Guest Post by Peter Lang. Peter is a retired geologist and engineer with 40 years experience on a wide range of energy projects throughout the world, including managing energy R&D and providing policy advice for government and opposition. His experience includes: coal, oil, gas, hydro, geothermal, nuclear power plants, nuclear waste disposal, and a wide range of energy end use management projects.
You can download a printable 16-page PDF version of this post here (updated 1 June 2010). This includes 7 appendices not included in the web-based version below.
Hazelwood Power Station is Australia’s most CO2 emission intensive power station. Replacing it with cleaner technology could reduce Australia’s CO2 emissions by 12 to 16 Mt/a. The NGO Environment Victoria recently commissioned a report by Green Energy Markets Pty Ltd to consider options. But the report has a pro-renewables bias, avoids the best option (gas only), and contains many inconsistencies.
Emissions saved per year: 12.2 Mt/a versus 11.8 Mt/a;
Capital cost: $6-$7 billion versus $2 billion;
Cost of electricity: $103/MWh versus $55/MWh;
CO2 avoidance cost: $64/t CO2 avoided versus $22/t CO2 avoided.
The renewables option for replacing Hazelwood is a poor one. It is high cost and yet yields only small extra emissions savings.
The significance of this analysis for governments is:
- It highlights the pro-renewables bias endemic in NGO environmental groups.
- It highlights the irrational decisions that some environmentalist advocates are causing.
- Federal and Victorian governments should reject the renewables option and implement the ‘gas only’ option.
- The nuclear option would be even better if it was available.
Hazelwood Power Station is Australia’s most emissions-intensive power station. The NGO Environment Victoria contracted Green Energy Markets Pty Ltd to consider options for replacing Hazelwood Power Station. The report “Fast-tracking Victoria’s clean energy future to replace Hazelwood Power Station”  was published by Environment Victoria in May 2010.
The project brief was (bold is my emphasis):
… to undertake an assessment into the options and opportunities for replacing the
Hazelwood Power Station by the end of 2012. This report assesses a combination of clean energy technologies to replace the generation capacity provided by Hazelwood in a way that maximise emissions reductions, whilst also maintaining energy security and minimising any increase in electricity bills.
Does it actually meet these criteria? The scenarios it considers are described as follows:
Scenario 1 – Supply side only option: this scenario involves bringing forward 1180 MW of combined cycle gas turbine plant running at 65 per cent capacity factor and 1500 MW of renewable generation (predominantly wind) at 30 per cent capacity factor; and
Scenario 2 – Supply side and demand side option: this scenario involves bringing forward 970 MW of combined cycle gas-fired generation running at 50% capacity factor initially, then declining over time, as well as 1500 MW of renewables. It also incorporates additional residential, commercial and industrial energy efficiency options that replace around 25 per cent of Hazelwood’s annual generation as well as 100 MW of Demand Side Management.
I have reviewed the report and my findings are provided in the following sections.
In short, the report does not meet its stated aims and contains many inconsistencies. Importantly, the report does not consider the alternative of replacing Hazelwood with just combined cycle gas (dispensing with the renewables component).
I have attempted to resolve the main inconsistencies (to the extent I can without having access to their assumptions and calculations). I have recalculated the CO2 emissions, CO2 emissions avoided, capital cost, cost of electricity and cost of emissions avoided. I have added a scenario for the case for combined cycle gas turbine plant only (no renewables).
Lastly, I considered a scenario where nuclear power is available as an established and mature option in Australia. Although nuclear is not an option on the time scale needed for replacing Hazelwood, it is important to recognise that if we continue to delay in allowing nuclear to be one of the options for replacing coal power stations in Australia in the future, we will be restricted as to how much emissions avoidance is achievable, the cost of emissions avoidance, and the cost pressures that will be applied to natural gas in the future.
My main criticisms of this report are:
- The energy efficiency and demand side management initiatives included in Scenario 2 – ‘Supply side and demand side option’ should not be considered a ‘replacement’ of Hazelwood. They are applicable to the whole electricity system, to all generators, and should be considered in their own right, not as a way to try to make the replacement of Hazelwood by renewable energy appear better and cheaper than it actually is. For this reason, I have not considered Scenario 2 any further in this critique.
