*Review* **Optimization of DC, AC, and Hybrid AC/DC Microgrid-Based IoT Systems: A Review**

**Belqasem Aljafari <sup>1</sup> , Subramanian Vasantharaj <sup>2</sup> , Vairavasundaram Indragandhi 2,\* and Rhanganath Vaibhav <sup>3</sup>**


**\*** Correspondence: indragandhi.v@vit.ac.in

**Abstract:** Smart microgrids, as the foundations of the future smart grid, combine distinct Internet of Things (IoT) designs and technologies for applications that are designed to create, regulate, monitor, and protect the microgrid (MG), particularly as the IoT develops and evolves on a daily basis. A smart MG is a small grid that may operate individually or in tandem with the electric grid, and it is ideal for institutional, commercial, and industrial consumers, as well as urban and rural societies. A MG can operate in two methods (stand-alone and grid-connected), with the ability to transition between modes due to local grid faults, planned maintenance, expansions, deficits and failures in the host system, and other factors. Energy storage is the process of storing and converting energy that can be used for a variety of purposes, including voltage and frequency management, power backup, and cost optimization. IoT is designed to deliver solutions for optimal energy management, security protocols, control methods, and applications in the MG, with numerous distributed energy resources (DER) and interconnected loads. The use of IoT architecture for MG operations and controls is discussed in this research. With the use of power grid equipment and IoT-enabled technology, MGs are enabling local networks to give additional services on top of the essential supply of electricity to local networks that operate simultaneously or independently from the regional grid. Additionally, this review shows how hybrid AC/DC MGs are advantageous compared to AC and DC MGs. The state-of-the-art optimization techniques and trends in hybrid MG research are included in this work.

**Keywords:** smart microgrid; optimization; hybrid renewable energy source; internet of things; cost of electricity; information and communication technology

**1. Introduction**

A MG is a group of electrical loads and small-scale generation resources that can meet all or part of the demand. MGs can be built individually (islanding mode) or in groups (connected to an upstream grid). If a MG is linked to the grid system, surplus intrinsic resource generation can be sold to the upstream grid to boost the MG profit. To increase efficacy, the majority of MG-producing units can be employed in a combined heat and power scheme [1]. The overview of MG generating and storage possibilities is presented in Table 1 [2].

**Citation:** Aljafari, B.; Vasantharaj, S.; Indragandhi, V.; Vaibhav, R. Optimization of DC, AC, and Hybrid AC/DC Microgrid-Based IoT Systems: A Review. *Energies* **2022**, *15*, 6813. https://doi.org/10.3390/ en15186813

Academic Editor: Emilio Lorenzani

Received: 20 July 2022 Accepted: 14 September 2022 Published: 18 September 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).


**Table 1.** Summary of MG generation and its storage possibilities.

Numerous hybrid approaches have been developed to examine the combined operation of the MG's micro-sources and storage facilities [11]. The MG administrator is in charge of the internal control of the MG's elements. The operators of the main grid, the market operator, or the regional transmission organization have no monitoring or control over the MG's micro-sources in this circumstance. A portion of the energy may be limited due to the MG's internal restrictions and inherent uncertainty.

The Internet of Energy (IoE) refers to the combination of IoT and MG technologies. The IoE takes advantage of the MG's bidirectional energy flow and information to gather data on power consumption and forecast future activities to improve energy efficiency and reduce net costs [12]. The MG relies on a number of IoT technologies. From the physical to the application layers, such technologies comprise the entire network protocols.

In 2017, the number of internet devices reached 8.4 billion, and by 2020, it is expected to reach 30 billion. The IoT is a system of these units that will communicate and share data. The IoT is at the zenith of its growth stage in the environment of MGs, with smart analytics promising a bright future. Energy-based analytic data sent from users to utilities have the ability to improve MG efficacy and minimize congestion, leading to increased power distribution reliability in a (future) 100% renewable energy paradigm.

