International Journal of Trend in Scientific Research and Development (IJTSRD)

Volume 8 Issue 5, Sep-Oct 2024 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470

Enhancing Voltage and Minimizing Losses in Power

Distribution Systems Using Fuzzy Based DG Integration
Phyo Thiha Lwin, Tin Tin Htay
Department of Electrical Power Engineering, Yangon Technological University, Yangon, Myanmar

ABSTRACT How to cite this paper: Phyo Thiha

In modern electrical distribution systems, addressing issues such as Lwin | Tin Tin Htay "Enhancing Voltage
high losses and maintaining voltage in acceptable limit is crucial. and Minimizing Losses in Power
One effective approach is integrating small-scale, decentralized Distribution Systems Using Fuzzy Based
DG Integration" Published in
generation units, known as Distributed Generation (DG), directly into
International Journal
the network. These DG units, placed closer to the end consumers, of Trend in
help reduce power losses and improve voltage profiles. This paper Scientific Research
explores how installing DG units affects a distribution circuit, and Development
focusing on how it impacts the network’s performance, particularly (ijtsrd), ISSN: 2456-
in terms of voltage and power losses. It introduces Fuzzy Algorithm 6470, Volume-8 |
for finding the optimal size and placement of DG units in a specific Issue-5, October IJTSRD70500
area, Phyu Township, to minimize losses and enhance bus voltage. 2024, pp.1144-
Using MATLAB software for simulation, the study demonstrates that 1150, URL:
with careful placement and sizing of DG units, it's possible to www.ijtsrd.com/papers/ijtsrd70500.pdf
significantly cut system losses and maintain bus voltage within
Copyright © 2024 by author (s) and
acceptable ranges.
International Journal of Trend in
KEYWORDS: Distribution system, DG (Distributed Generation), Scientific Research and Development
Fuzzy, Voltage improvement, Power losses Journal. This is an
Open Access article
distributed under the
terms of the Creative Commons
Attribution License (CC BY 4.0)
(http://creativecommons.org/licenses/by/4.0)

I. INTRODUCTION
Power utilities today face significant challenges, the units. A range of techniques has been developed to
including growing electricity demand, dwindling solve the challenges of DG placement and sizing,
fossil fuel reserves, and rising environmental from analytical methods to more advanced approaches
concerns. As transmission and distribution networks like evolutionary computing. Methods such as Genetic
approach capacity, this increased demand causes Algorithms and Ant Colony Optimization are known
voltage drops and higher power losses, occasionally for delivering strong results, but they often involve
resulting in blackouts. One potential solution is to heavy computational efforts. To address this,
construct new power plants to boost voltage levels, engineers have long pursued simpler, more intuitive
reduce losses, and ensure a reliable power supply. solutions that, while easier to implement may not
However, relying on aging, overloaded networks always produce optimal results. Fuzzy logic-based
ultimately degrades system performance. methods offer a middle ground, reducing
Furthermore, building new plants or upgrading computational complexity by using fuzzy membership
existing infrastructure proves both costly and functions to model the system. The most challenging
environmentally burdensome. Consequently, power aspect of these methods lies in selecting the
companies are increasingly focusing on appropriate membership functions. Typically, these
improvements in distribution systems through the approaches rely on indicators like node voltage levels
integration of Distributed Generation (DG). and power loss to guide the optimal placement of DG
DG units are frequently incorporated into power units in distribution networks.
distribution systems to lower energy losses and II. A CASE STUDY OF DISTRIBUTION
maintain voltage levels within acceptable limits. The SYSTEM
effectiveness of DG integration depends on various Phyu Township is experiencing ongoing power
factors, including location, size, type, and quantity of disruptions due to frequent scheduled and

