INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Reliability Assessment of Major Feeders of the Atoabo Substation,

Tarkwa Using Autorecloser-Based ETAP Simulation

J. K. Annan^1*and J. M. Yevunya²

¹University of Mines and Technology, Tarkwa, Ghana

²Electricity Company of Ghana, Ahomaso, Kumasi, Ghana

DOI: https://dx.doi.org/10.51584/IJRIAS.2025.101100092

Received: 30 November 2025; Accepted: 06 December 2025; Published: 19 December 2025

ABSTRACT

Reliable Medium Voltage (MV) distribution networks are critical to economic activities in Ghana, particularly

in mining-intensive municipalities such as Tarkwa, where prolonged outages impose substantial operational

losses. This study evaluates the reliability performance of the 11 kV Town 1, Town 2 and Manganese feeders

supplied from the Atoabo Bulk Supply Point (BSP), a strategically important node feeding high-value industrial

and residential loads. Six years of outage data (2016–2021) were analysed and used to calibrate an ETAP

probabilistic reliability model, addressing the absence of simulation-based reliability evaluation and automation-

planning studies for Ghanaian MV distribution networks. The calibrated ETAP model replicated historical SAIFI

and SAIDI values within ±5–10%, confirming strong model fidelity. Simulation results show that ACR

deployment yields significant reliability improvement at SAIFI reduction of 35–40% on Town 2, SAIDI

reduction of 32–35% on Manganese feeder, and overall reliability improvement of 25–30% on Town 1. The

findings demonstrate that targeted MV automation at Atoabo BSP provides a cost-effective and high-impact

reliability intervention, capable of reducing cumulative annual customer interruption duration by over 100 hours

per feeder. This work provides an investable pathway for Electricity Company of Ghana to achieve Public

Utilities and Regulatory Commission’s reliability benchmarks in similar radial distribution environments.

Keywords: Reliability Analysis, Distribution Automation, Automatic Circuit Recloser (ACR), ETAP

Simulation, Power Distribution Network, System Interruption Indices, Tarkwa Municipality.

INTRODUCTION

Electric power reliability remains a key determinant of socio-economic development in many developing

economies, where interruptions in distribution networks impose large costs on industry, households and public

services. In Ghana, distribution-level outages are a major contributor to unreliable supply and elevated customer

interruption indices (SAIDI, SAIFI), particularly in mining municipalities where continuous power is critical for

operations and safety. The present study uses six years of outage data from the Atuabo Bulk Supply Point

(Atoabo BSP) to evaluate reliability of the 11 kV Town 1, Town 2 and Manganese feeders and to test the

reliability benefit of targeted Distribution Automation (DA) interventions using ETAP simulations.

Recent research indicates that distribution automation which includes automatic circuit reclosers (ACRs),

sectionalizers, remote fault indicators and SCADA-enabled feeder control can materially reduce outage

frequency and duration when deployed in contexts with high rates of transient and vegetation-related faults. Case

studies and modelling work from the region demonstrate measurable SAIFI/SAIDI improvements following the

installation of reclosers and sectionalizers, and highlight that even modest automation investments can yield

substantial reliability and commercial benefits in utilities with limited operational staff and manual switching

processes.

However, the distribution automation literature also emphasises that the performance gains and cost-

effectiveness of DA are highly context dependent: feeder topology, fault mix (transient versus permanent),

communication infrastructure, and existing switching points determine where devices like ACRs produce the

Page 971

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

greatest benefit. Comparative studies from Nigeria and South Africa suggest that DA implementations should

be guided by localized reliability data and simulation-based evaluation before large-scale rollout.

This study therefore integrates historical outage analysis for the Atoabo BSP (2016–2021) (ECG, 2022) with

ETAP-based simulation experiments to quantify the reliability impact of selected automation schemes (ACRs

and sectionalizers) on the Atoabo 11 kV feeders. The objective is to provide an evidence-based, simulation-

driven justification for prioritised automation investments in the Tarkwa Municipality that are aligned with

PURC benchmarks and local operational constraints.

Reliability Challenges in Ghana’s Distribution Networks

Ghana’s national studies on distribution performance provide a useful backdrop, but reliability challenges vary

substantially between metropolitan and mining municipalities. Tarkwa’s Atoabo BSP supplies several large

mining customers and dense residential clusters via three principal 11 kV feeders (Town 1, Town 2 and

Manganese). Field data from ECG’s Atoabo records show elevated SAIFI and SAIDI on these feeders (average

SAIFI and SAIDI over 2016–2021 for the Manganese feeder are 84 and 88.9 hr/customer·yr respectively), and

the network operates without SCADA or automated feeder devices. These locally observed characteristics

including high transient-fault share, long restoration times and absence of feeder automation motivate a targeted,

simulation-based evaluation of ACR and sectionalizer placements at Atoabo rather than relying on generic

country-level prescriptions.

Technological Interventions for Reliability Improvement

Modern reliability enhancement strategies are increasingly centered on automation and smart grid technologies.

Devices such as Automatic Circuit Reclosers (ACRs), sectionalizers, and Fault Path Indicators (FPIs) enable

remote fault detection, automatic isolation, and rapid service restoration (Elkadeem et al., 2017).

Autoreclosers, in particular, have become vital components of self-healing distribution networks. They

automatically open during a fault, test the circuit, and reclose after a short delay, thereby restoring supply if the

fault was transient (Rones and Vittal, 2013). Studies by Mehdi et al. (2021) and Singh (2017) demonstrated that

autoreclosers can reduce outage duration by up to 60% and prevent cascading failures in radial systems. In the

Ghanaian context, the integration of autoreclosers at strategic points in the network has shown potential to

significantly improve reliability and operational efficiency.

Simulation software such as ETAP (Electrical Transient Analyzer Program) has become a standard tool for

analyzing, modeling, and improving power system reliability (Idowu et al., 2021). The Reliability Assessment

Module (RAM) of ETAP allows engineers to compute probabilistic reliability indices, perform contingency

analysis, and evaluate the effect of equipment upgrades like autoreclosers or sectionalizers.

LITERATURE OVERVIEW

Overview of Electric Power Systems

An electric power system, which consists of three interdependent subsystems namely generation, transmission,

and distribution, is aimed at ensuring reliable and economical energy delivery to end-users (Kumar et al., 2018).

While generation and transmission systems are generally well protected and redundant, the distribution network

is more vulnerable to faults because of its extensive exposure to environmental, mechanical, and human

interferences (Chandhra et al., 2017).

Distribution systems typically operate at medium voltages (11 kV – 33 kV) for bulk supply and low voltages

(400/230 V) for end-user consumption. They are mostly radial in topology, making them highly susceptible to

single-point failures. As a result, more than 80% of customer outages originate at the distribution level (Ghiasi

et al., 2019).

Page 972

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Power System Reliability Concepts and Reliability Indices

Reliability in electrical systems refers to the probability that a power network will perform its intended function

without interruption for a specified period under defined conditions (IEEE Power & Energy Society, 2012).

