People’s Democratic Republic of Algeria

Ministry of Higher Education and Scientific Research

University M’Hamed BOUGARA - Boumerdes

Institute of Electrical and Electronic Engineering

Department of Electronics

Project Report Presented in Partial Fulfillment of the Requirements

of the Degree of:
‘LICENCE’
In Electrical and Electronic Engineering

Title:

Automation of Network Analysis Techniques

Based on Graph Theory

Presented By:
AZZAZ Aissa

Supervisor:
Pr. Abdelmadjid Recioui

Registration Number: 181834022005/2021

 Abstract
The aim of this project is to build a program to demonstrate the different network analysis techniques
based on graph theory. All the techniques will be implemented in MATLAB. A comparative study is
done on the different techniques.
The project studies and demonstrates the nodal, loop and cutset network analysis techniques.The
performance assessment of these techniques is done by comparing the results with some known sim-
ulators such as NI Multisim.
We have come to a conclusion that the nodal analysis method is the easiest one to perform since it
could be applied directly without the need to extract any further matrices except the incidence matrix.

1
Contents

Title Page 1

Abstract 1

Contents 2

Introduction 4

1 Introduction to graph theory 5

1.1 Electrical networks and graph theory . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Matrix representation of a graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.1 The incidence matrix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.2 The fundamental loop (Tie-set) matrix B . . . . . . . . . . . . . . . . . . . . 12
1.3.3 The fundamental cut set matrix D . . . . . . . . . . . . . . . . . . . . . . . 15

2 Network analysis techniques 18

2.1 Generalized branch model in electrical networks . . . . . . . . . . . . . . . . . . . . 18
2.2 Impedance Z and admittance matrices Y . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.1 Example 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2.2 Example 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3 Analysis Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.1 Nodal analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.2 Loop Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.3 Mesh analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.4 Cutset analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3 Implementation 27
3.1 MatLab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2
Contents Contents

3.2 Block diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.3 Simulation program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.3.1 Get AA , A, B, D functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.3.2 Get tree function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.4 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4.1 Example 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4.2 Example 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.4.3 Example 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Conclusion 37

References 39

3
Introduction

The field of graph theory is a vastly used and well established mathematical field. The usefulness of
graph theory lies in its ability to model many different science areas and solve their fundamental prob-
lems. Graph theory finds use cases in network traffic, neuronal networks, social networks, biological
networks and computer networks. [1] [2] [3]
Electrical networks theory and analysis often uses graph theory principles. The simple and fundamen-
tal laws of electrical circuits (KVL, KCL) could handle simple networks very easily and perfectly, but
when it comes to huge electrical networks such as those modeling the behavior of large systems, the
simple laws and approaches could become very hard and cumbersome to apply especially if they are
applied manually.
Almost all electrical simulators make use of graph theory principles in which they extract information
from the circuit being simulated and then apply the graph theory principles to obtain the different
network parameters. An example of this, is NI Multisim[4] simulation software which uses the nodal
analysis method implemented on the SPICE (Simulation Program with Integrated Circuit Emphasis)
engine[5] to perform circuit simulation.[6]
In this project, we examine some of the main principles of graph theory in the first chapter along
with the basic representations of graphs through matrices. In the second chapter, we illustrate and
approach network analysis using the graph theory principles obtained earlier by explanting how elec-
trical networks are represented through graphs. In the third chapter, we give a brief explanation on
how to build a program implementing the different network analysis techniques presented in earlier
chapters using MatLab, along with a comparison with NI Multisim.

4
Chapter 1

Introduction to graph theory

1.1 Electrical networks and graph theory

In general terms, any network is a combination of elements. The interconnection of two or more
simple elements (Active and passive) is called an electrical network. Any electrical network could be
converted into a graph, this is known as topology. [7]
Graphs are useful abstraction layers that provide us with easier ways to solve realistic problems by
applying the different graph theory rules. In electrical engineering, those problems are concerned
with the analysis of electrical networks to solve for different network parameters. [8] [9] [10] [11]
[12]

0.5 mH
5
2 mH 1 kΩ
1 2 3
2 4
1 2 3
3 kΩ

0.25 kΩ 2 nF 3
1 6
+
v(t) −
0
0

Figure 1.1: Circuit to graph transformation

1.2 Graphs
In order to proceed any further in the topic of graph theory, we have to define some basic definitions
and concepts that are necessary and useful for a proper understating [13] [14]:

5
1.2. Graphs Chapter 1. Introduction to graph theory

Graph: a collection of nodes and branches that connect these nodes. Figure 1.2 shows examples of
graphs.
Node: an end point of a line segment or an isolated point. It is also called a vertex or junction.
Branch: a line segment which represents a network element or a combination of elements connected
between two points. Each branch joins two distinct nodes. It is also called an edge.
An incident branch izs a branch that starts or ends at that node. In the example of Figure 1.2(b),
branch ’1’ is incident to nodes "1" and "2", whereas branch 8 is incident to nodes "2" and "6".
Path: any sequence of nodes connected by branches. A path maybe closed or open.In the graph
Figure 1.2(b), the path connecting nodes 1,2,3 and 5 by the branches 1,2,3 and 4 is a closed path
whereas the path connecting nodes 2,3,5 and 4 by the branches 2,3 and 6 is an open path.
Loop: any closed path. In the graph Figure 1.2(b), the path joining nodes 5,6 and 4 using the branches
7,5 and 6 is a closed path.
Mesh: a loop that doesn’t contain any other loops within it.
Connected graph: a graph in which at least one path could be found between any two nodes of it.
Directed (Oriented) graph: a graph in which all the nodes and branches are numbered or a direction
is assigned to the branches by an arrow. Figure 1.2 shows an example of a direct and undirected graph.
Subgraph: a subset of the original graph that contains fewer nodes and branches than the original
one. The example of Figure 1.3(a) shows an arbitrary subgraph of the graph on Figure 1.2(b).
Tree: a connected subgraph with no loops. The subgraph of Figure 1.3(b) shows an arbitrary tree
whereas the one on Figure 1.3(a) does not because it contains a loop. The tree contains all nodes of
the graph and only one branch (twig) is connecting any two nodes, hence the number of twigs in a
tree is one less than the number of nodes (N − 1).
The branches of a tree are called twigs and the remaining ones are called links. Branches 1,2,3,5 and
7 of the graph on Figure 1.2(b) generate a tree and the corresponding links are the branches 4,6 and 8.
Co-tree: the subgraph formed by the set of links (the complement of a tree). The subgraph of Fig-
ure 1.3(c) shows the corresponding co-tree of the tree shown in Figure 1.3(b).

