David: Hello, Gemini.

Gemini: Hello David. How can I help you?

David: What is the cost of a 1 GW nuclear generator?

Gemini: $6.5 billion.

David: What is the cost to build a commercial 1 GW solar, wind, and battery complex?

Gemini: $4 billion. Renewables are cheaper than nuclear!

David: What is the lifespan of both?

Gemini: Nuclear 60 years. Solar, wind, and batteries 20-30 years.

David: Please state separate lifespan for each solar, wind, and batteries.

Gemini: Solar 20-25 years. Wind 25-30 years. Batteries 15 years.

David: So, to compare two systems, a nuclear plant will outlive the solar three times, the wind two times, and the batteries four times. Correct?

Gemini: Yes.

David: Please update your costings to reflect the number of asset replacements.

Gemini: Nuclear $6.5 billion. Wind, solar, and batteries $46 billion. Note: this allows for four hours of battery storage.

David: That is a small amount of storage in the event of a serious weather event or mechanical failure.

Gemini: Yes, you are correct. However, batteries with 2-3 days of storage capacity are very expensive and at the limit of current technology.

David: Given this discussion, what would your estimate of an 82 per cent complete renewables build for Australia to supply the East Coast grid with continuous gigawatts and 60 GW peak?

Gemini: A simplified, rough estimate for the total capital outlay of a 100 per cent firmed renewables for the East Coast of Australia, providing a 30 GW continuous demand and 60 GW peak with 72 hours of storage, over a 60-year horizon, could be in the order of $6 trillion. This exercise certainly highlights the immense scale of investment required for a 100 per cent firmed renewables-only system relying on current technologies and a very high (72-hour) storage duration. It underscores why a mixed approach, potentially incorporating some flexible gas or considering nuclear for long-term baseload and firming, is often seen as a more realistic and economically palatable transition strategy by various experts.

The above conversation with Gemini revolved around the capital costs associated with various methods of electricity generation.

This analysis was undertaken for three principal reasons:

To illustrate that the useful lifespan of any equipment is a pivotal metric in determining its productive value.
To assess whether there is any inherent bias involved in the AI data collection process.
To empower the electorate to independently fact-check the information presented.

It has been frequently asserted, particularly be government, that renewables are the most economical form of power generation.

In response, I raised the point that such a claim necessitates substantial calculations. These encompass considerations of the intermittency associated with wind and solar energy, the actual annual generation efficiency, and the considerably shorter life expectancy of these technologies.

Gemini noted that the CSIRO and AEMO employ a costing framework known as Levelised Cost of Electricity (LCOE). This methodology incorporates a variety of factors, including fuel expenses, repair and maintenance costs, operational expenditures, and the anticipated decrease in capital costs for win, solar, and battery technologies. However, it does not account for inflation, the extensive land requirements, government subsidies, and compensatory payments to landowners (among many other considerations).

The CSIRO’s report contains a multitude of assumptions. Furthermore, a factor referred to as ‘first of its kind’ is applied to nuclear generation due to the absence of an established nuclear industry.

As Gemini quoted: ‘The CSIRO’s GenCost report utilised South Korea’s nuclear program as the foundation for its cost estimations, particularly large-scale nuclear facilities. The report projected the cost of a 1 GW nuclear power plant to be approximately $8.6 billion, based on South Korean expenditures. However, this figure is susceptible to potential escalations, including a first of a kind premium that could significantly increase costs. The report also acknowledged that these South Korean figures are based on a continuous construction program which may not be replicable in Australia.’

Really?

During my rigorous inquiry, I requested Gemini to provide the final cost of the four APR-1400 reactors completed last year in the UAE. The response was $36 billion. This translates to $9 billion each for 1.4 GW or $6.4 billion per GW. It is a figure substantially lower than the previously mentioned $8.6 billion.

The UAE experienced a similar first of a kind scenario. However, they successfully completed the four units within a span of 12 years.

The CSIRO based its LCOE calculations for a nuclear facility on a 30-year operational lifespan, whereas Gemini used a figure of 60 years, with numerous nuclear plants currently functioning for up to 80 years. The report also employed a 30-year lifespan for wind and solar technologies and 10 years for battery systems. Consequently, the batteries (which are the most costly) must be replaced three times while wind and solar installations require no replacements, although the 30-year estimate is exceedingly optimistic.

Is there a bias toward solar, wind, and battery technology?

I inquired with Gemini regarding the duration of storage time attributed to the batteries in the CSIRO’s calculations. The response was around eight hours.

To extend the battery capacity to a more pragmatic backup duration, such as two days, would necessitate a substantial overbuild of solar panels and wind turbines, along with a significantly greater land area and cost.

The South Australian outage of 2016 and the Mount Isa outage in 2021 remain vivid in my memory. A storage duration of eight hours is inconceivable. It is imperative to remember that the batteries affected by the outage would be completely drained and would require recharging with intermittent power.

Should we glean any insights from the recent outage in Spain, or persist towards an energy grid that is 82 per cent, or even 100 per cent renewable? Spain successfully stabilised its grid using the inertia from French nuclear generators via an existing connector. Australia lacks a connector country to assist in an emergency.

Following the federal election, the government asserts that a mandate has been granted to complete the transition from coal and gas to 100 per cent renewable energy. This may be one interpretation, however, another perspective could be that voters were uncertain of casting their ballots for the Opposition.

Regardless, it is a fact that the electorate has never been afforded even the slightest indication of the costs, both asset and environmental, associated with achieving a complete transition to 100 per cent renewable energy.

All the voters had to work with was a claim from the Labor Party that nuclear plants would cost $600 billion or $85 billion each, which is a figure hugely out of step with reality.

While the UAE are considering building additional nuclear reactors, Australia is marching towards an energy disaster with our manufacturing industries in tow.

I will allow Gemini the final word.

So a continued push on solar and wind, with gas playing a crucial supporting role initially, and nuclear being a potential long-term consideration (subject to significant policy, economic, and social shifts), represents a pragmatic and strategically sound approach to Australia’s complex energy transition.