Renewable Energy and Electricity
(Updated September 2016)
- There is widespread popular support for using renewable energy, particularly solar and wind energy, which provide electricity without giving rise to any carbon dioxide emissions.
- Harnessing these for electricity depends on the cost and efficiency of the technology, which is constantly improving, thus reducing costs per peak kilowatt, and per kWh.
- Utilising electricity from solar and wind in a grid becomes problematical at high levels for complex but now well-demonstrated reasons. Supply does not correspond with demand.
- Back-up generating capacity is required due to the intermittent nature of solar and wind, but at high levels the economics of this are compromised.
- Policy settings to support renewables are generally required to confer priority in grid systems and also subsidise them, and some 50 countries have these provisions.
- Utilising solar and wind-generated electricity in a stand-alone system requires corresponding battery or other storage capacity.
- The possibility of large-scale use of hydrogen in the future as a transport fuel increases the potential for both renewables and base-load electricity supply.
Technology to utilise the forces of nature for doing work to supply human needs is as old as the first sailing ship. But attention swung away from renewable sources as the industrial revolution progressed on the basis of the concentrated energy locked up in fossil fuels. This was compounded by the increasing use of reticulated electricity based on fossil fuels and the importance of portable high-density energy sources for transport – the era of oil.
As electricity demand escalated, with supply depending largely on fossil fuels plus some hydro power and then nuclear energy, concerns arose about carbon dioxide emissions contributing to possible global warming. Attention again turned to the huge sources of energy surging around us in nature – sun, wind, and seas in particular. There was never any doubt about the magnitude of these, the challenge was always in harnessing them so as to meet demand.
Today we are well advanced in meeting that challenge, while also testing the practical limits of doing so. Wind turbines have developed greatly in recent decades, solar photovoltaic technology is much more efficient, and there are improved prospects of harnessing tides and waves. Solar thermal technologies in particular (with some heat storage) have great potential in sunny climates. With government encouragement to utilise wind and solar technologies, their costs have come down and are now in the same league as the increased costs of fossil fuel technologies due to likely carbon emission charges on electricity generation from them.
However, the variability of wind and solar power does not correspond with most demand, and as substantial capacity has been built in several countries in response to government incentives, occasional massive output from these sources creates major problems in maintaining the reliability and economic viability of the whole system.
Demand for clean energy
There is a fundamental attractiveness about harnessing such forces in an age which is very conscious of the environmental effects of burning fossil fuels, and where sustainability is an ethical norm. So today the focus is on both adequacy of energy supply long-term and also the environmental implications of particular sources. In that regard the near certainty of costs being imposed on carbon dioxide emissions in developed countries at least has profoundly changed the economic outlook of clean energy sources.
A market-determined carbon price will create incentives for energy sources that are cleaner than current fossil fuel sources without distinguishing among different technologies. This puts the onus on the generating utility to employ technologies which efficiently supply power to the consumer at a competitive price. Wind, solar and nuclear are the main contenders.
Sun, wind, waves, rivers, tides and the heat from radioactive decay in the earth's mantle as well as biomass are all abundant and ongoing, hence the term "renewables". Only one, the power of falling water in rivers, has been significantly tapped for electricity for many years, though utilization of wind is increasing rapidly and it is now acknowledged as a mainstream energy source. Solar energy's main human application has been in agriculture and forestry, via photosynthesis, and increasingly it is harnessed for heat. Until recently electricity has been a niche application for solar. Biomass (eg sugar cane residue) is burned where it can be utilised, but there are serious questions regarding wider usage. The others are little used as yet.
Turning to the use of abundant renewable energy sources other than large-scale hydro for electricity, there are challenges in actually harnessing them. Apart from solar photovoltaic (PV) systems which produce electricity directly, the question is how to make them turn dynamos to generate the electricity. If it is heat which is harnessed, this is via a steam generating system.
If the fundamental opportunity of these renewables is their abundance and relatively widespread occurrence, the fundamental challenge, especially for electricity supply, is applying them to meet demand given their variable and diffuse nature*. This means either that there must be reliable duplicate sources of electricity beyond the normal system reserve, or some means of large-scale electricity storage (see later section).
* The main exception is geothermal, which is not widely accessible.
Policies which favour renewables over other sources may also be required. Such policies, now in place in about 50 countries, include priority dispatch for electricity from renewable sources and special feed-in tariffs, quota obligations and energy tax exemptions.
In 2015 over 140 countries submitted to the UN’s Framework Convention on Climate Change (UNFCCC) secretariat their Intended Nationally Determined Contributions (INDCs) to combat climate change. Together, these would lead to an 8% per capita reduction in CO2 emissions by 2025 and 9% by 2030. The role of India and China INDCs is noteworthy here. Regarding solar capacity, India has pledged 246 GWe and China 352 GWe by 2030 on top of present world 178 GWe. Regarding wind, China has pledged 345 GWe and India 78 GWe capacity by 2030 on top of today’s 370 GWe world capacity.
The prospects, opportunities and challenges for renewables are discussed below in this context.
This load curve diagram shows that much of the electricity demand is in fact for continuous 24/7 supply (base-load), while some is for a lesser amount of predictable supply for about three quarters of the day, and less still for variable peak demand up to half of the time; Some of the overnight demand is for domestic hot water systems on cheap tariff. With overnight charging of electric vehicles it is easy to see how the base-load proportion would grow, increasing the scope for nuclear and other plants which produce it. Source: Vencorp
Most electricity demand is for continuous, reliable supply that has traditionally been provided by base-load electricity generation. Some is for shorter-term (eg peak-load) requirements on a broadly predictable basis. Hence if renewable sources are linked to a grid, the question of back-up capacity arises, for a stand-alone system energy storage is the main issue. Apart from pumped-storage hydro systems (see later section), no such means exist at present on any large scale.
However, a distinct advantage of solar and to some extent other renewable systems is that they are distributed and may be near the points of demand, thereby reducing power transmission losses if traditional generating plants are distant. Of course, this same feature sometimes counts against wind in that the best sites for harnessing it are sometimes remote from population, and the main back-up for lack of wind in one place is wind blowing hard in another, hence requiring a wide network with flexible operation.
In Europe, at the end of 2014, the five largest electricity markets (UK, Germany, France, Italy and Spain) had 97.5 GWe of installed wind capacity and 77.6 GWe of installed solar capacity. In the first half of 2015 this produced 107.6 TWh from wind and 60.1 TWh from solar, giving (based on these figures) 25% capacity factor for wind and 17.7% for solar (20.8% in July).
Rivers and hydro electricity
Hydro-electric power, using the potential energy of rivers, is by far the best-established means of electricity generation from renewable sources. It supplies over 16% of world electricity (99% in Norway, 58% in Canada, 55% in Switzerland, 45% in Sweden, 7% in USA, 6% in Australia) from over 1060 GWe installed capacity (2015). Half of this is in five nations: China (212 GWe), Brazil (82.2 GWe), USA (79 GWe), Canada (76.4 GWe), and Russia (46 GWe). Apart from those four countries with a relative abundance of it (Norway, Canada, Switzerland and Sweden), hydro capacity is normally applied to peak-load demand, because it is so readily stopped and started. The individual turbines of a hydro plant can be run up from zero to full power in about ten minutes. This also means that it is an ideal complement to wind power in a grid system, and is used thus most effectively by Denmark (see case study below). In 2011, hydro supplied about 3565 GWh (40% capacity factor), underlining its generally peak use.
Hydropower using large storage reservoirs is not a major option for the future in the developed countries because most major sites in these countries having potential for harnessing gravity in this way are either being exploited already or are unavailable for other reasons such as environmental considerations. Growth to 2030 is expected mostly in China and Latin America. China has commissioned the $26 billion Three Gorges dam, which produces 22.5 GWe and has a major role in flood control, but it has displaced over 1.2 million people.
The chief advantage of hydro systems is their capacity to handle seasonal (as well as daily) high peak loads. In practice the utilisation of stored water is sometimes complicated by demands for irrigation which may occur out of phase with peak electrical demands.
Run-of-river hydro systems are usually much smaller than dammed ones but have potentially wider application. Some short-term pondage can help them adapt to daily load profiles, but generally they produce continuously, apart from seasonal variation in river flows. Small-scale hydro plants under 10 MWe represent about 10% of world capacity, and most of these are run-of-river ones.
Pumped storage is discussed below under: Renewables in relation to base-load Electricity Demand.
Utilization of wind energy has increased spectacularly in recent years, with annual increases in installed capacity of around 20% in recent years. The 39 GWe increment in 2010 represented an investment of €47 billion (US$ 65 billion), and it was followed by a 41 GWe increase in 2011, 45 GWe in 2012, 35 GWe in 2013, 51 GWe in 2014 (23 GWe of this in China) and 63 GWe in 2015 (30.7 GWe of this in China). This brought total world wind capacity to 433 GWe, with tens of thousands of turbines now operating. However, all this has to be backed up with conventional generating capacity, due to low (20-30%) utilization and intermittency. (See later sections on this aspect.)
In 2014 political and regulatory uncertainty especially in the EU caused the rate of wind farm installations to drop markedly there as well as in some other countries. Investments were deterred by uncertainties and changes to renewable energy policies in several previously large wind markets. The rate of installations plunged by 90% in Denmark, 84% in Spain and 75% in Italy; and only 12.8 GWe was added in Europe, mostly in Germany, taking the European total to 129 GWe. Of this 8 GWe is offshore. Early in 2016 the EU total was 142 GWe, with 11 GWe of this offshore – 90% of the world offshore total.
Wind turbines of up to 6 MWe are now functioning in many countries, though most new ones are 1-3 MWe. The power output is a function of the cube of the wind speed, so doubling the wind speed gives eight times the energy potential. In operation such turbines require a wind in the range 4 to 25 metres per second (14-90 km/hr), with maximum output being at 12-25 m/s (the excess energy being spilled above 25 m/s). While relatively few areas have significant prevailing winds in this range, many have enough to be harnessed effectively and to give better than a 25% capacity utilisation.
Where there is an economic back-up which can be called upon at very short notice (e.g. hydro), a significant proportion of electricity can be provided from wind. The most economical and practical size of commercial wind turbines is now about 2 MWe, grouped into wind farms up to 200 MWe. Depending on site, most turbines operate at about 25% load factor over the course of a year (European average), but some reach 40% offshore. There is a distinct difference between onshore and offshore sites, though the latter are more expensive to set up and run. For the UK, in 2015, onshore wind averaged 30% capacity, and offshore 41%.
China led the field at the end of 2015 with 129 GWe installed according to its National Bureau of Statistics, or 145 GWe according to GWEC, overtaking the EU with 142 GWe. The USA has 74 GWe, Germany has 45 GWe, Spain has 23 GWe, India 25 GWe, the UK 14 GWe Canada 11 GWe, and France with 10 GWe at the end of 2015. World total then was 432 GWe. (Data taken from Global Wind Statistics 2015 published by the Global Wind Energy Council.)
