Mathematical Modelling of Engineering Problems

Vol. 11, No. 1, January, 2024, pp. 18-26

Journal homepage: http://iieta.org/journals/mmep

Power Flow Analysis Using Numerical Computational Methods on a Standard IEEE 9-Bus
Test System
Adedeji Tomide Akindadelo1,2 , Folorunsho A. Shodiya1 , Ayodeji Olalekan Salau3,4* ,
Olawale Joshua Olaluyi5 , Jeremiah Oluwatosin Bandele6 , Sepiribo Lucky Braide7
1
Department of Basic Sciences, Babcock University, Ilishan-Remo 121103, Ogun State, Nigeria
2
Department of Science, The Sheffield College, Sheffield S22RL, United Kingdom
3
Department of Electrical/Electronics and Computer Engineering, Afe Babalola University, Ado-Ekiti 360001, Nigeria
4
Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Tamil Nadu 600124, India
5
Department of Electrical and Electronics Engineering, Bamidele Olumilua University of Science, Education, and Technology,
Ikere-Ekiti 250, Nigeria
6
Department of Electrical/Electronic Engineering, School of Science and Technology, Pan-Atlantic University, Ibeju-Lekki
Lagos 73688, Nigeria
7
Department of Electrical and Electronics Engineering, Rivers State University, Port Harcourt 5080, Nigeria

Corresponding Author Email: [email protected]

Copyright: ©2024 The authors. This article is published by IIETA and is licensed under the CC BY 4.0 license
(http://creativecommons.org/licenses/by/4.0/).

https://doi.org/10.18280/mmep.110102 ABSTRACT

Received: 5 May 2023 Load flow is an important tool for studying, designing, and analyzing power systems.
Revised: 22 August 2023 It allows power system engineers to determine whether the operation and configuration
Accepted: 10 September 2023 of the power system is safe under varying loading conditions. It is necessary to model
Available online: 30 January 2024 and simulate such a system in order to determine the power flow and losses. This
research paper focuses on using numerical methods such as Newton Raphson and Gauss
Seidel power flow equations for load flow analysis to calculate bus voltage magnitudes,
Keywords: phase angles, real and reactive power of each bus of an IEEE 9-bus test system. Newton
power flow, bus, load flow analysis. loss Raphson’s computation offers fast, accurate convergence but demands complex
minimization, gauss-seidel, Newton-Raphson, implementation, whereas Gauss Siedel is simpler but converges slower with lower
voltage magnitude, phase angle, real power, accuracy. The analysis was carried out using a MATLAB program. By manipulating
reactive power, convergence variables such as power injections, voltage magnitudes, and phase angles, it solves
nonlinear equations iteratively to establish stable operating points which aids in
enhancing power system analysis. The line losses for the two methods are compared
and the system's total load and generation power are also displayed. The consideration
of line losses and assessment of total load generation is crucial for maintaining system
efficiency, reliability and preventing voltage instability and equipment damage. The
results are also used to generate a directed graph which shows the interconnected nature
of the power system, aiding engineers in understanding power flow paths, identifying
potential issues, and making informed decisions about system operations. The Newton
Raphson method yields the lowest loss, with 4.585MW and 10.789Mvar. In
comparison, the Gauss Seidel method achieved 4.809MW and 10.798Mvar.

1. INTRODUCTION the system's components, ensuring that each generator

operates at its optimal performing level that consumer’s needs
Power flow analysis, also known as load flow analysis, is are achieved without overloading the facilities, and that plans
the foundation of voltage-current system analysis and design. for maintenance can be carried out without jeopardizing the
The study of power flow analysis is essential to understanding system's network. Load flow investigations also provide a
problems in voltage-current system operation and distribution. mathematical methodology for calculating different bus
It is also a fundamental technique in electrical engineering that voltages, phase angles, real power (A) and reactive power flow
enables the planning, operation, and optimization of power (B) via various nodes under steady-state conditions [1]. It also
systems by determining steady-state conditions, ensuring aids in determining the impact of a single generating station or
reliability, efficiency, and the integration of renewable sources, transmission path failure on the load system. Moreover, it
and facilitating informed decision-making for grid stability provides a balanced steady-state operation state of the load
and resource utilization. The goal of such an analysis is to system while ignoring a non-steady process system. This
determine how safely the system can work, i.e., whether there means that any load flow problem's mathematical formulation
are installation overloads or extremely high or extremely low is a nonlinear algebraic equation system without differential
voltages nodes. It also provides more information about all of equations. As a result, using prepared algorithms and

