Matrix Analysis of Networks – J. R.

Lucas

Used to have a compact and neat form of solution.

 necessary to know the structure of a network, and
 formulate the problem based on the structure
Because Large networks are
 tedious to analyse using normal equations
 easier/more convenient to formulate in matrix form.

Matrix Analysis of Networks – Professor J R Lucas 1 May 2011

Topology
Deals with structure of an interconnected system
Formulates the problem based on non-measurable
properties of network.
Geometric structure of the interconnection of network
elements completely characterises
 number of independent loop currents
 number of independent node-pair voltages
that are necessary to study the network.

Matrix Analysis of Networks – Professor J R Lucas 2 May 2011

 R1
C
L1

(a) (b)
Figure 1 – Structure of the network
1(a) and (b) have the same structure (or topology).
However elements are quite different.
Matrix Analysis of Networks – Professor J R Lucas 3 May 2011
Graph of Network

Figure 2 – Circuit

(a) Network (b) Graph of Network

Figure 2
Circuit of figure 2(a) also has same topology.
Figure 2(b) shows structure corresponding to all 3
circuits.

Matrix Analysis of Networks – Professor J R Lucas 4 May 2011

Does not indicate any of the elements in the networks.
Known as the graph of the network
– Has all the nodes of the original network
In obtaining the graph,
 each element of the network is represented by a line
 each voltage source by a short-circuit
 and each current source by an open circuit.

Matrix Analysis of Networks – Professor J R Lucas 5 May 2011

Tree of a Network
Which of the diagrams would also represent a normal
tree (without leaves) ? and why ?

Only first diagram would fully satisfy the requirements.

Second diagram has branches closing on itself
 only a tree like Nuga might appear to close on itself
Third diagram has branches in mid air not joined to main tree.
Matrix Analysis of Networks – Professor J R Lucas 6 May 2011
Properties associated with trees.
1. All branches must be part of the tree
2. There cannot be closed loops formed from branches
3. There cannot be branches isolated from the tree
Same properties apply in defining a tree of a network
 can be many trees associated with a given network.
 need not have a trunk coming from the ground and
branches coming from the trunk.
 a reduced graph of network with some of the links
removed so as to leave all the nodes connected
together by graph, but not to have any loop left.
Matrix Analysis of Networks – Professor J R Lucas 7 May 2011
Possible trees for the Graph

Graph of
Network Some of the possible trees
When a tree of the network is removed from graph, what remains
is called the co-tree of the network.

Co-tree is graph of removed links –compliment of the tree.

A co-tree may contain closed loops, and disconnected branches.
Matrix Analysis of Networks – Professor J R Lucas 8 May 2011
Analysis structure of network
A single branch is required to join two nodes.
Node 1 Node 2
Joining each additional node would require an
additional branch. Node 1 Node 2

New Node
Let b = number of branches in the network
n = number of nodes in the network
l = number of independent loops
Thus number of branches in tree = n – 1
number of links removed = b – (n – 1) = b – n + 1
Matrix Analysis of Networks – Professor J R Lucas 9 May 2011
Formation of Independent Loops
If any one of removed links are added to the tree, then
a new loop is formed.
 number of links removed from graph to form the
tree is equal to the number of independent loops.
l=b–n+1
Oriented Graph 1
 Numbered branches with assigned 2 3
directions to currents.
4
 Voltage considered to increase in 5
6
direction opposite to flow of current
Oriented Graph
Matrix Analysis of Networks – Professor J R Lucas 10 May 2011
Matrix Analysis of Networks
To solve circuit problems,
need to write the equations corresponding to
 Ohm’s Law, and
 Kirchoff’s Current Law
 Kirchoff’s Voltage Law
Same is true even when there are a large number of
branches.
 use matrix analysis

