International Journal of Applied Power Engineering (IJAPE)

Vol. 13, No. 1, March 2024, pp. 11~19

ISSN: 2252-8792, DOI: 10.11591/ijape.v13.i1.pp11-19  11

Single-phase distributed generations for power balanced using

adaptive real coded genetic algorithm

Umar Umar1, Faanzir Faanzir1, Iswan Iswan1, Ramly Rasyid1, Muhammad Natsir Rahman1,
Amal Khairan2, Haryati Haryati3
1
Department of Electrical Engineering, Faculty of Engineering, Universitas Khairun, Ternate, Indonesia
2
Department of Informatics Engineering, Faculty of Engineering, Universitas Khairun, Ternate, Indonesia
3
Faculty of Engineering, Universitas Khairun, Ternate, Indonesia

Article Info ABSTRACT

Article history: Easy access to distributed generation (DG) technology is promoting the
utilization of single-phase DGs for residential purposes. Surplus energy
Received Apr 14, 2023 generated by household DGs can be shared or sold to other communities
Revised Jun 20, 2023 through existing networks. However, the interconnection of single-phase
Accepted Jun 26, 2023 DGs from residential generators to the distribution network requires careful
handling to secure the reliability and quality of the electric power system.
This paper focuses on the optimal placement of single-phase multi-type DGs
Keywords: on unbalanced distribution systems that are connected to nonlinear loads.
The objective of this study is to minimize the power losses and voltage
Adaptive unbalances in the distribution networks. To verify the efficacy of this
Distributed generation method in reducing voltage unbalances and harmonics, an optimization
Harmonic approach is also presented using three-phase DG. Optimal placement of DG
Multi-type is performed on a modified Kaliasin distribution system using an adaptive
Real-coded real coded genetic algorithm (ARC-GA). Simulation results demonstrate that
Voltage-unbalanced the installation of single-phase DGs is highly effective in reducing power
losses and voltage unbalances in the distribution networks.
This is an open access article under the CC BY-SA license.

Corresponding Author:
Umar Umar
Department of Electrical Engineering, Faculty of Engineering, Universitas Khairun
Street Pertamina, Campus II Gambesi, 97719, Ternate, Maluku Utara, Indonesia
Email: [email protected]

1. INTRODUCTION
The need for inexpensive and reliable electricity has led to the widespread use of energy-based
distributed generation (DG) systems. In 2016, the global installation of DG systems saw a significant growth
of 183% [1], with these systems contributing 20 to 30% of the total electricity production [2]. The recent
changes in power system management regulations have shifted the control of power plants away from the
government, allowing communities to have their own generators and even sell surplus energy to local areas
using established networks. This development is anticipated to become a prevailing trend in the future.
DG in a distribution system can be used to minimize active power loss [3], reduce voltage
deviation [4], and minimize costs [2], [5]. The integration of DG into a distribution system is commonly
performed using a balanced system approach, where harmonics and load imbalances are often disregarded.
Electrical power system in general is an unbalanced system [6]. Modifying an unbalanced system to a
balanced system does not represent a real system. A significant voltage unbalance in the distribution system
causes various issues, such as overheating, harmonics and interference on the three-phase equipment [7]. The
problem of load imbalance has been discussed by several researchers with different approaches, positive
voltage ratio [8], reinforce learning [9], Cuckoo search [10], improved particle swarm optimization [11],

Journal homepage: http://ijape.iaescore.com

12  ISSN: 2252-8792

crisscross optimization algorithm [12], optimal power flow [13], open distribution system simulator [14], and
symbiotic organisms search (SOS) [15]. These studies do not discuss nonlinear loads. However, the number
of nonlinear loads in a power system can reach 40-41% of the load [16]. Harmonics in the distribution system
can increase power loss [17], decrease insulation life, increase temperature, and reduce power factor [18].
The interconnection of DG into the distribution system considering harmonics and imbalances has
been performed by several researchers with different methods: genetic algorithm [19], real-coded genetic
algorithm (RC-GA) [20], harmony search [21], biogeography-based optimization [22], and modified group
experience of teaching learning-based optimization [23]. All these studies use three-phase DG so that the
injection of three-phase power into an unbalanced distribution system has a less significant effect on voltage
unbalance, even in some cases, it can increase voltage unbalance and harmonic.
The effect of single-phase DG in an unbalanced distribution system has been studied by
Amirullah et al. [24], Pinthurat and Hredzak [25], and Jiao et al. [26], but DG type and harmonics are not
considered. This paper explores the most advantageous positioning of various types of single-phase DG units
in unbalanced distribution systems. The objective of this study is to minimize power losses, reduce the
voltage unbalanced, and maintain the voltage and harmonics at an acceptable limit. The contributions of this
research are: i) optimal placement of DG in addition to considering the imbalance also considers the
harmonics, ii) using single-phase DG and it compares with three-phase DG for harmonic and unbalanced
mitigation, and iii) the optimization method uses the ARC-GA method which has better performance than the
standard GA. The remaining sections of this paper are organized as follows: in the second section, the ARC-
GA, objectives, DG modeling and optimization algorithms are discussed thoroughly. The third section
discusses the optimization results and the performance of the methods. The conclusions are discussed in the
last section.