Potentially the world’s largest wind farm is that planned by Forewind, a consortium of four major energy companies, for the Dogger Bank in the North Sea, costing some £30 billion. Stage 1 is 2.4 GWe, followed by 4.8 GWe, to give 7.2 GWe, which Forewind says will supply some 25 billion kWh/yr to the UK grid at projected 40% annual capacity factor. In the USA, the $8 billion, 3 GWe Anschutz Corp plant in Wyoming is planned to send power 1200 km via Utah and Nevada to the Californian grid near Las Vegas.
In 2016 the Dutch government auctioned the first development rights for the Borssele offshore wind farm, for a record low price – €72.70 per MWh – to Dong Energy plus about €14/MWh connection fee. It is auctioning 700 MWe per year to 2020.
With increased scale and numbers of units, generation costs decreased but have now stabilised. They are still greater than those for coal or nuclear, and allowing for backup capacity and grid connection complexities adds to them. Wind is intermittent, and when it does not blow, back-up capacity such as hydro or quick-start gas is needed. When it does blow, and displaces power from other sources, it may reduce the profitability of those sources and hence increase prices.
One approach to mitigate intermittency is to make hydrogen by electrolysis and feed this into the gas grid. E.On is building a pilot plant to produce up to 360 m3/hr of hydrogen at Falkenhagen, Germany, to feed into the Ontras gas grid, which can function with 5% hydrogen. It has been suggested that all electricity from wind might be used thus, greatly simplifying electrical grid management. Vattenfall at Prenzlau in Germany is also experimenting with hydrogen production and storage from wind power via electrolysis. Also in Germany, near Neubrandenburg, WIND-Projekt is using surplus electricity from a 140 MWe wind farm to make hydrogen, storing it, and then burning it in a CHP unit to make electricity when demand is high. However, there is an 84% loss in this double RH2-WKA demonstration process.
However, government policies in many countries ensure that power from wind turbines is able to be sold (see Appendix). The Global Wind Energy Council claimed that world capacity of 121 GWe at the end of 2008 would produce 260 TWh per year (ie 24.6% capacity factor). Applied to the 238 GWe at the end of 2011 that is 511 TWh/yr. Wind is projected to supply 3% of world electricity in 2030, and perhaps 10% in OECD Europe.
Wind turbines have a high steel tower to mount the generator nacelle, and have rotors with three blades up to 50m long. Foundations require a substantial mass of reinforced concrete. Hence the energy inputs to manufacture are not insignificant. Also siting is important in getting a net gain from them. In the UK the Carbon Trust found that small wind turbines on houses in urban areas often caused more carbon emissions in their construction and fitting than they saved in electrical output (CT 7/8/08).
Bird kills, especially of raptor species, are an environmental impact of wind farms. In the USA half a million birds are killed each year, including 83,000 raptors (hawks, eagles, falcons etc) according to reports of a peer-reviewed published estimate in Wildlife Society Bulletin. A similar estimate comes from the US Fish & Wildlife Service. Other figures are based on 2.1 fatalities per turbine per year. There is particular concern regarding birds covered by the US Migratory Bird Treaty Act and the Bald and Golden Eagle Protection Act, which make bird fatalities illegal.
New wind farms are increasingly offshore, in shallow seas. The UK had 3300 MWe wind capacity offshore, more than the rest of the world combined as of early 2013. The London Array, 20 km offshore Kent, has 175 turbines of 3.6 MWe, total 630 MWe, on a 245 km2 site and claims to be the world's largest offshore wind farm.
Replacing old turbines is becoming an issue – repowering the wind capacity. In Europe wind provides 11.5% of electricity early in 2016, from 142 GWe of capacity. Half of this will be retired by 2030, and needs to be replaced with larger turbines, likely without subsidies. The repowering priority is at the best sites. Decommissioning involves removal of old towers and foundations, not simply turbines.
Solar energy is readily harnessed for low temperature heat, and in many places domestic hot water units (with storage) routinely utilise it. It is also used simply by sensible design of buildings and in many ways that are taken for granted. Industrially, probably the main use is in solar salt production – some 1000 PJ per year in Australia alone (equivalent to two-thirds of the nation's oil use). It is increasingly used in utility-scale plants, mostly photovoltaic (PV), and by mid-2013 some 15 GWe of utility-scale solar power was installed, 3.1 GWe of this in China, and 2.9 GWe in each of Germany and USA. Domestic-scale PV is widespread.
The Renewables 2016 Global Status Report had 227 GWe of solar PV and 4.8 GWe of concentrating solar power (CSP), but no indication of output.
Three methods of converting the Sun's radiant energy to electricity are the focus of attention.
Photovoltaic (PV) systems
The best-known method utilises light, ideally sunlight, acting on photovoltaic cells to produce electricity. Flat plate versions of these can readily be mounted on buildings without any aesthetic intrusion or requiring special support structures. Solar photovoltaic (PV) has for some years had application for certain signaling and communication equipment, such as remote area telecommunications equipment in Australia or simply where mains connection is inconvenient. Sales of solar PV modules are increasing strongly as their efficiency increases and price falls, coupled with financial subsidies and incentives. In 2012 world installed PV capacity reached the 100 GWe milestone, with 30.5 GWe installed that year. In the World Energy Outlook 2011 New Policies Scenario, 553 GWe of new solar PV capacity (and 81 GWe of CSP) would be added by 2035.
Thin-film PV modules using silicon or cadmium telluride are at least 20% less costly than crystalline silicon-based ones, but are less efficient. Even working on 1 kilowatt per square metre in the main part of a sunny day, intensity of incoming radiation and converting this to high-grade electricity is still relatively inefficient – typically 10% in commercial equipment or up to 30% in more expensive units. But the cost per unit of electricity – at least ten times that of conventional sources – limits its unsubsidised potential to supplementary applications on buildings where its maximum supply coincides with peak demand.
In Germany, 1.5 million solar PV installations have combined capacity of 40 GWe and deliver 940 equivalent full-load hours of electricity per year – 10.7% capacity factor. On summer weekdays, solar irradiation covers up to 35% of German grid requirements, and nearly 50% on weekends.
More efficiency can be gained using concentrating solar PV (CPV), where some kind of parabolic mirror tracks the sun and increases the intensity of the solar radiation up to 1000-fold. Modules are typically 35-50 kW. In Australia a 2 MWe demonstration plant followed by a 102 MWe dense-array CPV power station was planned by Silex SolarSystems for Mildura in Victoria, with A$ 125 million government support promised. Anticipated cost of power is under 15c/kWh. Silex claimed 34.5% conversion efficiency, with a target of 50%. This project was abandoned in 2015 due to failure to find investment capital.
In the USA Boeing has licensed its XR700 high-concentration PV (HCPV) technology to Stirling Energy Systems with a view to commercializing it for plants under 50 MWe from 2012. The HCPV cells in 2009 achieved a world record for terrestrial concentrator solar cell efficiency, at 41.6%. CPV can also be used with heliostat configuration, with a tower among a field of mirrors.
In 2011 several Californian plants planned for solar thermal changed plans to solar PV – see mention of Blythe, Imperial Valley and Calico below.
Many solar PV plants are connected to electricity grids in Europe and USA, and now China. The OECD IEA reported 23 GWe of solar PV capacity in 2009, 17 GWe of this in Europe. China's 200 MWe Golmud solar park was commissioned in 2011 and is claimed to produce 317 GWh/yr (18% capacity factor). The 100 MWe Perovo solar park in Ukraine was commissioned in 2011 also, with 15% capacity factor claimed. EdF has built the 115 MWe Toul-Rosieres thin-film PV plant in eastern France. There is a 97 MWe Sarnia plant in Canada. In Italy, SunEdison plans to build a 72 MWe solar PV plant near Rovigo, for $342 million. India’s 214 MWe Gujarat Solar Park was commissioned in 2012 and aims for eventual 1000 MWe capacity. The Indian government announced the 4 GWe Sambhar project in Rajasthan in 2013, expected to produce 6.4 TWh/yr, i.e. capacity factor of 18% from almost 80 sq km. The initial 1 GWe is expected to operate from 2016, costing Rs 7,500 crore ($1.2 billion). In Australia the 102 MWe Nyngan solar PV array cost A$290 million and is expected to produce 230 GWh/yr from 2015, i.e. 26% capacity factor.
In the USA, the 550 MWe Desert Sunlight solar farm in the Mojave Desert opened early in 2015, using cadmium telluride thin film technology and financed with a $1.46 billion federal loan guarantee. MidAmerican’s Antelope Valley plants in California comprise a 579 MWe development with Sunpower as EPC contractor and due to be complete at the end of 2015. Its panels will track the sun, giving 25% more power. MidAmerican Solar owns the 550 MWe Topaz Solar Farms in San Luis Obispo County, Calif., and has a 49% interest in the 290 MWe Agua Caliente thin-film PV project commissioned in 2014 by First Solar in Yuma County, Arizona. Many PV plants are over 20 MWe, and quoted capacity factors range from 11% to 27%.
A South Korean consortium has commissioned 42 MWe PV capacity at two plants in Bulgaria, which are expected to produce 61 GWh/yr (16.5% capacity factor), their cost being €154 million (€3667/kW). Research continues into ways to make the actual solar collecting cells less expensive and more efficient. In some systems there is provision for feeding surplus PV power from domestic systems into the grid as contra to normal supply from it, which enhances the economics. The 2000 MWe Ordos thin-film solar PV plant is planned in Inner Mongolia, China, with four phases – 30, 100, 870, 1000 MWe – to be complete in 2020. Over 30 others planned are over 100 MWe, most in India, China, USA and Australia. A 230 MWe solar PV plant is planned at Setouchi in Japan, with GE taking a major stake in the JPY 80 billion project expected on line in 2018. Serbia plans a 1 GWe solar PV project costing €1.3 billion which is expected to deliver 1.15 TWh/yr to Enerxia Energy from 2015, a 13% capacity factor, without any feed-in tariff. (That output at €50/MWh would return €57.5 million pa. After €20 million pa maintenance, it is less than 3% pa return on capital.)
In recent years there has been high investment in solar PV, due to favourable subsidies and incentives. In 2011 Italy saw 9000 MWe of solar PV installed, and Germany 7500 MWe of solar. In Germany, solar PV capacity reached 32.4 GWe at end of 2012 (7.6 GWe installed during the year) and generated 28 billion kWh, increasing 45% over 2011, but apparently only 11% capacity factor. In Italy, feed-in tariffs range from 15-27 euro cents/kWh, depending on size, giving a 2011 cost to consumers of nearly €6 billion. In the World Energy Outlook 2011 New Policies Scenario, 553 GWe of new solar PV and 81 GWe of CSP capacity would be added by 2035. Solar PV capacity at the end of 2011 was 67 GWe.
In Nigeria, the federal government and Delta state have set up a $5 billion public-private partnership with SkyPower FAS Energy to build 3 GWe of utility-scale solar PV capacity, with the first units coming on line in 2015. A feed-in tariff regime will support this.
A serious grid integration problem with solar PV is that cloud cover can reduce output by 70% in the space of one minute. Various battery and other means are being developed to slow this to 10% per minute, which is more manageable. The particular battery system required is designed specifically to control the rate of ramp up and ramp down. System life is ten years, compared with twice that for most renewable sources.
The manufacturing and recycling of PV modules raises a number of questions regarding both scarce commodities, and health and environmental issues. Copper indium gallium selenide (CIGS) solar cells are a particular concern, both for manufacturing and recycling. Silicon-based PV modules require high energy input in manufacture, though the silicon itself is abundant.
