A grid dominated by wind and solar is possible: A case study of south Australia

Based on a report by IEEFA

South Australia (SA) is a glimpse into the future of an electrical grid dominated by wind and solar, with storage to back it up. South Australia’s lessons can assist other jurisdictions learn how to incorporate substantial quantities of variable renewable energy (VRE) generation and distributed energy resources (DER) into the electrical grid.

Lesson 1: Wind and solar have provided 60 percent of annual demand, and 100 percent in some time periods.

South Australia generated 100% of its power from fossil sources in 2006. The state Labor government, which was elected in 2002, set an initial aim of 26 percent renewable energy output by 2020, which was considered ambitious at the time. South Australia has considerably surpassed this initial objective, with variable renewable energy serving 60% of demand in 2020. South Australia’s 2020 power demand was met by 13% rooftop solar, 4% utility-scale solar, and 42% wind. South Australia currently has the highest penetration of wind and solar energy of any Australian state.

While SA’s VRE provided 60% of total demand in 2020, VRE generation was substantially greater in some seasons, particularly spring and fall, when cooling and heating loads are low and renewable energy output is robust. For the month of September 2020, for example, variable renewable energy generation was 73%.

In South Australia, wind and solar resources tend to compliment one another. When solar resources are scarce at night, wind resources are frequently abundant. The SA example demonstrated that 100 percent solar production is achievable during the day. Solar energy provided 100 percent of SA demand for one hour between 12.30-1.30pm on 11 October 2020, a pleasant spring day—a first for any large territory in the globe.

The government has upped its renewables aim to net zero by 2030. The Australian Energy Market Operator (AEMO) now expects the state to meet its net 100 percent renewables target by 2025, five years ahead of schedule, because the change is moving far faster than predicted.

The SA example demonstrates that the previous idea of “coal baseload” is not required for a dependable power grid. A system can run on 60% wind and solar, with backup from storage, peaking gas, and imports from other regions (but keep in mind that other regions currently have coal plants).

It is feasible for solar PV output to provide 100 percent of a state’s energy needs.

Lesson 2: Renewable Energy Adoption Can Be Driven by Government Policy and Market Factors

The South Australian Labor government, elected in 2002, set an initial aim of generating 26 percent of South Australia’s power from renewable sources by 2020. In addition, the Australian Federal Government established a national Renewable Energy Target (RET) in 2001, which encouraged renewable energy projects through a Renewable Energy Certification programme. The national RET was increased in 2009, with the goal of generating 20% of the nation’s electricity.

Much of South Australia’s finest onshore wind resources are located near existing transmission lines that extend 300 kilometres from Port Augusta to Adelaide. As a result, wind energy projects in South Australia were frequently more appealing than in other states. As a consequence, SA may compete on a national scale for RET-incentivized projects. By 2015, wind has provided 30% of SA demand.

Contracts with the state government also aided renewable energy initiatives throughout the state. To satisfy the government’s power demands, the South Australian government signed a 10-year supply agreement in 2020 for 280MW from the Cultana solar project and 100MW from the Playford Utility battery.

By 2016, all coal-fired power plants in South Australia had been phased out. Playford B (240MW) was decommissioned in 2012, while Alinta Energy deactivated Northern Power Station (520MW) in 2016. Prior to shutting down, both facilities were largely used in the summer to satisfy peak energy demand. Wind generating gains, along with decreased demand, have rendered coal facilities unprofitable to operate around the clock.

SA had the highest wholesale energy rates in the mainland NEM prior to the closure of the final remaining coal facilities in 2016. The closure of the Northern facility resulted in a 7.7 percent increase in wholesale prices. Since 2017, wholesale power costs have fallen by 65 percent (2017 – 2021 YTD), and renewables’ share of demand has increased from 43 percent to 60 percent (2017 – 2020). By refusing to subsidise the ageing and unprofitable Northern (coal-fired) Power Station, the SA government facilitated a faster transition to a renewables-dominated grid.

The South Australian experience demonstrates that financially backed government policies (for example, the RET, the state feed-in tariff, and government supply contracts) may promote renewables installations. Allowing unprofitable fossil fuel facilities to shutter when market conditions demand, rather than giving subsidies to keep them open for longer, can lead to increased renewables uptake and a drop in wholesale power prices.