2. METHOD
2.1. Adaptive real coded genetic algorithm
Studies indicate that the genetic algorithm does not always ensure convergence to the global
optimum. To address this issue, the ARC-GA method is proposed, which adjusts the probabilities of
crossover (𝑝𝑐 ) and mutation (𝑝𝑚 ) based on changes in individual fitness values [2]. When individual
variability is low, pc and pm are increased. This increase in pc and pm results in the generation of more new
individuals, thereby enhancing individual variability. Conversely, if individual variability is high, 𝑝𝑐 and 𝑝𝑚
are reduced. This decrease in 𝑝𝑐 and 𝑝𝑚 reduces the rate at which new individuals are created through
crossover and mutations. The probability formulas for crossover and mutation are as (1) and (2) [2].

𝐹𝑎𝑔 𝑁
𝑝𝑐 = 𝑝𝑐𝑜 (1 + 𝑎 ) (1)
(𝐹𝑚𝑎𝑥 −𝐹𝑎𝑔 )𝑁+𝐹𝑎𝑔 𝑁

𝑜 𝐹𝑎𝑔 𝑁
𝑝𝑚 = 𝑝𝑚 (1 + 𝑏 ) (2)
(𝐹𝑚𝑎𝑥 −𝐹𝑎𝑔 )𝑁+𝐹𝑎𝑔 𝑁

𝑝𝑐𝑜 and 𝑝𝑚
𝑜
are crossover and mutation probabilities. Fmin, Fmax, and Favg are the minimum, maximum, and
average fitness of individual respectively. N, a, and b are real numbers.

2.2. The objective

2.2.1. Active power loss
The power losses in the distribution networks are influenced by two main factors: the current
flowing through the system and the impedance of the lines. The line current is influenced by the composition
of the load, while the line impedance is determined by various factors such as the material used, the length of
the distribution lines, and their diameter. The active power loss in an unbalanced distribution system can be
formulated as (3).

(𝐹) (𝐹) (𝐹) (𝐹) (𝐹) (𝐹)

𝑃𝑙𝑜𝑠𝑠 = ∑𝐹=𝐴,𝐵,𝐶 (∑𝑛𝑖=1 ∑𝑛𝑗=1 𝑉𝑖 𝑉𝑗 𝑌𝑖𝑗 (𝐶𝑜𝑠(𝜃𝑖 − 𝜃𝑗 − 𝛿𝑗 )) (3)
𝑖

P, V, and Y are power, voltage, and admittance respectively. While n, θ, and δ are the number of buses, phase
voltage angle, and phase admittance angle. DG placement is intended to reduce power loss on the distribution
system. The objective function associated with power loss can be defined in (4).

𝑃𝐷𝐺
𝐹1 = 𝑚𝑖𝑛 [ ] (4)
𝑃𝐵

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 11-19
Int J Appl Power Eng ISSN: 2252-8792  13

PB and PDG are power loss before and after DG installation.

2.2.2. Voltage profile

In addition to reducing power losses, DG can also improve the voltage level at each bus. The phase
voltage on each bus is expected to be close to or equal to 1 pu. The objective function related to the voltage
profile on each bus is formulated as (5).

𝐹2 = 𝑚𝑖𝑛( |𝑉𝑟𝑒𝑓 − 𝑉𝑃_𝑖 |) (5)

𝑉𝑟𝑒𝑓 and 𝑉𝑃_𝑖 are voltage reference and phase voltage at bus i.

2.2.3. Voltage unbalance

The unbalanced loading in the distribution system leads to an imbalance of inter-phase voltages. The
objective function related to the voltage unbalanced is formulated as (6).