@ IJTSRD | Unique Paper ID – IJTSRD70500 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 1144
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
unscheduled load shedding, which are a result of an optimally, there is a noticeable improvement in bus
insufficient power supply and low voltage levels. The voltage and a marked reduction in power losses. These
imbalance between the growing electricity demand results highlight how important precise optimization
and the limited capacity of the power infrastructure techniques are for enhancing the efficiency of modern
leads to frequent disruptions in scheduled electricity power distribution systems.
and occasional power outages. Causing these
IV. MODELLING OF DISTRIBUTION
difficulties is by the township’s geographic location,
SYSTEM
nestled between two edges of 33 kV power lines,
Figure 2 illustrates the modeling of the distribution
making it harder to maintain voltage stability for its system. The simulation began by analyzing the radial
residents and businesses.
distribution network without any distributed
As Phyu Township is on the point of significant generation (DG) connected. This initial step was
growth, with numerous restaurants and electric conducted to determine the voltage profiles at each
vehicle (EV) charging stations being established, the bus within the system. Following this, the model was
demand for electricity is set to rise even further. This tested under four distinct conditions:
growth highlights the urgent need to upgrade the Condition 1: Simulating the power flow from the
power infrastructure to support the town's increasing Taungoo substation to the Phyu substation under
energy needs. Addressing the issues of low voltage minimum load conditions.
and inadequate power supply is crucial to ensuring
the electricity network meets the demands of this Condition 2: Simulating the power flow from the
development. Taungoo substation to the Phyu substation under
maximum load conditions.
To tackle the ongoing challenges of load shedding
and voltage instability, strategically integrating Condition 3: Simulating the power flow from the
Distributed Generation (DG) at optimal locations and Tharyar Gone substation to the Phyu substation under
sizes within the township is essential. By doing so, minimum load conditions.
Phyu Township can bolster its power supply, reduce Condition 4: Simulating the power flow from the
the strain on the existing infrastructure, and improve Tharyar Gone substation to the Phyu substation under
bus voltage and reduce power losses. Proper maximum load conditions.
placement and sizing of DG units provides electricity
more efficiently, ensuring that the growing demands These conditions is used to evaluate how different
of the community are met. In this way, incorporating load conditions affect the voltage profiles and overall
DG is a key step toward solving the township's performance of the distribution network.
electricity challenges and supporting its long-term
growth and prosperity.

Figure. 1: Single line diagram of 33kV

distribution system.
III. METHODOLOGY
Figure.2: The modelling diagram of the power
This paper explores the use of MATLAB to model
system
and simulate a 33 kV distribution system, focusing on
how Fuzzy logic is used to optimize the placement V. LOAD FLOW ANALYSIS OF
and sizing of distributed generation (DG) units. The DISTRIBUTION SYSTEM (WITHOUT
Fuzzy algorithm helps pinpoint the best locations and DG)
capacities for these units, all with the goal of boosting The simulation results are displayed in Figures 3(a),
the system's performance. By carefully placing DG (b), (c), and (d), which show the voltage levels at
units, the method aims to cut down on power losses each bus under the different conditions mentioned
and improve voltage stability throughout the earlier. These graphs provide a clear visual
distribution network. A comparison of system representation of how voltage is distributed across the
performance for both with and without DG integration network, making it easier to analyze and compare the
shows that when DG units are placed and sized system's performance under varying load scenarios.

@ IJTSRD | Unique Paper ID – IJTSRD70500 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 1145
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
In Conditions 1 and 2, the Phyu bus voltage starts at
around 27.07 kV when the load is minimal but drops
significantly to 21.38 kV under maximum load. This
sharp decline suggests that the system faces higher
losses or struggles with voltage regulation as the load
increases. On the other hand, in Conditions 3 and 4,
the voltage at the Phyu bus is higher under minimal
load, around 28.98 kV, but drops even more, down to
20.34 kV, under maximum load. While the voltage is
better under light load compared to Conditions 1 and
2, it suffers a steeper drop when the system is heavily
loaded.
(a)
When considering where to place and size DG units,
it’s important to look at both the voltage levels and
the power losses in each condition. Tables 1 to 4
provide a detailed breakdown of the power losses for
each scenario.
Table I. Loss Report for Min Load (TGO-Phyu)
SN Name P (MW) Q (MVAR)
1 Total generation 7.71 3.14
2 Total PQ load 0 0
3 Total Z shunt 7.26 2.47
4 Total losses 0.45 0.67
5 Loss percentage 5.84 21.34
(b) Table II. Loss Report for Max Load (TGO-
Phyu)
SN Name P (MW) Q (MVAR)
1 Total generation 15.58 7.41
2 Total PQ load 0 0
3 Total Z shunt 13.46 3.46
4 Total losses 2.12 3.95
5 Loss percentage 13.61 53.31
Table III. Loss Report for Min Load (TYG-
Phyu)
SN Name P (MW) Q (MVAR)
1 Total generation 12.82 4.93
(c) 2 Total PQ load 0 0
3 Total Z shunt 12.35 4.18
4 Total losses 0.47 0.74
5 Loss percentage 3.67 15.01
Table IV. Loss Report for Max Load (TYG-
Phyu)
SN Name P (MW) Q (MVAR)
1 Total generation 27.26 12.18
2 Total PQ load 0 0
3 Total Z shunt 25.06 6.42
4 Total losses 2.19 5.76
5 Loss percentage 8.03 47.29
The active power losses on the Taungoo to Phyu line
(d) are about 5.84% under minimum load and rise to
Figures 3: Bus voltage under (a) condition 1, (b) 13.61% under maximum load. In comparison, the
condition 2, (c) condition 3 and (d) condition 4 Tharyar Gone to Phyu line experiences lower losses,

