do not necessarily reflect the views of UKDiss.com.
The thesis aims to analyse the performance and viability of solar-based microgrids in isolated agricultural areas. It goes through the benefits and limitations of a microgrid compared to traditional “pole and wire” transmission as well as how a microgrid in isolated Western Australia would perform. This thesis aims to show that there is a point where the cost of adding or replacing current transmission infrastructure is outweighed by choosing a stand-alone power system.
Nomenclature
This section contains a listing of all words and abbreviations that are used through this thesis.
AEMO Australian Energy Market Operator
C20 20-hour battery discharge rate
COE Cost of Energy ($/kWh)
DC Direct Current
DOD Depth of Discharge
HOMER Software for microgrid and distributed generation power system design and optimization
O&M Operations and Maintenance
PV Photovoltaic
SOC State of Charge
SPS Stand-alone Power System
STC Standard test conditions, for solar PV array [1kW/m2, 25°C]
SWER Single Wire Earth Return
SWIS South-West Interconnected System
UPS Uninterruptible Power System
WEM Wholesale Electrical Market
Contents
4.2.1 Grid Electricity Information
4.2.3 Wind Turbine Information
4.2.4 Inverter/Charger Information
4.2.5 Battery Storage Information
4.4.1 Configuration 1: Hybrid System
4.4.2 Configuration 2: PV-Battery System with Generator Backup
4.4.3 Configuration 3: Hybrid System with no Generator Support
4.4.4 Configuration 4: PV- Battery System with no Generator Support
List of Figures
Figure 1 Western Australian power distribution showing the SWIS region[2].
Figure 2 Solar Insolation for Koorda, WA
Figure 3 Wind speed averages for koorda, WA
Figure 4 Scaled Average Load Curve
Figure 5 Average Solar Radiation for 18th June, Koorda
Figure 6 Average Wind Speed for May 8th, Koorda
Figure 7 Turbine Power Output, May 8th
Figure 8 Graphical Representation of Storage Methods(ref)
Figure 9 Graph showing the common region supported by PV of the load
Figure 10 Graph showing the supported region from the hybrid outputs of the load
Figure 11 HOMER Schematic Used
Figure 12 Power output of Bergey Excel 1kW DC turbine
Figure 13 Average Electrical Production Distribution – Hybrid System
Figure 14 Generator Output – Hybrid System
Figure 15 Battery Status- Hybrid System
Figure 16 Grid Extension versus SPS – Hybrid System
Figure 17 Average Electrical Production Distribution -PV-Battery with Backup
Figure 18 Generator Output – PV-Battery System with Backup
Figure 19 Battery Status – PV-Battery System with Backup
Figure 20 Grid Extension versus SPS – PV-Battery System with Backup
Figure 21 Average Electrical Production Distribution – Hybrid System with no Backup
Figure 22 Battery Status – Hybrid System with no Backup
Figure 23 Grid Extension versus SPS – Hybrid System with no Backup
Figure 24 Average Electrical Production Distribution – PV-Battery System with no Backup
Figure 25 Battery Status – PV-Battery System with no Backup
Figure 26 Grid Extension versus SPS – PV-Battery System with no Backup
Figure 27 Cost Comparison of each System
Figure 28 COE Comparison of the Systems
Figure 29 Cost of Systems versus Power Bill Cost to Determine Payback Period
Figure 30 Cost of an SPS versus Cost of Grid Extension
List of Tables
Table 1 Comparison of different storage technologies (ref)
Table 2 System Configuration Options
Table 4 Hybrid System Component List
Table 5 PV-Battery with Generator Backup Component List
Table 6 Hybrid System without Generator Support Component List
1.0 Introduction
Justification of the problem and the application of solutions
Renewable Energy is coming to the forefront of electrical generation methods. Traditional methods are outdated and environmentally unfriendly, the attitudes towards power generation need a dramatic shift. This is why we need renewable energy generation, to maintain our electrical future and security. Although there has been an increase in the uptake of renewable energy generation, there needs to be further investment particularly in the storage of the energy produced. Renewable energy generation up until recently meant that the consumer would generate their power, use what they needed, then export it to the grid. This isn’t a very economical way to use their power as large amounts of excess energy were just fed into the grid for little return. This is where electrical storage comes into play. By storing the excess power generated, consumers are able to further reduce their grid dependence by being able to access their stored power when the renewable sources aren’t producing enough to cover their usage. Storage is the key to future energy sustainability and allowing a consumer to no longer depend on an aging and sometimes unreliable power grid.
The Power grid in Australia, particularly regional Western Australia, is nearing the end of its designed life span. Large lengths of line built in the boom periods of the 1970s and 80s are coming to a point of needing replacement. With approximately 101,097km[1] of line in the SWIS, as seen in figure 1, it quickly becomes a costly exercise to maintain and replace these sections of lines.
Figure 1 Western Australian power distribution showing the SWIS region[2].
Western Power have a huge task in the foreseeable future of having to make the decision of what to do with the power lines that need replacement. Long rural feeders pass over huge sections of open land with a constant risk of poles being damaged and the risk of bush fires and power interruptions always apparent. This coupled with plans to change the way WEM works to more closely resemble the national regulatory framework from AEMO, now is a perfect time to explore options for power in Western Australia.
So considering the current state of the Western Australian power grid and the direction the regulators are taking, how can renewable energy benefit the system? The opportunity for stand-alone power systems are already being explored by Western Power as a viable alternative to the replacement or upgrade of a long feeder with few consumers at the end of it. The concept of an SPS is a relatively straightforward one consisting of a renewable generator such as a solar array and/or a wind turbine, coupled with a battery bank and a backup diesel generator for when storage is depleted over long periods of low renewable generation. The goal for this is to design a system that can offer the reliability of the electrical grid without being connected to it. The advances in renewable energy components and subsequent decreases in cost allow for a system to be installed at a cost that is competitive with a power line.
• Statement of personal achievements
• Thesis outline
2.0 Background
• Critique of current practice.