18
programs for load flow analysis is critical for performing flow analysis. Voltage on the load bus can vary within certain
recent load system analysis. boundaries, say 5%. The reason why the bus voltage is
Over the years, different methods have been used for insignificant and therefore not mentioned. This bus is in
calculating load flow. The use of these methods is primarily charge of distributing consumer power.
driven by the fundamental requirements of power flow
calculation, which include its iteration properties, computing Table 1. BUS type
accuracy and storage requirements, as well as its convenience
and flexibility of implementation [2-5]. Type Known Parameter
Unknown
The power flow problem is a system of nonlinear algebraic Parameter
equations that must be solved mathematically. Its answer will Load BUS A, B (Real and Reactive Power) |V|, ∠
almost always require some iteration. As a result, the most Generator BUS |V|, A B, ∠
crucial condition for a load flow calculation method is reliable Swing BUS |V|, ∠ B, A
convergence. The dimension of load flow equations grows
increasingly large as the size of the load system expands. Thus, 2.2 Voltage controlled or generator bus
the power flow analysis is becoming increasingly important
for equations with such high dimensions. Swing or Reference Bus are other names for Slack Bus. This
In this work, power flow analysis using the Newton in reality does not come to play but it is believed to account
Raphson method and Gauss-Seidel on an IEEE standard 9 bus for losses that arises when transmitting power. In the power
test system is compared. The 9-bus test system is a simplified system, real power is specified for only two buses: the Load
model of an electrical power network used for analyzing Bus and the Generator Bus. Because the real power delivered
power system techniques. It comprises 9 buses, including a by Generator Bus differs from the real power consumed by
slack bus with known values, load buses with demand, a Load Bus, the difference gives the power loss. This loss is then
generator bus with both load and generation, and branches calculated after the load flow problem has been solved. To
representing connections between buses. Loads and compensate for the loss, an additional generator bus is
generation are specified at certain buses, and the system is considered, with |V|, and phase angle specified, and real and
used to study power flow, voltage profiles, and stability under reactive power to be solved. Some additional buses that are
various conditions. considered in performing load flow analysis.
In the realm of power system analysis, methods for solving
the power flow equations play a pivotal role in ensuring the 2.3 Isolated (or dummy) bus
efficient and reliable operation of electrical networks. Two
widely employed techniques for solving these equations are An isolated bus is used to represent an unconnected or
the Newton-Raphson and Gauss-Seidel methods. This study isolated portion of the power system. It is used when there are
aims to compare and evaluate the performance of these areas in the network where there are no direct connections to
methods in the context of power flow analysis. By other buses.
systematically assessing their convergence characteristics,
computational efficiency, and accuracy, we seek to gain 2.4 Slackless (or zero injection) bus
insights into their respective strengths and limitations. This
study compares the Newton-Raphson and Gauss-Seidel In certain situations, such as load flow analysis with
methods for power flow analysis on the IEEE 9-bus test constant power loads, a slackless bus can be used to represent
system, focusing on convergence, accuracy, and efficiency. the system's behavior more accurately. This type of bus helps
The simulation setup involves replicating the system, avoid some of the inaccuracies that can arise when using a
implementing methods with defined criteria, and running traditional slack bus. The Bus data table for the IEEE 3-bus
analyses. Through this analysis, we aim to provide valuable system is presented in Table 2.
guidance for selecting the most suitable method based on
specific application scenarios, contributing to the Table 2. Bus data table for the IEEE 3-bus system
advancement of power system analysis techniques and the
optimization of electrical grid operation. Voltage Voltage Active Reactive
Bus Bus
Magnitude Angle Power Power
Number Type
(pu) (degrees) (MW) (MVAR)
1 Swing 1.0 0 Adjustable Adjustable
2. BUS DEFINITIONS 2 PV 1.05 Adjustable 1.0 Adjustable
3 PQ Adjustable Adjustable 0.8 0.4
The node that connects more lines, more loads, and more
generators is termed BUS. In a load system, every bus is The IEEE 3-bus system's bus data table outlines essential
accompanied with the following parameters: |𝑉| , voltage characteristics for each of the three buses. It includes bus
phase angle (<), real power (A), and reactive power (B) in numbers, types (swing, PV, PQ), voltage magnitudes, angles,
which two parameters are known and the other must be and power injections/consumptions (active and reactive). The
determined by solving equation [3]. The buses are categorize swing bus has adjustable power injections for stability, the PV
based on the parameters specified, as shown in Table 1. bus maintains a specified active power and adjusts reactive
power, and the PQ bus adjusts voltage to balance active and
2.1 Load bus (AB) reactive power consumption. This data forms the foundation
for load flow analysis, a process to calculate voltages and
The real power A and reactive power B of the load bus are ensure power system stability.
specified, while the |V| and phase angle(<) of the voltage are In the Newton-Raphson method, initial voltage guesses are
not specified [6, 7]. The bus voltage is determined using power iteratively adjusted using the Jacobian matrix and mismatch

19
equations to achieve power balance and stability. The Gauss- Bus i is included in the above equation as j. When we
Seidel method updates voltage estimates sequentially, express this equation in polar form, we get:
considering neighboring buses' values, until convergence is
𝑛
reached. Both methods ensure that bus voltage magnitudes and
angles meet power balance requirements based on the given 𝐼𝑖 = ∑|𝑋𝑖𝑗 | |𝑉𝑗 |∠𝜃𝑖𝑗 + 𝛿𝑗 (7)
bus data in the system. 𝑗=0

Then complex load at bus i is:

3. METHOD OF POWER FLOW ANALYSIS
𝐴𝑖 − 𝑗𝐵𝑖 = 𝑉𝑖∗𝐼𝑖 (8)
Understanding the analysis in the solution of nonlinear
algebraic simultaneous equation serves as the foundation in Substituting for Ii in Eq. (7) from Eq. (8), we have
solving nonlinear equations in digital power system flow
𝑛
analysis [3]. The main idea in power flow study is to get the
X-bus admittance matrix using the transmission and input data 𝐴𝑖 − 𝑗𝐵𝑖 = |𝑉𝑖 |∠ − 𝛿𝑖 ∑|𝑋𝑖𝑗 | |𝑉𝑗 |∠𝜃𝑖𝑗 + 𝛿𝑗 (9)
transformer. The system’s equation for a power with an X bus 𝑗=1
is:
Separating the real and imaginary parts,
𝐼 = 𝑋BUS 𝑉 (1)
𝑛

