ZCA2020 Stationary Energy Synopsis v1

Embed Size (px)

Citation preview

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    1/17

    ENERGY

    > A ten year roadmap for 100% renewable energy> Baseload energy supplied by renewable sources

    > Affordable at $8 per household per week

    Australian Sustainable Energy

    Zero Carbon AustraliaStationary Energy PlanSypsis

    A researchcollaboration

    Markel redondo

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    2/17

    | 2Sypss ZCa2020 Sttiy egy P

    Pt 1: Itcti vviw

    The world stands on the precipice of significant change.Climate scientists predict severe impacts from even

    the lowest estimates of global warming. AtmosphericCO2 already exceeds safe levels. A rational response tothe problem demands a rapid shift to a zero-fossil-fuel,zero-emissions future. The Zero Carbon Australia 2020Stationary Energy Plan (the ZCA2020 Plan) outlines atechnically feasible and economically attractive way forAustralia to transition to 100% renewable energy within tenyears. Social and political leadership are now required inorder for the transition to begin.

    This document is a synopsis of the ZCA2020 Plan. For fullreferences and analysis, see the ZCA2020 Plan full report.

    With a combination of energy efficiency, fuel switchingfrom gas and oil to electrified energy services, then usinga combination of commercially available renewable energytechnologies, Australias energy needs can be met with 100%renewables. Wind and Concentrating Solar Thermal (CST)with Molten Salt Storage are the two primary technologiesused, with minimal contingency backup from biomass andexisting hydro. Modelling on a half-hourly timescale showsthat this combination can ensure 100% reliable supply.Implementing the proposed renewable infrastructure overa ten-year timescale would require a small percentage ofAustralias industrial capacity, in terms of resources andlabour force. The required investment of $37 Billion/year

    is the equivalent of 3% of GDP. Estimated funding cost,including generation and grid upgrades, would be equivalentto an additional 6.5cents/kWh on delivered electricity.

    Why w P f z missis?

    We know that climate change is a real threat, and weknow what to do about it. The cause of the problem isthat the level of atmospheric CO2 is already too high. Foratmospheric CO2 to stabilise, human activities must stopemitting greenhouse gases particularly from our use offossil fuels for energy.

    The threat comes from dangerous tipping pointmechanisms, which can be triggered by excessivetemperatures and would prevent us from returning to safeclimate conditions. To avoid this threat, we need to reduceatmospheric CO2 from the present level of 390 ppm towell below 350 ppm significantly closer to pre-industrialconcentrations of 285 ppm.

    Evidence of a tipping point is the sudden decline in Arcticsea ice, which has accelerated beyond the worst-caseforecasts of the IPCC (Figure 1).

    To avoid more severe risks, such as that presented by themelting of sub-artic permafrost, our path is clear - we need

    to attain a zero-emissions economy. That transition is thefocus of the ZCA2020 Project. The aim of the Project asa whole is to outline how each sector of the Australian

    economy can achieve zero or negative greenhouseemissions: in Stationary Energy, Transport, Buildings,Industrial Processes, and Land Use. This Stationary EnergyPlan is the first of the installments.

    Why t ys?

    The premise of a ten-year transition is based on The BudgetApproach from the German Advisory Council on GlobalChange. In order to have a 67% chance of keeping globalwarming below 2oC above pre-industrial temperatures, on abasis of equal allocation of emissions on a per-capita basis, itwould be necessary for the USA to reduce emissions to zeroin 10 years. Australia has about the same per-capita emissionsas the USA, and would need to pursue the same goal.

    Figure 2 shows a global budget per-capita (the blueblock in the background), which limits emissions across allpopulations of the world to 110 tonnes of CO2 per-capita(2.75 tonnes per capita per annum over 40 years) from

    2010-2050. With decisive action beginning now, the targetof zero emissions by 2020 for high emitting countries is arealistic and necessary goal.

    Should the start of transition be deferred, then the Planstill serves as a blueprint that could see 100% renewableenergy achieved at a later date. However delays pose anever greater risk of exceeding safe limits within the climatesystem.

    Giig pmt: cmmciy-vib-w

    To make an immediate transition, we can use only thesolutions that are available to us today. The ZCA2020 Plan

    has specified the 100% renewable grid based on proventechnologies that are already commercially available andthat have already been demonstrated in large industries.

    FIGure 1

    actic S Ic extt

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    3/17

    | 3Sypss ZCa2020 Sttiy egy P

    The ZCA2020 Plan is only one scenario in which 100%renewable energy could be achieved, not necessarily the onlyway, and is not dependent upon promising technologies thatare on the horizon. The choices for the Plan were made inorder for the proposed scenario to be rigorously assessed,and to show that all the solutions necessary are alreadyavailable. Should new zero-emissions technologies becomeviable, cost-competitive and available by 2020, their inclusionmay reduce costs and increase benefits even further.

    Pt 2: dsigig th systm

    Under the ZCA2020 Plan, it is projected that 2020 gridelectricity demand will be over 40% higher than today, from228TWh/yr up to 325TWh/yr. This increase in electricitydemand is due largely to the switch of services currentlyprovided by gas and oil to be supplied more efficiently, withrenewable electricity. The increase is significantly offsetby ambitious but achievable electrical energy efficiencytargets. This is more electricity demand than would berequired under Business-As-Usual growth.