Matrix Analysis of Networks – Professor J R Lucas 11 May 2011

Kirchoff’s current Law in matrix form
For any node k
 i1  i2  i4 + i6 + i7 = 0
or +1 i3
i7 -1 0
i1  i2  i4 = i6 + i7 i1 k -1
or i6 i5
i1 + i2 + i4 – i6 – i7 = 0 i2 +1 0
+1 i4
or
+1 . i1 + 1 . i2 + 0 .i3 + 1 .i4 + 0 . i5 – 1 .i6 – 1 .i7 = 0
Last form is preferred for matrix implementation
 all currents in network are included in equation with
different coefficients.
Matrix Analysis of Networks – Professor J R Lucas 12 May 2011
For computer implementation, there must be a unique
method (convention) of obtaining the coefficients ajk.
th
Ij – current in j branch
th
j branch
 directed away from k node: ajk = +1
th

 directed towards k node: th

ajk = –1
 not incident on the k node: th
ajk = 0
th
Kirchoff’s current law may be written, for the k node
a1k . i1 + a2k . i2 + a3k . i3 + a4k . i4 + ...... ....... ..... a7k . i7 = 0
b
 a jk  i j  0 th
or j 1 at k node, for all k
Matrix Analysis of Networks – Professor J R Lucas 13 May 2011
Collection of equations, for each node k, would give

A t I  0
b
(b1) ( n1)
( nb )
t
In [A] , row vectors are dependant, since sum is zero.
t
[A] written with one row less, giving only (n-1) rows.
t
[A] – node-branch incidence matrix, (n-1)b.
[A] – branch-node incidence matrix, b (n-1)
th th
ajk = +1 if j current is directed away from the k node
ajk = 1 if jth current is directed towards the kth node
th th
ajk = 0 if j current is not incident on the k node
Matrix Analysis of Networks – Professor J R Lucas 14 May 2011
Kirchoff’s voltage Law in matrix form
b 0
 brs  vr  0 0 +1 -1 0 0
r 1
th
for s loop, for all s; -1 s
where brs = 1, 0, or +1 -1 0

B b
t V  0
( b1) ( l 1) 0
+1
0
( l b )

[B] – mesh-branch incidence matrix, (lb)

t

[B] – branch-mesh incidence matrix, (bl)

th th
brs = +1 if r current is in same direction as s loop
brs = 1 if r current is in opposite direction to s loop
th th

brs = 0 if rth current is in the not part of the sth loop

Matrix Analysis of Networks – Professor J R Lucas 15 May 2011
Ohm’s Law in matrix form

vk egk
ik ik+ igk Zk

igk

Figure - General branch

for all branches k = 1, 2, .... ... b
vk = – egk + Zk igk + Zk ik
Either voltage source or current source would normally be used.
Matrix Analysis of Networks – Professor J R Lucas 16 May 2011
Conversion with either Thevenin’s Theorem or Norton’s Theorem.
With a voltage source only vk
vk = – egk + Zk ik egk
for all branches k = 1, 2, ....... b ik Zk

and in matrix form as

Vb   E gb  Z b  I b vk
With a current source only ik Yk

ik = Yk vk – igk igk
for all branches k = 1, 2, .... ... b
and in matrix form as
I b   I gb  Yb Vb , where [Y ] = [Z ]-1
b b
Matrix Analysis of Networks – Professor J R Lucas 17 May 2011
In Summary
From Kirchoff’s Laws
A( nb )
t I  0
b
( b1) ( n1) (1) (n-1) independent equations
B t V
( l b )
b
( b1)
 0
( l 1) (2) l independent equations
and from Ohm’s Law
Vb   E gb  Z b  I b (3) b independent equations
I b   I gb  Yb Vb
or (3)* b independent equations
Thus total number of independent equations is
n–1+l+b=b+b=2b
Matrix Analysis of Networks – Professor J R Lucas 18 May 2011
2b independent equations
2b unknowns (b branch currents and b branch voltages)
Can be solved.
Not usual to solve for both current and voltage
simultaneously.
Reductions can be done in two ways.
1) Eliminate voltages and solve for currents
 mesh analysis
2) Eliminate currents and solve for voltages.
 nodal analysis.

Matrix Analysis of Networks – Professor J R Lucas 19 May 2011

Mesh Analysis
 Eliminate the branch voltages from the equations.
 Reduce remaining currents to a minimum using
Kirchoff’s current law.
Apply Kirchoff’s voltage law for solution.
Define a set of mesh currents, I m .
Branch currents I b I
related to mesh currents m by an
algebraic summation.
I b  BI m (4)
Eliminate Vb from the equations,
Matrix Analysis of Networks – Professor J R Lucas 20 May 2011
 t
Pre-multiply equation (3) by [B] .
B Vb  B E gb  B Z b  I b
t t t

t
from equation (2), [B] Vb = 0.

