3.1. System Configuration and Energy Flow Analysis of the Optimum PVS
Based on the 2017 meteorological recorded data and the OCAS power consumption, the simulation for many considerable configurations were performed to decide an optimal system configuration, as shown in
Figure 8 and
Figure 9. The results obtained by HOMER show that the optimum system comprises PV array (12.2 kW, 47 panels), battery bank (40 units, 10 stings), charge controller (2 units, 8 kW) and inverter (1 unit, 3.8 kW). Although the percentage of maximum annual capacity shortage to annual power demand of the OCAS was set at 0%, an extremely small load, totaling to approximately 0.259 kWh (0.00312% of the total load) was generated during the simulated year. In previous studies on the design of stand-alone power system [
36,
37,
38], such small capacity shortage has been generally considered to have no impact on the operation of the power system. However, in the aquaculture in the OCAS, it must be carefully considered a possibility that the serious damage to the fish occur by the interruption of aeration in rearing the tank [
39]. Therefore, the second optimal system configuration with no unmet load was selected. This system is considered safe for power supply, which comprises PV array (12.5 kW, 48 panels), battery bank (40 units, 10 stings), charge controller (2 units, 8 kW) and inverter (1 unit, 3.8 kW). The performance of this optimum PVS is discussed in
Section 3.1,
Section 3.2 and
Section 3.3, and the economic analysis on this PVS is provided in
Section 3.4.
The energy flow of the stand-alone PVS is shown in
Figure 10. The electricity supply is originally given by PV output (19,106 kWh) and battery storage depletion (about 29 kWh). It is obtained that the amount of electricity of 8291 kWh is consumed by the OCAS. If the amount of PV power generation is more than total amount of power demand and battery charging, the excess electricity is supplied to the dump load at that time. It is found that the excess electricity transferred to the dump load is 43% (8278 kWh) of the total PV power generation. However, this also implies that if the power demand of OCAS or storage capacity of PVS is reasonably increased, it is possible to spend the excess electricity for other operations instead of being transferred to the dump load. Therefore, it results in improving the energy use efficiency of the system [
40].
3.2. Operating Performance of PVS Components
A summary data of the technical and economic performance of PV array, battery bank and inverter is shown in
Table 6. The operation hours for PV power generation is 4156 h and the annual PV power generation is 19,106 kWh. The average power output of the PV array is 2.18 kW, and the average daily accumulated power generation is 52.32 kWh. The levelized cost for the PV array is
$0.066/kWh. The levelized cost of the battery bank is not calculated because the battery bank is charged by the only the PV array. However, the battery wear cost of
$0.196/kWh becomes significantly higher, calculating close to 45% of the COE (
$0.438/kWh) (the simulated value of COE is explained in
Section 3.4). The battery cost is more expensive than that of the PV array, which explains why users sometimes stop utilizing the battery bank and install more solar panels [
41]. This leads to reduction in the initial equipment cost for system design of PVS to the same power demand. However, in the OCAS process, the circulation pump and the aeration pump must be worked for rearing the fish and the dehumidifier is also operated for recovering water from the air through the nighttime. Consequently, it must give preference to install more battery banks for ensuring the safe power supply although it results in increasing the equipment cost [
42]. It is also simulated that the capacity of the inverter of the optimum PVS is 3.8 kW. The actual output of the inverter changes with power consumption of the OCAS. According to the simulation results, the annual amount of electricity imported to the inverter was 8821 kWh, and the annual amount of electricity exported from the inverter was 8291 kWh. The average output of the inverter was 0.947 kW with a minimum output of 0.135 kW and maximum output of 1.66 kW.
3.4. Economic Evaluation of the PVS
In this study, the cost of the stand-alone PVS was evaluated using the total NPC and the levelized COE [
50]. The total NPC of
$46,993 for the proposed PVS is obtained from this simulation. Especially, the total NPC includes several types of cost such as initial capital, O&M, replacement and so on. The breakdown of NPC by type of cost and equipment for the PVS is shown in
Figure 19. It is found that the initial capital cost of the PV system is
$26,115 and reaches to about 56% of the total NPC. It is also remarkable that the cost for the battery bank reaches to about 59% through initial capital and replacement cost. This indicates that the cost of the power storage system in a stand-alone PVS is dominant. The cost of the 12.5 kW PV array occurs at the initial capital and O&M cost. It accounts for 31% of the total NPC of the PVS, while the cost of the inverter accounts for about 6% of the NPC of the entire system. The lowest cost item is the charge controller, which accounted for 4% of the NPC. At the end of the 25-year project, a total remaining value of
$1696 for each facility is calculated.
