Operation and Assessment of a Microgrid for Maldives: Islanded and Grid-Tied Mode

Abstract

This research work examines the prospect of a dispatch strategy governed hybrid renewable energy microgrid for the proposed location in Maldives for both off and on grid conditions. The techno-environmental-economic-power system responses of the proposed microgrid have been evaluated. The techno-environmental-economic analysis of the proposed microgrid has been conducted utilizing HOMER Pro and the power system response analysis has been conducted using DIgSILENT PowerFactory software platforms. The evaluation shows that, for both on and off grid modes, cycle charging strategy has the worst performance having net present costs (NPC) of $132,906 and $147,058 and cost of energy (COE) of 0.135 $/kWh and 0.213 $/kWh respectively. During on grid mode, generator order performs the best having NPC of $113,137, COE of 0.166 $/kWh. In off grid mode, load following strategy performs the best with NPC of $141,448 and COE of 0.024 $/kWh. The active power and voltage responses of the microgrid shows the stable operation of the proposed system by implementing dispatch techniques and voltage Q-droop and input mode P-Q controller. A comparison section is also presented for demonstrating the significance of the research work. The research work has been conducted considering a location in Maldives but provides an overall idea about establishing a microgrid in anywhere in the world having similar meteorological and load conditions.

1. Introduction

The rapid growth of the demand of electrical energy all over the world cannot be satisfied by the conventional fossil fuel based generation for long. As the conventional sources of energy are going to be diminished very soon, they are injurious to the environment and due to their higher prices, the world is diverting to the renewable based alternatives [1]. Renewable energy based eco-friendly alternatives offer a better and sustainable solution with negligible environmental and ecological effect. However, still today, some of the major disadvantages of the renewable energy based sources confine their usage to a limited scale. The higher cost, lower efficiency, larger space requirement and intermittent nature still discourage people to switch to renewable based solutions [2]. The inconsistency in supply (like inconsistent solar irradiance and wind speed) makes the utilization of renewable resources more challenging. For this purpose, a very well planned energy management system for a hybrid renewable energy systems (HRES) is necessary for the renewable resources to perform in an optimum fashion.

Before establishing any HRES, it is very much important to perform a comprehensive performance test of the proposed system for the efficient operation and integration [3]. The optimum operating configuration of different microgrid components and determination of estimated various costs is important for the reliable and sustainable operation of the HRES. Many software tools are readily available for assessing the design of a HRES and ensuring the optimum performance of the system. Among many other software simulation platforms for HRES designing, HOMER Pro is one of the finest, simple and a very well accepted one within the researchers’ community [4]. HRES can be efficiently integrated within a microgrid facility where the dispatchable and non-dispatchable generators and loads are interconnected in a small scale [5].

Microgrids can offer optimized generation and consumption of electrical power in a small scale on top of dispatch strategy based control. Within a microgrid it is possible to ensure distributed generation as well as higher penetration of renewable sources [6]. Dispatch strategy in this regard works on the optimum dispatch of the electrical power for the load demand satisfaction offering reduced system costs and system performance inside a microgrid [7]. Researches around the world have focused in the optimum designing and analyzing the microgrid considering different dispatch strategies [8,9]. A PV/WT/Hydro Kinetic Turbine/DG based hybrid system for remote island electrification is proposed by researchers in [8] which is designed for grid isolated operation. For two dispatch strategies (load following (LF) and cycle charging (CC)) the performance of the proposed microgrid have been analyzed and it has been determined that the CC has superior performance than LF for the specific location and load in terms of present cost. In terms of harmful gas emission LF performed better on the other hand. In [9], an optimal design of PV/Wind/Battery/diesel based microgrid is proposed for the electrification in the location of Rohingya refugee camp in Bangladesh. The work presents the techno-economic-environmental perspectives of the design as well as the estimated power system (voltage and frequency) responses of the hybrid system. Here, for the proposed location and load profile, LF was found to be the best strategy among the others offering about 89.08% less energy cost than other designs. An analysis to find the most appropriate dispatch approach for telecommunication and domestic load demands in India, has been presented in [4]. The analysis specially focuses on the technical, environmental, economical as well as social (TEES) parameters in finding the optimum solution. HRES performs better than other dispatch systems when using a predictive dispatch approach. To arrive at a more workable design, a sensitivity analysis of the HRES’s costs for components, wind resource, load profile, and macroeconomic factors, solar radiation, is also carried out in the research work. A distributed economic dispatch methodology has been proposed by researchers in [10]. The methodology has been specially modelled for multi-storage device systems. It used deconstructed formulations for resource dispatch. The study offers further study on autonomous energy and power exchange scheduling for each of the storage units. Independent and decoupled dispatching formulations has also been utilized for any other resource that does not offer support services for power following. According to numerical findings, this dispatching method can operate at the lowest possible cost to match centralized dispatching.