6
1.2. Graphs Chapter 1. Introduction to graph theory

1
1
1

2
2
2 4

3
6
8 3

5
3

4 6 6

5 4
(a) Undirected Graph (b) Directed Graph

Figure 1.2: Types of graphs [15]

1
1
1
2
2 2
2
2 4
3
3
3
3 8
5
5 5
7
7
6 6 6 6
6
5
5
4 4
4
(c) Co-tree
(a) Subgraph
(b) Tree

Figure 1.3: Examples of graph elements

7
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

1.3 Matrix representation of a graph

For a given oriented graph, there are several representative matrices. They are extremely important
in the analytical studies of a graph, particularly in the computer aided analysis and synthesis of large
scale networks.

1.3.1 The incidence matrix A

The complete incidence matrix AA

A directed graph may be described completely in a compact and convenient matrix, this is known as
the complete incidence matrix AA . It describes how nodes and branches of a graph are interconnected.
To obtain A we shall first obtain the complete node to branch matrix AA . For a graph containing N + 1
nodes numbered from 0 to N and B branches, AA is defined as:

+1 if branch j is directed away from node i

ai j = 0 if branch j is not incident to node i

−1 if branch j is directed towards node i


The graph in Figure 1.4 has 5 nodes and 8 branches, hence the construction of AA can be organized in
Table 1.1.

2 4
1 2 3

1 7
3 6

0 4
8
Figure 1.4: Graph Example

Table 1.1: Construction of AA

Branches
1 2 3 4 5 6 7 8
Nodes
0 -1 0 -1 0 0 0 0 -1
1 1 1 0 0 0 0 0 0
2 0 -1 1 1 1 0 1 0
3 0 0 0 -1 -1 1 0 0
4 0 0 0 0 0 -1 -1 1

8
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

So, the complete incident matrix AA is:

 
−1 0 −1 0 0 0 0 −1
1 1 0 0 0 0 0 0
 
AA =  0 −1 1
 1 1 0 1 0
0 0 0 −1 −1 1 0 0
0 0 0 0 0 −1 −1 1

Construction of the incidence matrix A

The complete incidence matrix is characterized by the fact that the sum of all elements along each
column is zero because a branch would leave a node and enter another one. As a result, eliminating
one row from AA wouldn’t affect the information conveyed about the graph. This elimination of one
row from AA would result in the incidence matrix A (also called the reduced node to branch incidence
matrix).
In the graph of Figure 1.4, eliminating the first row corresponding to node 0 we could obtain the
incidence matrix as:  
1 1 0 0 0 0 0 0
0 −1 1 1 1 0 1 0
A= 0 0 0 −1 −1 1

0 0
0 0 0 0 0 −1 −1 1

Partitioned incidence matrix

The incidence matrix A could be partitioned such that N columns correspond to the twigs of an
arbitrary selected tree and the remaining B − N columns represent the co-tree links. Hence, A would
be partitioned to square sub-matrix At and a rectangular sub-matrix Ac which results in:
 
A = At Ac

In At at least one column has only a non zero entry that is +1 or −1 because any branch is incident to
two nodes. Hence, the matrix At is a non-singular (invertible) matrix and det(At ) = ±1.
In the graph of Figure 1.4, if branches 2,4,6 and 8 are selected to form an arbitrary tree then the
incidence matrix could be partitioned as follows:
 
1 0 0 0 1 0 0 0
   −1 1 0 0 0 1 1 1 
A = At Ac =   0 −1 1 0 0 0 −1 0
 ; det(At ) = +1

0 0 −1 1 0 0 0 −1

The matrix AAT

The matrix AAT is an N order square matrix with entries:

(
Diagonal elements aii = number of branches incident to node j
Off-diagonal elements ai j = −(total number of branches linking to node i to node j)

9
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

Using the Binet-Cauchy theorem:

det(AAT ) = sum of the squares of all majors of A
we would obtain the following:
det(AAT ) = Total number of trees in a graph

In the graph of Figure 1.4 and the incidence matrix obtained earlier, we get :
 
2 −1 0 0
 −1 5 −2 −1 
AAT =   0 −2 3 −1  ; det(AA ) = 34
 T

0 −1 −1 3
which means there are 34 possible trees in the graph shown of Figure 1.4.

Branch to node voltage transformation

In an electric network, V denotes the column vector of branch voltages and E the column vector that
represents the node voltages with respect to the reference node (the one eliminated from the complete
incidence matrix ). A relation linking V, E and the incidence matrix A is formed as:

V = AT E (1.3.1)

vk

+ −
i kth branch element j
+ +
ei ej

− o −

Figure 1.5: Branch to node voltage transformation

Kirchhoff’s current Law (KCL)

KCL states that the algebraic sum of the currents entering a node is equal to the algebraic sum of
currents leaving it [16]. In other words, the sum of currents incident to a node is zero. If I denotes the
column matrix of branch currents, that is:
AI = 0 (1.3.2)

The nonzero entries of any row of A reflect branch incidence at the node corresponding to the row
and the sign is one to one correspondence to branch directions.
If I is partitioned for some tree and co-tree as:
 
It
I = · · ·
Ic

10
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

Then and since At is invertible ,we get :


  It
AI = At Ac · · · = At It + Ac Ic = 0
Ic

It = −At−1 Ac Ic (1.3.3)

In the graph in Figure 1.6, taking the branches 1,3 and 6 as a tree results in the following partitioned
incidence matrix:
 
−1 0 0 1 0 1 0
A =  0 1 0 −1 1 0 0
0 0 1 0 −1 −1 1

2 4
1 2 3

3
1 6
7
0

Figure 1.6: A graph for a certain circuit

As a result, the tree current could be expressed in terms of the co-tree currents as follows:
  
 i2
i1 = i2 + i5
  
i1 1 0 1 0   
−1 i 4
It = i3  = −At Ac Ic = 1 −1 0 0  
i5  ⇒ i3 = i2 − i4

i6 0 1 1 −1 
i6 = i4 + i5 − i7
i7

11
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

1.3.2 The fundamental loop (Tie-set) matrix B

The complete loop to branch incidence matrix BA

The complete loop to branch incidence matrix BA conveys information about the branches direction
when traversing the loops obtained from a selected tree of a graph.
The complete loop-to-branch incidence matrix BA has entries as follows:

+1 if branch j is oriented in the same direction as the path of loop i

bi j = 0 if branch j is not incident with loop i i

−1 if branch j is oriented in opposite direction as the path of loop i


The graph of Figure 1.7 contains 6 loops:

{1; 2; 5}, {2; 3; 4}, {4; 6; 5}, {1; 2; 4; 6}, {1; 3; 4; 5}, {2; 3; 6; 5}, {1; 3; 6}. The loops are organized in
Table 1.2.