Also
I b  B  I m


B Egb  B Zb BI m
t t

[B] Vb = 0  sum of voltages around a loop is zero.

t

i.e. [B] Vb  sum of voltages around a loop.

t

 [B] Egb  sum of source voltages around a loop.

t

Defined as mesh source voltage vector Egm .

Matrix Analysis of Networks – Professor J R Lucas 21 May 2011
i.e. Egm  [B] Egb t

Egm = B Z b BI m  Z m I m

t

where [Zm] = B Z b B corresponds to l equations
t

[B] also known as the tie-set matrix

(as its elements tie the loop together)
Unknowns are l values of current Im
Original 2b equations and 2b unknowns reduced to
l equations and l unknowns.
Elements of [Zm] can be obtained either from above
mathematics, or by inspection as follows.
Matrix Analysis of Networks – Professor J R Lucas 22 May 2011
Simple evaluation of [Zm] and Egm
zjj = self impedance of mesh j
= sum of all branch impedances in mesh j
zjk = mutual impedance between mesh j and mesh k
= sum of all branch impedances common to mesh j
and mesh k and traversed in mesh direction
 sum of all branch impedances common to mesh
j and mesh k, and traversed in opposite direction
ej = algebraic sum of the branch voltage sources in
mesh j in mesh direction.

Matrix Analysis of Networks – Professor J R Lucas 23 May 2011

Example 1
j20  20  j6 

10 
E1 10 
-j120 
10000 E2
V 100300 V

Solve the circuit using Mesh matrix analysis.

Work from first principles.
Solution
Number the branches and the loops.

Matrix Analysis of Networks – Professor J R Lucas 24 May 2011

 j20  20  j6 
i1 i4 i6
i5 i2
i3 10 
E1 –j120  10 
10000 E2
V I1 I2 I3 10036.870

Write the loop currents in terms of the branch currents.

i1 = I1  i1  1 0 0 
i  0 0  1
i2 = – I3  2    I1 
i3 = I1 – I2 or in matrix form
i3  1  1 0   
  I 2 
i4 = I2 i4  0 1 0   I 
i5 = I2 – I3 i5  0 1  1  3 
   
i6 = I3 i6  0 0 1 
Matrix Analysis of Networks – Professor J R Lucas 25 May 2011
This gives the Branch-Mesh incidence matrix [B].
t
Mesh–Branch incidence matrix [B] can also
independently by writing the relation between the
mesh direction and the branch direction.
1 0 1 0 0 0
B t  0 0  1 1 1 0
0  1 0 0  1 1
Notice that this corresponds to the transpose of the
earlier written matrix.

Matrix Analysis of Networks – Professor J R Lucas 26 May 2011

  10000 
 0
10036.87 
 0 
Vector of branch source voltages is E gb  
 0 
 0 
 
 0 
Branch impedance matrix is
 j 20 0 0 0 0 0
 0 10 0 0 0 0 

 0 0  j120 0 0 0
Zb    
 0 0 0 20 0 0
 0 0 0 0 10 0
 
 0 0 0 0 0 j 6
Matrix Analysis of Networks – Professor J R Lucas 27 May 2011
 t t
Egm = [B] Egb , and [Zm] = [B] [Zb] [B]
 1000 0 
 0
10036 .87
1 0 1 0 0 0  
   0   1000 0 
E gm  0 0  1 1 1 0     
0  1 0 0  1 1  0   0 
 0 , Egm =  10036.87 0 
   
 0 
 j 20 0 0 0 0 0  1 0 0 
 0 10 0 0 0 0  0 0  1
1 0 1 0 0 0 
   0 0  j120 0 0 0  1  1 0 
Z m   0 0  1 1 1 0   
 0 0 0 20 0 0  0 1 0 
0  1 0 0  1 1
 0 0 0 0 10 0  0 1  1
  
 0 0 0 0 0 j 6 0 0 1 

Matrix Analysis of Networks – Professor J R Lucas 28 May 2011

  j 20 0 0 
 0 0  10
1 0 1 0 0 0   j100 j120 0 
   j120 0 
Z m   0 0  1 1 1 0 
j120
  j120 30  j120  10 
0  1 0 0  1 1 
0 20 0   
 0 10  10
=  0  10 20  j 6
 
 0 0 j 6 

Both Egm and Zm could have been written by inspection.