Max(|𝑉𝑖𝑘 -V𝑖mean |)
𝐹3 = ×100% (6)
|𝑉𝑖mean |

Vi, Vimean, and k are voltage bus i, average voltage bus i, and phase respectively.

2.2.4. Harmonic
The presence of harmonics can negatively impact the precision of measuring instruments,
particularly single-phase devices that rely on a disc mechanism. These instruments tend to exhibit
approximately 2% faster movement when subjected to harmonic loads. Harmonics can also lead to alterations
in currents, frequencies, and impedances, which may trigger the operation of protective equipment like relays
even in the absence of actual disturbances [27]. The IEEE-519 standard sets the permissible individual
harmonic distortion (iHD) at 3%, while the total harmonic distortion (THD) must remain below 5% [19]. The
levels of total harmonics and individual harmonics are formulated as (7) and (8).

1, if iHD𝑗 ≤3
𝑁iHD𝑗 = { 𝑒𝑥𝑝(𝛼|3−iHD (7)
𝑗 |)if iHD𝑗 >3

1, if THD ≤5
𝑖
𝑁THD𝑖 = { 𝑒𝑥𝑝(𝛼|5-THD |) if THD >5 (8)
𝑖

NTHD and NiHD are total harmonics and individual harmonics level. i, j, and α are individual, order harmonics,
bus number and constanta respectively. The individual and total harmonics limit violations can be formulated
as (9) and (10).

𝑁𝑖 = ∏𝑘𝑗=1 𝑁iHD (9)

𝑁𝑇 = ∏𝑛𝑖=1 𝑁THD (10)

The objective function pertaining to harmonics can be computed utilizing as (11).

𝐹4 = ∏𝑘𝑗=1 𝑁iHD × ∏𝑘𝑖=1 𝑁THD (11)

If both THD and iHD meet the specified standards, then F4 is assigned a value of 1. However, if either THD
or iHD fails to meet the standards, F4 will yield a significantly large number. The objective function in this
study is a composite of active power loss, voltage profile, voltage unbalance, and harmonics, formulated as (12).

𝐹 = 𝑤1 𝐹1 +w2 𝐹2 +w3 𝐹3 +w4 𝐹4 (12)

𝐹 is the objective function while 𝑤1 , w2 , w3 , and w4 are weights.

2.3. DG modelling
DG performance in the power system is highly determined by location, type, and rating of DG. In
this paper, single-phase DG is modeled into four types [3]: type I DG: supplying active power only; type II
DG: supplying reactive power only; type III DG: supplying active and reactive power; and type IV DG:

Single-phase distributed generations for power balanced using adaptive … (Umar Umar)
14  ISSN: 2252-8792

supplying active power but absorbing reactive power. The maximum size of DG type I and type II are
100 kVar and 100 kW respectively. The maximum size of DG type III is 100 kVA with power factor 0.95
leading, while maximum size of DG type IV is 100 kVA with power factor 0.95 lagging. The maximum
penetration of DG into the power system is 30% and the maximum number of DG is 6.

2.4. Optimization algorithm

The optimization of the location and size of single-phase DG is performed simultaneously.
Individual is modeled into four strings, namely location, phase, type, and size. The location string represents
the bus number where the DG will be located. This string contains integers representing the number of buses
in the distribution system. The second string is a phase string representing the phase in which DG will be
placed. This string contains integers 1, 2, and 3 representing phase a, b, and phase c respectively. The third
string is the DG type string. This string contains integers between 1 and 4 according to the number of DG
types. The fourth string is DG size string. This string contains a real number between 0 and 1. 0 and 1
indicate the minimum and maximum size of DG. The DG size is obtained through the decoding process using (13).

𝑅𝐷𝐺 = 𝑟𝑓 𝑥 𝑅𝑚𝑎𝑥 (13)