Solar thermal systems, concentrating solar power (CSP)
Solar thermal systems need sunlight rather than the more diffuse light which can be harnessed by solar PV. A solar thermal power plant has a system of mirrors to concentrate the sunlight on to an absorber, the energy then being used to drive turbines – concentrating solar thermal power (CSP). About 2.55 GWe of CSP capacity worldwide (end of 2012), three-quarters of this in Spain, supplies a proportion of the solar electricity. More CSP is under development.
The concentrator may be a parabolic mirror trough oriented north-south, which tracks the sun's path through the day. The absorber is located at the focal point and converts the solar radiation to heat in a fluid such as synthetic oil, which may reach 700°C. The fluid transfers heat to a secondary circuit producing steam to drive a conventional turbine and generator. Several such installations in modules of up to 80 MW are now operating. Each module requires about 50 hectares of land and needs very precise engineering and control. These plants are supplemented by a gas-fired boiler which generates about a quarter of the overall power output and keeps them warm overnight.
A simpler CSP concept is the Fresnel collector using rows of long narrow flat (or slightly curved) mirrors tracking the sun and reflecting on to one or more fixed linear receivers positioned above them. The receivers may generate steam directly.
In mid-2007 Nevada Solar One, a 64 MWe capacity solar thermal energy plant, started up. The $250 million plant is projected to produce 124 million kWh per year and covers about 160 hectares with 760 mirrored troughs that concentrate the heat from the desert sun on to pipes that contain a heat transfer fluid. This is heated to 390°C and then produces steam to drive turbines. Nine similar units totaling 354 MWe have been operating in California as the Solar Energy Generating Systems. More than twenty Spanish 50 MWe parabolic trough units including Andasol 1-3, Alvarado 1, Extresol 1-2, Ibersol and Solnova 1-3, Palma del Rio 1-2, Manchasol 1-2, Valle 1-2, commenced operation in 2008-11. Andasol, Manchasol and Valle have 7.5-hour heat storage.
Other US CSP parabolic trough projects under construction include Abengoa's Solana in Arizona, a 280 MWe project with six-hour molten salt storage enabling power generation in the evening. It has a 778 ha solar field and started operation in 2013. The $2 billion cost is offset by a $1.45 billion loan guarantee from the US Department of Energy. Abengoa's 280 MWe Mojave Solar Project near Barstow in California also uses parabolic troughs in a 715 ha solar field and came on line in 2014. It has a $1.2 billion federal loan guarantee.
In 2010 California approved construction of the $6 billion, 968 MWe Blythe CSP plant by Solar Trust, the US arm of Solar Millennium at Riverside, Calif., using parabolic trough technology in four 250 MWe units occupying 28.4 sq km and funded partly by US Dept of Energy. The company has a $2.1 billion loan guarantee and a 20-year power purchase agreement with SC Edison, from 2013. However, this has now become a solar PV project, apparently due to difficulty in raising finance. Also in California, Imperial Valley (709 MWe), and Calico (663 MWe) are Stirling engine systems (see below), though the new owners of Calico are switching 563 MWe of it to PV, and Imperial Valley is re-permitting for PV. Abu Dhabi commissioned its 100 MWe Shams parabolic trough CSP plant in 2013; it cost $600 million.
Another form of this CSP is the power tower, with a set of flat mirrors (heliostats) which track the sun and focus heat on the top of a tower, heating water to make steam, or molten salt to 1000°C and using this both to store the heat and produce steam for a turbine. California's Solar One/Two plant produced 10 MWe for a few years. Abengoa’s Solucar complex in Spain has the 11 MWe PS10 power tower plant with 624 mirrors, each 120 m2 and the 20 MWe PS20 adjacent, with 1255 mirrors, producing steam directly in the tower. Solucar also has three parabolic trough plants of 50 MW each. By 2015 Spain expected to have 2000 MWe of CSP operating. The 500 MWe Guzman CSP plant at Palma de Rio was opened in 2012. Power production in the evening can be extended fairly readily using gas combustion for heat.
The US Department of Energy awarded a $1.37 billion loan guarantee to BrightSource Energy to build the 392 MWe Ivanpah Solar Power complex in the Mojave Desert of California, essentially a gas-fired plant with major subsidized solar supplement. It comprises three CSP Luz power towers which simply heat water to 550°C to make steam, using 300,000 heliostat mirrors in pairs each of 14 m2 per MWe, in operation from 2013 as the world's largest CSP plant. The steam cycle uses air-cooled condensers. There is a back-up gas turbine, and natural gas is used to pre-heat water in the towers. It was expected to generate 940 GWh/yr, but mid-2014 to mid-2015 only managed 59% of this. It burned 915 TJ of gas in 2014 which resulted in 46,000 tonnes of CO2 emissions that year. On its own this gas would produce 250 GWh of electricity, so it appears to be 45% gas-fired rather than solar in 2014, and 27% gas-fired at full solar performance. BrightSource plans a similar 500 MWe plant nearby in the Coachella Valley.
BrightSource Energy is partnering with Alstom and NOY Infrastructure & Energy Investment Fund to build the 121 MWe Ashalim Solar Thermal Power Station in Israel’s Negev desert. It will use BrightSource’s CSP tower with more than 50,000 computer-controlled heliostats tracking the sun on two axes and reflecting sunlight on to a boiler on top of a 240-meter tower.
Using molten salt in the CSP system as the transfer fluid which also stores heat, enables operation into the evening, thus approximating to much of the daily load demand profile. Spain's 20 MWe Gemasolar (formerly Solar Tres) plant has 2500 mirrors/ heliostats, each 115 m2 and molten salt storage, claiming to be the world's first near base-load CSP plant, with 63% capacity factor. Its cost is reported to be $33,000 /kW. Spain's 200 MWe Andasol plant also uses molten salt heat storage, as does California's 280 MWe Solana and Nevada’s 110 MWe Crescent Dunes plant with power tower and 10-hour heat storage claimed. The salt used may be 60% sodium nitrate, 40% potassium nitrate with melting point 220°C. Andasol stores heat at 400°C and requires 75 t of salt per MW of heat. Its condensers require 5 L/kWh for cooling. Spain's Gemasolar employs 6250 tonnes of salt. Solana uses 125,000 tonnes of salt, kept at 277°C. In Colorado the 2x100 MWe SolarReserve plant in San Luis Valley will use molten salt.
An 810 MWe plant occupying 13 sq km with six power towers is being built in Qinghai province in northwest China, by BrightSource with Shanghai Electric Group. It will have heat storage using molten salt. Phase 1 of this Qinghai Delingha Solar Thermal Power Project is two 135 MWe CSP plants using BrightSource power towers with up to 3.5 hours of heat storage and producing 628 GWh/yr, hence 26.55% capacity factor. Majority ownership is by Huanghe. The project will apply to NDRC for feed-in tariff. It is part of an international collaboration.
In Morocco the 580 MWe Noor-Ouarzazate CSP plant is a major project due for completion in 2018, with its first 160 MWe phase, Noor 1, commissioned early in 2016 and supplying power at $0.19/kWh. It uses parabolic trough collectors heating oil or salt which produces steam in a secondary circuit. The full plant will be the size of the country’s capital city.
A small portable CSP unit – the Wilson Solar Grill – uses a Fresnel lens to heat lithium nitrate to 230°C so that it can cook food after dark.
Another CSP set-up is the Solar Dish Stirling System which uses reflectors to concentrate energy to drive a stirling cycle engine. A Tessera Solar plant of 709 MWe is planed at Imperial Valley in California. The system consists of a solar concentrator in a dish structure with an array of curved glass mirror facets which focus the energy on the power conversion unit's receiver tubes containing hydrogen gas which powers a Stirling engine. Solar heat pressurizes the hydrogen to power the four-cylinder reciprocating Solar Stirling Engine and drive a generator. The hydrogen working fluid is cooled in a closed cycle. Waste heat from the engine is transferred to the ambient air via a water-filled radiator system. The stirling cycle system is as yet unproven in these large applications, however.
With solar input being both diffuse* and interrupted by night and by cloud cover, solar electric generation has a low capacity factor, typically less than 15%, though this is partly addressed by heat storage using molten salt. Power costs are two to three times that of conventional sources, which puts it within reach of being economically viable where carbon emissions from fossil fuels are priced.
Large CSP schemes in North Africa, supplemented by heat storage, are proposed for supplying Europe via high voltage DC links. One proposal is the TuNur project based in Tunisia and supplying up to 2000 MWe via HVDC cable to Italy. A related and more ambitious one was Desertec, with estimated cost of €400 billion, networking the EU, Middle East and North Africa (MENA) with 20 transmission lines of 5 GW each, to provide 15% of Europe's electricity and much of that in MENA by 2050. The Desertec Foundation was set up in 2009 as an NGO to promote the Desertec concept.
The Desertec Industrial Initiative GmbH (Dii) “Desertenergy” is a Europe-based consortium founded in 2009 to advance the grand vision and work towards creation of a market for desert power in EU and MENA. It comprised 55 companies and institutions and is active in Morocco, Algeria and Tunisia. The first Dii-fostered project was to be the Noor-Ouarzazate 580 MWe CSP plant in Morocco (see above). Morocco is the only African country to have a transmission link to Europe. In mid-2013 the Desertec Foundation left the Dii consortium. Bosch and Siemens had left it in 2012. The Desertec Industrial Initiative then announced that it would focus on consulting after most of its former backers pulled out. The remaining members of the Munich-based consortium are Saudi company ACWA Power, German utility RWE and Chinese grid operator SGCC. The new network “Supporters of Desert Energy” became operational early in 2015 in Dubai to “identify practical hurdles for projects and offer solutions in interaction with the public sector and the civil society.”
The Mediterranean Solar Plan (MSP) targets the development of 20 GWe of renewables by 2020, of which 5 GWe could be exported to Europe. Total investment would be of the order of EUR 60 billion. The OECD IEA's World Energy Outlook 2010 says: The quality of its solar resource and its large uninhabited areas make the Middle East and North Africa region ideal for large-scale development of concentrating solar power, costing 10 to 13.5 c/kWh ... in 2035. Solar power could be exported to Europe (at transmission costs of 2 to 5 c/kWh) and/or to countries in sub-Saharan Africa. The report projects that the actual CSP generation cost in North Africa could be the same as EU wholesale electricity price in 2035 – about 10 c/kWh.
CSP boost to fossil fuel power, hybrid systems
Solar energy producing steam can be used to boost conventional steam-cycle power stations. Australia's Kogan Creek Solar Boost Project will be the largest solar integration with a coal-fired power station in the world when it is operational in 2013. A 30-hectare field of Areva Solar's compact linear Fresnel reflectors at the existing Kogan Creek power station will produce steam which will be fed to the modern supercritical 750 MWe coal-fired power station, helping to drive the intermediate pressure turbine. The solar boost at 44 MW (peak sunshine) will add 44 million kWh annually, about 0.75% of output, for $105 million – equivalent to $19,000/kW of base-load capacity. The 2000 MWe Liddell coal-fired power station has a 2 MWe equivalent solar boost (9 MW thermal addition).