2 4
1 2 3

5
1 6

Figure 1.7: Graph Example

Table 1.2: Construction of BA

branches
1 2 3 4 5 6
Loops
L1 : {1; 2; 5} 1 1 0 0 1 0
L2 : {2; 3; 4} 0 -1 1 -1 0 0
L3 : {4; 6; 5} 0 0 0 1 -1 1
L4 : {1; 2; 4; 6} 1 1 0 1 0 1
L5 : {1; 3; 4; 5} 1 0 1 -1 1 0
L6 : {2; 3; 6; 5} 0 -1 1 0 -1 1
L7 : {1; 3; 6} 1 0 1 0 0 1

Then, the complete loop to branch incidence matrix BA is:

 
1 1 0 0 1 0
0 −1 1 −1 0 0
 
0 0 0 1 −1 1
 
BA = 1 1 0 1 0 1

1 0 1 −1 1 0
 
0 −1 1 0 −1 1
1 0 1 0 0 1

12
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

The number of nonzero entries along each column represents the number of loops in which that branch
appears, while the number of nonzero entries in each row is equal to the number of branches in the
loop corresponding to that row.
The number of nonzero entries in each row is equal to the number of branches in the loop correspond-
ing to that row.

Construction of the fundamental loop matrix B

The fundamental loop matrix B f is a submatrix of BA that defines the minimum number of indepen-
dent loops. B is constructed based on trees and co-trees. It is constructed by selecting a tree from the
graph, then deducing the corresponding independent loops by taking the corresponding links one at a
time. The construction of B f proceeds by rearranging the columns so that the twigs appear first then
the links.
Each link results in a fundamental loop, as a result the number of fundamental loops is equal to the
number of links that is B − N + 1.
It happens that the co-tree submatrix Bc is square with one nonzero element in each row and column.
Since B is not unique and no restrictions are made on the choice of row and column order selection;
one may choose to make B appear in the normal fundamental form or simply the fundamental loop
matrix as follows:  
..
B = Bt . U

where Bc is reduced to the identity matrix U. This is done by starting the numbering of the funda-
mental loops from the links.
In the graph of Figure 1.7, choosing the branches 2,4 and 5 as a tree, will lead to branches 1,3 and
6 being the links. Then, the obtained fundamental loops (shown in Figure 1.8) are L1 : {1; 2; 5}, L2 :
{2; 3; 4}, L3 : {4; 6; 5} which results in the following reduced fundamental loop matrix B:
 
    1 0 1 1 0 0
B = Bt Bc = Bt U =  −1 −1 0 0 1 0 
0 1 −1 0 0 1

3
2 4 2 4 2 4 2 4
1 2 3 1 2 3 1 2 3 1 2 3

5 5 5 5
1 6
0 0 0 0
(a) Tree (b) L1 (c) L2 (d) L3

Figure 1.8: The fundamental loops

Kirchhoff’s current Law (KVL)

KVL states that the algebraic sum of the voltage drop is equal to the algebraic sum of the voltage rise
in any closed path [16]. In other words, the algebraic sum of branch voltages along any loop is zero.

13
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

If V is the vector of branch voltages, this yields:

BV = 0 (1.3.4)

By partitioning V for some tree and co-tree as:

 
Vt
V = · · ·
Vc

we get the following :  

Vt  
BV = · · · Bt U = Bt Vt +Vc = 0
Vc

Vc = −Bt Vt (1.3.5)

In the example of Figures 1.7 and 1.8, applying this rule would result in:
  
v1 = −v2 − v5
  
v1 −1 0 −1 v2 
Vc = v3  = −Bt Vt =  1 1 0  v4  ⇒ v3 = v2 + v4
v6 0 −1 1 v5

v6 = v4 − v5


Orthogonality between A and B

The matrices A and B are orthogonal which means:

ABT = 0 (1.3.6a)
BAT = 0 (1.3.6b)

and this is also true for the complete non-reduced versions of A and B, that is:

AA BTA = 0
BA ATA = 0

. .
If A is known and written in partitioned form: A = [ At .. Ac ] and B = [ Bt .. U ], applying equation 1.3.6a
and since At is invertible, we obtain:
 T
 Bt
AB = At Ac  · · ·  = At BtT + Ac = 0 ⇒ B = −(At−1 Ac )T U
T
  

Bt = −(At−1 Ac )T (1.3.8)

Similarly, if B is known, the equation 1.3.8 would result in:

A = At Ac = At −At BtT = At U −BtT

     

14
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

The last formula is not useful as we still need to determine At to find A.

In the graph of Figure 1.4, applying equations 1.3.8 will deduce B:
 
−1 −1 −1 −1 1 0 0 0
−0 −1 −1 −1 0 1 0 0 
B = −(At−1 Ac )T U = 
  

 −0 −1 −0 −0 0 0 1 0 
−0 −1 −1 −0 0 0 0 1

1.3.3 The fundamental cut set matrix D

The cut-set

The cut set is a minimal set of branches of a graph, removal of which cuts the graph into two parts.
It separates the nodes of the graph into two groups. The cut-set consists of only one tree branch and
remainders are links. Each branch of the cut-set has one of its terminal incident at a node in one
group and its other end at a node in the other group and its other end at a node in the other group. The
orientation of the cut-set is the same as the orientation of the tree branch.
The fundamental cut set matrix D is just the reduced incidence matrix written such that: At = U, so:
   
D = U Dc = U Ac ; DI = 0 ⇒ It = −Dc Ic

Construction of the fundamental cut-set matrix D

.
To write D as D = [U .. Dc ] , Dc must correspond to a co-tree. A tree should be selected first. Then D
is constructed with entries as follows and the reference direction is taken as the direction of the twig:

+1 if branch j belongs to cut set i and reference direction agree

di j = −1 if branch j belongs to cut set i and reference direction opposite

0 if branch j doesn’t belong to cut set i


In the graph of Figure 1.9(a), by selecting branches 1,4 and 5 as a tree, we obtain three fundamental
cut-sets:{1; 2; 6}, {2; 3; 4}, {2; 3; 5} which are shown in Figure 1.9(b).

