Optimal Design and Model Predictive Control of Standalone HRES: A Real Case Study for Residential Demand Side Management

Conventional electricity generation is one of the greatest sources of CO2 emissions. For a successful transformation of conventional energy systems into non-polluting and renewable energy systems, technology-focused traditional systems and economics must be combined for a more accurate holistic viewpoint with consideration of socio-political, technical, economic and environmental factors. Hybrid energy systems are considered the most feasible solution to the stochastic nature of renewable energy resources (RERs). Different renewable sources such as wind, solar, and hydrogen fuel cells can be integrated to form hybrid systems. An energy management strategy (EMS) is a strategy for power flow coordination among different components, by considering power demand and other constraints. The choice for an accurate EMS is the key element of a hybrid system as it is instrumental in providing an optimum solution of the hybrid system design and operation management. The objective of the optimization is to find suitable configurations for cost-effective solutions. Optimization and EMS must be treated as one entity to completely understand the system design. This study focuses on a techno-economic analysis with an optimized sizing of a hybrid renewable energy system (HRES) components to meet the residential load demand of a specific area in Pakistan. Nine different scenarios based on the PV-wind-diesel-BSS-converter system are investigated in terms of total net present cost (TNPC), Levelized cost of energy (LCOE), and greenhouse gas (GHG) emissions to find the optimal system design. HOMER Pro software is used to develop the HRES model and for simulation analysis, with optimal sizing of each component for an economical solution. Simulation studies established that PV-wind-BSS-converter is the best suitable choice for the given location, and the optimal component sizes were determined. The TNPC of this system is $47,398 and the LCOE is $ 0.309/kWh. This represents an 81.7 % decrease in overall cost, compared to the base case (diesel only) and a 100% reduction in harmful gases while satisfying 100 % of the energy requirement with a 63.9 % of the surplus. MATLAB/Simulink model is developed for the optimum HRES system design. Its validity is tested by maintaining bus voltages (dc and ac), the secure operation range of storage SOC and real power balance among different components of the hybrid renewable energy system (HRES), and an effective ac voltage, irrespective of external perturbations. Model predictive control (MPC) is regarded as a high-performing algorithm. Since power converters are largely applied in microgrids (MGs), the problem formulation with MPC for a reconfigurable bidirectional voltage source converter (VSC) is applied in this work for hybrid MG. The inevitable fluctuations due to the linear and non-linear loads and the nature of renewable sources are addressed. The regulation of ac voltage is implemented through a finite control set model predictive control (FCS-MPC) based active front end (AFE) rectifier, while direct power MPC (DPMPC) is used to control the power during grid-connected operation. The regulation of an ac load voltage is done through voltage based MPC (MPVC) in the islanding operation of the MG. Moreover, the HRES transition from grid-tied to grid-isolated mode is comprehensively analyzed. MATLAB/Simulink® software certified the robustness and evaluated the performance of the proposed HRES model under different varying loads viz. balanced, unbalanced, and nonlinear. The proposed strategy offers superior performance with low total harmonic distortion (THD), compared to previously developed strategies. The output waveform of voltage and current have THD of 0.28 % compared to 3.71 % with the conventional strategy. The contributions of this paper lie in the sequential use of HOMER as well as MATLAB tools and in the validation of the suggested HRES plan for the considered location; along with the implementation of FCS-MPC for a reconfigurable bidirectional VSC.

MPC (MPVC) in the islanding operation of the MG. Moreover, the HRES transition from grid-tied to grid-isolated mode is comprehensively analyzed. MATLAB/Simulink R software certified the robustness and evaluated the performance of the proposed HRES model under different varying loads viz. balanced, unbalanced, and nonlinear. The proposed strategy offers superior performance with low total harmonic distortion (THD), compared to previously developed strategies. The output waveform of voltage and current have THD of 0.28 % compared to 3.71 % with the conventional strategy. The contributions of this paper lie in the sequential use of HOMER as well as MATLAB tools and in the validation of the suggested HRES plan for the considered location; along with the implementation of FCS-MPC for a reconfigurable bidirectional VSC.
INDEX TERMS Distributed energy system, energy management, finite control set model predictive control, residential load, demand side management, net present value, cost of energy, techno-economic design optimization, reconfigurable bidirectional VSC, hybrid microgrid, voltage regulation. With the growing demand for energy worldwide, caused by population growth and advancements in technology, the predictions of negative impacts of current energy systems are alarming. Most of the energy is being generated from fossil and nuclear fuels which create harmful emissions, greenhouse gases, and are proved to be dangerous to the environment. On the other hand, it is expected that a minimum or zero CO2 discharge could minimize global warming up to 2 • C annually [1].
In Pakistan, renewable energy has seen tremendous growth in recent years. According to IRENA ''renewable energy capacity statistics'', the total renewable energy capacity installed in Pakistan was 13 Wind may be available all day, but with large variations over short intervals and therefore wind generation is not reliable to supply the specific load. Solar energy is more predictable but it is not available at night. Aside from their intermittent nature, a high cost is also a major constraint on the use of renewable energy sources. To minimize these drawbacks, new energy systems need to combine several resources in a hybrid system with backup storage units for reliable and optimal power supply [3].
Currently, a large percentage of the global population is living in remote rural areas without access to the electricity grid. Two billion people worldwide have no access to grid-based electricity [4], while a total of 17% of the world population (1.2 billion people) have no electricity for their domestic use [5], [6]. For many remote areas, conventional energy generation units cannot fulfill the load demand due to the high cost of fuel and the infeasibility of grid extension because of geographical complexity. Moreover, conventional energy resources are the main sources of polluting GHG emissions. Access to electricity is related to economic, social, FIGURE 1. A general overview of energy production, distribution, utilization and control [11].

