Delivering on the ISP – risks and opportunities for future iterations

The Integrated System Plan (ISP) released last week by the Australian Energy Market Operator (AEMO) maps out what it sees as the optimal development path (ODP) for generation, storage and transmission investments to hit the country’s net zero by 2050 target.

The ISP and its ODP are predicated on a range of Federal and state government policy settings and reforms and on a lot of scenarios being successful. Contrary to how it has been characterised over the last week, the ISP points to the scale and scope of the energy transition and what the energy system could look like if the ODP is achieved. Whilst it models what the government targets require for the system to change, it doesn’t seek to forecast what will happen in practice.

As with all modelling exercises, the ISP is based on a range of inputs and assumptions, all of which can, and do, change.  As noted by AEMO in its disclaimer, modelling “inherently requires assumptions about future behaviours and market interactions, which may result in forecasts that deviate from future conditions. There will usually be differences between estimated and actual results, because events and circumstances frequently do not occur as expected, and those differences may be material.” As a result, there will be a range of views on how realistic those assumptions, modelling and outcomes are.

AEMO itself has highlighted several risks to achieving its ODP, which we summarise below. We also look at a couple of areas where the ODP could face challenges.

The Expectations

The 2024 ISP expects all of Australia’s coal generation fleet to close by 2038, removing more than 125,000GWh from Australia’s generation mix or 46 per cent of the total annual generation[i]. At the same time electricity consumption from the grid is expected to nearly double.

To deal with this shift in generation, the ISP sees a need for:

• Storage to increase significantly from 3GW to 22GW by 2030 and to 49GW by 2050, with 66 per cent expected to come from coordinated Consumer Energy Resources (CER) (behind the meter batteries).
• Gas-fired generation capacity to increase from 11.5GW now to 15GW in 2050, including replacement of 9.3GW of the current capacity that is expected to retire.
• Maximum daily gas-fired generation demand to increase from 1,001TJ/day now to 2,461TJ/day in 2050.
• Grid-scale wind and solar to increase from 21GW to 55GW (2030) to 127GW (2050).

Rooftop solar increasing from 21GW to 36GW (2030) to 86GW (2050).

Some of the Risks

The ISP outlines four sets of risks that require market settings to be in place if the National Electricity Market (NEM) is to be ready for a very high penetration of renewables and for coal plant retirements.

• Risk of insufficient infrastructure investment
• Risk of early retirements of coal-fired generators
• Risk that market and power system operations are not ready for high penetration of renewables; and
• Risk that insufficient consumer energy resources are integrated into grid operations.

AEMO also points to the risks arising from a lack of a social licence and supply chain and adequate skills workforce risks.

The retirement of coal, five years earlier than previously forecast, highlights the urgent need for investment in infrastructure to ensure this gap in generation is filled.

Current estimates forecast Australia needing three times as much generation as it currently has installed by 2050, further underlining the need for significant investment. Any delay in investment will risk the energy transition and increase the likelihood of interruptions and higher costs, according to AEMO. Project approvals are proving to be one of many challenges. Renewable energy projects are taking years to be approved, with large-scale renewable projects taking an average of 746 days to progress through the approval process in NSW^[i]. Cost pressures, investment decision uncertainty, social licence issues, supply chain issues and workforce shortages are also proving an issue in projects being developed to schedule.

A report published last year by IMF Investors, in partnership with Australian superfunds -AustralianSuper, Cbus, Hostplus, CareSuper, HESTA and UniSuper - found Australia will need an average of at least $12 billion a year in investment to transition to net zero by 2050, but with delays in development and uncertainty on when projects will be operational, these funds and other investors are looking towards the US, UK and Europe for investment, potentially jeopardising renewable infrastructure in Australia.^[ii]

Gas Generation

The ISP expects nearly 13GW of new gas-fired generation to be needed in the next 25 years. With the Capacity Investment Scheme (CIS) excluding gas investments to help firm the grid it is not yet clear what investment signals will drive the necessary investment in new gas plant.