In a generalized pattern for n number of bus: 𝐴𝑖 = ∑|𝑉𝑖 ||𝑉𝑗 ||𝑋𝑖𝑗 | cos(𝜃𝑖𝑗 − 𝛿𝑖 + 𝛿𝑗 ) (10)
𝑗=1
𝐼𝑖 = 𝑉𝑖 ∑𝑛𝑗=0 𝑋𝑖𝑗 − ∑𝑛𝑗=1 𝑋𝑖𝑗 𝑉𝑗 For i=1, 2, 3… n (2) 𝑛

At BUS i, real and reactive power is: 𝐵𝑖 = ∑|𝑉𝑖 ||𝑉𝑗 ||𝑋𝑖𝑗 | sin(𝜃𝑖𝑗 − 𝛿𝑖 + 𝛿𝑗 ) (11)
𝑗=1

𝐴𝑖 + 𝐵𝑖 = 𝐼𝑖 ∗ 𝑉𝑖 (3)
Eqs. (10) and (11) form a set of nonlinear algebraic
Or equations in terms of |V| per unit and in radians. For each load
bus, use the two equations from Eqs. (10) and (11). Also, the
𝐴i +𝐵i controlled-voltage bus is solved by Eq. (10). The following set
=Ii (4)
V∗i of linear equations is derived by enlarging Eqs. (10) and (11)
in Taylor's series about the initial guess and leaving out all
Substituting for I in terms of Ai & Bi in the equation gives: terms in higher order.
𝑛 𝑛 (𝑘)
𝐴i + 𝐵i ∆𝐴2
= 𝑉𝑖 ∑ 𝑋𝑖𝑗 − ∑ 𝑋𝑖𝑗 𝑉𝑗 (5) .
Vi∗ .
𝑗=0 𝑗=1 (𝑘)
∆𝐴𝑛
(𝑘)
The mathematical formulation of the load flow problem, ∆𝐵2
.
derived from the above equation, yields a set of non-linear .
algebraic equations that must be solved iteratively. As a result, (𝑘)
[∆𝐵𝑛 ]
the Newton Raphson and Gauss Seidel solution methods must 𝜕𝐴2
(𝑘)
𝜕𝐴2
(𝑘)
𝜕𝐴2
(𝑘)
𝜕𝐴2
(𝑘)

be reviewed. 𝜕𝛿2
⋯
𝜕𝛿𝑛 𝜕|𝑉2 |
⋯
𝜕|𝑉𝑛 | (𝑘) (12)
∆𝛿2
⋮ ⋱ ⋮ ⋮ ⋱ ⋮ .
(𝑘) (𝑘) (𝑘) (𝑘)
3.1 Newton-Raphson method 𝜕𝐴𝑛 𝜕𝐴𝑛 𝜕𝐴𝑛 𝜕𝐴2 .
⋯ ⋯ (𝑘)
( 𝜕𝛿2 𝜕𝛿𝑛 ) ( 𝜕|𝑉2 | 𝜕|𝑉𝑛 |) ∆𝛿𝑛
=
The method described above was named after Isaac Newton 𝜕𝐵2
(𝑘)
𝜕𝐵2
(𝑘)
𝜕𝐵2
(𝑘)
𝜕𝐵2
(𝑘) (𝑘)
∆|𝑉2 |
⋯ ⋯
and Joseph Raphson. This numerical method can be traced 𝜕𝛿2 𝜕𝛿𝑛 𝜕|𝑉2 | 𝜕|𝑉𝑛 | .
back to the 1960s [4]. Taylor's series is used in this method to ⋮ ⋱ ⋮ ⋮ ⋱ ⋮ .
(𝑘) (𝑘) (𝑘) (𝑘) (𝑘)
approximate a series of nonlinear algebraic equations to a 𝜕𝐵𝑛 𝜕𝐵𝑛 𝜕𝐵2 𝜕𝐵𝑛 [∆|𝑉𝑛 |]
⋯ ⋯
series of linear algebraic equations. This method has a more [( 𝜕𝛿2 𝜕𝛿𝑛 ) ( 𝜕|𝑉2 | 𝜕|𝑉𝑛 | )]
powerful convergence characteristic than any other alternative
process and has also proven reliable because, unlike any other Because the swing bus variable |𝑉| and < are already
iterative process, it can solve a case of divergence [6]. The known, they are omitted from Eq. (12). After expressing the
number of iterations needed to reach a solution is unaffected partial derivatives of Eqs. (10) and (11) that give a linearized
by the size of the system, making it more robust and relationship between small changes in |𝑉| and <, the element
completely efficient. of the Jacobian matrix is derived. In matrix form, the equation
Eq. (1) gives the current flowing into the bus I for a notable is as follows:
bus system. Thus, the bus admittance matrix, is as follows:
∆𝐴 𝐽 𝐽2 ∆𝛿
[ ]=[1 ][ ] (13)
𝑛 ∆𝐵 𝐽3 𝐽4 ∆|𝑉|
𝐼𝑖 = ∑ 𝑋𝑖𝑗 𝑉𝑗 (6)
𝑗=1
where, J1 … J4 are the elements of the matrix.