The future MG will be made possible by the transition of a device-electric grid into a smart, self-healing bidirectional intelligent system [13]. Energy suppliers seem to be more interested in delivering efficient power, minimizing CO<sup>2</sup> emissions, helping to bring in green energy, and lowering prices while maximizing utility profits with these modern technologies. This IoT-enabled MG enterprise contributes to the global smart city mission. Table 2 shows the primary capabilities of a MG.

**Table 2.** Primary capabilities of a MG.


The MG design [14] necessitates constant device monitoring, examination, and total management of the overall grid, in which large numbers of monitoring equipment of various sorts are placed at several power plants, transmission and distribution regions, and at the customer's side [15]. The IoT is described as collections of physical objects that are linked together via the internet [14,16].

Despite the commitment and availability of IoT technology, a MG will be impossible to accomplish in the future. Interconnectivity via communication devices, such as mobile phones, allow for quick decision-making through social cooperation and lowering application TCO. There are numerous advantages to cloud computing from a financial standpoint, where the TCO of a product is calculated from its acquisition, taking into account both service and running expenditures. The utility receives detailed information from smart meters and sensors, allowing it to prepare a compressed service order and the closest work to be delivered. Once the power goes out in the modern era of IoT and the MG junction, a message from the power line sensor is delivered quickly to the utility providers, who can then monitor the transformer operation. The IoT allows for more seamless activity and interactions between the two parties, resulting in a more effective wireless system. The primary contributions of this paper are to illustrate the benefits of hybrid AC/DC MGs over AC and DC MGs, to discuss the role of the IoT in the design and development of smart MGs, including benefits, challenges, and risks, and to expose a number of technologies, architectural designs, and applications that use the IoT with the goal of preserving and regulating innovative smart microgrids in accordance with contemporary optimization features and regulations. Figure 1 depicts the framework of this study.

**Figure 1. Figure 1.**  Framework of the paper. Framework of the paper.

**Table 2.** Primary capabilities of a MG.

ergy consumer interaction and cost planning has been made.

linked together via the internet [14,16].

Self-healing A MG has the ability to assess, respond, and discover serious flaws very quickly. Smart metering systems are used to

Resist attack The main challenges that a MG can readily combat are cyber-attacks and physical attacks. For MGs, several data con-

The MG design [14] necessitates constant device monitoring, examination, and total management of the overall grid, in which large numbers of monitoring equipment of various sorts are placed at several power plants, transmission and distribution regions, and at the customer's side [15]. The IoT is described as collections of physical objects that are

Despite the commitment and availability of IoT technology, a MG will be impossible to accomplish in the future. Interconnectivity via communication devices, such as mobile phones, allow for quick decision-making through social cooperation and lowering application TCO. There are numerous advantages to cloud computing from a financial standpoint, where the TCO of a product is calculated from its acquisition, taking into account both service and running expenditures. The utility receives detailed information from smart meters and sensors, allowing it to prepare a compressed service order and the closest work to be delivered. Once the power goes out in the modern era of IoT and the MG junction, a message from the power line sensor is delivered quickly to the utility providers, who can then monitor the transformer operation. The IoT allows for more seamless activity and interactions between the two parties, resulting in a more effective wireless system. The primary contributions of this paper are to illustrate the benefits of hybrid AC/DC MGs over AC and DC MGs, to discuss the role of the IoT in the design and development of smart MGs, including benefits, challenges, and risks, and to expose a number of technologies, architectural designs, and applications that use the IoT with the goal of preserving and regulating innovative smart microgrids in accordance with contemporary optimization features and regulations. Figure 1 depicts the framework of this study.

Consumers motivation Consumers can choose their suitable tariffs and more efficiently manage their energy usage. The case for enhanced en-

Improved power quality Constant voltage is the most common consumer demand across all domestic, commercial, and industrial sectors. The MG has the ability to keep a constant voltage, therefore improving power quality.

**Functionality Microgrid Description**

servation strategies have been implemented.

identify faulty circumstances and blackout scenarios.

#### **2. Optimization of MG**

Clean and renewable energy is advancing in order to achieve energy sustainability and harmonious growth in the economy and society. MGs are important tools for implementing clean and renewable energy. MG operation optimization has grown in importance as a study area. This study examines the recent improvements in MG operation optimization.