Lesson 3: Ambitious Renewable Energy Plans (500%) Can Drive Economic Growth

While SA is on track to attain 100 percent renewables by 2025, the SA government is now aiming for 500 percent renewables by 2050, with surplus renewable energy exported to neighbouring states and worldwide via green hydrogen and other low-emission goods. This is projected to fuel regional economic growth. To meet the 500 percent renewables target, additional transmission projects will need to be built. There are proposals to build new Renewable Energy Zones (REZs) around the NEM in locations with strong wind and/or solar potential, as well as transmission lines to connect the REZs to load centres.

The lofty SA goal of reaching 500 percent renewable energy is supported by government investment in decarbonization efforts. This involves fostering an atmosphere that attracts foreign investment while also producing local jobs.

SA now has a number of state government schemes in place to assist with the transition, including:

• $100 million Home Battery Scheme.
• Grid Scale Storage Fund of $50 million.
• Renewable Technology Fund of $150 million.
• A $2 million grant with a $20 million loan for a Virtual Power Plant.
• $18.3 million Electric Vehicle Action Plan.

Furthermore, the federal and state governments have agreed to invest $1 billion in hydrogen, carbon capture and storage (CCS), electric cars, and natural gas. While CCS has been chastised for its high cost and gas for its high emissions, green hydrogen and electric cars are projected to be critical facilitators of a low-emissions grid.

Additional investment in new renewable energy generating projects, green hydrogen facilities, and local low-emission product manufacturing will be necessary to attain 500 percent renewables. The South Australian example demonstrates that state-based decarbonisation policies and initiatives have the ability to create considerable economic development.

Lesson 4: Wind and solar power help to lower wholesale electricity prices.

The increased use of renewables in South Australia and the NEM is lowering wholesale power spot costs. Wind and solar have the lowest short-run marginal cost (SRMC) among the NEM’s major generators. This is mostly owing to their zero-fuel cost as well as their cheap operating and maintenance expenses.

In addition, battery storage and large-scale solar PV have the lowest new build costs of any energy-generating technology in the NEM.

Because the marginal cost of solar and wind power facilities is so low, they are typically bid into the market at low, zero, or even slightly negative prices.

Prices in the NEM have been shown to fall in accordance with the merit order effect when low-cost wind and solar are incorporated into the market. Wind and solar bid so low into the market that they put downward pressure on wholesale power spot prices.

In 2017, wholesale energy prices reached an all-time high across the NEM. Prices in the NEM were $103/MWh over this time period, whereas in SA they were $126/MWh. Year-to-date (YTD) prices in the NEM in 2021 were $40 and in SA $44. The NEM has dropped by 61 percent, while SA has dropped by 65 percent (2017 – 2021 YTD).

These significant wholesale energy price decreases are indicative of rising quantities of renewables in the system, with SA witnessing a greater decrease than the NEM generally. Prices in South Australia have now reached their lowest levels since 2015-16. Low prices in SA (and across the NEM) are a result of an increasing quantity of renewables in the system pushing down prices. In the first quarter of 2021, SA had negative spot prices 17% of the time, with the average spot price during peak solar production (10am-3.30pm) being negative $12/MWh.

Rates are falling in the middle of the day, during the highest solar insolation time, when supply is provided at negative prices as generators battle to stay online. In South Australia, for example, prices reached -$150 in the middle of the day on September 18, 2020.

The cost of producing electricity accounts for around 39 percent of the average residential consumer’s bill; the balance is made up of transmission, distribution, retail, and other costs. Residential power rates in South Australia are anticipated to fall by 10.8 percent between 2019/20 and 2022/23, thanks to lower wholesale pricing and lower regulated network and environmental expenses.

The SA example demonstrates that increasing the quantity of variable renewable energy output leads to reduced wholesale power costs, particularly during the day when there is a lot of solar PV output. This translates into retail electricity savings for customers.

Lesson 5: In a high renewables grid, system reliability and security may be maintained.

SA is facing a transition challenge: integrating significant volumes of variable renewable energy and increasingly dispersed energy supplies while preserving system stability and security. Australia has a very high dependability requirement, with 99.998 percent of customer demand satisfied each year. Except for 2008-09, when high temperatures in Victoria and SA limited interconnector availability, and Victorian generators provided 0.004 percent and 0.0032 percent of unserved electricity, respectively, SA has fulfilled its dependability criteria for the previous 15 years.

