RELIABILITY OF POWER

ELECTRIC SYSTEMS
Chapter 3: Basic Systems Assessment

Power Electrical Engineering

Cotopaxi Technical University
Semester: October 2017 - February 2018
Lecturer: Edwin M. Lema G, MSc.

October 2017-February 2018

Latacunga-Ecuador

1
 THE PROCESS: OPERATION-FAIL-RECOVERY-OPERATION
(renewable)

• All element is subjected to this cycle.

• If observed for a sufficiently long time, it is
possible to estimate expected values. Assume
there are N cycles:

mi
operation

failure ri

2
 THE PROCESS: OPERATION-FAIL-RECOVERY-OPERATION
(renewable)

• The equivalent cycle is :

m
operation

failure r

3
 MTTF MTTR

N N
 mi  ri
1 1 T mr
m 1  r   1
N  N 
The fraction of the time the element is operating is called
FORCED AVAILABILITY.

FORCED UNAVAILABILITY is the fraction of the time that the element

was in the failure state

m m r r
A  U  1 A  
T mr T mr
4
 Availability is understood as the long
range probability of finding the element
operating.

It can be shown that :

 P(O,0)  P( F ,0)  (   )t

P(O, t )   e
 
 P(O,0)  P( F ,0) (    )t
P( F , t )   e
 

5
Example
• A generator is planned to enter in operation in the
interval from 07h00 to 21h00, producing the peak
load at 18h00. It has been established as a policy
that, once a forced failure occurs, it is immediately
repaired. If the reparation finishes before 21:00 it
is commissioning again. For this generator: =20
f/year; r = 2 days. Calculate: the average times that
would be in fault in the peak hour, 2200
commissionings are planned in the next 10 years,
assuming that it does not fail in the start.

6
 Exmple (cont…)
• The probability of failure must be calculated at a given time (11
hours).
• Is required .
r = 2 = 2 * (1/365)
 = (1/r) = 182,5 (1/año)

20 20 *1  182,5 * 0 ( 20182,5)11
P( F , t  11)   e
20  182,5 20  182,5

P( F , t  11)  0,0222

7
Example (cont…)
• Now, the common definition of probability :
• P(F) = Number of failures / number of attempts
• P(F) = Nf/Ni
• Nf = Ni * P(F) = 2200* 0,0222 = 49

8
METHODS OF ASSESSMENT OF RELIABILITY:

•Reduction model of network:

• State Model
•F&D

9
 MODEL REDUCTION OF NETWORKS
• Many systems are physical networks or can be
represented by networks.
• In these networks the elements are in series,
parallel, enmeshed or some combination.
• THE SYSTEM AND ITS RELIABILITY MODEL
NECESSARILY DO NOT HAVE THE SAME
TOPOLOGY.
• SERIES SYSTEM: All of its components needs to
work for the success of the system, or only one
needs to fail to the failure of the system.
• PARALLEL SYSTEM: One component needs to
work for the success of the system, or all must fail
to the failure of the system.
• SERIES SYSTEM: is non redundant system.
• PARALLEL SYSTEM: redundant system.

10
Serial system:
• The elements are independent
• Ri Is the probability of successful operation.
• The probability of success of the system is :

N
R s   Ri
i1
N
Q s  1  R s  1   Ri
i1
• Reliability decreases as the number of serial components
increases.

11
 1,0
Ri = 0,999
0,9
0,8
reliability

0,7
0,6
Ri = 0,99
0,5
0,4 Ri = 0,98
0,3
Ri = 0,95
0,2
Ri = 0,9
0,1
0,0
0 10 20 30 40 50 60 70 80 90

Number of components

12
 Parallel system :

N N
Q s   Qi R s  1  Q s  1   Qi
i1 i1
MODEL OF STATES
•Consider each element with two states; UP y DOWN.
• Output rates are  y .


UP DOWN


• The evaluation will depend on the serial or parallel components.

13
The steps are :
• Enumeration of system states:
• It is determined from the combinations of
individual states.
• Determination of transition rates:
• It is determined by observing how fast you
travel from one state to another.
• Determination of state probabilities:
• The rule of the product of individual
probabilities applies.
• Determination of reliability:
• It depends on whether the components are
in series, parallel or some combination of
them; Additionally you must define Success
Status and System Fault Status.

14
Example 3.1: System with two independent elements.

1
1U 1D
2U 1 2U 2
1

2 2 2
2

1
1U 1D
2D 3 2D 4

1

• N = Number elements
# State  2 N

15
Example 3.1 (Cont…)

State Probability

1 A1* A2 =

2 U1* A2 =

3 A1* U2 =

4 U1* U2 =

16
 FREQUENCY & DURATION METHOD

• The above method allows to determine the

probabilities of the states, as well as the
availability and unavailability of the system.
• However, other Reliability Indices are of
interest to evaluate:
• Frequency of finding status,
• Average duration of residence in the
state.

17
To illustrate the evaluation is considered a system with an element,
which has two states.

1
f   A  U
T
It is the frequency of finding a state = probability of being in the
state x state output rate

A U
m r
f f
It is the average duration of residence in the state

18
Example 3.2: Determine the frequency and duration for a two element
system

State Probability F D

19
• Evaluation of Accumulated State Indices:
• States that lead to the same system outcome
(success or failure of the system) must be
accumulated. They are obtained by summing the
probabilities of the states.
• The frequency of finding the accumulated state is
obtained by considering the transitions across the
boundary that surrounds the accumulated state,
applying the previously mentioned concept of
frequency.
• For the example of the system with two elements,
the frequency of finding the accumulated state 34 is
given by:

f34  P3 2  P4 2

20
 APPROXIMATE EVALUATION OF SYSTEM RELIABILITY

Serial system :

 1 ; 1  2 ; 2 = s ; s

1 2 s
 s  1   2 s 
1 2  1 2   21

1 2 s 1r1   2r2

s  rs 
1 2   21 s

s
Us  1  A s    srs  1r1   2r2
s  s
21
• In general, for serial systems you have:

Equivalent failure rate: Average repair time:

n
n
 s   i
  iri
i1 rs  i1
s

Annual Time Out of

n
Us    iri
Service (Unavailability)

i1

22
Parallel system :

 1 ; 1

s s
= ;
 2 ; 2

Us r1r2
rs  
fs r1  r2
1 2 (r1  r2 )
s   1 2 (r1  r2 )
1  1r1   2r2
Us   srs  1 2r1r2
These results do not extend to n element systems.

23
Example 3.3: A feeder is composed of an air section of 2 km in length;
An underground section of 1 km and two cable terminals. Determine
the failure rate, equivalent repair time, and annual unavailability if you
have the following transition speeds:
Section  r
Aerial 0,1 failures/km/year 4h
Underground 0,1 failures/km/year 24 h
Terminals 0,002 failures/ter/year 4h

24
Example 3.4: Two hydro generators of 75 Mw have the same
characteristics of forced disconnections: =0.00488 f/day; r=1,066 days.
Determine the equivalent system parameters for: a) When the load must
be supplied by either unit; And, b) when the load has to be supplied by
the two units.

25