𝑅𝐷𝐺 , 𝑟𝑓, and 𝑅𝑚𝑎𝑥 are actual size, string size, and maximum DG size.
ARC-GA is employed to enhance the performance of the genetic algorithm (GA). Adaptations in the
crossover and mutation operators are introduced to be responsive to variations in individual fitness. When
individual variability is high, the likelihood of crossover and mutation is reduced, and vice versa. The
optimization approach for determining the optimal placement, phase, size, and type of DG is outlined as
follows:
− Step 1: Read parameters of GA, DG types, DG number (nDG), bus data and load data
− Step 2: Run fundamental and harmonic load flow to find the initial condition of the system, including bus
voltages, network losses, IHD, THD and voltage unbalanced
− Step 3: Initialize population with size (nDG, nPopulation). Individual is represented by four number
strings: DG location, phase, type, and size
− Step 4: Determine DG size through decoding process using (13)
− Step 5: Read system data and injection of harmonic currents
− Step 6: Update system data by size, location, phases, and DG type of individual chromosome
− Step 7: Run harmonic load flow to determine voltages, network losses, harmonics and voltage unbalanced
− Step 8: Determine the fitness of all individuals using (12)
− Step 9: Verify if the termination condition is met. If it is met, go to step 14. Otherwise, go to step 10
− Step 10: Update the worst individual using the fittest individual from the population
− Step 11: Individual selection using roulette whee method
− Step 12: Exchange of parental chromosomes using adaptive crossover
− Step 13: Change a few genes in a chromosome using adaptive mutations
− Step 14: Determine the optimal location, phase, size, and type of DGs

3. RESULTS AND DISCUSSION

3.1. Data test system
The modified Kaliasin radial distribution system comprises 10 buses and 9 lines. The medium
voltage network is linked to five load buses via 20/0.380 kV transformers. The overall load connected to the
system amounts to 2.2068 MW and 1.468 MVAR. All loads on the distribution system are supplied through
bus 1. The information regarding bus data, line data, transformer data, and harmonic current injection of the
nonlinear load is taken from [20].

3.2. Before placement of DG

Load flow analysis reveals that all system bus voltages meet the standard 1±0.05 pu. The highest
voltage on bus 1 which is 20 kV (1 pu) and the lowest is on bus 10 phase a (10a) which is 19.907 kV. Active
and passive power losses are 5.352 kW and 2.556 kVar respectively. THD on all buses and phases are less
than 5% so they meet the IEEE standards. The highest THD is 4.2398% on bus 7(b). The lowest THD is
3.237% on bus 1(a). Individual harmonics of the order 7, 11, and 13 are all below 3%. Individual harmonic
order 5 in phase b of all buses exceeds 3% so it does not meet IEEE standards.

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 11-19
Int J Appl Power Eng ISSN: 2252-8792  15

3.3. Single-type DG placement

Figure 1 illustrates the optimization outcomes for the arrangement of six single-phase DG units. It is
evident from Figure 1 that each DG type exhibits a distinct best fitness value. Specifically, DG types I, II, III,
and IV possess the best fitness values of 9.4803, 8.5002, 8.2450, and 8.9737, respectively. The optimization
results highlight that DG type III demonstrates superior performance compared to the other four DG types in
terms of reducing power losses, voltage unbalance, and harmonics.

Figure 1. Single and multi-type DG convergence characteristic

Table 1 presents the optimal size and location of four types of DG at penetration rates of 13.6% and
27.27%. At 13.6% penetration rate, placement of three DG type III on bus 10(a), 10(b), 10(b) able to reduce
active and passive power loss of 22.08% and 21.93% respectively. The same outcomes are observed at a
penetration rate of 27.27%. DG type III provides the best performance in reducing power losses. DG type III
can reduce active and passive power loss by 40.30% and 40.17% respectively.

Table 1. Single-phase DG optimization result

DG Scheme Location Size Ploss Qloss Ub iHD
penetration (kW) (kVar) (%) Violation
Base case - - - 5.352 2.556 0.35400 169.76
Type I 8b, 10b, 10b 3×100 kVar 4.7471 2.2693 0.33475 2.031
Type II 10a, 10b, 10b 3×100 kW 4.3595 2.0837 0.31607 1.000
13.60% Type III 10a, 10b, 10b 3× (95 kW+31.23 kVar) 4.1704 1.9954 0.31195 1.000
Type IV 8b, 10b, 10b 3× (95 kW-31.23 kVar) 4.6891 2.2365 0.32404 1.005
Multi-type 10a, 10b, 10b 3× (95 kW+31.23 kVar) 4.1704 1.9954 0.31195 1.000
Type I 8b, 8b, 10a, 10b, 10b, 10c 6×100 kVar 4.2762 2.0461 0.31545 1.000
Type II 8b, 10a, 10b, 10c, 10a, 10b 6×3100 kVar 3.5596 1.7012 0.27826 1.000
27.27% Type III 8b, 10a, 10b, 10c, 10b, 10b 6× (95 kW+31.23 kVar) 3.1952 1.5293 0.27004 1.000
Type IV 8b, 8b, 10a, 10b, 10b, 10c 6× (95 kW-31.23 kVar) 4.1642 1.9829 0.29410 1.000
Multi-type 8b, 10a, 10b, 10c, 10b, 10b 6× (95 kW+31.23 kVar) 3.1952 1.5293 0.27004 1.000