In the USA the federal government has a SunShot initiative to integrate CSP with fossil fuel power plants as hybrid systems. Some $20 million is offered for two to four projects. The US Department of Energy says that 11 to 21 GWe of CSP could effectively be integrated into existing fossil fuel plants, utilizing the turbines and transmission infrastructure.
While CSP is well behind solar PV as its prices continue to fall and utilities become more familiar with PV. However, CSP can provide thermal storage and thus be dispatchable and it can provide low-cost steam for existing power plants (hybrid set up). Also, CSP has the potential to provide heating and cooling for industrial processes and desalination.
Solar updraft tower
Another kind of solar thermal plant is the solar updraft tower, using a huge chimney surrounded at its base by a solar collector zone like an open greenhouse. The air under this skirt is heated and rises up the chimney, turning turbines as it does so. The 50 MWe Buronga plant planned in Australia was to be a prototype, but Enviromission's initial plans are now for two 200 MWe versions each using 32 turbines of 6.25 MWe with a 10 square kilometre collector zone under a 730 metre high tower in the Arizona desert. Thermal mass – possibly brine ponds – under the collector zone means that some operation will continue into the night. A 50 kWe prototype plant of this design operated in Spain 1982-89. In China the 27.5 MWe Jinshawan solar updraft tower is under construction.
A significant role of solar energy is that of direct heating. Much of our energy need is for heat below 60oC, eg. in hot water systems. A lot more, particularly in industry, is for heat in the range 60-110oC. Together these may account for a significant proportion of primary energy use in industrialised nations. The first need can readily be supplied by solar power much of the time in some places, and the second application commercially is probably not far off. Such uses will diminish to some extent both the demand for electricity and the consumption of fossil fuels, particularly if coupled with energy conservation measures such as insulation.
With adequate insulation, heat pumps utilising the conventional refrigeration cycle can be used to warm and cool buildings, with very little energy input other than from the sun. Eventually, up to ten percent of total primary energy in industrialised countries may be supplied by direct solar thermal techniques, and to some extent this will substitute for base-load electrical energy.
Where hot underground steam can be tapped and brought to the surface it may be used to generate electricity. Such geothermal sources have potential in certain parts of the world such as New Zealand, USA, Philippines and Italy. Global installed capacity was about 12.9 GWe at the end of 2015, including 2600 MWe in California, 1900 MWe in Philippines and 1200 MWe in Indonesia, and in 2013 geothermal produced 72 billion kWh worldwide. In Japan 500 MWe of capacity produces 0.3% of the country's electricity. In New Zealand 420 MWe produces over 7% of the electricity, and Iceland gets one-quarter of its electricity from 200 MWe of geothermal plant (90 MWe more is planned), and also most of its district heating. Mexico has 958 MWe geothermal, Italy 843 MWe and Nevada 470 MWe. In Italy, ENEL got 8769 TWh from geothermal in 2012, in Iceland 4974 TWh. Lihir Gold mine in Papua New Guinea has 56 MWe installed, the last 20 MWe costing US$ 40 million – about the same as annual savings from the expanded plant. Geothermal electric output is expected to triple by 2030. The largest geothermal plant is The Geysers in California, producing about 1000 MWe, but diminishing. See also Geothermal Energy Association website.
There are also prospects in certain other areas for hot fractured rock geothermal, or hot dry rock geothermal – pumping water underground to regions of the Earth's crust which are very hot or using hot brine from these regions. The heat – up to about 250°C – is due to high levels of radioactivity in the granites and because they are insulated at 4-5 km depth. They typically have 15-40 ppm uranium and/or thorium, but may be ten times this. The heat from radiogenic decay* is used to make steam for electricity generation. South Australia has some very prospective areas. The main problem with this technology is producing and maintaining the artificially-fractured rock as the heat exchanger. Only one such project is operational, the Geox 3 MWe plant at Landau, Germany, using hot water (160ºC) pumped up from 3.3 km down (and maybe should be classed as conventional geothermal). It cost EUR 20 million. A 50 MWe Australian plant is envisaged as having 9 deep wells – 4 down and 5 up.
Ground source heat pump systems or engineered geothermal systems also come into this category, though the temperatures are much lower. Generally the cost of construction and installation is prohibitive for the amount of energy extracted. The 1997 Geoscience Australia building in Canberra is heated and cooled thus, using a system of 210 pumps throughout the building which carry water through loops of pipe buried in 352 boreholes each 100 metres deep in the ground. Here the temperature is a steady 17°C, so that it is used as a heat sink or heat source at different times of the year. See 10-year report (pdf).
This falls into three categories – tidal, wave and temperature gradient, described separately below.
Collectively they are receiving more attention, especially in the EU, where a 2015 study suggested that ocean energy might provide 10% of power by 2050, given collaborative effort. The European Commission's Strategic Energy Technology (SET) plan acknowledges the potential role of ocean energy in Europe's future energy mix and suggests enhancing regional cooperation in the Atlantic region. The EU Ocean Energy Forum is being formed by 2016 and will develop a roadmap by 2020.
Harnessing the tides with a barrage in a bay or estuary has been achieved in France (240 MWe in the Rance Estuary, since 1966), Canada (20 MWe at Annapolis in the Bay of Fundy, since 1984), South Korea (Sihwa, 260 MWe, since 2011), and Russia (White Sea, 0.5 MWe), and could be achieved in certain other areas where there is a large tidal range. The trapped water can be used to turn turbines as it is released through the tidal barrage in either direction. Worldwide this technology appears to have little potential, largely due to environmental constraints.
Natural Energy Wyre (NEW) in UK has set up a consortium to develop the Eco-THEP, a 90 MW tidal barrage plant with six turbines on the River Wyre near Fleetwood in northwest England by 2020. It predicts annual output of 220 GWh.
However, tidal arrays, placing free-standing turbines in major coastal tidal streams appears to have greater potential, and this is being explored.
Currents are predictable and those with velocities of 2 to 3 metres per second are ideal and the kinetic energy involved is equivalent to a very high wind speed. This means that a 1 MWe tidal turbine rotor is less than 20 m diameter, compared with 60 m for a 1 MWe wind turbine. Units can be packed more densely than wind turbines in a wind farm, and positioned far enough below the surface to avoid storm damage.
A 300 kW turbine with 11 m diameter rotor in the Bristol Channel can be jacked out of the water for maintenance. Based on this prototype, early in 2008 the 1.2 MWe SeaGen twin turbine was installed in Strangford Lough, Northern Ireland, billed as the first commercial unit of its kind the world’s largest grid-connected tidal stream turbine. It produces power 18-20 hours per day and is operated by a Siemens subsidiary. The next project is a 10.5 MWe nine-turbine array off the coast of Anglesey. An 86 MWe tidal turbine project in Pentland Firth, between Orkney and the Scottish mainland has been approved, and MeyGen’s initial 9 MWe demonstration array of six turbines is expected on line in 2015, using Atlantis and Andritz technology. The first Atlantis 1MWe prototype was deployed at the European Marine Energy Centre at Orkney in 2011, and a 1 MWe Andritz Hydro Hammerfest prototype is also deployed there, as is a 2 MWe turbine from Scotrenewables mounted under a barge. At the North Shetland tidal array in Bluemull Sound, Nova Innovation is installing three 100 kW turbines, the first already supplying power to the grid.
In France, two pilot 1 MWe tidal turbines are being commissioned by EDF off the Brittany coast at the end of 2015. They are 16 m diameter and will pilot the technology with a view to installation of seven 2 MWe tidal turbines in the Raz Blanchard tidal race off Normandy in 2018.
French energy company Engie has announced plans to build a tidal energy project on the western coast of the Cotentin peninsula in northwest France. It aims to install four tidal turbines with a total generating capacity of 5.6 MWe by 2018 in a region with the strongest marine currents in Europe.
Some tidal stream generators are designed to oscillate, using the tidal flow to move hydroplanes connected to hydraulic arms sideways or up and down. A prototype has been installed off the coast of Portugal.
Another experimental design is using a shroud to speed up the flow through a venturus in which the turbine is placed. This has been trialled in Australia and British Colombia.
A major pilot project using three kinds of tidal stream turbines is being installed in the Bay of Fundy's Minas Passage, about three kilometers from shore. Some 3 MWe will be fed to the Canadian grid from the pilot project. Eventually 100 MWe is envisaged. The three designs are a 10m diameter turbine from Ireland, a Canadian Clean Current turbine and an Underwater Electric Kite from USA.
Tidal power comes closest of all the intermittent renewable sources to being able to provide a continuous and predictable output, and is projected to increase from 1 billion kWh in 2002 to 35 billion in 2030 (including wave power).
Harnessing power from wave motion has the potential to yield significant electricity. The feasibility of this has been investigated, particularly in the UK. Generators either coupled to floating devices or turned by air displaced by waves in a hollow concrete structure (oscillating water column) are two concepts for producing electricity for delivery to shore. Other experimental devices are submerged and harness the changing pressure as waves pass over them.
The first commercial wave power plant is in Portugal, with floating rigid segments which pump fluid through turbines as they flex at the joints. It can produce 2.25 MWe. Another – Oyster – is in UK and is designed to capture the energy found in nearshore waves in water depths of 12 to 16 metres. Each 200-tonne module consists of a large buoyant hinged flap anchored to the seabed. Movement of the flap with each passing wave drives a hydraulic piston to deliver high-pressure water to an onshore turbine which generates electricity. The 315 kW demonstration module being tested in the Orkney Islands is expected to have about a 42% capacity factor.
In Australia Carnegie Wave Energy has three 240 kW CETO 5 units delivering power to the grid. The CETO 5 system consists of buoys that are fully submerged and their movement drives seabed pump units to deliver high pressure fluid via a subsea pipe to standard hydroelectric turbines onshore. A three-unit plant using 1 MW CETO 6 units is planned – these generate power inside the buoyant actuator attached to a pump tethered to the seabed. Another submerged ocean-surge design is AW-Energy’s WaveRoller on the seabed 8 to 20 metres deep. A 350 kW full-scale module is to be installed near Peniche in Portugal in 2017, supported by €10 million from the European Investment Bank.
Numerous practical problems have frustrated progress with wave technology, not least storm damage.
Ocean thermal energy
Ocean thermal energy conversion (OTEC) has long been an attractive idea, but is unproven beyond small pilot plants up to 50 kWe, though in 2015 a 100 kWe closed cycle plant was commissioned in Hawaii and connected to the grid. It works by utilising the temperature difference between equatorial surface waters and cool deep waters, the temperature difference needing to be about 20ºC top to bottom. In the open cycle OTEC the warm surface water is evaporated in a vacuum chamber to produce steam which drives a turbine. It is then condensed in a heat exchanger by the cold water. The main engineering challenge is in the huge cold water pipe which needs to be about 10 m diameter and extend a kilometre deep to enable a large water flow. A closed cycle variation of this uses an ammonia cycle. The ammonia is vapourised by the warm surface waters and drives a turbine before being condensed in a heat exchanger by the cold water. A 10ºC temperature difference is then sufficient.