20
 (𝑘) (𝑘)
The terms ∆𝐴𝑖 and ∆𝐵𝑖 are the difference between the Kirchhoff's Current Law suggested that current entering the
scheduled and calculated values. This is called the power bus will be positive. Thus, for buses where real and reactive
residuals, given by powers are inserted into the bus, such as generator buses, 𝐴𝑖𝑠𝑐ℎ
and 𝐵𝑖𝑠𝑐ℎ are positive. For load buses where real and reactive
∆𝐴𝑖
(𝑘)
= 𝐴𝑖𝑠𝑐ℎ − 𝐴𝑖
(𝑘)
(14) powers are flowing out from the bus, 𝐴𝑖𝑠𝑐ℎ and 𝐵𝑖𝑠𝑐ℎ are
negative. Solving the power flow equation in (5) for 𝐴𝑖 and 𝐵𝑖 ,
(𝑘) (𝑘) we have:
∆𝐵𝑖 = 𝐵𝑖𝑠𝑐ℎ − 𝐵𝑖 (15)
𝑛 𝑛
The new estimates for bus voltages are: 𝐴𝑖
(𝑘+1)
= ℜ {𝑉𝑖∗
(𝑘)
[𝑉𝑖
(𝑘)
∑ 𝑋𝑖𝑗 − ∑ 𝑋𝑖𝑗 𝑉𝑗
(𝑘)
]} 𝑗 ≠ 𝑖 (21)
𝑗=0 𝑗=1
(𝑘) (𝑘)
𝛿 (𝑘+1) = 𝛿𝑖 + ∆𝛿𝑖 (16)
𝑛 𝑛
(𝑘+1) (𝑘) (𝑘) (𝑘+1) (𝑘) (𝑘) (𝑘)
|𝑉𝑖 | = |𝑉𝑖 | + ∆|𝑉𝑖 | (17) 𝐵𝑖 = −ℑ {𝑉𝑖∗ [𝑉𝑖 ∑ 𝑋𝑖𝑗 − ∑ 𝑋𝑖𝑗 𝑉𝑗 ]} 𝑗 ≠ 𝑖 (22)
𝑗=0 𝑗=1

The proposed Newton-Raphson method procedure for

power flow is presented as follows: The power flow expression is typically expressed in terms
(1) For load buses, where 𝐴𝑖𝑠𝑐ℎ and 𝐵𝑖𝑠𝑐ℎ are known, of the bus admittance matrix elements. Since the off-diagonal
magnitudes of the voltage and phase angles are set the same as elements of the bus admittance matrix, 𝑋𝑏𝑢𝑠 , shown by
(0)
the swing bus values, or 1.0 and 0.0, i.e., |𝑉𝑖0 | = 1.0 and 𝛿𝑖 = uppercase letters, are 𝑋𝑖𝑗 =−𝑋𝑖𝑗 , and the diagonal 𝑋𝑖𝑖 = ∑ 𝑋𝑖𝑗 ,
Eq. (20) gives
0.0. For voltage-regulated buses, where |𝑉𝑖 | and 𝑃𝑖𝑠𝑐ℎ are
known, phase angles are set the same as the swing bus angle,
(0) 𝐴𝑖𝑠𝑐ℎ − 𝑗𝐵𝑖𝑠𝑐ℎ (𝑘)
or 0 i.e., 𝛿𝑖 = 0. + ∑𝑗≠𝑖 𝑋𝑖𝑗 𝑉𝑗
(𝑘+1) 𝑉𝑖∗
(𝑘) (𝑘)
(2) For load buses, 𝐴𝑖 and 𝐵𝑖 are solved by (10) and (11). 𝑉𝑖 = (23)
𝑋𝑖𝑖
(𝑘) (𝑘)
While ∆𝐴𝑖 and ∆𝐵𝑖 are solved by (14) and (15) 𝑗≠𝑖
respectively.
(𝑘) (𝑘) 𝑛
(3) Controlled-voltage buses, 𝐴𝑖 and ∆𝐴𝑖 are solved (𝑘+1)
𝐴𝑖 = ℜ {𝑉𝑖∗
(𝑘)
[𝑉𝑖
(𝑘)
𝑋𝑖𝑖 + ∑ 𝑋𝑖𝑗 𝑉𝑖
(𝑘)
]}
using Eqs. (10) and (11) respectively. 𝑗=1 (24)
(4) Elements of the Jacobian matrix (J1, J2, J3, and J4) are 𝑗≠𝑖
solved.
(5) The linear simultaneous equation in Eq. (13) is 𝑛
(𝑘+1) (𝑘) (𝑘) (𝑘)
calculated directly by optimally ordered triangular 𝐵𝑖 = −ℑ {𝑉𝑖∗ [𝑉𝑖 𝑋𝑖𝑖 + ∑ 𝑋𝑖𝑗 𝑉𝑗 ]}
factorization and Gaussian elimination. (25)
𝑗=1
(6) The new |𝑉 (𝑘+1) | and phase angles are calculated using 𝑗≠𝑖
Eqs. (16) and (17).
(𝑘)
(7) The procedure is repeated until the residuals ∆𝐴𝑖 and
(𝑘)
∆𝐵𝑖 are less than the specified accuracy, i.e., 4. RESULTS AND DISCUSSION