@ IJTSRD | Unique Paper ID – IJTSRD70500 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 1146
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
around 3.67% at minimum load and 8.03% at Index (VSI). The output variable is the DG Suitability
maximum load. This shows that the Taungoo to Phyu Index (DGSI). The Power Loss Index (PLI) ranges
line consistently has higher losses under both loading from 0 to 1, the per-unit nodal voltage spans from 0.9
conditions. to 1.1, and the DG Suitability Index also falls
Integrating Distributed Generation (DG) into the between 0 and 1. The formulation of the PGI is
Taungoo to Phyu line could help improve the system provided below.
by reducing these power losses and stabilizing X (i ) − Y
PGI (i ) =
Z −Y
voltage, especially when the load is high. With
strategic placement and sizing of DG units, along Where, X = Power Generation improvement at bus i
with careful coordination with the existing grid, the Y = Minimum generation improvement
overall performance of the line could be greatly Z = Maximum generation improvement
improved. n = Number of nodes
VI. LOCATION AND SIZING OF DG BY For the PGI, we have defined seven triangular
USING Fuzzy Algorithm membership functions: Very Low (VL), Low (L),
This paper presents a fuzzy logic approach to Low Medium (LM), Medium (M), High Medium
determine the optimal placement and sizing of (HM), High (H), and Very High (VH), as illustrated
Distributed Generation (DG) units in a distribution in Fig. 4. Similarly, the voltage has its own set of
system. The fuzzy algorithm is designed to handle the seven triangular membership functions: Very Low
uncertainties and variability in system parameters (VL), Low (L), Low Normal (LN), Normal (N), High
such as load demand, voltage levels, and power Normal (HN), High (H), and Very High (VH),
losses. It works by taking into account multiple input depicted in Fig. 5. Finally, for the DG Suitability
variables, such as the Voltage Stability Index (VSI), Index, we also use seven membership functions: Very
Power Loss Index (PLI), and Power Generation Index Low (VL), Low (L), Low Medium (LM), Medium
(PGI). These variables are modeled using fuzzy (M), High Medium (HM), High (H), and Very High
membership functions to reflect the imprecision (VH). However, these membership functions are
inherent in the system. defined as Gaussian, as shown in Figure 6.
The core of the method is a Fuzzy Inference System
1
(FIS) that applies a set of rules to evaluate the
suitability of each bus in the distribution network for
DG integration. The primary objectives of this 0.5
approach are to minimize real power losses and
maintain voltage levels within permissible limits. The
0
FIS generates a ranking of candidate nodes,
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
identifying the most suitable locations for DG Membership function plot for PGI
placement, along with the optimal DG size. By Figure. 4, Input 1 PGI Membership function
achieving a balance between voltage stability and
reduced losses, this fuzzy approach offers a flexible
and adaptive solution for DG integration, even in
complex and dynamic operational environments. For
instance, it is logical to prioritize sections of a
distribution system that exhibit high losses and low
voltage for the placement of Distributed Generation
(DG). In contrast, sections with low losses and stable
voltage levels are generally less suitable for DG
integration. Figure. 5, Input 2 VSI Membership function
VII. IMPLEMENTATION OF FUZZY
METHOD
This paper examines two input variables and one
output variable to determine DG placement
suitability. The first input variable is the Power
Generation Index (PGI), while the second is the per-
unit nodal voltage (V), which represents the
proportion of received voltage relative to a reference
voltage and is also known as the Voltage Stability Figure. 6, Output DGSI Membership function