Beyond traditional direct uses for cooking and warmth, growing plant crops particularly wood to burn directly or to make fuels such as ethanol and biodiesel has a lot of support in several parts of the world, though mostly focused on transport fuel. More recently, wood pellets for electricity generation have been newsworthy. The main issues here are land and water resources. The land usually must either be removed from agriculture for food or fibre, or it means encroaching upon forests or natural ecosystems. Available fresh water for growing biofuel crops such as maize and sugarcane and for processing them may be another constraint.
Burning biomass for generating electricity has some appeal as a means of utilising solar energy for power. However, the logistics and overall energy balance usually defeat it, in that a lot of energy – mostly oil based – is required to harvest and move the crops to the power station. This means that the energy inputs to growing, fertilising and harvesting the crops then processing them can easily be greater than the energy value in the final fuel, and the greenhouse gas emissions can be similar to those from equivalent fossil fuels. Also other environmental impacts can be considerable. For long-term sustainability, the ash containing mineral nutrients needs to be returned to the land.
In southeastern USA, 1.75 million tonnes of wood pellets were exported to Europe in 2012, and the figure is projected to grow to over 5 Mt in 2015. Most of this comes from low-value woodchips which formerly went into papermaking. Three 660 MWe units of Drax, Britain’s largest coal-fired power station, are being converted to burn wood, most of it imported (like the coal of higher heat value that it replaces). Early in 2015 two units were converted and delivering 630 MWe each, on 75% load factor and contributing to system balancing. No CO2 emissions are attributed to the actual burning, on the basis that growing replacement wood balances out those emissions, albeit in a multi-decade time frame. Drax figures show 121 g/kWh CO2 for harvesting, preparing and transporting wood pellets to UK, compared with mined and delivered coal 32 g/kWh. Unlike coal, the wood needs to be stored under cover. Drax expects too use 10 Mt/yr of pellets by the end of 2016.
In Australia and Latin America sugar cane pulp is burned as a valuable energy source, but this (bagasse) is a by-product of the sugar and does not have to be transported. In the EU in 2010 over 11 million tonnes of wood pellets were used. In 2012 Europe imported 4.36 million tonnes of pellets and in 2015, 15 million tonnes import is projected, two thirds from North America (as in 2012). The pellets are made from sawmill residues preferably, but also forest residues and low-value timber. UK demand is expected to reach 11 million tonnes by 2015, equivalent to twice that amount of fresh wood.
In 2015 biomass and waste provided 429 TWh of electricity worldwide, from 101 GWe of capacity. By 2030 biomass-fuelled electricity production was projected to triple and provide 2% of world total, 4% in OECD Europe, as a result of government policies to promote renewables. However, such projections are increasingly challenged as the cost of biofuels in water use and pushing up food prices is increasingly questioned. In particular, the use of ethanol from corn and biodiesel from soybeans reduces food production and arguably increases world poverty. The cost in subsidies is also increasingly questioned: in the OECD US$ 13-15 billion is spent annually on biofules which provide only 3% of liquid transport fuel.
In 2008 about 100 million tonnes of grain (enough feed nearly 450 million people) was expected to be turned into fuel. This includes a legislated 40% of the US corn crop, aided by heavy subsidies. In 2012 the US corn crop amounted to about 360 million tonnes, so about 140 Mt would have been used for ethanol. Meanwhile basic food prices have risen sharply, leading the UN Food & Agriculture Organisation in mid-2012 to call for the USA to halt its biofuel production to prevent a food crisis. In any case, the energy return on investment (EROI) of corn ethanol in the USA is strongly questioned, and a consensus that it is below the minimum useful threshold is reported.
A new technology, Pavegen, uses pavement tiles about one metre square to harvest energy from pedestrian traffic. A footfall on a tile will flex it about 5mm and result in up to 8 watts of power over the duration of the footstep. Electricity can be stored, used directly for lighting, or in other ways.
In recent years there has been discussion as to whether nuclear power can be categorised as “renewable”. In the context of sustainable development it shares many of the benefits of many renewables, it is a low-carbon energy source, it has a very small environmental impact, similarities that are in sharp contrast to fossil fuels. But commonly, nuclear power is categorised separately from ‘renewables’. Nuclear fission power reactors do use a mineral fuel and demonstrably depletes the available resources of that fuel.
In the future nuclear power will make use of fast neutron reactors. As well as utilizing about 60 times the amount of energy from uranium, they will unlock the potential of using even more abundant thorium as a fuel. In addition, some 1.5 million tonnes of depleted uranium now seen by some people as little more than a waste, becomes a fuel resource. In effect, they will ‘renew’ their own fuel resource as they operate. The consequence of this is that the available resource of fuel for fast neutron reactors is so plentiful that under no practical terms would the fuel source be significantly depleted.
‘Renewables’, as currently defined, would offer no meaningful advantage over fast neutron reactors in terms of availability of fuel supplies. Most also tend to make very large demands on resources to construct the plant used for harnessing the natural energy – per kilowatt hour produced, much more than nuclear power.
Centralised state utilities focused on economies of scale can easily overlook an alternative model – of decentralized electricity generation, with that generation being on a smaller scale and close to demand. Here higher costs may be offset by reduced transmission losses (not to mention saving the capital costs of transmission lines) and possibly increased reliability. Generation may be on site or via local mini grids.
Electricity storage at utility scale
In some places pumped storage is used to even out the daily generating load by pumping water to a high storage dam during off-peak hours and weekends, using the excess base-load capacity from low-cost coal or nuclear sources. During peak hours this water can be used for hydro-electric generation. Relatively few places have scope for pumped storage dams close to where the power is needed, and overall efficiency is 70 to 75%, but about 150 GWe pumped storage is installed worldwide, including 21 GWe in the USA and 38 GWe in Europe.* This amounts to some 500 GWh of storage capacity.
Pumped storage comprised 95% of the world’s large-scale electricity storage in mid-2016, and 72% of that which was added in 2014. Battery storage, however, was deployed strongly through 2014, with a record 400
MWe of additions, more than doubling the installed base in 2013. Building-scale power storage emerged in 2014 as a defining energy technology trend. The market grew by 50% year-on-year, with lithium-ion batteries prominent, but flow batteries showing promise. In 2015 battery storage costs were around $400/kWh, and 1.6 GWe was installed or planned.
In 2015 lithium-ion batteries accounted for 51% of newly announced energy storage system (ESS) capacity and 86% of deployed ESS power capacity. An estimated 1,653 MW of new ESS were announced around the world in 2015, with just over one-third coming from North America. Li-ion batteries are the most popular technology for distributed energy storage systems.
Following a two-year study by the California Public Utilities Commission, the state in 2010 passed legislation requiring 1325 MWe of electricity storage (excluding large-scale pumped storage) by 2024. In 2013 it brought forward the deadline to 2020, then having 35 MWe total. The legislation specifies power, not storage capacity (MWh), suggesting that the main purpose is frequency control. The stated purpose of the legislation is to increase grid reliability by providing dispatchable power from an increasing proportion of solar and wind inputs, replace spinning reserve, provide frequency control and reduce peak capacity requirements (peak shaving). The storage systems can be connected with either transmission or distribution systems, or be behind the meter. The main focus is on battery energy storage systems (BESS). Energy arbitrage may enhance revenue, buying off-peak and selling for peak demand. Southern California Edison has plans for 260 MWe of electricity storage to offset the closure of a nuclear plant. However, 1.3 GWe in the context of the state’s 50 GWe will not provide much dispatchable power.
In 2015 lithium-ion batteries accounted for 51% of newly announced energy storage system (ESS) capacity and 86% of deployed ESS power capacity. An estimated 1,653 MW of new ESS capacity was announced around the world in 2015, with just over one-third coming from North America. Li-ion batteries are the most popular technology for distributed energy storage systems (Navigant Research). Lithium-ion batteries have a 95% round trip direct current efficiency, falling to 85% when the current is converted to alternating current for the grid. They have a 10-20 year lifespan, depending on use.
A World Energy Council report in January 2016 projected a significant drop in cost for the majority of storage technologies from 2015 to 2030. Battery technologies showed the greatest reduction in cost, followed by sensible thermal, latent thermal and supercapacitors. Battery technologies showed a reduction from around €100-700/MWh in 2015 to €50-190/MWh in 2030 – a reduction of over 70% in the upper cost limit in the next 15 years. Sodium sulfur (NaS), lead acid and lithium-ion technologies lead the way.
Early in 2016 the UK’s National Grid got a strong response to a tender for 200 MWe enhanced frequency response. It was offering four-year contracts for capacity able to provide 100% active power output in a second or less of registering a frequency deviation. Some 888 MWe of battery capacity was offered, 150 MWe of interconnection, 100 MWe of demand-side response and 50 MWe of flywheel capacity. UK Power Network has a 6MW/10 MWh demonstration battery operational at Leighton Buzzard. In May 2016 renewables energy company RES signed a four-year contract with National Grid to provide 20 MW of dynamic frequency response from lithium ion battery storage, to be operational in 18 months. RES already has more than 100 MWe/60 MWh of battery storage in operation, mostly in North America. Nearly 1 GWe of battery storage is forecast on the UK grid by 2020.
In the USA 221 MWe of new storage capacity was deployed in 2015, and this is expected to grow to 1700 MWe by 2020.
Battery energy storage systems (BESS)
Over one-third of the 1.5 GWe ‘battery storage’ in 2015 was lithium-ion batteries, and 22% was sodium-sulfur batteries. The International Renewable Energy Agency (IRENA) estimates that the world needs 150 GWe of battery storage to meet its desired target of 45% of power generated from renewable sources by 2030.
Probably the largest BESS is a 40 MWe/20 MWh Toshiba Li-ion system at the Tohoku Electric Power Company’s Nishi-Sendai substation in Japan, commissioned early in 2015, but STEAG Energy Services has started a 90 MWe storage program in Germany and Edison is setting up a 100 MWe facility in Long Beach, California.
In August 2015 GE was contracted to build a 30 MWe/20 MWh Li-ion battery storage system for Coachella Energy Storage Partners (CESP) in California, 160 km east of San Diego. The facility will aid grid flexibility and increase reliability on the Imperial Irrigation District network by providing solar ramping, frequency regulation, power balancing and black start capability for an adjacent gas turbine.
A large utility-scale electricity storage is a 4 megawatt sodium-sulfur (NaS) battery system to provide improved reliability and power quality for the city of Presidio in Texas. The 4 MWe set-up was energized early in 2010 to provide rapid back-up for wind capacity in the local ERCOT grid. Another 4 MWe NaS system was commissioned in May 2013. The $18 million Yerba Buena BESS Pilot Project in San Jose, California, was set up by PG&E with $3.3 million support from the California Energy Commission. PG&E operates a similar 2 MWe BESS near its Vaca-Dixon solar plant in Solano County, California. Sodium-sulfur batteries are widely used elsewhere for similar roles.
A large project is Southern California Edison’s $50 million Tehachapi 8 MWe/32 MWh lithium-ion battery storage project in conjunction with a 4500 MWe wind farm, using 10,872 modules of 56 cells each from LG Chem, which can supply 8 MWe over four hours. A lithium-ion battery storage system of 500 kWh and delivering 2 MWe is operating in the UK, on the Orkney Islands. This Kirkwall power station uses Mitsubishi batteries in two 12.2m shipping containers, and stores power from wind turbines. In Germany, a 10 MWe/10.8 MWh lithium-ion battery storage system was commissioned in 2015 at Feldheim, Brandenburg. It has 3360 Li-ion modules from LG Chem in South Korea. The €13 million battery unit stores power generated by a local 72-MWe wind farm and was built to stabilise the grid of TSO 50Hertz Transmission. It also participates in the weekly tendering for primary control reserve.