(𝑘) 4.1 Results

|∆𝐴𝑖 | ≤ 𝜖 (18)
The IEEE library is used to obtain the input data, which
(𝑘)
|∆𝐵𝑖 | ≤ 𝜖 (19) consists of the target bus system's line and load data. For IEEE
9 test cases, the simulation results from the NR and GS
3.2 Gauss seidel method computational methods, as well as the line flow and losses on
each transmission line in the system, are obtained using
This is a method for solving a set of nonlinear algebraic MATLAB.
equations that is iterative [7]. The method employs an initial
guess for the value of voltage to get a derived value of a
specific parameter. A derived value replaces the initial guess
value. After that, the process is repeated until the convergence
of the iteration [8-12]. The initial guess has a large effect on
the convergence time. However, the method has a poor
convergence characteristic [8, 13-16].
This is an iterative method used to solve (5) for the value of
𝑉𝑖 , and the iterative series becomes:

𝐴𝑖𝑠𝑐ℎ − 𝑗𝐵𝑖𝑠𝑐ℎ (𝑘)

+ ∑ 𝑋𝑖𝑗 𝑉𝑗
(𝑘+1) 𝑉𝑖∗ (20)
𝑉𝑖 = 𝑗≠𝑖
∑ 𝑋𝑖𝑗

Figure 1. One line diagram for IEEE 9 bus system

21
 The load and line data used for the MATLAB simulation Seidel approaches, the obtained results completely solve for
were obtained from the IEEE library and are shown in Tables all unknown values for each bus in the system. Tables 4 and 5
3 and 4. The base MVA, iteration value (accuracy), and show the simulation results for Newton Raphson and Gauss
maximum number of iterations are all defined. Figure 1 Seidel, respectively. The line flows and line losses calculated
depicts a one-line diagram of the IEEE 9-bus System. using the Newton Raphson and Gauss Seidel methods are
Tables 5 and 6 show the computed results from the shown in Tables 6 and 7. The directed graph of the obtained
numerical methods. Using the Newton Raphson and Gauss results is shown in Figures 2 and 3.

Table 3. IEEE 9 bus system load data

Load Data
Voltage Load Generation
Bus Type of Bus |V| (P.U) δ(θ) A (M.V) B (Mvar) A (MW) B (Mvar) Bmin Bmax
1 1 1.040 0.000 0.000 0.000 0.000 0.000 0.000 0.000
2 2 1.025 0.000 0.000 0.000 163.000 6.700 -99.000 99.000
3 2 1.025 0.000 0.000 0.000 85.000 -10.900 0.000 99.000
4 0 1.0 0.000 0.000 0.000 0.000 0.000 0.000 0.000
5 0 1.0 0.000 125.000 50.000 0.000 0.000 0.000 0.000
6 0 1.0 0.000 90.000 30.000 0.000 0.000 0000 0.000
7 0 1.0 0.000 0.000 0.000 0.000 0.000 0.000 0.000
8 0 1.0 0.000 100.000 35.000 0.000 0.000 0.000 0.000
9 0 1.0 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Table 4. IEEE 9-bus system line data

Line Data
Bus No. Bus No. R, (PU) X, (PU) ½ B, PU Transformer Tap
1 4 0.0000 0.0576 0.000 1
4 5 0.0100 0.0850 0.0880 1
4 6 0.0170 0.0920 0.0790 1
6 9 0.0390 0.1700 0.0179 1
5 7 0.0320 0.1610 0.0153 1
9 3 0.0000 0.0586 0.0000 1
7 2 0.0000 0.0625 0.0000 1
9 8 0.0119 0.1008 0.1045 1
7 8 0.0085 0.0720 0.0745 1

Table 5. Simulation result for IEEE 9 bus system using Newton Raphson load flow solution

Bus Voltage Mag. Angle Load Generation

Injected Mvar
No. (pu.) Degree MW Mvar MW Mvar
1 1.030 0.000 0.000 0.000 71.945 50.044 0.000
2 1.019 -9.477 0.000 0.000 163.000 27.796 0.000
3 1.012 -4.789 0.000 0.000 85.000 12.480 0.000
4 1.027 -2.254 0.000 0.000 0.000 0.000 0.000
5 1.050 -4.037 125.000 50.000 0.000 0.000 0.000
6 1.020 -3.674 90.000 30.000 0.000 0.000 0.000
7 1.021 3.846 0.000 0.000 0.000 0.000 0.000
8 1.030 0.780 100.000 35.000 0.000 0.000 0.000
9 1.016 2.055 0.000 0.000 0.000 0.000 0.000
Total 315.000 115.000 319.945 90.321 0.000