Power system reliability is generally divided into two conditions:

i.

The system’s ability to meet load demands under normal operating conditions; and

ii. The ability of the system to withstand sudden disturbances such as faults, line outages, or equipment failures

(Kumar et al., 2018).

Distribution Automation in Developing Economies

Recent studies have examined Distribution Automation (DA) in low and middle income settings and reported

the following consistent findings:

i.

Simulation and field studies indicate that well-placed ACRs and sectionalizers produce significant

reductions in SAIFI and SAIDI on radial feeders dominated by transient faults. ETAP and similar platforms

are commonly used to quantify expected gains prior to deployment;

ii. The magnitude of improvement depends on feeder topology, fault type composition (momentary vs

permanent), and communication or maintenance capability; and

iii. In many utilities in Sub-Saharan Africa, limited SCADA, weak communications and constrained operations

and maintenance budgets slow adoption despite clear technical benefits.

Comparative Regional Studies

In Nigeria, several utility-level and academic studies report reliability improvements after selective automation

and protection upgrades in Nigerian distribution networks. These works highlight the practical benefits of

reclosers and sectionalizers in networks where vegetation and temporary faults are dominant causes. They also

underscore the need for communication and maintenance planning to ensure devices remain effective (Akande

et al., 2021). In South Africa, Eskom-oriented research and municipal studies emphasise automation integrated

with network reconfiguration and improved switching practices; South African work typically stresses a

combined techno-economic approach (which included automation, network reconfiguration, and targeted

maintenance) to maximise benefit under constrained budgets (Gumede and Saha, 2022).

Although regional studies document the technical potential of distribution automation, there is a lack of site-

specific, simulation-based assessments for the Atoabo feeders that combine the actual ECG outage record (2016–

2021), realistic device models and placement optimisation using ETAP. Existing Ghanaian reliability work has

largely been descriptive or pilot-scale; it has not produced a combined historical-plus-simulation analysis

tailored to a mining-intensive feeder set such as Manganese/Town 1/Town 2 at Atoabo. In short: a simulation-

based assessment of automation solutions for Atoabo feeders (quantifying SAIFI/SAIDI/EENS reductions and

energy savings under realistic device placements and coordination settings) is missing.

Many case studies report substantial benefits from ACRs but also emphasise that field performance depends on

the proportion of temporary or transient faults, correct coordination and setting of shots/dead-time and

communication and maintenance to prevent device misoperation (Mandefro and Mabrahtu, 2020; Gumede and

Saha, 2022). These lessons map directly to Atoabo where the historical data indicate ~49% transient faults.

Resources and Methods Used

Objectives, Data Review and Methods

The objectives of this research are to quantify baseline reliability of the Atoabo BSP 11 kV feeders (Town 1,

Town 2 and Manganese) from ECG outage records (2016–2021) and benchmark against PURC standards, and

Page 973

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

then evaluate the reliability impact of targeted distribution automation (automatic circuit reclosers and

sectionalizers) using ETAP-based simulation scenarios and compute expected changes in SAIFI, SAIDI, CAIDI

and EENS.

Operational data were obtained from the Electricity Company of Ghana (ECG) Regional Office in Takoradi and

the Tarkwa District Office. The dataset covered a six-year period (2016–2021) and included feeder interruption

frequencies, outage durations, causes, restoration times, load profiles, and network topology. Field visits were

conducted to the Atuabo Bulk Supply Point (BSP) to verify network parameters and record equipment

specifications, including transformer ratings, conductor sizes, and circuit breaker types. Additional data were

gathered on customer population, installed capacity, and load demand patterns.

The study focused on the 11 kV distribution feeders made up of Town 1, Town 2, and Manganese feeders

supplied from the Atuabo BSP in the Tarkwa Municipality of Ghana’s Western Region. The BSP receives 33

kV supply from GRIDCo and steps it down to 11 kV through a 20 MVA, 33/11 kV transformer. The 11 kV

network operates on a radial configuration using 120 mm² aluminum conductors and serves approximately

21,500 customers via 117 distribution transformers with a total installed capacity of 22.67 MVA. Circuit

protection is provided by Vacuum Circuit Breakers (VCBs) on the 11 kV side and SF₆ circuit breakers on the 33

kV side. The system currently operates without SCADA or automation facilities.

Reliability analysis was conducted using the historical assessment method, employing standard indices such as

System Average Interruption Frequency Index (SAIFI), System Average Interruption Duration Index (SAIDI),

Customer Average Interruption Duration Index (CAIDI), Average System Availability Index (ASAI), Expected

Energy Not Served (EENS), and Average Energy Not Served (AENS). These indices were computed based on

outage data using Equations (1)–(8) defined in IEEE Std. 1366 (2012). The computed values were benchmarked

against the PURC thresholds (SAIDI ≤ 48 hr/customer·yr; SAIFI ≤ 6 interruptions/customer·yr).

Outages were categorized into planned and unplanned types. Planned interruptions involved scheduled

maintenance or installations, while unplanned outages resulted from faults such as transient disturbances,

lightning strikes, equipment failures, and conductor breakages. Fault classification followed standard

distribution fault categories including single line-to-ground, line-to-line, double line-to-ground, and three-phase

faults, as summarized by Gupta (2019) and Grainger and Stevenson (2016).

Rationale for Data Timeframe (2016–2021)

The six-year dataset (2016–2021) was selected because it represents the most complete, consistent and validated

outage record available in the Atoabo BSP logbooks, covering both pre- and post-network modification periods.

The timeframe includes periods of abnormal system stress (notably 2016), transitional operational improvements

(2017–2018), and relative supply stability (2019–2021), offering a statistically representative window of the

feeder behaviour.

Although more recent data (2022–2023) are desirable, these years exhibited incomplete logging due to COVID-

19 staffing constraints and SCADA-less manual reporting inconsistencies, issues reported for several

developing-economy utilities (Ogunjuyigbe et al., 2021; Kahimba and Mbuli, 2022). The 2016–2021 series

therefore provides the most reliable and continuous basis for trend computation, index averaging, and ETAP

model calibration.

Reliability Indices in Distribution Systems

Reliability indices are standardised metrics that help utilities quantify outage performance, compare feeder

conditions, and monitor compliance with regulatory benchmarks (Manandhar, 2013). Common indices are

presented in Table 1.

Page 974

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Table 1 Summary of Reliability Indices

Rel. Index Definition

Description

Unit of Measure

SAIFI

System Average Interruption Mean number of sustained interruptions a fr/cust.yr

Frequency Index

customer experiences at a predefined period

in the system.

SAIDI

CAIDI

MAIFI

ASAI

System Average Interruption Total duration of interruption for the average hr/cus.yr

Duration Index

customer during a predefined period.

Customer

Average The average time required to restore service.

hr/cust.

Interruption Duration Index

interruption

Momentary

Average This is a measure of the average frequency of fr/cust. yr

Interruption Duration Index

momentary interruptions.

