Complex short circuit MVA method for power

system studies

T.-H. Chen

Indexing terms: Complex short circuit M V A method, Ohmic method, Per unit method, Powerflow, Short circuit

Second, by using the proposed method the conversion

Abstract: A ‘complex short circuit M V A method’ results and all other system parameters are unique. This
is proposed as an alternative to the standard per feature is of potential value in the teaching and learning
unit method for power system studies. It removes process. For students it is simpler and less susceptible
the need for base quantities, and adopts the repre- to unintentional errors than current methods and for tea-
sentation of power system elements in terms of chers it is easy to grade learning and testing assignments.
M V A instead of the conventional per unit quant- Third, by using the proposed method most of the
ities. The proposed approach is simpler and less system parameters are represented in terms of their
susceptible to unintentional errors than current actual quantities, especially in power flow analysis.
methods. It has potential value in training and Therefore, it avoids the need for converting these kinds of
industry. The basic methodology and applications parameters to per unit quantities and, after completing a
to short-circuit calculations are presented. calculation, converting back to the corresponding actual
quantities as required by the per unit method. Hence, the
approach is more straightforward than that found in con-
ventional methods. Compared with the standard per unit
1 Introduction method, it reduces the source and object program codes
and improves the efficiency of developing and utilising
The standard per unit method is normally used in power the programs.
system studies [I-lO]. Base quantities are arbitrarily Finally, the proposed method is similar to the per-unit
selected and, in practice, it is found to be more con- method. All the methodologies of the per unit method
venient to select base kV and base kVA. The base values can be easily applied to the proposed method. Therefore
of other quantities are thus automatically fixed. All parts no difficulty exists for those who are familiar with the per
of an electric circuit (or system) are expressed in per unit unit method. However, although the proposed method
terms on the basis of these selected base quantities. appears like a special case of the per unit method, it is
However, because the base quantities are arbitrarily not. Also, because the proposed method seldom requires
selected, an unsuitable selection is possible and this may formulae to be memorised, it is easy-to-learn and easy-to-
result in truncation or round-off errors. use, especially for hand calculations.
Another method for solving electric circuit problems is
the Ohmic method. Converting impedance from one 2 Conversion formulae
voltage to another is necessary for the multivoltage
system, i.e. an electric power system. The proposed method is basically a modification of the
Conversion formulae used for both .the Ohmic and the M V A method [ll]. Like the M V A method, the complex
per unit methods are complex and not easy to memorise. short circuit M V A method is used by separating the
Furthermore, most of the data obtained must be con- circuit into components, and calculating the complex
verted back to their corresponding actual quantities after short circuit M V A of each component with its own infin-
the calculation has been completed. An alternative ite bus according to the scheme in Fig. 1.
method, called the ‘complex short circuit M V A method‘,
is introduced here to perfect the calculation procedure. infinite bus
The proposed method disperses with the need for base
quantities, and adopts the representation of power
system elements in terms of M V A instead of the conven- generator,
tional per unit quantities. Some advantages are briefly
described below. conductor,
First, as stated above, the proposed method dispenses etc.
with the need for base quantities and so the data input is
simplified and the memory-space and CPU-time require- short circuit
ments are reduced. Although the improvements are not to ground
considerable, it nevertheless improves the user-friendly Fig. 1 Scheme
feature of commercial programs and is therefore of value
to industry. 2.1 Generators.synchronous motors, induction
motors, transformers,etc.
0IEE, 1994 The conversion formula for generators, motors, and
Paper 9865C (Pll), received 16th April 1993 transformers is as follows
The author is with the National Taiwan Institute of Technology, Rated M V A
Department of Electrical Engineering and Technology, 43, Keelung MVA,= MVA
Road Section 4, Taipei, 106, Taiwan, Republic of China L, PU