#### *2.1. DC Microgrid*

A DC MG has a DC bus that provides power to the DC loads coupled to it. Cell phones, internet routers, DVD players, battery-powered vacuum cleaners, wireless phones, and laptops are examples of low-power electronic devices. In a DC MG configuration, resources with DC output are simply coupled to the DC bus [17]. There are few converter elements necessary since these are more DC-generating RESs than AC-generating RESs. It increases the total efficacy of the DC MG.

#### 2.1.1. Concept and Features

In this environment, the use of DC-operated technology in regular life has increased dramatically [18]. DC loads are generally linked to AC inputs because of an absence of independent DC supply networks at the consumer's end. Multiple conversions are required because the AC power is adjusted by converters for various DC load demands. Conversion losses and harmonics created by converters are steadily increasing, contaminating the power grid. The average power loss from these conversion procedures is 10–30% [19]. Regarding the principle of a MG—it was developed in response to an increased usage of DC systems and to handle low-powered DG resources. It complements the development of MG operations and improves the BESS [20]. Figure 2 shows the circuit of DC MG.

DC MG.

**2. Optimization of MG** 

optimization.

*2.1. DC Microgrid* 

2.1.1. Concept and Features

Clean and renewable energy is advancing in order to achieve energy sustainability

A DC MG has a DC bus that provides power to the DC loads coupled to it. Cell

In this environment, the use of DC-operated technology in regular life has increased

dramatically [18]. DC loads are generally linked to AC inputs because of an absence of independent DC supply networks at the consumer's end. Multiple conversions are required because the AC power is adjusted by converters for various DC load demands. Conversion losses and harmonics created by converters are steadily increasing, contaminating the power grid. The average power loss from these conversion procedures is 10– 30% [19]. Regarding the principle of a MG—it was developed in response to an increased usage of DC systems and to handle low-powered DG resources. It complements the de-

phones, internet routers, DVD players, battery-powered vacuum cleaners, wireless phones, and laptops are examples of low-power electronic devices. In a DC MG configuration, resources with DC output are simply coupled to the DC bus [17]. There are few converter elements necessary since these are more DC-generating RESs than AC-generat-

ing RESs. It increases the total efficacy of the DC MG.

and harmonious growth in the economy and society. MGs are important tools for implementing clean and renewable energy. MG operation optimization has grown in importance as a study area. This study examines the recent improvements in MG operation

**Figure 2.** Circuit of DC MG. **Figure 2.** Circuit of DC MG.

If a BESS is attached to standard AC ports, massive conversion losses occur because of multiple conversion processes. As a result, BESS reliability has diminished. However, in a MG, the BESS is (most of the time) run within the DC bus. As a result, the losses experienced in the charge operations are considerably reduced, and the BESS behavior in If a BESS is attached to standard AC ports, massive conversion losses occur because of multiple conversion processes. As a result, BESS reliability has diminished. However, in a MG, the BESS is (most of the time) run within the DC bus. As a result, the losses experienced in the charge operations are considerably reduced, and the BESS behavior in a MG configuration is significantly enhanced. Electricity is presently unavailable to approximately 1.1 billion people [21], the majority of whom live in rural parts of Sub-Saharan Africa and Southeast Asia, as well as, to a smaller extent, the Middle East, Central Asia, and Latin America [22].

## 2.1.2. Optimization in DC MG

Thousands of SHSs [23] have been built in distant areas as a result of rural electrification schemes, typically in areas with no electricity grid, no regular wired telecommunication networks, and (mostly) poor availability by ordinary transport. SAPV systems are SHSs. Typically, crystalline-silicon PV modules are used in these setups. The most typical battery type used in a battery backup unit is lead-acid, and many tiny SHSs use charge controllers with PWM to optimize the charge current to the battery [24]. The lack of appropriate SHS monitoring and, hence, the inability to recognize O&M issues, can result in a severe reduction in the lifespans of PV systems, or even their removal from use [25]. The contributions of different optimization methods for DC MG are discussed in the various research works mentioned in Table 3.


