EE5712 Power System Reliability

:: Review of Probability Theory

Panida Jirutitijaroen
What is reliability?
Motivations
Issues with uncertainty

REVIEW
 What is Reliability?
• Ability of a system to perform its intended
function
• Within a specified period of time
• Under stated condition
→ Qualitative sense in terms of performance
function, time, and surrounding conditions
How to quantify ‘reliability’?
 Reliability
• Relate to the absence of failures, that due to
random phenomenon
• Define numerically as ‘average’ or mean value
• Can be treated as a parameter
• Can be traded off with other parameters such
as cost
Motivations for Quantitative Reliability
• To evaluate system performance
• System design purpose
• Trade-off reliability with cost
• Increasing complexity of systems
• Competitiveness
• Establish standard in operation procedure
 Issues with Uncertainty
• Reliability is quantified by probabilistic value.
• Need to study a bit about probability theory
– To model uncertainty
– To calculate probabilistic values such as probability
that a system will fail given a failure probability of
each component in the system
– To model characteristics of a component
• Failure rate/repair rate (coming up in the few weeks)
Some definitions
Probability rules
Application to power systems
Random variables
Some distributions

OUTLINE
Sample space or state space
Events
Union of events
Intersection of events
Disjoint events
Complement of an event
Independent events
Probability of an event
Conditional probability

DEFINITIONS
 Sample Space or State Space

The set of all possible outcomes of a

random phenomenon

• Example
– Outcome of coin flipping, state space = {H,T}
– Outcome of dice rolling, state space = {1,2,3,4,5,6}
– A status of a generator, state space = {Up, Down}
 A Status of Two Transmission Lines

Line 1 Line 1

Line 2 Line 2

(1 Up, 2 Up) (1 Up, 2 Down)

Line 1 Line 1

Line 2 Line 2

(1 Down, 2 Up) (1 Down, 2 Down)

State space = { (1U,2U), (1U,2D) , (1D,2U) , (1D,2D) }

 Events

A set of outcomes or a subset of a sample

space.

• Example
– Event: rolling a 1 on one dice, E = {1}
– Event : a generator fails, E = {Down}
An Event : One Transmission Line Fails

Line 1 Line 1

Line 2 Line 2

(1 Up, 2 Up) (1 Up, 2 Down)

Line 1 Line 1

Line 2 Line 2

(1 Down, 2 Up) (1 Down, 2 Down)

E = { (1U,2D) , (1D,2U) }
 Union of Events

Union of event E1 and E2 (E1 U E2 ) contains

outcomes from either E1 or E2 or both.

• Example
– E1 is an event that at least one line is up,
E1 = { (1U,2U), (1U,2D) , (1D,2U)}
– E2 is an event that at least one line is down,
E2 = { (1U,2D) , (1D,2U) , (1D,2D) }
– Then, union of event E1 and E2 is,
E = E1 U E2 = { (1U,2U), (1U,2D) , (1D,2U) , (1D,2D) }
 Intersection of Events

Intersection of events E1 and E2 ,(E1∩E2 )

contains outcomes from both E1 and E2.

• Example
– E1 is an event that at least one line is up,
E1 = { (1U,2U), (1U,2D) , (1D,2U)}
– E2 is an event that at least one line is down,
E2 = { (1U,2D) , (1D,2U) , (1D,2D) }
– Then, intersection of event E1 and E2 is,
E = E1 ∩ E2 = { (1U,2D) , (1D,2U) }
 Disjoint Events or Mutually Exclusive

Events that can not happen together.

• Example
– E1 is an event that two lines are up,
E1 = { (1U,2U)}
– E2 is an event that two lines are down,
E2 = { (1D,2D) }
– Then, E1 and E2 are disjoint events.
 Independent Events

Events that happen independently.

• Example
– Failure of a generator in one location and failure
of another generator in a distance location
– Failure of line 1 and failure of line 2
– (Latter on we will learn about “common mode
failure”– failure as a result of the same cause)
 Complement of an Event
The set of outcomes that are not included in an
event.
• Example
– E1 is an event that two lines are up,
E1 = { (1U,2U)}
– Ē1 is a complement of E1,
Ē1 = { (1U,2D) , (1D,2U) , (1D,2D) }
 Probability of an Event

Number of times that an event occurs

divided by total number of occurrences

• Also,
– 0 ≤ P(E) ≤ 1
– P(Impossible event) = 0
– P(Sure event) = 1
 Conditional Probability

The probability of E2 given E1, P(E2 | E1), is

the probability that event E2 occurs given
that E1 has already occurred.

