Comparative Analysis of Power Quality of Solar PV Power Only, Hybrid Solar PV–Grid Power, and National Grid Power Only for SMEs: The Case of a Laundry Shop in Ghana

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

1. Introduction

Per the technical details of the Electricity Company of Ghana Ltd. (ECG), alternating current (AC) at a frequency of 50 Hz is provided to all customers. The typical voltages are 400 V with a ±10% error margin for three-phase (four lines) and 230 V with a ±10% error margin for a single-phase (two lines). Customers requiring more than 800 kW of power may get it at 11 or 33 kV [1]. This implies that a single-phase user should operate with 50 Hz at voltages ranging between 207 and 253 V. Ghana Grid Company Limited (GRIDCo), the nation’s electricity transmission corporation, aims to provide a top-notch product via a partnership with the utilities, ECG, Northern Electricity Distribution Company (NEDCo), and Enclave Power Company (EPC). The pivotal role of utilities in generating several breakthroughs in electricity quality sometimes goes unnoticed within the fervent debates. In a publication released in 2020 by the Electricity Supply Plan Committee, stakeholders addressed issues such as inadequate available transfer capacity, poor customer end power factor, long feeder and overaged equipment, generation capacity, and fuel supply as major challenges of power quality [2]. These entities are however protected from competition by state regulations.

There are several complaints of appliance failure by end users and repairers alike. It is believed that the fault is due to the poor power quality of the power from the national grid [3]. At Ayeduase in Kumasi, an area mostly populated by university students at the Kwame Nkrumah University of Science and Technology (KNUST), there are several small and medium-sized enterprises (SMEs) that depend on the national grid for power. These SMEs are engaged in business activities such as laundry, electronics sales, printing, groceries, and barbering. For most of these SMEs, appliance failure is rampant due to their reliance/dependence on the national grid. This situation causes financial burden and frustration among business owners, which has a negative impact on productivity. It has been determined that poor power quality has several negative effects on businesses, including high production costs, a high staff redundancy rate, higher maintenance costs, and damage to plants, equipment, appliances, and perishable goods [4–7]. It has been proven that unreliable energy supply hinders SMEs’ production and quality, resulting in low sales and profitability [7].

Though electrical power quality issues are assumed to be the responsibility of the utility companies, end users also have a role in protecting their appliances. To handle power disruptions, most electricity-intensive industries have resorted to the usage of backup generators, reducing the expense of maintenance and fueling with staff [8]; this however, has not been able to fully resolve its negative effects on equipment. Other efforts made by end users include the use of equipment protection devices, such as surge protectors, stabilizers, and automatic voltage regulators (AVRs) to help prevent equipment failure [9, 10]. Despite these efforts, some equipment has been reported to fail during extreme conditions of power instability.

For instance, a laundry shop at Ayeduase, considered as the case study for this research, has had key equipment like washing machines failing, even under the protection of the said protection devices. Owners of this shop have reported a decrease in productivity due to unplanned repair expenses and the inability to meet customer demand during equipment downtime. This means productivity and effective functioning of a business rely on the unobstructed accessibility of energy and power supplies [11, 12]. SMEs see an improvement in labor productivity as a result of gaining access to electricity [13]. To address the power quality issues, this study adopted the notion of using solar PV power during the day when most machines will be in use, and the national grid at night, when most of the electrically sensitive equipment would not be in use.

This study investigated three modes of power operation for a hybrid solar PV system installed at the laundry shop. For each mode, the power delivered to the equipment is critically and comparatively analyzed to determine the presence of power quality issues like voltage sag, voltage swell, voltage interruption, frequency, and flicker [14]. Power quality and the integration of renewable energy have been the subject of several studies. However, the majority have focused on feed-in PV grid-connected systems, which claim that the quality of grid electricity is negatively impacted by renewable energy [15]. In the studies conducted by Sandhu et al. [16], Canova et al. [17], Lin et al. [18], Enslin [19], Roy et al. [20], and Adak et al. [21] on renewable energy integration, they affirm that renewable energy grid integration impacts the grid negatively.

Aditya et al. [22] reviewed key challenges in grid-integrated solar photovoltaic (GIPV) systems, focusing on power quality and technical issues. As reliance on renewable energy grows, GIPV systems face challenges such as voltage fluctuations, harmonics, power factor deviations, and frequency instability. Raj and Bhattacharjee [23] examined a 100 kW double-stage three-phase grid-connected solar PV (GCSPV) system designed to meet the growing demand for cost-effective renewable energy. An efficient passive LCL filter is implemented, reducing total harmonic distortion (THD) in the grid output to below 1.17%. A PLL system ensures synchronization with the grid. The system is modeled in MATLAB, analyzing converter and inverter performance, grid voltage, current, power, and THD levels. Raj et al. [24] analyzed a double-stage grid-connected solar photovoltaic (GCPV) system with enhanced control and improved power quality. A DC–DC boost converter stabilizes the variable DC voltage, while a voltage source inverter (VSI) converts it to AC. To mitigate harmonics that degrade power quality, a passive LCL filter is implemented at the grid interface. The designed filter significantly reduces THD, lowering grid voltage THD from 0.59% to 0.08%, grid current THD from 10.71% to 1.17%, and inverter current THD from 10.70% to 4.9%.

