Robert Mullen Reynolds

Abstract:

This report looks at use of solar photovoltaic (pv) technology for various stand alone applications (parking meters, lighting systems, remote or mobile locations..etc) where a grid connection may not be feasible. The report introduces the basic principles of solar pv and the different types of pv-related technology available. Global market trends for both on-grid and off-grid systems are analyzed and comparisons are made between Ireland and its EU counterparts. Technology trends are touched upon and future prospects for each technology type are considered. Using simulation software a computer model was designed to simulate a solar pv powered car parking meter located in Drogheda, Co. Louth. Improvements in the systems design showed an increase in the systems capacity to meet the required demands.

1.0 Introduction:

Over the last number of years renewable sources of energy have earned greater acceptance in power generation for a multitude of devices and systems. Such utilization of renewable energy sources is important for Ireland given its lack of resources for fossil fuels [1], the only indigenous fossil fuel being peat. In 2006 Ireland’s energy imports reached 91% [2] making it the most imported energy dependant country in the European Union (EU). Such evidence highlights the importance of renewable energy sources for securing Ireland’s energy needs for the future.

Solar photovoltaic (pv), a technology that converts solar energy directly into electricity has experienced rapid development in the last decade. Advances in manufacturing and design has resulted in improved pv efficiencies, reduced costs and increased productivity [3]. Currently, solar technology in Ireland is mostly limited to low temperature systems, such as solar thermal for heating water in domestic applications. The intermittent nature of direct sunshine in Ireland makes any significant electricity generation from solar pv difficult. However, for small low-powered applications where a grid connection may not be feasible, solar pv is proving to be a cost effective way for power generation.

1.1 Basic principles of Solar Photovoltaic (Pv):

Solar pv systems consist of an array of cells usually made from a semiconductor material such as silicon to convert incident sunlight in to electricity. This is process is known as the photovoltaic effect. Each cell comprises of two dissimilar semiconductors (usually P-doped Silicon and N-doped Silicon) joined together to form a PN junction (as shown in figure a). By P-doping silicon, impurities are added (usually Boron) that cause the material to have a lack of free electrons. This lack of electrons causes the material to have more holes (more positively charged). For N-doping silicon, the added impurities (usually phosphorus) cause the material to have excess free electrons or be more negatively charged.

Figure (a): Current flow in a photovoltaic cell

Image adapted from M. W. Davidson, ‘Solar Cell Operation’

[4]

When photons (light particles) of an appropriate wavelength hit the PN junction they transfer some of their energy to the electrons in the material. This transfer of energy allows for the electrons to move freely to higher energy levels. Electrons in the N-silicon layer are swept across the PN junction in to the P-silicon layer to recombine with holes. Similarly, holes in the P-silicon layer are swept across the PN junction to recombine with electrons in the N-silicon layer. This causes the region around the PN junction to deplete of charge carriers (both electron & holes) forming a depletion region. This depletion region makes it difficult for additional electrons to cross the junction. Once an equilibrium condition is reached, both sides of the junction are separated by a fixed electric field. Connecting of an external circuit to the positive and negative contacts of the cell allows the electrons to flow out of the semiconductor and return to the other layer. This produces a current that is maintained once sunlight is hitting the solar cell.

1.2 Different types of Solar Pv:

Solar pv cells can be made from a range of different semiconductor materials, each exhibiting unique characteristics that influence their use. Characteristics such as crystallinity (how well the atoms are ordered in the material), bandgap (minimum amount of energy needed to make an electron a mobile charge carrier) and absorption coefficient (how far light of a specific wavelength can travel in the material before being absorbed) define a solar cells efficiency and suitability for a specific application. Different manufacturing complexities and costs exist for each material, making some materials more attractive than others.

Monocrystalline Silicon Cells:

These types of cells are made from a single slice of pure silicon, often fabricated using the Czochralski method. This process involves extracting large cylindrical silicon ingots from molten silicon, from which thin wafers may be cut. Monocrytsalline silicon cells have high efficiencies, but are expensive to produce.

Polycrystalline (multicrystalline) Silicon Cells:

Polycrystalline or multicrystalline silicon cells are made from a single slice of silicon, but contain a large number of crystals compared to monocrystalline silicon. Made from square blocks of liquid silicon, different sized crystalline structures form during solidification giving rise to defects. These defects make the solar cell less efficient, but are less expensive to produce.