FIGURE 2.
Power levels for the applications of renewable energy [11]. and environmental problems of remote communities especially in developing countries like Pakistan.
Distributed generation is considered as the safest and most viable approach to solving these decade-long problems. Due to the depletion of fossil fuels and the losses of transmission and distribution, standalone MG through distributed generations (DGs) [3] is seen as an attractive solution to fulfill the local energy demand of remote areas electrification, where access to the grid is obsolete or expensive. An increase in energy transfer among different countries is expected in the future due to globalization, distributed microgrid (MG) configurations and high penetration of distributed energy resources (DERs). The transformation from old centralized FIGURE 3. Global statistics of renewable energy [11] (a) Global renewable energy changes (b) Global installed wind capacity, total installed capacity in 2018 is 594GW with 53.9GW newly added capacity(c) Global installed PV capacity, total installed capacity in 2018 is 505GW with 100GW newly added capacity.
systems into decentralized systems with the same fossil fuels makes the system more complex, without addressing its sustainability.
Incorporating renewable energy resources (like wind, and solar) offers a non-polluting solution as part of a distributed generation. The World is now looking for generation models with improved sustainable energy solutions, maximum system performance, and minimum resource usage. Such systems may integrate the wind, solar, hydro and hydrogen sourced through small and medium wind turbines (WT), photovoltaics (PV), micro-hydro turbines, and hydrogen cells (FC). Geothermal energy, ocean energy, biomass, and biogas are also primary energy resources, which can be incorporated with the common renewables. However, biomass and biogas need constant fuel supply and therefore are not suitable for small residential or industrial loads. Also, incorporating FC into a system makes the system more complex and this component needs to be particularly considered during the design phase [3].
Modern diesel generators (DGs) and DC loads furnished with modern power electronics (PE) interface have unfolded the new interesting purview in DC energy systems which has led to hybrid MG modeling. The hybrid MG reduces the energy conversion stage, through a single-stage bidirectional converter, while offering the benefits of a hybrid system by including both AC & DC grids. The hybrid MG function is vital in the standalone mode -during main grid fault, and in supplying remote locations.
Energy storage systems (ESS) are necessary to handle the stochastic nature of the major RERs (wind and solar) and to supply the load profile while ensuring stability and reliability. In the short-term, ESS is used as a grid-supporting entity to fulfill the varying power demand and to maintain the system security and energy supply during load variations. In the longterm, ESS is used to meet the load demand during times of inadequate generation capacity. Currently, batteries are generally used for storage in medium and small-scale energy systems to overcome short-term energy shortage problems. High discharge and high load capacity of batteries make them suitable for coping with generation defects and deficiencies under variable load conditions. Moreover, the battery storage system (BSS) can be incorporated to handle the abrupt load or dynamic behavior of the HRES. Batteries have a high energy density, while supercapacitors (SC) and superconducting magnetic storage (SMES) can compensate for its low power density. The low storage capacity of SC restricts their use. Ultimately, batteries are superior in use to other ESS and are preferred worldwide [1]. Long-term ESS is based on non-renewable resources like diesel generators which have high maintenance costs and negative environmental effects [7].
MG operations are normally equipped with advanced control structure which needs modern control strategies as opposed to conventional control schemes used for conventional DGs. An inner control loop is considered as the main control of DGs to regulate the voltage/frequency in MGs under varying load demand and intermittent DGs. The droop control is often used in this regard. This primary droop control is usually applied as voltage and current loops. The voltage control loop is simple, with the speedy reaction which includes few control loops, but it is not adequate to serve the need for DG protection under fault conditions and high fluctuating current. The current control loop with the additional property of current control resolves this problem. During the off-grid mode of the MG, the disturbances imposed on the DG may have dangerous effects on the DG functioning and can lead to unstable operation in terms of voltage and frequency. Thus, for reliable and economical operating conditions, there needs to be quick and precise control under variable loading conditions, during short time intervals [8].
Model predictive control (MPC) has shown superior performance for online optimization by handling problems with multiple variables through future prediction of control actions, and it is the most suitable and most developed method to address large and complex power system problems [9]. Finite control set model predictive control (FCS-MPC) is the most commonly used MPC due to its simplicity and accuracy. Predictive control for the control of power converters is a unique and powerful approach. Recently, MPC has seen growing applications in power electronics and industry. The quick dynamic reaction of MPC with its online receding horizon and its ability to handle uncertainties [10] make it a preferred method for the control of power converters.
The remainder of the paper is structured as follows. A literature review, as well as a comprehensive and comparative investigation, are elaborated and discussed in section II, and the main objectives, scope, and implications of the proposed strategy are outlined. Section III includes a detailed description of the HRES including feasibility studies and components modeling. Techno-economic studies using HOMER are presented in section IV, including design optimization, objective formulation, and constraints description, with a detailed flow chart. Section V describes the employment of converters control in MATLAB/Simulink, to arrive at an optimal configuration model. Section VI depicts the simulation outcomes, with a detailed analysis of the proposed model. Section VII presents the verification of the key features of the proposed system and its comparison with other works in the literature. The conclusion is presented in section VIII.

II. RESEARCH BACKGROUND SYNOPSIS
Many research studies are focused on standalone HRES. PV-battery based DC microgrid for Nigeria is studied in [12]. In [13], optimum sizing with hierarchical energy management of standalone PV-hydrogen-battery based DC grid is analyzed. PV-MPPT droop control is stabilizing the DC bus voltage while fuel cell (FC) output power fluctuations are handled by using sliding mode control (SMC) with current control mode with feedback linearization.
Application of MPC is found in current literature including FCS-MPC in [14] for primary control of voltage and frequency with large available bandwidth in off-grid systems, while FLC is applied as a secondary control to handle steady-state errors. Due to high cost, active power filter (APF) for unbalance and harmonics compensation, continuous control set MPC (CCS-MPC) with parameter tuning as the      main focus is proposed in [15] and trade-off among voltage quality and injected current, stable operation and quick response is analyzed. In [16], regulating frequency for the microgrid network is investigated with voltage adjustment for voltage-sensitive loads by using distributed MPC (D-MPC). In [17], a decentralized backstepping algorithm for voltage control as well as load sharing is applied for the standalone DC microgrid system. Coordination between the intermittent nature of wind energy and the charging load of plug-in electric vehicles (PEVs) is handled through hierarchical MPC [18]. The control strategy of grid-forming converters with transient analysis under high-signal disturbances is analyzed in [19]. In [20], the adaptive voltage control strategy is proposed to control DG and very good performance and response are observed. The adaptive scheme of static compensator (STATCOM) is investigated in [21]. PI conventional control strategy is applied in [22] for the execution of direct power control (DPC). In [23], DPC with a look-up table is applied for switching states to improve the performance of the converter. Switching harmonics is addressed with a fuzzybased DPC scheme in [24]. Sliding mode control (SMC) is used for the power regulation of a grid-connected system in [25]. Table 3 demonstrates a comprehensive literature work in terms of their contribution, drawbacks, HRES configurations, the applied methodologies as well as the main objectives. Results show that most literature is focused on the investigation of PI traditional schemes but rigorous analysis of the optimal design of HRES components is not considered. In this paper, the control unit of a wind turbine comprises an uncontrolled rectifier as a cheap solution. Further, it needs no IGBTs and control for generating signals for its operation and because of that, it reduces the complication and system cost.
Most published studies were not thoroughly investigated by incorporating maximum power tracking algorithms (MPPT) of WT and solar, buck-boost control to regulate dc voltage, THD investigation, and the comparative study between PI and MPC for bidirectional VSC of hybrid MG systems. Maximum literature studies have focused on either optimal HRES design components or management. This work focuses the design optimization with the best desirable and upgraded EMS scheme. First, the elaborated techno-economic investigation with multiple HRES configurations for DG, PV, WT, BSS, and converter is used for the optimal HRES design. The obtained scheme is then applied to meet the consumer's requirements of an off-grid location selected in Pakistan with detailed feasibility analysis. Afterward, MATLAB R is employed for the execution of the proposed EMS. The most feasible HRES is then tested with the suggested EMS and the operation analysis of the presented scheme is validated for the improvement of energy quality, steady-state and transient behavior with fluctuating renewables and varying loads.