#### **Table 3.** Optimization methods for DC MG.

Grid-connected PV plants typically require large expenditures, and the related data collecting systems allow for monitoring key variables and the execution of required maintenance operations without considerably increasing the overall cost of installation. Nevertheless, it is extremely complex to monitor the functioning of SHSs, owing to the reality that the necessary commercial data loggers are costlier in comparison to the overall system cost. As a result, more precise and independent external sources of data collecting systems must be developed at a smaller cost. Analytical control has progressively been implemented in small PV systems in recent years. Monitoring has been highlighted as one of the variables that lead to the viability of rural electrification programs since these efforts improve the lifetime of the system and reduce PV system failure, enhancing the user's confidence in the system. Table 4 lists the parameters in real-time PV systems that should be monitored.


**Table 4.** PV System parameters to be monitored in real-time.

## *2.2. AC Microgrid*

An AC bus system connects the numerous energy-producing sources and loads in an AC MG network. AC MGs are often made up of dispersed generating units, such as renewables and traditional power production sources, such as engine-based generators. Such distributed generators are linked to an energy storage media, such as BESS, via an AC bus system. DC output is generated by renewable generators, such as solar PV and wind turbines. Through power electronic-based converters, this output can be transformed to AC.

#### 2.2.1. Concept and Features

Wind energy has emerged as an essential alternate energy resource for power generation, owing to the diminishing reserves of global real-world resources and the progressive development of low-carbon and environmental protection principles. Wind energy is useful to the world's natural resources and ecology [40]. It is also conducive to sustainable economic development as a non-polluting and clean energy source [41]. According to studies, wind power generated roughly 12% of global electricity production in 2020 [42]. Wind energy is also expected to account for 22% of the worldwide power supply in 2030 [43]. Wind speed fluctuations and intermittency can have negative impacts on the stability and reliability of power grid operations, resulting in high costs and low efficiency. To increase the accuracy and reliability of WSP, it is critical to build strong prediction techniques. Physical techniques [44], traditional statistical strategies [45], spatial correlation strategies [46], and AI strategies [47] are the four basic kinds [48] of WSP methods that have been established in the last several decades [49]. Figure 3 depicts the AC MG circuit.

#### 2.2.2. Optimization in AC MG

The following are the shortcomings of the forecasting strategies:


time wind energy system are given in Table 6.

4. Despite the usage of alternative methods, AI technologies were thoroughly investigated and are now being utilized to handle complex relationships and make accurate assumptions. These techniques can be used to capture the actual series in non-linear

The contributions of different optimization methods for AC MG are discussed in the various research works mentioned in Table 5; the parameters to be monitored in a real-

**Figure 3.** Circuit of AC MG. **Figure 3.** Circuit of AC MG.

**Table 5.** Optimization methods in AC MG. **Reference Year Optimization Algorithm/Method Contribution Drawbacks** The contributions of different optimization methods for AC MG are discussed in the various research works mentioned in Table 5; the parameters to be monitored in a real-time wind energy system are given in Table 6.

Does not consider a grid-connected PV

system.

[52] 2020 Techno-economical Single-axis and dual-axis solar trackers are used to test and optimize the performance of a stand-alone solar PV power **Table 5.** Optimization methods in AC MG.

patterns.



**Table 5.** *Cont*.

**Table 6.** Factors to be monitored in real-time wind energy (a survey of cyber-physical advances).


#### *2.3. Challenges, Need for a Hybrid AC/DC MG, and Its Implementation*

The main hurdles for successfully implementing hybrid AC/DC generation systems are described in detail in this section. In addition, the current scenario's solutions to the difficulties are offered. The second portion of the section discusses the need for AC/DC MG integration as well as its advantages. Figure 4 shows the hybrid renewable energy generation by source, measured in terawatt-hour (TWh).

[62] <sup>2017</sup> Day-ahead MG EM optimiza-

[63] <sup>2018</sup> Iterative consistency algo-

tion

rithm

*2.3. Challenges, Need for a Hybrid AC/DC MG, and Its Implementation*

generation by source, measured in terawatt-hour (TWh).

