CALCULATION METHODS
CHANAN SINGH
DEPARTMENT OF ELECTRICAL & COMPUTER
ENGINEERING
TEXAS A&M UNIVERSITY
COLLEGE STATION, TX
May, 2006
INTRODUCTION TO QUANTITATIVE
RELIABILITY
RELIABILITY RELATES TO THE ABILITY OF A
SYSTEM TO PERFORM ITS INTENDED FUNCTION
IN A QUALITATIVE SENSE, PLANNERS AND
DESIGNERS ARE ALWAYS CONCERNED WITH
RELIABILITY
WHEN QUANTITATIVELY DEFINED, RELIABILITY
BECOMES A PARAMETER THAT CAN BE TRADED
OFF WITH OTHER PARAMETERS LIKE COST
NECESSITY OF QUANTITATIVE RELIABILITY:
EVER INCREASING COMPLEXITY OF SYSTEMS
EVALUATION OF ALTERNATE DESIGNS
COST COMPETITVENESS AND COSTBENEFIT
TRADE OFF
MEASURES OF RELIABILITY
BASIC INDICES
PROBAILITY OF FAILURE LONG RUN
FRACTION OF TIME SYSTEM IS FAILED
FREQUENCY OF FAILURE EXPECTED OR
AVERAGE NUMBER OF TIMES PER UNIT TIME
MEAN DURATION OF FAILURE MEAN
DURATION OF A SINGLE FAILURE
COST VS RELIABILITY
Total Cost
Investment Cost
Failure Cost
To Customer
ROPT
4
Optimum
SEQUENCE OF PRESENTATION
DESCRIBE AN EXAMPLE THAT WILL BE USED TO
ILLUSTRATE CONCEPTS
INTRODUCE ANALYTICAL METHODS OF
RELIABILITY EVALUATION
INTRODUCE MONTE CARLO SIMULATION
METHODS OF RELIABILITY EVALUATION
EXAMPLE SYSTEM
Generators
Transmission
Load
Generators:
Each generator either has full capacity of 50 MW or 0 MW when failed.
Failure rate of each generator is 0.1/day and meanrepairtime is 12 hours.
Transmission Lines:
The failure rate of each transmission line is assumed to be 10 f/y during the
normal weather and 100 f/y during the adverse weather. The mean down time
is 8 hours. Capacity of each line is 100 MW.
Weather:
The weather fluctuates between normal and adverse state with mean duration
of normal state 200 hours and that of adverse state 6 hours.
Breakers:
Breakers are assumed perfectly reliable except that the pair B1&B2 or
B3&B4 may not open on fault on the transmission line with probability 0.1.
Load:
Load fluctuates between two states, 140 MW and 50 MW with mean duration
in each state of 8hr and 16hr respectively.
FOR THE DESCRIBED SYSTEM,
HOW CAN YOU CALCULATE
RELIABILITY INDICES ?
1. Loss of load probability
2. Frequency of loss of load
3. Mean duration of loss of load
THE
FOLLOWING
BASIC
Markov Processes
Using
8
UP
2
DN
10
where
MUT= mean up time
Also
Transition rate from down to up state = repair rate
= n 21 / T2
= 1/( T 2 / n21 )
=1 / MDT
where
MDT = mean down time of the component
Concept of Frequency
Frequency of encountering state j from state i is the expected (mean)
number of transitions from state i to state j per unit time.
Fr(ij) = steady state or average frequency of transition from state i to
j
= nij / T
= ( Ti / T) (nij / Ti )
= pi ij
where
pi = long term fraction of time spent in state i
= steady state probability of system state i
11
10MW
0 MW
UP
DN
2 State Model for a 10MW Generator
1u
1 2u
30 MW
2
20 MW
< 10MW
2D
3u
1u
7
10 MW
1 u
2u
3 D
2D
20 MW
1 D
2u
3 D
6
10 MW
3 D
0 MW
1u
3
1 D
2D
3 U
10 MW
20 MW
3u
1 D
2u
3 U
1 D
2D
3 D
p
n
i 1
14
Other indices:
Mean cycle time of an event (MCT) = 1 / frequency of the event
Mean duration of the event = MCT x Prob. of the event
Thus
Mean cycle time between failures = 1/ freq of failure
Mean down time (MDT) = prob of failure / freq of failure
Mean up time = MCT  MDT
= prob of system up / freq of failure
15
Disjoint Subsets
16
2. Find the expected transition rate into the boundary or out of the
boundary.
