Compared to the other sources of renewable energy, photovoltaic electricity was a late starter. Although the photoelectric effect had been studied since the late 19th Century, it wasn’t until 1954, in the Bell Labs in the US, that the first functional solar cell was produced.
At first very expensive, the technology only started to become commercialized in the late 1950s–mainly for satellites that needed low levels of power for long periods of time. In 1958, Vanguard I & II, Explorer III, and Sputnik-3 satellites were launched–each powered by solar PV arrays. In 1959, Explorer VI was launched with an array of 9600 cells.
Terrestrial applications were few and far between because photovoltaic cells were expensive. Communication systems in isolated areas was an early niche market. In 1962, Bell Labs launched Telstar–the first telecommunication satellite. A year later, Japan installed a 212-Watt PV array on a lighthouse—the world’s largest array at that time. 
Since that time the pace of photovoltaic power deployment has been fast and furious. In 2017, 98 gigawatts (that’s 98,000 megawatts) of solar photovoltaic capacity was added worldwide—equivalent to the installation of more than 40,000 solar panels every hour . By the end of that year, global solar PV capacity totalled just over 400 GW. The chart shows the exponential increase in PV installed capacity worldwide from 2006 to 2017 .
China once again leads the way, followed by the US, Japan, India and Turkey—together these five countries accounted for about 84% of additional capacity. While China dominates both the use and the manufacturing of solar PV, emerging markets on all continents have begun to contribute significantly to global growth. By the end of 2017, 29 countries had more than 1 GW installed. Per capita, the leaders are Germany, Japan, Italy, Belgium and Australia.
In the US, solar PV was the country’s leading source of new generating capacity in 2017. More than 10 GW of capacity was brought online for a total of 51 GW. California led the field for capacity added (5.2 GW), followed by North Carolina (1 GW) and Florida (0.4 GW). 
Photovoltaic technology and applications exist at many levels: from residential kilowatt-scale systems up to centralized gigawatt-scale installations operated by utilities managing transmission and distribution networks providing electricity to millions of customers. The degree of penetration of residential systems depends strongly on the incentives provided by governments, and other factors such as the cost of electricity from the utility, and the cost and availability of financing. Where Feed in Tariffs (FITs) are a financially attractive option, households can be quick to take advantage of the policy. However, in many countries, FITs have been scaled back, and although net metering is available in many jurisdictions, this policy has not been as influential as feed in tariffs in inducing households to install residential photovoltaic systems.
In Germany, the solar-plus-storage market is growing rapidly as consumers shift from FITs to self-consumption. The share of newly installed residential systems paired with storage rose 14% in 2014 to more than 50% in 2016, when that country represented about 80% of Europe’s home energy storage market. Australia’s market has also been predominantly residential. By late 2017 almost 1.8 million rooftop solar PV installations (residential and commercial) were operating in that country.
In addition to Australia, Germany and Japan, interest in solar-plus-storage is picking up in other developed countries (e.g. France, Italy, and the UK) for on- and off-grid systems where incentives are persuasive and the option is financially attractive. 
Although demand is rising rapidly for off-grid solar PV in Africa and other regions, grid-connected systems continue to account for the majority of existing and new installations. In terms of the fraction of added capacity, decentralized rooftop grid-connected systems have been declining–particularly with the transition from FITs and net metering to self-consumption—although it ticked up slightly in 2017.
Centralized large scale projects by contrast have comprised a rising share of annual installations –particularly in emerging markets, and now represent the majority of annual installations .
Floating PV installations are also growing in number and scale. Since 2015, more than 100 plants have started up, floating on hydropower reservoirs, industrial water sites, aquaculture ponds, and other areas of water. The benefits of floating PV modules include increased efficiency (because the modules are cooler) and reduced evaporation from the reservoirs. Japan leads in terms of the number of installations due in part to the country’s FIT policy combined with limited roof and ground space. Other countries with projects include China, India, South Korea and Brazil.