Renewable energy resources are in abundance in Australia. Solar and wind being at the forefront of the renewable resources. This renewable resource is being harvested more and more in recent years, particularly in regional Western Australia, either in an off-grid form where a grid connection is either unwanted or difficult to implement, or in a grid connected form where excess power generation can be fed back to the grid. These grid connected forms often lack the ability to store energy produced and rely on expensive grid electricity during peak periods where renewable energy cannot supply the required demand.
Evidently the addition of storage to a system would create additional costs but these costs could be justified by analysing the benefit of this additional storage capability. This thesis aims to complete this analysis and establish the actual benefits of installing a storage system and compare a stand-alone system to a grid connection. This chapter will look at the sizing requirements for a system to power a larger than average power consumer in the Wheatbelt region of Western Australia.
Data for this analysis was collected for the town of Koorda in the Wheatbelt region of Western Australia, approximately 236km from Perth. Load data was taken from Energy Made Easy [6] where a large household average usage for the year and per season can be obtained. This could then be scaled as necessary for agricultural operations. Load curves could then be produced based on seasonal trends. Weather data was obtained for the previous 25 years in that region, an average was created then used for prediction of future weather conditions. Weather data was collected from the Bureau of Meteorology[3]. This weather data could then be overlaid with the load profiles to understand sizing requirements of the system and how the solar and wind aspects would operate, considering the periods of low renewable energy density. The following figures demonstrate the renewable energy densities over the period of the year.
Figure 2 Solar Insolation for Koorda, WA
Figure 3 Wind speed averages for koorda, WA
As can be seen from figure 1, the solar energy density is lowest over the middle part of the year from May to July. Wind speeds however from figure 2 can be seen to be reasonably consistent, particularly in the afternoon.
2.0.1 PV Power
A PV array can be defined as an electrical device that produces DC electricity proportional to the solar radiation on the modules. Power output from a PV array can be calculated using a relatively straight forward equation 1.
PowerPV=CPVηPVIns.incidentIns.STC[1+βTPV-TPV,STC] (1)
However if there is no effect from temperature on the PV array, i.e. the temperature coefficient of power is equal to 0, then only the first half of the equation is used.
PowerPV=CPVηPVIns.incidentIns.STC (2)
Where , CPV – PV array rated capacity under STC [kW], ηPV – Efficiency of the array, i.e. de-rating factor [%], ins.incident/ins.STC – Incident solar radiation on the array [kW/m2]/Incident solar radiation under standard testing conditions [1kW/m2], β – Temperature coefficient of power [%/ °C], TPV – Temperature of the PV module [°C], TPV,STC – Temperature under standard test conditions [25°C]. The de-rating factor is a key indicator of the modules performance and depends on factors such as temperature, mismatched modules and dirt built up on the modules.
2.0.2 Wind power
By using a wind turbine we can convert kinetic wind energy into electrical energy through a rotor then a gearbox and then a generator. Wind energy is a very unreliable form of power generation, this is dimply due to wind never blowing at a constant velocity. Also if the wind speed is too high or too low the wind turbine would not spin. A prime example of this is the blackouts in Adelaide in September 2016 [4], where a massive storm came through causing the majority of the wind farms to drop production significantly as they could not cope with the huge wind gusts. The back-up then also failed as too many faults occurred in a short period of time eventually causing power loss to the majority of the state.
Also much like solar panels, wind turbines are not able to convert all energy supplied to it into electrical energy. Wind turbines are limited to a maximum efficiency of 59.3%. This is known as Betz’ Law [5] after Albert Betz, a noted German physicist, found that a turbine cannot convert more than this percentage of wind energy into mechanical energy through the rotor. This gives us a coefficient of power for calculating the maximum electrical power produced through wind energy. This can be obtained through the following equation 3:
Power= 12×αpρ×π(d2)2×V3 (3)
Where, αp – coefficient of power [0.59], ρ – Density of air [1.225kg/m3 at sea level at 15°C], d – Diameter of the rotor [m], V – wind velocity [m/s].
2.1 Daily load profiles
To create daily load profiles for the purposes of simulation there are several forms of data that were required. These included average power usage for the area, average power usage for agricultural purposes and seasonal load curves. The average power consumption for a large family home in the area is given as 22.9kWh/day or 8356kWh/year [6]. Home sites usually only make up 20 to 30% of total farm electricity usage according to a study made by Hydro Tasmania whilst auditing power usage of farms in Tasmania [7]. Assuming a similar case for farms in Western Australia with the addition of required agricultural equipment, i.e. irrigation pumps for the wheat farms, this would boost power consumption to approximately 92.9kWh/day. Although this tends to vary due to seasonal changes. Taking this into consideration, the load curves generated by HOMER look as follows:
Figure 4 Scaled Average Load Curve
As can be seen from the load curves, the seasonal impact can be quite significant, with a higher power distribution in summer months compared to spring. The winter curves show a “duck curve” effect where there are two distinct peaks in power in both the morning and evening periods. Summer has the highest loading, this is due to the extra run times of equipment such as air conditioning and also pumping equipment to make sure the crops are kept adequately irrigated through the hot summer months.
2.2 Daily Renewable Energy
The availability of solar radiation would vary not only between seasons but also across the period of the day. This is why renewable energy can be difficult to make use of. When designing a solar system the best practice is to find the period when solar radiation is at its lowest and make sure the system can support the full load with the minimum input. In the Wheatbelt region sunlight hours generally range between 10 hours in winter months to 13 hours in summer months (ref). For Koorda in the Wheatbelt region it was found that the solar radiation at its highest was 7.99kWh/m2 and at its lowest was 2.87kWh/m2 with an average of 5.405kWh/m2. For the purposes of estimating the solar radiation and then designing the system the worst month is chosen, June. Figure 4 shows the solar radiation of the course of 24 hours for the 18th June.