    To meet this demand, a combination of wind and solarthermal with storage are proposed as the primary electricitygeneration technologies. Biomass and hydro are utilisedas contingency, supplying about 2% of annual electricity.This limited use is due to capacity constraints imposed byenvironmental considerations.

    Fcstig m

    There are several key considerations in forecasting theevolution of demand up to 2020 and beyond: Business-As-Usual projections of energy consumption,

    driven by population growth, and growth in per-capitaconsumption; Energy-efficiency improvements for existing electricity

    use, a cost-effective strategy that reduces the need toincrease installed generating capacity;

    A fuel switch from gas and oil for end-use applications torenewable electricity. Electrical systems for heating and

    transport can have significantly higher efficiency; Improved thermal insulation for buildings to flatten

    demand peaks and seasonal variations in demand,reducing the need for peak generators.

    Energy supplied from onsite generation, which displacesgrid electricity.

    Replacing inefficient fossil fuels and introducing energyefficiency measures decreases total energy consumptionin Australia by more than 50%, from 3,915 PJ/yr in 2008to 1,660 PJ/yr in 2020 (Figure 3). This is accompanied bya 40% increase in electricity demand, from the present228 TWh/yr to 325 TWh/yr. The provision of energy

    services shows no decline, and may even increase, a resultachievable through significantly improved energy efficiencythroughout the system.

    egy fficicy

    Australia is an energy-intensive country, both in terms ofenergy consumption per capita and per dollar of GrossDomestic Product (GDP). An efficiency gain of 33% percapita is assumed compared to current energy services.This is an achievable target that will bring Australia in linewith other modern economies.

    The ZCA2020 Plan calls for energy-efficiency measures toprogressively reduce electricity used for current services.Examples include efficient appliances, improved building

    FIGure 2

    Gb Cb Bgt f emissis

    18

    20

    rm n

    8

    10

    12

    14

    16

    CO2percapita

    USA (& Australia)

    China

    India

    Burkina Faso

    Global budget per

    capita assuming

    constant annual

    emissions

    0

    2

    4

    6

    2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

    t

    FIGure 3

    asti egy Scs: Pst uZCa2020

    2008 ZCA2020

    0

    500

    1000

    1500

    2000

    2500

    3000

    3500

    4000

    4500

    Oil & LPG

    Industrial Coal

    Natural Gas

    Bioenergy

    Direct Solar

    Electricity

    PJ/y

    ear

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    4/17

    | 4Sypss ZCa2020 Sttiy egy P

    design, retrofitting insulation, double- and triple-glazing, aswell as improved industrial efficiency.

    In Figure 4, the red curve shows Australias per capitaelectricity use under Business-As-Usual, projected tocontinue rising over the next 20 years. The yellow curveshows Germanys per capita use, which is currently 30%less than Australia, will reduce over the next 20 years

    due to ongoing official government programs. Germanyis a modern industrial economy, with a high standard ofliving, manufacturing and metals production (including fivealuminium smelters). The blue curve shows the effect ofproposed energy efficiency measures under the ZCA2020Plan, of 3.5-4% per capita per year until 2020.

    F-switch fm i gs t cticity

    The switch of end-use applications from gas and oil toelectricity yields substantial efficiency gains. These resultin a reduction in total energy use, accompanied by alesser increase in renewable electricity use (Figure 5). Key

    switched services include: Low-temperature gas heating (e.g. space and hot water) is

    replaced with highly efficient electrical heat pump heating,bringing an average gain in energy efficiency of around 3:1;

    High-temperature heating applications (e.g. cooking,industrial processes) can be converted to inductionor electric resistance heating. This is typically moreefficient, with no losses from hot flue gases caused byfuel combustion;

    Replacement of oil with electricity for cars, freightrail, and passenger rail transport brings a high gainin overall efficiency. This occurs partly through theinherent efficiency of electric motors (>80%) relative to

    internal combustion engines (

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    5/17

    | 5Sypss ZCa2020 Sttiy egy P

    the CST plants to accumulate more stored energy forrelease at night.

    Hyctic gts. Existing hydroelectricgenerators have a useful role in providing first-dispatchbackup and peaking power.

    Bimss ctigcy. Backup from crop-residualbiomass for the CST plants offers energy security,protecting against occasions when a combination oflow wind and low daily solar radiation occurs.

    lif-Cyc emissis

    Life-cycle analysis of energy technologies measures thetotal CO2-e emissions resulting from the construction of theplant, its continuing operation and the final decomissioning.

    Wind generators and solar thermal generators have thelowest life-cycle emissions of all available technologies.Coal Carbon Capture and Storage, in contrast, has life-cycleemissions at least 25 times larger than CST. Similarly,average estimates of nuclear power life cycle emissionsare several times larger than CST due mainly to fuel mining,processing, transporting and handling.

    In addition, both Coal Carbon Capture and Storage as well asNuclear plants take considerably longer to commission thanwind and CST plants. This lengthens the time span overwhich the old fossil fuel plants must continue to operateand emit.

    Cmpt aud$, B

    CST $175Backup Heaters $8

    Bioenergy supply $6

    Wind $72

    Transmission $92

    ToTal $353

    Off-grid CST + Backup $17

    ToTal + offgi $370

    TaBle 1

    Ivstmt cst f ZCa2020 wb gi

    FIGure 6

    Pps Pw Gi f 100% rwb ecticity

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    6/17

    | 6Sypss ZCa2020 Sttiy egy P

    Pt 3: astis 100% wbgy sppy

    Australias projected on-grid electricity demand in 2020 is325TWh/yr. The ZCA2020 plan proposes that this can beprovided through a combination of renewable power sources.