Average System Availability

Index

Fraction of time that a customer has received Per unit

power during reporting period.

EENS

ACCI

Expected Energy Not Served

Mean energy not supplied per customer per MWh/yr

year.

Average

Customer kWh of connected load interrupted for each kVA/ customer

affected customer in one year.

Curtailment Index

ECOST

IEAR

Expected Interruption Cost Cost of energy not supplied at load point.

Index

$/yr

Interrupted

Energy This is the cost per unit of unserved energy.

$/kWh

Assessment Rate

Utilities use the indices of Table 1 to assess feeder-level performance and prioritise reliability investments. The

PURC of Ghana stipulates benchmark values of SAIDI ≤ 48 hr/customer·yr and SAIFI ≤ 6

interruptions/customer·yr (PURC, 2022).

Equation 1 through to Equation 8 give the mathematical definitions of the most frequently used reliability indices

mentioned in Table 1.

푛

푖

∑

휆 푁

푖

푆퐴퐼퐹퐼 =

(1)

푁

푡

where,

n = total number of load points

λ_i= avg. failure rate of each segment i (f/yr)

N_i= number of customers interrupted

N_t= total number of customers served

푛

∑

푈 푁

푖

푆퐴퐼퐷퐼 =

(2)

푁

푡

Page 975

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

where ꢀ_ꢁis the average annual outage time at load point i (f/yr).

푛

푖

∑

푈 푁

ꢂꢃꢄꢅꢄ

ꢂꢃꢄꢆꢄ

푖

퐶퐴퐼퐷퐼 =

푀퐴퐼퐹퐼 =

=

(3)

(4)

푛

푖

∑

휆 푁

푖

푛

푖

∑

휆 푁

푖

푁

푖

ASAI is the ratio of customer hours of service availability to customer hours of service demand.

푛

∑

8760푁 −

푟 푁

푖 푖

1−ꢂꢃꢄꢅꢄ

푡

푖

퐴푆퐴퐼 =

=

(5)

8760푁

8760

푡

where ꢇ = ꢈ표ꢈ푎푙 ꢇ푒푠ꢈ표ꢇ푎ꢈꢉ표ꢊ ꢈꢉ푚푒.The number 8760 represents the number of hours in a regular year.

ꢁ

퐴푆ꢀ퐼 = 1 − 퐴푆퐴퐼

(6)

퐸퐸ꢋ푆 = ^ꢌ푃 ꢀ_ꢁ

∑

(7)

ꢁ

EENS is basically interpreted as a product of the average load and the output duration.

ꢍꢍ푁ꢂ

퐴퐸ꢋ푆 =

(8)

푛

∑

푁

푖

AENS represents the ratio of the expected energy not served to the total number of customers served.

Faults in Distribution Systems

Faults in electrical distribution systems are unintended deviations from normal operating conditions that cause

abnormal currents or voltages in power circuits. They can result from equipment failure, insulation breakdown,

weather conditions, mechanical damage, or human error. Distribution systems, being the final stage of power

delivery to consumers, are particularly vulnerable to such disturbances due to their wide geographical spread

and exposure to environmental factors.

Understanding the nature, causes, and effects of various faults is essential for the design of reliable protection

schemes, ensuring continuity of supply, safety of personnel, and longevity of equipment. Table 2 summarises

the major types of faults that occur in electrical distribution networks, outlining their causes, effects, and common

protection or mitigation measures employed in modern power systems.

Table 2 Summary of Faults in Electrical Distribution Systems

Type of Fault Meaning

Typical

Causes

Effects on System

Method

Detect

to Ways

Mitigate

to

One

Line-

phase Insulation

Unbalanced

currents,

dips, over-voltages residual

Earth

voltage relays,

fault Earth fault relay

+ circuit breaker

Single

conductor touches breakdown,

earth or grounded lightning,

to-Ground

(L–G)

surface

moisture,

pollution

on healthy phases

current

detection

Two

conductors

into contact

phase Tree branches, High fault current, Over-current

Over-current

or differential protection, phase

relay segregation

Line-to-Line

(L–L)

come conductor

swing,

unbalanced

voltages,

damaged

overheating

insulation

Page 976

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Two

phase Flashovers,

Severe current flow Earth fault and Ground

fault

Double Line-

to-Ground

(L–L–G)

conductors contact heavy

each other and the lightning,

through

path,

ground over-current

voltage relays

protection, surge

arresters

ground

mechanical

failure

unbalance

All

three Severe

Very high current, Over-current

High-speed

circuit breakers,

fuses

Three-Phase

(Balanced)

Fault

L)

conductors shorted insulation

maximum

relays,

together

(L–L–

failure,

equipment

breakdown

mechanical stress

differential

relays

All three phases Major

shorted to ground equipment

Extremely

fault current, system and

collapse

high Difference

Auto- reclosing

Three-Phase-

to-Ground

(L–L–L–G)

over- breakers,

current relays clearing

fast

same time

failure,

lightning

One

or

more Mechanical

Unbalanced

Under-voltage Reclosing

single or negative scheme,

open phasing in motors sequence maintenance

relays inspection

Open Circuit

Fault

conductors broken damage, broken voltages,

(no current flow)

lines,

switches

Conductor touches Fallen

poor conductor conductors,

such as asphalt or degraded

Low but dangerous Special

fault current, fire detection

risk

HIF Arc

detectors,

algorithms isolation relays

fault

High

Impedance

Fault (HIF)

dry soil

insulation

Electric

between

conductors or to damaged

ground insulation

arc Loose

connections,

Voltage

equipment

degradation and fire relays,

harmonic

analysis

flicker, Arc

detection

fault Arc

restraint

Arcing Fault

coils, insulation

care

(Sources: Gupta, 2019; Grainger and Stevenson, 2016)

Atuabo Bulk Supply Point

The Atuabo Bulk Supply Point (BSP) is a substation located in the Tarkwa Municipality of the Western Region

in the Republic of Ghana. It is one of the substations in the Region where ECG takes bulk of its electric power

supply to serve customers in Tarkwa, Abosso, Prestea, Damang and surrounding communities. Mining

companies such as Goldfields Ghana Limited, Ghana Manganese Company and African Mining Services as well

as institutions like UMaT, Fiaseman and Tarkwa Senior High Schools also take supply from the Substation.

There are four feeders from the national grid to the substation. These feeders supply four transformers namely

9T1, 9T2, 9T3, and 9T4 with voltage levels of 161 kV. Each feeder is connected to a 26/33 MVA, 161/33 kV,

step-down transformer at the GRIDCo side of the system.

The 33 kV incoming feeders from GRIDCo is connected to the ECG busbar through Sulphur hexafluoride (SF6)

gas circuit breakers and isolators. Potential Transformers (PTs), Current Transformers (CTs), relays, and other

auxiliary equipment are used to design the protection and metering systems. ECG again steps down the voltage

through a 20 MVA, 33/11 kV transformer for distribution.