Figure 5 Average Solar Radiation for 18th June, Koorda
The next renewable resource that is to be looked at is wind energy. Wind energy, much like solar, varies over the course of the day and has different profiles for different seasons. Again the worst month for wind speeds was used to ensure a correctly designed system. To ensure an output from the wind turbine, wind speeds have to be above 4m/s. The month wind speed averages are all above that mark with the highest being November with 17.75m/s and lowest is May with 13.2m/s. A low wind speed day in May looks as follows in Figure 5. The converted Power then for the same day is shown in Figure 6.
Figure 6 Average Wind Speed for May 8th, Koorda
Figure 7 Turbine Power Output, May 8th
2.3 Why Storage is Important
There are many benefits to incorporating energy storage into a system. There are also several limitations that come with the different forms of storage. Although, having access to storage and the subsequent ability to store excess produced power would allow greater independence from conventional forms of Generation. There are many different forms of energy storage now available to cover most applications, whether it be large scale utility sized storage or at the consumer level. Table 1 demonstrates some of the different technologies available and some of their attributes and limitations. Figure 7 then shows these different technologies graphically.
| Type | Attributes | Limitations | ||
| Efficiency (%) | Life time | Response (s) | ||
| Flywheel | 85 | 20 | 0.1 | Large standby losses, Low energy Density |
| Pumped Hydro | 80 | 50 | 10 | Has to be in right location, expensive |
| Compressed Air | cef=1.3a fhr=4300kJ/kWhb | 25 | 360 | Has to be in right location, expensive |
| Hydrogen | 50 | 25 | 360 | Extremely Flammable |
| Thermal Energy Storage | 75 | 30 | Tens of minutes | Storage is expensive |
| Batteries | 80 | 10 | 0.01 | Still a developing technology |
| Capacitor | 80 | 10 | 0.01 | Low energy density |
| Superconducting Magnetic Energy | 90 | 30 | 0.01 | Low energy density, expensive |
Table 1 Comparison of different storage technologies (ref)
a) Charge energy factor, b) Fuel heat rate
Figure 8 Graphical Representation of Storage Methods(ref)
The choice of energy storage system for a particular application is generally based on the required discharge time. That is why energy storage systems are divided into three main categories, these being Power Quality, Grid Support and Bulk Storage. Large scale storage is however very expensive and requires a lot of space, especially when looking at pumped hydro which requires the building of a dam and reservoir to work, note Snowy Hydro for example. The bulk of research is being done in battery technology. This is a means to create inexpensive and efficient battery technologies that can suit a wide range of applications.
By the effective use of storage systems on a distributed generation level we can actively reduce dependence of grid systems and ultimately reduce the requirements from large scale power plants. This would have a positive environmental effect and help create clean renewable energy. Although the use of renewable energy in a large power plant scale is still a relatively new technology and difficult to get the same power outputs from, the future of this technology couple with an advanced storage technique would see us being able to future-proof energy supplies.
The energy storage systems help to combat the number one issue with renewable sources which is intermittency. Again with an energy storage system we can then smooth out the intermittent renewable source and provide a range of environmental and financial benefits with the cost expected to decline over 50% in the next 10 years.
2.4 Grid Connections
Generally speaking an SPS or microgrid is usually an alternative to a grid connection. Some remote communities may have no other choice than to implement an SPS for their power requirements but usually there would be a means to a grid connection accessible or already connected. This is why it is important to assess whether it would be beneficial to construct an SPS over a power line. Environmentally it would always be better to go with a renewable source but economic viability is the main prohibitor of these renewable sources. The cost of a SWER line with wooden poles would cost approximately $30,000 per km (ref). This is the cheapest and most basic transmission method. There would then be ongoing costs for the provision of the electricity provided unlike a SPS where the initial cost may be high but running costs are minimal with only minor maintenance required and the supply of diesel fuel to a generator if required.
• Review of the theoretical issues (and others) relevant to the problem (where appropriate)
• Detailed problem definition and solution requirements.
3.0 Problem Solution
• Overall and specific arguments indicating how and why the solution has been identified or chosen.
This is to refer to appropriately presented supporting data such as graphics, tables and so forth.
To come up with a viable solution we need to make sure that the PV, wind and storage are all sized adequately. Along with assessing the requirement for a backup generator for any unmet load, particularly in winter months. It is also necessary to try and avoid wasted energy. Wasted energy is energy that is neither used by the load nor able to be stored by the battery as the battery would be exceeding its maximum SOC. When it comes to designing an SPS the storage sizing is practically the most important part of it. Establishing the performance of the battery based on its voltage, SOC, DOD and capacity is the first step to sizing a storage system.
There are several established practices and standards for the sizing of a lead-acid batteries integrated into an SPS. These include IEEE Std-1013-2007 (ref) which provides a method for determining the energy-capacity requirements of lead-acid batteries in an SPS. This standard however only covers the case of a PV connection to a battery bank not a hybrid system. For that we turn to IEEE Std-1561-2007 (ref), this standard is a guide for optimising the performance and life span of lead-acid batteries in remote hybrid power systems.
3.1 System Requirements
3.1.1 PV-Battery System
Usually a PV array would be sized based on the average load over the number of sun hours available. Most systems, being grid connected, only provide 10-60% of the energy requirements of a system. This will need to be scaled up to provide up to 100% of the load in the case of a PV-Battery only system. For this thesis, the property chosen has an approximate daily load of 92.9kWh. This means the PV array needs to be able to support up to 100% of this load. Therefore over the minimum 10 hours of sunlight that the Koorda area receives during winter its required AC capacity would be as follows:
PPV AC= Load(kWh/day)Sunlight Hours=92.910=9.29kW
The required DC capacity would then be calculated from multiplying the AC capacity with the system efficiency, this is taken as 90%
PPV DC= PPV ACEff.=9.290.9=10.32kW
These calculations are based on the system drawing all the power from the PV. However, with the inclusion of a battery system the PV is also required to charge the battery. As this is dealing with an SPS the PV array should be sized in such a way that its output is 1.3 times the loading (ref). Therefore, the new equivalent DC loading is:
PDCAdj.=1.3×PPV DC=1.3×10.32=13.42kW
For this property, with correct storage, a 13.42kW PV array could support 24 hours of load. To be able to accurately size the battery, it is required to calculate the amount of power that will be produced by the system for the design month. This requires finding the surface area of the PV using the panel efficiency, insolation and adjusted DC load. Efficiency in this case is 16.7% using a Jinko Solar Eagle 72 polycrystalline panel (ref):
Area= PDCAdj.1sun×eff.panel=13.421×0.167=80.36m2
This requires a quite a large surface area and this would have to be considered when putting a system in. Using the Jinko panels this would equate to 42 panels. The surface area is calculated to estimate the total energy given by the PV and then to calculate the approximate storage required to support the load.