In May 2016 Fortum in Finland contracted French battery company Saft to supply a €2 million megawatt-scale lithium-ion battery energy storage system for its Suomenoja power plant as part of the largest ever BESS pilot project in the Nordic countries. It will have a nominal output of 2 MWe and 1 MWh of capacity, to be offered to the TSO for frequency regulation and output smoothing. It is similar to the system operating in the Aube region of France, linking two wind farms, total 18 MWe. Saft has deployed over 80 MWe of batteries since 2012.
The 98 MWe Laurel Mountain wind farm in West Virginia employs a multi-use 32 MWe/8 MWh grid-connected BESS. The plant is responsible for frequency regulation and grid stability in the PJM market as well as arbitrage. The lithium-ion batteries were made by A123 Systems, and when commissioned in 2011 it was the largest Li-ion BESS in the world.
US generator AES has completed a 10 MWe/5 MWh energy storage array at its Kilroot power station in Carrickfergus, Northern Ireland. The system consists of over 53,000 Li-ion batteries arranged in 136 separate nodes with control system which responds to grid changes in under a second. It is the largest advanced energy storage system in the United Kingdom and Ireland, and the only such system at transmission scale according to AES. The company wants to build the storage array up to 100 MW, providing £8.5 million in system savings per annum “by displacing out of merit thermal back up plant and facilitating fuller integration of existing renewables,” it said.
In December 2015 EDF Renewable Energy commissioned its first BESS project in North America, with 40 MWe flexible (20 MWe nameplate) capacity on the PJM grid network in Illinois to participate in the regulation and capacity markets. The lithium-ion batteries and power electronics were supplied by BYD America, and consist of 11 containerized units totaling 20 MWe. The company has more than 100 MWe of storage projects under development in North America.
SolarCity is using Tesla’s 52 MWh Powerpack lithium-ion storage system for its 13 MW Kaua’i Island solar PV project in Hawaii, to meet evening peak demand. Power is supplied at 14.5 cents/kWh.
Toshiba has supplied a large BESS for Hamilton, Ohio, comprising an array of 6 MWe/2 MWh Li-ion batteries. Lifetime of over 10,000 charge-discharge cycles is claimed.
In the UK, Statoil has commissioned the design of a 1 MWh lithium-ion battery system, Batwind, as onshore storage for the 30 MWe offshore Hywind project. From 2018 it is to store excess production, reduce balancing costs, and allow the project to regulate its own power supply and capture peak prices through arbitrage.
RedFlow has a range of zinc bromide flow battery modules (ZBM) which can be installed in connection with intermittent supply and is capable of daily deep discharge and charge. They are more durable than Li-ion type, and expected energy throughput for smaller ZBM units ranges to 44 MWh. Large-scale battery (LSB) units comprise 60 ZBM-3 batteries that deliver peak 300 kW, continuous 240 kW, at 400-800 volts and supply 660 kWh.
Duke Energy is testing a hybrid ultracapacitor-battery storage system (HESS) in North Carolina, close to a 1.2 MW solar installation. The 100 kW/300 kWh battery uses aqueous hybrid ion chemistry with salt water electrolyte and synthetic cotton separator. The rapid-response ultracapacitors smooth the load fluctuations.
Lower-cost lead-acid batteries are also in widespread use at small utility scale, with banks of up to 1 MWe being used to stabilise wind farm power generation. A 0.5 MWe Purewave Storage Management system with 1280 advanced lead-acid batteries was commissioned in September 2011 at PNM's Mesa Del Sol, Albuquerque New Mexico, by S&C Electric Co. The GS Batteries are capable of up to 4000 deep discharge cycles.
Stanford University is developing an aluminium-ion battery, claiming low cost, low flammability and high-charge storage capacity over 7500 cycles. It has an aluminium anode and graphite cathode, with salt electrolyte, but produces only low voltage.
Ontario's ISO has contracted a 2 MWe zinc-iron redox flow battery from ViZn Energy Systems.
Avista Corp in the northwest USA Washington state is purchasing a 3.6 MWe vanadium flow battery to load balance with renewables.
(Flow batteries have two chemical components dissolved in liquids and separated by a membrane.)
In May 2015 Tesla announced a household battery storage unit of 7 or 10 kWh for storing electricity from renewables, using lithium-ion batteries similar to those in Tesla cars. It will deliver 2 kW and works at 350-450 volts. The Powerwall system would be sold to installers at $3000 for a 7 kWh unit or $3500 for 10 kWh, though the latter option was promptly discontinued and the former downrated to 6.4 kWh storage capacity and 3.3 kW power. While this is clearly domestic-scale, if widely taken up it will have grid implications. Tesla claims 15 c/kWh to utilize the storage, plus the cost of that renewable energy initially, with 10-year, 3650-cycle warranty covering diminishing output to 3.8 kWh at year five, 18,000 kWh total.
In the UK, Powervault supplies diverse batteries for household use, mainly with solar PV but also with a view to savings with smart meters. Its 4 kWh lead-acid battery is the most popular product at £2900 installed, although the actual batteries need replacing every five years. A 4 kWh Li-ion unit costs £3900 installed, and other products range from 2 to 6 kWh, costing up to £5000 installed.
Compressed air storage (CAES)
Energy storage with compressed air (CAES) in geological caverns is being trialled, using gas-fired or electric compressors, the adiabatic heat being dumped. When released (with preheating to compensate for adiabatic cooling) it powers a turbine, up to 300 MWe, with overall about 70% efficiency. CAES capacity can even out the production from a wind farm and make it partly dispatchable. Two CAES systems are in operation, in Alabama and Germany, and others trialled or developed elsewhere in the USA. CAES capacity in the USA was 450 MWe in 2010 and is expected to grow to 6000 MWe by 2020. Nearly one-third of the 1.5 GWe ‘battery storage’ in 2015 was CAES.
Duke Energy and three other companies are developing a 1200 MWe, $1.5 billion project in Utah, ancillary to a 2100 MWe wind farm and other renewable sources. This is the Intermountain Energy Storage Project, using salt caverns. It is targeting 48-hour duration for discharge, hence apparently over 50 GWh.
Toronto Hydro with Hydrostor has a pilot project using compressed air in bladders 55m underwater in Lake Ontario to yield 0.66 MWe over one hour.
Other electricity storage
As described above in the solar thermal subsection, some CSP plants use molten salt to store energy overnight. Spain's 20 MWe Gemasolar claims to be the world's first near base-load CSP plant, with 63% capacity factor. Spain's 200 MWe Andasol plant also uses molten salt heat storage, as does California's 280 MWe Solana.
In Germany Siemens has commissioned a 6 MW hydrogen storage plant using proton exchange membrane (PEM) technology to convert excess wind power to hydrogen, for use in fuel cells or added to natural gas supply. The plant in Mainz is the largest PEM installation in the world. In Ontario, Hydrogenics partnered with German utility E.ON to create a 2 MW PEM facility that came on in August 2014, turning water into hydrogen through electrolysis.
Ontario’s ISO has contracted for a 2 MWe flywheel storage system from NRStor Inc.
Another form of energy storage is ice. Ice Energy has contracts from Southern California Edison to provide 25.6 MW of thermal energy storage using its Ice Bear system, attached to large air conditioning units. This makes ice at night when power demand is low, then uses it to provide cooling during the day instead of the aircon compressors, thus reducing peak demand.
The US Department of Energy Global Energy Storage database has more information.
Renewables in relation to base-load demand
It is clear that renewable energy sources have considerable potential to meet mainstream electricity needs. However, having solved the problems of harnessing them there is a further challenge: of integrating them into the supply system. Obviously sun, wind, tides and waves cannot be controlled to provide directly either continuous base-load power, or peak-load power when it is needed, so how can other, dispatchable sources be operated so as to complement them?
If there were some way that large amounts of electricity from intermittent producers such as solar and wind could be stored efficiently, the contribution of these technologies to supplying electricity demand would be much greater – see following subsection. The only renewable source with built-in storage and hence dispatchable on demand is hydro from dams.
There is some scope for reversing the whole way we look at power supply, in its 24-hour, 7-day cycle, using peak load equipment simply to meet the daily peaks. Conventional peak-load equipment can be used to some extent to provide infill capacity in a system relying heavily on renewables. Its characteristic is rapid start-up, usually (apart from dammed hydro) with low capital and high fuel cost. Such capacity complements large-scale solar thermal and wind generation, providing power at short notice when they were unable to. This is essentially what happens with Denmark, whose wind capacity is complemented by a major link to Norwegian hydro (as well as Sweden and the north German grid).
Case study: West Denmark
West Denmark (the main peninsula part) is the most intensely wind-turbined part of the planet, with 1.74 per 1000 people - 4700 turbines totaling 2315 MWe, 1800 MWe of which has priority dispatch and power must be taken by the grid when it is producing. Total system capacity is 6850 MWe and maximum load during 2002 was 3700 MWe, hence a huge 81% margin. In 2002, 3.38 billion kWh were produced from the wind, a load factor of 16.8%. The peak wind output was 1813 MWe on 23 January, well short of the total capacity, and there were 54 days when the wind output supplied less than 1% of demand. On two occasions, in March and April, wind supplied more than total demand for a few hours. In February 2003 during a cold calm week there was virtually no wind output. Too much wind is also a problem - over 20 m/s output drops and over 25 m/s turbines are feathered. Generally, a one metre/second wind change causes a 320 MWe power change for the whole system.
However, all this can be and is managed due to the major interconnections with Norway, Sweden and Germany, of some 1000 MWe, 600 MWe and 1300 MWe respectively. Furthermore, especially in Norway, hydro resources can normally be called upon, which are ideal for meeting demand at short notice. (though not in 2002 after several dry years). So the Danish example is a very good one, but the circumstances are far from typical.
Case study: Germany
The 2006 report from a thorough study commissioned by the German Energy Agency (DENA) looked at regulating and reserve generation capacity and how it might be deployed as German wind generation doubled to 2015. The study found that only a very small proportion of the installed wind capacity could contribute to reliable supply. Depending on time of year, the gain in guaranteed capacity from wind as a proportion of its total capacity was between 6 and 8% for 14.5 GWe total, and between 5 and 6% for 36 GWe total projected in 2015. This all involves a major additional cost to consumers.
Case study: UK
The performance of every UK wind farm can be seen on the Renewable Energy Foundation web site. Note particularly the percentage of installed capacity which is actually delivering power averaged over each month.
If hydro is the back-up and is not abundant, it will be less available for peaking loads. If gas is the back-up it this will usually be the best compromise between cost and availability. But any conventional generating plants used as back-up for wind and solar renewables has to be run at lower output than designed to accommodate the intermittent input then the lower capacity factor can make them uneconomic, as is now being experienced with many GWe of gas and coal capacity in Germany. This incidentally has adverse CO2 emission implications. (See sections below).