There are three outgoing 11 kV feeders and three outgoing 33 kV feeders. These are Town 1, Town 2 and

Manganese for the 11 kV lines and Bonsa, Abosso 1 and Aboso 2 for the 33 kV feeders.

Page 977

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

The busbar arrangement at the Atoabo substation is a sectionalized single busbar system. In this arrangement a

circuit breaker and isolating switches are used to sectionalize the bus. This arrangement enables maintenance to

be carried out on one part of the system without a complete shutdown of the entire station. Fig. 1 shows a single

line diagram of the Atoabo Substation modelled with ETAP 19.0 software.

Fig. 1 Busbar Arrangement at Atoabo Substation

The 33 kV busbars are outdoor, and the 11 kV busbars are indoor. Monitoring and operation are carried out

through indoor panels for both voltage levels. There is no SCADA facility at the ECG side of the substation.

Circuit Breakers and Other Infrastructure

Two types of circuit breakers are employed at the Atoabo substation: Gas Insulated Switches (GIS) containing

sulfur hexafluoride (SF₆) circuit breaker for the 33 kV system and Vacuum Circuit Breaker (VCB) for the 11 kV

system. Fig. 2 shows the outdoor SF₆circuit breakers at the substation. Fig. 2 and Fig. 3 show the indoor panels

of both 33 kV and 11 kV system while Fig. 4 indicate the rectifier and batteries used in the substation.

Fig. 2 Atoabo Substation 33 kV Indoor Panels

Fig. 3 Atoabo Substation 11 kV Indoor Panels

Page 978

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Fig. 4 Atoabo Substation Rectifier and Batteries

11 kV Distribution Network

The 11 kV distribution system has three outgoing feeders which supply power to communities such as

Brahabebome, Bankyekrom, Efuanta, Tamso, Senyakrom, Kwabedu and other communities in the Tarkwa

Municipality. There are two isolators, five Ring Main Units (RMU) and six Extensible Oil Switches (EOS) at

different locations. These switches are not automated. There are Normally Open Points (NOP) between the

feeders for purpose of transferring load from one feeder to the other and to isolate portions of the network for

maintenance activities. The network has a total of 117 distribution transformers with an installed capacity of

22,670 kVA.

The transformers are mostly pole mounted (PMTs) and few ground mounted (GMTs). The average current

recorded on the 11 kV feeders were as follows:

i.

ii.

Town 1 Feeder: 342 A;

Town 2 Feeder: 302 A; and

Manganese Feeder: 421 A.

iii.

The network was constructed with 120 mm²bare aluminum conductor (with a cross-sectional area of 120 mm²

supported on 11-meter wooden and concrete poles. The feeders are predominantly overhead lines (OHL),

however portions are constructed with 3×185 mm²Cross Linked Polyethylene (XLPE) aluminum underground

cable. Table 3 presents data for feeders and types of related circuit breakers for both 11 kV and 33 kV outgoing

feeders.

Table 3 Name of Feeders and Type of Circuit Breakers at Atoabo Substation

SN

1

Feeder Name

Town 1

Voltage (kV) Type of CB

Average Current (A) Circuit Length (km)

11

33

VCB

SF₆

342

302

421

122

35

33.42

45.24

41.67

67.43

8.75

2

Town 2

3

Manganese

Bonsa

4

5

Aboso 1

Aboso 2

SF₆

6

SF₆

35

8.75

Page 979

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

The number of distribution transformers, customer population, installed capacity (kVA), and conductor size on

the 11 kV feeders are shown in Table 4.

Table 4 Distribution Transformers and Customer Population on 11 kV Feeders

Feeder

No. of Distribution No.

of Total

Rated Size

of

Aluminum

Transformers

Customers

Capacity (kVA)

Conductor (mm²)

Town 1

Town 2

Manganese

Total

54

7 876

6 917

6 710

21 503

8 230

5 875

8 565

22 670

120

24

39

117

Table 5 indicates the parameters of the step-down power transformer at the Atoabo substation. The transformer

is delta-wye connected with the neutral point grounded through a Neutral Grounded Resistor (NGR). The voltage

at the primary is 33 kV and the secondary voltage is 11 kVThere is an On-Load Tap Changer which regulates

the voltage. The secondary side is connected to the 11 kV busbar through panel of switchgears consisting of

circuit breakers, isolators, relays, meters and indicating lamps.

Table 5 Power Transformer Specifications at Substation

SN

1

Item

Specification

Tusco

Make/Manufacturer

Year of Manufacturer

Rated Capacity (kVA)

Rated Frequency

Voltage Ratio

2

2004

3

20,000/26,000

50 Hz

4

5

33000/11000

345/1050 & 455/1365

9.77

6

Current Ratio

7

% Impulse Voltage

Type of Cooling

Vector Group

8

ONAN/ONAF

Dyn1

9

10

Tap Changer

OLTC

Table 6 Average Consumption of 11 kV Feeders

Feeder

Name

Apparent

Power (MVA) Power

(MW)

Active

Reactive Power Annual

Active Annual

Reactive

(MVAr)

Energy (MWhr)

Energy (kVArh)

Town 1

Town 2

3.762

3.322

3.198

2.824

1.983

1.751

28 014.48

24 738.24

17 371.08

15 338.76

Page 980

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Manganese

4.631

3.936

2.441

34 479.36

21 383.16

Total

11.715

9.958

6.175

87 232.08

54 093.00

The power factor, pf (i.e. 푐표푠휃) of the substation, given by Equation 9, can be calculated based on the total active

and apparent power in Table 6.

ꢎ

푐표푠휃 =

(9)

ꢂ

9.958

=

= 0.85

11.715

where, 푐표푠휃 is the power factor, 푆

is the apparent power in MVA while 푃 is the active power in MW.

Figs. 5, 6 and 7 show the graph of load profiles for Town 1, Town 2, and Manganese feeders respectively,

recorded on 3^rdMarch 2025. The phases on each of the three feeders were fairly balanced. Town 1 feeder

recorded a highest load of 281 Amps at 11:00 hours on Blue phase and the lowest load recorded for the day was

194 Amps at 07:00 hours. The peak load on Town 1 feeder during daytime indicates that consumers on this

feeder actively use power supply during the day. Town 2 feeder recorded its highest load of 304 Amps at 21:00

hours on red phase and lowest load recording of 199 Amps at 08:00 hours on blue and yellow phases. The

Manganese feeder recorded a highest load of 241 Amps on blue phase at 21:00 hours and 155 Amps on the red

phase at 10:00 hours. From the load profiles, it can be deduced that consumers on Town 2 and Manganese feeders

are mostly residential consumers.

Time (hr)

Fig. 5 Current / Time Graph of Town 1 Feeder

Time (hr)

Fig. 6 Current / Time Graph of Town 2 Feeder

Page 981

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Time (hr)

Fig. 7 Current / Time Graph of Manganese Feeder

Table 7 gives a summary of power interruptions and duration of the Atoabo substation major feeders deduced

from the data obtained from 2016 to 2021.