A. NOVELTIES OF THE PROPOSED SCHEME
After the comprehensive review of the literature, the summarized points are discussed as follows: • The HRES models are believed to be the best solution in terms of reliability and economics as well as workable choices with low emission of toxic gases. Although multiple hybrid configurations of HRES are applicable for any specified area based on load trends, the potential of renewable energy resources, and environmental circumstances.
• The main focusing point in the literature studies is either a size optimization with a feasible plan in terms of economics or on energy management schemes. Two viewpoints are seldom presented and investigated at the same instant. Moreover, the comprehensive sensitivity analysis with eighteen (18) study cases including the ROI factor is a crucial prospect for investors in terms of economics which is not taken into consideration in the literature during design optimization of the HRES. Based on the authors' knowledge, the suggested scheme with concurrent analysis of design optimization and energy management for the PV-wind-DG-BSS-converter model with the FCS-MPC scheme of reconfigurable bidirectional VSC for a grid-isolated system is not comprehensively analyzed in the researcher's world. Further, to the best of the authors' knowledge, FCS-MPC based four modes namely. AFE rectifier mode, grid-tied inversion operation, grid-disconnected inversion operation, transitional operation from grid-connected to grid-disconnected) are not comprehensively and simultaneously considered in the literature. Based on the literature studies, the main control for interlinking converters should be developed by keeping in view the dynamic nature of the load for the improved steady-state and transient operation with different and varying loads. Further, efficient-dynamic performance is obtained with MPC while the dedicated optimal design is also possible by improving and to regulate the voltage and frequency of DGs under grid-disconnected mode and active power control in grid-tied operation. Bidirectional VSC is the primary part of hybrid MG to deal with two-way control of active power while maintaining system stability. This work proposed an MPC based scheme of main interlinking VSC in rectification and inverter modes with the provision of robust control operation under transient and steady-state conditions with the considerations of different varying loads.
The main contributions along with the aims and scope of the suggested study are described below: (1) To meet the increasing electricity demand in the world, the specified remote area in Pakistan is selected as a case study to analyze feasible HRES options with PV and wind.
(2) An integrated and generic methodological analysis for the optimal sizing of the HRES components with management for the grid-disconnected system is demonstrated. The suggested scheme is certified with a residential case study in Pakistan. grid-connected VSI mode (V2G mode) and mathematical modeling is also described. (10) Voltage control based MPC (MPVC) is applied for the regulation of ac voltage in grid-disconnected mode. (11) Transitional state from grid-connected to grid-disconnected mode is thoroughly analyzed. (12) The robustness and effectiveness of the considered system are validated through the applied scheme during varying loading conditions viz. balance, unbalance, and non-linear.

B. ADVANTAGES OF THE PROPOSED SCHEME
MPC is the most preferred option for handling the complexity of control problems [56]. The merits of the presented scheme are explained below: • The primary significance is the simple algorithm and the conceptions are based on a heuristic rule. Further, it is intuitive as well perceivable.
• The proposed scheme can be applied more properly for multi-variable models to extend consequently.
• The presented scheme is practicable for real and on-line HRES with non-linear models.
• The presented strategy with a comprehensive outcome including optimized design and management for the selected area is not previously conducted.
• The proposed FCS-MPC controller with HRES can be easily extended from residential electrifications to commercial and agriculture load demands.
• The presented scheme chooses the most viable hres plan by addressing the objectives and constraints of the system. Hence, this scheme fulfills the need of the customers in terms of the economic advantages with fast performance while predicting the dynamic response of linear and non-linear multivariable systems.
• The applied mpc based scheme is also applicable for grid-connected applications in addition to grid-disconnected operations under transient studies.
while the most widely used PI control schemes are only applicable to grid-disconnected operations.
• The settling time is minimum in the case of FCS-MPC control and is more efficient in minimizing the errors and eliminating the noises from the output signals.
• The applied scheme has no requirement of any PWM strategy and it is also capable of variable switching frequencies.
• The performance analysis during steady-state response is more efficient and better in terms of all reference frames while the complex design level is also low and its control is easily implementable in experiments.
• The suggested HRES with detailed analysis includes the incorporation of sensitivity analysis and bidirectional reconfigurable VSC for four modes of operations.

III. SYSTEM MODELING AND COMPONENTS DESCRIPTION
Fig. 5 [57] elaborates the presented scheme for the optimized sizing of the hybrid model for grid-disconnected system and the possible energy management scheme for grid-disconnected as well as grid-connected modes of operation. Initially, the step involves comprehensive energy planning and feasibility analysis including meteorological data and the load demand of the selected location. The first step is followed by specifying the optimization objectives and constraints, viable plans of the HRES, the detailed model analysis of different HRES components is also achieved. The most viable configuration plan of HRES is suggested based on the elaborated methodology with the judgment of three prospects namely technological, economic and environmental. The succeeding step is the performance analysis of the proposed scheme for the feasibly selected model. The last step involves the proof of the suggested overall scheme and its comparative analysis with the traditional PI method. The elaboration of the suggested model and the description of the components are described below: The proposed HRES comprises of diesel, wind, PV, BSS, and converter. The first three components including PV, wind, and the diesel generator are taken as the main generating resources to fulfill the consumer's electricity demand. The energy storage system is considered to feed the consumers throughout the intermittent duration of solar and WT generation under different conditions which include steady-state, and transients' conditions. The power converters are considered as a substantive component of the HRES model to exchange power between buses. It is pertinent to acknowledge here that the dc-dc converters are applied to extract maximum power from solar and WT generating systems. Whereas the battery converter regulates the dc voltage.