Megawatt-scale photovoltaic power plants are getting larger—particularly where there is plenty of both sunshine and available land.
The table on the right lists the world’s largest solar photovoltaic power plants with a capacity of 300 MW or greater. 
Perhaps sensing the end of the road for their free-flowing oil, Saudi Arabia is diversifying its energy resources and exploiting the solar energy it has in abundance. The country’s first solar project is planned to be a 300 MW installation at Sakaka in north-western Saudi Arabia. The contract was awarded in 2018 to the Saudi energy group, ACWA, whose bid came in at under US 3 ¢/kWh. The 300 MW plant is expected to involve a total private sector capital investment of about $300 million and create 400 jobs .
Clearly, Saudi Arabia is thinking big. In early 2018 it was reported that Saudi Arabia and Japan’s SoftBank Group Corp. had signed a memorandum of understanding to build a 200 Gigawatt (that’s 200,000 MWp) photovoltaic power plant.
If its built, the PV power plant would almost triple Saudi Arabia’s electricity generation capacity, which stood at 77 GW in 2016. About two thirds of that amount is generated by natural gas, with the rest coming from oil. The gigantic PV project, which includes both power generation and module and equipment manufacturing, will create as many as 100,000 jobs, and is expected to shave $40 billion off conventional power costs. The development will reach its maximum capacity by 2030 and cost close to $1 billion per Gigawatt.
How’s it work?
A typical silicon photovoltaic (PV) cell is composed of a layer of phosphorus-doped (N-type) silicon on top of a thicker layer of boron-doped (P-type) silicon. An electrical field is created near the top surface of the cell where these two materials are in contact—called the P-N junction. When sunlight strikes the surface of the PV cell this electrical field generates light-stimulated electrons, resulting in a flow of current when the solar cell is connected to an electrical load.
A silicon PV cell produces only about 0.5 – 0.6 volts DC under open-circuit, no-load conditions. The current and power of a PV cell depends on its efficiency and surface area and is proportional to the intensity of sunlight striking the surface of the cell. Under peak sunlight conditions a typical commercial PV cell with a surface area of 160 cm2 will produce about 2 Watts of power. 
Photovoltaic cells are connected in series and parallel circuits to produce higher voltages, currents and power levels. The PV cells are assembled into modules (solar panels), which for large installations, are then interconnected into an array—which may consist of thousands of individual panels.
The performance of PV modules and arrays are generally rated according to their maximum DC power output under standard conditions—which approximates the maximum insolation possible on the surface of the Earth. PV panels are rated according to the maximum power they will produce under these conditions. For example, a PV panel rated as being 200 peak watts (written as 200 Wp) will produce 200 watts of direct current (DC) power under the maximum level of sunshine possible. Since a PV panel in a fixed position will not receive this level of insolation for more than an hour or so a day (if it is perfectly oriented perpendicular to the sun’s rays), the average power output of a PV array is a lot less than its rated power.
The capacity factor of a PV system is obviously low since for at least half the time the module is not producing any power (at night), and even during the day it reaches its peak power only for a few hours around midday. So how much energy does a solar panel actually produce?
Although in most places the sun shines for about 12 hours a day, only for about 4 or 5 hours will the PV panel be producing at close to its rated output. For instance, a 100-peak-watt panel (100 Wp) will produce about 400 Watthours of electricity when installed at a site with average insolation. In a sunnier location, you should get about 500 Wh of energy from the same system. This is a just a rule-of-thumb—but it enables you to make a quick estimate of the energy produced by an array of PV panels.
Let’s say you are thinking of installing a 6000 Wp (6 kilowatts peak) PV system on the roof of your house in Florida. Florida is pretty sunny, so let’s reckon that we get on average 4.5 hours x 6 kW = 27 kWh of electricity a day from the modules. That’s 9,855 kWh of electricity a year—which gives you a first cut at roughly the amount of electricity you might get out of your system. It’s a ball park figure. But it might be good enough for you to decide whether or not to go ahead with a more detailed evaluation of the economics of the proposal.