[61] <sup>2016</sup> Neuro-fuzzy controller Frequency control without any storage. The structure is not totally interpreta-

of energy, are all part of the target function.

ing agents to achieve a global optimum.

Environmental

Mechanical

Electrical

Temperature

The O&M costs of lithium batteries and fuel cells, the interruptible compensation of interruptible loads, and the price

Increases the AC MG control's durability and flexibility, and only requires little communication between surround-

**Table 6.** Factors to be monitored in real-time wind energy (a survey of cyber-physical advances). **General Parameters Specific Parameters**

Fluid Pressure, level, flow

The main hurdles for successfully implementing hybrid AC/DC generation systems are described in detail in this section. In addition, the current scenario's solutions to the difficulties are offered. The second portion of the section discusses the need for AC/DC MG integration as well as its advantages. Figure 4 shows the hybrid renewable energy

ble.

Sensitive to communication failure.

Wind Humidity Lighting Icing

Positions Speed Angle Stress Strain

Voltage Current Power factor Frequency Faults

Bearings Oil Windings Electronic components


**Figure 4. Figure 4.** Hybrid renewable energy generation. ( Hybrid renewable energy generation. (**Source: Source:** Statistical review of world energy). Statistical review of world energy).

#### 2.3.1. Concept and Features

One of the most intriguing strategies in the evolution of the MG principle in the present distribution network is the hybrid AC/DC MG. Figure 5 depicts a typical hybrid MG structure, with the AC and DC networks. Several devices can be observed in the diagram: PV, fuel cell, a diesel generator, DG and ESS units, VSD, AC and DC loads, etc. Interlinking converters have been used to connect AC and DC sub-grids. This arrangement confines greater interfacing, which in turn minimizes the cost and improves overall efficiency. This architecture is the most appealing option for a future MG framework. *Energies* **2022**, *15*, 6813 10 of 28

**Figure 5.** Hybrid AC/DC MG with interlinking converter. **Figure 5.** Hybrid AC/DC MG with interlinking converter.

(i) Designing objectives with ESS. The following are the most significant advantages of these MGs:

The imbalance between peak demand and generation can be smoothed out by energy storage. As a result, matching electricity generating sources for the setup of an HRES is simply a designer's optimization challenge with several limitations to meet. The design Integration: Devices that are powered by AC or DC are directly interconnected with the fewest interfacing devices possible, eliminating conversion stages and, thereby, energy losses.

techniques must strike a balance between reliability and cost [72]. Supply security issues were also taken into account in the design of the MG design in a recent study [73], which defined a probability adequacy index. Leou [74] analyzed installation costs, O&M costs, Synchronization: Because the generation and storage devices are directly coupled to the AC or DC network, there is no need for them to be synchronized. As a result, the device's control method is simplified.

and income, considering energy price arbitrage for minimizing transmission access costs and postponing facility construction. Sundararagavan [75] conducted a cost analysis of eleven types of energy storage technology for essential applications linked with a wind

assist designers in determining the best design of a hybrid wind–solar–diesel generator– battery power system for independent- or grid-connected applications. Batteries, at relatively low costs and with widespread availability, are the most widely used components in hybrid systems. Tewari et al. [77] looked at a nNaS battery system for transferring power generation from off-peak to on-peak, as well as ramp rate limitation to smooth out wind output. Using model predictive control theory, Khalid [78] devised a new semi-distributed approach that efficiently lowers the BESS capacity required, lowering the total system cost. Brekken et al. [79] employed flow batteries in conjunction with an ANN method to handle the uncertainty in wind output and reduce energy prices even further. The uncertainties in an HRES integrating PV–wind–diesel and a hydrogen-based ESS

The first was a typical strategy relying on the battery's state of charge, and the second was an enhanced ANN algorithm that was evaluated based on energy storage costs. According to the findings, the hybrid battery–hydrogen system storage costs 48% less than a hydrogen-alone system and just 9% less than a traditional battery-only system. Katsigiannis, on the other hand, used NSGA-II optimization and discovered that the hydrogen-based system had higher LCE and emissions than the lead–acid-based system [81]. Choi [82] aimed to reduce battery charging current variations and energy losses in supercapacitors by optimizing a battery/super-capacitor hybrid ESS. Thounthong et al. [83] developed a novel approach for combining a super-capacitor with a hybrid PV–fuel cell

power plant as an additional source and short-term storage units.

farm combined with an electric grid.

were investigated by Giannakoudis et al. [80].