The shifting dynamics of demand and supply in South Australia are posing operational and planning problems for AEMO in terms of how to manage the grid when demand is exceptionally low. ‘Minimum operating demand’ is reaching new lows in South Australia, most recently on 11 October 2020 at 1pm, when it was 290MW, a new low. AEMO anticipates that minimum demand will continue to fall and may reach nil by 2024-25.

Due to consistently low energy prices below their cost of generating, AEMO directed SA’s gas powered generators (GPGs) for a record 70% of the quarter in 2021. SA is researching and developing numerous technical solutions to support system security and enhance consumer value, such as installing four synchronous condensers (purchased by Electranet at a cost of $166 million to SA consumers over the next 40 years), enhancing connectivity with neighbouring states to improve import and export capabilities, getting contingency frequency reserves from renewable energy large-scale battery storage, as well as putting in place rapid-start and rapid-response technologies.

Regulatory solutions are also being explored or implemented in SA and the wider NEM to support system security, including:

• introducing a mandatory requirement for generators to activate any existing capability to provide primary frequency response,

• modification of inverter requirements guidelines for smaller distribution-connected generation in order to “optimise and maintain a safe power system under high levels of DER penetration, enhance energy affordability, and allow customers to seek personalised services.”

• implementing a wholesale demand response mechanism in the NEM (beginning in 2022), “to allow consumers to sell demand response in the wholesale market directly or through aggregators,”

• limiting imports into SA when SA’s Under Frequency Load Shedding schemes are “ineffective to prevent cascading failures and potential system blackouts.”

Other potential electricity market reform options being investigated by the Energy Security Board (ESB) include mechanisms for resource adequacy, managing ageing thermal generator retirement, procuring essential system services, integrating distributed energy resources and demand side participation, and managing transmission and access.

Many of these rules and market processes will contribute to system security and assist customers in gaining value from their DER resources.

According to a recent IEEFA report, investments in synchronous condensers and the establishment of a market for procuring inertia should be evaluated since they will be redundant in a post-transition grid. To avoid investing in assets that may become outdated in the future, the energy business should plan for a zero-inertia, zero-emissions future dominated by inverter-based resources (IBRs).

The emphasis should be on building the digital technologies required to manage an increasingly complex power grid. To deal with a growing number of variable and dispersed energy resources, the future grid control system will need to be more dynamic and adaptable.

The SA example demonstrates how, by implementing multiple measures, reliability and system security can be maintained in a high renewables grid.

Planning for a digitalized, zero-inertia, zero-emissions future grid is critical to avoiding investment in assets, markets, or laws that may become outdated, as well as to establishing the digital systems necessary for intelligent future grid management.

Lesson 6: Batteries can aid in the maintenance of system reliability and security.

Following multiple power outages in South Australia in 2017, Tesla CEO Elon Musk collaborated with Neoen to build the Hornsdale Power Reserve battery. It was a successful experiment, saving South Australians more than $150 million in the first two years of operation. At the time of its commissioning, the Hornsdale Power Reserve was the world’s biggest lithium-ion battery, measuring 100MW/129MWh. The battery’s extension to 50MW/64.5MWh was completed in 2020, and it is co-located with the Hornsdale Wind Farm.

South Australia presently has four large-scale batteries. In the NEM, distributed batteries may also be used to deliver services. Virtual Power Plants (VPPs) pool dispersed batteries (or other DERs) to provide auxiliary frequency control services (FCAS) and energy market services. AEMO, in collaboration with Tesla, is conducting a VPP experiment in SA to determine how VPPs may be best integrated. The study intends to link 50,000 SA houses to the VPP system.

Lesson 7: Rooftop Solar Penetrations Can Be Managed with Distribution Network Innovation

Rooftop Solar Can Achieve High Penetration

The South Australian government implemented Australia’s first subsidised solar feed-in tariff, which was valid for installations constructed between 2008 and 2013. This early government backing permitted the establishment of a rooftop solar business, which has resulted in the creation of employment and the expansion of an industry that is still growing.

Australia has the world’s most utility and rooftop solar Watts per capita (786W per person as at the end of 2020). Australia presently has over 2.7 million rooftop solar systems with a total capacity of more than 13GW (out of a total PV capacity of more than 20GW). In the calendar year 2020, there was a record pace of 3GW installation, representing a 39 percent rise year on year and the highest capita install rate globally. 