The current study, however, focuses on zero-export grid-tied PV systems and the potential of solar PV technology to solve power quality issues for end users in Ghana and most African countries. Figure 1 presents an overview of the strategy used. To the best of the authors’ knowledge, no studies have been conducted on the use of solar PV to address the grid power qualities in this way, especially within the context of Ghana and the sub-Saharan region. The methodology employed and the results obtained are presented in the following sections.

[figure(s) omitted; refer to PDF]

2. Methodology

2.1. Study Location

The study was carried out at a laundry shop located at Ayeduase, a business town located on the south side of KNUST. The laundry shop has three washing machines for its laundry activities, as shown in Figure 2a. Other appliances include a dryer, an iron, a fan, lights, a television, a submersible water pump, and two surface pressure pumps (Figure 2b).

[figure(s) omitted; refer to PDF]

2.2. Sizing the PV System

This section describes the process, motives, and decisions pertaining to the sizing of the PV system under study. Three main components were sized: the solar PV, the batteries, and the inverter. The solar PV size was selected based on the limited space on the building’s roof available for mounting the panels. The inverter size was selected based on the size of the solar PV and the estimated load size. The battery size was solely based on the size of the load hence necessitating a load assessment procedure. Table 1 presents the sizes of the various components.

Table 1

Laundry shop equipment load and rated voltage.

Equipment	Quantity	Load (W)
Front load washer (motor)	1	250
Front load washer (heater)	1	1350
Top load washer automatic	1	200
Top load washer semiautomatic	1	675
Dryer	1	1500
Iron	1	1500
Submersible water pump	1	1500
Surface pressure pump	2	720
Lights	1	150
Fan	1	100
TV	1	100
Total	11	8045

Load assessment refers to the process of evaluating and analyzing the electrical load requirements of a system, facility, or device. This assessment helps determine the amount of electricity consumed by various electrical appliances, equipment, and systems within a given context. A load assessment procedure was carried out in the laundry shop to determine the individual power ratings of equipment and the total power demand of all equipment at the laundry. The total power was computed using Equation (1) [25]. The outcome of the load assessment is presented in Table 1, showing the rated power and voltage of the various equipment.

Having assessed the total rated load of the laundry, the total average actual power was estimated based on certain assumptions: (1) the dryer is not often operated, owing to the fact that Ghana falls within the tropical climate zone; thus drying of clothes is usually done by the sun; (2) though the rated capacity of the front load washer is seen to be 1600 W, it normally operates at 250 W as it is only when the heater is activated (i.e., consumes full power), which is about 20% in an hour cycle; (3) the submersible water pump is activated when there is a need for water supply into the reservoirs, which is often 30 min a day; and (4) the surface pressure pump (2 units, each 360 W) detects a drop in water pressure and starts one of the pumps when a washing machine is in use and requires water input. Taking these assumptions into consideration, the typical load at the laundry will be 2500 W or less, which is used to size the battery.

The operational hours of the laundry are usually from 8:00 to 18:00, indicating that the laundry shop operates for about 10 h a day. The estimated consumption was therefore 25 kWh per day using Equation (2). Equation (3) was then used to convert the battery size in kWh to Ah considering a battery system voltage of 48 V. $$\begin{array}{}\text{(1)}& \text{Total}\text{power}=P_T=P_1+P_2+P_3\cdots P_n\end{array}$$where $P_T$ is the total power of all devices that are operational in an instant and $P_n$ is the power of individual devices. $$\begin{array}{}\text{(2)}& \text{Energy}\text{demand}=E=\sum _0^{T}n\ast P\ast \Delta h\end{array}$$where $E$ is the energy demand, $T$ is the equipment usage period in hours, $n$ is the number of equipment, $P$ is equipment-rated capacity, and $\Delta h$ is the operational hours of each piece of equipment.

The battery capacity in Ah of the laundry was calculated using Equation (3): $$\begin{array}{}\text{(3)}& {\text{Batt}}_{\text{Ah}}=\frac{1000\ast {\text{Batt}}_{\text{kWh}}}{v}\end{array}$$where $\text{Bat}{\text{t}}_{\text{Ah}}$ is the battery’s energy capacity, $\text{Bat}{\text{t}}_{\text{kWh}}$ is the expected energy demand from the battery, and $V$ is the voltage of the battery.

2.3. Description of Hybrid PV Inverter System

From the load assessment, the laundry’s energy architecture was redesigned to have a grid-tied zero-export PV system to support its energy needs. As opposed to net-metering grid-tied PV systems, zero export indicates that this system draws power from the grid without feeding power back into the grid. The system consists of a hybrid inverter that doubles as a central control for the PV system, determining which mode of operation is best at any time based on the availability of power from the grid, PV, or battery. These days, the term hybrid as used here alludes to the inverter having several functions apart from just converting the DC power from the PV to AC for load consumption. It also acts as a rectifier, converting AC from the grid to DC and combining it with PV power (if and when available) to charge the battery. It also has a charge controller, optimally regulating the charging of the batteries to prevent overcharging and undercharging. Figure 3 shows the hybrid inverter used at the laundry shop, while Figure 4 shows a schematic diagram of the grid-tied PV system. Table 2 presents the various components of this grid-tied solar PV system.