A. WIND ENERGY SOURCE
The WECS consists of WT, PMSG, and MPPT mechanism. Wind turbine model which is taken from [34] is of 1 kW capacity, whereas both capital and replacement price are taken as $ 900/kW [59], O&M is $ 10/kW while project lifetime is 20 years.
The desirable hub height (V hub ) is found from the real wind speed data at the corresponding height by applying the following logarithmic law [50] where, V hub , and V act are wind speed relevant to desirable hub height (H hub ), and measurable wind speed relevant to reference height (H act ) respectively. L o is the length of surface roughness.
Wind mechanical power P mec is represented as: The formula for the blade tip speed ratio is shown as Fig. 6 represents the maximum power coefficient (Cp) with reference to the tip speed (λ) value. Normally, the pitch angle is taken as zero during P mec less than the nominal value. Therefore, Cp is the part of λ and its value is high i.e. C pmax at a relevant λ. Line with zero angles represents the tip speed with a peak value of coefficient which is taken in this paper. At this instant, the wind turbine works at maximum power with the optimal rotor speed i.e. ω Dopt and at the specific wind speed i.e. v w in (5). By substitution of (5) in (2), we get: The maximum power extracting scheme of WT and the pitch angle scheme are applied for keeping power at the VOLUME 8, 2020 FIGURE 5. A suggested scheme for the optimal sizing, and management system [58]. desired level. The reference value of power i.e. Pis found with the MPPT algorithm. The relationship to reach the optimal value of the rotor speed is represented below: By considering the fast-electronic function, the power (P) of the WT is considered the same as the reference power.

B. PV SOLAR ENERGY SOURCE
The solar energy is relatively not expensive when equated with WT energy, the reason is that no wear and tear exists in the case of PV generation [9]. For MPPT, the IC scheme is employed for extracting maximum power from solar PV generator with improvement in steady-state and dynamic response and well operation during fast-alteration of atmospheric situations.  The clearness index (K t ) is given by [61] The total PV output current is computed as follows [6] The PV output power is computed as [26] where P SC , and G irr (t) are the output power at standard conditions, and PV irradiance on tilted panel surface at hour t respectively, F dirt is constant which is about 0.9, α is temperature coefficient in %/ . T cell (t) is cell temperature in which is expressed as: where T a (t)and T nom are respectively the ambient and nominal temperatures of PV cell in . The PV Simulink diagram is shown in Fig.   FIGURE 8. PV Simulink diagram.  The photovoltaic current (I PV ) is linked with insolation (λ S ) as follows [62]  power. R1 and Ro are the two different load lines where Ro refers to the load where the maximum power point overlaps with the operating point. Fig. 11 shows the graph between voltage (Voc) and current (Isc). Voc is linearly dependent on cell temperature and is logarithmical with irradiation, and Isc is linearly dependent on irradiation. Fig. 12 shows the PV system with the schematic diagram for DC-DC converter, where Rin and Ro are the converter input resistance and output or load resistance respectively. Approximate ranges of Rin for boost converter is shown in Fig. 13.

C. BATTERY BANK
The excess energy which is available from generating energy resources is stored and then fed to the load under low generation hours. The battery capacity can be expressed as [12] The formula for battery energy is represented as The total number of batteries in series (N S ) and parallel (N P ) are calculated by using the following relationship.
The battery energy [26] in terms of production and absorption is expressed as.
with battery charging during: (19) while the battery is discharging during: (20) where C Bat (t) is the available power of the battery bank at duration t in hours while C Bat (t−1) shows the same power at the previous time (t−1). The symbol σ represents the selfdischarge rate. The term P l (t) shows the load requirement at the duration t in hours. The battery/inverter efficiencies are expressed as η Bat and η inv respectively. The limits for C Bat (t) are taken as the permissible energy point between the minimum and maximum range as: For prolonged services of the battery, the maximum charging rate is defined as: where C Bat and N Bat are nominal battery capacity and a total number of batteries respectively. Similarly, the minimum Number of batteries is calculated as [50] The battery terminal voltage is described as [64].
The battery SOC is calculated as [50] SOC(k)

Subject to
The relationship to design the inductor is expressed as [65].

D. POWER CONVERTER
A bidirectional voltage source converter is applied to exchange bidirectional power between both buses. The power transmitted through the converter is [26] P in = P chg η inv (29) where η inv and P chg are converter efficiency and load demand (W) in hours respectively. The relationship for VSC capacity (C) is shown as [6] C = (3 × L ind ) + L res (30) where L ind and L res are respectively inductive load and resistive load.

E. DIESEL GENERATION
A DG is used to supply a high amount of energy shortfall under the intermittent and fluctuating behavior of the generating sources. The fuel price is $ 0.78 per liter. The fuel utilization of DG (F dg ) in L/hr can be represented in Eq. (36) [66] as: where α and β are coefficients of consumption in L/kWh, P nom , and P out are DG nominal and output capacities in kW. The DG efficiency (η dg ) is defined as where H fuel is the value of fuel consumption heating in kWh/L (ranges from 10-11.6). The values of α and β are 0.246 and 0.08145 respectively [26].

IV. TECHNO-ECONOMIC INVESTIGATION USING HOMER
The comparison of different optimal sizing tools is described in [67], while these tools are used in the research studies for the sizing of HRES models. Out multiple available tools [68], [69]- [73], HOMER (Hybrid Optimization of Multiple Electric Renewables) tools are the best often applied and extremely commended in different literature works.
HOMER is a robust design optimization software for design optimization [32] which takes into account the adaptable circumstances to investigate different configurations and optimal HRES models [32].
Its algorithms of design optimization permit the designer and planners to judge the feasible scenarios under different technical criteria related to the changing technological trends and resource accessibility. Hence, HOMER is selected in this study for the techno-economic investigation to determine the most feasible option to meet the load requirements for the selected area.
HOMER assesses viable configuration schemes among various possible solutions with frequent and easy process. The suggested flow chart of the optimization design is demonstrated in Fig. 14 [34]. It optimizes the component sizing [6]. Its design optimization is purely based on the input parametric values which include load requirements, gener- ation profiles, economic and technical circumstances, modeling constraints, suggested energy management, scheme, GHG evaluation parameters [6], [32] with TNPC as the main objective of design optimization scheme [32]. HOMER calculates a one-year optimal design to evaluate different aspects of HRES which includes technical, environmental, and economic considerations [74]. Subsequently, extrapolation of costs operates for the remaining years throughout the project life, to ensure reliable generation and consumption balance on an hourly basis [32]. After finishing the testing stage for all possible configurations, the viable plans are selected and ranked in terms of design objective.