Table 6. Simulation result for IEEE 9 bus system using Gauss Seidel

Bus Voltage Mag. Angle Load Generation

Injected Mvar
No. (pu.) Degree MW Mvar MW Mvar
1 1.040 0.000 0.000 0.000 75.279 49.698 0.000
2 1.025 9.193 0.000 0.000 163.000 27.548 0.000
3 1.025 4.544 0.000 0.000 85.000 12.364 0.000
4 1.013 -2.343 0.000 0.000 0.000 0.000 0.000
5 0.0974 -4.188 125.000 50.000 0.000 0.000 0.000
6 0.989 -3.813 90.000 30.000 0.000 0.000 0.000
7 1.013 3.594 0.000 0.000 0.000 0.000 0.000
8 1.002 0.538 100.000 35.000 0.000 0.000 0.000
9 1.019 1.838 0.000 0.000 0.000 0.000 0.000
Total 315.000 115.000 323.279 89.610 0.000

22
 The IEEE 9 bus system, as shown in Table 3, consists of transformer tap of each transmission line connected together.
one swing bus, six load buses connected to a load, and two After obtaining the required parameters through the load
generator buses connected to a generator. flow solutions provided by the NR and GS metho das
The IEEE 9 bus system consists of nine line data as presented in Tables 5 and 6. The line flows and losses can be
presented in Table 4 and Table 5, which shows the values for computed as presented in Tables 7 and 8.
resistance, reactance, half susceptance per unit and

3 =1.025V , 4.79°
-1
00
M
W
,-
35
M
VA
R

9 =1.019V , 2.06° 0 M W , 0 M VAR

8 =1.002V , 0.78°

0M
R

W,
M VA

0M
VA
R
-5 0
MW ,

VAR 2 =1.025V , 9.48°

-90 M W , -30 M
-1 25

7 =1.013V , 3.85°

6 =0.989V ,-3.67° R
VA
M
0
,
W
M
0
85
MW
,1 2
.5
M

5 =0.974V ,-4.04°
VA
R

R
8 M VA
W ,2 7.
16 3 M

4 =1.013V ,-2.25°
R
VA
0M
,5
MW
.9
71

1 = 1.04V , 0°

Figure 2. A directed graph plot of the system obtained from Newton Raphson’s solution (1)

3 =1.025V , 4.54°
-1
0
0
M
W
,-
35
M
VA
R

9 =1.019V , 1.84° 0 M W , 0 M VAR

8 =1.002V , 0.54°

0M
R

W,
M VA

0M
VA
R
-5 0
MW ,

VAR 2 =1.025V , 9.19°

-90 M W , -30 M
-1 25

7 =1.013V , 3.59°

6 =0.989V ,-3.81° R
VA
M
0
,
W
M
0
85
MW
,1 2
.4
M

5 =0.974V ,-4.19°
VA
R

5 M VAR
W ,2 7.
16 3 M

4 =1.013V ,-2.34°
AR
MV
.7
49
MW,
.3
75

1 = 1.04V , 0°

Figure 3. A directed graph plot of the system obtained from Newton Raphson’s solution (2)

23
Table 7. Line flow and losses of IEEE 9-bus system obtained 6 for the NR and GS methods, respectively. When the values
from Newton Raphson from the two tables are compared, there are minor differences
between them. This difference isn't all that significant. This
Line Flow and Losses demonstrates that the two methods produce roughly
Line Line Flow Line Loss comparable results (as shown from the total load and
From Bus To Bus
A (MW) B (Mvar) MW Mvar generation power).
(f) (t)
1 4 71.945 50.044 0.000 4.090
4 5 41.086 33.095 0.000 -14.514
4 6 30.859 12.859 0.231 -14.590
6 9 -59.371 -2.551 1.405 2.516
5 7 -84.251 -2.391 2.394 9.023
9 3 -85.000 -8.363 0.000 4.117
7 2 -163.000 -11.531 0.000 16.265
9 8 24.223 3.296 0.090 -20.589
7 8 76.355 0.118 0.488 -10.997
Total
4.608 -24.679
Loss

Table 8. Line flow and losses of IEEE 9-bus system obtained

from Gauss Seidel

Line Flow and Losses

Line Line Flow Line Loss
From Bus To Bus
A (MW) B (Mvar) MW Mvar
(f) (t)
1 4 74.945 49.749 0.000 4.298
4 5 42.319 32.892 0.346 -14.454
4 6 31.806 12.714 0.239 -14.549
6 9 -58.618 -2.697 1.369 2.358
5 7 -83.248 -2.601 2.336 8.730
9 3 -84.162 -8.344 0.000 4.036
7 2 -162.088 -11.499 0.000 16.082
9 8 24.662 3.249 0.092 -20.571
7 8 76.139 0.183 0.485 -11.022
Total Loss 4.867 -25.092 Figure 4. Multiple graphs of the voltage magnitude, load
power and generation power on each bus for the IEEE 9 bus
4.1.1 Comparison of the computational time, maximum power system obtained from Newton Raphson’s solution
mismatch and the iteration number of the two methods
Table 9 presents the comparison of computational time,
maximum power mismatch, and iteration number.

Table 9. Comparison of computational time, maximum

power mismatch and iteration number

Newton Raphson Gauss Siedlel

Maximu Iteratio Maximu Iteratio
Computatio m power n Computatio m power n
nal Time (s) mismatc Numbe nal Time (s) mismatc Numbe
h (MW) r h (MW) r
−3 −3 0.02045
4𝑥10 4 7 3𝑥10 11
82

4.1.2 Directed graph

This section presents the directed graph of the 9-bus system
resolved from the NR and GS methods. It shows the node
voltages and phase angle on each bus as well the real and
reactive power loss on each transmission line. The pink dots
represent the buses. The index of the bus is shown beside it.
The lines represent the transmission line. The arrows represent
the loads.