[figure(s) omitted; refer to PDF]

Table 2

Grid-tied solar PV system components.

Component	Specification
Solar PV panels	3 kW
Hybrid inverter	5 kVA
Gel batteries	570 Ah 48 V
Data logger	WatchPower software on a laptop

2.3.1. Modes of Operation of the Hybrid Inverter

The hybrid inverter serves as the central control point of the PV system, determining the system’s mode of operation based on the available sources of energy. The grid-tied solar PV system is therefore operated in four main modes of operation, namely, battery mode, line mode, standby mode, and shutdown mode. The main controller, embedded in the hybrid inverter, is responsible for switching these modes, based on the power availability of the grid, PV, and battery. Table 3 briefs on the various modes of operation.

Table 3

Mode of operation of the grid-tied PV system.

Mode of operation	Functioning
Line mode	Output power solely from the national grid
Battery mode	Outpower from a DC source, either PV or batteries
Standby mode	When both national grid and DC sources are not available
Shutdown mode	The unit powers down

2.3.2. Functioning of Operation Modes

The laundry is configured to operate in the battery mode and line mode depending on conditions as stated below. In the battery mode, the controller is configured to serve load with energy from the DC sources during the day when the sun is up, usually between 6:00 a.m. and 6:00 p.m. [26]. During this mode, should the load exceed the power supply from the DC sources that is battery mode, the power supply is switched to the national grid, activating the line mode. After sundown, power is provided solely from the national grid. In the case when the national grid is not available, stored energy from the battery is supplied to the load.

2.4. Physical Parameters

Data were logged per minute using the WatchPower software which communicates with the hybrid inverter. Parameters monitored and recorded by the software include (1) time, (2) AC voltage grid, (3) AC frequency grid, (4) output apparent power, (5) output active power, (6) output voltage, and (7) output frequency. These were used to compute power quality parameters for analysis, specifically, voltage level with percentage error, sag/undervoltage, swell/overvoltage, voltage interruption, flicker, and frequency. The percentage error was determined using Equation (4). $$\begin{array}{}\text{(4)}& \text{Percentage}\text{error}=\frac{\text{actual}\text{value}-\text{expected}\text{value}}{\text{expected}\text{value}}\ast 100\%.\end{array}$$

3. Results and Discussion

3.1. Modes and Output

To assess the performance of the system, various power modes were run and analyzed for 24 h to account for all the power variations within a day properly. The power modes include PV power only, hybrid PV-grid power, and grid power only. The “PV power only” alludes to the battery mode of operation where the load is supplied with power from the PV and battery. The “hybrid PV–grid power” in this case means that, in the 24-h period, the power is switched between PV only and grid only; it does not suggest a combined power supply from both sources. It is important to note that the hybrid inverter of the system under study does not combine power from the PV and grid instantaneously to supply the load. Finally, grid power only where the load is supplied with power from the grid which is line mode.

3.1.1. Solar PV Power Only

Figure 5 presents a plot of apparent power and AC voltage against time for the “solar PV power only” mode. Apparent power here represents the power in volt-amperes (VA) consumed by the laundry shop’s equipment, which is also seen as the final output of the hybrid inverter. The AC voltage is the voltage measured on the power delivered to the load. It could be seen from the plot that, throughout the day, voltage was maintained at 230 V with very little variations. The few variations observed were when a very sharp change occurred in consumption, causing a variation between 217.7 and 233.5 V. Even still these values fall within the acceptable working range specified by ECG and IEEE standards and could be asserted that no instances of voltage sag or swell occurred. The observed profiles indicate that solar PV provides a stable supply of power.

[figure(s) omitted; refer to PDF]

3.1.2. Hybrid Solar PV–Grid Mode

Figure 6a shows the profile of apparent power and AC output voltage against the time of the laundry’s PV system when operated in the “hybrid PV–grid” mode. Figure 6b takes the same data and plots them against the system’s mode of operation (line or battery) per instance. A sample of the raw data from which the figures were plotted is presented in Table 4 to give a numerical appreciation of the interplay of studied parameters with other parameters like battery state of charge. It can be seen from Figure 6a that, between 12:00 a.m. and 6:00 a.m. and 6 p.m. and 11:59 p.m., the output voltage from the inverter had high fluctuations above the nominal 230 V. This is seen as a period that the system operated in the line mode as shown in Figure 6b. Conversely, the output voltage is seen to have a very stable voltage during the day when the system is operating in battery mode, that is, receiving power from the PV and battery. The only notable fluctuation is seen around 8:00 a.m. and 12:00 p.m., which are times that the inverter had switched to line mode. Taking the first fluctuation as an example as presented in Table 4, it could be seen that the battery had been supporting the PV in power supply to the load as observed in the decrease in the state of charge up until 8:15:36. The hybrid inverter is set to give the battery a depth of discharge of about 50%; hence, the mode of operation was switched to line mode, drawing power only from the grid to serve the load and the power from solar PV is used to charge the batteries. At this time, voltage fluctuations increased, raising the voltage to hit a maximum of 244.4 V. At 8:25:37, operation was switched to battery mode, stabilizing the voltage to its nominal 230 V. These voltage fluctuations, as observed typically on the Ghana national grid, are an undesirable condition that leads to the shortened lifespan of equipment/appliances, leading to unexpected expenses for the users.