Thus
 1000 0   j100 j120 0   I1 
   j120 30  j120  10   I 
 0    2 
 10036.87 0  =  0  10 20  j 6  I 3 
 
Equations may be solved by inversion or otherwise.
Matrix Analysis of Networks – Professor J R Lucas 29 May 2011
 I1  (30  j120)(20  j 6)  10 2  j120(20  j 6)  j120  10   1000 0 
I   1   j120(20  j 6)  j100(20  j 6)  j100  10

0

 2    
 I 3    j120  10  j100  10 2  0
 j100(30  j120)  ( j120)   10036.87 

 I1   1220  j 2220  j 2400  j 720  j1200   100 
 I   1  j 2400  720  j 2000  600  j1000  0 
 2    
 I 3    j1200  j1000  j 3000  12000  14400  80  j 60
 = (1220 – j2220)(–j100) + (720 – j2400) (j120) + (–j1200) 0
= – j122000 – 222000 + j 86400 + 288000
= 66000 – j 35600 = 74989-28.34o
I1 = (122000 – j 222000 + 0 + j 96000 – 72000)/74989-28.34o
= (50000 – j 126000)/ 74989-28.34o
= 135558-68.36o/74989-28.34o = 1.808-40.02o A
[Note: Inversion has not been checked so answers may be in error.]
Currents I2 and I3 can be similarly determined.
The branch currents i1, i2, ..... may then be determined from the matrix equation.
[Normally branch 6 would have been marked as part of branch 2]
Matrix Analysis of Networks – Professor J R Lucas 30 May 2011
Nodal Analysis
 eliminate branch currents from the equations.
 Reduce number of remaining voltages to a minimum
using Kirchoff’s voltage law.
Apply Kirchoff’s current law for solution.
Define a set of nodal voltages, V N which are node pair
voltages (i.e. voltage across a pair of nodes)
Branch voltages Vb are related to nodal voltages V N by
an algebraic summation.
Vb  AVN (5)
[A] too does not have the reference node.
Matrix Analysis of Networks – Professor J R Lucas 31 May 2011
 t
Pre-multiply equation (3)* by [A] .
A I b  A I gb  A Yb Vb
t t t

t
from equation (1), [A] Ib = 0 .
Substituting from (5)
0  A I gb  A Yb AVN
t t

A I gb  A Yb AVN

t t

IgN = [YN]VN
I
where gN  At
I gb , and

YN   A Yb A t

Matrix Analysis of Networks – Professor J R Lucas 32 May 2011

Source nodal current vector IgN and the nodal
admittance matrix [YN] could be written by inspection.
yii = sum of all branch admittances incident at node i
yij = negative of the sum of all branch admittances
connecting node i and node j .
Reason for negative sign can be understood as follows:
ik = yk vk = yk (Vi – Vj) yk
ik
At any node i,
i
injected current Igi ik =  yk (Vi – Vj) vk
j
N N
I gi   y k Vi   y k V j
j 1 j 1
j i j i for all j
Matrix Analysis of Networks – Professor J R Lucas 33 May 2011
Since Vi is a constant for a given i,
N 
N
 N
 N
I gi  Vi  y k   y k V j    y k Vi   ( y k )V j
j 1 j 1  j 1  j 1
j i j i  j i  j i

N N
I gi  yiiVi   yiiV j   yiiV j
j 1 j 1 corresponds to nodal equation
j i
As in the case of mesh analysis,
IgN = [YN]VN
is first solved to give VN and the branch voltages
and branch currents then obtained using the matrix equations.

Matrix Analysis of Networks – Professor J R Lucas 34 May 2011

Example 2
V1 20  V2 j6 
i1 i4 i2
j20  i3 i5
10 
5-900 A
-j120  10  8.5755.910 A

Example 1 has been reformulated as a problem with

current sources rather than with voltage sources.
[If voltage sources are present, they would first have to
be converted to current sources].
Matrix Analysis of Networks – Professor J R Lucas 35 May 2011
Network may also be drawn in terms of admittances.
0.05
V1 S V2
i1 i4 i2
i3 i5 0.0735 – j
-j0.05
0.0441 S
S 0.1 S
5-900 A j0.00833 8.5755.910 A
S

The branch-node incidence matrix [A],

branch injected current Igb, and
branch admittance matrix may be written,
with reference selected as earthed node as follows.