A. OBJECTIVE FUNCTION AND CONSTRAINTS
The past literature studies have taken into account different evaluation standards reported in [34] which are applied to determine the viable configuration plan of HRES. In this study, the annualized cost of the system (ASC) is used as the main objective function with LLP and emission as the main constraints are described as follows.
Annualized system cost (ASC) or annualized total cost (ATC) in $/yr is the sum of system capital cost per annum, annualized maintenance cost, and annualized replacement cost for each model component which is expressed as follows ASC = A capt,cos t +A maint,cos t +A repl,cos t + A fuel,cos t (33) with A capt,cos t = C capt · CRF(i, n) (34) where C capt is the capital value for each HRES component (in US$), i is annualized interest and n is project life. CRF is described as where i is expressed as Operation and maintenance cost are represented as A maint,cos t (n1) is the first-year cost of the corresponding component. Replacement cost is related as where C repl is the replacement cost of each component (in US$), n repl is the component of life in years. SFF is given below.
LCOE in $/kWh is the ratio of ASC and generated electricity which is derived as LCC in $ is the sum of recurring and non-recurring costs during the system lifespan, which include operating, installation, capital, upgrading, maintenance costs, and salvage value which is written as LPSP is mentioned as.
LLP is defined as where LLP ≤ ε LLP , Edeficit, and E demand are energy shortage and unmet load respectively. LOA is the reliability index which is defined as one minus a total hour when a load has not served to the system operation hours as expressed below.
ELF is the ratio between forced outage duration and the total system operation time (hours). ELF is less than 0.1 for an offgrid system which is expressed as follows.
The mathematical relation for renewables fraction [?] is written as The power balance constraint is [75].
The limits of generation and battery operation [75] are.
The DG creates pollution in the form of toxic gases and carbon dioxide is the primary toxic gas and is taken into account in this work which is expressed as follows [26].
where CO 2W should be less than or equal to ε CO2 (i.e. CO 2W ≤ ε CO2 ) while F em is the emission factor (in kg/L) based on DG type with its engine attributes and ranges from 2.4-2.8.

V. TECHNICAL ANALYSIS USING MATLAB A. SYSTEM DESCRIPTION
A microgrid comprises of energy resources including distributed generators (DG), battery storage system (BSS), and renewable energy sources (RESs). The DG output power in an ac grid is calculated as: where P ref g and Q ref g represent active and reactive powers respectively. k p and k q are the droop coefficients of two curves viz. P − ω and Q − V (see Fig. 15) respectively. ω and V demonstrate the values of frequency and voltage magnitude respectively. For the dc grid, the output power of DGs is expressed as. Fig. 16 shows the general concept of the MPC scheme. The prediction step of the system is incorporated in the optimization model to address the issue which is based on the predicted time horizon as shown in Fig. 17.    FCS-MPC is implemented for a bidirectional VSC converter to regulate the load voltage. FCS-MPC scheme with 2-level 3-phase VSC is suggested in this work. The applied control strategy chooses the most feasible state from seven switching states for minimizing the cost function. Simulation results demonstrate compensation of disturbance from loads, and sources while the load voltage is in continuous mode for tracking the reference signal of the voltage. A buckboost control is used for the regulation of the dc voltage. The renewables converters of PV and WT are implemented to extract maximum power. The following section discusses the control as well as management strategies. Fig. 19 depicts the applied FCS-MPC concept. The parameters required to address and resolve the particular issue is highlighted with symbol v. The external perturbation is symbolized with d. The control signal is highlighted with symbol c. The forecasting parameter is expressed with a symbol w. With the measurement values of the current system (v) and forecasting parameter (d), the simulation is executed with an optimizer. MPC strategy is used for main VSC control. Contrary to the conventional controls, FCS-MPC does not need to apply the control with inner and outer loops or PWM techniques The working rule of the suggested methodology is described in [57].   explained as follows: where L s , R s are respectively the inductance and resistance of the RLC filter. The parameters like v s v r and i s are representing respectively the voltages of source, rectifier and input current. Fig. 21 shows the generated voltage vectors of VSC. Discrete-time state-space modeling for the prediction of current is represented as: where T s shows the sampling time and the following relationships represent the predicted value for instantaneous powers (active and reactive): The following cost function is applied for the predictive power control.
where Q * i , P * i are the reference values of reactive and active powers respectively. Fig. 22 shows the flow chart of the FCS-MPC algorithm which is applied for the control of AFE VSC.

2) VOLTAGE MAGNITUDE/FREQUENCY CONTROL OF VSI DURING GRID-DISCONNECTED MODE
The filter model is shown in Fig. 23 and predictive voltage control is described in Fig.24 with the following relationship.
where V c , V i , I f , R f , L are respectively the voltage vectors on the load side of VSC, VSC output voltage, current, resistance, and inductance of RLC filter. Continuous state-space  modeling of the above-mentioned relationship is expressed as follows.
To determine the voltage value at the next sampling time, the discrete state-space model is expressed as follows By solving the above-mentioned relationship for the discrete state space, we get the following equation.
and V i represe nts the voltage vector with seven switching states for VSC switches viz. S a , S b , and S c with the following relation.
The load disturbance observer is applied by defining the objective function as, Fig. 25 demonstrates the flow chart of an algorithm for VSC as an inverter operation in the grid-disconnected state.
The real power (P) is controlled with the VSC output voltage and the imaginary power (Q) is controlled with the frequency of the VSC. The output voltage (V nom ) and frequency (ω nom ) of VSC are used to control the powers (P and Q) as mentioned in the following expressions.
the coefficients namely m and n are chosen so that the system stability is ensured with the following relationship.
where P max and Q max are the maximum values of powers furnished by VSC. The other parameters like ω and V demonstrate the deviations with an upper limit for frequency and voltage amplitude of the VSC output signal. The parameters, m, and n are set to 0.0014 and 0.0008 respectively.

3) DIRECT POWER CONTROL (DPC) WITH GRID-TIED MODE
For the development of the DP-MPC model in the gridconnected state as demonstrated in Fig. 26, the following relationships show the output voltage and field current vectors in the αβ reference frame [76].
For balanced loading conditions, the relation for line current is expressed as follows.
where v cαβ , v iαβ , i f αβ , and i o are respectively the voltage vectors of output, and input of VSC, the current values of filter, and load. The exchangeable power (P and Q) with the grid is determined with the following relationship.
Substituting (78) and (79) in (80) and then taking the derivative, the following mathematical relationship is found.
For load voltage underbalanced and sinusoidal waveform, the following expression is obtained. To calculate the instantaneous powers (P and Q), (21) can be modified by substituting (76) and (83) into (81) as where a distinctive space vector is expressed by − → v ii in αβ frame of reference with the following relation.
where S ia , S ib , S ic are the switching states of the VSC. For constant dc bus voltage, the power prediction for the next instant i.e. k+1 during the end of the sampling period (T s ) is expressed as follows.
The feasible switching state with corresponding minimum voltage ripples is chosen for VSC switching with the expression of the following cost function.