4.2 Discussion

4.2.1 Simulation results Figure 5. Graph of the voltage magnitude, load power and
The solutions to some required unknown variables obtained generation power on each bus for the IEEE 9 bus system
after using both load flow methods are shown in Tables 5 and obtained from Newton Raphson’s solution

24
 Figure 4 depicts the generation and load power, as well as I-Line Current, R-Resistance of transmission line, X-
the voltage magnitude on each bus, as calculated using Reactance of transmission line, A_loss-Real Power Loss,
Newton Raphson's power flow solution. Figure 5 depicts the B_loss-Reactive Power Loss.
same plot, but for the Gauss Seidel's power flow solution. The By conducting this analysis, a more profound
two figures show that the voltage magnitude in bus 5 of the GS comprehension of energy dissipation arises, facilitating the
solution is significantly lower or attenuated when compared to efficient administration and enhancement of the power
the NS solution. distribution system.
The knowledge acquired from computing line flows and
4.2.2 Line flow and line losses losses through the application of the Newton-Raphson and
In the domain of power system analysis, understanding how Gauss-Seidel techniques has significant implications for both
electricity moves and where energy is lost in a power system power system analysis and operation. Through a
is essential for keeping things running smoothly. Using the comprehensive understanding of current distribution, apparent
Gauss-siedel and Newton-Raphson methods, the line flows power, and energy losses, engineers and analysis can
and losses are derived from the solution data yielded from the anticipate congestion points in advance, optimize power
load flow solutions of each of these numerical methods. transmission processes, and improve network stability.
By using the results from the power analysis, the line Tables 6 and 7 show the line flow and losses on each bus
current flowing in different parts of the system can be system line obtained from the nr and gs power flow solutions.
determined using the famous Ohm’s law equation: Figure 4 depicts the line loss resulting from the ns and gs
solutions. The nr method clearly shows that line 2 of the bus
S system (the line connecting buses 4 and 5) has a significant
I= (26)
V drop when compared to the gs method.

where, 4.2.3 Tolerance

I-Line Current, In the realm of power system analysis utilizing numerical
V-Voltage Magnitude, methods such as Newton-Raphson and Gauss-Seidel,
S-Apparent Power. convergence refers to the attainment of a stable and precise
The apparent power underscores the interplay of electrical solution. The determination of convergence hinges upon
magnitudes. meeting specific criteria, which are evaluated by assessing
tolerance-a pivotal parameter that significantly influences
𝑆 = 𝐴 + 𝑗𝐵 (27) these criteria.
The Newton-Raphson method uses iterative corrections to
By utilizing the voltage magnitudes and angles from the improve the accuracy of calculated values of bus voltages and
result of power flow analysis from the two methods, line angles. Accuracy is influenced by initial guesses and
currents are carefully calculated iteratively for each convergence criteria set for maximum change in bus voltage
transmission line. magnitudes and angles. Lower tolerance values result in higher
Line power flows show how electricity is moving between accuracy, but setting it too low may increase computational
places. It is a pivotal metric for line performance. To calculate time without significant improvement.
this, we look at the voltage and the current, and do some math Similar to the Newton-Raphson procedure, precision in
to see how much power is moving around. With this Gauss-Seidel is impacted by the convergence criteria.
information, we can see which parts of the system are busy and Decreasing tolerance values result in more precise solutions,
which ones might need attention. but akin to Newton-Raphson, excessively low tolerance
The magnitude of line power flow is established as the settings may cause excessive computational time without
product of the sending-end voltage, the conjugate of the line significant improvements in accuracy.
current: The simulation's tolerance iteration value was set to 0.001.
This means that using a high tolerance value for such analysis
S = Vi ∗ I ∗ (28) increased the accuracy of the solution.

By applying this equation in conjunction with voltage 4.2.4 Computational time

magnitudes and calculated line currents, the real-time Table 8 shows the computation time for load flow solutions
assessment of power propagation becomes discernible for using a selected iteration value of 0.001 for the NR and GS
every transmission line. methods. For the 9-bus system, the computational time for the
Incorporating the concept of power losses adds depth to GS and NR methods is very similar.
power system evaluations. Real and reactive power losses
manifest as outcomes of line current interaction with line 4.2.5 Convergence
resistances and reactances, respectively by employing the Convergence is used to calculate how quickly a power flow
following expressions: arrives at its solution. The rate of convergence is calculated by
graphing the maximum power mismatch versus the number of
A_loss = I 2 ∗ R (29) iterations.
Newton-Raphson convergence is based on the max change
B_loss = I 2 ∗X (30) in voltage magnitudes/angles. If the change is below tolerance,
it has converged. A smaller tolerance means tighter
where, convergence, more iterations, and better accuracy.