C3

6 C2 6
2 2
1 2 C1 1 2

3 3
1 4 1 4

4 3 4 3
5 5
(a) Graph (b) Cut-sets

Figure 1.9: Graph and its cut-sets

15
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

The construction of the fundamental cut-set matrix C for the graph of Figure 1.9 is organized in
Table 1.3.
Table 1.3: Construction of D

Branches
1 2 3 4 5 6
f-cut-sets
C1 : {1; 2; 6} 1 1 0 0 0 -1
C2 : {2; 3; 4} 0 -1 -1 1 0 0
C3 : {2; 3; 5} 0 -1 -1 0 1 1

By rearranging the columns so that the twigs appear first, we get:

 
  1 0 0 1 0 −1
D = U Dc =  0 1 0 −1 −1 0 
0 0 1 −1 −1 1

Kirchhoff’s current and voltage Law

Since the fundamental cut set matrix D is just the reduced incidence matrix written such that: At = U,
then KCL could be applied using D as follows:

AI = DI = 0 (1.3.9a)
V = DT E = 0 (1.3.9b)

If I is partitioned for some tree and co-tree as:

 
It
I = . . .
Ic

Then:  
  It
DI = U Dc · · · = UIt + Dc Ic = 0 ⇒ It = −Dc Ic
Ic

In the graph of Figure 1.9, and by taking the fundamental cut set matrix obtained previously, we get:

  
i1 = −i2 + i6
  
i1 1 0 −1 i2 
It = i4  = −Dc Ic = − −1 −1 0  i3  ⇒ i4 = i2 + i3
i5 −1 −1 1 i6

i5 = i2 + i3 − i6


Orthogonality of B and D

The matrices D and B are orthogonal which means:

DBT = 0 (1.3.10a)
BDT = 0 (1.3.10b)

16
1.3. Matrix representation of a graph Chapter 1. Introduction to graph theory

Applying equation 1.3.10a to the partitioned forms of B and D would result in:
 T
 Bt
DBT = U Dc  · · ·  = BtT + Dc = 0


Dc = −BtT (1.3.11)

Using equation 1.3.8, we could conclude that:

Dc = (At−1 Ac ) (1.3.12)

Using equations 1.3.5 and 1.3.11 and if the twig voltages are known, all branch voltages can be
found if the fundamental cutset matrix is known as follows:
     
Vt U U
V =  · · ·  =  · · ·  Vt =  · · · 
−Bt Vc −Bt DTc

V = DT Vt (1.3.13)

Using equations 1.3.3 and 1.3.12 and if the link currents are known, all branch currents can be found
if the fundamental loop matrix is known as follows:
 −1     t
−At Ac Ic −Dc Bt
I=  ···  = · · · Ic = · · · Ic
  
Ic U U

I = BT Ic (1.3.14)

17
Chapter 2

Network analysis techniques

2.1 Generalized branch model in electrical networks

The first step in the formulation of the analysis of the electrical network is to represent the network
by a graph with nodes and branches.
The generalized branch model is used to determine the nodes and branches of the graph .

0
ik

+ +

vk

0 − ik
jk vk

− vKE
+
−

Figure 2.1: The generalized branch model

In the generalized branch of Figure 2.1:

• v0k : the branch voltage

• i0k : the branch current

• bk : the branch element

• bk : the branch element

• vk : the element voltage

• ik : the element current

18
2.2. Impedance Z and admittance matrices Y Chapter 2. Network analysis techniques

• vkE : an external voltage source

• jk : an external current source

Applying KVL and KCL on the branch of Figure 2.1, would result in the following:

v0k = vk − vkE

i0k = ik − jk

By generalizing the above equation to all network branches, we ge:

V 0 = V −VE (2.1.1a)
I0 = I − J (2.1.1b)

From equation 1.3.2, 1.3.4 and 1.3.9a , we get :

BV 0 = B(V −VE ) = 0
AI 0 = A(I − J) = 0
DI 0 = D(I − J) = 0

which would result in the set of equilibrium equations between the input and output variables:

BV = BVE (2.1.2a)
AI = AJ (2.1.2b)
DI = DJ (2.1.2c)

If Ic is the vector of link currents, Vt is the vector of twig voltages and E is the vector of node-to-
reference voltages, applying equation 1.3.1, 1.3.14 and 1.3.9b would result in:

I 0 = BT Ic ⇒ (I − J) = BT Ic
V 0 = DT Vt ⇒ (V −VE ) = DT Vt
V 0 = AT E ⇒ (V −VE ) = AT E

which would result in the following set of equations:

I = J + BT Ic (2.1.3a)
V = VE + DT Vt (2.1.3b)
V = VE + AT E (2.1.3c)

2.2 Impedance Z and admittance matrices Y

A representation of an electrical network using a graph is not enough to solve for any parameter
because the graph conveys only information about the network connection.[17]

19
2.2. Impedance Z and admittance matrices Y Chapter 2. Network analysis techniques

In order to represent and convey information (value and transfer functions ) about the network ele-
ments, the b × b impedance matrix Z is introduced with element as follows:
(
Diagonal elements zii = the impedance along branch i
zi j =
Off-diagonal elements zi j = the transimpedance of branch j on branch i

The admittance matrix Y of a network is the inverse of the impedance matrix Z ,that is:

Y = Z −1 (2.2.1)

The information about the external voltage and current sources is conveyed by the b vectors VE and J
as follows:

0 if the branch doesn’t contain any external voltage sources

vEi = +vEi if the external voltage sources contributes current to the branch current

−vEi if the external voltage sources opposes current to the branch current



0 if the branch doesn’t have an external current sources

ji = + ji if the external current sources contributes current to the branch current

− ji if the external current sources opposes current to the branch current


The matrices Y and Z could be used to obtain the network parameters based on Ohm’s law, that is:

V = ZI (2.2.2)
I = YV (2.2.3)