When sizing a battery system, it is important to consider the number of days of autonomy. A grid system connected system wouldn’t require more than one days’ worth of storage where as an SPS could be up to 12 days depending on the power reliability and available investment into the system.
To estimate the load on the battery the solar power produced is first overlaid onto the load curve to analyse how much power is being supported by the PV, how much power is redundant, and how much power should be supported by the battery.
Figure 9 Graph showing the common region supported by PV of the load
From Figure 9 it can be determined that the common area of load supported by the PV system, during which the battery is charging, would be 51.45kW/day. This selected day, being in winter, has a slightly lower average power usage for the day of only 89.33kWh for the day. This leaves a total of 37.88kWh/day that must be supported by the storage system. This is the minimum amount of load that could be placed onto the storage. The maximum load is the difference between solar output and the supported load. This would give a total amount of 63.99kWh/day.
As this is a large system it was elected to use a 48V DC system voltage. This would maximise the system voltage whilst attempting to reduce the overall storage required. The required storage is then calculated based on the following equations:
Minimum Storage Load=Minimum LoadSystem voltage=37.88kWh48V=789.17Ah/day
Maximum Storage Load=Max LoadSystem Voltage=63.99kWh48V= 1333.125Ah/day
The selection of the battery for this usage was a lead-acid, deep cycle variant. This allowed cheap storage whilst maintaining a larger number of cycles at a deeper discharge. Lead-acid batteries generally have a MDOD of approximately 80% (ref). This can however be affected by heat and over factors. Although assuming an 80% MDOD and required storage of 3 days the battery storage would be:
Storage Min=Ahload×Days of StorageMDOD=789.17×30.8=2959.39Ah/day
Storage Max=Ahload×Days of StorageMDOD=1333.125×30.8=4999.2Ah/day
At standard ratings (25°C/C20), there is a 0.96 multiplier onto the required storage. Bringing the required storage to:
Standard CapacityMin=Storage (Min)Cap Multiplier=2959.390.96=3082.7Ah/day
Standard CapacityMax=Storage (Max)Cap Multiplier=4999.20.96=5207.5Ah/day
3.1.2 Hybrid System
To further improve the SPS, a hybrid system can be introduced. A hybrid system would include more than one RE source, in this case a PV array would be coupled with a wind turbine and supported, where necessary, by a generator. This many different components can lead to a costlier system and potentially one that takes up more space. There could also be a slightly higher maintenance cost also. Although these things will be considered when selecting the optimal system to use. As the renewable source is intermittent and varies seasonably, it is more likely a generator would be required during the winter months for longer periods of overcast days. Wind is significantly more variable than solar irradiation, therefore it will only comprise approximately 10% of the power generated, the remaining 90% would come from the PV array.
Following the same steps as the previous section:
90% Solar load:
92.9×0.9=83.61kW
The capacity of the array is:
PPV AC= Load(kWh/day)Sunlight Hours=83.6110=8.361kW
The required DC capacity would then be calculated from multiplying the AC capacity with the system efficiency, this is taken as 90%:
PPV DC= PPV ACEff.=8.3610.9=9.29kW
The DC load is then adjusted to cover 1.3 times the load of the system. This becomes:
PDCAdj.=1.3×PPV DC=1.3×9.29=12.077kW
This means that a PV array of 12.077kW is sufficient to cover 90% of the load and charge the battery.
Using the Jinko Solar Eagle 72 polycrystalline panels, the PV array surface area would be:
Area= PDCAdj.1sun×eff.panel=12.0771×0.167=72.32m2
This new area reduces the required panels from 42 to 38 panels and saves approximately 8m2 of space.
The wind turbine will then be able to pick up the remaining 10% of the load equivalent to 9.29kW. The turbine capacity that is required would be:
PACWind=LoadHours of wind=9.2914=0.664kW
Then coupled to the inverter with the 90% efficiency, the DC capacity is then:
PDCWind=PACWindeff.=0.6640.9=0.737kW
The design of the system requires the wind turbine to be rated 1.3 times the load. Therefore, the adjusted size would become:
Padj=1.3×PDCWind=1.3×0.737=0.958kW
When considering the size of the wind turbine and finding the diameter, an average wind speed of 12m/s was used as that was the lowest average for the region. This returned a turbine diameter of 1.4m.
Figure 10 Graph showing the supported region from the hybrid outputs of the load
In the hybrid configuration, it was found the renewable sources supported 60kWh of the daily 89.33kWh. This means 29.33kWh is required to be supported by the battery storage. This is the minimum required. The maximum is calculated from supporting full load with the system. This would give a value of 34.56kWh.