    40% of this electricity (130TWh/yr), will be supplied fromwind power, which is to be provided through 50,000MW of wind turbine capacity. The rest of the electricitywill be supplied from 42,500 MW of Concentrating SolarThermal (CST) with storage, providing reliable, 24-hourdispatchable power through molten salt thermal storage.The investment costs of these technologies have beenbased on expected cost reductions from significant scaleup of the renewable industry, coming to a total cost of$370 billion for the ten-year Plan, as shown in Table 1.

    The plants are located at sites around Australia that areselected for their wind availability, solar incidence, abilityto support economies of scale, transmission costs, andgeographical diversity giving 23 sites for wind, and 12sites for CST (Figure 6). Extra generation capacity is alsocosted for remote off-grid power generation, as shownin Table 1.

    Contingency will be provided by existing hydropowercapacity, and from biomass fired heaters attached tosome of the CST plants.

    The supply of Tasmanias electricity demand is includedin this scenario, however future work will separatelydetail Tasmanian wind installations.

    Ccttig s thm with stg (CST)

    There are four main types of CST systems: power tower,dish, linear fresnel and parabolic trough systems. Powertowers are selected for their high temperatures andefficiency, high year-round solar collection performance,and easy integration with molten salt thermal storage.

    Solar thermal power towers with storage have been provenat commercial scale during three years of operation of theSolar Two tower in the USA, from 1996 to 1999. There are

    a number of companies building and operating utility-scalesolar thermal with storage around the world, includingTorresol, SolarReserve and ACS Cobra.

    The proposed CST power towers consists of a central towerreceiver surrounded by a field of 18,000 heliostats, as shownin Figure 7. The heliostats are two-axis tracking mirrorsthat follow the sun, reflecting focused sunlight on thereceiver heating it to at least 565oC. Importantly, the ability

    FIGure 7

    S Thm Pw Tw

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    7/17

    | 7Sypss ZCa2020 Sttiy egy P

    of heliostats to track both the daily passage and seasonalelevation of the suns position in the sky means they haveincreased solar energy collection efficiency compared tosystems with less tracking ability.

    The receiver is similar to conventional boiler tubes, exceptthat instead of water as the working fluid, molten salt is used.The molten salt flows through the receiver accumulatingthermal energy from the focus and then flows into the hot salttank (shown in red). When there is a demand for electricity,the hot molten salt is passed through a heat exchangerto generate high-temperature steam for the turbine. Thesteam spins the turbine to generate the required amountof electricity. The heat extracted from the molten salt coolsit to 290oC (still molten), and it is returned to the cold tank(shown in yellow) where it waits to be pumped up the toweragain for heating.

    These power towers are able to operate at 70-75%annual capacity factor, similar to conventional fossil fuelplants. They are capable of dispatching power 24 hours aday, and are as reliable as conventional baseload power.Furthermore, they are in fact more flexible to meetingvarying demand. The steam heat-exchange and turbinesystems are specifically designed for rapid start up, allowingCST to balance changing electricity demand patterns, andproviding firming power to more variable wind generation.

    Sc f th CST its

    Each of the 12 CST sites has an operational capacity of just

    over 3,500MW. The CST power towers are optimally sizedas Solar 220 generating units, rated at 217 MW with aircooling. Thirteen Solar 220 units are to be installed at eachsite, along with an allowance for smaller rated modules ininitial years of scaleup.

    CST sit ctis

    CST is suited to large installations in areas of high solarincidence, and as it can store energy, it is less affectedby daily weather patterns as are more variable renewable

    energy sources.As the locations with high solar radiation are typicallyinland, at a distance from Australias coastal centreswith high electricity demand, the CST plants require highvoltage transmission to connect them to the locations ofdemand. Building a smaller number of high-power-ratingtransmission cables is more economical than a largenumber of lower power transmission cables. This has beentaken into account in determining the number of solar sitesand their locations.

    Furthermore, the impact of seasonal solar variations canbe mitigated by the choice of geographically diverse sitesfor CST plant locations. The annual average daily insolationcollected across all of the 12 sites is 7.95 kWh/m2/day, withwinter-time energy availability 25% below average. Thesolar thermal capacity is sized to make sure that demandduring the winter is met.

    l wt s

    Each 3,500MW CST site requires an area of 230 km2 theequivalent of just over 15km x 15km. This would ideally besituated in areas of marginal farming land. In total, the landuse for CST sites would be 2,760km2, less than 0.04% of

    Australias total land mass. It is less than the land areaof the largest single cattle station, or about the size ofKangaroo Island.

    Water use is minimised by using air cooling. This reducesthe rated plant output by 1.3%, but requires only 12% of the

    0.0

    2.0

    4.0

    6.0

    8.0

    10.0

    12.0

    14.0

    2004

    Near Term

    2006 2010

    Mid Term

    2015 2020

    Long Term

    centsperkWh

    S&L-Trough

    S&L - Tower

    Sunlab - Tower

    Sunlab -Trough

    6.2 cents/kWh

    4.3 cents/kWh

    5.5 cents/kWh

    3.5 cents/kWh

    FIGure 8

    Cst rcti Tjcty f Ccttig S Thm. Vs i uSd. (Sgt & ly Cstig, llC)

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    8/17

    | 8Sypss ZCa2020 Sttiy egy P

    water of a conventional power station. The total CST waterconsumption is projected to be 0.4% of Australias totalwater consumption, significantly less than the water usedby existing fossil fuel plants. A preliminary study of waterresource availability at each of the 12 sites has found thatthere is ample water available for the proposed CST usageof 6.3GL/yr per site. Even in Kalgoorlie, mining activitiescurrently use twice this.