Dispatch strategies have overall impact on the microgrid system’s techno-economic-performance parameters [11]. Due to this much impact, performance of dispatch strategy controlled isolated hybrid microgrid on basis of techno-economic-environmental and power system responses like microgrid’s voltage and frequency is evaluated in [12]. The researchers in [13] have considered two different microgrid configurations to evaluate their techno-economic-environmental performance under the control of different power dispatch strategies. In this work the cost for energy has been optimized and a comparison between the results has also been done to find out the optimal hybrid system. In [14] the performance evaluation of an optimum grid isolated hybrid microgrid is presented for a location in Iraq. A new dispatch control strategy is proposed in this research work utilizing the MATLAB Link tool within HOMER platform. The performance of the proposed architecture has been also compared with the built in CC (cycle charging) dispatch strategy of HOMER and a significant amount of reduction in different costs and technical features. The other dispatch strategies like LF (load following), PS (HOMER predictive dispatch), CD (combined dispatch), and GO (generator order dispatch) strategies have not been considered in the study. As grid isolated microgrids have higher instability risks, considering this issue, a two layers EMS (energy management system) is proposed by researchers in [15]. The analysis has been conducted for a HRES in off grid mode with energy storage and non-controllable and controllable generation units.

PV/Diesel/Wind power based microgrid has been proposed by researchers in [16] proposing an optimal dispatch strategy appropriate for grid tied mode of microgrid operation. In this work, a game theory based demand response algorithm has been used for microgrid’s energy management. This work considers the grid tied mode along with a sensitivity analysis and ignores the islanded mode of operation. As the grid extension seemed more expensive, the performance of a grid isolated hybrid system based on dispatch strategies has been analyzed in [17]. The core focus of this study has been the evaluation of the performances of Li-ion and Lead-acid batteries for various microgrid combinations for different dispatch controls. The study has also confined their analysis to only the off-grid mode, not taking the grid tied mode into account. A study for assessing optimum design of a grid connected HRES has been done in [18]. The study considers the optimum design of the proposed microgrid utilizing HOMER platform. Also, the study evaluates the design by considering the active power response and active power losses for the HRES. The consideration of different dispatch strategies has been ignored in the research which is the core research gap of the study.

A dispatch strategy based off grid HRES design and evaluation has been presented in [19]. PSO (particle swarm optimization) based optimal dispatch solution has been proposed for distributed microgrids by the researchers in [20]. The economic feasibility and sustainability of hybrid systems has been investigated in [21] in conjunction with HOMER and DIgSILENT PowerFactory. The load following dispatch control had been adopted while finding optimal solution in HOMER. A novel dispatch methodology has been developed by the researchers in [22] utilizing the MATLAB link feature within HOMER Pro platform. The HRES has been optimized to work in a grid connected mode. The proposed strategy has been compared with other built in dispatch strategies within HOMER and a significant amount of improvement in cost perspective has been obtained. A technical and economic feasibility assessment of the microgrid based electrification scheme has been proposed in [23]. For the assessment of different microgrid configurations HOMER has been adopted. The microgrid’s voltage loss and drops have also been estimated as technical factors of the system. DIgSILENT PowerFactory has been implemented in this regard.