The most recent gas plant development was Tallawarra B, a 320MW open cycle plant built as part of the existing Tallawarra facility in NSW. It was the first in the state for more than a decade. Snowy Hydro’s Hunter Project at Kurri Kurri, which will involve two large open-cycle gas turbines (OCGTs) (660MW initially, up to 750MW) is expected to come online in December this year. AEMO currently lists 3068MW of proposed and anticipated OCGT plants plus 207MW of combined cycle gas turbine (CCGT) generation that it expects to come into the grid.

There have been questions raised about Australia’s gas supply, particularly for Victoria and New South Wales. AEMO has warned there could be annual deficits from 2028 without new supply, and suggestions we might already be late to bring new gas online[iii].  It also recently issued a risk/threat notice for the east coast gas system. Even with good gas availability, there have also been questions about the capacity of the system to deliver it, particularly during winter and in situations where there is a renewables drought and cold weather pushing up demand. This year the existing pipeline from Queensland to southern states has been running at capacity. The ISP notes winter peaks are forecast to test the limitations of the current gas supply network and suggests on-site storage or other solutions will be needed to address when there are network constraints.

Consumer Energy Resources

As can be seen above, there is an expectation that CER will play a major role in delivering the energy transition. Aside from electricity generation for rooftop solar, there is an expectation in the ISP that coordination of consumer batteries will be able to deliver the bulk of the NEM’s storage capacity by 2050. The ISP sees coordinated CER increasing from 0.2GW now to 3.7GW in 2029-30 and 37GW in 2049-50. ‘Coordinated CER storage’ usually refers to batteries at homes and businesses, which are aggregated into Virtual Power Plants (VPPs). Retailers and aggregators would provide for orchestration of battery charging and discharging as part of a VPP.

Figure 1: Storage capacity by type and year

Source: AEMO ISP 2024

The ISP notes that hitting the projections will need a mix of financial incentives, technology and communication standards, customer preferences, and market or policy arrangements as well as greater engagement from consumers, retailers, networks and other market players.

The ISP recognises there is a risk of CER not eventuating at the scale envisaged and there are a number of barriers that will need to be overcome.

Given success in orchestrating, CER will be a factor in offsetting other investments such as the scale of network and utility scale infrastructure needed to maintain reliability, security and affordability, AEMO developed a low CER orchestration sensitivity to investigate the impacts on power system needs of a lower uptake relative to the Step Change scenario. Its verdict is it would require $4.1 billion in additional grid-scale investment.

Some of the challenges to large-scale CER are highlighted by Project EDGE, a 3-year cross-industry collaboration, part funded by ARENA, that was undertaken to consider the potential for coordinated CER and VPPs. This involved   households and businesses and showed price responsive CER was technically feasible.

Based on assessments to date, while households like the idea of VPPs, they appear less interested in becoming involved (see figure 2). The motivation for most households to invest in solar PV, batteries and EVs is to manage their energy costs and be more energy self-reliant. The final Project EDGE report notes consumers are “open but lukewarm about joining VPPs”. It confirms consumers will need incentives to join a VPP and will need to be confident they will be better-off by participating in VPPs.

Figure 2: Consumer views on VPPs

Source: Project EDGE Final Report, Customer Insights Study Summary Report

The Project EDGE final report refers to three potential barriers that could affect the motivations of consumers joining a VPP. These are:

• Unfair financial returns.
• Insufficient stored energy to cover consumers’ own consumption.
• Data security.

While consumers appear positive about the concept of coordinated CER, there are financial and trust barriers to getting involved, and they will need not just incentives but also confidence they will actually be better off to get them to participate at the scale required. Given the importance of CER to the current ISP, there is and will need to be continuous government, market body and market participant attention given to how the scale of CER envisaged can be delivered.

Peak demand and reliability

Peak demand is forecast to be met within the reliability standard throughout the ISP’s forecast period, through combinations of renewable generation and storage, supported by gas-powered generators. But renewable droughts can occur and are most common during winter months when there is less solar available.