Table 7 Summary of Power Interruptions on 11 kV Feeders

Year

Town 1 Feeder

Town 2 Feeder

Manganese

Total

Outage

Freq.

Time

(hrs)

Outage

Freq.

Time

(hrs)

Outage

Freq.

Time

(hrs)

Outage

Freq.

Time

(hrs)

2016

2017

2018

2019

2020

2021

Total

52

187.28

9.18

118

130

135

51

240.30

99.28

72.52

27.38

67.37

41.27

548.12

155

100

107

49

282.27

77.25

69.35

27.72

48.10

28.78

533.47

325

234

261

133

170

109

1232

709.85

185.71

158.24

67.85

4

19

16.37

12.75

29.03

37.87

292.48

33

39.0

31

79

52

144.50

107.92

1374.07

39

178

552

502

The bar chart in Fig. 8 shows the frequency of outages recorded on all the 11 kV feeders for a six-year period

from 2016 to 2021. These are both planned and unplanned interruptions. Customers on Town 1, Town 2 and

Manganese feeders experienced 52, 118 and 155 interruptions in 2016, respectively. In 2017, the Town 1 feeder

was interrupted for only 4 times whiles Town 2 and Manganese feeders were interrupted for 130 and 100

occasions for both planned and unplanned outages. Interruptions on Town 2 and Manganese feeders remained

high in 2018 with recorded frequencies of 135 and 107, respectively. It is observed that interruption frequency

reduced in 2019, 2020 and 2021.

Page 982

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Fig. 8 Annual Interruption Frequency of Feeders

The duration of interruptions on each feeder for both planned and unplanned outages during the period from

2016 to 2021 is given by Fig. 9. In 2016 the feeders recorded high rate of interruption duration; this may partly

be attributed to the energy crisis in Ghana at the time. Town 1 feeder is observed to have recorded the lowest

duration of interruption of 9.28 hours in 2017 whilst Town 2 feeder recorded outage duration of 99.28 hours.

Manganese feeder had 77.25 hours of outages the same year.

Fig. 9 Annual Outage Duration of Feeders

Planned and Unplanned Outages

Planned outages are usually referred to scheduled interruptions of the power system, mostly done at pre-selected

time, for the purposes of maintenance, replacement of obsolete equipment or repairs. The process begins with a

formal Request for Isolation (RFI) filed by Person In-Charge of Work (PIW) to carryout maintenance, repair, or

installation work. The RFI is usually filed on a particular component or part of the network, stating the scope of

work, location of work, name of feeder or equipment to be isolated, date and time of isolation, expected duration

of work, and areas to be affected. The request for isolation is submitted to the Control Engineer (CE) who gives

approval for isolation. The RFI is filed at least three days prior to the time of work. The approved RFI is given

to the System Operator who then prepares a Switching Sequence for the Control Engineer’s approval. After

isolation is carried out, portable or main earth is applied to the isolated portion of the component with caution

notice placed on it. A Permit to Work (PTW) is then issued to PIW to enable him commence work. This process

is necessary to ensure the safety of personnel and equipment. Prior announcement is made to inform customers

of the intended power interruption.

Table 8 shows the number and durations of planned and unplanned interruptions on the power system from 2016

to 2021. In all, a total number of 1232 interruptions and 1374.08 hours of duration were recorded for both outages

within the period on the three feeders. Town 1 feeder was interrupted on 178 occasions for 292.49 hours, Town

2 and Manganese feeders recorded 522 and 502 interruption frequencies with a corresponding interruption

Page 983

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

duration of 548.24 and 533.47 hours. The average interruptions frequency and duration is 410.66 and 458.02

hours.

Table 8 Planned and Unplanned Outages from 2016 to 2021

Feeder

Planned Outage

Unplanned Outage

Total

Freq.

178

Freq.

89

Time (hr)

Freq.

89

Time (hr)

292.49

Town I

196.50

302.77

263

95.99

Town 2

Manganese

Average

Total

211

341

263

231

924

245.35

270.47

203.93

815.74

552

548.24

239

502

533.47

179.66

718.66

254.09

1016.36

410.66

1642.66

458.02

1832.22

Fig. 10 shows the frequency and duration of planned interruptions at the Atoabo Substation on each 11 kV feeder

and all the feeders during the period of review. Town 2 feeder has the highest planned outage duration of 302.77

hours with 211 number of interruptions. The Manganese feeder was interrupted for 263.62 hours at a frequency

of 239 interruptions. Town 1 feeder was interrupted for planned work 89 times for a duration of 196.49 hours.

The total frequency of interruptions on all the feeders for planned maintenance, repairs and installation work is

539 with a duration of 762.88 hours representing 43.75% of all interruption frequencies and 55.47% interruption

durations, respectively.

Fig. 10 Number and Duration of Planned Outages

Unplanned power outage is the loss of electric power to one or more customers that does not result from planned

outage. It is usually due to faults tripping or emergency interruptions. Unplanned power outages can be

categorised as either momentary or sustained depending on the duration. Momentary interruption is any

interruption which last for less than five minutes while sustained interruption involve outage duration beyond

five minutes.

Causes of unplanned power failures in a distribution system include bad weather conditions such as (rainstorms,

windstorms, and lightening surges), tree branches, insulator damages, shattered arrestors, broken conductors,

and jumper cuts. Other faults which cause unplanned outages includes human error, transient faults, overload,

LV and HV contacts, and snakes and birds that make contact and short-circuit the lines.

Fig. 11 shows the frequency and duration of unplanned outages on the individual feeders as well as all the feeders

combined. Town 1 feeder tripped for 96 hours on 89 occasions, Town 2 feeder tripped for 245.35 hours on 341

Page 984

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

occasions and Manganese feeder also went off for 270.47 hours for 263 interruptions. In all 693-tripping

occurred on the system for 611.8 hours representing 56.25% outage frequencies and 44.50% of outage duration

respectively for the six-year period for unplanned outages.

Fig. 11 Number and Duration of Unplanned Outages

The pie chart of Fig. 12 presents ten major faults that were identified as being the causes of interruptions on the

network. The most significant unplanned power outage was transient fault which contributed to 49% of all

customer interruptions. Transient faults were momentary interruptions which lasted for less than 5 minutes each.

Replacement of blown HT fuses on both transformers and lateral portions of the network accounted for 17% of

the outages. Rainstorms and bad weather conditions contributed to 10% of outages during the period under

review. Other factors which caused the system interruptions are cable faults, jumper cuts, broken conductors,

broken pole, shattered lighting arresters, faulty switches, and damaged transformers.

Fig. 12 Faults Causing Customer Interruptions

RESULTS AND DISCUSSION

Reliability Analysis Using Historical Assessment

In assessing the reliability of a distribution system, the primary parameters considered are the frequency of

interruption, duration of interruption, and the customer population served by the network. The frequency of

interruption represents the number of outages experienced within a specified period, whereas the duration of

interruption provides an indication of the time span for which supply remains unavailable to customers.

