1. Introduction
Ensuring access to sustainable and reliable energy is crucial to achieving social and economic development [
1]. A large percentage of the global energy demand is covered by fossil fuels, which have a negative environmental impact. When fossil fuels are burned, toxic greenhouse gasses are released into the atmosphere, endangering people’s health and causing climate change [
2]. Furthermore, over the past years, there has been increasing concern about running out of fossil fuels because they are non-renewable sources [
3]. For these reasons, it is vital to cover the energy demand using renewable energy resources (RESs) such as wind, hydro, solar, geothermal, biomass, etc. RESs offer viable indigenous alternatives to conventional energy sources since they are secure, economically competitive, clean, and sustainable [
4]. The use of RESs in producing electricity would help minimize energy imports and fossil fuel use, which are presently utilized in several applications, such as providing power for community and household buildings, pumping, irrigation, and other purposes [
5]. The high rate of renewable energy capacity expansions was maintained in 2021, with an additional value of 270 GW. Moreover, this expansion is likely to continue in 2022, with an expected value addition of around 280 GW. Around 90% of the power capacity increase in 2021–2022 comes from RESs. This increment is about 50% higher than the capacity additions of 2017–2019. Net renewable capacity additions from 2019 to 2022 are depicted in
Figure 1 [
6].
Uncertainty and variability are known characteristics of RESs. Backup components, such as battery energy storage and diesel generators, are used with these sources to overcome these problems. HESs that consist of RESs and backup components are capable of improving system reliability and reducing reliance on national grids [
7].
The optimal design of HESs is a vital challenge to achieve a cost-effective, reliable, and clean system. The components capacity of the HESs, such as wind turbines, PV, generators, and energy storage, must be optimally designed to meet all constraints and maximize/minimize the objective functions [
8].
Optimizing HESs using different techniques has been discussed in several studies. In [
9], different optimization algorithms were developed to find the optimal design of an off-grid PV/wind/biomass/battery HES to cover the electrical load of a rural area in Egypt. It was found that the HES consisting of wind turbines, biomass, and batteries using the Slime Mould algorithm was the optimal configuration in terms of cost of energy (COE), net present cost (NPC), and loss of power supply probability. The authors in [
10] used HOMER software to investigate the optimal design of a grid-connected PV system for a household in South Africa. The results showed that the suggested combination is a feasible option from the technical, environmental, and economic point of view. The optimal design of a grid-connected HES comprising PV, wind turbine, and fuel cell used for a case study in Qingdao, China, was presented in [
11]. A modified seagull optimization technique was used to minimize the power generation cost. The results concluded that utilization of the proposed HES is an attractive option despite the intermittency in RESs. García-Vera et al. [
12] conducted a techno-economic feasibility study using different stand-alone HESs consisting of PV, wind turbines, hydraulic turbines, diesel generators, and batteries to supply electricity for different locations in Colombia using iHOGA software. The study demonstrated that access to energy problems could be solved using the proposed isolated HESs.
The authors in [
13] performed an optimization study to find the optimal sizes of a grid-connected PV/fuel cell hybrid generation system for the electrification of a small shopping complex at an Indian university campus. Artificial bee colony, particle swarm optimization, and a combination of both were used to find the optimal design based on the least NPC. It was found that using PV and hydrogen could reduce reliance on the overburdened national grid. In [
14], a techno-economic and environmental assessment was performed using RETScreen software to find the best capacity size of a grid-connected PV system for a case study in Nigeria. The optimization analysis showed that the suggested system was an attractive choice for all selected sites. In [
15], two different control strategies, load following (LF) and cycle charging (CC), were used to control the operation of an HES proposed to cover the load of a certain area in Turkey. The optimization was performed using HOMER software. CC strategy achieved better economic performance than the LF strategy. An optimization study to find the best control strategy based on power prediction of PV/wind HES was carried out [
16]. Predicting the power of the proposed system was achieved using the WPNN model. The results indicated that the suggested technique was efficient in forecasting the power. Fares et al. [
17] applied different metaheuristics optimization techniques to explore the optimal configuration of an off-grid wind/PV/battery to cover an electrical load in Algeria. The authors found that simulated annealing and flower optimization algorithms presented higher accuracy and robustness than the other techniques.
According to the literature mentioned above, the optimal planning and design of the proposed systems were performed by considering different objectives for a certain project lifetime. However, the simulation was conducted for only one year, and the results were extended over the rest of the project’s lifetime. Therefore, changes that occur over the project’s lifetime, such as PV and battery degradation, were not considered in these studies.