Rooftop solar generates the most electricity in SA of any NEM area, accounting for 13% of total output in 2020. On average, 40 percent of SA houses have rooftop solar, according to local government areas (LGAs) (some LGAs have even higher amounts).

PV systems between 100kW and 30MW, known as PV Non-Scheduled Generation (PVNSG), which are typically installed on commercial and industrial (C&I) rooftops and small solar farms (including on agricultural properties), have experienced rapid growth since 2015, with a 79 percent increase in capacity since 2018-19 to 129MW on 30 June 2020.

Rooftop solar PV power may be integrated into the grid at penetration rates of more than 40% of houses, as demonstrated by the SA example. It is critical to design for high rooftop solar penetration and multi-way flows inside distribution networks.

Voltage rise risk was exaggerated, perhaps leading to questionable investments and regulations.

There have been concerns that high levels of rooftop solar penetration might cause grid voltage to increase. When rooftop solar exports connect into distribution networks, this can happen. The ESB commissioned a research by the University of New South Wales (UNSW), which concluded that even in the absence of rooftop solar, high voltages persist across the NEM and SA.

The nominal voltage standard for the NEM is 230V, with a maximum voltage limit of 253V. Over the course of the year, average maximum voltages in SA are frequently near to the top limitation of 253V. The historic 240V nominal voltage standard, from which distribution networks are still adapting, is to blame for high voltage in the distribution network.

A variety of new rules and regulations will have an influence on rooftop solar exports. SA inverters must be able to be remotely deactivated as instructed by the system operator, according to rules enacted in 2020.

On July 1, 2020, a ‘solar sponge’ network tariff was also implemented. Electricity network prices are now 25% down between 10 a.m. and 3 p.m. (3.6c/kWh vs. 18c/kWh peak). This should promote higher demand throughout the day, when solar output is plentiful. However, benefits have yet to be observed, and persistent problems include the daily application of the ‘solar sponge’ tariff.

It is expected that the adoption of batteries, particularly electric vehicles (EVs), will have the greatest impact on rooftop PV exports, potentially significantly reducing the volume of solar that flows back into the grid and, in the long run, all but eliminating any problematic distributed solar-related voltage rises. Concerns regarding the voltage increase impact of Distributed Energy Resources (DER) on the grid can be overdone, as the SA example demonstrates, and reliable data and experiments are essential to guarantee dangers are not overblown and unneeded solutions are not invested in.

Voltage monitoring, transformer tap changes, and DOEs are all more effective ways to manage rooftop solar

Improved voltage monitoring is required to address pre-existing problems in the SA distribution network. In the Adelaide metropolitan region, SA Power Networks is installing and commissioning permanent remote monitoring at a sample set of roughly 1,300 multi-customer Low Voltage (LV) distribution transformers. At a net cost of roughly $4.5 million over five years, the goal aims to enhance capacity planning in its LV network.

DOEs for each connection point at SA Power Networks are established dynamically at five-minute intervals, 24 hours in advance. This has enabled the 5MW capacity to be expanded to a potential 10MW while maintaining the network safe and secure in a Virtual Power Plant (VPP) experiment spanning 1,000 batteries.

When building or upgrading distribution networks and network management systems, distribution networks should plan for high levels of solar PV penetration, as shown in the SA example. This should incorporate network visibility planning, intelligent management tools, and software to build dynamic operating envelopes.DOEs can control exports to stay within distribution network constraints.

Rooftop Solar Management Isn’t Expensive: In South Australia, it now costs less than 1% of network revenue

SA Power Networks has been allocated $3,914 million in income from its customers by the Australian Energy Regulator (AER) for the period 1 July 2020 to 30 June 2025. SA Power Networks would invest an estimated $32 million in capital expenditures to develop capabilities to roll out DOEs as a regular connection service. Other Australian network firms have similar statistics, with less than 5% of revenue being spent on developing smarts to control rooftop solar for at least the next five years.

Australian distribution firms have given no indication that future rooftop solar integration prices would skyrocket. Indeed, according to Energy Networks Australia’s (ENA) Electricity Network Transformation Roadmap, a renewables-only NEM could save $16 billion in infrastructure expenses by 2050, as well as $414 per year in typical home bills, a decrease of about 30% over 2017. The SA example demonstrates that rooftop solar PV management is not costly.

The above report extract is based on IEEFA’s report on South Australia: A Grid Dominated by Wind and Solar Is Possible. The full report can be accessed by clicking here