C. BATTERY CONTROL UNIT
To regulate the dc bus voltage, Fig. 27 demonstrates the BSS model and Fig. 28 represents the buck-boost controller. The net power is maintained and is described as follows.

D. WIND MPPT CONTROLLER
A WT with a constant (zero) pitch angle and fluctuating wind input is applied, and the parameters detail of PMSG and wind is given in [34]. Fig. 29 shows the MPPT boost controller for a wind energy conversion system.

VI. CASE STUDY
Pakistan is located in the territory of South Asia and further detailed investigation of the territory of Pakistan is comprehensively discussed in [57] along with site description, energy resource assessment, and electric load estimation. The domestic area of Kharan district is shown in Fig. 30 and is located at the latitude and longitude of 28 • 44.0 N and 65 • 3.3 E respectively as shown in Fig 31 with wind generation details. And the annual radiation is shown in Fig. 32. The wind speed, solar irradiance, and temperature are shown in Fig. 33, and Fig. 34 respectively. The estimated peak load is 6.03 kW and the annual energy demand is 35.94 kWh/day. Fig. 35 shows the load profile for two households with detailed load estimation in Table 6.

VII. RESULTS AND DISCUSSIONS A. DESIGN OPTIMIZATION AND OPERATION INVESTIGATION USING HOMER
The economic prospects and technological analysis are investigated for each component of HRES which includes PV, wind, diesel, battery, and VSC with load profile, real-time solar irradiance, temperature and wind speed. The suppositions and constraints are taken into considerations for the development of the HRES model and simulation studies as demonstrated in Fig. 36 and described as follows.
• Project lifetime is set to be 20 years to find the best computations of the HRES model.
• The nominal discount value is taken as 10 % which is based on the routine trends in Pakistan while inflation is set as 4 %.
• To guarantee the maximum system reliability, zero capacity shortage is considered during the simulation.   • A 10 % reserve is assumed to cover the fluctuating load. The wind and solar energy reserves are chosen to be 15% for compensating random variations in wind speed and solar irradiance. • A GHG emission penalty of $ 20/tons is considered as one of the constraints.
• Power flow among different components of HRES is controlled by using the proposed dispatch scheme. Fig. 37 [36] demonstrates the flow chart to produce adequate electricity from a diesel generator which can meet the load requirement. To charge the battery and to serve the deferrable load, low priority objectives are set which depend on renewable power generation capacity. LF strategy is applied to achieve the maximum profit from RERs as well as to reduce fuel consumption. LF preserves the battery by VOLUME 8, 2020 avoiding over-charging and over-discharging [36]. Feasible outcomes of all possible configuration plans of the presented HRES models are shown in Table 9 in terms of techno-economic investigation and are ranked based on TNPC. The concluding remarks about these above-mentioned results are described as follows.
• Among all available HRES configurations, PV-WT-BSS-converter is the most viable option with higher   Table 7 Fig. 38 shows the electricity produced and shared between PV and WT. MPPT is applied for extracting maximum power for meeting the load requirement. It is noticed that the generation from PV source is high in July due to maximum radiations while June has next-top value and these observations are also justified in Fig. 34.
The on-line energy scheduling scheme including generation and load profiles is demonstrated in Fig. 39 for December 15. The critical analysis of Fig. 39 reveals many important prospects for the suggested configuration plan. It is observed that both solar and wind energy generations VOLUME 8, 2020  are available from 0800 onward and are not available after 1700. During these nine hours, solar power is more than the wind power and all the load is supplied by renewable generation. Moreover, excess energy is also used to charge the battery units. During this time, the battery remains in idle state and hence the life of the battery is improved. Due to the non-availability of PV and wind energy before 0800 and after 1700, all the load is supplied by the battery storage units. In this way, the optimized operation of the HRES components is ensured with proper dispatch and management strategy. Table 7 concludes the costs of the component of the proposed configuration plan. The PV initial cost is about four times the initial cost of WT. And BSS has 1.5 times cost as compared to wind. The O&M cost of the BSS is slightly higher than that of PV, and it is six times greater as compared to wind. Based on the total cost, PV and BSS have almost the same values while wind has three times lower cost than that of PV and four times lower cost than BSS. Fig. 40 shows the optimal solution set and the cash flow summary of two cases viz. base case (DG only) and the optimized HRES (PV-WT-BSS) for a 20 years plan. The optimal system maintains minimum cash flow during the entire lifetime of the project, whereas the cash flow for the base     Table 11 shows the sensitivity parameters which are used for the comprehensive analysis and impact of important factors on different cost s and emissions. Fig. 42 shows the TNPC by varying solar irradiance and wind speed while other parameters are kept constant. It is observed that TNPC has a direct relationship with wind speed and solar irradiance for the selected area. An increase in either of the two parameters will reduce the TNPC. Fig. 43 shows the impact on LCOE with the variations of nominal discount rate and fuel price while keeping another parameter constant. LCOE is reduced with a low discount rate while diesel fuel has no impact due to the case when DG is not present. Fig. 44 shows the impact on TNPC with the variations of load and fuel price. There is the pivotal value of loading i.e. 35.94 kWh/d where the optimal configuration plan is changed from 100 % renewables (Case-7) to diesel-based HRES system (Case-8. Therefore, it is important to mention here that the load will decide whether a 100 % renewables system is possible for the selected location or not. In this study, 100 % renewables are only possible up to 35.94 kWh daily load. While DG needs to be incorporated above this specified load. It means that the present study is conducted for two households while the system can be extended for more households while keeping the DG part of the MG system. Table 12 shows the impact of different parameters on the system performance in terms of techno-economic and environmental context. The discount rate has a minor impact on  PV sizes while WT size is reduced from six to four units as the discount rate increases while battery size increases two units. TNPC is reduced while LCOE is increased with an increase in discount. Initial capital cost remains almost the same while O&M cost is reduced to about 75 %. Variations of solar irradiance have more impact on PV sizes while battery units remain the same. TNPC and LCOE are reduced significantly with the increase in solar irradiance. During the variation VOLUME 8, 2020  of load, PV and WT sizes and TNPC reduced significantly with a decrease in load while LCOE increases. High wind speed reduced the PV and battery sizes as well as TNPC and LCOE while WT size remains almost the same. Variations of diesel price have no significant impact on PV and WT sizes while battery size is reduced significantly with the increase of TNPC and LCOE. The impact of SOC with the incorporation of temperature effects is interesting. With the increase in temperature, while keeping the value of SOC same, there is no major change in PV, while a minor decrease in WT sizes and a minor increase in battery sizes are observed. TNPC and LCOE are increased with an increase in temperature. During SOC variation from 10 % to 70 %, while keeping the temperature constant, PV size remains almost the same till 50 % SOC while it is reduced drastically with the increase of 20 % SOC. TNPC and LCOE are increased and directly proportional to the SOC while DOD has the opposite relation as compared to SOC. Battery string sizes have a major impact   on PV, wind and battery sizes while TNPC is increased slightly with the increase of LCOE. PV capital cost has no impact on PV size while TNPC is reduced and LCOE is increased with a 75 % reduction in PV capital cost. The same but more significant impact on TNPC and LCOE is observed during wind capital cost. By reducing wind capital costs up to 75 %, the PV and battery sizes are reduced significantly while WT units are increased in a significant number. The capital cost of battery banks has no major impact on PV and WT sizes while a drastic reduction in TNPC and LCOE is observed. Therefore, storage cost has a major impact on the overall system cost. More efficient and low-cost storage systems will help to develop the models of HRES in a costeffective manner. The inflation rate has a direct impact on TNPC while indirectly proportional to the LCOE. Table 13 shows the effect of load perturbation on different parameters for the two cases when DG is available for one case. without DG i.e., case-7, the highest value of ROI is found when a load is decreased to about half of the main value i.e. 15 kWh/d with no emission while LLP is the same as the    Table 14 shows the effect of load perturbation on components sizes. Table 15 shows the effect of load perturbation on different costs and have a directly proportional relationship. When a load is decreased to about three times i.e. 10 kWh/d, the new case with PV-battery is found to be more economical without wind and diesel units. When the load is increased beyond the base value, diesel generator becomes part of the HRES. Table 16 shows the effect of emissions on different costs. Table 17 shows the effect of LLP on different costs. TNPC and LCOE are drastically reduced VOLUME 8, 2020  with the increase of LLP value from initially zero to 1.2 % while the cost reduction is more than half of the base value.