The present work fills the existing research gap by investigating the optimal configuration of a PV battery system in a grid-connected residential home in Karbala, Iraq. HOMER software was utilized as a simulation and optimization tool by considering the techno-economic performance of the HES using two different control strategies, LF and CC. The simulation was carried out using a multi-year module by running a simulation for every year of the project’s lifetime. Hence, some important phenomena that cannot be captured in a single-year simulation, such as PV and battery degradation, were considered in this study. As per the author’s knowledge, exploring the multi-year effects on the optimal sizing and control strategy of HES represents an original contribution.
3. Results and Discussion
Results of using different control strategies to find the optimal HES design for the residential house are presented in this section. The comparison analysis considers technical, economic, and environmental performance. The simulation was carried out with and without the multi-year effects of PV and battery degradations throughout the project’s lifetime. First, the optimal design of the proposed HES based on the economic performance was investigated for different control strategies, and then the technical and environmental characteristics were discussed in detail. A sensitivity analysis is also provided in this section to explore the effects of critical parameters on the optimization results. The values of critical parameters required to run the simulation are provided in
Table 4.
3.1. Optimal Design
This study examined the optimal design of the grid-connected PV system to electrify a residential house based on the lowest NPC. HOMER software was used to simulate and optimize all possible combinations. An hourly time series simulation was performed for each feasible HES to examine the technical and economic feasibility to obtain the best feasible economic scenario. The optimization results for the HES using different control strategies are depicted in
Table 5. It is evident from the results that the optimal HES is achieved by using the CC strategy. When considering the effects of the multi-year parameters, the optimal design using the CC strategy consists of 7 kW PV, 20 batteries, 4 kW converter with average power grid purchases of 11,675 kWh/year. In this scenario, NPC is calculated as
$33,102, which is 23.7% more expensive than the system using the CC strategy without considering the multi-year parameters that comprise 6 kW PV, 18 batteries, and a 4 kW converter with average power grid purchases of 11,995 kWh/year. On the other hand, the LF strategy is much more expensive than the CC strategy. The system using the LF strategy with the effect of multi-year parameters consists of 23 kW PV, 36 batteries, 4 kW converter, and average power grid purchases of 6215 kWh/year. The NPC of this option is estimated at
$55,113, which is 35.7% higher than the NPC of the system using LF strategy without considering the multi-year parameters that consist of 19 kW PV, 30 batteries, 4 kW converter with power grid purchases of 6299 kWh/year. This finding is mainly due to the fact that by considering the effects of multi-year parameters, such as PV and battery degradation, higher PV and battery capacity are required to compensate for the loss of power resulting from the degradation. Moreover, the study assumed that the battery must be replaced when the degradation exceeds 30% of the battery capacity, which leads to an increase in the replacement cost. All these factors result in increasing the NPC. The cost summary of the suggested hybrid generation configuration using LF and CC strategies with and without the effect of multi-year parameters is provided in
Figure 10.
3.2. Technical and Environmental Evaluations
Table 6 shows the technical specifications of the optimal scenarios. Note that with the multi-year effects, the value of each parameter represents the average amount for 20 years. It is clear from the results that the PV production in the LF strategy is higher than that of the CC strategy. This is mainly because the national grid in the LF strategy is not capable of charging the battery, which can be charged only by the excess power of the PV. Therefore, a large PV capacity is required, which results in high PV production and hence a high renewable fraction. On the other hand, the national grid in the CC strategy can feed the load and charge the battery. This leads to an increase in grid purchases, reduces the PV capacity, and reduces the renewable fraction. For the LF strategy with the effect of a multi-year module, the renewable fraction is evaluated at 69.6%, while this value is reduced to 68.7% by neglecting the multi-year effects. On the other hand, the renewable fraction for the system with and without the multi-year module for CC strategy is calculated as 37.6% and 34.7%, respectively. These findings are based on the argument that by considering the multi-year effects, such as PV and battery degradation, higher renewable components capacity is required to cover load demand.
The electrical load which cannot be covered by the HES is referred to as unmet load. It occurs when the output power of the HES is lower than the required load demand. Minimizing the unmet load enhances system reliability. The results show that optimal designs with multi-year effects have a lower unmet load than those without. The unmet load for LF and CC with multi-year effects is calculated as 114 and 121 kWh/year. For the system without multi-year effects, these values are estimated at 155 and 129 kWh/year.
Battery autonomy is an important technical indicator in the HES. It refers to the period in which the battery alone can satisfy the load in the event of power shortages. The analysis demonstrates that the LF strategy outperforms the CC strategy regarding autonomy. The LF strategy for the system with and without multi-year effects has battery autonomies of 13.2 and 11.8 h, respectively, while for the CC strategy, these values are evaluated at 7.23 and 7.06.