In practical terms non-hydro renewables are therefore able to supply up to some 15-20% of the capacity of an electricity grid, though they cannot directly be applied as economic substitutes for most coal or nuclear power, however significant they become in particular areas with favourable conditions. Nevertheless, they make an important contribution to the world's energy future, even if they cannot carry the main burden of supply. The Global Wind Energy Council expects wind to be able to supply between 10.8 and 15.6% of global electricity by 2030.
A 2005 report on wind energy in Germany by E.On, the country's largest grid operator, pointed out that: "As wind power capacity rises, the lower availability of the wind farms determines the reliability of the system as a whole to an ever increasing extent. Consequently the greater reliability of traditional power stations becomes increasingly eclipsed. As a result, the relative contribution of wind power to the guaranteed capacity of our supply system up to the year 2020 will fall continuously to around 4%. In concrete terms, this means that in 2020, with a forecast wind power capacity of over 48,000 MW in Germany (Source: DENA grid study), 2,000 MW of traditional power production can be replaced by these wind farms." Hence "traditional power stations with capacities equal to 90% of the installed wind power capacity must be permanently online in order to guarantee power supply at all times." Wind energy cannot replace conventional power stations to any significant extent, and this has been borne out by E.On's experience since that report. In the event, almost 45,000 MWe of wind capacity was installed by the end of 2015, operating at about 22% capacity factor.
In 2014 the OECD International Energy Agency (IEA) published a report on this issue: The Power of Transformation, wind, sun and the economics of flexible power systems. It says that the cost-effective integration of variable renewable energy (VRE) has become a pressing challenge for the energy sector. However, it finds that large shares of VRE can be accommodated in some circumstances, though usually cost-effective integration calls for a system-wide transformation. Each country may need to deal with different circumstances in achieving such a transformation. The study is based on seven case studies involving 15 different countries. It assesses four ‘flexible resources’ that enable VRE integration – flexible power plants (notably gas-fired), enhanced grid infrastructure, electricity storage (costly and inefficient) and demand-side integration (particularly distributed thermal storage).
At less than 10% VRE, integration poses few challenges, since this is within the range of natural variability of any system. But the study showed that annual VRE shares of 25% to 40% might be achieved from a technical perspective, assuming current levels of system flexibility and sufficient capacity in the system, and assuming that some curtailment of VRE output was accepted (rather than guaranteed priority access to grid for VRE). “However, mobilising system flexibility to its technical maximum can be considerably more expensive than least-cost system operation.” The study goes on to show that integrating large shares of VRE really requires system-wide transformations and results in higher costs.
Simply adding 45% VRE to a normal system resulted in system costs increasing by 40%, though this modeling excluded diseconomies in established generating plants due to lower utilization. CO2 emission costs of $30 per tonne were factored in to the model (a $30/MWh burden on black coal). Adding the 45% (or 30%) while progressively closing down a lot of base-load plant, adding peaking capacity and boosting grid capacity cost only 13% (or 7%) more than the base case. With significant demand-side integration the system cost increment can be reduced further. However, the system is then deprived of the major part of its low-cost base-load capacity.
The study says that “in order to deal efficiently with short-term variability and uncertainty, market operations need to facilitate trading as close as possible to real-time.” This is antithetical to the economics of base-load capacity, where assured full-load operation is optimal, and this needs to be long-term ahead in order to justify large capital commitments required for new plant. So there is a fundamental difficulty relating investment in subsidized VRE capacity which has guaranteed priority grid access with investment in base-load capacity which requires guaranteed minimum prices and high grid access. According to the IEA study the only non-VRE investments should be in additional flexible resources – essentially gas-fired, and hence contributing to CO2 emissions.
Meanwhile Germany provides a case study in accelerated integration of VRE into a stable system, with both politically- and economically-forced retirement of conventional generating capacity. The IEA study also mentions Italy and Spain in this regard. It notes that economically-forced retirement of conventional plant “can raise concerns about security of supply.” Indeed.
In the USA over 1992-2013 a production tax credit (PTC) applied for wind, finally at $23/MWh net, compared with a wholesale price usually not much above that. Thus the PTC meant that intermittent wind generators could dump power on the market to the extent of depressing the wholesale price so that other generators were operating at a loss. This market distortion has created major problems for the viability of dispatchable generation sources upon which the market depends.
Intermittency and grid management
Grid management authorities faced with the need to be able to dispatch power at short notice treat wind-generated power not as an available source of supply which can be called upon when needed but as an unpredictable drop in demand. In any case wind needs about 90% back-up, whereas the level of back-up for other forms of power generation which can be called upon on demand is around 25%, simply allowing for maintenance downtime.
The OECD IEA's World Energy Outlook 2010 considered the likely integration costs of variable renewables in 2035 under three headings: interconnection costs, balancing costs and maintaining supply adequacy. For the EU these together, for onshore and offshore wind, CSP and large scale PV, amounted to 1.63 c/kWh, and for USA 1.73 c/kWh.
Modeling done by the UK National Grid Corporation shows the effect of wind's unreliability on the required plant for achieving the 20% UK renewables target:
|Contribution from wind
% of 400 TWh
|Wind capacity GWe
||Conventional capacity GWe
||Spare capacity GWe
Thus, building 25 GWe of wind capacity, equivalent to almost half of UK peak demand, will only reduce the need for conventional fossil and nuclear plant capacity by 6.7%. Also, some 30 GWe of spare capacity will need to be on immediate call continuously to provide a normal margin of reserve and to back up the wind plant's inability to produce power on demand - about two thirds of it being for the latter.
Ensuring both secure continuity of supply (reliably meeting peak power demands) and its quality (voltage and frequency control) means that the actual potential for wind and solar input to a system is limited. Doing so economically, as evident from the above UK figures, requires low-cost back-up such as hydro, or gas turbine with cheap fuel. For the UK, with little interconnection beyond its shores, a 20% renewables target is difficut.
In a March 2004 report Eurelectric and the Federation of Industrial Energy Consumers in Europe pointed out what is now evident, that "Introducing renewable energy unavoidably leads to higher electricity prices. Not only are production costs substantially higher than for conventional energy, but in the case of intermittent energy sources like wind energy, grid extensions and additional balancing and back-up capacity to ensure security of supply imply costs which add considerably to the end price for the final consumer." "Reducing CO2 by promoting renewable energy can thus become extremely expensive for consumers," though both organisations fully support renewables in principle. The economic disadvantage referred to will also be reduced as carbon emission costs become factored in to fossil fuel generation.
Because wind turbine output is so variable, for planning purposes its potential output is discounted to the level of power that can be relied upon for 90% of the time. In Australia that figure comes to 7% of installed wind capacity, in Germany it is 8%, which is all that can be included as securely available (ie 90% of the time).* On the 90% availability basis, other technologies can be counted on for much higher reliability, and hence the investment cost per kilowatt reliably available is much less.
* Figures from NEMMCO and E.ON respectively.
A 2006 report by the UK Energy Research Council looked at the system implications and costs of intermittent inputs from renewables whose variability was uncontrollable. It found that intermittent sources meeting up to 20% of electricity demand need not compromise reliability, but was likely to have a significant cost. The report looked at system balancing impacts, in managing fluctuations from system balancing reserves, and reliability impacts which affected ability to meet peak demand and also required a greater system margin (15-22% higher). It costed the former at £2-3/MWh and the latter at £3-5/ MWh - total £5-8 (0.5p to 0.8p/kWh) with 20% wind input.
Nuclear power plants are essentially base-load generators, running continuously. Where it is necessary to vary the output according to daily and weekly load cycles, for instance in France, where there is a very high reliance on nuclear power, they can be adapted to load-follow. For BWRs this is reasonably easy without burning the core unevenly, but for a PWR (as in France) to run at less than full power for much of the time depends on where it is in the 18 to 24-month refueling cycle, and whether it is designed with special control rods which diminish power levels throughout the core without shutting it down. So while the ability on any individual PWR reactor to run on a sustained basis at low power decreases markedly as it progresses through the refueling cycle, there is considerable scope for running a fleet of reactors in load-following mode. Generation III plants have more scope for load-following, and as fast neutron reactors become more established, their ability in this regard will be an asset.
If electricity cannot be stored on a large scale, the next logical step is to look at products of its use which can be stored, and hence where intermittent electricity supply is not a problem.
System Integration Costs of Intermittent Renewable Power Generation
Power generation technologies generally compete with each other both in regulated and deregulated markets to supply electricity through a ‘merit order’ based on availability and marginal cost of production for any given period. Fossil fuel, nuclear, biomass and hydro power generators can all to varying degrees supply electricity ‘on demand’, in other words supply from these sources can be called upon or adjusted to meet demand. In contrast to renewable hydro, the feed-in of wind and solar output is uncontrollably intermittent due to the uncertainty of meteorological conditions. In grid management terms they are not dispatchable. Therefore the energy system needs backup capacity from the on-demand-sources to bridge periods with high or low generation from renewables. The targeted rapid increase of power supply from intermittent renewable sources in many countries presents a fundamental challenge to the smooth functioning of many electricity supply systems.
Wind and solar power are the forms of renewable power that are growing most rapidly. They accounted for 35% of EU renewable capacity in 2009, a percentage that the IEA in World Energy Outlook 2011 expected to increase to 55% in 2015 in its central 'new policies' scenario. By 2030 the IEA expects wind and solar to constitute 34% of total EU electrical capacity, up from 11% in 2009. However, the intermittency of wind and solar generation meant that the amount of electricity supplied from wind and solar capacity in 2009 was less than 5% of the total; 133 TWh was generated by EU wind turbines in 2009 which equated to a capacity factor of 20%. In the same year, nuclear had a capacity factor of 73%; thus to generate an equivalent amount of potential power to nuclear on the basis of these load factors, it is necessary to install three or four times as much wind capacity*. But that is not the main problem.
Wind and solar power supply is largely governed by wind speed and the level of sunlight, which can only loosely be related to periods of power demand. It is this feature of intermittent renewable power supply that results in the imposition of additional costs on the generating system as a whole, which will implicitly be paid for either by other generators, consumers or taxpayers.
The IEA disaggregates these system costs into three components:
- Adequacy costs: the cost of ensuring that the power system has sufficient capacity to meet peak loads.
- Balancing costs: the cost of ensuring that the power system can respond flexibly to demand changes at any given time.
- Interconnection costs: the cost of linking sources of supply to sources of demand.
Required adequacy and balancing capacity for intermittent renewables
As noted above, additional back-up capacity is needed to meet demand rapidly when meteorological conditions result in insufficient wind and solar power generation. The adequacy and balancing capacity must itself have a high degree of availability, ie, it should be from a dispatchable source. This reserve or backup capacity is most likely to be needed during periods of high demand and lack of wind and solar, for instance on a calm winter’s evening. In such a situation significant levels of dispatchable backup capacity are needed to ensure security of supply.
Estimates of the wind/solar reserve requirement suggest that the reserve capacity ratio is expected to increase exponentially in proportion to the reliance of the system on intermittent power generation*. Germany is planning to increase wind and solar input to 20% of the total by 2020 and about 35% by 2030**. The EU as a whole is targeting a 27% renewable contribution to power supply by 2030. The adequacy and balancing capacity requirements for such high levels of intermittent renewable penetration can only be estimated theoretically.