• If E2 and E1 are independent, then

P(E2 | E1) = P(E2).
Addition rule
Multiplication rule
Conditional probability rule
Complementation rule

PROBABILITY RULES
 Addition Rule
A method of finding a probability of union of two
events.
P(E1 U E2) = P(E1) + P(E2) – P(E1 ∩ E2)

• If E1 and E2 are mutually exclusive, then

P(E1 U E2) = P(E1) + P(E2).

E1 E2 E1 E2
 Multiplication Rule
A method of finding probability of intersection of
two events.

P(E1 ∩ E2) = P(E1)×P(E2 | E1)

• If E1 and E2 are independent, then

P(E1 ∩ E2) = P(E1)×P(E2).
 Conditional Probability Rule
If an event E depends on a number of mutually
exclusive events Bj, then
P(E) = Σj [ P(E | Bj)×P(Bj) ]

B3
B1
E B5
B2 B4
 Complementation Rule
Probability of the set of outcomes that are not
included in an event.
P(Ē) = 1 – P(E)

• Example
Probability of success = 1 – Probability of failure
Example
State space
Loss of load probability

APPLICATION TO POWER SYSTEMS

 Example

• 3 generators
• Each 50 MW
1 • Identical probability of failure
2 Load
= 0.01
• Assume that each generator
3 fails independently.
• Find probability distribution
of generating capacity.
 State Space
1 1 1 1
Load Load Load Load
2 2 2 2
3 3 3 3
(1U,2U,3U) (1D,2U,3U) (1U,2D,3U) (1U,2U,3D)

1 1 1 1
Load Load Load Load
2 2 2 2
3 3 3 3
(1D,2D,3U) (1D,2U,3D) (1U,2D,3D) (1D,2D,3D)

State space = {(1U,2U,3U),(1D,2U,3U),(1U,2D,3U),

(1U,2U,3D),(1D,2D,3U),(1D,2U,3D),(1U,2D,3D), (1D,2D,3D)}
 Generating Probability Distribution
• Find probability associated with each generating
capacity level
• 4 capacity levels, 0 MW, 50 MW, 100 MW, and 150
MW
• Let
– E0 be an event that generating capacity is 0 MW
– E1 be an event that generating capacity is 50 MW
– E2 be an event that generating capacity is 100 MW
– E3 be an event that generating capacity is 150 MW
 Capacity 0 MW
• E0 = {(1D,2D,3D)}
P(E0) = P(1D ∩ 2D ∩ 3D)
• Using multiplication rule,
P(E0) = P(1D)×P(2D)×P(3D)
P(E0) = 0.01×0.01×0.01 = 0.000001
 Capacity 50 MW
• E1 = {(1D,2D,3U),(1D,2U,3D),(1U,2D,3D)}
P(E1) = P{(1D,2D,3U)U(1D,2U,3D)U(1U,2D,3D)}
• Using addition rule,
P(E1) = P{(1D,2D,3U)} + P{(1D,2U,3D)} + P{(1U,2D,3D)}
P(E1) = P(1D ∩ 2D ∩ 3U) + P(1D ∩ 2U ∩ 3D) + P(1U ∩ 2D ∩ 3D)

• Using multiplication rule,

P(E1) = P(1D)×P(2D)×P(3U) + P(1D)×P(2U)×P(3D) + P(1U)×P(2D)×P(3D)

• Using complementation,
P(1U) = 1 – P(1D) = 1 – 0.01 = 0.99

• Then, P(E1) = 0.000297

 Capacity 100 MW
• E2 = {(1D,2U,3U),(1U,2D,3U),(1U,2U,3D)}
P(E2) = P{(1D,2U,3U)U(1U,2D,3U)U(1U,2U,3D)}
• Using addition rule,
P(E2) = P{(1D,2U,3U)} + P{(1U,2D,3U)} + P{(1U,2U,3D)}
P(E2) = P(1D ∩ 2U ∩ 3U) + P(1U ∩ 2D ∩ 3U) + P(1U ∩ 2U ∩ 3D)

• Using multiplication rule,

P(E2) = P(1D)×P(2U)×P(3U) + P(1U)×P(2D)×P(3U) + P(1U)×P(2U)×P(3D)