AEMO tested its ODP to see if it would meet the reliability standard through VRE drought conditions across the southern regions of the NEM for eight days, a period it reports is at least twice as long and more severe than any since 2010, and possibly since 1980. It notes historical weather patterns suggest longer ‘dark and still’ periods of up to three days covering a wide geographical area are rare. AEMO’s ISP modelling applied weather patterns from 2010-11 to 2019-20 to analyse the future operability of the power system. It notes that wind records back to 1980 “show similar average wind and a comparable spread of extremely high or low wind speed events”.

Figure 3: Operability through eight-day renewable drought, NEM except Queensland

Source: AEMO ISP 2024

The test showed the power system would remain reliable and secure, but unsurprisingly “reliability risks would be elevated, particularly if major generator or transmission outages occurred”.  It also has assumed new transmission is delivered as planned, to allow renewable generation to be shared. The modelling excluded Queensland, the second largest NEM jurisdiction, because it assumed there is low risk of a NEM-wide event. The chart covers a ten-day period with the first two days showing high levels of wind generation.

Reliability of supply in a higher renewables grid is a real risk which needs to be carefully managed.  Already in the second quarter of this year we have seen four significant periods of wind lulls. Although none were NEM-wide, one of the recorded events involved reduced output from wind farms in Queensland, South Australia and Tasmania. The lulls occurred between 13 April and 11 June and ranged in duration from 4.5 days to 7 days. The wind lulls are well outlined by energy analysts WattClarity and you can read their detailed analysis here.  Wind shortages were considered periods where the capacity factor in rolling totals fell below 15 per cent on consecutive days. WattClarity also reported the aggregate yield from wind farms across the NEM this year is “easily the worst of the 5 years sampled” and based on previous analysis of 16 years of weather data, they noted significant and sustained bad wind periods occurred every five years.

All of this serves to highlight the fluctuating nature of the output from the current NEM wind generation, underscoring the challenges in meeting energy needs as well as the ramping requirements from gas plants and/or long duration storage in a future grid with higher penetration of renewables.

The reliability analysis in the outer years of the ISP raises questions around how the energy system would cope with renewable droughts. Ideally, future ISPs should deal with this at a more granular level to provide confidence that the energy transition will continue to deliver an affordable and reliable system as emissions come down.

Conclusion

Developing such a broad assessment of what the future grid could look like based on current government policies and reforms is always going to be extremely challenging as it is open to events and circumstances changing. To flag this in the current ISP, the market operator has considered a range of risks to the delivery of the ODP, which include supply chain constraints, cost pressures, social licence issues and shortages of skilled labour.

It is a significant piece of work, but the sensitivities could be improved by more fully quantifying the impact of all the various risks it has identified and by assessing the compounding impacts of these. If multiple risks materialise, a combined sensitivity analysis could articulate what these challenges might look like, and ideally identify an alternate path.  

While the ISP includes some further sensitivity analysis, for example in relation to CER, it would be useful to include further such analysis which would help policy makers understand trade-offs. For example, if the policy settings for the identified investments were not in place, and the required investments did not materialise, what’s the next best alternative and what’s the impact on the net market benefits? More sensitivity analysis could help the ISP become something that evolves and plays a role in supporting both market participants and policy makers in their decision making to best support the energy transition.

[i] https://www.energy.gov.au/publications/australian-energy-statistics-table-o-electricity-generation-fuel-type-2022-23-and-2023#:~:text=The%20largest%20source%20of%20renewable,3%20percentage%20points%20on%202022.

[ii] https://www.abc.net.au/news/2024-02-29/renewable-approval-delays-slowing-cleanenergy-transition/103517718

[iii] https://www.theguardian.com/australia-news/2023/dec/01/industry-super-funds-warn-slow-transition-to-net-zero-puts-australia-at-risk-of-losing-attractive-investments

[iv] Does gas-rich Australia really need to start importing it?, The Age 3 July 2024