2.2.1 Example 1
The graph on Figure 2.2, could be extracted from the network of Figure 2.3.

2 4
1 2 3

3
1 5

Figure 2.2: The resulting graph from the network of Example 1

20
2.2. Impedance Z and admittance matrices Y Chapter 2. Network analysis techniques

R4

v4 (t) i3 (t) C1 v5 (t)

L1

−
+

+
−
R1 R2 R3
i1 (t) i2 (t)

−
+ +
v1 (t) − v2 (t) + v3 (t) −

Figure 2.3: Network of Example 1

The different matrices describing the network are:

1 

R1 0 0 0 0 0

R1 0 0 0 0 0 
−V s1
 
Is1

0 1
 0 L1 s 0 0 0 0  L1 s 0 0 0 0 −V s2   0 
1
     
0 0 R2 0 0 0 
0 0 0 0 0
  V s 3
  0 
Z =  ; Y = R2  ; VE = 
1 −V s4  ; J =  0 
  
0 0 0 0 0 0

0 0 C1 s 0 0
 C1 s     
0 0 0 0 R3 0 

0 1   V s5  −Is2 
0 0 0 R3 0
0 0 0 0 0 R4 0 0 0 0 0 1 0 −Is3
R4

2.2.2 Example 2
The graph on Figure 2.4, could be extracted from the network of Figure 2.5.

3 4
1 2 3

5
4
1 2
7
0

Figure 2.4: The resulting graph from the network of Example 2

21
2.2. Impedance Z and admittance matrices Y Chapter 2. Network analysis techniques

C2

Is2

L1 L3

R1 L2 (t)
Is1 R3 gm vC (t)

+ +
Vs − vC C1
−

Figure 2.5: Network of Example 2

The different matrices describing the network are:

1
   
R1 0 0 0 0 0 0 R1 0 0 0 0 0 0
 0 R2 0
 0 0 0 − gCm1Rs2 

0

1
R2 0 0 0 0 gm  
0 1
0 0 L s 0
 1 0 0 0    0 L1 s 0 0 0 0 
0 0 0 L3 s 0 0 0  ; Y = 1
Z= 0 0 0 0 0 0 
 
L3 s
0 0
 0 0 L2 , s 0 0  

0 0 0 0 1
0 0 

1 L2 s
0 0 0 0 0 0 
   
C2 s 0 0 0 0 0 C2 s 0 
1
0 0 0 0 0 0 C1 s 0 0 0 0 0 0 C1 s
 
−Vs  
 0  −Is1
   0 
 0   
   0 
 0   
 0  ; J = −Is2 
VE =    
   0 
 0   
   0 
 0 
0
0

22
2.3. Analysis Techniques Chapter 2. Network analysis techniques

2.3 Analysis Techniques

Different analysis techniques could be used to analyze electrical networks and each would result in
a different network parameter but nevertheless all the parameters are related to each other so once a
parameter is found the others could be derived using the network description.

2.3.1 Nodal analysis

From the set of equations , 2.1.2b , 2.1.3b and 2.2.3, we get:

AI(s) = AJ(s) ⇒ AY (s)V (s) = AJ(s)

AY (s) VE + AT E = AJ(s)


AY (s)AT E = A {J(s) −Y (s)VE }


(2.3.1)

By denoting the nodal current vector that follows through the nodes due to external current source as
JN (s) and the nodal admittance matrix as YN (s), we get the following:

YN (s) = AY (s)AT (2.3.2a)

JN (s) = A {J(s) −Y (s)VE } (2.3.2b)
YN (s)E(s) = JN (s) (2.3.2c)
E(s) = YN−1 (s)JN (s) (2.3.2d)

In the example of Figure 2.5, the resulting matrices are:

 
Vs

1 0 −1 0 0 −1 0 − Is1 
R
  1
1 −1 0  ; JN (s) =  −Is2 
0 0 1 0  
A=
0 1 0 −1 0 1 0 
 Is2 
 
0 0 0 0 1 0 −1 0
 
C s + R11 + L11 s − L11 s −C2 s 0
 2 
− L11 s 1
+ L12 s + L13 s − L13 s − L12 s
 
L1 s
 
YN (s) = 
 
−C2 s − L13 s C2 s + R12 + L13 s −gm

 
 
0 − L12 s 0 C1 s + L12 s

2.3.2 Loop Analysis

The loop analysis would output the loop the tree voltages and currents. From equations 2.1.2a and
2.2.2, we conclude that:
BV = BVE (s) ⇒ BZ(s)I(s) = BVE (s)

Using equation 2.1.3a and if B is partitioned for some tree, we get:

BVE (s) = BZ(s)I(s) = BZ(s){J(s) + BT Ic (s)} (2.3.3)

23
2.3. Analysis Techniques Chapter 2. Network analysis techniques

BZ(s)BT Ic (s) = B{VE (s) − Z(s)J(s)} (2.3.4)

By denoting the column vector of link currents as Ic (s), the co-tree impedance matrix as Zc (s), and
co-tree voltage vector as Ec we get the following:

Zc (s) = BZ(s)BT (2.3.5a)

Ec = B{VE (s) − Z(s)J(s)} (2.3.5b)
Ec (s) = Zc (s)Ic (s) (2.3.5c)
Ic (s) = Zc−1 (s)Ec (s) (2.3.5d)

Once the link currents are known, all branch currents can be found using equation 1.3.14.
Fom equations 2.3.4 and 2.3.5c, it is evident that to do a loop analysis of a certain network the
external sources vectors J and VE along with the impedance and admittance matrices Z and Y should
be reordered such that they maintain the same branch order as in the loop matrix B.
In the example of Figure 2.5, by selecting the tree {3; 4; 5; 7} as shown in Figure 2.6, the resulting
matrices and modified vectors are:
 