[figure(s) omitted; refer to PDF]

Table 4

Hybrid operation transition logs.

Device mode	Time (t)	Grid AC voltage (V)	PV input power (W)	Apparent power (VA)	Battery state of charge (%)	Solar AC voltage (V)
Battery mode	8:05:38	244.4	552	552	75	230
Battery mode	8:06:37	245.1	553	506	76	230.2
Battery mode	8:07:36	244.9	553	506	76	230.1
Battery mode	8:08:38	245.3	567	482	76	229.9
Battery mode	8:09:37	245.8	568	781	70	229.9
Battery mode	8:10:36	246.2	555	299	76	230
Battery mode	8:11:36	244.6	569	529	73	230.1
Battery mode	8:12:38	245.1	564	1241	74	229.9
Battery mode	8:13:38	243.4	562	782	63	230.1
Battery mode	8:14:37	242.3	560	827	58	229.6
Battery mode	8:15:36	242.3	577	622	56	230.5
Line mode	8:16:37	239	576	2533	89
Line mode	8:17:36	239	593	2270	99
Line mode	8:18:37	237.7	594	2186	100
Line mode	8:19:36	237.2	580	1992	100
Line mode	8:20:38	238.7	620	2458	100
Line mode	8:21:37	244.1	632	805	100
Line mode	8:22:37	243.8	633	804	100
Line mode	8:23:36	242.9	616	1020	100
Line mode	8:24:36	244.4	634	831	100
Line mode	8:25:37	242.9	662	1165	100
Battery mode	8:26:38	242.8	660	529	85	230.2
Battery mode	8:27:38	243.3	663	878	77	231.1
Battery mode	8:28:38	243.1	655	758	88	229.9
Battery mode	8:29:36	243.9	645	414	92	230.2
Battery mode	8:30:36	244.4	669	870	92	229.7
Battery mode	8:31:38	243.4	675	758	94	229.7
Battery mode	8:32:36	242.5	665	758	95	229.8
Battery mode	8:33:36	242.5	669	781	93	229.9
Battery mode	8:34:36	244	656	666	95	229.9
Battery mode	8:35:38	243.6	671	620	96	229.6
Battery mode	8:36:38	244	691	690	96	230.1
Battery mode	8:37:37	242.8	699	782	99	230.3
Battery mode	8:38:38	243	715	736	100	229.8

3.1.3. National Grid Power Only

Figure 7 depicts the voltage of the national grid should the load run completely in Grid mode; that is, power is supplied completely from the grid. Observing the figure, the voltage level from the national grid was inconsistent and highly fluctuating throughout the day, with some voltage dips. A voltage dip can be seen from 242 to 215 V at 9:24 a.m. A steady decrease in voltage occurred from 243 to 225 V at 12:05 p.m. and continued to 04:07 p.m. A sudden rise in voltage occurred from 239 to 244 V at 7:30 p.m. Lastly, a dip in voltage occurred from 238 to 225 V at 08:15 p.m. Hence, the voltage level as seen here is very wavy and of poor power quality.

[figure(s) omitted; refer to PDF]

3.2. Power Quality Parameters

3.2.1. Voltage Level

Ghana’s electricity power generation is by the Voltage River Authority (VRA), Bui Power Authority (Bui), and the Independent Power Producers (IPPs). The National Interconnected Transmission System (NITS) transmits electricity; its operation and maintenance are by GRIDCo. Distribution of power is by the ECG for the southern electricity distribution zone of Ghana which is the largest distribution company in Ghana; the NEDCo is responsible for electricity distribution in the northern parts of Ghana and lastly EPC the only private electricity distribution company which operates within the Tema Free Zones Enclave and Dawa Industrial Zone. Ghana operates at a frequency of 50 Hz. The voltage levels from generation by power producers through transmission and distribution by utilities to the end user are illustrated in Table 5.

Table 5

Voltage levels for the grid power system for Ghana.