Matrix Analysis of Networks – Professor J R Lucas 36 May 2011

  1 0   5  90 o 
 0  1  o
  8 . 575 5 . 91 
A   1 0   0 
   
 1  1 ,Igb =  0 ,
 0 1   
 0 
 j 0.05 0 0 0 0
 0 0.0735  j 0.0441 0 0 0 

Yb    0 0 j 0.008333 0 0
 
 0 0 0 0.05 0 
 0 0 0 0 0.1
As in mesh analysis, nodal current injection vector and nodal
admittance matrix may be written from first principles.
Left as an exercise for you to work out.
Matrix Analysis of Networks – Professor J R Lucas 37 May 2011
This is worked by inspection.
 5  90   j 0.05  j 0.00833  0.05  0.05 
I   
o

, Y   
  0.05 0.05  0.1  0.0735  j 0.0441
gN o N
8.575 5.91  

 5  90 o   j 0.05  j 0.00833  0.05  0.05  V1 

 o
  V 
8.5755.91    0 .05 0.05  0.1  0.0735  j 0.0441 2 

V1  1 0.05  0.1  0.0735  j 0.0441 0.05   5  90 o 

     o
 2
V   0 .05  j 0.05  j 0.00833  0.05 8.5755.91 
 = (–j0.05+j0.00833+0.05)(0.05+0.1+0.0735–j0.0441) – 0.052
= (0.05 – j 0.04167)(0.2235 – j 0.0441) – 0.0025
= 0.06509-39.81o0.2278-11.16o – 0.0025
= 0.01483-50.97 – 0.0025
= 0.00934 – 0.0025 – j 0.01152 = 0.00684 – j 0.01152
= 0.0134-59.30o
Matrix Analysis of Networks – Professor J R Lucas 38 May 2011
V1 = (0.2278-11.16o5-90o+0.058.5755.91o)/0.0134-59.3o
= (– 0.2205 – j 1.1175 + 0.4265 + j 0.04415) /0.0134-59.30o
= (0.2060 – j 1.0733)/0.0134-59.30o
= 1.093-79.14o/0.0134-59.30o
o
V1 = 81.6-19.84 V
100  81.6  19.84 o 23.3  j 27.68

branch current i1 = j 20 j 20

i1  1.384  j1.165  1.809  40.09 A

o

which is the same answer (to calculation accuracy) that

was obtained in example 1.

Matrix Analysis of Networks – Professor J R Lucas 39 May 2011

Conversion of Ideal sources
(a) Ideal Voltage sources
No impedance directly in series with voltage source
Z3 Z3 Z3
Z1 Z1
Z2 E Z2
 Z1 E E E
E Z2 or E
Z4 Z5 Z4 Z5 Z4
Z5

Ideal voltage sources are distributed to branches

connected to one of the nodes of original ideal source.

Matrix Analysis of Networks – Professor J R Lucas 40 May 2011

(b) Ideal Current sources
No admittance appears directly in parallel with current source
Is
Is
Is


Is
Ideal current source has been distributed around a loop
connecting the two points of original source.

Matrix Analysis of Networks – Professor J R Lucas 41 May 2011

Port
Pair of nodes across which a device can be connected.
Voltage is measured across the pair of nodes.
Current going into one node is the same as the current
coming out of the other node in the pair.
These pairs are entry (or exit) points of the network.
Compare with an Airport or a Sea Port.
Entry and exit points to a
country.
Planes that enter at a given port
are the ones that take off from

Matrix Analysis of Networks – Professor J R Lucas 42 May 2011

same port.