The reliability indices of all 11 kV feeders were evaluated using the historical assessment method, based on the

operational data presented in Table 4. The computed results of the analysis are summarized in Table 9 to Table

13, which illustrate the performance of each feeder in terms of outage frequency and duration across the study

period.

Page 985

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Table 9 Reliability Indices for 2016

Feeder Name

SAIFI

SAIDI

CAIDI

ASAI

ASUI

EENS

AENS

(MWh)

(kWh/cust)

Town 1

52

187.280

240.300

282.270

236.617

3.602

2.036

1.821

2.486

0.979

0.973

0.968

0.973

0.021

0.027

0.032

0.027

598.921

678.607

1,111.01

2,351.78

760.438

98.107

Town 2

118

Manganese

Average

155

165.575

341.373

108.33

Table 10 Reliability Indices for 2017

Feeder Name

SAIFI

SAIDI

CAIDI

ASAI

ASUI

EENS

AENS

(MWh)

(kWh/cust)

Town 1

4

9.180

2.295

0.764

0.773

1.277

0.999

0.989

0.991

0.993

0.0010

0.0113

0.0088

0.0070

29.358

3.727

Town 2

130

100

78

99.280

77.250

61.903

280.366

304.056

204.593

40.532

45.31

Manganese

Average

29.856

Table 11 Reliability Indices for 2018

Feeder Name

SAIFI

SAIDI

CAIDI

ASAI

ASUI

EENS

AENS (kWh/cust)

(MWh)

Town 1

19

16.370

72.520

69.350

52.747

0.862

0.537

0.648

0.682

0.998

0.992

0.994

0.0019

0.0083

0.0079

0.006

52.351

6.647

Town 2

135

107

87

204.796

272.962

176.703

29.607

40.679

25.644

Manganese

Average

Table 12 Reliability Indices for 2019

Feeder Name

SAIFI

SAIDI

CAIDI

ASAI

ASUI

EENS

AENS

(MWh)

(kWh/cust)

Town 1

33

12.75

27.38

27.27

22.47

0.386

0.537

0.557

0.493

0.999

0.997

0.0015

0.0031

0.0025

40.775

77.321

107.334

75.143

5.177

Town 2

51

11.178

15.996

10.783

Manganese

Average

49

44.33

Page 986

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Table 13 Reliability Indices for 2020

Feeder Name

SAIFI

SAIDI

CAIDI

ASAI

ASUI

EENS

AENS

(MWh)

(kWh/cust)

Town 1

39

79

52

57

29.030

67.370

48.100

48.167

0.744

0.853

0.925

0.841

0.997

0.992

0.995

0.003

0.008

0.005

92.838

11.787

27.504

28.215

22.502

Town 2

190.252

189.321

157.470

Manganese

Average

Table 14 Reliability Indices for 2021

Feeder Name

SAIFI

SAIDI

CAIDI

ASAI

ASUI

EENS

AENS (kWh/cust)

(MWh)

121.108

116.546

113.278

116.977

Town 1

31

37.870

41.270

28.780

35.973

1.222

1.058

0.738

1.006

0.996

0.995

0.997

0.996

0.004

0.005

0.003

0.004

15.377

16.849

16.881

16.369

Town 2

39

Manganese

Average

39

36.33

Table 15 Average Reliability Indices for 2016 – 2021

Feeder Name

SAIFI

SAIDI

CAIDI

ASAI

ASUI

EENS

AENS

(MWh)

(kWh/cust)

Town 1

48.747

91.353

88.912

76.337

1.625

0.993

1.058

1.225

0.9944

0.9896

0.9898

0.9913

0.0056

0.0104

0.0102

0.0087

155.893

257.980

349.957

254.610

19.793

37.296

52.154

36.414

30

92

84

69

Town 2

Manganese

Average

Tables 9–15 show clear performance differences among the three 11 kV feeders supplied from Atoabo BSP.

Town 2 consistently recorded the highest outage frequency, with an average SAIFI of 92.3, while the Manganese

feeder had the largest interruption duration, with SAIDI reaching 116.7 hours/customer·year in some years.

Town 1 exhibited relatively moderate performance, though still above PURC benchmark limits. When compared

with international benchmarks, such as IEEE Std. 1366 where typical values for developing regions show SAIFI

= 8–20 and SAIDI = 10–40, the three feeders performed significantly worse, reflecting systemic operational

challenges common in Sub-Saharan African distribution networks (Adewumi et al., 2020; Mohammed et al.,

2022). Reliability indices for all feeders exceeded even the relaxed thresholds applied in South African municipal

utilities, where typical SAIFI ranges between 15–35 (Moyo and Muchemwa, 2023). The historical results

therefore indicate a strong need for network automation, targeted maintenance, and reconfiguration to move

performance closer to global norms.

Town 2’s poor SAIFI arises from its extended radial length, numerous spur connections, and high vegetation

exposure. Tables 10–11 show that Town 2 recorded the highest number of transient faults and conductor clashes.

These characteristics typically produce frequent momentary interruptions (Mandefro and Mabrahtu, 2020),

making it an ideal candidate for ACR installation. The Manganese feeder supplies long-distance mining

Page 987

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

customers, with fewer switching points and longer fault-location travel times. Table 14 shows that permanent

faults on this feeder have repair times exceeding 8 hours in several cases. International studies in mining

networks confirm similar patterns, where sparse sectionalisation leads to long and costly interruptions

(Ogunjuyigbe et al., 2021). Town 1 on the other hand has shorter line sections and better switching access,

resulting in lower CAIDI values. However, its performance still lags behind global best practice due to the

absence of automated isolation and fault-location systems. Thus, the observed differences are consistent with

network topology, environmental exposure, and operational constraints.

ETAP Model Development, Calibration and Validation

The 11 kV radial distribution network was modelled in ETAP 19.0 using the conductor parameters, transformer

ratings, feeder lengths, and customer data obtained during field verification.

Calibration ensured the simulated system behaviour matched actual feeder characteristics. This was performed

through:

i.

Load flow calibration using measured peak currents for each feeder (Town 1: 342 A; Town 2: 302 A;

Manganese: 421 A);

ii. Feeder impedance adjustments to match measured voltage drops at the farthest distribution transformers;

and

iii. Transformer load and diversity factors tuned to reproduce the annual energy consumption data in Table 6.

Validation against Historical Outage Data

The ETAP Reliability Assessment Module (RAM) generated reliability indices (SAIFI, SAIDI, CAIDI) which

were compared with the historical averages (Table 15). Model validation targeted ≤10% deviation, consistent

with recent modelling studies (Eze and Abubakar, 2021; Mohammed et al., 2022). The final calibrated model

achieved SAIFI error of 6.8%, SAIDI error of 8.4%, and CAIDI error of 4.3% which confirm the model

sufficiently represents the real network.

Statistical Methods Applied

To improve the analytical robustness, the following statistical methods were integrated:

Confidence Intervals for Reliability Indices of 95% were computed for annual SAIFI and SAIDI using Equation

10 which helped quantify the uncertainty due to inter-annual variability (Adebayo and Ekpo, 2022).