Power producers		Transmission		Distribution		User
VRA, Bui, and IPPs		NITS (GRIDCo)		ECG, NEDCo, and EPC		Ayeduase
11 kV		69 kV		33 kV		$$400\text{V}\pm 10\%$$
14 kV		161 kV		11 kV
15 kV		225 kV		$$400\text{V}\pm 10\%$$		$$230\text{V}\pm 10\%$$
34.5 kV		330 kV		$$230\text{V}\pm 10\%$$

Figure 8a presents samples of typical voltage data of the laundry shop plotted against the mode of operation and the time of operation. The graphs show the voltage profile for the grid power and the solar output power. It is evident from the graph that there are significant differences between both plots. The solar output, deliverable through the batteries, is seen to be very stable, supplying a constant voltage to the load, ensuring a safe and smooth operation of the laundry devices. Meanwhile, the grid voltage is observed to be sporadic and unstable in its operation, changing rapidly and sharply at certain times during the system’s operations. For instance, in Figure 8b, the voltage is seen to decrease sharply from around 240 to 215 V. While these values may still be within the allowable ±10% error, such fluctuations could still cause some equipment damage, leading to wear and tear, malfunctioning of sensitive devices, and inefficiencies in the operation of equipment. Per Bollen, the susceptibility of the equipment to voltage fluctuations has heightened as modern power electronic equipment not only experiences vulnerability to voltage fluctuations but also causes disruption to other users [27]. These findings reinforce the benefit of relying on solar power systems for power supply in SMEs.

[figure(s) omitted; refer to PDF]

Some efforts taken by consumers include the use of AVR to help mitigate fluctuations. However, this attempt seems to fail under certain conditions. On some occasions at the laundry, it was noticed that the AVR only worked when the voltage from the national grid was above 255 V; in other instances, it gave a high voltage error and disconnected the national grid, halting operations. Images of instances not recorded by the software have been attached in Figure 9a,b. At a voltage above 293 V, the ECG meter disconnected power to the facility, as seen on the ECG meter display console in Figure 9a. However, there were instances whereby the meter did not disconnect; thus, high voltage would have reached the devices in the laundry shop. This was rectified by the presence of the AVR, which disconnected the power from reaching the devices at the laundry shop in Figure 8b. Finally, at a voltage below 290 V, the AVR was able to regulate the voltage to 220 V, as seen in Figure 9c.

[figure(s) omitted; refer to PDF]

3.2.2. Power Factor

Figure 10 presents a sample power profile of the laundry shop’s power system, displaying the apparent power, active power, and power factor of the same load operating in both the line mode and the battery mode. During the period from 4:00 a.m. to 7:00 a.m. (Figure 10a), the load consistently used a steady amount of active power in both line and battery modes. During the period from 4:00 a.m. to 5:36 a.m., the load used an average apparent power of 243 VA in line mode with a total of six spikes observed during this timeframe. In battery mode, the average apparent power was 229 VA, and there were two instances of spikes for a similar duration which occurred between 5:36 a.m. and 7:00 a.m. The power factor exhibited a better performance in battery mode compared to line mode for the same load (Figure 10b).

[figure(s) omitted; refer to PDF]

3.2.3. Frequency

Figure 11 shows the frequency of the laundry power system when operated in the solar PV and grid modes, respectively. Frequency in power systems refers to the rate at which AC oscillates or alternates direction. It represents the number of cycles per second of the AC power. Ideally, frequency is kept constant to facilitate the smooth operation of a power system. Comparing the plot for solar PV (Figure 11a) and grid (Figure 11b) frequency, it is evident that the solar PV power, after inversion, maintained a fairly constant frequency while that of the grid showed some level of inconsistency. A constant frequency is crucial for the smooth operation of a device, especially motors which are designed to operate at set frequencies. Changing frequencies can affect the reliability and longevity of these devices.

[figure(s) omitted; refer to PDF]

3.2.4. Power Interruption

To have a more comprehensive appreciation of the rate of grid power interruptions, a different data logging aspect of the WatchPower software was initiated from December 7, 2023, to January 5, 2024. A decrease in power interruptions is usually anticipated in Ghana during the Christmas and New Year months. This is because utility companies are urged by the government and security services to put in extra effort to ensure a stable and consistent power supply due to the high criminal activities experienced during such festive seasons. Despite this, it could be gleaned from Table 6 that there were 26 instances of power interruptions within the period monitored, sometimes lasting as much as 4 h.

Table 6

Outage data from December 7, 2023, to January 5, 2024.