Voltage transformation: Use of such transformers on the AC side can be used to modify voltage levels in a straightforward manner. On the DC side, DC–DC converters are used to conduct the conversion.

Economic feasibility: By adding a power converter to the existing distribution grid and a communication network for linked devices, a hybrid MG can be created. With the use of a power converter, the net prices are higher than that of AC MGs. Although, if the number of connected devices grows, the expenditure will be repaid more quickly because the overall number of interfacing converters will decrease.

#### 2.3.2. Optimization of Hybrid AC/DC MG

Integration of renewable energy resources, such as solar panels, batteries, and other energy storage devices with low voltage systems, will be a viable method for reducing multiple energy conversion losses in the proposed system.

The utilization of RESs has rapidly increased to address the critical concerns of increasing energy demand and global warming [64,65] as a result of increasing energy consumption, which is expected to reach 53% by 2035 [66]. For instance, HMGS [67] delivers energy delivery to rural places where essential T&D amenities are not accessible or costlier to install. When three features are met, DERs can be considered a MG: electrical boundaries are set, an EMS is included, and the power generation capacity must surpass the peak critical load [68,69]. In [70], a novel load flow algorithm for AC/DC distribution systems that makes use of matrix algebra and graph theory is presented. To find load flow solutions, four designed matrices (the loads beyond the branch matrix, the path impedance matrix, the path drop matrix, and the slack bus to other buses drop matrix), as well as basic matrix operations, are used. Similarly, in [71], AC/DC hybrid distribution systems provide a single load flow (LF) model (DSs). The suggested approach can be used in hybrid distribution systems (DSs) that have AC/DC buses and AC/DC lines configured in a wide variety of ways. Additionally, a new DS bus category is presented for LF analysis. There are a number of important factors to consider when making a decision. First, the COE that relates to the cost of operation to fulfill the load demand. Second the stability needs to be at the highest level so that the power supply breakdowns are prevented. The LPSP, which will be discussed in the following sessions, is one of the metrics that can be used to describe HMGS reliability.

#### (i) Designing objectives with ESS.

The imbalance between peak demand and generation can be smoothed out by energy storage. As a result, matching electricity generating sources for the setup of an HRES is simply a designer's optimization challenge with several limitations to meet. The design techniques must strike a balance between reliability and cost [72]. Supply security issues were also taken into account in the design of the MG design in a recent study [73], which defined a probability adequacy index. Leou [74] analyzed installation costs, O&M costs, and income, considering energy price arbitrage for minimizing transmission access costs and postponing facility construction. Sundararagavan [75] conducted a cost analysis of eleven types of energy storage technology for essential applications linked with a wind farm combined with an electric grid.

Chedid et al. [76] developed the core of a computer-assisted evaluation tool that can assist designers in determining the best design of a hybrid wind–solar–diesel generator– battery power system for independent- or grid-connected applications. Batteries, at relatively low costs and with widespread availability, are the most widely used components in hybrid systems. Tewari et al. [77] looked at a nNaS battery system for transferring power generation from off-peak to on-peak, as well as ramp rate limitation to smooth out wind output. Using model predictive control theory, Khalid [78] devised a new semi-distributed approach that efficiently lowers the BESS capacity required, lowering the total system cost. Brekken et al. [79] employed flow batteries in conjunction with an ANN method to handle the uncertainty in wind output and reduce energy prices even further. The uncertainties

in an HRES integrating PV–wind–diesel and a hydrogen-based ESS were investigated by Giannakoudis et al. [80].