In addition to reducing power losses, single-phase DG placement in the distribution system is also
expected to reduce the voltage unbalance significantly. The optimization results show that DG type III gives
the best performance in reducing voltage unbalance and the worst is DG type I. DG type III and type I can
reduce the voltage unbalance of 5.44% and 11.88% respectively at 13.60% penetration level. For penetration
rate of 27.27% DG type III and type I reduce voltage unbalance by 10.89% and 23.72% respectively.
Figure 2 shows that the penetration of single-phase DG significantly increases the voltage at each
bus. The largest increase is observed in phase A, while the other two phases show a similar increase.
Figures 3 and 4 display the iHD and THD before and after the DG placement respectively. In the base case,
the iHD at phase B of all buses is greater than 3%, while the other two phases are less than 3%. As for the
THD, in the base case, all buses have a THD value smaller than 5%, which means they meet the IEEE-519
standard. However, when single-phase DG is deployed, it results in an increased fundamental bus voltage,
which in turn leads to lower iHD and THD levels. All single-phase DG deployment schemes meet the IEEE-

Single-phase distributed generations for power balanced using adaptive … (Umar Umar)
16  ISSN: 2252-8792

519 standard, except when DG type I reaches a penetration level of 13.6%. At this point, there is an iHD
violation of 2.031 indicating the presence of a bus with iHD greater than 3%.

Figure 2. Bus voltage before and after placement of 6 single-phase DG

Figure 3. iHD before and after DG placement Figure 4. THD before and after DG placement

3.4. Multi-type DG placement

Multi type optimization results are shown in Table 1. Table 1 indicates that optimal placement of
multi type DG and DG type III into distribution systems has the same results regarding location, size, and
influence on the reduction power losses, voltage unbalanced and harmonics. The only difference lies in the
speed of ARCGA in reaching the optimum point. The multi-type of DG placement scheme is slower to reach
the optimum point than the single DG type scheme since the multi-type of optimization problem becomes
more complex with the inclusion of DG types in the optimization scheme.

3.5. Three phase DG optimization

Three-phase DG placement optimization is performed to compare single-phase and three-phase DG
performance in reducing voltage unbalances, power losses and harmonics. Three-phase DG optimization is
executed at the same penetration level as single-phase DG. The three-phase DG optimization results are
shown in Table 2. Table 2 shows that for a penetration level of 13.60%, all DG optimization schemes do not
meet iHD standard. The iHD violation for three-phase DG is greater than single-phase DG at the same
penetration level. In single-phase DG scheme, only type I and type IV schemes do not meet the standard. The
iHD violation values for type I and type IV are 2.031 and 1.005. Unbalanced voltage and power losses for the
three-phase DG scheme are slightly lower than the single-phase DG scheme. This difference is due to the
priority level of each objective function. In this optimization scheme, harmonics occupy the main priority,
then unbalanced voltage, voltage profile, and power losses. For penetration levels 27.27%, single-phase DG
provides better results in reducing harmonics, voltage unbalance and losses than three-phase DG scheme.

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 11-19
Int J Appl Power Eng ISSN: 2252-8792  17

Table 2. Three phase DG optimization result

DG Ploss Qloss Ub iHD
Scheme Location Size
Penetration (kW) (kVar) (%) Violation
Base case - - - 5.3520 2.556 0.35400 169.762
Type I 10 300 kVar 4.7171 2.2571 0.33470 29.354
Type II 10 300 kW 4.3965 2.102 0.31607 10.699
13.60% Type III 10 285 kW+93 kVar 4.2231 2.021 0.31195 6.588
Type IV 10 285 kW-93 kVar 4.6741 2.2323 0.32398 23.285
Multi-type 10 285kW+93 kVar 4.2231 2.021 0.31195 6.588
Type I 8, 10 2×300 kVar 4.2426 2.0322 0.31556 6.595
Type II 8, 10 2×300 kW 3.5879 1.715.3 0.27840 1.000
27.27% Type III 8, 10 2×(285 kW+93.69 kVar) 3.2458 1.5556 0.27020 1.000
Type IV 8, 10 2×(285 kW-93.69 kVar) 4.1384 1.9735 0.29420 4.056
Multi-type 8, 10 2×(285 kW+93.69 kVar) 3.2458 1.5556 0.27020 1.000

4. CONCLUSION
ARCGA is used in this paper to determine the location, phase, size, and type of four single-phase
DG types in the distribution system. The simulation results show that the placement of 6 single-phase DG
type III provides the best performance in reducing voltage unbalances, power losses and harmonics compared
to the other three DG types. The multi-type of single-phase DG optimization provides the same result as DG
type III scheme. Single-phase DG optimization also performs better in reducing harmonics, voltage
imbalances and power losses than the three-phase DG scheme at the same penetration level.