Example:
For the case of three 2state units,
Fr(capacity 10) = p2 ( 2 + 3 ) + p3 ( 1 + 3 ) + p4 (1 + 2 )
=p5( 1 + 2 ) +p7 ( 2 +3) + p6 (1 + 3)
This frequency is typically called cumulative frequency.
17
18
Down
State
0 MW
Figure 2: Each Generator has two possible states
Up State
50 MW
1
730 / year
12
8760
19
(1)
State 2
State 3
State 1
12
34
23
Two
UP,
one
DN
One
UP, two DN 43
32
All UP 150MW 21
100MW
50MW
State 4
All DN  0MW
12G
P1 P1 P1
3
P1
21G
P2 P3 P4
P2 P3 P4
23G
2 P2 2 P3 2 P4
2
P2 P3 P4
32G
(2)
2 P 2 P6 2 P7
5
2
P5 P6 P7
34G
P5 P6 P7
P5 P6 P7
43G
P8 P8 P8
3
P8
RG
3
( 2)
2
0
0
2
(2 )
3
0
0
(3)
If we substitute values for and obtained in (1) into the matrix (3), transition rate matrix
for the generator system is:
RG
109.5
730
109.5
803
1460
0
73
1496.5
2190
0
36.5
2190
1. 2. Transmission Lines
20
(4)
Up State
100 MW
Down
State
0 MW
(5)
If all the breakers are perfectly reliable, for the twotransmissionline system, there will be 4
states.
State 1
1U 2U
200 MW
State 2
1D 2U
100 MW
State 4
1D 2D
0 MW
State 3
1U 2D
100 MW
If breakers
open on command:
State may
5 not
0.1
1D 2U
0 MW
State 1
1U 2U
200 MW
State 6
1U 2D
0 MW
0.1
0.9
21
0.9
State 2
1D 2U
100 MW
State 4
1D 2D
0 MW
State 3
1U 2D
100 MW
Merging of states:
State 1
12
one UP one DN 21
0MW
State 2
two UP
200MW
32
23
State 3
34
one UP one DN 43
100MW
State 4
two DN
0MW
22
12
P5 P6
P5 P6
21
P1 (0.1 0.1 )
0.2
P1
23
P1 (0.9 0.9 )
1.8
P1
32
P2 P3
P2 P3
34
P2 P3
P2 P3
43
P4 ( )
2
P4
1.2.1
(6)
Weather
State 4
1 UP, 1
DOWN
0 MW
S
0.2
N
State 8
1 UP, 1
DOWN
0 MW
State 1
2 UP
200
MW
S
1.8
State 5
2 UP
200
MW
State 2
1 UP, 1
DOWN
100 MW
S
State 6
1 UP, 1
DOWN
100 MW
State 3
2
DOWN
0 MW
S
N
State 7
2
DOWN
0 MW
0.2
1.8
Figure 9: Equivalent
Eight State Transition Diagram Transmission System
23
1
43.8 / year
200
8760
1
1460 / year
6
8760
RT
0
0.2
 (2 N) 1.8
 ( N)
0
0
2  (2 N)
0
0
0
 ( N)
S
0
0
0
0
S
0
0
0
0
S
0
0
0
0
S
0
0
0
N
 (2'S) 1.8'
0
0.2'
 ( 'S) '
0
0
2  (2 S)
0
0
0
 ( S)
N
0
0
0
0
N
0
0
0
0
N
0
(7)
24
1.3
Load
State 1
140 MW
21
State 2
50 MW
1095 / year
8
8760
21
1
547.5 / year
16
8760
1095 1095
RL
547.5 547.5
(8)
25
2. 1.
Generation System
P1G
P
T
R G 2G
P 3G
P4G
0
0
i 1
iG
(9)
Using the RG we obtained in equation (4), solving equations (9), we get the steady state
probability of each state.
If generators are independent probabilities can be calculated by product rule also.
Probabilities calculated in either way are the same.
0.95238
Pd
0.047619
Pu
P1G= Pu * Pu * Pu
P1G= 3*Pu * Pu * Pd
P1G= 3*Pu * Pd * Pd
P1G= Pd * Pd * Pd
2. 2.