Dispatch strategy based analysis can be implemented with both off-grid and grid tied modes of microgrids. The presented literature in this domain of research are mainly focusing on the islanded mode of operation where, a study of both the modes is important. Some of the researches consider the grid tied mode of study where the proposed microgrid is connected to the conventional utility grid. But the techno-economic analysis based on both grid tied and isolated mode along with power system response based assessment on top of dispatch strategy is not much available in research domain. For this purpose, in this research work, the optimum performance of the modelled microgrid is investigated in terms of dispatch strategies for both grid isolated and grid tied modes to demonstrate their performances in terms of techno-economic-environmental and electrical power system performances. For assessing techno-economic-environmental aspects, HOMER Pro HRES simulation platform has been implemented and for power system performance assessment, DIgSILENT PowerFactory is integrated. Finally the proposed work has been suitably compared with other available works in this research domain to highlight the significance of this proposed work. This research might help in estimating the optimum performance of the proposed microgrid and the techno-economic-environmental aspects as well as securing a stable and sustainable hybrid HRES system for uninterrupted power supply comparing both grid tied and off grid mode.

The residual part of this research article is decorated as: Section 2 describes the methodological approach, Section 3 illustrates the findings and Section 4 finally concludes the research work.

2. Methodology

This section clarifies the methodological approach adopted in this research work with the help of several relevant subsections. The HOMER Pro platform was utilized to estimate the optimal system combination with related techno–economic–environmental perspectives under five dispatch strategies. After that, the optimum system configurations were integrated into the DIgSILENT environment to evaluate the active power and voltage responses for the HRES under different dispatch methods. The following subsections will help to clarify the methodology of the research work.

2.1. System Configuration

The modeled HRES i.e., microgrid system was composed of a wind turbine (WT), solar PV system (SPVS), emergency backup diesel generator (DG) unit, battery storage device (BSD), loads, necessary inverters/converters and controllers, as illustrated in Figure 1. The 230 V, 50 Hz conventional utility grid was tied with the microgrid model when the microgrid was in grid-connected mode. In grid-isolated or islanded modes of microgrid operation, the utility grid was kept disconnected from the HRES. The SPVS and BSD were connected to the DC bus. Necessary control, conversion and filtration were performed to meet the system requirements. The output from WT was rectified and put through filtration and rectification to be connected to the DC bus. Although the output from PMSG was AC, due to the intermittent nature of wind speed, it was not directly fed to the AC bus. The DG and AC loads were connected to the AC bus. The wind speed and solar irradiation measurements were borrowed from [24]. The proposed site chosen for this research work is located in Maldives near 3.2028° N, 73.2207° E with a load demand of 165.59 kWh/day and 23.31 kW peak demand.

2.2. Control Approaches

In this research work, while modeling the HRES for the proposed location and load profile, a total of five dispatch control strategies were implemented within the HOMER Pro platform. HOMER predictive dispatch (PS), cycle charging dispatch (CC), load following dispatch (LF), generator order dispatch (GO) and combined dispatch (CD) were taken into account. After that, according to their performance, the best and worst control cases for the microgrid were determined. After examination, it was found that, for our case, the CC strategy is the worst case for both the grid-tied and off-grid modes. GO has the best performance in grid-tied mode and LF has the best performance in off-grid mode. So, to keep things simple, in this section, only the best and worst dispatch cases are discussed.

When using the CC approach, the generator always operates at maximum capacity. So, in this instance, surplus electricity is generated. The storage unit is charged using the extra electricity. This system’s operating method is the same as the one used by systems that employ LF dispatch. However, this method differs from the LF plan in that the generator operates at its full rated capacity to provide the net load and charge the battery with additional energy when it is turned on. Figure 2 depicts the flowchart for the WT/SPVS/DG/BSD HRES CC dispatch technique [25].

The GO strategy selects the quickest generator combination among the specified generator combinations to meet the load requirement. Figure 3 illustrates the GO dispatch methodology flowchart for the WT/SPVS/DG/BSD HRES. This dispatch employs the first order in this figure to reach the necessary operating capacity. If this dispatch enables using an earlier row, the battery might be applied to the system. The GO dispatch employs the storage device to handle the load when it is practical to do so. To put it another way, the SoC only provides power up to its maximum capacity before selecting a GO combination to provide the remaining electrical demand. If the demand is greater than the generation of renewable energy, the generators or other components charge the battery as much as they can at each time step. This technique is simpler than the previous ones due to the generators’ obvious priority; as a result, it could handle the hybrid system even in emergency situations [26].