B. OPERATIONAL INVESTIGATION OF OPTIMIZED MODEL
An EMS with FCS-MPC based control strategies is applied by considering the load fluctuations and generation perturbations as shown in Fig 45. The PI controllers' parameters including wind boost and battery buck-boost and the electrical parameters of the suggested model are shown in Table 18. The case studies for the validation of the proposed strategies are discussed one by one as follows.

4) AFE RECTIFIER
The first operational model is shown in Fig 55 for the implementation of the AFE rectifier to feed the dc load from the ac supply which can be ac generation from distributed energy resources or from the ac grid.
A constant dc load with fixed resistance is applied for the performance of the AFE in rectification mode. The FCS-MPC based control of AFE rectifier tracks the reference value of dc bus voltage at 510 V. The constant three-phase ac voltage is depicted while the three-phase ac source current shows some    high magnitude transients during the initial 30 ms due to the rise time for the real dc voltage to track the reference dc voltage through the FCS-MPC control. The reactive power is   zero while the real power of 5.3 kW is fed through the rectifier to the dc load.
A dc bus voltage changed from 510 V to 400 V at 0.5 s to check the performance of the suggested AFE rectifier.      is reduced to 38.7 % (3.25 kW). the reactive power (Q) remained zero. dc load resistance is changed from 50 to 75 ohms at 0.5 s to the controller performance for regulating the dc bus voltage. During the abrupt change in dc load, a 3.5 % change in the dc voltage (i.e. 328 V) is observed as shown in Fig 57 while the controller is tracking the dc reference voltage after the abrupt load variation. The 33.3 % reduction in three-phase ac current while real power is reduced from 5.3 kW to 3.5 kW.
The dc voltage is kept constant while the active power of dc load is changed from 2 kW to 3.5 kW as shown in Fig 58.

5) GRID-CONNECTED VSI FOR DPC
The second case study includes the grid-connected operation of the VSC in grid-feeding mode. The DC constant voltage   The DC voltage is fixed at 510 V while the AC active power is fixed at 8 kW as shown in Fig 15. The reactive power (Q) in this case is taken as zero. It can be observed that the constant voltage and current set by the infinity bus (i.e. utility grid) is not disturbed while the direct power transfer from DC source to the AC grid is carried out.
A DC bus voltage is fixed at 510 V while the active power flow is varied from 8 kW to 12 kW at 0.5 s by keeping the reactive power fixed at zero as shown in Fig 60. Constant three-phase AC voltage is observed while variation in three-phase AC load due to the variation in active power validate the proposed strategy of reconfigurable microgrid after direct power flow control from DC source to AC grid. The DC bus voltage remained constant at 510 V while both active and reactive powers are changed from 8 kW to 12 kW and 0 kW to 5 kW respectively as shown in Fig 61. Constant three-phase AC voltage while the variation in magnitude of three-phase AC current is more due to the reason that the reactive power transfer of 5 kW is also included in this case. The unbalanced three-phase load with each phase resistance of 40, 200 and 7500 ohms respectively is taken. The DC voltage is again fixed at 510 V while the active power is set to 8 kW (see Fig 62) with zero reactive power. The three-phase AC voltage and the unbalanced load can be seen from threephase unbalanced AC current. The active power of 8 kW is successfully supplied to the grid in a reconfigurable fashion.

6) ISLANDED VSI FOR VOLTAGE REGULATION
The third case study includes the standalone microgrid configuration to control the voltage magnitude and its frequency in grid-forming mode and voltage control-based FCS-MPC (FCS-MPVC) strategy is used in this case.
The three-phase balanced and constant load is taken with zero reactive power at a fixed DC voltage of 510 V. The regulated voltage is demonstrated in Fig 63 and the load current. The three-phase AC active power of 3 kW is fed to the load. The three-phase load is varied from 20 ohms to 10 ohms at 0.5 s while the DC source voltage is fixed at 510 V. The real power is changed from 1.69 kW to 3.37 kW at 0.5 s while the imaginary power is zero. Fig 64 shows the three-phase load voltage which is regulated by the FCS-MPC control strategy.