    Tt cst f CST pts

    The total cost of the CST plants is $190 Billion, as shownin Table 2. The initial power plants are priced in line withdetailed current project costs in the United States, with laterinstallations becoming cheaper following the cost reductiontrajectory as projected by reports from the U.S. Departmentof Energys solar thermal program.

    As detailed by the cost curve from this report (Figure 8)

    it is projected that at an installed power tower capacity of8,700 MW, the price of electricity will drop to 3.5c/kWh (US,

    2003 dollars). This translates to about 5-6c/kWh in todaysAustralian dollars, which is competitive with the price ofconventional coal power. The key factors in achieving thesecost reductions are economies of scale of construction (withan ongoing pipeline of projects) and mass-manufacture of

    heliostat mirrors.

    Wi pw tis

    The ZCA2020 Plan proposes that 40% of Australias totalestimated electricity demand be met by wind power. TheDanish target for wind power is 50% of annual electricityby 2025. The Australian grid has greater geographicaldiversity than the Danish grid, so a 40% level of wind powerfor Australia is considered conservatively low.

    Due to the variable nature of the wind resource, windturbines do not operate at full capacity all of the time. Basedon the annual energy generated by a particular wind farm,the equivalent percentage of the time operating at full ouputis called the capacity factor. In Australia capacity factorsfor operating wind farms are in the range of 30-35%. Weestimate an average capacity factor of 30% for proposedwind farms, as only a few wind farms will be located inthe best high wind resource areas. The ZCA2020 Planalso proposes new transmission lines to access high windresource areas that currently have low proximity to the grid.

    At a minimum, the reliable instantaneous output is 7,500MW from the wind generators - 15% of the combinedrated capacity (50,000 MW) of all wind generators

    installed across the country. This firm wind output is asreliable as conventional baseload power. The 15% figureis conservatively estimated based on detailed studiesof interconnected wind farms carried out in the U.S., bythe National Renewable Energy Laboratories. This studyfound that depending upon the levels of transmissioninterconnection, a grid of onshore wind farms produced14-27% of nameplate capacity with firm reliability. As ourwind farms are planned to operate at 30% annual averagecapacity factor, we can expect half of the system- wideelectricity produced from wind to be firm.

    Sc f wi its

    To supply 130TWh/yr from wind power, 50,000MW of windcapacity is proposed. This requires the construction of anadditional 48,000MW of wind turbines on top of Australias2010 wind capacity. We propose that 6,400 high quality,technologically advanced 7.5MW wind turbines be utilised.E126 Enercon turbines are specified as they are currentlythe only commercially available 7.5MW turbines. Largeturbine sizes enables the extraction of more energy froma given site by tapping into stronger and more consistentwind resource at higher altitudes.

    Wi pw sits

    The Plan proposes 23 sites for wind power, eachconsisting of either 2,000 or 3,000 MW of turbines. Thegeographically dispersed wind sites exploit the diversityec e-126 Tbis i Bgim. Source: Steenki

    Phs Cst (2009 aud$)

    Phase one First 8700 MW $60 BillionPhase two 156 x Solar 220 $115 Billion

    Total Ongrid CST $175 Bii

    Off-grid CST 4,475 MW $15 Billion

    All CST sites for ZCA2020 + off-grid $190 Bii

    TaBle 2

    CST istti cst tb

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    9/17

    | 9Sypss ZCa2020 Sttiy egy P

    5 years of the Stationary Energy plan, the capital costs ofwind turbines are expected to transition to the forecast 2015European amount $1.65 Million/MW, based on estimatesfrom the European Wind Energy Association. It is expectedthat the final 5 years of the plan will benefit from the influence

    of Chinese manufacturers on the market, with capital costsexpected to drop by 25% to approximately $1.25 Million/MWin Australia. This is based on capital costs estimates for theGansu Province Three Gorges of Wind 20,000MW project,currently under construction.

    Hyctic bimss bcp

    There is the possibility of simultaneous cloud cover overseveral of the CST sites coinciding with a period of low windand high demand. To completely ensure energy security underthe plan, a system of biomass co-firing of the CST plantsis incorporated, utilising only waste biomass. The proposed

    system incorporates a simple biomass-fired thermal heateralongside the molten salt tanks at the actual CST sites.Biomass is burnt, and the energy used to heat the molten saltreservoirs, so that the existing steam power cycle, turbineand transmission can be utilised. This means the only extraexpenditure is for the biomass heater, minimising the extracost of the backup system.

    A combination of 5 GW of existing hydro capacity and 15GW electrical equivalent of biomass-fired backup heaterswill be sufficient to ensure reliable supply of electricity evenat times of low wind and solar radiation at several powerplants/sites. ZCA2020 does not propose additional hydro

    dams in Australia. Future work will investigate the feasibilityof upgrading existing hydro dams to pumped storage hydrofor long-term seasonal supply, which would replace biomassbackup. If further analysis and optimisation demonstrate thatmore or less biomass backup generation is required, this isnot expected to affect the overall system cost significantly,as the biomass backup heaters represent only 2% of overallinvestment for the ZCA2020 Plan.