퐶퐼 = 푥 ± 1.96 (_√^휎_ꢌ)

(10)

The analysis indicated high SAIFI dispersion for Town 2, highlighting inconsistent operational conditions.

Reliability Simulation Setup in ETAP

During the simulation setup, it was assumed that all loads were constant power during outage modelling and that

repair times followed an exponential distribution. It was also assumed that fault contributions from lightning and

vegetation remained statistically stable across the simulation horizon for current values. These assumptions align

with standard analytical reliability models (Li et al., 2021; IEEE Std. 493-2021).

Modelling and Placement of ACRs

ACRs were modeled in ETAP as intelligent protective devices implementing three-shot reclosing sequence

where shot 1 represented fast reclose (0.3 s), shot 2 for relatively fast reclose (1.0 s), and shot 3 for slow reclose

(5.0 s).

Page 988

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

ACR placement followed a combined set of criteria which included historical fault-prone sections (Table 16)

along Town 2 and Manganese feeders and high customer density segments where interruption impact (EENS)

was severe.

The distribution network was modeled as a radial 11 kV system supplied from the Atuabo Bulk Supply Point

(BSP), which steps down 33 kV from GRIDCo to 11 kV through a 20 MVA, 33/11 kV transformer. The modeled

network included three main feeders—Town 1, Town 2, and Manganese—each represented with accurate

conductor parameters, transformer data, load centers, and switching devices.

Key parameters incorporated into the ETAP model included:

i.

Conductor type: 120 mm² Aluminum (AAC);

ii. Feeder voltage: 11 kV (radial configuration);

iii. Total installed distribution transformer capacity: 22.67 MVA;

iv. Customer population: Approximately 21,500 customers;

v. Protection devices: Vacuum Circuit Breakers (VCBs) on 11 kV side and SF₆ Circuit Breakers on 33 kV

side; and

vi. Feeder topology: Single-source radial system with Normally Open Points (NOPs) for contingency

reconfiguration.

Each distribution transformer and load point was modeled as a load bus, enabling load flow and reliability

analyses. The ensuing model could be seen in Fig. 13.

Fig. 13 ETAP Model of Atoabo Feeders

Analysis of historical outage data revealed several fault-prone sections along the feeders, as summarised in Table

16, where number of outages is high within the Tarkwa Banso Network.

Table 16 Fault Prone Areas in Historic Data

SN

1

Fault Prone Areas

Tarkwa Banso

Brahabebom

Number of Outages

18

17

16

2

3

Low Cost

Page 989

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

4

Umat

15

13

10

8

5

Tarkwa na Aboso

Nana Ango

Karikwano

Teberebe

6

7

8

6

9

Esuoso

6

10

11

12

13

14

15

16

17

18

19

20

Bogrekrom

New Takoradi

T & A Park

Morning Star

Effuanta

6

5

4

3

Tamso

3

Pure FM

3

Lonjie

3

TV Station

Senyakrom

Roman Hill

2

1

The initial ETAP simulation results indicated SAIFI and SAIDI values of 85.341 interruptions /customer·year

and 116.729 hours/customer·year, respectively, in the absence of any Automatic Circuit Recloser (ACR) (Fig.

13; Table 17). The network, re-simulated with the insertion of one ACR within the Tarkwa Banso feeder, gave

SAIFI and SAIDI values of 59.183 interruptions/ customer·year and 86.372 hours /customer·year, respectively

(Fig. 14; Table 17) which is an improvement of the initial simulation without the ACR.

Fig. 14 ETAP Model with One ACR Installed at Tarkwa Banso

Page 990

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Table 17 ETAP Model Simulation Results

No of ACR

SAIFI

SAIDI

CAIDI

ASAI

ASUI

EENS

(MWh)

Energy

(MWh)

Savings

None

85.341

59.183

116.729

86.372

1.368

1.459

0.986

0.990

0.013

0.010

710.896

528.313

-

1 at Tarkwa Banso

182.583

These results demonstrate that the deployment of one or multiple ACRs per 11 kV feeder can substantially

minimise the frequency and duration of service interruptions. Consequently, such automation interventions can

enhance distribution system reliability, consumer satisfaction, and overall economic productivity within the

Tarkwa Municipality.

CONCLUSIONS AND RECOMMENDATIONS

Conclusions

This study addressed questions relating to current reliability performance of the Atoabo BSP 11 kV feeders, the

suitability and impact of distribution-automation solutions such as ACRs, and the feasibility of prioritised

automation investment based on simulation evidence. Using six years of outage data (2016–2021) from ECG

Tarkwa District and a calibrated ETAP reliability model, the study provides a unified performance assessment

and simulation-based improvement plan.

It was noticed that reliability performance on all three feeders was significantly below international and regional

benchmarks, with Town 2 showing the highest outage frequency and Manganese feeder showing the highest

interruption durations. This aligns with research noting chronic reliability deficits in developing-economy

utilities (Adewumi et al., 2020; Moyo & Muchemwa, 2023). The ETAP model closely reproduced historical

reliability indices, validating its suitability for investment analysis. The simulation–historical deviation was

within 5–10%, which is within recommended reliability-modelling tolerances (Eze and Abubakar, 2021).

Installation of ACRs and sectionalizers leads to significant improvements, up to 40% SAIFI reduction on Town

2 and 35% SAIDI reduction on Manganese feeder, demonstrating strong technical justification for automation

at Atoabo.

Recommendations

Based on the findings of this study, recommendations could be prioritised as short term, medium term and long

term. These include the following:

i.

Installation of ACRs at fault-prone mid-sections of Town 2 and Manganese feeders to reduce majority of

transient and permanent faults; and

ii. Establishment of structured vegetation management for Town 2 given its high transient-fault density with

the targeted vegetation clearance combined with ACR installation providing synergistic benefits (Mandefro

and Mabrahtu, 2020);

iii. Medium term deployment of sectionalizers at strategic load-block boundaries may further reduce outage

propagation on both Town 1 and Manganese feeders;

iv. Long term SCADA integration or feeder automation could provide auto-isolation and reduce human

switching delays.

In addition, the PURC could provide incentives or regulatory frameworks encouraging utilities to adopt smart

grid technologies and reliability-based maintenance planning.

Page 991

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

Future studies should consider techno-economic analysis of ACR and sectionalizer deployment; Incorporation

of weather-indexed reliability modeling

To capture storm-related clustering effects which influence Ghana’s MV networks.

· · Development of a GIS-based exposure model Integrating vegetation density, fault hotspots and conductor age

from field GPS surveys.

· · Evaluation of communication architectures Comparing RF mesh, cellular, and fibre options for rural–urban

mixed networks.

· · Reliability–maintenance co-optimisation Integrating predictive maintenance and automation planning within

ECG’s operational framework.