- There are many inconsistencies in the report (see below for more on this).
- The report overstates the emissions savings to be gained from a mix of wind power and gas turbines (see below for more on this).
- The option that is clearly the best for the immediate replacement of Hazelwood, combined cycle gas turbines alone, was not considered (or if it was, the results are not presented in the report).
- The report seems biased towards promoting a pro-renewable energy solution despite the much higher costs and negligible additional emissions savings. This is not an objective, scientific or transparent way to plan energy policy.
Inconsistencies in the report
The report contains many inconsistencies. The numbers in the Executive Summary, Section 3 and Attachment 3 do not agree. Numbers in tables do not agree with related text. For example:
Page 12 states:
Hazelwood’s power contribution to the NEM can therefore be summarised as follows:
• Total generation – 11,770 GWh on a gross basis (effective capacity factor of 84 per cent). This needs to be reduced by the extent of its auxiliary electricity consumption (10 per cent) and its transmission loss factor (3 per cent). This means that 10,240 GWh per annum of generation needs to be replaced.
• Contribution to meeting peak summer demand of 1350 MW on gross basis with 1175 MW after auxiliary power use and transmission losses.
• Emissions intensity 1.37 tonnes/MWh on a gross basis (1.53 on a sent out basis after adjusting for auxiliary power use).
But these figures do not agree with Attachment 3. Table 1 shows three different figures for the annual generation that must be replaced.
When any of these figures are multiplied by the emissions intensity (1.53 t/MWh) the result does not match the total emissions figure quoted in the Attachment 3, which is. 16,166 kt/a. Multiplying the highest of these figures by the emissions intensity gives the total emissions as 15,900 kt/a. I suspect the 16,166 kt/a is the correct figure because it is derived from the 11,770 GWh (gross) generation multiplied by the gross emissions intensity (1.37 t/MWh). Therefore, I suspect the report has an error in the calculation of the ‘sent out’ energy. It appears, the report has understated the net energy that needs to be replaced. I calculate the sent out energy to be 10,566 GWh/a (see Appendix 2).
The peak capacity of the replacement generators is also understated. The report states 1350 MW (gross) of peak generating capacity is required to replace Hazelwood. I calculate 1500 MW (gross) is required. Why the difference? Figure 3 shows that Hazelwood provides 1600 MW peak power (gross). However, the report has used the average power output over the summer rather than the peak power output. Then this figure was reduced by 10% (allowance for the internal energy use) and also by 3% for transmission losses. However, transmission losses should not be deducted in calculating the ‘sent out’ power. It seems transmission losses have been deducted in calculating Hazelwood’s net peak capacity, but not included when calculating the gross capacity of renewables and gas generators required to replace Hazelwood. The transmission losses from the wind farms would be higher than from the coal and gas power stations. The transmission losses from coal and gas should be similar. The transmission losses should not be included in the calculation of the capacity. I calculate, to replace the 1600 MW gross capacity of Hazelwood, we would need 1500MW gross capacity of combined cycle gas turbine if air cooled, or 1462 MW if water cooled. The report states the replacement capacity required is 1350 MW (gross). This significantly understates the peak capacity required to replace Hazelwood. (See Appendix 2 for basis of calculations.)
Emissions Savings overstated
The emissions savings that would be achieved from the proposed combination of wind power and gas turbines is overstated.
- Combined Cycle Gas Turbines (CCGT) cannot back up for wind power on their own. A mix of CCGT and Open Cycle Gas Turbines (OCGT) would be required.
- Both types will have to operate in a cycling mode and both will run at below their optimum output. As a consequence, both will produce higher emissions than they would if running at optimum output and if not cycling to follow the changing wind power output. , , 
- The gas turbines will spend more time in start up, spinning reserve and cool down, than if they were not backing up for wind power.
- The Kent Hawkins calculator provides some guidance on the additional fuel used and emissions involved in shadowing for wind power .
Calculations on Emissions Savings
I have recalculated the figures for Scenario 1 using what I believe are more realistic assumptions and inputs. I have also calculated the figures for the option with CCGT only (with no renewables). Appendix 1 compares the figures in the original Scenario 1, my revised Scenario 1, and the ‘CCGT only’ scenario. The following table compares the main results.