The generator only supplies the exact amount of electricity needed to meet the load demand in the LF approach. The extra power (if any) is then delivered for battery charging, which is only accomplished by employing renewable sources once the principal loads have been satisfied.

The WT/SPVS/DG/BSD HRES LF dispatch methodology flowchart is presented in Figure 4.

2.3. DIgSILENT PowerFactory Microgrid Model

The DIgSILENT PowerFactory model shown in Figure 5 was utilized to replicate the HRES-based power system network to extract the voltage and power responses of the microgrid for the evaluation of the microgrid on the basis of active power and voltage responses. The model shows that the considered microgrid is a 4-bus system with 3 low-voltage (LV) buses. A total of 5 delta–delta transformers were utilized between different buses for necessary voltage level conversions. Within buses LV 3 and LV 4, necessary DC/AC conversions were incorporated. The loads were connected to bus 2, which has a voltage profile of 230 V AC and a frequency of 50 Hz. For the various components within the microgrid model such as PV, wind turbine etc., necessary control algorithms i.e., voltage q droop (Q-V droop) and input mode P-Q controller were implemented within the DIgSILENT PowerFactory platform.

Droop controls are utilized to control the drooping characteristics within a system. Droop controls can be effectively used in avoiding circulating current between distributed generation units connected in parallel within a microgrid [27,28]. When the equivalent impedance linking distributed generators to the utility grid within a microgrid is inductive, Q-V droop control is typically utilized [29]. The research in [30] suggests a reliable and simple to use droop control mechanism that is ideal for high-voltage microgrids. Equations (1) and (2) demonstrate the basic theoretical fundamentals for Q-V droop control [30]:

$$P=\frac{EV}{X}\delta$$

(1)

$$Q=\frac{V}{X}(E-V)$$

(2)

where P and Q refer to active and reactive powers, respectively. $\delta$ refers to the rotor angle, X is inductive reactance, E is the voltage output of generator and V is the voltage for AC bus.

As a result, distributed generators’ output voltage may be controlled by implementing Equation (3) [30]:

$$E=E^{*}-nQ$$

(3)

where $E^{*}$ refers to the rated voltage of the microgrid and n refers to the voltage droop gain.

Output frequency can be governed by Equation (4) [30]:

$$\omega ={\omega }^{*}-mP$$

(4)

where ${\omega }^{*}$ refers to the rated frequency of the microgrid; $\omega$ and m refer to the output frequency and frequency droop gain, respectively.

Now, the active power from the various distributed generators i (where i = 1, 2, …up to the no. of generators) can be formulated as Equation (5) [30]:

$$m_1{P}_1=m_2{P}_2=\dots .=m_i{P}_i$$

(5)

In islanded mode, it is crucial to regulate the voltage and frequency of each power converter connected to each distributed generator (referred to as the VF control). On the other hand, each distributed generator must have its output active and reactive powers regulated or P-Q controlled in grid-connected mode [31]. In this context, input mode P-Q control was adopted in this study for grid-tied mode analysis within the DIgSILENT platform.

3. Results and Discussion

This section summarizes the findings of this research work and related discussion. This section is divided into several subsections to properly illustrate the findings.

3.1. Techno–Economic–Environmental Evaluation

The techno–economic–environmental prospects of the modeled HRES are summarized in

Table 1. Results of optimization process.