• Using complementation,
P(1U) = 1 – P(1D) = 1 – 0.01 = 0.99

• Then, P(E2) = 0.029403

 Capacity 150 MW
• E3 = {(1U,2U,3U)}
P(E3) = P(1U ∩ 2U ∩ 3U)
• Using complementation,
P(1U) = 1 – P(1D) = 1 – 0.01 = 0.99
• Using multiplication rule,
P(E3) = P(1U)×P(2U)×P(3U)
P(E3) = 0.99×0.99×0.99 = 0.970299
Generating Probability Distribution

Capacity (MW) Probability

0 0.000001
50 0.000297
100 0.029403
150 0.970299
 Loss of Load Probability
• Assume that load has
distribution shown on Load (MW) Probability
right, find loss of load
probability. 50 0.20
• Let E be the event that 100 0.75
system suffers loss of
load, then 150 0.05
E = {Generation < Load}
• Let
– B1 be an event that load is 50 MW
– B2 be an event that load is 100 MW
– B3 be an event that load is 150 MW
• Then, using conditional probability rule,
P(E) = P(E|B1)×P(B1) + P(E|B2)×P(B2) + P(E|B3)×P(B3)
P(E) = P(E|B1)×0.20 + P(E|B2)×0.75 + P(E|B3)×0.05
 Load 50 MW
1 1 1 1
50 50 50 50
2 2 2 2
3 3 3 3
150 100 100 100
(1U,2U,3U) (1D,2U,3U) (1U,2D,3U) (1U,2U,3D)

1 1 1 1
50 50 50 50
2 2 2 2
3 3 3 3
50 50 50 0
(1D,2D,3U) (1D,2U,3D) (1U,2D,3D) (1D,2D,3D)

P(E|B1) = P{ (1D,2D,3D) } = 0.000001

 Load 100 MW
1 1 1 1
100 100 100 100
2 2 2 2
3 3 3 3
150 100 100 100
(1U,2U,3U) (1D,2U,3U) (1U,2D,3U) (1U,2U,3D)

1 1 1 1
100 100 100 100
2 2 2 2
3 3 3 3
50 50 50 0
(1D,2D,3U) (1D,2U,3D) (1U,2D,3D) (1D,2D,3D)

P(E|B2) = P{(1D,2D,3U),(1D,2U,3D),(1U,2D,3D), (1D,2D,3D) }

P(E|B2) = 0.000297 + 0.000001 = 0.000298
 Load 150 MW
1 1 1 1
150 150 150 150
2 2 2 2
3 3 3 3
150 100 100 100
(1U,2U,3U) (1D,2U,3U) (1U,2D,3U) (1U,2U,3D)

1 1 1 1
150 150 150 150
2 2 2 2
3 3 3 3
50 50 50 0
(1D,2D,3U) (1D,2U,3D) (1U,2D,3D) (1D,2D,3D)

P(E|B3) = 1 - P{ (1U,2U,3U) } = 0.029701

• Loss of load probability is,
P(E) = P(E|B1)×0.20 + P(E|B2)×0.75 + P(E|B3)×0.05
P(E) = 0.000001×0.20 + 0.000298×0.75 + 0.029701×0.05
P(E) = 0.00170875
 Example

Magic number?? • 3 generators

• Each 50 MW
1 • Identical probability of failure
2 Load
= 0.01
• Assume that each generator
3 fails independently.
• Find probability distribution
of generating capacity.
Probability density function
Cumulative distribution function
Expected value
Variance
Covariance
Moments
Survival function
Hazard function

RANDOM VARIABLES
 Random Variable (RV)

A function that assigns numerical values

to all possible outcomes of a state space

• Discrete RV assigns discrete value.

– Ex: Number of components down in a power
system
• Continuous RV assigns continuous value.
– Ex: Time to failure of a component
Discrete RV: Number of T- Lines Down

Outcomes Random Variable

Line 1
(1 Up, 2 Up) 0
Line 2
Line 1
(1 Up, 2 Down)
Line 2 1
Line 1
(1 Down, 2 Up)
Line 2
Line 1
(1 Down, 2 Down) 2
Line 2
 Continuous RV: Time to Failure

A generator start working at

time x = 0 G

Outcome: time to It can fail at any time, x ≥ 0

failure

Random Variable, X x≥0

 Probability Density Function
A function that gives probabilities associated
with all possible values of a random variable.
Pr x  X  x  x 
f x   lim
x 0 x
• Discrete RV: • Continuous RV:
Exponential distribution function
Pr  X  1  0.1
f x  0.5e0.5 x , x  0
0.5
0.3 0.3

Pr 2  X  3   0.5e 0.5 x dx
3
0.2
2
0.1 0.1

1 2 3 4 5 1 2 3 4 5
 Cumulative Distribution Function
A function that gives a probability that a random
variable X takes on value less than or equal to x.