Another important parameter is the battery life. High battery life leads to a more cost-effective system since it reduces the replacement cost of the battery. It is obvious from the results that the battery life, when considering the multi-year effects, is shorter and lower than the battery life without multi-year effects. This is because when considering the multi-year effects, the battery degradation reduces the battery lifetime. With multi-year effects, the battery life for the LF and CC strategies is calculated as 9.92 and 7.75 years, respectively. In the case of neglecting the multi-year effects, these values become 15 and 10.2 years. It is clear that the LF strategy has higher battery life than the CC strategy. This is mainly because the battery in the LF strategy is not charged by the national grid, which results in reducing the charging/discharging cycles per year and hence increasing the battery life. On the other hand, the battery in the CC strategy can be charged by both PV and national grid, and this increases the charging/discharging cycles per year and hence decreases the battery life.
Greenhouse gases are the major driver of global warming. It is expected that there will be a 3.2–5.4 °C increase in the global temperature by the end of the century based on the current greenhouse gas emissions. The combustion of fossil fuels releases CO
2, which makes up most of greenhouse gas emissions [
29,
30]. In this study, the national grid is the only source of emissions. The environmental analysis shows that the LF strategy is more environmentally friendly than the CC strategy. This is mainly due to the high grid purchases in the CC strategy compared to the LF strategy. The CO
2 emissions for LF and CC with multi-year effects are calculated as 3928 and 7379 kg/year, respectively [
31]. On the other hand, for a system without multi-year effects, these values are estimated as 3981 and 7581 kg/year.
3.3. Year–Year Analysis
The multi-year model takes into account the changes that occur throughout a project. The simulation is carried out every year during the project’s life. When neglecting the multi-year effects, a one-year simulation is carried out, and the results are extrapolated over the remainder of the project’s lifespan.
Figure 11 and
Figure 12 show the year–year analysis using the LF and CC strategies, respectively. The results indicate that by neglecting the multi-year effects, the values of PV production, grid purchases, battery output energy, and CO
2 do not show any variation for the project’s lifetime. On the other hand, when considering the multi-year effects, PV production decreases over the years, mainly due to the effects of degradation. This leads to increasing the grid purchases and hence increasing the CO
2 emissions. For the LF strategy, the PV production decreases from 34,882.46 to 31,713.61 kWh/year after 20 years, leading to an increase in grid purchases from 6188.8 to 6241.54 kWh/year and CO
2 emissions from 3911.2 to 3944.65 kg/year. In the case of the CC strategy, the PV production decreases from 10,616.4 to 9651.97 kWh/year, leading to increased grid purchases from 11,483.6 to 11,771.9 kWh/year and CO
2 emissions from 7286.29 to 7439.84 kg/year with some fluctuations among years. Moreover, the results show that the battery output energy decreases for certain years due to the battery degradation and then suddenly increases again, which relates to the battery replacement.
3.4. Sensitivity Analysis
The effects of some critical parameter variations on the NPC concerning the multi-year effects are discussed in this section. The sensitivity analysis of the solar radiation, average load, and battery degradation limits considering −20% and +20% variations of the original value of each parameter are depicted in
Figure 13. It is obvious that the variation of the solar radiation from −20% to +20% of its original value leads to a decrease in the NPC by 13.8% and 5.9% for LF and CC, respectively. The same variation rate in the average load increases the NPC by 47.1% and 48.6% for LF and CC, respectively. Moreover, the NPC decreases by 11.2% and 14.2% for LF and CC, respectively, with the same variation ratio in the battery degradation limit.
4. Conclusions
Integrating RESs into the grid can provide reliable, economical, and environmentally friendly power sources. The best component selection and control technique are critical considerations in the design of the HESs. The present study investigates the optimal design of a grid-connected PV/battery system to electrify a residential house in Karbala, Iraq, using HOMER software. LF and CC are used to control the energy flow among components. A detailed technical, economic, and environmental assessment is performed with and without the multi-year effects, such as PV and battery degradation. The results demonstrate that the CC strategy is more economical than the LF strategy for the proposed HES. Under the CC strategy, the multi-year effects increase the required PV size from 6 kW to 7 kW and the required number of batteries from 18 to 20, leading to an increase in the net present cost from $26,750 to $33,102 and a decrease in the CO2 emissions by 2.66%. While, for the LF strategy, considering the multi-year effects leads to an increase in the required PV size from 19 kW to 23 kW and the required number of batteries from 30 to 36, resulting in an increase in the net present cost from $40,607 to $55,113 and decrease the CO2 emissions by 1.33%. Variations in some critical parameters, such as solar radiation, average load, and battery degradation limits, affect the optimization results dramatically. The findings of this work show the importance of the multi-year effects and control strategy in the optimal design of HES.