The first was a typical strategy relying on the battery's state of charge, and the second was an enhanced ANN algorithm that was evaluated based on energy storage costs. According to the findings, the hybrid battery–hydrogen system storage costs 48% less than a hydrogen-alone system and just 9% less than a traditional battery-only system. Katsigiannis, on the other hand, used NSGA-II optimization and discovered that the hydrogen-based system had higher LCE and emissions than the lead–acid-based system [81]. Choi [82] aimed to reduce battery charging current variations and energy losses in super-capacitors by optimizing a battery/super-capacitor hybrid ESS. Thounthong et al. [83] developed a novel approach for combining a super-capacitor with a hybrid PV–fuel cell power plant as an additional source and short-term storage units.

#### (ii) Sizing objectives.

The optimal sizing of producing units is critical for efficiently and economically utilizing RESs. With an appropriate and complete utilization of the HRES components, the sizing optimization approaches can help to ensure the cheapest investment. Economic and environmental objectives are the most typical goals considered while sizing an HRES. Nehrir et al. [84] examined several methods for system setup, unit sizing, control, and energy management of hybrid systems under investigation. Details about HRES initiatives being implemented around the world were also compiled. Bernal-Agustin et al. [85] and Zhou [86] have also provided their analyses on HRES design, simulation, and control employing PV, wind, and diesel with battery storage. Luna-Rubio provided a review of sizing approaches, including several metrics that were adjusted for maximum performance at the lowest cost [87]. Elma and Selamogullari [88] investigated a stand-alone hybrid system that met the electrical requirements of a residential house.

Numerous research has taken into account economic system parameters, such as LPSP, LCOE, and fuel costs while sizing. With LPSP as the main restriction, Hongxing built and studied a hybrid solar–wind–battery system optimal model for lowering system costs [89,90]. Ekren [91] investigated the difficulty in scaling a PV/Wind/BESS system for use in a GSM station in Turkey. RSM was used to address the sizing problem, and a minimum energy cost of USD 37,033.9 was attained. A siting strategy was devised by comparing this COE to transmission line expenses using the break-even analysis [92].

Using a controlled elitist GA, Reference [93] proposed a triple multi-objective optimization technique to assist developers in taking into account both environmental and economic issues. LCC, EE, and LPSP indicators were merged in the optimal solution. Di-Silvestre et al. [94] established a multi-objective optimum operation using a distinct layered technique. In Reference [95], decision support tools based on the fuzzy technique for order preference by similarity to ideal situation (TOPSIS) and level diagrams were used to build HRES. Arnette [96] devised a multi-objective linear programming approach for HRES planning that allows the decision maker to balance generation costs and emissions under a variety of operating situations. As a result, the wind and solar potential capabilities were assessed separately, taking into account the sizing objectives. Zhang et al. [97] introduced a unique technique for optimizing power dispatch simulations in a PV–battery–diesel system by reducing LCE, which also took into account maintenance costs, capital depreciation costs, pollution damage costs, and fuel costs. Tan et al. [98] introduced a new optimization model for DG siting and size that took into account technical factors including grid VA need, voltage profile, real power losses, and so on.

#### 2.3.3. HRES Control and Energy Management

In order to attain the needed quality power at predefined costs, optimization approaches play a critical part in the functioning of an HRES. Any portion of the HRES can benefit from optimization. The main functional areas for optimization are generation controls such as power dispatch control, energy management decision-making controls, operation controls for power quality and cost control, and MPPT control systems. The following are some examples of the state-of-the-art in control and management.

(i) Power quality and cost control.