    Pt 4: Mig th ZCa2020 100%wb gy mix

    The ZCA2020 electricity generation mix has beenquantitatively modelled in fine detail using real-world data,with the results demonstrating that Australias electricitydemand can be met by 100% renewable energy.

    The data underlying the model are derived from realmeteorological recordings of solar insolation at the 12proposed CST sites, real wind speeds at existing windfarms, and grid demand based on actual demand in theNational Electricity Market (NEM). Modelling has beenperformed on a half-hourly timescale for the years 2008and 2009.

    The wind data used in the model come from existingwind farms in South Eastern Australia. As a result, themodel overstates wind variability, which would in reality

    in weather systems that occur simultaneously across theAustralian continent and counter localised wind variability.Attempts have been made to select wind sites accordingto highest average winter wind speeds rather than highestaverage annual speeds, to accommodate the projected

    winter peak.

    Mgig wi vibiity

    Wind power integrates well with CST, because CST hasstorage that can readily dispatch power, providing back-up for wind power during lows in energy production.Similarly, periods of high wind output offset CST electricityproduction, and allow the CST plants to direct more heat tostorage for later use.

    40% penetration of wind is considered possible, based onstudies by British Energy Consultant David Milborough andDanish grid operator Energinet. These studies show thatoversupply of power, above the amount which the grid canhandle, is only 4%, at penetration levels of 40%.

    Tt Cst f Wi pw pts

    The total investment for wind farms is $72 Billion, asdetailed in Table 3.

    Y $Mii/MW

    CstctCpcity

    (MW)

    optiCpcity

    (MW)

    Csts($Mii)

    2011 2.2 1,250 2,000 $2,750

    2012 1.9 3,250 2,500 $6,175

    2013 1.9 5,500 4,500 $10,450

    2014 1.65 6,000 9,000 $9,900

    2015 1.65 6,000 15,000 $9,900

    2016 1.25 6,000 21,000 $7,500

    2017 1.25 6,000 27,000 $7,500

    2018 1.25 6,000 33,000 $7,5002019 1.25 6,000 39,000 $7,500

    2020 1.25 2,000 45,000 $2,500

    2021 50,000

    Total Capital Costs $71,675

    TaBle 3

    ZCa2020 Pjct a Cpit Csts f Wi(aud 2010 pics)

    Currently the average Australian capital cost for wind farmsis $2.2-2.5 Million/MW. This cost is relatively high comparedto other regions such as Europe and America as Australiahas seen a slower growth in wind power.

    The ZCA2020 Plan involves a large scale roll out of windturbines that will require a ramp up in production and assista subsequent reduction in wind farm capital costs, bringingAustralian costs in line with the world market. For the first

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    10/17

    | 10Sypss ZCa2020 Sttiy egy P

    be significantly lower due to the highly dispersed sitesoutlined in the Plan.

    The demand data derived from the NEM has been scaledto account for the extra demand for electricity under thePlan, when transport and present fossil fuel applicationsare converted to renewable electricity.

    rsts fm th m

    Figure 9 shows the daily averages of electrical outputfrom the model, running over a full year of data (2008).The black line between the red and orange areas signifiesthe electricity demand. The output from wind generatorsis shown as blue at the bottom, while the red area isdispatched energy from CST turbines. The light orange

    area is the excess energy that is available from the solarand wind sites but is unused.

    The Plans proposed combination of 42.5 GW of CST and50 GW of wind provides enough energy to cover 98% ofthe electricity demand modelled, with the remaining 2%being supplied by existing hydro power (5 GW) and heatfrom biomass co-firing on the equivalent of 10 GW of theCST plants, though spread out over many of the plants.This is represented by the small green and navy blueareas in Figure 9.

    Thm stg

    Figure 10 shows how the CST thermal storage operateson a randomly selected normal day. The light orange linesshow the solar input during the day, and the purple lineshows the reserve thermal energy stored in the molten salt

    Jan08

    Feb08

    Mar0

    8

    Apr0

    8

    May08

    Jun08

    Jul0

    8

    Aug08

    Sep08

    Oct0

    8

    Nov0

    8

    Dec08

    0

    200

    400

    600

    800

    1,000

    1,200

    1,400

    1,600

    Excess

    Biomass

    Hydro

    CST

    Wind

    GWh/day

    FIGure 9

    ZCa2020 M rsts (iy vg t). excss is cmbi s wi gy i xcss f m.

    GWh(reservoir)

    FIGure 11

    Hy isptch with bimss bcp

    FIGure 10

    Hy isptch with xcss sppy i Fby 2009

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    11/17

    | 11Sypss ZCa2020 Sttiy egy P

    2. Interconnection of the three main existing grids.3. Transmission upgrade of existing grids to provide

    resilience and reinforcement.4. More active demand-side load management.

    Grid upgrades use High Voltage Direct Current (HVDC) andHigh Voltage Alternating Current (HVAC) transmission lines,which are already in widespread use around the world.

    The proposed National Grid is illustrated in Figure 12, minusthe plug in transmission lines from the generators.

    HVDC is the most economical technology for long-distancebulk power transmission links. However the high cost ofconverter stations means that HVAC is specified for shorterlengths and if multiple connections to the existing AC gridare needed along their length.

    High-voltage transmission lines have been designed toconnect into the key distribution nodes of the existing grid,enabling an easy merger with present infrastructure andsecure distribution of electricity around the country.