In recalculating, I took the gross generation, total annual emissions (16,166 kt/a), the emissions intensity (1.53 t/MWh) as correct. To calculate the net generation I ignored the 3% transmissions losses and I changed the capacity factor slightly (from 84% to 83.76. Making these changes gives the Hazelwood net generation as 10,566 GWh/a. Note the emissions saved are less than stated for the original Scenario 1.
Table 3 compares the two options, wind and gas versus gas only, on the key criteria of capital cost, cost of electricity, emissions avoided and cost of emissions avoided. The basis of estimates is in Appendixes 3, 4 and 5.
Replacing Hazelwood with wind and gas generators (Scenario 1) is only 3% better than the gas only option for the amount of emissions avoided. However, the wind and gas option (Scenario 1) is much more costly than the gas only option – see Table 3. The wind and gas option is 3.7 times the capital cost, 3 times the emissions avoidance cost, and, importantly for most people and industry, the cost of electricity is nearly double that of the gas only option. Thus, their stated criteria of “minimising any increase in electricity bills” is not satisfied.
On this basis it is clear that the wind and gas option should not be considered further. For currently available replacement technology in Australia, the gas only option is by far the cheaper option, and has only slightly (3%) higher emissions.
I also considered a ‘Nuclear’ option. It is informative to consider this option because it demonstrates why we should not continue to delay the decisions to allow nuclear to be an option for new electricity generation capacity in Australia. Had the Hawke Government not banned nuclear from consideration during the Ecologically Sustainable Development work 20 years ago, we could have five operating nuclear power stations by now, be past the period of FOAK (first of a kind) costs, and have nuclear power providing clean electricity at a competitive cost. In this case our emissions from electricity generation would be near 20% lower than they are today. The clear message from this is we should not delay the decision to allow nuclear as an option for generating our electricity in the future.
If nuclear was an available option, replacing Hazelwood with nuclear would reduce emissions by 16 Mt/a. If the cost of electricity from nuclear power was the same as for the new nuclear power plants in Europe , the cost of electricity would be about $4/MWh (8%) more than the combined cycle gas plant option now, and much less as gas prices rise in the future. (Gas price is the main factor in the cost of electricity from gas generation, but fuel cost is a very small component of the cost of electricity from nuclear). Table 4 lists the key results:
1Capital cost and electricity cost from NEEDS , p3, converted to A$ and escalated to 2010 $.
Implications for governments
The renewables option for replacing Hazelwood is a poor one. It is high cost and yet yields only small extra emissions savings.
The report demonstrates an obvious pro-renewables bias in the advice being provided to governments by the environment NGOs.
Such bias is causing irrational decisions that are forcing high cost electricity on Australia.
Federal and Victorian governments should reject the renewables option and implement the gas only option.
Governments should recognise subsidising renewables is irrational and costly.
Australian governments should implement the policy, legislative and regulatory changes necessary to allow nuclear power to be implemented at least cost (consistent with appropriate safety requirements) in the shortest practicable time.
 Green Energy Markets (2010). Fast-tracking Victoria’s clean energy future to replace Hazelwood Power Station. Environment Victoria.
 Lang, P, (2009), Cost and quantity of greenhouse gas emissions avoided by wind generation.
 Hawkins, K, (2010) Wind Integration: Incremental Emissions from Back-Up Generation Cycling (Part V: Calculator Update).
 Lang, P, (2010), Emission cuts realities – electricity generation: Cost and CO2 emissions projections for different electricity generation options to 2050.
 NEEDS (2007) Final report on technical, costs and lifecycle inventories of nuclear power plants.
 ACIL-Tasman (2009), Fuel resource, new entry and generation costs in the NEM.
 Mills, A. et al, (2009), The Cost of Transmission for Wind Energy: A Review of Transmission Planning Studies. Lawrence Berkeley National Laboratories, Environmental Energy Technology Division.
 ABARE, (2009), Electricity generation; Major development projects – October 2009 listing
Download a printable 16-page PDF version of this post here. This includes 7 appendices not included in the web-based version above.