ACKNOWLEDGEMENTS
The author expresses gratitude to the anonymous reviewers for their valuable time and insightful
suggestions that have contributed to enhancing the quality of this paper. Additionally, the author extends
thanks to the editorial team of IJAPE for their support and permission to publish this article.

REFERENCES
[1] A. Colmenar-Santos, C. Reino-Rio, D. Borge-Diez, and E. Collado-Fernández, “Distributed generation: A review of factors that
can contribute most to achieve a scenario of DG units embedded in the new distribution networks,” Renewable and Sustainable
Energy Reviews, vol. 59, pp. 1130–1148, 2016, doi: 10.1016/j.rser.2016.01.023.
[2] Umar, Firdaus, A. Soeprijanto, and O. Penangsang, “Optimal expenditure and benefit cost based location, size and type of DGs in
microgrids systems using adaptive real coded genetic algorithm,” Telkomnika (Telecommunication Computing Electronics and
Control), vol. 16, no. 1, pp. 10–17, 2018, doi: 10.12928/telkomnika.v16i1.6761.
[3] Firdaus, Umar, R. S. Wibowo, and O. Penangsang, “Optimal Placement of Intermittent Dg Renewable Energy and Capacitor
Bank for Power Losses Reduction and Voltage Profile Improvement in Microgrids Systems,” Journal of Engineering Science and
Technology, vol. 17, no. 5, pp. 3539–3554, 2022.
[4] Rekha and S. C. Byalihal, “Optimal allocation of solar and wind distributed generation using particle swarm optimization
technique,” International Journal of Electrical and Computer Engineering, vol. 13, no. 1, pp. 229–237, 2023, doi:
10.11591/ijece.v13i1.pp229-237.
[5] F. Faanzir, M. Ashari, S. Soedibyo, S. Suwito, and U. Umar, “The Design of DC Micro Grid with a Load-Based Battery
Discharge Method for Remote Island Electrification Utilizes Marine Currents and Solar Photovoltaic,” Kinetik: Game
Technology, Information System, Computer Network, Computing, Electronics, and Control, vol. 7, no. 4, pp. 409–420, 2022, doi:
10.22219/kinetik.v7i4.1576.
[6] P. U. Reddy, S. Sivanagaraju, and P. Sangameswararaju, “Power Flow Analysis of Three Phase Unbalanced Radial Distribution
System,” International Journal of Advances in Engineering & Technology, vol. 514, pp. 514–524, 2012.
[7] R. Yan and T. K. Saha, “Investigation of voltage imbalance due to distribution network unbalanced line configurations and load
levels,” IEEE Transactions on Power Systems, vol. 28, no. 2, pp. 1829–1838, 2013, doi: 10.1109/TPWRS.2012.2225849.
[8] N. Hayat, S. R. Ghatak, S. R. Pani, and S. Mishra, “Performance Improvement of Three Phase Unbalance Distribution System
Using DG and DSTATCOM,” in 2020 IEEE Calcutta Conference, CALCON 2020 - Proceedings, 2020, pp. 200–205, doi:
10.1109/CALCON49167.2020.9106473.
[9] Y. Yoon and S. Choi, “Application of Reinforcement Learning to Operate the Unbalanced Distribution System,” in International
Conference on ICT Convergence, 2022, pp. 347–349, doi: 10.1109/ICTC55196.2022.9953017.
[10] T. Wang, Y. Wang, Y. Hou, F. Gu, S. Hou, and W. Jin, “Optimization of Unbalanced Active Distribution Systems Using Cuckoo
Search Algorithm,” in 2021 9th IEEE International Conference on Smart Energy Grid Engineering, SEGE 2021, 2021, pp. 129–
133, doi: 10.1109/SEGE52446.2021.9534957.
[11] R. Bonela, S. R. Ghatak, N. Hayat, S. C. Swain, and A. Mohapatra, “Optimal planning framework of three phase unbalanced
distribution system using multiple DG and DSTACOM under different loading condition,” 7th IEEE Uttar Pradesh Section
International Conference on Electrical, Electronics and Computer Engineering, UPCON 2020, 2020, doi:
10.1109/UPCON50219.2020.9376508.
[12] L. Luan, H. Peng, W. Mo, Y. Wang, and Z. Xu, “Location Method for Three Phase Unbalance Sources in Distribution Network
Based on CSO Optimized Deep Belief Network,” in Proceedings - 2021 International Conference on Power System Technology:
Carbon Neutrality and New Type of Power System, POWERCON 2021, 2021, pp. 1986–1990, doi:
10.1109/POWERCON53785.2021.9697787.