In World Energy Outlook 2010 the IEA estimates that adequacy and balancing costs for intermittent renewable power supply range from 0.18-0.8c/kWh of intermittent renewable power supply in the USA and from 0.13-0.65c/kWh in the EU in the New Policies Scenario in 2035*. The New Policies Scenario envisages that intermittent renewable power supply would constitute 13% of the 2035 total power supply in the US and 22% in the EU; these levels of penetration are relatively modest and the adequacy and balancing costs could be expected to increase significantly at higher levels of penetration.
IEA estimates* of generating (busbar) costs of wind in the USA and EU vary between 7.0-23.4c/kWh (solar is even more costly), so these adequacy and balancing cost estimates are equivalent to no more than 11% of the levelised wind generating cost. However, the costs of backup capacity clearly depend on the type of backup capacity envisaged. Pumped storage is often cited as an ideal renewable form of flexible backup but is relatively expensive (between 1.1-23.2 c/kWh); gas turbines would be generally cheaper, but still quite expensive given that they would be operating only for part of the time and therefore suffer low load factors (probably less than 20%). Gas turbines would also emit greenhouse gases as it would likely be prohibitively expensive to install CCS for sources with such low load factors. It is possible that where supply reliability cannot be guaranteed, many electricity users placing a premium on reliability (eg, hospitals) will invest in expensive local generating capacity (typically diesel generators).
Interconnection costs for wind and solar
The third category of intermittent renewable integration cost is grid interconnection. Wind and solar farms are ideally sited in areas that experience high average wind speeds and high average solar radiation respectively. These sites are often, even typically, distant from areas of electricity demand. Transmission and distribution networks will often need to be extended significantly to connect sources of supply and demand - this is a current challenge in UK and North Germany. The IEA estimates these costs at 1.2 c/kWh in the USA and 0.9 c/kWh in the EU, which again must be recovered either from other producers, consumers or taxpayers.
Impacts of wind and solar on base-load generators
The three categories of intermittent renewable integration costs are estimated by the IEA to vary approximately in sum from 1.1 to 1.7 c/kWh in the EU and 1.3 to 1.9 c/kWh in the USA. As footnoted above, these estimates are theoretical and pertain to the penetration of intermittent renewables forecast by the IEA in their ‘New Policies Scenario’ for 2035.*
However, there is another category of costs that result from the operation of renewables and these can be described as the external costs borne by other power producers, in particular base-load power producers, as a result of intermittency. The structure of wind and solar levelised generation costs is characterised by high capital, significant O&M costs and zero fuel costs. As a result, the operating costs for these sources are very low and when power is generated they undercut and are able to displace all other sources of power in a utility’s merit order. In a situation of high levels of wind and solar power penetration and during periods of low demand, baseload generators will be displaced in the merit order. This is also required by legislation in many countries.
The impact of high levels of intermittent, low cost power will be to reduce the load factors of base-load power generators, and thereby increase their unit costs per kilowatt-hour. Given the high capital costs of nuclear, such an impact will significantly increase the levelised generation costs of nuclear. For example, a 15% decrease in the capacity factor of a nuclear power plant could increase its levelised cost by about 24%. In a situation similar to that targeted by Germany for 35% of supply to be intermittent renewable by 2030, renewable capacity when fully utilised would provide up to100% of supply. Inevitably there would be periods when a great deal of base-load capacity would be forced off-line.
The Hydrogen Economy
Hydrogen is widely seen as a possible fuel for transport, if certain problems can be overcome economically. It may be used in conventional internal combustion engines, or in fuel cells which convert chemical energy directly to electricity without normal burning.
Making hydrogen requires either reforming natural gas (methane) with steam, or the electrolysis of water. The former process has carbon dioxide as a by-product, which exacerbates (or at least does not improve) greenhouse gas emissions relative to present technology. With electrolysis, the greenhouse burden depends on the source of the power.
With intermittent renewables such as solar and wind, matching the output to grid demand is very difficult, and beyond about 20% of the total supply, apparently impossible. But if these sources are used for electricity to make hydrogen, then they can be utilised fully whenever they are available, opportunistically. Broadly speaking it does not matter when they cut in or out, the hydrogen is simply stored and used as required.
A quite different rationale applies to using nuclear energy (or any other emission-free base-load plant) for hydrogen. Here the plant would be run continuously at full capacity, with perhaps all the output being supplied to the grid in peak periods and any not needed to meet civil demand being used to make hydrogen at other times. This would mean maximum efficiency for the nuclear power plants, and that hydrogen was made opportunistically when it suited the grid manager.
About 50kWh is required to produce a kilogram of hydrogen by electrolysis, so the cost of the electricity clearly is crucial.
Renewable energy sources have a completely different set of environmental costs and benefits to fossil fuel or nuclear generating capacity.
On the positive side they emit no carbon dioxide or other air pollutants (beyond some decay products from new hydro-electric reservoirs), but because they are harnessing relatively low-intensity energy, their 'footprint' - the area taken up by them - is necessarily much larger.
Whether Australia could accept the environmental impact of another Snowy Mountains hydro scheme (providing some 3.5% of the country's electricity plus irrigation) is doubtful. Whether large areas near cities dedicated to solar collectors will be acceptable, if such proposals are ever made, remains to be seen. Beyond utilising roofs, 1000 MWe of solar capacity would require at least 20 square kilometres of collectors, shading a lot of country.
In Europe, wind turbines have not endeared themselves to neighbours on aesthetic, noise or nature conservation grounds, and this has arrested their deployment in UK. At the same time, European non-fossil fuel obligations have led the establishment of major offshore wind forms and the prospect of more.
However, much environmental impact can be reduced. Fixed solar collectors can double as noise barriers along highways, roof-tops are available already, and there are places where wind turbines would not obtrude unduly.
APPENDIX: Government Support for Renewables Deployment
In an open market, government policies to support particular generation options such as renewables normally give rise to explicit direct subsidies along with other instruments such as feed-in tariffs, quota obligations and energy tax exemptions. In the EU, feed-in tariffs are widespread.
Corresponding to these in the other direction are taxes on particular energy sources, justified by climate change or related policies. For instance Sweden taxes nuclear power at about EUR 0.6 cents/kWh.
European Environment Agency figures in 2004 gave indicative estimates of total energy subsidies in the EU-15 for 2001: solid fuel (coal) EUR 13.0, oil & gas EUR 8.7, nuclear EUR 2.2, renewables EUR 5.3 billion.
The Global Wind Energy Council (2008) reported that "In the pursuit of the overall target of 21% from renewable electricity by 2010, the Renewable Electricity Directive 2001 gives EU Member States freedom of choice regarding support mechanisms. Thus, various schemes are operating in Europe, mainly feed-in tariffs, fixed premiums, green certificate systems and tendering procedures. These schemes are generally complemented by tax incentives, environmental taxes, contribution programs or voluntary agreements."
France had a feed-in tariff of EUR 8.2 c/kWh to 2012, which then woiuld decrease.
Germany's Renewable Energy Sources Act gives renewables priority for grid access and power dispatch. It is regularly amended to adapt feed-in tariffs to market conditions and technological developments. For wind energy an initial tariff applies for up to 20 years and this then reduces to a basic tariff of EUR 5.02 c/kWh. The initial tariff is EUR 9.2 c/kWh for onshore wind and 15 c/kWh for offshore wind from January 2009. The combined subsidy from consumers and government totals some €5 billion per year – for 7.5% of its electricity.
Denmark has a wide range of incentives for renewables and particularly wind energy. It has a complex 'Green Certificate' scheme which transfers the subsidy cost to consumers. However, there is a further economic cost borne by power utilities and customers. When there is a drop in wind, back-up power is bought from the Nordic power pool at the going rate. Similarly, any surplus (subsidised) wind power is sold to the pool at the prevailing price, which is sometimes zero . The net effect of this is growing losses as wind capacity expands.
Italy in 2008 legislated to provide EUR 18 c/kWh on a quota system for wind power.
Spain has different levels of feed-in tariffs depending on the technology used. A fixed tariff of EUR 7.32 c/kWh is one option, or a fixed premium of 2.93 c/kWh on the market price (but with a floor of 7.13 cents) is the other, as of 2008. The tariffs for renewables are adjusted every four years.
Greece has a feed-in tariff of 6.1-7.5 c/kWh, whereas the Netherlands relies on exemption from energy taxes to encourage renewables.
The UK has not used any feed-in tariff arrangement, but is to do so from 2010. Meanwhile a specific indication of the cost increment over power generation from other sources is given by the 4.5-5.0 p/kWh market value for the Renewables Obligation, by which utilities can cover the shortfall in producing a certain proportion of their electricity from renewables by paying this amount and passing the cost on to the consumer. In addition there is a Climate Change Levy of 0.43 p/kWh on non-renewable sources (at present including nuclear energy, despite its lack of greenhouse gas emissions), which corresponds to a subsidy.
Sweden subsidises renewables (principally large-scale hydro) by a tax on nuclear capacity, which works out at about EUR 0.67 cents/kWh from 2008. For wind, there is a quota system requiring utilities to buy a certain amount of renewable energy by purchasing certificates.
In Norway the government subsidises wind energy with a 25% investment grant and then production support per kWh, the total coming to NOK 0.12/kWh, against a spot price of around NOK 0.18/kWh (US$ 1.3 cents & 2 cents respectively).
In the USA the wind energy production tax credit (PTC) of 1.5 c/kWh indexed to inflation (now about 2.1 c/kWh) has provided incentive, though this expires every two years before being renewed by Congress.
Canada provides a production incentive payment of 1c/kWh for wind power, plus feed-in tariffs.
In Australia energy retailers are required to source specified quantities of power from new (non hydro) renewables. The obligation is tradeable and there is a fallback tax of AUD 4 c/kWh for retailers failing to comply.
In India ten out of 29 states have feed-in tariffs, eg 2.75 times the tariff for coal-generated power in Karnataka, plus a federal incentive scheme paying one third of the coal-fired tariff.
Small-scale PV input is encouraged by high feed-in tariffs, eg 48 c/kWh in Germany and 50 c/kWh in Portugal.
Boyle, G (ed), 1996, Renewable Energy – Power for a Sustainable Future, Open University, UK
OECD IEA (1987) Renewable Sources of Energy
European Wind Energy Association + Greenpeace (2002), Wind Force 12.
Duffey & Poehnell, 2001, Hydrogen production, nuclear energy & climate change, CNS Bulletin 22,3.
Laughton, M.A. 2002, Renewables and the UK Electricity Grid Supply Infrastructure, Platts Power in Europe.
Sharman, H. 2003, Danish Lessons, www.countryguardian.net
Eurelectric 25/3/04, www.eurelectric.org
OECD IEA 2008, 2009, 2010 World Energy Outlook
OECD NEA 2012, Nuclear Energy and Renewables – system effects in low-carbon electricity systems.
UK Energy Research Centre, 2006, The Costs and Impacts of Intermittency.
E.On Netz Wind report 2005
DENA 2006 Grid Study.
Global Wind Energy Council, 2008 Annual Report
REN21 Renewables Global Status Report, 2013
International Energy Agency, Energy Technology Perspectives 2016
Re biofuel wood pellet export from the USA to Europe