For an accurate evaluation of the output from a PV system we need to run a detailed simulation of the insolation at the proposed site over the course of a year. These data need to be modelled together with the energy losses associated with the equipment necessary for the operation of the system—particularly the inverters which convert direct current (DC) electricity to alternating current (AC), and the batteries (if these are included). One of the best (and free) PV evaluation tools is the Photovoltaic Geographical Information System (PVGIS) developed and supported by the Joint Research Centre of the European Commission. It’s available at this site: //re.jrc.ec.europa.eu/pvg_tools/en/tools.html.
Running the simulation for a site in Miami with a 6kWp photovoltaic installation, the model produces the output shown below.
The model takes into account all the system losses and calculates the monthly energy output from the PV panels–assumed to be optimally set up at a slope of 26 degrees and an azimuth of -7 degrees (oriented slightly to the south east).
The monthly energy output varies between about 700 kWh in September to just over 900 kWh in March. The total energy output for the year is estimated as 9400 kWh—about 5 % less than our quick calculation based on the rule-of-thumb method.
PVGIS can be used to model the performance of much larger photovoltaic systems—even utility-scale PV power plants can be simulated by the software, which will also calculate the levelized cost of the electricity produced by the power plant.
Take the example of a 30 MWp photovoltaic power plant to be installed in Djibouti in east Africa. The capital cost of the system is estimated at about $60 million. If we run this simulation for a site at Grand Bara in Djibouti, we get the results shown in the chart below.
Output data for a 30 MWp photovoltaic system in Djibouti
If the interest on capital is set at 5% p.a. and assuming a 30 year lifetime, the levelized cost of energy is calculated as 9¢/kWh . That’s pretty good for Djibouti—where electricity is hugely expensive.
One of the most useful characteristics of photovoltaic energy is the extraordinary range of the services that PV energy can provide—everything from a tiny cell producing a fraction of a watt for a pocket calculator, to arrays of thousands of PV modules generating megawatts of power. In between these extremes is an application that has the potential to completely change the way electricity is provided and used by residences, communities, commerce, and industry.
In 2017, the US solar market installed 10.6 GW of PV capacity. This was 30 % less than the record-breaking year before, but in line with the upward trend in PV installations that has been rising exponentially for the last decade. About 60% of this capacity was for utility-scale PV power plants; the other 40 % was for distributed solar. This segment covers both residential rooftop systems and non-residential systems—the latter fraction including community-based PV systems and PV systems installed on commercial buildings. 
The installation of thousands of small solar PV systems on residences and commercial buildings has largely been driven by governments’ determination to reduce emissions of greenhouse gases on both sides of the supply-demand equation. So in western Europe, megawatt-scale PV and wind power is replacing coal-fired and nuclear power plants, while at the same time strong financial incentives have been provided in order to induce households, businesses and communities to install kilowatt-scale PV systems to reduce the aggregate demand for electricity from the centralized distribution system.
In Europe, this policy was pioneered by Germany which introduced Feed-in-Tariffs (FITs) to encourage the use of new energy technologies including photovoltaic energy in 2001. FITs provide a revenue to any renewable energy installation, including residential solar, that feeds electricity into the grid. At first set close to 50 cents/kWh, a level than induced hundreds of thousands of households in Germany to install rooftop PV systems, the tariff has been progressively reduced–to around 15 cents/kWh, which is where it stood in 2018.
In Ontario, Canada, the FIT program for PV electricity started in 2009 and offered extraordinarily high FIT rates to rooftop and ground-mounted PV installations. Rates varied between 44.3 and 80.2 ¢/kWh  with the upper rate applicable for rooftop installations less than 10kW. In addition, homeowners enjoyed the exceptional security of a 20-year contract !  Perhaps unsurprisingly, the cost of this generous program was excessive, and Ontario’s FIT program was eventually discontinued.