Reiterating as this is a large system it was elected to use a 48V DC system voltage. This would maximise the system voltage whilst attempting to reduce the overall storage required. The required storage is then calculated based on the following equations:
Minimum Storage Load=Minimum LoadSystem voltage=29.33kWh48V=611.04Ah/day
Maximum Storage Load=Max LoadSystem Voltage=34.56kWh48V= 720Ah/day
The selection of the battery for this usage was a lead-acid, deep cycle variant. This allowed cheap storage whilst maintaining a larger number of cycles at a deeper discharge. Lead-acid batteries generally have a MDOD of approximately 80% (ref). This can however be affected by heat and over factors. Although assuming an 80% MDOD and required storage of 3 days the battery storage would be:
Storage Min=Ahload×Days of StorageMDOD=611.04×30.8=2291.4Ah/day
Storage Max=Ahload×Days of StorageMDOD=720×30.8=2700Ah/day
At standard ratings (25°C/C20), there is a 0.96 multiplier onto the required storage. Bringing the required storage to:
Standard CapacityMin=Storage (Min)Cap Multiplier=2291.40.96=2386.875Ah/day
Standard CapacityMax=Storage (Max)Cap Multiplier=27000.96=2812.5Ah/day
3.2 System Options
By tabulating the information in to table 3, the options for configuration and subsequent implementation can be seen.
| System
Design |
Capacity of the system | Storage Required at 48V | ||
| Minimum Storage | Maximum Storage | |||
| PV-Battery | 13.42kW of PV at 80.36m2 | 3082.7 Ah/day | 5207.5 Ah/day | |
| Hybrid | 12.077kW of PV at 72.32m2 with a 0.958kW 1.4m Diameter Turbine | 2386.875 Ah/day | 2812.5 Ah/day | |
Table 2 System Configuration Options
It can be seen from the table that significantly less storage is required in the hybrid configuration and there is also fewer panels. However, there is the addition of the wind turbine and stand by generator that may negatively impact this option’s viability. There is also the option of a grid extension/connection. Whereby the grid is extended to the property and powered via the grid.
4.0 Implementation
• Overview of the implementation or simulation of an implementation
The selected method of implementation was using HOMER software. HOMER is a design and optimisation tool for the creation of microgrids (ref). HOMER, standing for Hybrid Optimisation Model for Multiple Energy Resources, allows a user to input a range of system constraints and system models to return the optimal solution for those parameters. The model was implemented as shown in Figure 11, using this model several optimal solutions were found that varied in cost and configuration.
Figure 11 HOMER Schematic Used
4.1 Data Sources
To simulate the model an assortment of data first had to be collected for the Koorda region. This included the data for weather conditions, loading, and components required.
4.1.1 Weather Data
The wind speed data and the data for solar irradiance was collected from the Bureau of Meteorology (ref). Solar irradiance was taken from hourly averages to provide accurate data over the period. Wind speed was also taken from hourly averages. The availability of the renewable sources is explained in chapter 2.2.
4.1.2 Load Data
The monthly load profiles can be found in figure 3. The average load for a large family house in the region was first collected (ref). It was then interpreted as comprising 25% of the overall load of a farming property. Particularly a wheat growing property with an array of pumping equipment (ref). This returned an overall load of 92.9kWh/day, this would then be a yearly load of 339.09MWh. The load was adjusted seasonally as a load tends to be distributed differently for winter compared to summer.
4.1.3 Component Data
The different components also had to be assessed, this included the PV modules, inverter, generator, wind turbine and battery storage. Selection of a battery was based on system requirements and battery availabilities. The chosen battery was Trojan L16P Battery.
4.2 Cost of the System
The following table 3 shows the cost of each of the components. This includes the PV modules, wind turbine, inverter, batteries and generator. The corresponding grid price is also included. Considering also the analysis of using an SPS as a replacement for a grid connection or to upgrade aging sections of line, the cost of the power line was also included. This was analysed as a SWER line with wooden poles (ref). Also required was the cost of fuel for the generator. This was taken from current average prices of regional WA.
| Component Description | Cost |
| Grid Electricity | |
| Capital cost per km | $29892 |
| Electricity Cost | $0.2647/kWh |
| O&M per km | $177.39/year |
| PV modules | |
| Life Time | 25 years |
| Capital Cost | $1820/kW |
| Replacement Cost | $1650/kW |
| O&M Cost | $50/year |
| Wind Turbine (Bergey Excel 1kW DC) | |
| Life Time | 25 years |
| Capital Cost | $5999 |
| Replacement Cost | $5000 |
| O&M Cost | $120 |
| Inverter | |
| Life Time | 15 years |
| Component Description | Cost |
| Capital Cost | $2142/kW |
| Replacement Cost | $2000/kW |
| O&M Cost | $25/year |
| Storage | |
| Capital Cost | $170/battery |
| Replacement Cost | $130/battery |
| O&M cost | $50/year |
| Generator | |
| Life Time | 15000 hours |
| Capital Cost | $2200/kW |
| Replacement Cost | $2000/kW |
| O&M Cost | $0.05/hour |
| Fuel Cost | $1.50/Litre |
** Explain components
4.2.1 Grid Electricity Information
When analysing the grid extension in HOMER it was assumed that a SWER line with wooden poles would be in use to power the property as described in chapter 2.4. This meant a cost per km of $29892. The rate was calculated from Synergy’s current tariff system (ref) and was found to be $0.2647 per unit (kWh) used. There is also an attributed supply charge. This was accounted by the O&M cost of $177.39/year.
4.2.2 PV Module Information
Cost of PV modules has significantly decreased in the last decade, from paying $3000/kw to now around $1800/kw (ref). Besides the battery system, the PV modules are one of the more costly parts of the system, particularly when using higher end tier 1 panels. Tier 1 panels are manufactured by only the top 2% of solar panel companies. They are built and designed completely in-house with components made solely by the company building them. The company should have over 5 years of experience in the industry and invest heavily in the R&D process. This is to ensure a high standard of product which would be expected to last well into the future.
4.2.3 Wind Turbine Information
In this system a Bergey Excel 1kW DC turbine is used. Designed with a rotor diameter of 2.5m and a peak output of 2kW it is perfect for the SPS application this thesis is analysing.
Figure 12 Power output of Bergey Excel 1kW DC turbine
A low cut in speed of 2.5m/s and 3m/s start up speed allows this turbine to be active in low wind situations. It also has the ability to charge a battery bank of 12-48V. Considering the battery bank in use is a 48V bank this makes it again perfect for this application. Wind technology however is costly as the investment into it hasn’t been as great as the investment into solar or other renewable technologies. The cost of the wind turbine is approximately $5999 (ref).
4.2.4 Inverter/Charger Information
Perhaps the most important part of the system, the inverter/charger regulates all the power of the system, converting the DC renewable sources into usable AC power for the consumer to use. As well as regulating the charge and discharge through the battery. SMA is a leading inverter producer particularly for off-grid usage, a Sunny Island inverter can cost approximately $4284 for a 2kW version (ref).