    The total cost of the upgrade is $92 Billion, which isconsidered an important investment in Australias futureenergy security.

    Ct f sppy m

    Total stationary energy demand reduces by conversion of gasheating to efficient electric heat pumps and building efficiencyprograms. The conversion of gas heating to electric results inan increase in total non-peak electricity demand.

    Peak demand is reduced through active load managementusing SmartGrid systems. The electrification of heating, inconjunction with active load management systems, enablesthe pre-emption of heating and cooling load to smooth outpeaks in demand. Deferral of electric vehicle charging alsoprovides a form of load management.

    reservoirs of the CST plants, which decreases when thesolar input is not available (i.e. during the night). The bluearea shows the electrical output from the wind generators,and the red area shows the electrical output from the CSTturbines, which work together to meet demand.

    Bimss gti f bcp

    Figure 11 shows how the hydro and biomass backupoperate over three days of low input from solar and windenergy. The purple line shows the reserve thermal energycycling with the daily solar input, but progressively runninglower.

    The pink line shows the input of heat energy from thebiomass backup system, which is triggered when thereserve energy falls below a set point. While the biomassfiring is operating, this adds heat into the thermal storagetanks of the CST system. In this example, the biomassfiring stops briefly when the reserve energy rises abovethe trigger on the afternoon of June 1. The blue area showsthe electrical output from the wind generators, which isunusually low. The red area shows the electrical outputcoming from the CST stored solar energy. The purplearea shows output from the hydro generators. Finally thegreen area shows the output that is drawn from the storedbiomass thermal energy.

    Although the modelling shows that 10 GW(e) of biomassbackup generation will suffice for expected scenarios thathave been modelled, the ZCA2020 Plan includes a total

    15GW of biomass capacity to conservatively provide forenergy security.

    M simpifictis fth w

    This model does not yet take into account transmissionconstraints within the ZCA2020 electricity grid. The gridconnections to individual generator sites are sized to takethe full output, but some of the internal connections mayplace some constraints on the transfer of energy.

    On the other hand, the model does not take account ofdemand management benefits that are included in theZCA2020 Plan, or other advantages yet to be accounted for,

    such as a much more spatially diversified wind portfolio.

    Pt 5: Gi mgmt cti f ti gi

    nw ti gi

    The Stationary Energy Plan proposes a comprehensiveupgrade to Australias electricity grid to allow full utilisationof the distributed renewable energy network.

    Creation of the new National Grid requires four mainupgrades to the existing grid:1. Plug in connections from proposed power generators

    to the electricity grid.

    Geraldton

    to Perth

    Kalgoorlie to Port Augusta

    Cleve to Port Augusta

    Naracoorte to Port Augusta (Adelaide bypass)Naracoorte - Portland (HVAC) upgrade

    Mildura to Mount Piper

    Mildura to Melbourne

    Broken Hill

    to Mildura

    Mt Isa upgrade

    Port Augusta

    to Mildura

    Port Augusta to

    Melbourne

    Roma - Armidale

    (HVAC) upgrade

    Geraldton to NWIS

    (HVAC) upgrade

    (via Carnarvon)

    Roma to

    Moree

    FIGure 12

    Gi Stgthig Itcctis

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    12/17

    | 12Sypss ZCa2020 Sttiy egy P

    Mfctig cpcity

    Due to the consistent demand generated from the Plansconstruction timeline, the wind and solar factorieshave a guaranteed pipeline of projects. This allows themanufacturing industry to run continuously, achievingefficient operations with significant economies of scale.

    At the peak of construction activity, the significant

    manufactured components required by the Plan willinclude 1000 wind turbines and 600,000 heliostats peryear. While this may seem like a significant demand on

    In the worst case scenario of low wind and low sun, thereis a minimum of 55GW reliable generation capacity, whichin combination with passive and active load management issufficient to meet peak demand.

    Bc by Sici kight Mz viwThe engineering firm Sinclair Knight Merz has reviewed thetransmission network of the generation scenario proposedin the ZCA2020 Plan. The review found:

    the transmission scenario proposed is technically feasible interms of capacity and reliability. In addition, the proposedtransmission uses mature technology with proven capabilityaround the world.

    Pt 6: Impmtig th tsiti

    This section presents an analysis of the feasibility of thePlan within the context of Australias industrial, human,and material resource profile. The analysis clearly showsthat Australia has more than enough capacity to meet thechallenge of achieving 100% renewable energy within tenyears. The roll-out would require approximately 8% ofAustralias existing construction workforce. Manufacturinghalf of the turbines and heliostats domestically wouldrequire 3% of Australias manufacturing workforce, alongwith a few additional factories.

    Th cstcti timiThe ten-year timeline proposed by the Plan projects agradual scale-up of the renewable industry that wouldsee most of the proposed infrastructure completed in thesecond half of the decade. Fast-track of permitting andplanning approvals will be required.

    Accordingly, the construction of CST plants ramps uprelatively slowly at first (Figure 13). Then, as economiesof scale are developed, the rate increases and 80% of thecapacity is completed from 2015-2020. It is not expectedthat any CST would become operational until 2014-2015.

    The CST plants are modular, and once the design andplanning for the first plant is complete then the processbecomes one of replication on each of the 12 sites.Construction thus becomes progressively faster in thelater stages as companies gain experience and becomemore efficient. Current power tower projects have aconstruction time of 2-2.5 years, though more matureparabolic trough projects of similar nameplate capacityare being completed in 1.5 years.