Grid-Tied Mode
DS	PV	WT	DG	Battery	Grid	Converter	NPC ($)	COE	Operating	Initial	RF	EP	ES	CO${}_2$
	(kW)	(kW)	(kW)	(kWh)	(kW)	(kW)		($/kWh)	Cost ($/yr)	Capital ($)	(%)	(kWh)	(kWh)	kg/yr
CC	25	4	1	1	999,999	18.8	132,906	0.135	7463	36,433	55.9	33,717	15,985	21,309
CD	No output results. HOMER was unable to find a solution
LF	25	4	1	1	999,999	18.8	132,522	0.135	7433	36,433	56.1	33,438	15,675	21,133
PS	No feasible solution. HOMER was unable to find a solution
GO	40	1	1	134	999,999	14.6	113,137	0.166	2086	86,175	100	0	0	0
Grid-Isolated Mode
DS	PV	WT	DG	Battery	Grid	Converter	NPC ($)	COE	Operating	Initial	RF	EP	ES	CO${}_2$
	(kW)	(kW)	(kW)	(kWh)	(kW)	(kW)		($/kWh)	Cost ($/yr)	Capital ($)	(%)	(kWh)	(kWh)	kg/yr
CC	25	15	2	87	-	15.5	147,058	0.213	5508	75,848	81.3	-	-	8107
CD	25	8	3	81	-	13.2	137,398	0.200	5635	64,553	78.4	-	-	9348
LF	25	13	3	82	-	14.9	141,448	0.024	5180	74,484	83.3	-	-	7547
PS	35	1	1	104	-	11.3	120,410	0.175	3807	71,191	89.7	-	-	4404
GO	40	1	1	134	-	14.6	113,137	0.166	2086	86,175	100	-	-	0

In this table, DS = dispatch strategy, RF = renewable fraction, EP = energy purchased, ES = energy sold; red color defines worst case, green defines best case.

. In grid-tied mode, where the HRES is connected with the utility grid, the GO strategy performs best on the basis of the minimum NPC of USD 113,137, operating cost of 2086 $/yr and zero emissions, and maximum renewable fraction (RF) of 100%. The GO strategy makes the grid-tied HRES a self-dependent one with no sold or purchased energy to or from the grid. The operating characteristics of the GO strategy can explain the result. In the GO strategy, the predefined generator combinations are considered and the fastest among them are brought into operation to satisfy the load demand. Here, the storage units are utilized efficiently whenever possible, to supply the load profile. The stored energy within the battery is utilized to satisfy the load demand in the first place; after that, the unsatisfied load is fulfilled by the GO combinations. An attempt is made to charge the batteries by utilizing the renewable sources in the first place. For this reason, the battery size and SPV size are estimated as the largest in the strategy. As the renewables and storages were utilized, the RF was 100% and emission was found to be zero. For the specific load demand in the HRES considered, the optimum configuration of 40 kW PV, 1 kW WT and 134 kWh battery was found to be just enough to satisfy the load demand; thus, no energy needs to be purchased or sold from/to the conventional grid.

In the PS strategy, the future load and resource profile is estimated to satisfy the load demand. Here, as the HRES is grid tied, any shortage in the microgrid can be filled up with the utility grid; thus, the PS strategy within HOMER did not support grid connectivity and, hence, offered no feasible solution in this circumstance. The CD strategy chooses between the LF and CC strategies in terms of the demand and resource profile. In CC, the charging of storage depends largely on the difference between the future renewable production and future net load profile. In the HOMER Pro platform, in grid-tied mode, no feasible solution is achieved in this circumstance [25].

In grid-tied mode, the CC strategy conversely demands the highest NPC, operating cost, energy purchase from grid and carbon emission. The RF was found to be 44.1% lower than that of the GO strategy. Although the optimum component sizes for CC were found to be comparatively lower than GO, according to the other important parameters, CC has the worst performance in grid-tied mode. CC has 17.5% more NPC and 2.6% more operating cost than that of GO. In off-grid mode, CC also offers the worst performance having larger component size requirements and carbon emission, the largest NPC and COE demands, and lower RF. In grid-isolated mode, as there is no provision to grid availability, energy is neither sold to nor purchased from the utility grid.

In this off-grid mode, as Table 1 shows, the best performance is offered by the LF strategy. LF offers a smaller component size than CC. LF offers 3.81% less NPC than CC in off-grid mode. Moreover, LF has 88.7% less COE than the CC strategy. The operating cost and initial capital are both significantly reduced in the LF compared with the CC dispatch. LF offers 6.9% less carbon emission with 2.4% more renewable penetration. The other dispatch strategies on the other hand have moderate performances between CC and LF. The reason for the lower performance of CC is that in this dispatch method, the generator is always operated with its full limit. In contrast, the generator in LF only delivers as much power as is necessary to meet the load’s demand.

