Transcript lecture2
Stochastic Optimization
Uncertainty and Reliability
Analysis
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Objectives
To learn the major elements of planning process
Uncertainty
Reliability
To discuss various methods for analyzing the uncertainty
To compute reliability using various methods
2
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Uncertainty
Most water resources decision problems face the risk of uncertainty
Due to the randomness of the variables that influence the performance of the systems
Uncertainty in water resources arises mainly due to the stochastic nature of
Hydrological processes such as rainfall, evaporation, temperature
Other variables like future population growth, per capita water usages, irrigation patterns
etc.
Depending upon the severity of uncertainty of the quantities involved, they are
replaced by either their expected value or some critical value (eg., worst case value)
and proceed with a deterministic approach.
The expected value or median value is used when the uncertainty is reasonably small
and the performance of the system is not much affected.
3
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Sensitivity Analysis
Simplest method for assessing the effect of uncertainty
System performance is analyzed by varying the magnitude of the more uncertain
parameters
This will help to identify the most sensitive parameter or variable in a system
For example let the cost required for a structure of capacity k is C(k) = akb
The parameter b represents the elasticity of costs and
b
k dC
C dk
or
dC
dk
b
C
k
For a value of b = 0.6, a change in capacity by 10% will result in a cost change of
only 6%.
4
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
First-order Analysis or Delta Method
Used to estimate the uncertainty in a deterministic model formulation in which the
parameters involved are uncertain
For example, in the estimation of weir discharge Q = CLH1.5
If the parameters C and H are not certain, then Q is also uncertain
Through first order analysis, one would be able to estimate the mean and variance of
a r.v. which is related to other variables which may also be random
Combined effect of uncertainty can thus be assessed
Consider a r.v. X which is a function of n other r.v. s
Mathematically,
X = f (Y)
where Y = (Y1, Y2, …, Yn) is a vector of n variables
5
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
First-order Analysis or Delta Method…
Through Taylor’s expansion about the means of n r.v., the first order approximation
of the r.v X can be expressed as (ignoring the second and higher order terms)
f
X f y
Yi y i
Y
i 1
i Y y
n
(1)
where y y1 , y2 ,..., yn and
f
is the sensitivity coefficient which is the rate of change of function value
Yi y
f (y) at Y y
The mean and variance of the r.v. X are approximated as
X EX f y
n f
VarX Var f y Var Yi yi
i 1 Yi y
6
Water Resources Planning and Management: M6L2
(2)
(3)
D Nagesh Kumar, IISc
First-order Analysis or Delta Method…
Var f y 0 since is calculated using the mean values of and is constant.
Therefore,
f
where ai
Yi y
n
VarX Var ai Yi yi
i 1
(4)
Eqn. (4) can also be expressed as
n 1
VarX a 2 ai a j CovYi ,Y j
n
i 1
2
i
2
i
n
(5)
i 1 j i 1
n
where VarX ai2 i2 is the variance of the r.v Yi.
i 1
For uncorrelated Yi s CovYi ,Y j 0
7
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
First-order Analysis or Delta Method…
Therefore,
n
VarX ai2 i2
(6)
i 1
The coefficient of variation can be expressed as
C
2
v X
yi
a
i 1
Y
n
2
i
Cv Y2
i
(7)
Equations (6) or (7) express the relative contribution of uncertainty of each
component to the total uncertainty
These can be used to minimize the effects of uncertainties in model output X
8
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Example:
Analyse the uncertainty in the channel discharge Q
1
AR 2 / 3 S 1 / 2 where the parameter
n
R is certain and the parameters n and S are uncertain.
Solution:
Since R and A are certain Q C n 1 S 1 / 2 where the constant C AR 2 /.3
First order approximation of Q is
Q
Q
Q Q n n S S
n n ,S
S n ,S
S 1/ 2
1
S S
Q C 2 n n C
1 / 2
n
2n S
n ,S
where Q C 1 S 1 / 2 ;
n
9
Q
Q
and
are sensitivity coefficients.
