AN OPTIMAL PORTFOLIO OF NEW POWER GENERATION - PowerPoint Presentation

Slide1
Slide2

AN OPTIMAL PORTFOLIO OF NEW POWER GENERATION

TECHNOLOGIES:AN ILLUSTRATION FOR SOUTH AUSTRALIA

Anthony D Owen and Ntasha BerryInternational Energy Policy InstituteUCL Australia, Adelaidetony.owen@ucl.ac.ukPresentation to the 32nd USAEE/IAEE North American ConferenceAnchorage, Alaska, 25-26 July 2013Slide3
Slide4

Australia’s National Electricity Market:

an energy-only market

Gross pool: obligatory marketPrice bids by generators every 5 minutes, averaged over half-hour trading intervals in five regional spot marketsTypical bid structureSome MW @ very low or –ve $/MWh: up to minimum loadSome MW at SRMC values: up to contracted positionRest at blue sky valuesAll dispatched power attracts the same marginal bid price, irrespective of the aboveFloor price = -$1000/MWhCeiling price = $13,100/MWhSlide5

The Cumulative Price Threshold

If the sum of the trading interval spot prices over a rolling seven day period total or exceed the Cumulative Price Threshold (CPT), then spot prices are capped

at the administrative price cap of $300/MWh. The CPT is currently $197,100, and is subject to annual indexation.Slide6

Characteristics of the SA market

Tail-end of longest connected grid in the world – 5000 km;

Electricity demand very “peaky” – peak load (in excess of 2,600 MW) occurs around 2% of year;Load ranges between 1,200 MW and 2,400 MW 90% of the year;Installed capacity (nameplate) 4,000 MW;Installed capacity of wind – 1,203 MW;Dominant generator – gas thermal (1,280 MW).Slide7

SA generation mix

Capacity Generation

Gas 55% 50%Wind 24% 26%Coal 15% 24%On particular days, wind has accounted for up to 65 per cent of total generation in the state, and up to 86 per cent of generation for a trading interval. However, wind generation is generally lower at times of peak demand; on average, it contributes to less than nine per cent of supply at any given time during summer. Slide8

Number of negative trading intervals in the NEM

2009-2010 86

2010-2011 2082011-2012 274Given the RECs wind receives, it can bid negative to ensure dispatch. Coal and gas may also bid negative to avoid costs associated with shutting down or operating inefficiently. Slide9

Electricity market intervention

Current government policy in the power sector is primarily focused on policies to support the development and deployment of non-nuclear low-carbon technologies to reduce their costs and thus reduce the long-term costs of de-

carbonising the sector. However, the current NEM market design may make low-carbon investments riskier than continued investment in fossil fuel technologies. Thus, even with a carbon price, investment in low-carbon technologies may be discouraged.Slide10

Carbon pricing

The marginal generator (that sets the wholesale price) is generally a fossil fuel generator. Thus, marginal bid pricing should include the carbon price. This should benefit low carbon technologies. However, the risks attached to recovery of fixed costs of new generation assets will vary considerably depending on the capital intensity of the different technologies. For large up-front cost low carbon technologies (such as geothermal, concentrated solar thermal, off-shore wind, CCS, etc.) these risks may more than offset the cost

minimisation criteria for investment choice.Slide11

Investment in power generation

It is intended that price “signals” should encourage investment in new capacity, but long-term investment decisions appear to conflict with the short-term (i.e. instantaneous) nature of the market and may impede the effectiveness of climate policies. Uncertainty on future electricity prices is compounded by other risks:

Long-term uncertainty on carbon price;Unclear competitive nature of low carbon technologies;Long lead times for high up-front cost low-carbon technologies: the chain of innovations is too long, too complex and too diverse;High levels of political and regulatory risks; andUncertain availability of finance.Slide12

Wholesale market without windSlide13

Wholesale market

with deployment of windSlide14

Overview of investment issues

Slide15

Classification of investment opportunities

I

nvestment opportunities in the NEM may be classified as:capacity-driven: short-fall in reserve margin reflected in high prices during peak demand, usually requires investment in OCGT;energy-driven: determined by longer term (average) price trends; andpolicy-driven: arise from Government policies, generally to encourage renewables. Slide16

Uncertainties in the SA electricity market

Demand (SA): mining expansion/industrial & manufacturing contraction + PV penetration.