@ IJTSRD | Unique Paper ID – IJTSRD70500 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 1147
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
To assess the suitability of DG placement at a specific profiles are depicted in Figures 8(a) and 7(b),
node, a set of fuzzy rules with multiple antecedents respectively.
has been formulated. These rules are summarized in
the fuzzy decision matrix in Table 5. In this study, 49
rules have been constructed.
Table V. Fuzzy Decision Matrix
VSI
VL L LN N HN H VH
PGI
VL M LM LM L L VL VL
L HM M LM LM L L VL
LM HM HM M LM LM L L
M H HM HM M LM LM L
HM H H HM HM M LM LM
H VH H H HM HM M LM
VH VH VH H H HM HM M
Optimal locations for DG placement are identified (a)
based on DGSI values. For this system, optimal
locations have been determined, with candidate
locations and their corresponding DGSI values
presented in Table 6.
Table VI. Buses with DGSI values
Bus PGI VSI DGSI
No (Input 1) (Input 2) (Output)
1 0.000 0.897 0.427
2 0.220 0.938 0.395
3 0.197 0.933 0.375
4 0.340 0.953 0.346
5 0.730 1.002 0.340
6 1.000 0.989 0.605
7 0.614 0.993 0.270
(b)
When integrating DG into a power distribution Figure 8; Bus voltage of power distribution
system, it is essential to size the DG units to meet the system after DG integration (a)with minimum
local power demand. To achieve this, the fuzzy load (b)with maximum load
algorithm is used to determine optimal DG sizing, The simulation results for the 33 kV Taungoo to Phyu
taking into account the connected load at each bus in line reveal a substantial improvement in bus voltage
the simulation model. For instance, the total demand levels following DG integration. Specifically, the
at bus 6 is set to 0.8 MW. voltage at the Phyu bus increases from 27.01 kV to
VIII. DISTRIBUTION SYSTEM WITH DG 31.91 kV under minimum load conditions and from
INTEGRATION 21.38 kV to 30.33 kV under maximum load
Figure 7 show the modelling of DG integration in the conditions. Figures 9(a) and 9(b) illustrate the
distribution system. comparison of bus voltages before and after DG
integration for both loading scenarios.

Figure 7: The modelling of DG integration in the

distribution system
The simulation results include the bus voltage profiles
for the distribution system with DG, showing both
minimum and maximum load conditions. These (a)