1 0 1 1 1 0 0
B= 0 1 −1 −1 0 1 0 
−1 −1 0 0 0 0 1

   1 
L1 s 0 0 0 0 0 0 L1 s 0 0 0 0 0 0
0 L3 s 0 0 0 0 0 1





 0 L3 s 0 0 0 0 0 

 0 0 L2 s 0 0 0 0  
0 0 1
0 0 0 0


1
  L2 s 
Zreordered = 
 0 0 0 C1 s 0 0 0  ; Yreordered = 
 
0 0 0 C1 s 0 0 0


 0 0 0 0 R1 0 0  
0 0 0 0 1
0 0

R1
   
gm R2
0 0 0 − C1 s 0 R2 0 1
   
   0 0 0 gm 0 R2 0 
1
0 0 0 0 0 0 C2 s 0 0 0 0 0 0 C2 s
   
0 0
0 −Is2 
     
0
 
 0 
  V s − Is1 R1
VE,reordered  0  ; Jreordered =  0  ; Ec =
=     Is2 L3 s 
Vs 
 
 Is1 
  −Is2 L3 s
0  0 
0 0

24
2.3. Analysis Techniques Chapter 2. Network analysis techniques

3 4
1 2 3

5
4
1 2
7
0

Figure 2.6: The selected tree of Example 2

2.3.3 Mesh analysis

In the loop formulation, care must be taken in the choice of the tree and the updating of the different
matrices with respect to this choice.
A different way to perform loop analysis is to consider currents flowing into meshes rather than
fundamental loops (hence the name mesh analysis). In order to do so, the current following into a
certain mesh is given an arbitrary direction and then the meshes matrix BM is constructed as follows:

+1 if branch j current contributes to the current following into i

bm i j = −1 if branch j current opposes to the current following into i

0 if branch j doesn’t belong to mesh i


Using the same derivation used to obtain equation 2.3.4, we could obtain the following:

BM Z(s)BTM IM (s) = BM {VE (s) − Z(s)J(s)} (2.3.6)

By denoting the column vector of mesh currents as IM (s), the mesh impedance matrix as ZM (s), and
mesh voltage vector as EM we get the following:
ZM (s) = BM Z(s)BTM (2.3.7a)
EM = BM {VE (s) − Z(s)J(s)} (2.3.7b)
EM (s) = ZM (s)IM (s) (2.3.7c)
−1
IM (s) = ZM (s)EM (s) (2.3.7d)

The meshes matrix BM has the branches in the same order of the graph as a result mesh analysis solves
the main problem arising from the loop analysis because the different network vectors and matrices
(J,VE , Z,Y ) don’t need to be updated.

2.3.4 Cutset analysis

From the set of equations , 2.1.2b and 2.2.3, we get:
DI(s) = DJ(s) ⇒ AY (s)V (s) = AJ(s)

25
2.3. Analysis Techniques Chapter 2. Network analysis techniques

By partitioning D for a certain tree with twig voltages Vt , we get:

DY (s) VE + DT Vt = DJ(s)


DY (s)DT Vt = D {J(s) −Y (s)VE }


(2.3.8)

By defining the tree current vector that follows through the a certain tree due to external current source
as as Jt (s) and the tree admittance matrix as Yt (s), we get the following:

Jt (s) = D {J(s) −Y (s)VE } (2.3.9a)

Yt (s) = DY (s)DT (2.3.9b)
Yt (s)Vt (s) = Jt (s) (2.3.9c)
Vt (s) = Yt−1 (s)Jt (s) (2.3.9d)

Once the twig voltages are found, all branch voltages can be found using equation 1.3.13.
Fom equations 2.3.9b and 2.3.9a, it is evident that to do a cutset analysis, the different network
vectors and matrices (J,VE ,Y ) should be reordered such that they maintain the same branch order as
in the cutset matrix D.
In the example of Figure 2.5, applying cutset analysis with the tree {3; 4; 5; 7} as shown in Figure 2.6,
would result in the following:
V 
s
− I
 
1 0 0 0 −1 0 1 R s1
 0 1 0 0 0 −1 1   1−I 
s2 
D=  0 0 1 0 −1 1 0  ; Jt (s) = 

Vs

 R1 − Is1 

0 0 0 1 −1 1 0 Vs
R1 − Is1
 
C s + R11 + L11 s C2 s 1 1
 2 R1 R1 
C2 s + R12 + L13 s − R12 − R12 − gm
 
 C2 s 
Yt (s) = 
 
1
− R12 1 1 1 1 1 
R1 R1 + R2 + L2 s R1 + R2 + gm
 
 
1
R1 − R12 1
R1 + R2
1
C1 s + R11 + R12 + gm

26
Chapter 3

Implementation

3.1 MatLab
The different analysis techniques described previously could be implemented using MatLab which
is a software package that is used for implementing, developing and analyzing algorithms and high-
performance technical computations. MATLAB is a matrix-based language hence it natively supports
vector and matrix operations that are fundamental to engineering and scientific problems.

3.2 Block diagram

The block diagram on Figure 3.1, explains the implementation of the project.

Obtain AA , A, B, D

netlist
Circuit Extract the graph Perform Analysis Results

Get Z, J,VE

Figure 3.1: Block diagram of the implementation

The circuit is provided by the user in the form of a netlist saved in a separate text file. After that, the
graph is extracted along with the different matrices that represent the graph and the network. Then,
they are fed into the analysis program to obtain the results.
Using a SPICE netlist as the link between the user and the simulation program enables the ability to
import different external SPICE netlists from other simulation packages along with the flexibility of
using the simulation program as a standalone simulator.

27
3.3. Simulation program Chapter 3. Implementation

The simulation program uses a variance on Pspice[18] netlist format where each line of the text file
represents an element defined as follows:

XXXX N1 N2 Val S1 S2

where each entry is defined as explained below.

• XXXX: the element and the first letter determines the element type.

• N1: the net connected to the first pin of the element.

• N2: the net connected to the second pin of the element

• Val: the value of the element.

• S1: the net connected to the first sense terminal of the element (only if it’s a controlled source).

• S2: the net connected to the second sense terminal of the element (only if it’s a controlled
source).

3.3 Simulation program

We shall explain some functionality of the simulation program

3.3.1 Get AA , A, B, D functions

The flowchart in Figure 3.2, explains how the different matrices representing the graph are obtained.
The circuit is transferred from a netlist to a structure named circuit containing the element branches
and nets which represent the elements and nets of the network. Then, the internal nets are detected
and replaced with the corresponding ones from the network. After that, the different matrices are
obtained along with a tree.

3.3.2 Get tree function

The flowchart in Figure 3.3, explains how the tree is obtained from the graph.

28
3.3. Simulation program Chapter 3. Implementation

Netlist

circuit.branches()
circuit.nets()

Find Internal nodes

Replace Internal nodes

Build graph

Extract tree Extract AA and A

Construct B, D

Figure 3.2: The get AA , A, B, D function

circuit.branches()
circuit.nets

Select (N − 1) arbitrary branches

No
Is tree ?