Count	Level	Date	Time	Duration	Event
26	Warning	5 Jan 2024	11:51:43 a.m.	0:08:59	AC recovery
	Warning	5 Jan 2024	11:42:44 a.m.		LINE_FAIL
25	Warning	3 Jan 2024	7:37:39 p.m.	0:29:36	AC recovery
	Warning	3 Jan 2024	7:08:03 p.m.		LINE_FAIL
24	Warning	3 Jan 2024	11:34:14 a.m.	0:03:40	AC recovery
	Warning	3 Jan 2024	11:30:34 a.m.		LINE_FAIL
23	Warning	3 Jan 2024	9:56:33 a.m.	0:06:14	AC recovery
	Warning	3 Jan 2024	9:50:19 a.m.		LINE_FAIL
22	Warning	2 Jan 2024	10:38:32 p.m.	4:01:53	AC recovery
	Warning	2 Jan 2024	6:36:39 p.m.		LINE_FAIL
21	Warning	2 Jan 2024	6:01:12 p.m.	0:00:17	AC recovery
	Warning	2 Jan 2024	6:00:55 p.m.		LINE_FAIL
20	Warning	2 Jan 2024	12:40:47 p.m.	0:18:49	AC recovery
	Warning	2 Jan 2024	12:21:58 p.m.		LINE_FAIL
19	Warning	2 Jan 2024	12:21:47 p.m.	0:08:49	AC recovery
	Warning	2 Jan 2024	12:12:58 p.m.		LINE_FAIL
18	Warning	2 Jan 2024	7:54:34 a.m.	0:10:45	AC recovery
	Warning	2 Jan 2024	7:43:49 a.m.		LINE_FAIL
17	Warning	26 Dec 2023	7:16:45 p.m.	0:05:22	AC recovery
	Warning	26 Dec 2023	7:11:23 p.m.		LINE_FAIL
16	Warning	26 Dec 2023	5:06:29 p.m.	0:03:40	AC recovery
	Warning	26 Dec 2023	5:02:49 p.m.		LINE_FAIL
15	Warning	26 Dec 2023	1:06:31 p.m.	0:43:27	AC recovery
	Warning	26 Dec 2023	12:23:04 p.m.		LINE_FAIL
14	Warning	26 Dec 2023	12:08:59 p.m.	0:02:26	AC recovery
	Warning	26 Dec 2023	12:06:33 p.m.		LINE_FAIL
13	Warning	26 Dec 2023	12:06:15 p.m.	0:05:09	AC recovery
	Warning	26 Dec 2023	12:01:06 p.m.		LINE_FAIL
12	Warning	25 Dec 2023	3:31:14 p.m.	0:31:20	AC recovery
	Warning	25 Dec 2023	2:59:54 p.m.		LINE_FAIL
11	Warning	17 Dec 2023	4:49:36 p.m.	0:00:20	AC recovery
	Warning	17 Dec 2023	4:49:16 p.m.		LINE_FAIL
10	Warning	17 Dec 2023	2:39:23 p.m.	0:06:14	AC recovery
	Warning	17 Dec 2023	2:33:09 p.m.		LINE_FAIL
9	Warning	16 Dec 2023	11:07:16 p.m.	0:09:22	AC recovery
	Warning	16 Dec 2023	10:57:54 p.m.		LINE_FAIL
8	Warning	16 Dec 2023	6:30:15 p.m.	3:25:41	AC recovery
	Warning	16 Dec 2023	3:04:34 p.m.		LINE_FAIL
7	Warning	16 Dec 2023	2:15:47 p.m.	0:41:48	AC recovery
	Warning	16 Dec 2023	1:33:59 p.m.		LINE_FAIL
6	Warning	15 Dec 2023	8:20:31 a.m.	0:45:11	AC recovery
	Warning	15 Dec 2023	7:35:20 a.m.		LINE_FAIL
5	Warning	13 Dec 2023	7:03:36 p.m.	0:06:30	AC recovery
	Warning	13 Dec 2023	6:57:06 p.m.		LINE_FAIL
4	Warning	9 Dec 2023	10:24:23 p.m.	3:23:08	AC recovery
	Warning	9 Dec 2023	7:01:15 p.m.		LINE_FAIL
3	Warning	9 Dec 2023	3:41:33 p.m.	0:00:27	AC recovery
	Warning	9 Dec 2023	3:41:06 p.m.		LINE_FAIL
2	Warning	7 Dec 2023	11:13:17 p.m.	0:01:53	AC recovery
	Warning	7 Dec 2023	11:11:24 p.m.		LINE_FAIL
1	Warning	7 Dec 2023	11:02:02 p.m.	0:01:13	AC recovery
	Warning	7 Dec 2023	11:00:49 p.m.		LINE_FAIL

3.3. Monthly Data

Figures 12 and 13 present monthly data on power and voltage for June and July, respectively. Looking at both figures, several incidents of power outages can be identified with the national grid, typified by the sudden drop of the grid voltage to zero. They provided a comprehensive perspective on the power quality parameters in Table 7. Throughout these, no instances of undervoltage were recorded; however, an overvoltage incidence was recorded (279.9 V) from the national grid on June 7, 2023, at 9:29 a.m. In the case where the system is operating in line mode, such a level of voltage could destroy any equipment connected to it.

[figure(s) omitted; refer to PDF]

Table 7

Power quality parameters.

Parameters	Solar AC	National Grid AC
Voltage level	Voltage appeared linear (216–235 V)	Voltage appeared sinusoidal (214–252 V)
Frequency	Stable	Fluctuating
Power factor	Better power factor	Poor power factor
Sag/undervoltage	Not observed	Not observed
Swell/overvoltage	Not observed	Recorded 279.9 V
Voltage interruption	Power was available throughout the day	Several power interruptions
Flicker	No flicker observed	Flicker observed

4. Conclusion

This study has assessed the optimal power quality for the functioning of SMEs by conducting a comparative analysis of three power sources: solar PV only, hybrid solar PV–grid power, and grid power only. The research findings indicated the following:

1. Solar PV only (battery mode) exhibited superior power quality metrics. The voltage level remained consistently steady at 230 V with a tolerance of ±5%. The frequency ranged between 49.9 and 50 Hz. The power factor was contingent upon the load and exhibited satisfactory performance. There were no instances of voltage interruption or flickering. Solar PV is the most effective method for safeguarding electrical equipment. The inconvenience caused by power outages, which result in the loss of working hours, is also prevented.