4.2.5 Battery Storage Information
Although most systems are now heading towards Li-Ion technology, the chosen battery for this SPS was a Lead-Acid variant. Lead-Acid is a proven technology that has been in use for a number of years, particularly in back-up power systems and UPS purposes. As seen in figure 8, it is also well suited for an SPS application. In this case the Trojan L16P was the selected battery. A 6V, 360Ah deep cycle battery. To get the 48V bus voltage it would mean 8 batteries would form a string and the number of parallel strings would be dependent on the required storage Ah.
4.2.6 Generator Information
An important aspect of an SPS is the generator, without the generator the intermittent nature of renewables coupled with the lack of grid may cause loss of power or increased cost to add additional storage to compensate. The idea is to minimise running times however. Cost of Diesel is based on local average prices, and currently for Perth it is approximately $1.50. The generator could then be optimised to the necessary size by HOMER.
4.3 System Configurations
The System was then simulated in HOMER, a number of different configurations were processed and several optimal solutions were outputted. This was to see the overall cost and performance of each variant of the system and then the subsequent trade off point between implementing an SPS compared to a grid extension/replacement.
The optimal configurations were:
- Configuration 1: Hybrid system – PV, Wind turbine, Generator and Battery Storage,
- Configuration 2: PV-Battery system with a Generator back up.
- Configuration 3: Hybrid system with no generator back up.
- Configuration 4: PV-Battery system with no generator back up.
Each of these configurations have their merits and drawbacks which is demonstrated.
4.4 Simulation of System
Throughout this section the performance and viability of the system is analysed. The optimal systems provided by HOMER are critiqued and compared with each other. Key performance indicators would be overall cost of the system, COE ($/kWh), excess power produced and the cost compared to grid. The four configurations previously mentioned are the optimal solutions analysed.
4.4.1 Configuration 1: Hybrid System
The hybrid system takes advantage of all components available to it to create an SPS. This includes a PV array, wind turbine, battery storage and a back-up generator. The construction of the system was as follows:
Table 4 Hybrid System Component List
| Component | Size | Cost |
| PV array | 27kW | $56,043 |
| Wind Turbine | 1x Bergey excel 1kW DC | $7,533 |
| Generator | 5kW Diesel AC | $17,281 |
| Battery | 120x Trojan L16P (15 strings) | $35,449 |
| Inverter | 10kW Inv/Chg | $29,490 |
| Total | $145,796 |
Analysing this configuration’s performance a better understanding of how a hybrid system works. Starting with the electrical load distribution
Figure 13 Average Electrical Production Distribution – Hybrid System
As can be seen, the system follows what was established in chapter 3.1.2 with a 10% load share from the wind turbine and the remaining 90% supported by the PV/Generator. In this case the generator is required during periods of low renewable resource, i.e. the winter months of June, July and August, as well as in times of high load demand, i.e. the summer months of January, February and December. The system however does produce excess electricity. Approximately 12,180kWh/year is produced that cannot be used by the load or stored by the battery. This would occur predominantly in spring and autumn months where renewable resource is quite high and the load may be slightly lower than the summer load. The excess power can be channelled into something useful such as a solar heat pump to provide hot water for the home.
Figure 14 Generator Output – Hybrid System
As established in chapter 3, the battery will charge during the day when the largest proportion of renewable resource is available, this way the excess solar energy can be used by the storage. This storage is then utilised during the morning and evening where there is almost no renewable resource. The battery usage is shown in figure 15.
Figure 15 Battery Status- Hybrid System
As can be seen the battery struggles to charge in the winter months where there is the minimum renewable resource. The autonomy of the system is just under two days. Additional autonomy can be achieved through an increase in battery size if necessary with an increase in cost of the system. As this system is 100% available it is not necessary to design a battery for longer autonomy. The battery also has an expected lifetime of 9.17 years. This means the battery system will have to be replaced at least twice during the course of the system life time of 25 years. The aim is to lessen the amount of time spent at deep discharge so as to increase the life span of the battery.
Figure 16 Grid Extension versus SPS – Hybrid System
By analysing the economics of a grid connection versus a standalone power system, it was established that the trade-off distance would be approximately 1km. This means that a property further than 1km away from an existing grid connection or a property with a line length of greater than 1km that required replacement would become economically viable for an SPS. The cost of power bills over the 25 year system life would be equivalent to:
Power bill cost=Power used ×rate+supply charge=92.9×365×25×0.27+177.39×25=$228,824.25
Meaning the payback period of the system would be equivalent to:
Payback=CoSCoB×system life=145796228824.25×25=15.929 years
This is a lengthy amount of time, where the benefits of a system may not be seen until the second half of its life. This is due to high capital investment. The capital investment for this system is $107,959. The remaining $37,837 is spread over the 25 years in operation cost and component replacement, mainly for batteries. Initial capital is still the most prohibitive aspect of an SPS.
4.4.2 Configuration 2: PV-Battery System with Generator Backup
This system is similar to the hybrid system but it drops the secondary renewable source. This system’s only renewable source is the PV array, the array charges a battery and then during periods of low renewable resource a generator is there to provide support for the system to ensure power availability.
Table 5 PV-Battery with Generator Backup Component List
| Component | Size | Cost |
| PV array | 35kW | $72,648 |
| Generator | 5kW Diesel AC | $16,472 |
| Battery | 128x Trojan L16P (16 strings) | $40,396 |
| Inverter | 10kW Inv/Chg | $29,490 |
| Total | $159,007 |
As with the hybrid system the initial area of interest is the electrical performance of the system.
Figure 17 Average Electrical Production Distribution -PV-Battery with Backup
As can be seen, the PV supports 99% of the load, with 1% supported by the generator. The generator only cuts in during winter and summer months where either the renewable resource is low or the load on the system is high. This system produces a higher amount of excess electricity due to its larger PV array. Approximately 20,329kWh/year excess is produced. Again this may be able to be channelled into something useful such as a solar heat pump so as to reduce this wasted energy.