The inducements and incentives for homeowners and communities to install small-scale photovoltaic systems on rooftops or adjacent spaces has been extremely successful where it has been instituted and supported by national and regional governments. There are two principal incentive mechanisms: net energy metering and feed-in-tariffs–although additional financial incentives such as rebates, discounts, tax breaks and other benefits are often included in the programs.
Net energy metering is an arrangement where the excess electricity produced by a roof-mounted (or adjacent) PV system (the amount of energy not required by the building itself), is fed back into the grid. In effect, the electricity meter runs backwards and the utility’s client—the home or business owner—is only charged for the difference between the amount of electricity he or she consumes and the amount injected into the grid. How the calculation is figured out varies: different power companies have different arrangements, and there are generally limits on how much electricity can be fed back into the grid.
Feed in-tariffs, as discussed briefly above, require the residence or building to have two electricity meters—one to measure the electricity supplied by the grid (at night for example), and one to measure the electricity produced by the PV system and fed to the grid during the day. The home or business owner will have to deal with two electricity tariffs: one for incoming electricity from the grid, and the other for electricity injected into the grid by the PV system. Both of these tariffs may vary according to the time of day.
Power purchase agreements (PPA) are also common. This is a contractual agreement between a utility and the owner of a solar photovoltaic system, where the utility agrees to purchase electricity from the PV system for a fixed price per kWh over a specified period of time.
In the US at the end of 2017, of the nine states that generated more than 1000 MW of solar PV electricity, three have more than half of that power output from distributed PV systems. All three states are in the NE: New Jersey, Massachusetts, and New York—where an impressive 87% of total solar electricity generation is from distributed sources. Yet the north-east region of the US has only modest solar resources. They key to the growth of distributed energy systems in this region is policy: in recent years the three states have expanded policies such as net metering and power purchase agreements that are favourable to distributed generation. It’s a reminder that solar energy can be economically viable even in regions with only average levels of insolation.
In California, which has a lot better sunshine and much more utility-scale PV power, distributed solar accounts for about 40% of installed PV capacity. This percentage will soon ramp up considerably: in 2018, the state passed a landmark solar homes rule that will mandate distributed solar on all new home construction starting in 2020.
Community shared solar
As distributed solar energy becomes more widespread, people are exploring new ways of using photovoltaic electricity. Not everyone is able to install panels on their roof due to unsuitable or insufficient roof space, living in a multi-unit condo building, or simply renting. However, there are some alternative business models are that are developing—like community solar and shared solar.
These business arrangements make it possible for people to invest in solar energy together. Shared solar falls under the community solar rubric allowing multiple participants to benefit directly from the energy produced by a single solar photovoltaic array. Shared solar participants typically benefit by owning or leasing a portion of a system or by purchasing kilowatt-hour blocks of renewable energy generation. The figure below shows schematically the three most common arrangements.
By aggregating customer demand, shared solar programs can reduce the financial and technical barriers to investing in residential solar energy. Instead of acting alone to purchase PV panels and hiring professionals to complete individual site assessments, shared solar programs divide these costs among all of the participants. Investments are even safe for those who may eventually move—their share of solar can be transferred to a new home within the same utility service area or sold to someone else.
Most onsite solar energy installations use net metering to account for the value of the electricity produced when the PV system produces more energy that is needed by the participants. Net metering allows customers to be credited for this excess electricity in the grid, usually in the form of kWh credits during a given period. The electricity meter runs backwards, and customers purchase fewer units of electricity from the utility, so the electricity produced from the PV system is effectively valued at the retail price of power.
Onsite energy storage
As the cost of batteries continues to fall, there is increasing interest in adding energy storage to distributed energy systems–which for solar photovoltaic power means residential and community shared solar including businesses and commerce.