Thin film (Amorphous Silicon) Cells:

Thin film cells are manufactured by depositing amorphous (non-crystalline) silicon on glass or another substrate material. If the substrate material is flexible then the whole solar pv cell is flexible due to the amorphous nature of the thin film. This type of solar pv cell is the least efficient, but is also the cheapest. They are normally found in low power applications such as digital watches and pocket calculators.

Multi-junction PV cells:

Developed during the 1980’s for space-based applications [5], multi-junction solar pv cells are the most mature of all photovoltaic technology for efficiencies over 30% (see appendix A). By utilizing different semiconductors layers for different parts of the solar spectrum, energy losses due to the thermalization of electron-hole pairs are greatly reduced. This allows for high efficiencies to be achieved when compared to single junction devices. Due to the manufacturing complexity these type of cells remain expensive to produce and are only considered for niche applications such as space satellites where power to weight ratio is important.

Silicon Spheres:

This type of technology uses small silicon spheres surrounded by hexagonal reflectors to form a cell. The sphere consists of a P-doped ball of silicon with an N-doped surface forming a PN junction. Compared to other pv cell technologies only one fifth of the raw silicon material is required [6], which greatly reduces costs. However, due to the curved surface of the spheres a lot of light gets reflected, which reduces their efficiency.

Nano Solar Cells:

Nano solar cells consist of a nanowire with concentric layers of N-doped and P-doped crystalline silicon. These layers form PN junctions similar to conventional solar cells that can be arranged in to large arrays. This technology is currently in its infancy and will probably be useful in niche applications such as nano sensors and robotics. Currently, the highest efficiency reported is 8.4%, but it is expected that efficiencies as high as 65% are possible with nano solar cells [7].

1.3 Market Trends:

Over the last decade the global solar pv market has experienced tremendous growth. Installed pv power capacity has grown from 0.1 giga-watts (GW) in 1992 to 14 GW in 2008 [8]. Figure b shows the global installed capacity for solar pv over the last two decades. During the 1990’s off-grid systems accounted for half of the global solar pv market. Due to the rapid growth of on-grid systems stimulated by various governmental incentive schemes [9], off-grid systems now only constitute less than 10% of the total market (see appendix B).It is unlikely that off-grid solar pv systems will experience a boom similar to that of on-grid systems, but modest growth would be expected. For the developing world off-grid systems could provide power to rural villages and act as an alternative to the unreliable electricity supplies in some urban areas. Unfortunately, for most developing countries affordability still remains a barrier, leading to the opting of cheaper, but more polluting energy sources.

Figure (b): Global installed capacity from 1992 - 2008 Figure (c): Leading countries of installed Solar Pv in 2008

Both images adapted from IEA, ‘International Solar PV Roadmap 2009’

[8]

As of 2008 four countries (Germany, Spain, Japan and US) accounted for over 80% of the global installed pv capacity (see Figure c). The introduction of new policies and support schemes means that some other countries including France, Italy and China are gaining market share. China especially is pursuing an aggressive strategy towards growth in the solar pv market. Currently ranked first in the world in exports for solar pv cells, domestic output expanded from 100 MW in 2005 to 2 GW in 2008 [10]. However, market demand within China remains small with over 95% of solar pv products exported. By the end of 2008, China’s total pv installed capacity was 150 MW [8] with over 40% accounting for off-grid systems for supply of electricity to remote villages not connected to the national grid. Market share of solar pv for industry and communications is increasing with the Chinese government offering subsidies for building-integrated photovoltaics (BIPV). It is predicted that installed pv capacity in China could reach 20 GW by 2020 [11].

1.4 Technology Trends:

In order for solar pv technology to be competitive with fossil fuel power generation in the future, significant investment in effective technology development is required. Achieving higher efficiencies and reducing costs will be key in improving current technologies and developing new ones. It is predicted that a range of technologies (both existing and emerging) will constitute the pv technology landscape for specific requirements of various applications (see Figure d). Crystalline silicon pv technologies are expected to continue as a dominant technology with market share forecasts of 50% by 2020 [12] and efficiencies of up to 23%. Thin film pv technologies are expected to experience rapid growth with advances in device structures, large area deposition techniques and roll-to-roll manufacturing. Increased R&D is required to overcome the industry’s little experience with thin film lifetimes. Emerging technologies such advanced inorganic thin films and organic pv cells have their use in niche applications, but have yet to make any impact on power applications. The use of novel pv concepts such as quantum wires is currently under research and is expected to be a significant technology within the next 10 years. Their market significance will depend if they are combined with existing technologies or used to develop completely new processes.