While differentiating between the off- and on-grid mode responses of the proposed microgrid according to the power system responses (active power responses and voltages, which will be demonstrated in the later sections), it was observed that both the best and worst strategies have similar stable and within ‘tolerable limit’ responses. Thus, both off- and on-grid designs can be declared feasible in terms of active power and voltage outputs. The techno–economic–environmental performances for both cases have different impacts, as shown in Table 1. As both techno–economic–environmental and power system responses were considered as the criteria for determining the best and worst dispatch strategies for the proposed microgrid for the specific load profile, and both the strategies have similar power system responses, the selection of best and worst cases was influenced mostly by techno–economic–environmental performances of the dispatch methodologies mentioned in Table 1.

3.2. Active Power Responses

The active power responses for the microgrid components were estimated by the DIgSILENT model analysis shown in Figure 5. Figure 6 shows the active power responses for the CC strategy (previously determined as the worst case for the proposed HRES and load condition) in grid-tied mode. The power responses were recorded for a time frame up to 5 s. From the illustration, it can be seen that the WT, DG, BSD and SPVS have stable power responses in the positive region after some fluctuations in the transient state. Figure 7 shows the active power responses for GO in grid-tied mode, which was determined as the best scenario, for a 5 s time frame. The responses seem stable and in the positive region, meaning the WT, BSD, DG and SPVS are supplying active power to the load demand.

In grid-isolated mode, the active power performances for WT, DG, BSD and SPVS under CC control are illustrated in Figure 8. Figure 9 shows the performances under the best dispatch strategy of LF in this mode. The responses observed are stable and within their prescribed limit. The responses were recorded for the time up to 5 s. It was found that for both modes (grid-tied and isolated) for the specific location, both the best and worst strategies have similar power system responses in terms of active power; thus, they both have equal points in the race to be the best strategy in terms of active power responses. The results in the mentioned figures demonstrate the stable and feasible active power responses, which both strategies perform (for both best and worst cases, which were determined according to the techno–economic–environmental performance presented in Table 1).

3.3. Voltage Responses

It is important to keep the voltage responses within a stable and permissible limit. High PV penetration causes a number of voltage problems, including overvoltage at the feeder’s end, voltage sag, flicker and swell; among them, overvoltage is the main problem [32]. The American standard (5% of nominal voltage, $V_N$), European standard (10% of $V_N$), Australian standard (6%/+10% of $V_N$), German standard (+3% of $V_N$) and Canadian standard (6% of $V_N$) are five regularly used voltage standards that were reviewed by Fatima et al. [33]. As rising PV penetration without adequate technical analysis violates the overvoltage range in the relevant network, a limitation on the aggregated PV capacity is required. Another problem is flickering, which develops as a result of the overabundance of renewable energy in the system and spreads to the feeder for distribution [34]. High PV penetration in a MG results in low system inertia and variable PV output, which frequently causes voltage sag and swell issues [35]. So, it is important to keep the voltage responses within a stable and acceptable range.

Figure 10 shows the voltage responses for the worst case of CC control under grid-connected mode. The WT, DG, BSD and SPVS responses were collected up to 5 s. Until 0.5 s, the responses had some oscillations; after that, they were stable. The WT and BSD had approximately zero and SPVS and DG had 1 p.u. voltage in the steady-state condition. According to CC, whenever the generators are operated, they are operated in its rated operating condition.

In a similar fashion, Figure 11 shows the best case GO strategy performance in grid-tied mode. The WT, DG, BSD and SPVS have stable responses from 0.5 s onward.

The CC strategy has similar stable and within-limit voltage performance in off-grid mode as in grid-connected mode, as Figure 12 demonstrates. Figure 13 shows the output for LF (the best case) control in grid-isolated mode. The oscillations before 0.5 s die out in the steady-state region, offering a stable response from 0.5 s onward.

From the analysis, it was found that for the proposed microgrid, for the considered load profile, both the off- and on-grid microgrids have similar responses in terms of voltage outputs. As this power system’s response criteria seems similar for both the dispatch cases, the impact of techno–economic–environmental performance was the most influential in determining the best and worst cases.

3.4. Comparison with Other Works

This part of our research presents a suitable comparison between the proposed work and other available works in this research domain.

Table 2. Comparison with other works.