=0.8638377
=0.1295725
=0.00647876
=0.000107979
(10)
Transmission System
P1 0
P
2 0
P 3 0
P 0
4
R TT 0
P5
P 0
6
P7 0
0
P8
P
i 1
(11)
Using the RT we obtained in equation (7), solving equations (11), we get the steady state
probability of each state:
26
RtT =
1.0e+003 *
0.0638
0.0180
0
0.0020
0.0438
0
0
0
1.0950
1.1488
0.0100
0
0
0.0438
0
0
0
2.1900
2.2338
0
0
0
0.0438
0
1.0950
0
0
1.1388
0
0
0
0.0438
1.4600
0
0
0
1.6600
0.1800
0
0.0200
0
1.4600
0
0
1.0950
2.6550
0.1000
0
0
0
1.4600
0
0
2.1900
3.6500
0
P1=0.9507726
P2=0.01787034
P3=0.0001196528
P4=0.0019820304
P5=0.02678843
P6=0.002162378
P7=0.000060686
P8=0.00024383
0
0
0
1.4600
1.0950
0
0
2.5550
(12)
We can also reduce the eightstate transmission transition diagram given in Fig 11 to a threestate diagram with respect to the capacities of the states:
State 1
200MW
21T
12T
State 2
100MW
32T
23T
31T
State 3
0MW
13T
12T
22.439339
0.97756103
P1' P5'
27
21T
P2' P6'
10
P2' P6'
23T
19.714808
0.020032718
P2' P6'
32T
P3' 2 P7' 2
'
1095
P3 P7' P4' P8'
13T
31T
2. 3.
2.493259
0.97756103
P1' P5'
P4' P8'
'
2090
P3 P7' P4' P8'
(13)
Load
P1L
0
0
P2 L
R TL
P
i 1
iL
(14)
Using the RL we obtained in equation (8), solving equations (14), we get the steady state
probability in each state:
P1L 0.3 3 3 3
P 2L 0.6 6 6 7
28
29
2. 4.
Steady state probability, frequency and mean time of loss of load could be found using the
following table:
P1G = 0.8638377
P2G = 0.1295725
P3G = 0.00647876
P4G = 0.000107979
System
State
P1T = 0.97756103
P2T = 0.020032718
P3T = 0.0024061992
P1L = 0.3333333
P2L = 0.6666667
Generation,
transmission, load
system state
111
Probability of
system state
Loss of
load
0.281485
2,3,4
( 12 T , 13T , 12 G )
No
2
3
4
5
6
7
8
9
10
11
12
13
14
121
131
211
221
231
311
321
331
411
421
431
112
122
0.562969
0.0115367
15( 13T )
2,15
( 21L , 23T )
15
16
132
212
0.0844453
17
222
0.0017305
4,18
( 21L , 13T )
5,18
( 21L , 23T )
18
19
232
312
0.00422226
20
322
0.000086525
21
22
23
24
332
412
422
432
7,21,22,
( 21L , 13T , 34G )
8,21,23
( 21L , 23T , 34 G )
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
Yes
No
No
Yes
No
No
Yes
Yes
Yes
Yes
We can calculate the probability of states having no load loss. Those probabilities are obtained
for the generators, transmission lines and loads as independent.
From Table 1, we can get the steady state probability of the loss of load as follows.
30
P=1(0.281485+0.562969+0.0115367+0.0844453+0.0017305+0.00422226+0.000086525)
P=0.053524715
(15)
i X j X
ij
(16)
P16 ( 21L 13T ) P17 (21L 23T ) P19 ( 21L 13T 34G ) P20 ( 21L 23T 34G )
F = 95.742635 /year
Values needed for F that are calculated previously:
23T 19.714808
12T 22.439339
12G 3 109.5
23G 2 73
34G 36.5
13T 2.493259
P
0.053524715
4.89726hours
F
95.742635 / year
31
(17)
32
where
i , ri
i j (ri rj )
1 i ri j rj
rcsk
ri rj
ri rj
where
i , j
ri , rj
Nr Nr
N
( i j i j i j)
N S N ri
N rj
33
r
Srj
N
r
Sri
(i i j
j j i
)
N S
N S ri
N S rj
Nrj
S
Nri
(i j
j i
)
N S
N ri
N rj
Srj
S
Sri
(i j
j i
)
NS
S ri
S rj
where
Component due to both failures occurring during normal
weather
Initial failure in normal weather, second failure in adverse
weather
Initial failure in adverse weather, second failure in normal
weather
Both failures during adverse weather.