Figure (d): Photovoltaic technology status and prospects

Image adapted from IEA, ‘International Solar PV Roadmap 2009’

[8]

2.0 Solar energy in Ireland:

The majority of Ireland’s domestic solar energy market comprises of low temperature applications that do not require direct sunlight such as water and space heating. As of 2006, installed solar heating capacity in Ireland was 11 MW, compared to 0.3 MW of installed capacity for solar pv (see appendix C). In contrast to other European countries these numbers are low given Ireland’s high GDP. Countries receiving low solar income would be expected to have low solar capacities, however Denmark has the fourth largest solar thermal capacity of the EU-15 countries and average annual solar irradiation in Ireland (see Figure e) is similar to much of Northern Europe (see appendix D). Although Denmark and Ireland are without the support mechanisms of their European counterparts it can be concluded that government incentives and subsidies are vital in promoting solar energy for market growth.

Figure (e): Average annual solar irradiation (Kwh/m²) in Ireland 2008

Image adapted from PVGIS © European Communities, 2001-2008

2.1 Typical Off grid system:

Off-grid stand alone solar pv systems are systems that operate independent of the national grid. These systems are often used where a grid access may be difficult and expensive to access such as an island or other remote locations. In some cases an off grid system can be cheaper than extending existing power lines. Off-grid systems vary in size depending on setup, but most domestic off-grid systems consist of: a solar panel, charge controller, battery system (storage) and an inverter (see Figure f). When light hits the solar panel a voltage travels to the charge controller. The charge controller is similar to a voltage regulator that regulates the voltage going to the battery. If there is no regulation the battery may suffer damage from over-charging due to the fluctuations in output voltage from the solar panel. The battery is used to store energy to provide electricity at night or on a cloudy day. The electricity generated by the solar pv panel is direct current (DC) therefore an inverter is required to transform it to alternating current (AC). A DC load may be connected before the inverter, but the majority of common household appliances use AC electricity.

Figure (f): Typical Off grid system

Image adapted from Remon Industrial Limited, ‘Off grid solar system’

[13]

2.2 Stand alone applications:

Like off-gird systems stand alone systems are totally self-sufficient with no connection to the grid. In most cases stand alone systems are designed for a specific application and are generally used in the capacity of lower power devices compared to an off-grid system. There are many stand-alone applications for solar pv in Ireland. These include car parking meters, street lighting, roadside information signs, speed cameras, offshore navigation beacons, weather stations, mobile phone transmitters, water pumps and mobile camping.

Car parking meters:

Over the last 5 years town councils across Ireland have chosen solar powered pay & display parking meters to replace older conventional meters. These new meters incorporate a small solar pv panel on top that supplies power to the device. Between 2005 and 2008 Drogheda Borough council installed 106 new Cale Mp 104 parking terminals (see Figure g) as a cost savings initiative in the town of Drogheda. These terminals are powered using solar pv and do not require any external power for operation. In addition to the cheaper running costs, money was also saved during installation since the town’s streets did not have to be dug up for laying of expensive mains power. The terminals also feature wireless connectivity and the ability to monitor the battery voltage level, power consumption and temperature using software. Other towns in Ireland who have installed solar powered pay & display parking meters include Bandon (Cork), Loughrea & Tuam (Galway) and Letterkenny (Donegal).

Figure (g): Car parking meter in Drogheda, Co. Louth

3.0 HOMER Simulation: Solar pv stand alone application in Drogheda, Co Louth

Hybrid Optimisation Model for Electric Renewables (Homer) [14] is a computer simulation software tool developed by the US National Renewable Energy Laboratory (NREL) [15]. It contains models for different renewable energy sources (solar pv, wind, hydro..etc) that allow for the economic and technical evaluation of renewable energy systems based on an input of resources and variables. The software allows for the design of both off-grid and grid connected systems for remote, stand alone and hybrid systems. Using the HOMER software a computer model was designed to simulate a solar pv powered car parking meter located in Drogheda, Co. Louth. Using technical specifications (see appendix E) from the MP104 terminals currently installed in Drogheda (see Figure h) and solar radiation data (see appendix F) from the NASA Atmospheric Science Data Center (ASDC) a complete model was simulated.