One reason for this interest is the notable increase in extreme weather events that have impacted north America over the last few years—especially in 2017. Severe weather is now the leading cause of power outages in the region.
For a solar system to provide electricity during a utility power outage, it must be designed to function as a stand-alone system that can isolate itself from the grid, continue to generate power and provide energy to the building, and also store excess electricity for later use.
For safety reasons, operating standards require that grid-connected solar PV systems automatically disconnect from the grid during a power outage. This is because most conventional rooftop PV systems are not designed to function as both a grid-connected and a stand-alone system. Instead they disconnect from the grid and completely cease power production during an outage. The figure shows the basic components of a solar PV system with energy storage.
Installing solar PV technology in conjunction with energy storage allows a solar PV system to provide power when the grid is down—in effect, it functions as a stand-alone system.
Batteries are the most commonly used storage technology for small distributed solar PV applications, although other types of energy storage systems (ESS) may be used for larger utility-scale systems. Batteries are linked to the PV panels though an inverter which automatically selects between charging the batteries, providing electricity to the onsite load, or feeding electricity into the grid. So it’s a more sophisticated piece of power conditioning equipment than the simple inverter used in a PV system that has no energy storage capacity. The inverter monitors the onsite load, the grid status, the state of charge of the batteries, and the power being generated by the PV system.
The principal barrier to the deployment of energy storage in distributed energy systems has been the cost of the batteries, but costs have been falling for the last decade. Lower battery prices, increased demand for backup power, and uncertainty surrounding the future cost of electricity from the grid are all factors stimulating interest in distributed energy systems with storage.
Residential PV systems with storage can generate benefits for grid operators. Storage can add value through:
- Demand side management to shave peaks in the load on the utility system
- Improved power quality by smoothing the variable output of a PV system
- Providing power to critical facility during an outage
- An increased ability to integrate higher levels of distributed energy systems into the grid system
- Ancillary grid services such as voltage control 
In order to provide these functions and services distributed energy systems with storage require very smart controls and meters.
The ability of a distributed energy system to operate as a stand-alone system in the event of a major outage means the building has considerable value to communities and local agencies in the aftermath of extreme weather.
For instance, Florida’s SunSmart Schools and Emergency Shelters Program has installed 115 solar PV systems with storage at Florida’s schools to create emergency shelters . The basic system in Florida schools consists of:
- 10 kW photovoltaic system on a ground mounted array (about 100 m2)
- 48 kWh battery backup energy storage
- 3-phase building electricity
- Utility grid-connected
- Net metering power
- Data monitoring
Using schools as emergency shelters is a strategy that should be adopted by island governments in the Caribbean and the Pacific. All islands have schools and kitting them out with a solar PV system with energy storage is relatively inexpensive and a potential life-saver during a major hurricane. Schools can also be used to store essential supplies (water and food) for local communities in the aftermath of hurricanes and cyclones.
Community based PV systems can also be designed and built with energy storage. With the ability to run as a stand-alone system in the event of an outage–‘islanded’ from the grid, the energy storage capacity enables the community to have electricity while other residences and buildings are without power.
The cost of PV electricity
Rapid declines in installed costs and increased efficiency have dramatically improved the economic competitiveness of photovoltaic power. The global weighted average LCOE of utility-scale PV plants is estimated to have fallen by over 70% between 2010 and 2017—from about $0.36 to $0.10 / kWh. LCOE costs vary by country. The Italian market experienced the largest LCOE reduction driven both by reductions in the cost of modules and a fall in the balance of systems (BOS) costs. In the US, costs are higher, but excellent solar resources mean than that the LCOE of utility-scale projects in the US is not significantly higher than in other markets. The LCOE of residential systems has also fallen at a rapid pace. In Germany, costs fell from $0.55/kWh in 2007 to $0.15/kWh in 2017. Data from India, China, Australia and Spain—which all have higher levels of insolation than Germany–show that lower LCOE costs can be achieved even if installed costs are sometimes higher. In these countries, the LCOE fell to between $0.08 and $0.12/kWh at the beginning of 2017. 