a
34
N S
12.621
NS
rcs1 8hr
load
8760
547.5 / year
16
rg 12hr .00137 yr
cs 2
rg rload
rg rload
35
4.8hr
Cut set 3:One line failure (breaker not stuck) and load
changes from 50 to 140
l av 0.9
rl 8hr
cs 3
rcs 3
l load ( rl rload )
15.03 / yr
(1 l rl load rload )
rl rload
4hr
rl rload
10 .9 9 / yr
100 .9 90 / yr
r 8hr .000913 yr
N 200hr .022831 yr
S 6hr .000685 yr
T cs1 cs 2 cs 3 cs 4 106.25 / yr
rT
T
1
rT
Frequency of failure =
Probability of failure =
36
T
T 100.45 / yr
T T
T
T T
= .0546
37
Introduction
The Monte Carlo method mimics the failure and repair history of
components and the system by using the probability distributions
of component states.
Statistics are collected and indices estimated by statistical
inference.
Two main approaches: random sampling , sequential simulation.
Random sampling
Sampling a component state:
38
State sampled
1
2
3
4
5
P(X=
1.0
1.0
0.9
0.9
0.8
0.8
0.7
0.7
x ) 0.6
0.6
0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0
1
Random Observation
1
39
< x)
F(x)=P(X
Now you can place the rn on the vertical axis and read the value of the
state sampled on the horizontal axis. It can be seen that this is equivalent
to proportional sampling.
P(3,3) = n/N
where
n = number of times state sampled
N = total number of samples.
From the table

P(3,3) = 4/14
40
=2/7
=.286
The actual prob of (3,3) is
.4x.4 = .16
If this sampling and estimation are continued the estimated value will
come close to .16.
You can appreciate on of the problems of Monte Carlo that the indices
obtained are estimates. So one must have some criterion to decide
whether the indices have converged or not.
.308
.116
.864
.917
.126
.157
.837
.029
.865
.603
.203
.823
.969
.509
41
.998
.209
.923
.883
.851
.135
.651
.034
.316
.525
.965
.427
.839
.434
.407
.355
.885
.931
.896
.680
.057
.922
.113
.852
.931
.538
.504
.925
42
Sequential Simulation
Initial State
0
1
Final State
0
0.3
0.4
1
0.7
0.6
event
stay in state 0
transit to 1
event
transit to 0
stay in 1
Step
1
2
3
4
5
6
7
8
9
10
RN
.947
.601
.655
.671
.791
.333
.345
.531
.478
.087
State
0
1
1
1
1
1
0
1
1
1
F(x) = P(X
< x)
RN
Value of RV
RV X
0 Z 0
f ( z ) 1 0 Z 1
0 Z 0
Similarly
0 Z 0
F ( z) z 0 Z 1
1 Z 1
Let F(x) be the distribution from which the random observations are to
be generated. Let
z = F(x)
Solving the equation for x gives a random observation of X. That the
observations so generated do have F(x) as the probability distribution
can be shown as follows.
Let be the inverse of F; then
x = (z)
Now x is the random observation generated. We determine its
probability distribution as follows
45
P( X x) 1 ex
where 1/ is the mean of the random variable X. Setting this function
equal to a random decimal number between 0 and 1,
z 1 e x
Since the complement of such a random number is also a random
number, the above equation can as well be written as
z ex
Taking the natural logarithm of both sides and simplifying, we get
x
ln( z)
46
0
5
9
49
51
101
112
161
Random Number
for Component
1
2
.946
.655
.670
.790
.332
.437
.601
.345
.531

47
Time to change
1
5
0/4
0/40
0/2
0/110
60
49
0/8
101
96
92
52
50
0/11
0/127
78
Compon
ent state
1
2
U
D
U
D
U
U
U
D
U
U
U
U
U
D
U
U
169
239
251
324
.087

.311
.693
.333
0/244
174
162
89
70
0/12
0/73
0/
U
U
U
U
indicates time causing change and is added to obtain the total time.
48
U
D
U
D
REFERENCES
1. C. Singh, Course Notes: Electrical Power System
Reliability, http://www.ece.tamu.edu/People/bios/singh/
2. C. Singh, R. Billinton, System Reliability Modelling and
Evaluation, Hutchinson Educational, 1978, London
3. R. Billinton, R. Allen, Reliability Evaluation of Power
Systems, Plenum Press, 1984
49