Matrix Analysis of Networks – Professor J R Lucas 43 May 2011

Two-Port Theory
Convenient to develop special methods for systematic
treatment of networks.
Single-port linear active networks
 Thevenin’s or Norton’s equivalent circuit.
Linear passive networks
 Convenient to study behaviour relative to a pair of
designated ports.
I1 I2
Linear
V1 V2
Port 1 Passive Port 2
Network
Matrix Analysis of Networks – Professor J R Lucas 44 May 2011
Definitions
Driving point impedance is defined as ratio of applied
voltage (driving point voltage) across a node-pair to
the current entering at the same port.
[input impedance of network seen from particular port]
Driving point impedance at Port 1 = V1/I1
Driving point impedance at Port 2 = V2/I2
Driving point admittance is similarly defined as the
ratio of the current entering at a port to the applied
voltage across the same node-pair.
Driving point admittance at Port 1 = I1/V1
Driving point admittance at Port 2 = I2/V2
Matrix Analysis of Networks – Professor J R Lucas 45 May 2011
Immittance is sometimes used to represent either an
impedance or an admittance
Transfer impedance is defined as the ratio of the
applied voltage across a node-pair to the current
entering at the other port.
Transfer impedance = V1/I2 , V2/I1
Transfer admittance is similarly defined as the ratio of
the current entering at a port to the voltage appearing
across the other node-pair.
Transfer admittance = I1/V2 , I2/V1

Matrix Analysis of Networks – Professor J R Lucas 46 May 2011

Transfer Voltage gain (or ratio) is defined as the ratio
of the voltage at a node pair to the voltage appearing at
the other node-pair.
Transfer voltage gain = V1/V2 , V2/V1
Transfer Current gain (or ratio) is similarly defined
as the ratio of the current at a port to the current at the
other port.
Transfer current gain = I1/I2 , I2/I1

Matrix Analysis of Networks – Professor J R Lucas 47 May 2011

Common Two-port parameters
External conditions of a two-port network can be
completely defined by currents and voltages at the 2 ports.
A general two port network can be characterised by four
parameters, derived from the network elements.
With symmetry, number of parameters will be reduced.
(a) Impedance parameters
(b) Admittance parameters
(c) Transmission Line parameters
(d) Hybrid parameters.

Matrix Analysis of Networks – Professor J R Lucas 48 May 2011

(a) Impedance Parameters (z-parameters)
or Open-circuit parameters
I1 I2
Linear
V1 Port 1 Passive Port 2 V2
Network

V1   z11
z12   I 1 
V    z
  
 2   21
z 22   I 2 
V1 = z11 I1 + z12 I2
If I2 = 0, then z11 = V1/I1
If I1 = 0, then z12 = V1/I2
Matrix Analysis of Networks – Professor J R Lucas 49 May 2011
It follows that,
V1 V1
z11  z12 
I1 I 2 0 , I 2 I1 0 ,
V2 V2
z21  z22 
I1 I 2 0 , I 2 I1 0 .
z-parameters correspond to the driving point and
transfer impedances at each port with the other port
having zero current (i.e. open circuit).
 open circuit parameters.
Matrix Analysis of Networks – Professor J R Lucas 50 May 2011
Example 3
Find impedance parameters of the two port T – network.
With port 2 on open circuit I1 Za Zc I2
V1
z11   Z a  Zb
I1 I 2 0 V1 Port 1 Zb Port 2 V2

V2
z21   Zb
I1 I 2 0
similarly with port 1 open,
z12 = Zb Z a  Zb Zb 
 Z   
z22 = Zb + Zc  Zb Z b  Z c 

Matrix Analysis of Networks – Professor J R Lucas 51 May 2011

(b) Admittance Parameters (y-parameters)
or Short-circuit parameters
I1 I2
Linear
V1 Port 1 Passive Port 2 V2
Network

 I 1   y11 y12  V1 

I    y y 22  V2 
 2   21
y11, y12, y21, y22 defined with either V1 or V2 zero.
y-parameters correspond to driving point and transfer
admittances at each port with the other port having zero
voltage (i.e. short circuit)  short circuit parameters.
Matrix Analysis of Networks – Professor J R Lucas 52 May 2011
Matrix Analysis of Networks – Professor J R Lucas 53 May 2011
Example 4
Find admittance parameters of the 2 port –network.
I1 Yb I2 I1 Yb I2

V1 V1 V2=0
V2 Ya Yc
Ya Yc

with port 2 on short circuit

I1
y11 = V1 V2 0 = Y a + Yb
Ya  Yb  Yb 
 Y 
I2 [Y] =  Yb  Yc 
y21 = V1 V2 0 = – Yb b