It is also obtained that the COE of the PVS is
$0.438/kWh, which is more than the current Mexican electricity rate (
$0.221/kWh) [
22]. However, because there is no power grid in the remote areas of drylands, comparing the COE of the PVS with the electricity cost of the areas with power grids may not reflect the actual costs of building the PVS. However, compared with a similar study [
34], the COE is still in a reasonable range, which is economically advantageous [
50] compared with the erection of long-distance transmission lines and utilization of diesel generators [
51]. When calculating COE through Equation (4), the amount of excess power was not considered in the amount of “total electrical load served per year”. In the same PVS, even if the cost is constant and the power consumption (“total electrical load served per year”) is larger than that in current simulation, it is possible to reduce COE. According to the results explained in
Section 3.1, 43% of the generated power is transferred to the dump load as excess power, which is not provided to the OCAS. The simulations of power supply and demand in
Section 3.3.3 show that the system generates excess power during the day. It is suggested that the selection of electrical appliances or the setting of the service time of electrical appliances in the PVS must be considered, and the surplus electricity should be reasonably used to further reduce the COE.
OCAS operators bear certain economic risks due to the high equipment investment and meticulous production technology. Therefore, to guide the reasonable operation of the OCAS, it is very important to establish an effective operation and management method of the OCAS suitable for the operators themselves. The specific management strategy of the OCAS is uncovered in this study. To provide a reference for operators, the production and sales volume of OCAS through field empirical experiments is estimated and discussed below.
Previous researchers systematically conducted LCA and economic analysis of the whole aquaponic systems [
52,
53,
54,
55]. According to previous studies, the market price changes on vegetables and fish are in the range of
$1.46–2.50/lb (
$0.66–1.13/kg) and
$1.46–5.00/lb (
$0.66–2.27/kg), respectively. The production costs reported for vegetables and fish are
$0.75–1.50/lb (
$0.34–0.68/kg) and
$1.46–4.99/lb (
$0.66–2.26/kg), respectively. However, as mentioned in the introduction of this paper, the productivity and energy consumption of the aquaponic system are related to the structure, the scale and the local environment. The design and operation methods of aquaponics also affect the production. Therefore, it is difficult to use the existing data in this work. Instead, a simple economic estimation by referring to the data of local surveys and production data of unpublished empirical experiments in La Paz is achieved. According to the field empirical experiment, taking the OCAS shown in
Section 2.1 as the research object, it can be estimated that the annual harvest is 225 kg tilapia, 78 kg Swiss chard and 48 kg pepper. Based on the market price survey data of fish and vegetables implemented in La Paz, the average sales prices in the last three years for tilapia, Swiss chard and pepper are respectively 51.80 pesos/kg (
$2.23/kg), 24.13 pesos/kg (
$1.04/kg) and 93.18 pesos/kg (
$4.01/kg). Therefore, it can be estimated that the annual sales revenue is
$775.35 (Tilapia:
$501.75, Swiss chard:
$81.12, Pepper:
$192.48). Assuming that the commodity price of these remains unchanged, the sales revenue will be
$19,383.75 within 25 years. If the sales revenue of 25 years is changed to NPC, further reduction of the total NPC may be expected. However, the cost of power equipment let alone the overall material and labor costs is much higher than the sales revenue of the OCAS system.
From the perspective of sales revenue of harvests, it seems even less economical to set up a stand-alone PVS for OCAS. However, compared with the previous aquaponics, OCAS can not only produce food but also alleviate the salinization of soil. In addition, the stand-alone PVS using clean energy has a series of beneficial effects on reducing greenhouse gas emissions [
56]. The important impact of the PVS on the environment should also be positively evaluated. In addition, if OCAS has a stand-alone power supply device, the applicability of OCAS in remote areas will be greatly promoted. Therefore, it is difficult to simply evaluate the budgeting of this system by only looking at the sales revenue. Combined with the values of environmental protection and mitigation of food shortage, the evaluation of the economic and environmental benefits of the whole system is a topic of future work.