Figure 18 Generator Output – PV-Battery System with Backup
Figure 19 Battery Status – PV-Battery System with Backup
Again it can be seen that the battery struggles to charge in the winter months where there is the minimum renewable resource. The autonomy of the system is just over two days. Additional autonomy can be achieved through an increase in battery size if necessary with an increase in cost of the system. As this system is 100% available it is not necessary to design a battery for longer autonomy. Due to the lack of a secondary renewable source, the system is more dependent on the battery and has a slightly larger battery than the hybrid system. It also has a higher throughput of the battery leading to a lessening of the battery’s expected life which was 8.20 years. This means the battery system will have to be replaced at least three times during the course of the system life time of 25 years leading to increased costs compared to the hybrid system once again. This is reflected in the comparison to the grid.
Figure 20 Grid Extension versus SPS – PV-Battery System with Backup
By analysing the economics of a grid connection versus a standalone power system, it was established that the trade-off distance would be approximately 1.41km. This means that a property further than 1.41km away from an existing grid connection or a property with a line length of greater than 1.41km that required replacement would become economically viable for an SPS. The cost of power bills over the 25 year system life would be equivalent to:
Power bill cost=Power used ×rate+supply charge=92.9×365×25×0.27+177.39×25=$228,824.25
Meaning the payback period of the system would be equivalent to:
Payback=CoSCoB×system life=159007228824.25×25=17.37 years
This is a longer amount of time than the hybrid system, benefits of a system would not be seen until the latter part of the system life. The capital investment for this system is slightly higher than the hybrid system at $117,880. The remaining $41,127 is spread over the 25 years in operation cost and component replacement, this is again higher than the hybrid system, mainly to accommodate 3 battery replacements.
4.4.3 Configuration 3: Hybrid System with no Generator Support
This system is the same as the hybrid system but it does not have the support of a backup generator, instead it completely relies on the renewable source and battery bank. This requires a larger battery bank to be implemented to ensure the system is reliable and able to support the load in times of low renewable resource.
Table 6 Hybrid System without Generator Support Component List
| Component | Size | Cost |
| PV array | 30kW | $62,270 |
| Wind Turbine | Bergey Excel 1kW DC | $7,533 |
| Battery | 144x Trojan L16P (18 strings) | $40,430 |
| Inverter | 15kW Inv/Chg | $44,235 |
| Total | $154,469 |
First area to analyse is the electrical performance of the system.
Figure 21 Average Electrical Production Distribution – Hybrid System with no Backup
As can be seen, the PV supports 91% of the load, with 9% supported by the wind turbine. The wind turbine allows a reduced PV size compared to the PV-Battery system as the wind availability compliments the solar resource. This allows charging of the battery over a larger window. This system produces a lower amount of excess electricity compared to the PV-Battery system. Approximately 16,434kWh/year excess is produced. Again this may be able to be channelled into something useful such as a solar heat pump so as to reduce this wasted energy. This system does require a much larger battery to support periods of lower renewable resource as there is no Backup.
Figure 22 Battery Status – Hybrid System with no Backup
The battery does struggle slightly to charge in the winter months where there is the minimum renewable resource but with the help of a wind turbine it maintains a higher average SOC than previous systems. The autonomy of the system is almost two and a half days offering a reasonably long period of availability. Additional autonomy can be achieved through an increase in battery size if necessary with an increase in cost of the system. As this system is 100% available it is not necessary to design a battery for longer autonomy. Due to the lack of a backup generator, the system is more dependent on the battery and the renewable sources and hence has a slightly larger battery than the hybrid system. The throughput of the battery is only slightly higher than the hybrid system, however due to the increased size the battery’s expected life was 10 years. This means the battery system will have to be replaced at least two times during the course of the system life time of 25 years similar to the hybrid system. The effect of this reflected in the comparison to the grid extension.
Figure 23 Grid Extension versus SPS – Hybrid System with no Backup
By analysing the economics of a grid connection versus a standalone power system, it was established that the trade-off distance would be approximately 1.27km. This means that a property further than 1.27km away from an existing grid connection or a property with a line length of greater than 1.27km that required replacement would become economically viable for an SPS. The cost of power bills over the 25 year system life would be equivalent to:
Power bill cost=Power used ×rate+supply charge=92.9×365×25×0.27+177.39×25=$228,824.25
Meaning the payback period of the system would be equivalent to:
Payback=CoSCoB×system life=154469228824.25×25=16.87 years
This is a longer amount of time than the hybrid system, benefits of a system would not be seen until the latter part of the system life and is only half a year shorter than the PV-Battery system. The capital investment for this system is slightly higher than the hybrid system at $117,209. The remaining $37,260 is spread over the 25 years in operation cost and component replacement, this is again higher than the hybrid system, mainly because of the larger inverter/charger and the larger battery that would require two replacements.
4.4.4 Configuration 4: PV- Battery System with no Generator Support
This system is the same as the PV-Battery system but it does not have the support of a backup generator, instead it completely relies on the renewable source and battery bank. This requires a very large PV array and battery bank to be implemented to ensure the system is reliable and able to support the load in times of low renewable resource.
Table 7 PV-Battery System with no Backup Component List
| Component | Size | Cost |
| PV array | 40kW | $83,027 |
| Battery | 152x Trojan L16P (19 strings) | $43,352 |
| Inverter | 15kW Inv/Chg | $44,235 |
| Total | $170,614 |
As in the previous sections the first area to analyse is the electrical performance of the system.
Figure 24 Average Electrical Production Distribution – PV-Battery System with no Backup
As can be seen, the PV must support 100% of the load, it must support a much larger portion of the load than the previous systems as well. Without a secondary renewable source or any back up the PV array is very oversized for lower load periods so it can cope with the peaks. This leads to a large amount of excess electricity produced and a very costly system. The system produces 28,121 kWh/year of excess electricity. Again it can be fed into a heat pump or other equipment but it is a much larger amount than previous systems. This system does require a much larger battery to make sure the system is available during periods of lower renewable resource as there is no Backup.