Matrix Analysis of Networks – Professor J R Lucas 54 May 2011

(c) Transmission Line Parameters (ABCD-parameters)
I1 I2
Linear
V1 Passive
Port 1 Port 2 V2
Network

V1   A B  V2 
 I   C D   I 
 1   2 
Parameters can be defined using either port 2 on short
circuit or port 2 on open circuit.
In case of symmetrical system, parameter A = D.
For a reciprocal system, A.D – B.C = 1
Matrix Analysis of Networks – Professor J R Lucas 55 May 2011
Example 5 Yc
I1 I2
Find ABCD parameters.
Yb  Yc 1
V1 V2
A= Yc , B= Yc Ya Yb
Ya Yb  YbYc  YcYa
C= Yc
Y a  Yc
and D = Yc [For symmetrical network, Ya = Yb , A = D].
Yb  Yc Ya  Yc 1 Ya Yb  Yb Yc  Yc Ya
  
A.D – B.C = Yc Yc Yc Yc
YbYa  YcYa  YbYc  Yc2  (YaYb  YbYc  YcYa )
= Yc2 =1
Matrix Analysis of Networks – Professor J R Lucas 56 May 2011
(d)
I1 I2
Linear
V1 Port 1 Passive Port 2 V2
Network

Hybrid Parameters (h-parameters)

Matrix Analysis of Networks – Professor J R Lucas 57 May 2011

The hybrid parameter matrix may be written as
V1   h11 h12   I1 
 I   h   
 2   21 h22  V2 
h-parameters can be defined as in other examples, and
are commonly used in some electronic circuit analysis.

Matrix Analysis of Networks – Professor J R Lucas 58 May 2011

Interconnection of two-port networks
(a) Series connection of two-port networks
Ir1 Ir2
Linear
Vr1 Passive
V1 Port r1 Network Port r2 Vr2
V2
r
Is1 Linear
Is2
Vs1 Passive
Port s1 Network Port s2 Vs2
s
Series properties are applied to each port
at port 1, Ir1 = Is1 = I1, and Vr1 + Vs1 = V1
at port 2 I r2 = Is2 = I2, and Vr2 + Vs2 = V2
[Z] = [Zr] + [Zs]
Matrix Analysis of Networks – Professor J R Lucas 59 May 2011
(b) Parallel connection of two-port networks
I1 Ir1 Ir2 I2
Linear
Passive Vr2
V1 Vr1 Network
r V2
Is1 Is2
Linear
Passive Vs2
Vs1 Network
s
at port 1, Ir1 + Is1 = I1, and Vr1 = Vs1 = V1
at port 2, I r2 + Is2 = I2, and Vr2 = Vs2 = V2
[Y] = [Yr] + [Ys]
Matrix Analysis of Networks – Professor J R Lucas 60 May 2011
(c) Cascade connection of networks
Output of one network becomes input to next.
Ir2 = Is1
Vr2 = Vs1
Ir1 Ir2 Is1 Is2
Linear Linear
Vr1 Port r1 Passive Port r2 Vr2 Vs1 Port s1 Passive Port s2 Vs2
Network Network

Vr1   Ar Br  Vr 2  Vs1   As Bs  Vs 2 

 I   C     I   C   
 r1   r Dr   I r 2  ,  s1   s Ds   I s 2 
V1   Ar Br   As Bs  V2 
 I   C     
 1  r Dr  C s Ds   I 2 
Matrix Analysis of Networks – Professor J R Lucas 61 May 2011
ABCD matrix of component networks
I1 I2 I1 I2
Z
V1 V2 V1 V2
Y

V1
A = V2 I 2  0 = 1, =1
V1
B = I 2 V2  0 = Z, =0
I1
C = V2 I 2  0 = 0, =Y
I1
D = I 2 V2 0 = 1, =1
Matrix Analysis of Networks – Professor J R Lucas 62 May 2011
In matrix form
A B 1 Z   1 0
C D  0 1  Y 1
  =  , =  
Consider example 5 again
I1 Yb I2 Yb

V1
Ya Yc V2  Ya Yc

A B 1 0 1 1  1 0
C D  Y   Yb   
  =  a 1 0 1  Yc 1
Simplification of matrix product would give the same answer as in example 5.
Matrix Analysis of Networks – Professor J R Lucas 63 May 2011