Power conditioning devices such as STATCOM and quality management procedures in distribution systems increase power quality [99–101]. It is critical to install them in the best possible location and execute them properly in order to save money and improve efficiency [102]. Serban [103] devised a system for optimization and testing of the frequency control mechanism in MGs, using BESS. By incorporating an ESS in small, isolated power systems, Sigrist [104] calculated the economic advantage of primary frequency control reserve and peak-shaving generation, resulting in a total cost savings of 23.2 Mio €/year and an internal rate of return of 7.25 percent. In Agios Efstratios, Greece, Vrettos et al. [105] investigated the infiltration of WT-ESS into an emerging diesel unit. Applying GA to optimize the LCOE, it was discovered that a 10–15% reduction in LCOE could be obtained with a 75% RES saturation level. In 2013, Zhao [106] addressed the lifetime properties of lead–acid batteries while doing multi-objective optimization to minimize power generation costs and increase the usable life of lead-acid batteries for a similar HRES MG. It has been demonstrated that a higher RES penetration level might result in a 30% drop in total expenditures. The control objectives defined by Younsi et al. [107] are to fulfill the power sought by the AC grid, control the transfer of energy among the hybrid system and the AC grid, optimize the use of wind energy, and minimize fuel costs of diesel generators. Arabali et al. [108] applied GA to a hybrid system that served a single HVAC load, analyzing the cost and efficiency of operation during different scenarios. With the introduction of commercial power system simulators that include powerful analytical and visualization tools, studying power flow controls in a MG has become considerably simpler [109,110].

(ii) Power dispatch control.

In power system applications, such as economic dispatch, unit commitment, and generation scheduling, optimization approaches are becoming increasingly popular. Conti et al. [111] developed an optimization approach for DG and ESS dispatch in a medium-voltage islanded MG with the goal of lowering emissions and operational costs. Zhang et al. [112] proposed a unique power scheduling technique for reducing utility costs of dispatchable loads, worst transaction costs due to renewable source uncertainty, and generation and storage costs. When combined with an HRES, CHP significantly increases efficacy and pays for itself. A CHP-based DG MG with ESS and three different forms of thermal power generation units and DRPs was studied in Reference [113]. Maa et al. [114] conducted a viability analysis on a residential MG system with a hybrid PV–WT and CHP generator. For power dispatching challenges, most new methods use commercial simulators, such as Power World [109]. It simplifies inquiries and the evaluation of complex market policies. MILP was a commercially available solver-based method that avoided the use of complex heuristics or decomposition methodologies [115].

#### (iii) Energy management control

The efficiency of HRES subsystems can be improved by appropriate resource management [116,117]. Zhao et al. [118] adjusted the reactive power output of a wind farm and the network infrastructure at the same time to reduce system actual power losses and bus voltage deviations, resulting in enhanced power control and voltage profile. Trifkovic et al. [119] used decentralized adaptive model prediction control and decisionmaking techniques to describe a power management strategy for a wind–PV–electrolyzer– fuel cell integrated standalone system. It was discovered that operating the electrolyzer at a lower power level increased the efficiency of the renewable energy produced, resulting in more hydrogen production. Table 7 shows the optimization methods used in a hybrid AC/DC MG.

CAES results in a 43% higher operating profitability and 6.7% less net load serving costs, even when capital expenses are not taken into account. For enlightening energy management techniques, a mixture of optimization techniques and AI methods has also been tried. Multi-objective smart power management tries to reduce a MG's operating costs and emissions while taking into consideration pre-operational variables such as future renewable energy supply and load demand. For optimal operation of an HRES system, controlling variables coming from renewable generation and load demand projections are also examined. Hong's PEM was used by Mohammadi et al. [120] to optimize a MG by modeling the uncertainty in load demands, market prices, and renewable energy generation. In Reference [121], rather than using Hong's estimate, an optimal stochastic approach was used, which included the usage of probability density functions for each unknown parameter and roulette wheel mechanism scenarios. The stochastic approach captured roughly three times the number of uncertainties as the deterministic approach.

**Table 7.** Optimization methods in Hybrid AC/DC MG.



**Table 7.** *Cont*.

## 2.3.4. Future Prospects of Hybrid AC/DC Microgrid

The flexible control capabilities of AC/DC hybrid DER systems are further utilized as one of the development directions of the power distribution system for the energy internet. In future studies, researchers might find great interest in investigating cooperative planning with gas, heat, and cooling systems within the context of multi-energy complementarity [141]. The important principles for the futuristic approach in an AC/DC microgrid environment for a smart and intelligent system with uninterrupted, secure, and safe power flow are listed below [142].