Yes

Extract cotree

Figure 3.3: The get tree function

29
3.4. Examples Chapter 3. Implementation

Firstly a number of (N − 1) is selected randomly from all branches that represent the graph then, they
are tested using the isdiag matlab function which tells whether a graph is directed acyclic graph (i,e
is it a tree or not). When the selected branches don’t form a tree another set of branches is selected
randomly. When a tree is obtained the corresponding co-tree branches are extracted.

3.4 Examples

3.4.1 Example 1
In this example, we shall demonstrate the nodal analysis method. The circuit in Figure 3.4 is imple-
mented by the netlist given in Listing 3.1. The results are shown on Listing 3.2. Using the probe
functionality of Multisim, we got the results shown on Figure 3.5.

C1 4 2 10 e−09
C2 4 3 3e−09
C3 5 0 1e−09
I1 3 4 1e−03
L1 3 5 30e−3
L2 5 4 10e−3
R1 3 1 5 e03
R2 4 0 1 e03
V1 1 0 10
V2 2 0 6
Listing 3.1: Circuit netlist

30
3.4. Examples Chapter 3. Implementation

I1
3 4

1mA
10kHz
0°

C2

3nF

L1 5 L2

30mH 10mH
R1
5kΩ
C1
1 10nF

2
V1
C3 R2
10Vpk V2
1nF 1kΩ
10kHz 6Vpk
0° 10kHz
0 0°

Figure 3.4: Example 1 circuit on Multisim

V_N3 = 3 . 6 2 2 3 Vrms
V_N4 = 2 . 0 1 2 Vrms
V_N5 = 2 . 4 7 6 9 Vrms
Listing 3.2: Nodal analysis output

31
3.4. Examples Chapter 3. Implementation

I1
3 4

1mA
10kHz
0°

C2

3nF

V L1 5 V L2 V V
PR1 PR2 PR3 PR4
30mH 10mH V(p-p): 6.03 V
V(rms): 2.08 V
R1 V(freq): 10.0 kHz
V(p-p): 8.05 V V(p-p): 6.07 V
5kΩ V(p-p): 10.4 V V(rms): 2.91 V V(rms): 2.08 V
V(freq): 21.5 kHz V(freq): 10.0 kHz
V(rms): 3.67 V C1
V(freq): 10.0 kHz
1 10nF

2
V1
C3 R2
10Vpk V2
1nF 1kΩ
10kHz 6Vpk
0° 10kHz
0 0°

Figure 3.5: Example 1 circuit on Multisim with the probes results

3.4.2 Example 2
In this example, we shall demonstrate the loop analysis method. The circuit in Figure 3.6 is imple-
mented by the netlist given in Listing 3.3. The results are shown on Listing 3.4. Using the probe
functionality of Multisim, we got the results shown on Figure 3.7.

C1 4 2 10 e−09
C2 4 3 3e−09
C3 5 4 1e−09
G1 4 0 5 5 0
I1 3 4 1e−03
L1 3 5 90e−3
L2 0 5 10e−3
R1 3 1 5 e03
R2 4 0 1303
V1 1 0 10
V2 2 0 6
Listing 3.3: Circuit netlist

32
3.4. Examples Chapter 3. Implementation

I1
3

1mA
10kHz
0°
C2

3nF

L1 C3

90mH 1nF
5

R1 C1
5kΩ 4 10nF

1 2
I2
L2 R2
5Mho
V1 10mH 1kΩ V2
10Vpk 0 6Vpk
10kHz 10kHz
0° 0°

Figure 3.6: Example 2 circuit on Multisim

I_branch_L1 = 0 . 0 0 0 2 2 4 5 6 Arms
I_branch_C2 = 0 . 0 0 1 1 2 3 5 Arms
I_branch_L2 = 8 . 2 9 6 e−07 Arms
I_branch_C1 = 0 . 0 0 2 8 0 1 1 Arms
I_branch_C3 = 0 . 0 0 0 2 2 4 5 5 Arms
I_branch_R1 = 0 . 0 0 1 3 4 3 Arms
I_branch_R2 = 0 . 0 0 3 7 8 3 6 Arms
Listing 3.4: loop analysis output

33
3.4. Examples Chapter 3. Implementation

I1
3

I(p-p): 1.23 mA
1mA I(rms): 434 uA
I(p-p): 636 uA I(freq): 10.0 kHz
I(p-p): 635 uA
I(rms): 225 uA 10kHz
I(rms): 225 uA
I(freq): 10.0 kHz I(freq): 10.0 kHz 0°
C2 A
PR1

3nF

A L1 C3
PR3
90mH A PR2
1nF I(p-p): 7.93 mA
A I(rms): 2.81 mA
PR5 5 I(freq): 10.0 kHz A
PR7
R1 I(p-p): 3.80 mA C1
I(rms): 1.34 mA
5kΩ I(freq): 10.0 kHz 4 10nF

1 2
G1
L2 R2
5Mho
V1 I(p-p): 2.44 uA 10mH 1kΩ I(p-p): 10.1 mA V2
I(rms): 866 nA
10Vpk I(freq): 10.0 kHz I(rms): 3.57 mA 6Vpk
0 I(freq): 10.0 kHz
10kHz 10kHz
0° A A 0°
PR4 PR6

Figure 3.7: Example 2 circuit on Multisim with the probes results

3.4.3 Example 3
In this example, we shall demonstrate the cutset analysis method. The circuit in Figure 3.8 is im-
plemented by the netlist given in Listing 3.5. The results are shown on Listing 3.6. Using the probe
functionality of Multisim, we got the results shown on Figure 3.9.