2. In the hybrid solar PV–grid power (battery and line mode) system, there is a trade-off in power quality, but it reduces the risk of downtime caused by adverse weather conditions. Solar PV (battery mode) runs sensitive equipment during daylight hours. The frequency range of 49.7–50.3 Hz is the most significantly impacted region, even while operating in battery mode. Furthermore, operations may continue to function during power outages by using the stored energy from the battery bank.

3. Grid power only is detrimental to the operations of SMEs. The grid power is plagued with power interruptions, undervoltage, and overvoltage, which can cause serious damage to equipment, causing a decrease in productivity and profitability in SMEs.

While this study is conducted in a specific geographical location, its findings can be adapted for broader applications in regions with different climates and grid conditions. The key to generalizing this approach lies in understanding how variations in solar irradiance, temperature, and grid stability influence system performance. Different climates affect solar energy generation due to varying levels of sunlight, temperature, and seasonal fluctuations. However, by incorporating climate-adjusted models and predictive analytics, the methodology presented in this study can be adapted to diverse environments. Additionally, integrating flexible control strategies, such as real-time grid monitoring and adaptive power electronics, can enhance system resilience under different grid conditions. For regions with weaker grid infrastructure, solutions such as energy storage integration and demand-side management can improve stability and efficiency. Furthermore, regulatory policies and market structures play a role in how renewable energy solutions are implemented; thus, case-specific modifications may be necessary. Future research should explore multiregional validation of this framework, considering both developed and developing energy markets. By tailoring the system design to local environmental and grid characteristics, the proposed solution can be effectively deployed in diverse geographical settings.

Author Contributions

Godwill Nkrumah Yeboah: conceptualization of idea, modeling, experimentation, data analysis, writing of original draft, design of system, experimentation, data curation and analysis, review of related research work. Richard Opoku: conceptualization of idea, modeling, experimentation, data analysis, writing of original draft, design of system, experimentation, data curation and analysis, writing of original draft, review of related research work, supervision, review of draft paper, editing of final manuscript. Felix Uba: design of system, experimentation, data curation and analysis, writing of original draft, review of related research work, data analysis, supervision, review of draft paper, editing of final manuscript. Gidphil Mensah: design of system, experimentation, data curation and analysis, writing of original draft, review of related research work, data analysis. Charles Kofi Kafui Sekere: review of related research work, data analysis.

Funding

No funding was received for this research.

Acknowledgments

The authors acknowledge the efforts of all contributors, especially during the field work.

References

[1] N. K. Fiasorgbor, R. A. T. Abdulai, M. Antwi-Yeboah, "A Review of the Characteristics and Challenges of Electricity Distribution in Ghana: Case Study of the Northern Electricity Distribution Company," Energy and Power Engineering, vol. 14 no. 11, pp. 615-663, DOI: 10.4236/epe.2022.1411033, 2022.

[2] B. K. Okyere, Financial Implication of Electricity Use and Distributional Losses on Industrial and Economic Performance in Ghana, 2021.

[3] J. I. Arthur, V. N. Arthur, E. Owusu, S. D. Arthur, "Electricity Power Production, Transmission and Distribution Losses in Ghana: the Role of Regulatory Quality," African Geographical Review,DOI: 10.1080/19376812.2024.2360620, 2024.

[4] N. A. B. Boakye, F. B. K. Twenefour, M. McArthur-Floyd, "The Impact of Power Outage ‘Dumsor’ on the Hotel Industry: Evidence From Ghana," Journal of Energy Technologies and Policy, vol. 6 no. 8, pp. 39-47, 2016.

[5] M. A. Nyasapoh, M. D. Elorm, N. S. A. Derkyi, "The role of renewable energies in sustainable development of Ghana," Scientific African, vol. 16,DOI: 10.1016/j.sciaf.2022.e01199, 2022.

[6] A. Sharma, B. S. Rajpurohit, S. N. Singh, "A Review on Economics of Power Quality: Impact, Assessment and Mitigation," Renewable and Sustainable Energy Reviews, vol. 88, pp. 363-372, DOI: 10.1016/j.rser.2018.02.011, 2018.

[7] F. Doe, E. S. Asamoah, "The Effect of Electric Power Fluctuations on the Profitability and Competitiveness of SMEs: A Study of SMEs Within the Accra Business District of Ghana," Journal of Competitiveness, vol. 6 no. 3, pp. 32-48, DOI: 10.7441/joc.2014.03.03, 2014.

[8] N. Kupzig, C. G. Ackah, The Effect of an Electricity Crisis on Firms – A Case Study of Dumsor in Ghana,DOI: 10.2139/ssrn.4440773, 2023.

[9] G. H. Hwang, "A Review on Equipment Protection and System Protection Relay in Power System," International Journal of Integrated Engineering, vol. 9 no. 4, 2017.

[10] N. M. Aljamali, A. J. Kadhium, A. H. J. Al-Jelehawy, "Review in Protection of Laboratory and Electrical Equipment in Laboratories and Institutions," Journal of Controller and Converters, vol. 6 no. 1, pp. 24-30, 2021.