    The timeline for wind power (Figure 14) has a somewhatearlier and faster start-up because Australia alreadypossesses some experience with wind technology, and

    because there are already 11,000MW of wind powerprojects are already in the pipeline.

    0

    5000

    10000

    15000

    20000

    25000

    30000

    35000

    40000

    45000

    50000

    MWc

    apacity

    Under construction

    Cumulative operational

    Jan20

    12

    Jan20

    14

    Jan20

    16

    Jan20

    18

    Jan20

    20

    Under construction

    0

    5000

    10000

    15000

    20000

    25000

    30000

    35000

    4000045000

    50000

    Jan20

    12

    Jan20

    14

    Jan20

    16

    Jan20

    18

    Jan20

    20

    MWc

    apacity

    Cumulative operational

    FIGure 13

    S Thm Istti Timi

    FIGure 14

    Wipw Istti Timi

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    13/17

    | 13Sypss ZCa2020 Sttiy egy P

    There are, however, great advantages to creating a localmanufacturing industrymost significantly, it presents anopportunity to participate in the global renewable energy boom.

    Jbs

    ZCA2020 modelling shows that many more jobs arecreated with the construction of a 100% renewable energygrid than are lost with the phasing out of coal and gasfrom the stationary energy supply chain. The Plan willcreate up to 80,000 jobs from installation of renewableenergy generation at the peak of construction, and over45,000 jobs in operations and maintenance that willcontinue for the life of the plant, as shown in Figure 15. Ifhalf the manufacturing for wind turbines and heliostats isdone domestically, this would create up to 30,000 jobs inmanufacturing.

    The level of expansion in jobs would boost overallemployment in the economy.

    Placing the job requirements for the ZCA2020 Plan incontext (Figure 16) with the existing Australian workforcein relevant industries shows the relatively modest labourrequirements that would need to be put towards the roll-out of the Plan. There are already just under 1 millionpeople employed in the Australian construction industry,with another 1 million people in manufacturing.

    In the five years leading up to 2008, the Australianconstruction industry was growing at the rate of 50,000new jobs per year. This has approximately flatlined

    since the Global Financial Crisis of 2008. Under mid-term government forecasts, growth in construction andmanufacturing is projected to be slow. The scale-up ofconstruction required for the plan is not as fast as whatAustralia has already shown itself capable of in the past.

    our manufacturing industry it is well within our capability.Across the globe there are examples of wind turbinemanufacture being ramped up quickly. Enercon hasestablished a wind turbine manufacturing hub in lessthan two years in Portugal. The manufacturing capacity

    required for the specified wind turbines in the Plan isequivalent to only 2-4 factories of similar size overseas.Manufacturing heliostats, on the other hand, is a muchsimpler process and is equivalent to establishing onemedium-to-large automobile factory of a similar size toexisting car factories overseas in Europe and Japan.

    Alternatively, turbines and heliostats could be imported.

    FIGure 16

    ZCa2020 Jbs i ctxt f xistig asti istis vt t th P

    Thousands

    ofJobs

    3,500

    3,000

    2,500

    2,000

    1,500

    1,000

    500

    0

    Professional,

    Scientific &

    Technical services

    Manufacturing

    Construction

    Mining

    Electricity, Gas,Water & Waste

    ZCA2020

    2002

    2000

    2004

    2006

    2008

    2010

    201

    2

    201

    4

    201

    6

    2018

    2020

    FIGure 15

    Tt Jbs ct by th ZCa2020 P

    No.ofJobs

    160,000

    140,000

    120,000

    100,000

    80,000

    60,000

    40,000

    20,000

    0

    Wind O&M

    CST O&M

    Heliostat manufacturing

    Wind manufacturing

    Wind Installation

    CST Installation

    Biomass Installation

    Line Installation

    Jan20

    12

    Jan20

    14

    Jan20

    16

    Jan20

    18

    Jan20

    20

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    14/17

    | 14Sypss ZCa2020 Sttiy egy P

    rscs qi

    The requirements for concrete and steel for the renewableenergy system are well within the capacity of Australiasresources.

    The requirement for concrete amounts to 7% of Australiasproduction over the 10 year timescale (Figure 18).

    The requirement for steel is 20-30% of Australias

    domestic production, but only 1.3% of Australias total ofiron ore exports and steel production (Figure 17).

    The main components of heliostat mirrors are glassand silver. The glass required is larger than Australiasproduction, but the demand can be met either by buildingone or two glass factories, or by imports.

    Silver is used as the reflective material in the mirror anda thin film is used on the back of the glass. The silverrequirements are very low with the Plan requiring lessthan 2% of Australias total production.

    emissis fm cstctiStationary energy, including electricity, is responsible formore than half of all Australian emissions. Under Business-As-Usual, the continuing emissions would be 2,950 Mt CO2-e

    MillionTonnes

    Potential Steel

    from ore exports

    FIGure 17

    ZCa St qimts icig St oexpts

    FIGure 18

    ZCa Cct rqimts

    Ts Gms CST Pw Tw cstcti i Spi

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    15/17

    | 15Sypss ZCa2020 Sttiy egy P

    Compared to Business-As-Usual, the ZCA2020 Plan ischaracterised by higher upfront investment, but lowerongoing costs. Under the Plan, by 2020, there would be$260 Billion more spent on the Stationary Energy sectorthan Business-As-Usual.