Single-phase distributed generations for power balanced using adaptive … (Umar Umar)
18  ISSN: 2252-8792

[13] M. A. A. Al-Ja’Afreh and G. Mokryani, “Voltage unbalance mitigation in low voltage distribution networks using time series
three-phase optimal power flow,” 2021 56th International Universities Power Engineering Conference: Powering Net Zero
Emissions, UPEC 2021 - Proceedings, 2021, doi: 10.1109/UPEC50034.2021.9548248.
[14] R. Bonela, S. R. Ghatak, N. Hayat, S. C. Swain, and A. Mohapatra, “Techno-Economic planning framework of three phase
unbalanced distribution system using multiple DG and Capacitor,” 1st Odisha International Conference on Electrical Power
Engineering, Communication and Computing Technology, ODICON 2021, pp. 1–6, 2021, doi:
10.1109/ODICON50556.2021.9428933.
[15] G. Setyawan, Umar, A. Soeprijanto, and O. Penangsang, “Symbiotic Organism Search for Sizing and Optimal Location of
Distributed Generation Using Novel Sensitivity Factor,” in IOP Conference Series: Materials Science and Engineering, 2019,
vol. 588, no. 1, doi: 10.1088/1757-899X/588/1/012021.
[16] A. Kalair, N. Abas, A. R. Kalair, Z. Saleem, and N. Khan, “Review of harmonic analysis, modeling and mitigation techniques,”
Renewable and Sustainable Energy Reviews, vol. 78, pp. 1152–1187, 2017, doi: 10.1016/j.rser.2017.04.121.
[17] M. J. Ghorbani and H. Mokhtari, “Impact of harmonics on power quality and losses in power distribution systems,” International
Journal of Electrical and Computer Engineering, vol. 5, no. 1, pp. 166–174, 2015, doi: 10.11591/ijece.v5i1.pp166-174.
[18] Z. Hu, Y. Han, A. S. Zalhaf, S. Zhou, E. Zhao, and P. Yang, “Harmonic Sources Modeling and Characterization in Modern Power
Systems: A Comprehensive Overview,” Electric Power Systems Research, vol. 218, 2023, doi: 10.1016/j.epsr.2023.109234.
[19] A. A. Abdelsalam, A. A. Zidan, and E. F. El-Saadany, “Optimal DG Allocation in Radial Distribution Systems with High
Penetration of Non-linear Loads,” Electric Power Components and Systems, vol. 43, no. 13, pp. 1487–1497, 2015, doi:
10.1080/15325008.2015.1043601.
[20] Umar, Firdaus, and O. Penangsang, “Simultaneously placement and sizing of multiple DGs and shunt capacitor banks in
unbalanced distribution systems using real coded GA,” in Proceeding - 2016 International Seminar on Intelligent Technology and
Its Application, ISITIA 2016: Recent Trends in Intelligent Computational Technologies for Sustainable Energy, 2017, pp. 487–
492, doi: 10.1109/ISITIA.2016.7828708.
[21] K. Muthukumar and S. Jayalalitha, “Harmony search approach for optimal capacitor placement and sizing in unbalanced
distribution systems with harmonics consideration,” in IEEE-International Conference on Advances in Engineering, Science and
Management, ICAESM-2012, 2012, pp. 393–398.
[22] N. Ghaffarzadeh and H. Sadeghi, “A new efficient BBO based method for simultaneous placement of inverter-based DG units and
capacitors considering harmonic limits,” International Journal of Electrical Power and Energy Systems, vol. 80, pp. 37–45, 2016,
doi: 10.1016/j.ijepes.2016.01.030.
[23] M. Kumawat, N. Gupta, N. Jain, and R. C. Bansal, “Optimal planning of distributed energy resources in harmonics polluted
distribution system,” Swarm and Evolutionary Computation, vol. 39, pp. 99–113, 2018, doi: 10.1016/j.swevo.2017.09.005.
[24] Amirullah, M. Ashari, O. Penangsan, and A. Soeprijanto, “Multi units of single phase distributed generation combined with
battery energy storage for phase power balancing in distribution network,” Jurnal Teknologi, vol. 78, no. 10–4, pp. 27–33, 2016,
doi: 10.11113/jt.v78.9887.
[25] W. Pinthurat and B. Hredzak, “Strategy for compensation of unbalanced powers in LV residential distribution networks using
distributed single-phase battery systems,” Electric Power Systems Research, vol. 211, 2022, doi: 10.1016/j.epsr.2022.108253.
[26] W. Jiao, Q. Wu, S. Huang, J. Chen, C. Li, and B. Zhou, “DMPC based distributed voltage control for unbalanced distribution
networks with single-/three-phase DGs,” International Journal of Electrical Power and Energy Systems, vol. 150, 2023, doi:
10.1016/j.ijepes.2023.109068.
[27] W. M. Grady and S. Santoso, “Understanding Power System Hannonics,” IEEE Power Engineering Review, vol. 21, no. 11, pp.
8–11, Nov. 2001, doi: 10.1109/MPER.2001.961997.