@ IJTSRD | Unique Paper ID – IJTSRD70500 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 1148
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
reducing system losses. As shown in the comparison
table, the total active power losses decrease from 2.12
MW without DG to 1.91 MW with DG integration.
Similarly, the reactive power losses reduce from 3.95
Mvar to 3.55 Mvar. This reduction in losses translates
to a decrease in loss percentages, from 13.59% to
12.18% for active power and from 53.27% to 50.12%
for reactive power. These results confirm that DG
(b) integration enhances system efficiency even under
Figure 9; Bus voltage comparison of power maximum load conditions, contributing to both active
distribution system with and without DG and reactive power loss reductions.
integration (a) minimum load condition (b) IX. CONCLUSION
maximum load condition. In conclusion, this research demonstrates the
Additionally, a detailed analysis of system losses was significant benefits of integrating Distributed
conducted, comparing the results before and after the Generation (DG) into the 33 kV Taungoo to Phyu
integration of DG. The impact of DG on overall line, utilizing MATLAB software and a Fuzzy
system efficiency and performance is clearly outlined algorithm for optimal placement and sizing of DG.
in Tables 7 and 8. The simulation results reveal substantial
enhancements in the performance of the power
Table VII. Load Flow and Losses Report distribution system, with the voltage at the Phyu bus
Comparison for Minimum Load increasing notably from 27.01 kV to 31.91 kV under
minimum load conditions and from 21.38 kV to 30.33
kV under maximum load conditions. These
improvements underscore the effectiveness of DG in
stabilizing and enhancing voltage profiles across
varying load scenarios.
Additionally, load flow analysis indicates a
significant reduction in total power losses due to DG
integration. Under minimum load conditions, active
Table VIII. Load Flow and Losses Report power losses decreased from 0.45 MW to 0.36 MW,
Comparison for Maximum Load while reactive power losses were reduced from 0.67
Mvar to 0.50 Mvar, resulting in percentage reductions
from 5.80% to 4.61% for active power and from
21.46% to 16.62% for reactive power. Similarly,
under maximum load conditions, active power losses
decreased from 2.12 MW to 1.91 MW and reactive
power losses from 3.95 Mvar to 3.55 Mvar,
translating to loss percentage reductions from 13.59%
to 12.18% for active power and from 53.27% to
The integration of Distributed Generation (DG) into 50.12% for reactive power.
the power distribution system significantly reduces Overall, the integration of DG not only enhances bus
total power losses, as evidenced by the load flow voltage stability but also significantly reduces both
analysis under minimum load conditions. As shown active and reactive power losses across different
in the comparison table, the total active power losses loading conditions. These findings highlight the
decrease from 0.45 MW to 0.36 MW after DG critical role of DG in improving the efficiency and
integration. Additionally, the reactive power losses reliability of power distribution networks, presenting
also reduce from 0.67 Mvar to 0.50 Mvar. This a viable solution to modern energy challenges. Future
improvement reflects a reduction in loss percentages studies should investigate additional optimization
from 5.80% to 4.61% for active power and from strategies and the impacts of various DG technologies
21.46% to 16.62% for reactive power, demonstrating to further maximize these benefits.
the positive impact of DG on system efficiency
ACKNOWLEDGEMENT
during minimum load conditions.
The author extends profound gratitude to his
The results for maximum load conditions also Chairperson, Dr. Soe Win, Professor and Head of the
demonstrate the positive effect of DG integration in Department of Electrical Power Engineering at

@ IJTSRD | Unique Paper ID – IJTSRD70500 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 1149
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
Yangon Technological University, and his Using Fuzzy Techniques” , (IJACSA) International
Supervisor, Dr. Tin Tin Htay, Professor of the Journal of Advanced Computer Science and
Department of Electrical Power Engineering at Applications, Vol. 1, No. 3, September 2010
Yangon Technological University, for their
[3] Zar Zar Tun, , “Optimal Capacitor Bank Sizing and
invaluable guidance and mentorship. He also wishes Placement for Loss Reduction in Distribution System
to acknowledge Sayar U Kyaw Zayar Win and all
by Using Fuzzy Techniques”, Conference Of the
educators who have imparted knowledge throughout
ASEAN Federation Of Engineering Organizations,
his academic journey, along with each individual who
CAFEO 30, December 2012.
contributed to the completion of this work. Heartfelt
appreciation is further extended to his family and [4] Ali Selim Mohamed Ibrahim, “Optimal Distributed
friends for their unwavering support, patience, and Generation Allocation Considering Uncertainty Of
encouragement, which proved instrumental in Power Generation And Time-Varying Load”, Jaén,
completing this endeavor. Lastly, the author sincerely Enero 2021.
thanks all those who, directly or indirectly, offered [5] Mostafa Elshahed, Mahmoud Dawod , Zeinab H.
ideas and assistance in various ways. Osman, “Optimization of Real Power Loss and
REFERENCES Voltage Stability Index for Distribution Systems with
[1] M. Padma Lalitha, V.C. Veera Reddy and N. Distributed Generation”, International Journal of
Sivarami Reddy , “Application of Fuzzy and ABC Engineering Research in Africa, ISSN: 1663-4144,
Algorithm for DG Placement for Minimum Loss in Vol. 33, pp 100-114.
Radial Distribution System”, Iranian Journal of [6] T. Adefarati, R.C. Bansal, “Integration of
Electrical & Electronic Engineering, Vol. 6, No. 4, Renewable Distributed Generators into the
Dec. 2010 Distribution System: A Review”, IET
[2] Sheeraz kirmani, Md. Farrukh Rahman, Chakresh Renewable Power Generation.
Kumar, “Loss Reduction in Distribution System

@ IJTSRD | Unique Paper ID – IJTSRD70500 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 1150