    This investment is of course paid back over a longertimeframe. One possible scenario where the ZCA2020renewable infrastructure is paid for through electricity

    revenues would lead to a price increase of 6.5c/kWh. Thisis based on a preliminary analysis and should not be takenas a recommendation of the Plan, but it provides a usefulbenchmark of the economic impact of the Plan. This pricerise would equate to only $420 per household per yearfor residential consumers, or $8 per household per week.

    nt pst cst cmpis

    The ZCA Plan is compared with continuing Business-As-Usual using Net Present Cost (NPC) over the period 2010to 2040. This comparison can be viewed in narrow termsof direct costs or in wider terms which take into account

    the fuel costs for transport and also the possibility of aprice on carbon emissions.

    In narrow terms of direct costs, NPC of Business-As-Usual is approximately the same as that of the ZCA2020Plan, shown in Figure 19. This is using a societal discountrate of 1.4%, an approach taken by the Stern Review onClimate Change.

    If oil costs for transport are taken into account, thenthere is a very clear financial advantage in converting thetransport system to renewable electricity. Even though thecomparison here does not include the cost of electrifyingtransport, it is clearly financially beneficial to replace oil as

    the major transport fuel.

    If the likely costs of emissions are included, then theZCA2020 Plan is clearly superior to Business-As-Usual.

    over 10 years. The emission embodied in the generatingand transmission infrastructure behind the ZCA2020 Plancomes to less than 3% of projected Business-As-Usualemissions, and the investment eliminates all on-goingenergy-related emissions thereafter.

    The most significant emissions from construction arise fromthe iron and steel for the solar plants - this comprises 60%of the total ZCA2020 construction emissions. While this issubstantial, it is dwarfed by the existing level of emissions:in total, the proposed 10 year roll out corresponds toapproximately 2 months of current Australian emissionsfrom the stationary energy sector.

    Pt 7 ecmic cmpiss

    The total investment required to implement the ZCA2020renewable energy system is $370 billion, averaging $37billion per year over the 10 year duration of the Plan. Thisis equivalent to 3% of GDP.

    The ZCA2020 Plan does not specify any particular fundingmechanism. The money could come from any mix of publicand private sources, depending upon a range of possiblepolicies that would lead to the implementation of the Plan.

    NetPresent

    Cost$A

    billion(real2010)

    Business as usual

    2,500

    2,000

    1,500

    1,000

    500

    0ZCA 2020

    Capital Investments

    Coal and Gas for

    electricity

    Operations and

    Maintenance costs

    Oil - BAU or ZCA2020

    ramp-down

    Heating Gas - BAU or

    ZCA2020 ramp-down

    Emissions (Stationary Energy)

    BAU electricity ramp-down

    operating and fuel costs

    (2011-2020, ZCA2020 case only)

    FIGure 19

    ecmic M Cmpis

    FIGure 20

    Cmtiv nt Pst Csts: Bau mis ZCa2020

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    16/17

    | 16Sypss ZCa2020 Sttiy egy P

    Cmpis f xpit with th s fth cmy

    The $37 billion annual cost for the ZCA2020 renewablegrid is a similar order of magnitude to other sectors ofthe economy (Figure 21 and Figure 22). This illustratesthat the Plan is neither beyond Australian capability norextraordinary for an infrastructure project.

    Implementing the ZCA2020 Plan requires investment ona scale that is well within Australian economic capacitiesand rapidly results in tremendous financial savings as

    well as other economic advantages such as eliminateddependence on oil imports, eliminated fuel price volatility,job and industry creation and of course creating a zero-emissions energy sector.

    Due to the nature of renewable energy sources havinghigh capital costs but low operating costs, the ZCA2020Plan requires a higher initial investment than BAU wouldrequire out to 2020. This additional expenditure, sourcedfrom both public and private investment, is offset byeliminating the costs associated with oil imports and likelyemissions costs. Economic pay-back occurs around 2040if emissions and oil costs are ignored (purple curve inFigure 20). When oil and emissions costs are included, thePlan achieves economic payback within a much shortertimeframe (blue and green curves in Figure 20).

    22.6

    66.6

    FIGure 21

    ZCa2020 Sttiy egy P cpit cst cmp t th cmic ctivity. GVa = Gss V a ftht sct, s ctibti t tt Gss dmstic Pct

    FIGure 22

    ZCa2020 Sttiy egy P cpit cst cmp t Tt asti GdP

  • 8/9/2019 ZCA2020 Stationary Energy Synopsis v1

    17/17

    | 17Sypss ZCa2020 Sttiy egy P

    Ccsi

    The ZCA2020 Stationary Energy Plan outlines a fully costedand detailed blueprint for transforming Australias energy

    sources to 100% renewable supply. This is achievableusing technology that is commercially available today, withno technical barriers to their deployment. Implementingthe proposed infrastructure in ten years is well within thecapability of Australias existing industrial capacity. Therequired investment is the equivalent of a stimulus to theeconomy of 3% of GDP.

    100% renewable energy in ten years is achievable andnecessary, ensuring Australias energy security, nationalsecurity and economic prosperity for the future. Australiahas some of the best renewable energy resources in theworld, and should be positioning itself as a leader in the

    emerging renewable energy economy. What is requiredto make this happen is leadership from policymakers andsociety, with firm decisions made quickly that will allow thistransition to occur.

    The time has come to aggressively accelerate thattransition... The time has come, once and for all, for thisnation to fully embrace a clean-energy future.

    - Barack oBama, PreSident of the united StateS of

    america, June 2010