Figure (h): Map of Drogheda, Co. Louth

3.1 System Design for solar powered car parking meter:

Before designing and simulating a solar pv system values for load and battery storage must be determined. The MP104 terminal has options to be configured for power by solar pv (with battery), mains operated or a hybrid of both. For this simulation solar pv (with battery) was used as the power source. The Mp104 can have two different power rated solar panels depending on geographical location. For northern climates with poor daylight a 25 W panel is used, in Ireland a 14W panel is sufficient to charge a 12V battery for storage. Paid parking in Drogheda is in place 6 days per week (Monday to Saturday), but the machines are powered 24-hours per day for a 7 day week. The machines are constantly in sleep mode and are only woken up when a customer pushes a coin into the coin slot. The device stays powered on for approximately 10 seconds after the purchase of a ticket and then goes back in to sleep mode to conserve power. Using this information load and battery calculations can be carried out.

Determine load:

In order to calculate the load the following formula is used:

Power / Voltage = Current

14 W / 12 V = 1.16 A

Determine required battery storage:

For the device to operate at the calculated load (1.16 A) and allowing for seven 24-hour days of battery storage, the capacity of the battery required can be calculated as follows:

Current x Hours per day x number of days = capacity of battery (Ah)

= 1.16 A x 24-hours x 7 days = 194.88 Ah

Expressed in Watt-hours (volts x amp-hours):

= 12 V x 194.88 Ah = 2338.56

= 2.34kWh

The HOMER software features a great deal of options for specifying the type of battery to be used in simulation. These include the battery’s voltage, capacity, lifetime and all associated cost curves. HOMER also contains built-in models for commercially available batteries that can be used in simulation. The battery found to be the best match based on the designs requirements was a Vision 6FM200D (see appendix G). This 12V battery has a capacity of 200Ah (2.4KWh), which is sufficient for this system. There are no AC powered components in the MP104 therefore no inverter is required. Figure I illustrates the system design for the parking meter to be simulated.

Figure (i): System design for parking meter

Solar pv panel:

After defining the specifications of the battery, parameters for the solar pv panel were assigned. HOMER provides a cost table where different pv array sizes can be entered in order for the software to determine the optimal system. Currently the cost of crystalline solar pv per watt is approximately €1.85 per watt [16]. For a 14W panel this equates to a cost of €25.90 ($20). The lifetime of the panel is 20 years and the replacement cost (cost to replace panel in 20 years) is estimated to be one quarter of the current cost. The derating factor is a scaling factor for the power output to account for reduced output in real-world operating conditions, for the simulation this was set to 80%. The slope is the angle at which the panel is mounted relative to the horizontal and the azimuth is the direction which the panel is pointing. The ground reflectance is the fraction of solar radiation incident on the ground that is reflected. By default HOMER sets ground reflectance to 20%. Since the system is based on a stationary object (car park meter) no tracking system was used.

Figure (j): Pv inputs for HOMER

Primary Load:

The primary load is the load that the system must meet immediately in order to avoid an unmet load. An unmet load is the electrical load that the power system is unable to provide. As previously stated the terminal is constantly in sleep mode and is only woken up when a customer pushes a coin into the coin slot. During sleep mode the machine draws a current of approximately 0.5 A. In order for HOMER to conduct the simulation hourly values for the load must be entered. The hourly load was calculated as follows:

Determine primary load:

0.5 A x 12 V DC = 6 W

Figure (k): Primary Load Inputs for HOMER

Determine solar radiation for location:

HOMER requires the user to specify the latitude and the amount of solar radiation available to the solar pv panel throughout the year. The software uses this data to calculate the output of the PV array in each time step. The latitude specifies the location on the Earth's surface. This is an important variable in solar calculations as the software uses it to calculate radiation values from clearness indices, and to calculate the radiation incident on a tilted surface. The clearness index is a measure of the clearness of the atmosphere. It is the fraction of the solar radiation that is transmitted through the atmosphere to strike the surface of the Earth. Drogheda in Co. Louth is located at a latitude of 53° 43’ and a longitude of 6° 21’. Solar radiation data as monthly averages was imported (see Figure j) from the NASA Atmospheric Science Data Center (ASDC) via the HOMER software.