I E E Proc.-Gener. Transm. Distrib., Vol. 141, No. 2, March 1994 81

where subscripts sc and pu refer to short circuit and per where the nominal kV is equal to base kV because the
unit, respectively. nominal voltages are selected as base voltages. Therefore,
It is important to note that all impedance or admit-
tance parameters of the devices listed above for short Y, = MVASC
circuit calculations need to be converted to their corres- Base M V A
ponding short circuit M V A quantities. Only the param- The above equations show that the relationship of the
eters of the transformers, however, need to be converted complex short circuit M V A and per unit admittance is
for power flow analysis. This is because the actual rated unique for all power system elements. That is
real and reactive powers of generators and motors are
used in the power flow studies by the proposed method. MV4,
Y, =
Base M V A
2.2 Conductors (feeders,cables) 3.2 Selected base voltages are not nominal voltages
The conversion formula for the conductors is If the selected base voltages are not the nominal voltages,
the relationship of these two values becomes
M V A , = kV:,,, M v A (2)
Z(n) (Base kV)' M VA,
Y, = (74
The conversion formulae, eqns. 1 and 2, corroborate with (Nominal kv)' Base M V A
the statements that the proposed method is an easy-to- or
learn and easy-to-use method.
M VA,
Y,=Cx
Base M V A
3 Relationship of complex short circuit M V A and In the per unit method the most convenient way is to
per unit quantities select the nominal kV as base k V . The relationship of the
The relationship of complex short circuit MVAs and per complex short circuit M V A and per unit admittance can
therefore be simply represented by eqn. 6. This relation-
unit admittances can be described as set out below.
ship results in the bus admittance matrix building algo-
rithm being directly applied to build the bus short circuit
M V A matrix. For convenience, the notation M V A , is
3.1 Nominal voltages are selected as base voltages
replaced with M , i.e. M = M V A X .
3.I . 1 Generators, synchronous motors, induction Based upon eqn. 6, the relationship of the bus admit-
motors: tance and bus short circuit M V A matrices, [Mbu,],is
Base M V A 1
zp = zp(at rated values)
Rated M V A CMbus1
= Base M V A
-zp rated values)
- (PI
x Base M V A Therefore,
Rated M V A
[ M b J = Base M V A x [&.,I (9)
- Base M V A
-
M VA,
(3) The respective elements of [&.,I and [ M b u J are in p.u.
and M V A . For a given system, [Mbu,] is unique, but
Therefore, [YhJ is not. The latter is dependent on the base quant-
ities selected.
M VAx
Y, = (4) If the inverse of the short circuit M V A has been used
Base M V A to build [Wbu,], the relationship of the bus impedance
where Zp(atratedva,ues) can be Z&.or Z d . If momentary matrix, [Z,,], and bus inverse short circuit M V A matrix,
duty is required, 2; is adopted, otherwise Z d is used for [Wbu,], can then be represented as follows
interrupting duty evaluation. This kind of conversion is [Zbu,]= Base M V A x [Wbu,] (10)
needed for short circuit analysis only.
where
3.12 Transformers: The conversion formula for trans- [~.,=
l [Mb.,l-' (1 1)
formers is the same as that of generators and motors
(eqn. 3). If the transformer impedance is given as a per- All the bus short circuit MVAs, bus inverse short circuit
centage, it needs to be divided by 100 to obtain its corre- MVAs, bus admittance, as well as the bus impedance
sponding per unit value. So, matrices are usually complex matrices. These matrices
can also be made real for the sake of simplicity. In this
case the conversion formulae and solution techniques are
still the same.
3.1.3 Conductors:
4 Applications on short circuit calculations

An interesting proportional relationship of the per unit

impedance and inverse short circuit M V A values is
- zfi
(Nominal kV)'
x Base M V A shown in eqn. 6. Based on this relationship, the solution
technologies used in the conventional per unit method
- Base M V A (5)
can be directly applied to the inverse short circuit M V A
M VA, method with only a little modification. As with the per
82 IEE Proc.-Gener. Transm. Distrib., Vol. 141, No. 2, March I994
unit method, short circuit problems can be solved by As conversion is being made, an inverse short circuit
hand calculations and computer programs as now M V A diagram is being developed. The one-line diagram,
described. Fig, 2, is replaced with an inverse short circuit M V A
diagram shown in Fig. 3.
4.1 Hand calculations
Short circuit problems for a small power system are con- ulllily line
veniently solved by hand calculations. How to combine short-circuit level
500 MVA
the inverse short circuit M V A values in series and in

i
parallel, and how to perform delta to wye conversions, or 13 8 k V
vice versa, are necessary work while conducting the short
circuit studies by hand calculations. Combinations of the
inverse short circuit M V A are easily seen from eqn. 6 to 13 8kV
be exactly the same as that of impedance. Combinations z=ooi51.10i5n
of inverse short circuit MVAs are therefore only a matter
of simple arithmetic. Procedures for the short circuit
hand calculations are as follows
( a ) All components are converted to inverse short
A ' '1 5000kVA
13 8kV-24kV

circuit MVAs.
(b) The inverse M V A diagram is reduced by combining
the inverse MVAs, following the same algorithms of com-
bining the per unit impedances.
(c) The short circuit M V A is calculated at the fault 2500 kVA
point. fault X''d 0.16 p.u
( d ) The total fault current is calculated.
Fig. 2 Three-bus system

4.2 Computer programs

Similar to the per unit method, the bus inverse short reduced
circuit M V A matrix has, however, been built instead of
the per unit bus impedance matrix. Procedures for the
short circuit calculation by computer program are as
follows I4
(a) All components are converted to inverse short
circuit MVAs.
[*)
'I[=]
+jO.O00793
(b) The bus inverse short circuit M V A matrix, W,,,, is
formed by following the same algorithm of the formation reduced $
of the bus impedance matrix.
(c) The short circuit M V A is calculated at each node of
the system by using 00120 /74*
1
MVA,,,,, =- i = 1, . . . , n
K,i
( d ) The total fault current is calculated as F
Fig. 3 Inverse short circuit M V A and reduced diagram