[11] A. Ado, M. M. Josiah, "Impact of Deficient Electricity Supply on the Operations of Small Scale Businesses in North East Nigeria," International Journal of Business and Economic Development (IJBED), vol. 3 no. 1, 2015.

[12] M. Bazilian, I. Onyeji, "Fossil Fuel Subsidy Removal and Inadequate Public Power Supply: Implications for Businesses," Energy Policy, vol. 45,DOI: 10.1016/j.enpol.2012.02.057, 2012.

[13] M. A. S. Chowdhury, S. Chuanmin, M. Sokolová, A. B. M. M. Rahman, A. Akbar, Z. Ali, M. Usman, "Unveiling the Nexus Between Access to Electricity, Firm Size and SME’s Performance in Bangladesh: New Evidence Using PSM," Energies, vol. 14 no. 20,DOI: 10.3390/en14206493, 2021.

[14] G. Bayrak, A. Yılmaz, "Assesment of Power Quality Disturbances for Grid Integration of PV Power Plants," Sakarya University Journal of Science, vol. 23 no. 1, pp. 35-42, 2019.

[15] X. Liang, "Emerging Power Quality Challenges due to Integration of Renewable Energy Sources," IEEE Transactions on Industry Applications, vol. 53 no. 2, pp. 855-866, DOI: 10.1109/TIA.2016.2626253, 2017.

[16] M. Sandhu, T. Thakur, "Issues, Challenges, Causes, Impacts and Utilization of Renewable Energy Sources-Grid Integration," International Journal of Engineering Research and Applications, vol. 4 no. 3, pp. 636-643, 2014.

[17] A. Canova, L. Giaccone, F. Spertino, M. Tartaglia, "Electrical Impact of Photovoltaic Plant in Distributed Network," IEEE Transactions on Industry Applications, vol. 45 no. 1, pp. 341-347, DOI: 10.1109/TIA.2009.2009726, 2009.

[18] C.-H. Lin, W.-L. Hsieh, C.-S. Chen, C.-T. Hsu, T.-T. Ku, C.-T. Tsai, "Financial Analysis of a Large-Scale Photovoltaic System and Its Impact on Distribution Feeders," IEEE Transactions on Industry Applications, vol. 47 no. 4, pp. 1884-1891, DOI: 10.1109/TIA.2011.2154292, 2011.

[19] J. H. R. Enslin, "Network Impacts of High Penetration of Photovoltaic Solar Power Systems," Proceedings of the IEEE PES General Meeting,DOI: 10.1109/PES.2010.5589675, 2010.

[20] N. K. Roy, M. A. Mahmud, H. R. Pota, "Impact of High Wind Penetration on the Voltage Profile of Distribution Systems," Proceedings of the 2011 North American Power Symposium,DOI: 10.1109/NAPS.2011.6025195, 2011.

[21] S. Adak, H. Cangi, "The Quality Problems at Low Irradiance in the Grid-Connected Photovoltaic Systems," Electrical Engineering, vol. 106 no. 5, pp. 6185-6197, DOI: 10.1007/s00202-024-02351-6, 2024.

[22] T. Aditya, R. D. A. Raj, M. R. Shinde, "Power Quality Analysis, Challenges and Feasible Solutions of 90 kWp Grid-Integrated PV System," Proceedings of the 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE),DOI: 10.1109/ic-ETITE47903.2020.452, 2020.

[23] R. D. A. Raj, S. Bhattacharjee, "Mitigation of Voltage And Current Harmonics In Double-Stage Grid-Connected PV Plant," Proceedings of the 2020 International Conference on Computer, Electrical & Communication Engineering (ICCECE),DOI: 10.1109/ICCECE48148.2020.9223024, 2020.

[24] R. D. A. Raj, T. Aditya, M. R. Shinde, "Power Quality Enhancement of Grid-Connected Solar Photovoltaic System Using LCL Filter," Proceedings of the 2020 International Conference on Power Electronics & IoT Applications in Renewable Energy and its Control (PARC), pp. 334-339, DOI: 10.1109/PARC49193.2020.236621, 2020.

[25] R. Opoku, E. A. Adjei, D. K. Ahadzie, K. A. Agyarko, "Energy Efficiency, Solar Energy and Cost Saving Opportunities in Public Tertiary Institutions in Developing Countries: The Case of KNUST, Ghana," Alexandria Engineering Journal, vol. 59 no. 1, pp. 417-428, DOI: 10.1016/j.aej.2020.01.011, 2020.

[26] R. Opoku, G. Mensah, E. A. Adjei, J. Bosco Dramani, O. Kornyo, R. Nijjhar, M. Addai, D. Marfo, F. Davis, G. Y. Obeng, "Machine Learning of Redundant Energy of a Solar PV Mini-Grid System for Cooking Applications," Solar Energy, vol. 262,DOI: 10.1016/j.solener.2023.06.008, 2023.

[27] M. H. Bollen, "Understanding Power Quality Problems," ,DOI: 10.1109/9780470546840, 2010.