C1 5 0 10 e−09
C2 4 2 3e−09
C3 5 4 1e−09
G1 4 0 2 5 0
I1 3 4 1e−03
L1 3 4 90e−3
L2 3 5 10e−3
R1 3 1 5 e03
R2 4 0 1 e03
V1 1 0 10
V2 2 0 6
Listing 3.5: Circuit netlist

34
3.4. Examples Chapter 3. Implementation

I1
3 4

1mA
10kHz
0°

L1

90mH

L2 5 C3

10mH 1nF

R1 C2
5kΩ 3nF

1 2
G1
R2
C1 2Mho
V1 1kΩ V2
10nF
10Vpk 6Vpk
10kHz 10kHz
0° 0 0°

Figure 3.8: Example 3 circuit on Multisim

V_L2 = 0 . 3 5 8 7 0 Vrms
V_L1 = 0 . 3 7 0 5 2 Vrms
V_C1 = 1 . 1 4 9 2 0 Vrms
V_C2 = 0 . 7 5 2 3 5 Vrms
V_C3 = 0 . 0 6 0 1 4 Vrms
V_R1 = 6 . 6 7 4 0 Vrms
V_R2 = 1 . 2 8 5 2 Vrms
Listing 3.6: cutset analysis output

35
3.4. Examples Chapter 3. Implementation

I1
3 4

1mA V(p-p): 1.01 V

V(rms): 361 mV
10kHz V(freq): 10.0 kHz
V(p-p): 1.08 V 0°
V(rms): 382 mV
V(freq): 10.0 kHz PR4
V L1 V
Ref4 Ref4

90mH

PR1 C3 PR3
Ref1 V L2 5 V Ref3
Ref1
10mH V 1uF
PR2
V Ref2 V(p-p): 3.65 V
V(rms): 1.29 V
R1 V(freq): 10.0 kHz C2
5kΩ V(p-p): 18.9 V 10uF
V(rms): 6.69 V
V(freq): 10.0 kHz

1 2
V G1
Ref2 R2
C1 2Mho
V1 1kΩ V2
10nF
10Vpk 6Vpk
10kHz 10kHz
0° 0 0°
V
Ref3

Figure 3.9: Example 3 circuit on Multisim with the probes results

36
Conclusion

Numerical simulation of electrical networks are a valuable tool in circuit analysis and design. Electri-
cal engineers should gain expertise in the underlying work of circuit simulators, because without an
accurate understanding of the equations that the programs use to simulate the networks, it is impossi-
ble to fully interpret obtained computational results properly.
The graph theory approach for network and circuit analysis provides an easy way to explore the design
of circuit simulators. Based on the different network analysis techniques, different parameters could
be extracted about the simulated network and the other parameters could be easily computed based
on the obtained ones as explored in Chapter 1 and 2.
The nodal analysis method is the easiest and most fast method to apply since it could be applied
directly without the need to extract a tree or any graph matrix rather than the incidence matrix.
The developed program did show approximately the same results provided by Multisim for the differ-
ent networks examined on Chapter 3. The difference in the results are a consequence of the slightly
different models of each element, because the developed program uses ideal models for each elements
whereas Multisim take into consideration different parasitics present on each element and their tem-
perature dependency along with their proper behavioral conditions(break down voltage, maximum
dissipated power, . . . ).
This project is still a "work under progress" where the following notes could be improved:

• Optimizing the circuit representation during the program execution by utilizing spares matrices.

• Cleaning the code using advanced data structures and object oriented programming tools.

• Extending the program to handle the non-linear components by introducing separate model
based simulation.

• Developing a GUI (graphical user interface) to ease the use of the program through the MatLab
AppDesigner[19], and making it a standalone application.

37
References

[1] J. A. Bondy and U. S. R. Murty, Graph theory with applications / J. A. Bondy and U. S. R.
Murty. Macmillan London, 1976, x, 264 p. : ISBN: 0333177916.
[2] N. Bhagya Jyoti, “Applications of graph theory in different branches of science,” International
Journal of Mathematics And its Applications, vol. 5, pp. 57–60, 3-A 2017.
[3] P. Anandhan, K. Uma, and J. Anuradha, “An overview of application of graph theory,” vol. 9,
pp. 242–248, Jan. 2016.
[4] NI, Multisim, version 14.2. [Online]. Available: https://www.ni.com/multisim.
[5] J. M. Rabaey, The spice page, http://bwrcs.eecs.berkeley.edu/Classes/IcBook/
SPICE/.
[6] P. Sekalski, “Electronic technology design & workshop,” Notes, Lodz University of Technol-
ogy, 2007, [Online]. Available: https://lux.dmcs.p.lodz.pl/etdw.
[7] A. Klos, Mathematical Models of Electrical Network Systems - Theory and Applications- An
Introduction, ser. Lecture Notes in Electrical Engineering. Springer, 2017, vol. 412, ISBN: 978-
3-319-52176-3. DOI: 10.1007/978-3-319-52178-7.
[8] R. J. Wilson, Introduction to graph theory, 4th. Prentice Hall, 1996, ISBN: 0582249937.
[9] J. L. Gross and J. Yellen, Graph Theory and Its Applications, 2nd. Chapman amp; Hall/CRC,
2005, ISBN: 158488505X.
[10] Reducible. (Jun. 2020). Introduction to graph theory: A computer science perspective, [Online].
Available: https://www.youtube.com/watch?v=LFKZLXVO-Dg.
[11] B. Xie, C. Qi, H. Ben, and W. Yu, “The applications of graph theory in electric network,”
in 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC),
2019, pp. 780–784. DOI: 10.1109/SDPC.2019.00148.
[12] F. Dörfler, J. W. Simpson-Porco, and F. Bullo, “Electrical networks and algebraic graph theory:
Models, properties, and applications,” Proceedings of the IEEE, vol. 106, no. 5, pp. 977–1005,
2018. DOI: 10.1109/JPROC.2018.2821924.
[13] M. Arshad, Network Analysis and Synthesis, 1st. Laxmi Publications Pvt. Ltd, 2006.
[14] Pr A. Recioui, “EE446: Network analysis and synthesis,” Lecture Notes, IGEE, Apr. 2021.
[15] E. R. Gansner and S. C. North, “An open graph visualization system and its applications to
software engineering,” Software: Practice and Experience, vol. 30, no. 11, pp. 1203–1233,
2000.
[16] W. Hayt, J. Kemmerly, and S. Durbin, Engineering Circuit Analysis, 7th. USA: McGraw-Hill,
Inc., 2007, ISBN: 0073366617.
[17] Dr. Boonying Charoen, “182304: Electrical network analysis notes,” KKU, 2018, [Online].
Available: http://eestaff.kku.ac.th/~jamebond/182304/.

38
References References

[18] OrCad/Cadence, Pspice. [Online]. Available: https://www.pspice.com/about.

[19] MathWorks, Matlab app designer, https://www.mathworks.com/products/matlab/app-
designer.html.

39