F  x   Pr  X  x    f t dt
x



• Discrete RV: 1.0

• Continuous RV:
0.9
Exponential distribution function

0.6

0.3

0.1

1 2 3 4 5 1 2 3 4 5

F 2   Pr  X  2   0.1  0.2  0.3

 Expected Value
Average amount of the outcome of a
random experiment; sum of probability of
each possible outcome multiplied by its
value
• Summation for discrete RV
E  X    xi P X  xi 
i
• Integration for continuous RV

E X    xf xdx

 Variance
Measure squared distance of the possible
values from expected value
Var X   EX  EX 
2

• For discrete RV,

Var X    xi  EX  PX  xi 
2

i
• For continuous RV,

Var X     x  E  X 2
f x dx

 Covariance
Measure deviations of two random
variables from their expected value
Cov  X , Y   E  X  E X Y  E Y 

• Give measure of tendency of two random

variables to vary together
• If X and Y tend to vary together, Cov is positive
• If X and Y vary in opposition, Cov is negative
 Raw Moments
Expectation of the kth power of a random
variable
 k  X   E X k 

• For discrete RV,

 k  X    xik P X  xi 
i
• For continuous RV,

k  X    x k f x dx

 Central Moments
Expectation of the kth power of the
deviation of a random variable from its
mean value

 k  X   E  X  E  X 
 k

• For discrete RV,
k  X    xi  EX  PX  xi 
 k

i
• For continuous RV,

k  X    x  EX  f xdx
 k


 Raw VS Central Moments
Raw moment to Central moment Central moment to Raw moment
k
k  k
k 
 k     i1k i  k     1k i i 1k i
i 0  i  i 0  i 

0  1
0  1 1  0
2  2  12 2  12  2
3  3  32 1  13 3  213  312  3
4  4  43 1  62 12  14 4  314  613 2  413  4
 Moment-Generating Function
Generate raw moments of the probability
distribution
 
M t   E etX
• Easy computation of raw moments
M t 
 
k
 M 0 
d
E Xk (k )

dt k t 0
• Related transforms: characteristic function,
probability-generating function, cumulant-
generating function.
 X: Time to Failure

A generator start working at

time x = 0 G

Outcome: time to It can fail at any time, x ≥ 0

failure

Random Variable, X x≥0

 Failure Distribution

Probability that a component fails at time x.

Pr x  X  x  x 
f x   lim
x 0 x
Exponential distribution function

f x  0.5e0.5 x , x  0
0.5

Pr 2  X  3   0.5e 0.5 x dx
3

1 2 3 4 5
 Cumulative Failure Distribution
A probability that a component fails before
or at time x.
F  x   Pr  X  x    f t dt
x



Exponential distribution function

1 2 3 4 5
 Survival Function
A function that gives probability of a
component survival beyond time x.

R x   Pr  X  x   1  Pr  X  x   1  F  x 
• Time to failure of a component is a random variable,
X.
• Commonly used in reliability theory


R4  Pr  X  4   0.5e 0.5 x dx
0.5
4

1 2 3 4 5
 Hazard Function
A function that gives a rate at time x, at which
a component fails ( i.e. failure rate), given that
it has survived for time x.

• Denoted by Ф(x),
Probability of a component fails between time x and x+Δx given
that it has survived for time x

Pr  x  X  x  x | X  x 
 x   lim
x 0 x
• From,
Pr  x  X  x  x | X  x 
 x   lim
x 0 x
• We have

Prx  X  x  x 
 x   lim
1

x 0 x Pr X  x 
f x 
 x  
R x 
 An Important Relationship
• From hazard function as Δx → 0,
  x x  Pr x  X  x  x | X  x 
• This gives probability of failure of a
component in interval (x, x+Δx).
 x 

Pr  x  X  x  x | X  x 

x x+ Δx
 Important Note
• Although , for simplicity, survival function and
hazard rate function have been described with
respect to a component failure, they apply to
any random variable.
• For example if the random variable is time to
repair, then Ф(x) represents the repair rate
Binomial distribution
Poisson distribution
Exponential distribution