25
 Gauss-Seidel checks convergence by comparing new and 127-138.
old voltage values. Lower tolerance means stricter [2] Xi-Fang, W., Yonghua, S., Malcolm, I. (2008). Modern
convergence but longer computation. power system analysis. New York: Springer
When compared to the Newton-Raphson method, the Science+Business Media. https://doi.org/10.1007/978-0-
Gauss-Seidel method has a slower convergence rate. Newton 387-72853-7
Raphson has the fastest convergence rate. The results of the [3] Hale, H.W., Goodrich, R.W. (1959). Digital computation
line graph of the line flow and line loss in the different lines of or power flow-some new aspects. Transactions of the
the IEEE 9-bus system are presented in Figure 6. American Institute of Electrical Engineers. Part III:
Power Apparatus and Systems, 78(3): 919-923.
https://doi.org/10.1109/AIEEPAS.1959.4500466
[4] Grainger, J.J., Stevenson Jr, W.D. (1994). Power system
analysis. McGraw-Hill Series in Electrical and Computer
Engineering.
[5] Kabisama, H. (1993). Electrical power engineering. New
York: McGraw-Hill.
[6] Milano, F. (2008). Continuous Newton's method for
power flow analysis. IEEE Transactions on Power
Systems, 24(1): 50-57.
https://doi.org/10.1109/TPWRS.2008.2004820
[7] Hadi, S. (2010). Power system analysis (3rd ed.). Edition,
New York: Mc-Graw Hill.
[8] Keyhani, A., Abur, A., Hao, S. (1989). Evaluation of
power flow techniques for personal computers. IEEE
Transactions on Power Systems, 4(2): 817-826.
https://doi.org/10.1109/59.193857
Figure 6. Line graph of the line flow and line loss in the [9] Aroop, B., Satyajit, B., Sanjib, H. (2014). Power flow
different lines of the IEEE 9-bus system analysis on IEEE 57 bus system using Mathlab.
International Journal of Engineering Research &
Technology (IJERT), 3.
5. CONCLUSION [10] Adejumobi, I.A., Adepoju, G.A., Hamzat, K.A.,
Oyeniran, O.R. (2014). Numerical methods in load flow
Both Newton-Raphson and Gauss-Seidel methods have analysis: An application to Nigeria grid system.
been extensively studied for their convergence behavior in International Journal of Electrical and Electronics
load flow analysis. Understanding convergence behavior has Engineering (IJEEE), 3.
led to more robust and reliable power system analysis tools. [11] Gilbert, G.M., Bouchard, D.E., Chikhani, A.Y. (1998). A
Investigations have shown that both methods can provide comparison of load flow analysis using distflow, gauss-
accurate results when applied correctly. However, the choice seidel, and optimal load flow algorithms. In Conference
of method, initial conditions, and convergence criteria can Proceedings. IEEE Canadian Conference on Electrical
impact accuracy. and Computer Engineering (Cat. No. 98TH8341), 2: 850-
Comparisons of the Newton-Raphson and Gauss-Seidel 853. https://doi.org/10.1109/CCECE.1998.685631
methods have also revealed the differences in computational [12] Afolabi, O.A., Ali, W.H., Cofie, P., Fuller, J., Obiomon,
efficiency. Gauss-Seidel is simpler but may require more P., Kolawole, E.S. (2015). Analysis of the load flow
iterations, while Newton-Raphson can converge faster but problem in power system planning studies. Energy and
involves more complex calculations. These findings have Power Engineering, 7(10): 509.
guided the selection of methods based on the size and http://dx.doi.org/10.4236/epe.2015.710048
complexity of the power system. [13] Aeggegn, D.B., Salau, A.O., Gebru, Y.W., Agajie, T.
The study found that computational/numerical methods can (2022). Mitigation of reactive power and harmonics in a
be used to calculate line flows and power losses in a power case of industrial customer. International Journal of
system. These methods derive the voltage magnitude, phasal Engineering Research in Africa, 60: 107-124.
angles, real and reactive power of the system's buses. The load https://doi.org/10.4028/p-j716jb
flow analysis methods of Gauss Seidel and Newton Raphson [14] Kassahun, H.E., Salau, A.O., Osaloni, O.O., Olaluyi, O.J.
were used to analyze an IEEE 9-bus test system, with results (2023). Power system small signal stability enhancement
showing only slight differences between the total line flow and using fuzzy based statcom. Przeglad Elektrotechniczny,
losses obtained from the two iterative solutions. A complete 2023(8): 27-32. https://doi.org/10.15199/48.2023.08.05
load flow analysis was performed on an IEEE standard 9-bus [15] Alyu, A.B., Salau, A.O., Khan, B., Eneh, J.N. (2023).
system using both computational methods. Hybrid GWO-PSO based optimal placement and sizing
of multiple PV-DG units for power loss reduction and
voltage profile improvement. Scientific Reports, 13(1):
REFERENCES 6903. https://doi.org/10.1038/s41598-023-34057-3
[16] Gebru, F.M., Salau, A.O., Mohammed, S.H., Goyal, S.B.
[1] Vijayvargia, A., Jain, S., Meena, S., Gupta, V., Lalwani, (2022). Analysis of 3-phase symmetrical and
M. (2016). Comparison between different load flow unsymmetrical fault on transmission line using Fortescue
methodologies by analyzing various bus systems. Theorem. WSEAS Transactions on Power Systems, 17:
International Journal of Electrical Engineering, 9(2): 316-323. https://doi.org/10.37394/232016.2022.17.32

26