Step 2 : The inverse short circuit M V A diagram is

4.3 Case results reduced by combining the inverse short circuit MVAs,
The three-bus system shown in Fig. 2 is used as a sample following the same rules of combining the per unit
system. impedances.
Step 3: The short circuit M V A is calculated at the
fault point. This is
4.3.1 By hand calculations
Step I : All components are converted to inverse short 1
MVA,, = = 83.33L -74.11" M V A
circuit M V A S . 0.012 L 74.1 1"
j Step 4 : The total fault current is calculated at the fault
W, = - = j0.002 M V A - '
point.
500

w,= 0.0151 + j0.151 83.33 M V A -

-
83.33 M V A
= 20.046 k A
(13.8 kV)2 I F= J(3) V J(3) x 2.4 kV

= 0.0000793 + jO.OOO793 M V A - '

4.3.2 By computer program
w,= 0.024 + j0.055 = 0.0048 + jO.011 M V A - '
5
Step 1 : All components are converted to inverse short
circuit MVAs.
Step 2 : The bus inverse M V A matrix, W,,,, is formed
j0.16 by following the same algorithm of the formation of the
W, = ~ = j0.064 M V A - '
2.5 bus impedance matrix.
I E E Proc.-Gener. Transm. Distrib., Vol. 141, N o . 2, March I994 83
 5 Conclusion
0.003 + j1.949 0.002 + j1.928
0.002 + j1.928 0.080 + j2.693 An effective straightforward method has been introduced
for solving power flow and short circuit problems. The
-0.103 + j1.639 -0.079 + j2.293 proposed method has provided power engineers with a
+j639
1
-0.103 good alternative method for conducting power system
analysis and simulation. Many advantages have been
-0.079 + j2.293 x
found when compared with the per unit and Ohmic
3.289 + jl1.554 methods. The proposed method has removed the need of
base quantities, simplified the conversion formulae and

i
1.94879L89.91 1.92835 L 89.93
unified the system parameters. All this suggests it could
= 1.92835 L 89.93 2.69394 ~ 8 8 . 3 0 be a valuable asset in the power industry and in educa-
1.64216L -93.59 2.29413L -91.96 tional training. Although improvements in memory
space, savings in computation time, etc. have not been
1.64216i -93.59
2 . 2 9 4 1 3 ~-91.96
12.01282 ~ 7 4 . 1 1
Step 3 : The short circuit MVAs are calculated as
1 x
significant, it has, nevertheless, become a reasonable
alternative to the standard per unit method convention-
ally practised by power engineers, students and teachers.

6 References
1 1
MVA,,,,, = -= 1 STAGG, G.W., and EL-ABIAD, A.H.: ’Computer methods in
Wl, 0.001 948 79 L 89.91
power system analysis’ (McGraw-Hill, New York, 1968)
= 513.14L -89.91 M V A 2 STEVENSON, W.D., Jr.: ‘Elements of power system analysis’, 4th
edn. (McGraw-Hill, New York, 1982)
1 1 3 WEEDY, B.M.: ‘Electric power systems’, 3rd edn. (Wiley, New
MVAs(2) = -- York, 1979)
W2, 0.002 693 94 L88.30 4 BROWN, H.E., PERSON, C.E., KIRCHMAYER, L.K., and
STAGG, G.W.: ‘Digtal calculation of three-phase short circuits by
= 371.12L -88.30 M V A matrix method‘, Trans. AIEE, Feb., 1961, Part 111, 79, pp. 1277-
1282
1 1 5 EL-ABIAD, A.H.: ‘Digital calculations of line-to-ground short cir-
MVAX(3) = - -
W3, - 0.0120 128 2 ~ 7 4 . 1 1 cuits by matrix method‘, Trans. AIEE, June 1960, Part 111, 79,
pp. 323-332
= 83.33i -74.11 M V A 6 ROY, L.: ‘Generalized polyphase fault-analysis program: calcu-
lation of cross-country fault’, Proc. IEE. Oct. 1979, 126, (lo), pp.
Step 4: Calculate the total fault currents 995-1001
7 GLIMN, A.F., and STAGG, G.W.: ‘Automatic calculation of load
flows’, Trans. A I E E , 1957, Part 111,76, pp. 817-828
8 TINNEY, W.F., and HART, C.E.: ‘Power flow solutions by
Newton’s method‘, Trans. IEEE, Nov. 1967, PAS-% p. 1449
9 BROWN, H.E.: ‘Solution of large networks by matrix methods’, 2nd
edn. (Wiley and Sons, New York, 1985)
10 GLOVER, J.D., and SARMA, M.: ‘Power system analysis and
design with personal computer applications’ (PWS Publishers,
Boston, 1987)
11 YUEN, M.H.: ‘Short circuit ABC‘ (Gas Industries Association, San
Francisco, 1975)

84 IEE Pro=.-Gener. Transm. Distrib., Vol. 141, No. 2, March 1994