unconditional probability of failure in (t, t +dt), f(t)dt = P [t < T < t+dt] h(t)dt: conditional probability of failure in (t, t +dt) given that the component has survived up to t. h(t)dt = P [t < T < t+dt/{ t < T}]
11
I
0 t1
II
t2
III
t
I II III
12
F( t ) = 1 e
R(t ) = e t
h( t ) = constant (no memory; the only pdf with this property) useful life on bathtub curve
F( t ) t
XT = (XAXB+XBXC+XCXA) - 2XAXBXC
14
For nominally identical components: P(XT) = 3q2 2q3 (slide 7 of this lecture) But q( t ) = 1 e t F( t ) with = 5x10 4 hr 1
System Unreliability:
FT ( t ) = 3(1 e t ) 2 2(1 e t ) 3
15
t = 0.36
q( t ) = 1 e 0.36 = 0.30
FT(720) = 3 x 0.302 - 2 x 0.303 = 0.216 Since = 0.36 > 0.1 the rare-event approximation does not apply.
Proof
dR )dt = tdR = MTTF = tf (t )dt = t( dt 0 0 0
and
R(t ) 0 faster than t
18
MTTF = e
0
1 dt =
19
YT = Yj
1
MTBF =
1 1 M system
20
Fs ( t ) = (1 e t ) 2
R s ( t ) = 1 Fs ( t ) = 2e t e 2t
2 1 3 = MTTF = 2 2
21
1 The MTTF for a single exponential component is: The 2-out-of-3 system is slightly worse.
RPRA 3. Probability Distributions in RPRA 22
1 1 5 + = MTTF = 2 3 6
Adjusting the value of b, we can model any part of the bathtub curve.
0.0035 0.003 0.0025 0.002 0.0015 0.001 0.0005 0 b=0.8 b=1.5 b=1.0 b=2.0
h ( t ) = b t R(t ) = e
b b 1
b
200
400
600
800
1000
(t )
23
A Simple Calculation
The average rate of loss of electric power in a city is 0.08 per year. A hospital has an emergency diesel generator whose probability of starting successfully given loss of power is 0.95. i. What is the rate of occurrence of blackouts at this hospital?
Loss of power
=0.08 yr-1
b = 0.004 yr
-1
ok = 0.076 yr
-1
24
A Simple Calculation (cont.) b = 0.08 (power losses per year)x 0.05 (Diesel fails given a power loss) = 4x10
per year.
-3
ii. What is the probability that the hospital will have no blackouts in a period of five years? Exactly one blackout? At least one blackout?
b t = 0 . 004 x 5 = 0 . 02
Use the Poisson distribution: P(no blackouts in 5 yrs) = exp(-0.02) = 0.9802 P(exactly one blackout in 5 yrs) = (0.02)x0.9802 = 0.0196 P(at least one blackout in 5 yrs) = P(1 or 2 or 3) = = 1 - P(no blackouts in 5 yrs) = 1 0.9802 = 0.0198
RPRA 3. Probability Distributions in RPRA 25
f (x) =
1 (x ) exp[ ] 2 2 2
< x< 0<< < <
Mean:
Standard Deviation:
RPRA 3. Probability Distributions in RPRA 26
X Z=
1 z2 ( z ) = exp[ ] 2 2
27
28
= 10,000 hr (MTTF)
= 1,000 hr
29
An Example
A capacitor is placed across a power source. Assume that surge voltages occur on the line at a rate of one per month and they are normally distributed with a mean value of 100 volts and a standard deviation of 15 volts. The breakdown voltage of the capacitor is 130 volts.
30
An Example (2)
i. Find the mean time to failure (MTTF) for this capacitor. sv = 1 per month Pd/sv = conditional probability of damage given a surge voltage = P (surge voltage>130 volts/surge voltage) =
31
An Example (3)
Therefore, the rate of damaging surge voltages is d = sv xPd / sv = 1x 0.0228 = 2.28x10 2 (month)1
Equivalently, the capacitors failure time follows an exponential distribution with the above rate. The mean time between failures of the capacitor is
MTBF = 1 2.28x10
2
= 43.86
months
32
An Example (4)
ii. Find the capacitors reliability for a time period of three months. R(3 mos) = exp(- d 3) = exp(-2.28 10-2x3) = 0.934
33
Observations
Events (shocks) occur in time according to the Poisson distribution [the losses of electric power on slide 24, the surge voltages on slide 30]. There is a conditional probability that a given shock will be lethal, i.e., will fail the component. This conditional probability was given on slide 24 as 0.05, while on slide 31 it was calculated from reliability physics, i.e., from the normal distribution of the voltage (2.28x10-2). We calculated the rate of lethal shocks as the product of the rate of shocks times the conditional probability of failure. The occurrence of lethal shocks is, then, modeled as a Poisson process with this rate.
34
*t
35
()
95 = e +1.645
95 50 95 EF = = = 50 05 05
RPRA 3. Probability Distributions in RPRA 37
Y ln
is a normal variable with parameters (mean) and
(standard deviation).
RPRA 3. Probability Distributions in RPRA 38
ln 95 = + 1.645
But, Y ln
95 = e +1.645
as in slide 37
39
= -6.91
and
=1.40
50 = exp(-6.91) 10-3
m = exp( + 2/2) = 2.65x10-3
40