Reference	Parameter	Grid-Tied Mode		Grid-Isolated Mode
		Proposed	Referred	Proposed	Referred
		Work	Work	Work	Work
Sood et al., 2020 in [36]	NPC	USD 113,137	-	USD 141,448	-
	COE	0.166 $/kWh	-	0.024 $/kWh	-
	Operating Cost	2086 $/yr	2298 $/24 h	5180 $/yr	2278 $/24 h
	Emission CO${}_2$	0 kg/yr	-	7547 kg/yr	-
	RF	100%	-	83.3	-%
	Power System Study	Active power	Active	Active power	Active power
		Voltage	power	Voltage
	Control Algorithm	Dispatch Strategy (GO)	PSO	Dispatch Strategy (LF)	PSO

shows the comparison of the proposed work with [36]. Here, the authors studied grid-tied and -isolated modes of HRES with PSO with active power responses. The comparison shows that different important parameters have not been studied by the referred work and, by implementing dispatch control, a significant of reduction can be achieved.

Table 3. Comparison with other works continued.

Reference	Parameter	Grid-Tied Mode		Grid-Isolated Mode
		Proposed	Referred	Proposed	Referred
		Work	Work	Work	Work
Ishraque et al., 2021 in [37]	NPC	USD 113,137	-	USD 141,448	USD 163,237
	COE	0.166 $/kWh	-	0.024 $/kWh	0.224 $/kWh
	Operating Cost	2086 $/yr	-	5180 $/yr	4327 $/yr
	Emission CO${}_2$	0 kg/yr	-	7547 kg/yr	4515 kg/yr
	RF	100%	-	83.3%	90.6%
	Power System Study	Active power,	-	Active power,	Active power, voltage
		Voltage		Voltage	frequency
	Control Algorithm	Dispatch Strategy	-	Dispatch Strategy	Dispatch Strategy
		(GO)		(LF)	(LF)

on the other hand shows the comparison of the proposed work with [37]. The researchers did not study the grid-tied mode of operation in the literature, although a full techno–economic–environmental and power system response (voltage, frequency, power) based assessment on top of dispatch control was presented in the work. So, the present research work has superior contribution and significant improvement compared with the previous works. As the referred works considered other specific locations with specific load demands, for some parameters the comparison might not replicate the scenario accurately. While choosing referred articles, we attempted to compare the proposed work with other works in which the proposed site and load demand are closely related to each other.

4. Conclusions

As the amount of fossil fuel is going to be diminished very soon, the prospect of renewable-energy-based electricity generation has become an important research topic. This research work performs a comparative analysis based on the performances of different dispatch strategies implemented on solar PV/diesel generator/battery storage/wind turbine based hybrid microgrid system in both grid-tied and off-grid modes. The performance evaluation of the different dispatch methodologies was conducted based on the techno–economic–environmental perspectives. The techno–economic–environmental performances were estimated using HOMER Pro. For the best and worst dispatch methodologies, the active power and voltage responses were anticipated using DIgSILENT PowerFactory microgrid simulation software. The analysis shows that for the proposed microgrid location in Maldives, the generator order strategy has the best performance in grid-tied mode and load following shows the best performance in grid-isolated mode. Cycle charging is observed as the worst strategy in both operation modes. In grid-tied mode, the CC strategy demands the highest NPC, operating cost, energy purchase from grid and carbon emissions. The RF was found to be 44.1% lower than that of the GO strategy. CC has 17.5% more NPC and 2.6% more operating cost than that of GO. In off-grid mode, the best performance is offered by the LF strategy. LF offers 3.81% less NPC than CC in off-grid mode. Moreover, LF has 88.7% less COE than the CC strategy. The operating cost and initial capital are both significantly reduced in LF compared with the CC dispatch. LF offers 6.9% less carbon emission with 2.4% more renewable penetration. The voltage and active power responses for the microgrid components exhibit satisfactory performance on the basis of stability measured in p.u. Furthermore, the comparison of this proposed work with other works demonstrates the significance of this study. This research work bridges the gap between grid-tied and off-grid modes of operation of a hybrid system by comparing their mutual performances, which might help in deciding the optimum choice for electrification through a hybrid microgrid.