Figure 25 Battery Status – PV-Battery System with no Backup
The battery is the best performing battery out of each system, as it must support the load when the renewable resource is unavailable, as a result it rarely drops to its minimum SOC of 30%. It does struggle slightly to charge in the winter months where there is the minimum renewable resource but due to the large battery size it is able to maintain its charge. The autonomy of the system is the longest of all the systems at 60 hours. Additional autonomy can be achieved through an increase in battery size if necessary with an increase in cost of the system. As this system is 100% available it is not necessary to design a battery for longer autonomy. Due to the lack of a backup generator, the system is completely dependent on the battery and the renewable source and hence it has the largest battery size out of all the systems. The throughput of the battery is comparative to the PV-Battery system with backup. Then due to the large size of the battery its expected life was 9.74 years. This means the battery system will have to be replaced at least two times during the course of the system life time of 25 years similar to the hybrid system. The increased costs of this are reflected in the grid extension.
Figure 26 Grid Extension versus SPS – PV-Battery System with no Backup
By analysing the economics of a grid connection versus a standalone power system, it was established that the trade-off distance would be approximately 1.77km. This means that a property further than 1.77km away from an existing grid connection or a property with a line length of greater than 1.77km that required replacement would become economically viable for an SPS. The cost of power bills over the 25 year system life would be equivalent to:
Power bill cost=Power used ×rate+supply charge=92.9×365×25×0.27+177.39×25=$228,824.25
Meaning the payback period of the system would be equivalent to:
Payback=CoSCoB×system life=170614228824.25×25=18.64 years
This is a longest amount of time of any of the systems, financial benefits of this system would not be seen until the last third the systems life. The capital investment for this system is again the highest of any system at $130,770. The remaining $39,844 is spread over the 25 years in operation cost and component replacement, this is not the highest running costs but it is more than the hybrid system, mainly because of the larger inverter/charger and having the largest battery that would require two replacements.
5.0 Verification
• Outline of the means devised to verify the solution meets requirements and the results achieved from that verification process.
The comparison of the implemented solutions will enable the verification of the results gained. The following table 8 compares the results based on the key indicators establish in chapter 4.4.
Table 8 Key Performance Indicator Comparison
| Configuration | Cost | COE ($/kWh) | Excess Power | Grid vs SPS distance |
| 1: Hybrid System | $145,796 | 0.340 | 12,180kWh/year | 1.00km |
| 2: PV-Battery with Generator | $159,007 | 0.371 | 20,329kWh/year | 1.41km |
| 3: Hybrid without Generator | $154,469 | 0.360 | 16,434kWh/year | 1.27km |
| 4: PV-Battery without Generator | $170,614 | 0.398 | 28,121kWh/year | 1.77km |
5.1 Excess Power Generation
Each system analysed produced a varying amount of excess power. Excess power is power that is neither consumed by the load or able to be stored as the battery SOC is at maximum. Without a grid connection to feed back into this power is essentially wasted across the system. Figure 27 shows the amount of excess power generated by each system.
5.2 System Costs
As previously stated, the main prohibtor of the implementation of an SPS is the cost of the system. These cost is constantly reducing but it is still an expensive technology. The systems compared all came in at over $145,000 as shown in figure 28.
Figure 27 Cost Comparison of each System
As can be seen the configurations vary by almost $25,000. The hybrid system is the cheapest at $145,796. Even though this system has more components than any other system it is able to reduce the storage and PV size to reduce costs. This is due to the presence of a secondary renewable source and a Backup generator to allow for a smaller storage. When a generator is not incorporated the price is increased heavily due to the larger storage required and subsequently larger PV arrays needed to ensure the storage is being adequately charged. The subsequent COE can then be found compared to the grid cost as shown in figure 29.
Figure 28 COE Comparison of the Systems
As can be seen from figure 29, the COE of each system is above that of the grid. This means that each system would have a payback period before any financial benefit is observed compared to having a grid connection. The payback periods are shown in figure 30.
Figure 29 Cost of Systems versus Power Bill Cost to Determine Payback Period
Although an SPS is an expensive system, so is the cost of a grid extension. The previous results have been made on the assumption of an existing grid connection. Where this no pre-existing grid connection the use of an SPS becomes more favourable. The cost of grid extension versus the cost of the SPS can be seen in figure 31.
Figure 30 Cost of an SPS versus Cost of Grid Extension
Once analysed in this way it is easier to see the benefits of an SPS for isolated agricultural areas. The cost of a hybrid SPS at a distance of greater than 1km from the grid, for a system of this size, will outweigh the cost of the power lines needed to service the property.
6.0 Conclusions
• Critical analysis of the outcomes.
• Outline of future development, highlighting particular areas for concern.
7.0 References
[1]”What we do – About – Western Power”, Western Power, 2017. [Online]. Available: https://www.westernpower.com.au/about/what-we-do/.
[2]”ERA”, Erawa.com.au, 2017. [Online]. Available: http://www.erawa.com.au/html/eraAnnualReport/eraAR06/report_on_operations/electricity_division.html.
[3]”Australia’s official weather forecasts & weather radar – Bureau of Meteorology”, Bom.gov.au, 2017. [Online]. Available: http://www.bom.gov.au/.
[4]”SA blackout: Why and how?”, ABC News, 2017. [Online]. Available: http://www.abc.net.au/news/2016-09-28/sa-power-outage-explainer/7886090.
[5]”The Betz limit and the maximum efficiency for horizontal axis wind turbines. Can it be exceeded and does it apply to vertical axis wind turbines?”, Wind-power-program.com, 2017. [Online]. Available: http://www.wind-power-program.com/betz.htm.
[6]”Understand and compare your home electricity usage | Energy Made Easy”, Energymadeeasy.gov.au, 2017. [Online]. Available: https://www.energymadeeasy.gov.au/benchmark.
[7]Hydro Tasmania Consulting, “Energy Self Audit Tool for Tasmanian Farmers”, Hydro-Electric Corporation, 2009.