SOME DISTRIBUTIONS
 Binomial Distribution Function
• Discrete distribution
• Number of success(or failure) in n
independent trials
• Each trial has two outcomes, success or failure
• Same probability of success, p
• Probability of getting x success in n trials is,
n x
Pr  X  x   f x; n, p     p 1  p 
n x
 
 x
E  X   np Var  X   npq
 Binomial Distribution Example
• 3 identical components, each has probability
of success, p.
1U 2U 3U
X = number of component in up state

 3
1D 2U 3U 1U 2D 3U 1U 2U 3D Pr  X  2  f 2;3, p     p 2 1  p 
 2

1D 2D 3U 1U 2D 3D 1D 2U 3D

1D 2D 3D
 What if # of trials continue to ∞ ?
• Let p = Λ/n, Λ = number of events in n trials,
• As n →∞, n x
        
x
 n
Pr X  x   f x; n, p   f  x; n,      1  
 n   x  n   n

   x e  
Pr  X  x   f  x; n,  
 n x!
• This leads to the next distribution.
 Poisson Distribution
• Discrete distribution
• Number of occurrences(failure or success) in a
fixed time
• Given that an average/expected number of
events is Λ,
• Probability of x occurrences in this interval is,
x 
e
Pr X  x   f x;   
x!
E  X    Var  X   
 Poisson Distribution Example
• A transmission line fails on average 2 times
per year, propose Poisson distributions model
for 5-year and 10-year time interval.
• What is Λ?
• What is the probability of having 2 failures in 5
years?
• What is the probability of having 2 failures in
10 years?
 X: Time to Failure

A generator start working at

time x = 0 G

Outcome: time to It can fail at any time, x ≥ 0

failure

Random Variable, X x≥0

 Exponential Distribution Function
• Non-negative continuous random variable
• Commonly used to represent up time of a
component
• Probability that a component will fail at time x
is,
f x  ex , x  0
• Mean time to failure (MTTF) is E X  
1

 Exponential Distribution Example

x1 x2 x3 x4 x5

x1  x2    xn
• Mean time to failure is 
n
• n = number of failures
1 n
 
MTTF Total observatio n time
• λ is called “Failure rate”.
 Constant Hazard Function
• If we assume that the up time of a component
is exponentially distributed, the failure rate is
constant.
f x  e  x
,x0 Rx  ex , x  0
f x 
 x    , x  0
R x 

• This is also called “memoryless property”

 Memoryless Property
• Observe a component up to time t. and a
component has not failed yet.
• Interested to find residual lifetime, Y = X-t.
FY  x   Pr Y  x | X  t   Pr  X  t  x | X  t 
• Distribution of a residual lifetime is independent
of the time that the component has been in
operations.
• This component does not age, only random
failure
Characteristics of Exponential Dist.

f x   e  x
Rx  ex  x   
Relationship with Poisson Distribution
• Constant failure rate, λ
• Consider time (0,t)
• Expected number of failure = λt
• Poisson distribution: probability of y failures in
time t.
Pr Y  y   f  y; t  
t  e
y  t

y!
• If no failure occur in time t,
PrY  0  et
• Recall survivor function of exponential
distribution,
Rt   et , t  0
• For Poisson failure, the distribution of time
between failures, y1, y2, y3,… are
exponentially distributed

y1 y2 y3 y4 y5
 Other Distributions
• Normal distribution
• Weibull distribution
• Gamma distribution
• Log-normal distribution

• Reference:
http://www.ece.tamu.edu/People/bios/singh/
sysreliability/ch2.pdf
Characteristics of Normal Dist.
Characteristics of Log-Normal Dist.
Characteristics of Weibull Dist.
Characteristics of Gamma Dist.
 Summary
• Review of
– Probability theory
– Some distributions, binomial, Poisson,
exponential.
– Probability rules (addition, conditional probability)
– Basic application on power systems
 Application
• To construct system probability distribution,
given that failure probability of each
component is known.
• For repairable systems, components are
characterized by failure/repair rates, the
system is described by the stochastic process,
coming up in the next few weeks.
• Exponential distribution has constant hazard
rate, we’ll be seeing a lot of it in this class
 Reading Materials
• Review today’s lecture
– Conditional probability rule
– Survival function
– Hazard function
– Relationship among distribution function, survival function
and hazard function
– Exponential distribution
• Chapter 2: The Preliminaries “System Reliability
Modeling and Evaluation”,
http://www.ece.tamu.edu/People/bios/singh/sysreliabi
lity/ch2.pdf
• Do homework!
 About Next Lecture
• Reliability Theory
– Physical meaning of hazard rate function
– Hazard rate function approximation
• Reliability measure
• Simple reliability evaluation techniques