Fuel price (International): domestic gas prices are relatively low by international standards, but will rise as LNG netback prices kick-in.Carbon policy (National): high degree of uncertainty with linking to EU ETS, but could be abandoned altogether!21 possible scenarios to reflect interactions Slide17

Least cost generation expansion modelling

This

study developed a least-cost generation expansion model to determine a set of long-term generation options from a set of possible technologies which minimised the cost over time, taking into account future demand, carbon price, and gas price uncertainties. For this application, the long-term planning simulation module of a commercially available software, Plexos, was used.The model emulates the process performed by the dispatch system in the NEM, which schedules generators that minimise the dispatch cost, subject to the criteria that the reliability and security of the system is maintained. The model executes half-hourly dispatch using security constrained optimum generation dispatch with each generator bidding at its short-run marginal cost.Slide18

Model constraints

An energy balance constraint, which ensures that the total electricity demanded at a given time is equal to the

un-served energy at that time plus the sum total of the electricity dispatched by all the generation units at that time;A feasible energy dispatch constraint, which ensures that the level of electricity dispatched by a given generator is not greater than the maximum generating capacity of that generator;A feasible build constraint, that ensures that the built capacity (in MW) for a given new entrant technology does not exceed the value of maximum built capacity (as set in the model) for that technology in a year (or over a given planning horizon); andA reliability standard constraint, which ensures sufficient generation capacity to meet demand.Slide19

Model inputs

The model considered the cost parameters of the existing generators in terms of their fixed and variable operations and maintenance costs, combined with their technical constraints: ramp-up and ramp-down rates, thermal efficiency, combustion emission factor, fugitive emission factor, minimum stable generation level, forced outage and planned outage rate, auxiliary load requirements and marginal loss factor over the planning horizon. Variation of nameplate capacity of generators due to seasonal factors was based upon past experience.

The model expanded the generation capacity within the region based on the inter-region transfer capability. The interconnector transfer capacity limits the transfer of energy between Victoria and South Australia during any load block and affects the reserve sharing capability between the two regions.Slide20

C

ost data for new entrant generators

Slide21

Results: Energy, capacity and policy-

driven investment opportunity in SA

The optimum energy, capacity, and policy-driven investment opportunities in SA in terms of capacity, technology and timing of additional generation over the next 20 years are given in the next slide. As a result of the closure of coal fired power plants (under the CEF), the reserve margin fell significantly and short-term investment in OCGTs was required. However, the results are dominated by the LRET constraint driving investment in substantial amounts of wind energy. Solar thermal and geothermal HSA technologies were also stimulated by the LRET, with initial build commencing in 2016 and 2019 and reaching maximum regional build by 2020.Slide22

Energy, capacity and policy-

driven investment opportunity in SASlide23

Conclusions I

As

might be expected, the LRET scheme, “peaky” demand, the high level of (intermittent) wind capacity, and volatile gas prices favoured wind and OCGT technology. Fuel price risk is the main source of concern for investments in CCGT technology, although generators with an upstream gas position have a natural hedge against this risk. In addition, the less capital intensive OCGTs can have their cycle “closed” when deemed appropriate. The “peaky” nature of demand in SA makes this an ideal strategy for meeting any future increase in base load demand. Slide24

Conclusions

II

The modelling also indicated that geothermal HSA technology was to be preferred on a cost basis to CCGT for base load, largely due to high gas price expectations and low operating capacity for the latter. However, this result is predicated upon geothermal becoming financially viable by 2020.