BIOGRAPHIES OF AUTHORS

Umar Umar is currently works as an associate professor in The Electrical

Engineering Department of Universitas Khairun, Indonesia. He completed his undergraduate
program in electrical engineering at Universitas Hasanuddin in 2000. He obtained his master’s
and doctoral degrees from the Electrical Engineering department of Institut Teknologi Sepuluh
Nopember, Indonesia in 2008 and 2022 respectively. His main research areas include artificial
intelligence, renewable energy, distribution automation, and power quality. He can be
contacted at email: [email protected].

Faanzir Faanzir is currently works as an Associate Professor in The Electrical

Engineering Department of Universitas Khairun, Indonesia. He completed his undergraduate
program at Universitas Gadjah Mada and master’s degree in Electrical Engineering at
Universitas Hasanuddin. He obtained his doctoral degree from the Electrical Engineering
department of Institut Teknologi Sepuluh Nopember, Indonesia in 2023. His main research
areas include renewable energy, marine current, and electrical materials. He can be contacted
at email: [email protected].

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 11-19
Int J Appl Power Eng ISSN: 2252-8792  19

Iswan Iswan has been a lecturer at Khairun University since 2006. He completed
his doctorate at the Department of Electrical Engineering, University of Indonesia in 2021.
Previously he completed undergraduate and master programs in Makassar in the Department
of Electrical Engineering. His current research scope is energy management, distribution
systems, and load forecasting. He can be contacted at email: [email protected].

Ramly Rasyid is currently working as an Associate Professor in the Electrical

Engineering Department at Khairun University, Indonesia. He completed his undergraduate
education at Universitas Muslim Indonesia, and a magister’s degree from Institut Teknologi
Sepuluh Nopember, Indonesia. During his career as a lecturer, he has held several important
positions at Khairun University, such as Dean of the Faculty of Engineering and Vice
Chancellor for student affairs. His main research includes electric machines and electric drives
controller. He can be contacted at email: [email protected].

Muhammad Natsir Rahman is currently works as an Assistant Professor in The

Electrical Engineering Department, Khairun University, Indonesia. He has completed his
undergraduate and master’s degree in electrical engineering at Hasanuddin University and
Institute of Technology Bandung, both in Indonesia, in 1999 and 2004, respectively. Now, he
pursues his doctor degree in Kyushu University, Japan. His main research areas include
renewable energy, optimization, and power quality. He can be contacted at email:
[email protected].

Amal Khairan presently holds the position of an assistant professor in the

informatics department at Universitas Khairun, Indonesia. He successfully completed his
undergraduate program in Electrical Engineering at Universitas Hasanuddin in 1999. In 2009,
He acquired his master's degrees from the Electrical Engineering Department of Gadjah Mada
University, Indonesia. Amal's primary research focus revolves around computer networks and
database. He can be contacted at email: [email protected].

Haryati Haryati is presently employed as a lecturer in Engineering Faculty at

Universitas Khairun Indonesia. She accomplished her bachelor of engineering degree from
Makassar State University in 2013 and completed her master’s degree at Institut Teknologi
Sepuluh Nopember, Indonesia, in 2016. Haryati's research concentrates on urban development
and the restoration of cultural heritage sites. She can be contacted at email:
[email protected].

Single-phase distributed generations for power balanced using adaptive … (Umar Umar)