Figure (l): Importing solar radiation data for Drogheda, Co. Louth

3.2 HOMER Simulation results:

Results from the HOMER simulation give a clear indication to how well the car parking meter operates for the given test conditions. As shown (see Figure m) the power delivered to the DC load corresponds to the incident solar that reaches the solar panel. The battery is charged to 100% at the start of the simulation, but quickly drops to 40% (see Figure n) for the remaining test time. On closer inspection (see appendix H) it can be noted that the operating DC capacity falls short of the required DC capacity for certain days in a given month. There are also some peaks in the operating DC capacity significantly larger than the required DC capacity. At first it was thought that some sort of seasonal fluctuations might cause the abnormal behaviour of the operating DC capacity, however this speculation was invalidated upon viewing of the data map (Dmap) (see Figure o). The Dmap displays the time of day on one axis and day of the year on the other, each time step of the year is represented by a rectangle which is colored according to the data value for that hour. As one would expect the solar pv power is strongest around mid-day and is more condensed during the sunnier months (May to August). By increasing the rating of the solar pv panel from 14 W to 25 W (see appendix I) a much more reliable system is produced. The operating DC capacity satisfies the required DC capacity much more compared to the system using the 14 W solar panel.

Figure (m): Incident Solar and DC load served

Figure (n): Battery Bank state of charge

Figure (o): DMap of PV power

4.0 Conclusion:

An accurate model of a stand-alone system utilizing solar pv as the main source of energy was developed. It was shown that such a system is technically and economically viable in Ireland, but choice in the correctly rated components (pv panel, battery..etc) is crucial in meeting the systems demand. Improvements where noted by increasing the power rating of the solar pv panel used. In order to perform accurate cost analysis of the system simulated further information (that was not obtainable) is required. Solar pv will play an important part in Ireland’s renewable future, especially in the role of stand-alone applications and off-grid systems. For stand-alone systems operating and installation costs can be dramatically reduced. In addition, the low associated maintenance costs and the falling price per watt makes solar pv an attractive option. For off-grid solar pv systems, unfortunately no incentive schemes currently exist in Ireland. Off-grid systems in rural areas are rarely supported, despite the fact that they may offer the only cost-efficient solution. Dedicated and realistic strategies to support off-grid pv systems need to be established.

References:

1. F.O. Rourke, F. Boyle & A. Reynolds. Renewable energy resources and technologies applicable to Ireland. Renewable and Sustainable Energy Reviews 2009. Vol: 13, Issue: 8, pp 1975 – 1984.

2. M. Howley, Dr. B. O Gallachoir & E. Dennehy. Energy in Ireland Key Statistics: 2008. Sustainable Energy Ireland. 2008

3. X. Gong & M. Kulkarni. Design optimization of a large scale rooftop photovoltaic system. Solar Energy 2005. Vol: 78, Issue 3, pp 362-374

4. M. W. Davidson. Solar cell operation. National High Magnetic Field Laboratory 2006. URL: http://micro.magnet.fsu.edu/primer/java/solarcell/

5. H. S. Wong, P. Peumans & Y. Nishi. Lateral Nanoconcentrator Nanowire Multijunction Photovoltaic Cells. Materials Science and Engineering, Stanford University.2006.

URL: http://gcep.stanford.edu/research/factsheets/lateral_nano_photovoltaiccells.html

6. D. G. Rowe. Focusing Light on Silicon Beads. MIT Technology Review, MIT.2007. URL: http://www.technologyreview.com/energy/19696/page1/

7. F. Shand. Improving the efficiency of nanowire solar cells. NANO Magazine UK. 2010.

URL: http://www.nanomagazine.co.uk/index.php?option=com_content&view=article&id=755:improving-the-efficiency-of-nanowire-solar-cells&catid=38:nano-news&Itemid=159

8. International Energy Agency (IEA). Technology Roadmap - Solar Photovoltaic Energy. France. 2010. URL: http://www.iea.org/

9. T. Couture & Y. Gagon. An analysis of feed-in tariff renumeration models: Implications for renewable energy investment.

Energy Policy 2010. Vol: 38, Issue: 2, pp 955 – 965.

10. S. Wang. China Solar Pv Report. 2008.

11. V. Mulky. Solar Pv Industry 2010: Contemporary scenario and emerging trends. 2010. India Semiconductor Association (ISA)

12. International Energy Agency (IEA). Energy Technology Perspectives 2008. France. 2008. p 370.

13. Remon Industrial Limited, ‘Off grid solar system’ URL: http://www.rmsolarpanel.com/html/51/2010-12-21/content-61.html

14. Hybrid Optimisation Model for Electric Renewables (Homer) Software. URL: http://www.homerenergy.com/

15. US National Renewable Energy Laboratory (NREL) URL: https://analysis.nrel.gov/homer/

16. Solar Buzz Module Pricing. December 2011 Update. URL: http://solarbuzz.com/facts-and-figures/retail-price-environment/module-prices