In the US, the SunShot Program is funding research and development intended to bring energy costs from PV systems down to as low as 3 ¢/kWh by 2030. The chart outlines the goals of the program for residential., commercial and utility-scale PV systems. .
Progress towards these targets has been impressive. In September 2017, the US DOE announced that the SunShot program had met the 2020 utility-scale target of 6¢/kWh three years earlier than expected—so a new target of 3¢/kWh was set for 2030.
Although the cost of photovoltaic systems and batteries continue to decline, they must eventually level out–they can’t drop to zero. But what is clear is that solar photovoltaic energy is poised to become the principal power technology of the 21st Century. It may share the stage with windpower–which has the significant advantage that it can be situated offshore, but the huge flexibility and range of solar energy in terms of its scale and applicability means that the number and size of PV installations will continue to increase both in industrial and emerging economies.
Small-scale photovoltaic systems are also the key to providing electricity to the millions of households in Africa and Asia that lack power.
Check out these sources for more information:
 See: The history of solar, US Department of Energy/Energy efficiency and renewable energy. At: //www1.eere.energy.gov/solar/pdfs/solar_timeline.pdf
 REN21. Renewables 2018 Global Status Report. Op.cit.
 See the Wikipedia article at: //en.wikipedia.org/wiki/List_of_photovoltaic_power_stations#World’s_largest_photovoltaic_power_stations
 See: ACWA wins Saudi Arabia’s 300 MW solar tender at //www.pv-magazine.com/2018/02/06/acwa-wins-saudi-arabias-300-mw-solar-tender/ Also: From oil to solar: Saudi Arabia plots a shift to renewables. At //www.nytimes.com/2018/02/05/business/energy-environment/saudi-arabia-solar-renewables.html
 See : Saudis, SoftBank Plan World’s Largest Solar Project. Accessed at:
 See: Florida Solar Energy Center website: //www.fsec.ucf.edu/en/consumer/solar_electricity/basics/index.htm
 This is the result of a simulation run on the PVGIS website with the input data as shown in the text. What is surprising is the optimal azimuth—which is angled significantly to the south-east–but the program takes into account the altitude of the physical horizon. The levelized cost of electricity is estimated as 9¢/kWh.
 See: US solar market insight, 2017 year in review. Available at: //www.greentechmedia.com/research/repoirt/us-solar-market-insight-2017-year-in-review#gs.zct2qTU
 See: Solar electricity handbook 2017 edition. At: //solarelectricityhandbook.com/canada-feed-in-tariff.html
 See: The state(s) of distributed solar – where are the biggest gains?. Accessed at: //www.renewablenergyworld.com/ugc/articles/2018/05/29/the-states-of-distributed-solar–2017-upgate.html
 Community and shared solar. Department of Energy, Office of energy efficiency & renewable energy. Available at: //www.energy.gov/eere/solar/community-and-shared-solar
 Community and shared solar. Ibid
 See: A guide to community shared solar: Utility, private, and nonprofit project development. Available at : //www.nrel.gov/docs/fy12osti/54570.pdf
 The diagram is from the Florida Solar Energy Center webinar notes available at : //www.cleanegroup.org/wp-content/uploads/ESTAP-webinar-slides-10.24.17.pdf
 See: Distributed solar PV for electricity system resiliency. National Renewable Energy Laboratory. Available at: //www.nrel.gov/docs/fy15osti/62631.pdf
 See the FSEC information at : //www.cleanegroup.org/wp-content/uploads/ESTAP-webinar-slides-10.24.17.pdf
 Renewable power generation costs, IRENA 2017. Op. cit
 See the Sunshot Initiative website at: //www.energy.goveere/solar/sunshot-initiative