7) INVERTER RECONFIGURATION FROM GRID-DISCONNECTED MODE TO GRID-TIED MODE
A seamless transformation from grid-isolated to gridconnected mode is implemented. Fig 66 shows the regulated DC voltage during the operation of the AFE rectifier in rectification mode. The voltage is tracking the reference value without any transient in voltage magnitude. The transformation from rectification to grid-tied mode is accomplished at 0.5 s while the three-phase voltage remained purely sinusoidal. The seamless transformation from grid-tied to grid-isolated mode is implemented at 1s where the zoomed view of voltage shows the magnitude variations of voltage phases for a short time while maintaining the voltage magnitude and its VOLUME 8, 2020  frequency at fixed reference peak value. The corresponding three-phase current is shown with the real and imaginary power flow during these transitions. The reactive power is zero during rectification mode while the active power of 8 kW is flowing. During grid-connected mode at 0.5 s, the power flow is opposite while the same active power of 8 kW is flowing to supply the grid. The reactive power is almost zero. During the third seamless transformation from grid-tied to grid-isolated mode at 1 s, the almost same active power of 8 kW is flowing due to the fixed load resistance, while reactive power of 4 kW is flowing to feed the AC load.   the THD values (0.28 %) of HMGS with MPC. THD of 3.71 % is recorded in the case of PI control as shown in Fig 68 and Fig 69. Table 23 summarizes the THD and voltage mismatch.  Table 22 compares the cost results of different optimal design models with the presented hybrid model. With the consideration of components sizes and capital cost, the TNPC is definitely not the same as all literature models as well  as the proposed model. Nevertheless, LCOE is an alternative parameter and judging tool as a crucial parameter to compare the cost of HRES based generated power.

VIII. PROMINENT ASPECTS OF PROPOSED STRATEGY AND RELATIVE ANALYSIS WITH LITERATURE
Iraq, Saudi Arabia, India, Egypt, Malaysia, China, Canada, Ethiopia, and Pakistan have the lowest LCOE in terms of comparison with the relevant parameter of other countries.   While LCOE in the Philippines, Cameroons, Spain, Turkey, Bangladesh, Morocco, and Nigeria are the highest. With the obtained solutions, the comparative analysis also shows reasonable trade-off with the literature studies and demonstrates a clear understanding of the feasibleness of economic aspects of HRES in Pakistan. Table 24. The THD value for the FCS-MPC scheme is found to be 0.28 % which is less than the relevant THD values mentioned in [57] under the same conditions. The probable reason is the different component sizing, dc/ac voltage level, and load profiles. Another model consisting of the wind-battery scheme which is analyzed in [34] has the same THD value under the MPC scheme without incorporating PV source. The comparison of load power ripples is examined between PI and MPC control schemes. Lower values of THD and power ripples during the MPC scheme show superior performance. It is relevant to describe the reason for very low values of THD during the MPC scheme for different case studies that can be examined from the output signals of voltage and current of Fig 46 and  Fig 52. The major outcomes of the proposed study are concluded below:

The comparative THD values are shown in
• The optimized model is obtained with PV-wind-BSSconverter and the TNPC is found to be $ 47,398 while the value of LCOE is $ 0.309/kWh. In this way, generation incorporation with 100 % renewable energy resources is ascertained, without any supply shortage.
• The value of LCOE is found to be high which is $ 1.690/kWh for the base case including DG only (case-9). Hence, it is examined that the diesel generator shows a critical role in terms of GHG emission.
• The TNPC of the proposed HRES is $ 47,398 that is about six times less than the value of base case which is $ 258,845. Hence, the stand-alone system with DG only is not commended, which is the usual case in Pakistan.
• No GHG emission is found and fuel consumption is not required for the presented HRES model whereas 19,364 kg/yr emission and fuel consumption of 7,398 L/yr is found for the base case.
• Remarkable outcomes are obtained during the implementation of the presented EMS scheme under the intermittent generation of PV, and wind with fluctuating load.
• With the MPC scheme, improved power quality and voltage regulation are obtained for a standalone system.
• The proposed MPC scheme is robust during transient analysis and fluctuating dc voltage level.
• Only 0.28 % of THD of the output ac voltage which is well below the value of 5 % as specified by IEEE standard 929, 2000 and IEEE-519 standards [6].
• The PI controllers' parameters for wind/solar boost and battery BSS converter are selected with a trial and error approach.
• The operation analysis of reconfigurable bidirectional VSC during generation intermittency and fluctuating load demand is implemented through simulations and different case studies.
• The proposed study will assist the government of Pakistan as well as developing countries to incorporate multiple standalone and grid-connected scheme with state-of-the-art schemes. Energy sector planners can implement appropriate action plans to integrate further renewable energy resources to ensure the system more reliable, economical, and environmentally safe. These analysis outcomes will assist new viewpoints for the designers, energy planners and researchers to design and apply the models in an effective and efficient way by penetrating more intermittent energy resources like PV and wind while handling the increasing demands of irregular and fluctuating load requirements.

IX. CONCLUSION
As determined by techno-economic analysis carried out through HOMER and MATLAB, a grid-disconnected model with PV-wind-BSS configurations is selected as the best suitable optimized plan, to fulfill the residential electricity requirements of Kharan region in the Baluchistan Province of Pakistan. The first step involves optimal components sizing through HOMER. Simulation studies for the region are carried out with practical and real load profile and weather data, with constraints including sixteen (16) sensitivity parameters. Out of nine possible models as examined in this work, PV-WT-BSS was found to be the best optimal choice with better economic considerations. The minimum value of TNPC is observed to be $ 47,398 while LCOE is recorded as $ 0.309/kWh. This represents an 81.7 % decrease in overall cost while a 100 % reduction in GHG emissions while fulfilling 100 % of the load requirements with 63.9 % of excess available energy. The suggested model consists of 21.1 kW PV, 5 kW WT, 5.96 kW power converter, and 38 BSS units with 2.37 kWh each capacity. For technical analysis, optimized sizes are obtained from the initial step, are therefore chosen and proposed management scheme in simulink with FCS-MPC for bidirectional VSC. The technical analysis which is based on different parameters including power balance among HRES components, constant level of DC and AC voltages, Safe operation of battery SOC, improved quality of an ac voltage. The analysis is carried out by observing the intermittent nature of PV irradiance, WT speed, and fluctuating load requirement. The result outcomes are formalized through simulation case studies. The total harmonic distortion (THD) was found to be 0.28 % which is well below the permissible values as specified by IEEE-929 as well as IEEE-519 standards, and considerably lower than 3.71 % THD with the conventional PI control scheme.
The suggested strategy could assist the governments, energy planners, and data analysts to inquire about proper action plans, to effectively and efficiently design of the hybrid model. Increased demand with external perturbations can be effectively treated with more generation penetration of renewable sources to achieve the system with better reliability, economically viable, and environmentally safe. The future work will incorporate biomass energy with a comparative analysis between sliding mode and model predictive control.