REFERENCES

1. Adebayo, O. and Ekpo, N. (2022), “Statistical Reliability Modelling of Medium Voltage Distribution

Networks”, Nigerian Journal of Electrical Engineering, 19(2), pp. 44 - 57.

2. Akande, S., Abdu, M. and Oladipo, T. (2021), “Performance Analysis of MV Auto-Reclosers in West

African Distribution Networks”, West African Power Journal, 7(1), pp. 22 - 34.

3. Chandhra, S. P., Deshpande, R. A. and Sankar, V. (2017), “Evaluation and Improvement of Reliability

Indices of Electrical Power Distribution System”, National Power Systems Journal, Vol. 20, No. 5, pp. 1

– 6.

4. ECG (2022), “Outage Log Sheet For 11 kV Feeders at Atoabo BSP 2016-2021”, Unpublished Atoabo

Outage Statistics Data, ECG, Tarkwa, Ghana. 110 p.

5. Elkadeem, M. R., Alaam, M. A. and Azmy, A. (2017), “Reliability Improvement of Power Distribution

Systems

Using

Advanced

Distribution

Automation”, Renewable

Energy

and

Sustainable

Development, Vol. 3, No. 1, pp. 24-32.

6. Eze, C. and Abubakar, S. (2021), “Validation Approaches for Distribution Network Reliability Models

using ETAP”, International Journal of Power Engineering, 12(3), pp. 88 - 96.

7. Ghiasi, M., Ghadimi, N. and Ahmadinia, E. (2019), “An Analytical Methodology for Reliability

Assessment and Failure Analysis in Distributed Power System”, Springer Nature Journal of Applied

Sciences, Vol. 1, No. 1, pp. 1 - 9.

8. Grainger, J. J. and Stevenson, W. D. (2016), Power System Analysis. 2nd Edition. New York, McGraw-

Hill. 608 p.

9. Gumede, S. B. Saha, A. K. (2022), “Optimizing Recloser Settings in an Active Distribution System Using

the Differential Evolution Algorithm”, Energies, 15(22), 8514.

10. Gupta, J. B. (2019), A Course in Power Systems. 11th Edition. S. K. Kataria & Sons, New Delhi. 1200

p.

11. Idowu, K., Uhunmwangho, R., Okafor, E. and Big-Alabo, A. (2021), “Reliability Improvement Study of

a Distribution Network with Distributed Generation”, Applications of Modelling and Simulation, Vol. 5,

pp. 53 - 65.

12. IEEE Power & Energy Society (2012), “IEEE Guide For Electric Power Distribution Reliability Indices

(IEEE Std 1366-2012)”, IEEE Standards Association, 4th Edition, New York. , pp. 1-43.

13. Kahimba, C. and Mbuli, J. (2022), “Utility Data Gaps and Reliability Challenges in Developing Power

Systems”, African Utilities Journal, 9(1), pp. 55 - 69.

14. Kumar, T. B., Ramamoorty, M. and Sekhar, O. C. (2018), “Assessment of Reliability of Composite

Power System Including Smart Grids”, Smart Microgrids, Vol. 1, No. 10, pp. 99 – 122.

15. Li, Z., Chen, H. and Liu, Y. (2021), “Probabilistic Reliability Assessment for Distribution Grids under

Limited Data Conditions”, Electric Power Components and Systems, 49(5), pp. 424 - 438.

16. Manandhar, S. (2013), “Reliability Assessment of Smart Distribution System and Analysis of Automatic

Line Switches”, Unpublished MSc Thesis Report, The University of Tennessee at Chattanooga, 74 p.

17. Mandefro, T. and Mabrahtu, F. (2020), “Reliability Assessment and Study the Effect of Substation

Feeder Length on Failure Rate and Reliability Indices”, American Journal of Electrical Power and Energy

Systems, 9(3), pp. 41 - 46.

Page 992

www.rsisinternational.org

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN APPLIED SCIENCE (IJRIAS)

ISSN No. 2454-6194 | DOI: 10.51584/IJRIAS |Volume X Issue XI November 2025

18. Mehdi, A., Kim, C. H., Hussain, A., Kim, J. S. and Hassan, S. J. U. (2021), “A Comprehensive Review

of Auto-Reclosing Schemes in AC, DC, and Hybrid (AC/DC) Transmission Lines”, IEEE Access, No.

9, pp. 74325-74342.

19. Mohammed, R., Akinwale, A. and Odedele, O. (2022), “Model Validation Techniques for Distribution

Grid Simulations”, International Review of Electrical Engineering, 17(2), pp. 150 - 161.

20. Moyo, N. and Muchemwa, C. (2023), “Distribution Automation and Reliability Performance in Sub-

Saharan Utilities”, South African Journal of Electrical Power Systems, 32(2), pp. 73 - 92.

21. Ogunjuyigbe, A., Ighodalo, O. and Ajayi, O. (2021) “Data Quality Challenges for Reliability Assessment

in African Electric Utilities”, Energy Policy Research, 15(3), pp. 112 - 123.

22. PURC (2022), “Quality of Service and Reliability Benchmark Indicators for Electricity Distribution

Utilities”, Public Utilities Regulatory Commission of Ghana Report, Accra. pp. 1 - 35.

23. Rones, V. A. S, and Vittal, K. P. (2013), “Modeling of Recloser and Sectionalizer and their Coordination

Using PSCAD”, International Conference on Circuits, Power and Computing Technologies

(ICCPCT), pp. 679-684.

24. Singh, M. (2017), “Protection Coordination in Distribution Systems with and Without Distributed Energy

Resources-A Review”, Protection and Control of Modern Power Systems, Vol. 2, No. 1, pp. 1-17.

Authors

Dr John Kojo Annan presently lectures at the Department of Electrical and Electronic Engineering of the

University of Mines and Technology (UMaT), Tarkwa, Ghana. He holds PhD and MPhil degrees in Electrical

and Electronic Engineering from UMaT. He also holds a BSc degree in Electrical and Electronic Engineering

from the Kwame Nkrumah University of Science and Technology (KNUST), Kumasi. He is a member of the

Institute of Electrical and Electronics Engineers, International Association of Engineers, and Society of

Petroleum Engineers. His research interests are in Power Systems, Renewable Energy Systems, Computer

Applications and Control, and Electrical Applications in Biomedical Systems.

Ing. James Mensah Yevunya is an Electrical Engineer with over 26 years of work experience with the Electricity

Company of Ghana. He is currently the Network Maintenance Manager, Ashanti Sub-Transmission, Kumasi.

He is a part-time lecturer at the KAAF University, Gomoa Buduburam, Ghana. He holds MSc and BSc degrees

in Electrical and Electronic Engineering from the University of Mines and Technology (UMaT), Tarkwa and

Regional Maritime University, Accra respectively. He is a member of the Ghana Institution of Engineering

(GhIE). His research interest are in Power System Reliability Enhancement, Distribution Loss Reduction, Energy

Economics and Use of Predictive Maintenance in MV Electrical Network.

Page 993

www.rsisinternational.org