S
n
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Example…
The uncertainty of Q assuming n and S as independent variables, can be expressed by
the variance operator as
Q
Q
Q2 n2 S2
n n ,S
S n ,S
2
2
Expressing uncertainty in the form on coefficient of variance gives (as per eqn. 7)
C
2
v Q
Q
n
2
n
Q
Cv 2n
S
Q
2
S
Q
Cv 2S
Cv n 0.25 Cv S
2
10
Water Resources Planning and Management: M6L2
2
D Nagesh Kumar, IISc
Reliability
Hydrosystems are expected to exhibit resistance against any external stresses
Resistance or strength of any system is its ability to function without failure against
the external stresses
Reliability is the probability of the system functioning satisfactorily
Consider the state of a system denoted by a random variable Xt at time t for t =
1,2,…,n.
If the possible outcomes of Xt can be divided into two sets:
(i) satisfactory outputs or successes, S and
(ii) unsatisfactory outputs or failures, F.
Then, reliability can be expressed as
PX t S
For example, the reliability of a water supply system depends on the conditions when
supply is greater than demand i.e., successes.
Reliability is the ratio of non-failures to the total periods.
11
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Reliability Computation using load-resistance
method
Reliability: Probability that the capacity of the system (or resistance) exceeds or
equals the loading
i.e., .
PL R
Risk is just the opposite of reliability
Risk is the probability of the loading exceeding the resistance
i.e., .
' PL R 1
In load-resistance method, reliability can be computed by
By direct integration
Using safety margin and safety factor
12
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Reliability Computation using load-resistance method…
Computation by direct integration:
Let the joint PDF of load and resistance be
f L ,R l , r .
Then reliability is
r
f L ,R l , r dl dr
0 0
If loading and resistance are independent, then,
0
r
f R r f L l dl dr
0
f R r FL r dr
0
where FL r is the CDF of the loading at L = r.
13
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Reliability Computation using load-resistance method…
Computation using safety margin and safety factor
Safety margin (SM): Difference between the system’s resistance and the anticipated
loading
SM = R – L
i.e.,
Reliability can be expressed as, PR L 0 PSM 0
This requires the PDF of SM
Assuming that the loading and resistance are independent normal r.v. s, the mean
and variance of SM are
SM R L
Then reliability is
and
SM SM SM
P
SM
SM
P Z SM
SM
14
2
SM
R2 L2
Water Resources Planning and Management: M6L2
SM
SM
D Nagesh Kumar, IISc
Example:
The average surface runoff to a sewer is 3 m3/s with a standard deviation of 1.2 m3/s.
The mean capacity of the sewer is estimated to be 4.5 m3/s with a standard deviation of
0.8 m3/s. Compute the reliability using safety margin approach.
Solution:
μL = 3;
μR = 4.5;
σL = 1.2;
σR = 0.8
μSM = μR - μL = 1.5
σSM2 = σR2 + σL2 = 2.08
Therefore,
Z
1.5
0.851
2.08
And risk is,
α’ = 1 – 0.851 = 0.149
15
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Reliability Computation using Time-to-Failure
Analysis
Instead of considering the resistance and loading, only one r.v
i.e., time is considered
Time-to-failure (T) of a system is the random variable with PDF fT(t) and is called
the failure density function.
Then, the reliability within a time interval [0, t] can be expressed as
t fT t dt
t
Risk can be expressed as
t
' t fT t dt
0
16
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Example
The time to failure of a pump in a water distribution system is assumed to follow an
exponential distribution with the parameter λ = 0.0137/day (5 failures per year).
Compute the reliability for 10 days operation.
Solution
The failure density function is an exponential distribution function
fT(t) = λ e- λt = 0.0137 e- 0.0137 t , t ≥ 0
The reliability is
t e t dt
t
e t e 0.0137t ,
For 10 days,
17
t0
α = e- 0.0137 *10 = 0.872
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc
Thank You
Water Resources Planning and Management: M6L2
D Nagesh Kumar, IISc