By Phil Kreveld
“The[re is] uncertainty in future dispatch patterns. While market modelling outcomes do provide some information about the future generation dispatch pattern and power flow conditions future generation dispatch pattern and power flow conditions could fall outside the operating envelope of the power system. These would require careful examination and nontrivial modifications to be translated for power system studies.” — AEMO Inaugural Transition Plan for System Security
The above quote from the Australian Energy Market Operator (AEMO) merits attention. AEMO is the one body which is critically important to the renewable energy sources transition. It has the ineluctable task of keeping the complete south-east coast alternating current (AC) grid not only operating reliably (minimum unserved electricity) but also securely (tightly controlled voltage and frequency).
Anyone who thinks the transition to ambitious renewable targets is only subject to regulations, and markets operating ‘properly’, with the physical side ‘falling into place’ to suit, should read on. A note of caution: there are selective quotes from the inaugural transition plan throughout the article but it is strongly advised that interested readers, read the AEMO report in its entirety.
Related article: Sailing between Scylla and Charybdis: the renewables Odyssey
The National Electricity Market (NEM) is an alternating current system: as such it is an airplane in flight. And like modifications to airplanes in flight, extreme care must be taken not to bring the system down in the process. The forerunner to the NEM and the present day version took off, decades and decades ago, running uninterruptedly ever since, and these days not only refuels ‘in flight’ with gas, coal and hydro, but also solar, and wind. Were it to land (in other words, cause a black out), getting up and running again, nationally could take days!
The consequences are:
- All coal-fired generators must be supplying power—they cannot all be turned off, and they cannot be just spinning without supplying power to the grid because that makes their steam control impossible. Nuclear generation would also have the same restrictions;
- Hydro and gas-fired generation can be turned off provided that there are sufficient other synchronous energy sources connected to maintain voltage and frequency stability.
Based on the points above, and in today’s grid technology, only renewable energy that can be turned on and off or curtailed without crashing the system are big batteries, big wind and solar, and consumer energy resources (CER), i.e. rooftop solar.
No guesses as to the priority order!
At present household and business CER, would have to be restrained from sending power into the transmission lines (although in sub-transmission, between terminal stations this can occur), potentially yielding repeats of Callide C (attempting to run a generator in reverse as a motor)—due to the malfunction of a reverse current relay.
So, what if the coal and gas were turfed out, and replaced with batteries? In principle that is not a problem—allowing power to flow both ways (charging as well as discharging).
Inverter engineers at the USA-based Electric Power Research Institute (EPRI), and the UNIFI voltage forming inverter consortium are very positive about replacing synchronous machines with inverter-based resources (IBR). However, controlling these in Australian grids with generators separated by hundreds and thousands of kilometres, is a major challenge to the system operator.
Please look at the illustration (Fig 1): in the bottom LH corner is an AC voltage waveform. A synchronous, 2-pole generator running at 3000 revolutions/minute would generate the sinewave with its peak and trough, spanning 20 milliseconds (360°) in one complete revolution. Inverter based resources (IBR) have to duplicate this via electronic control circuits in order to reliably (co)-operate:
- (with synchronous machines) already in the system (AEMO terms this ‘Horizon-1’);
- with much reduced synchronous generator capacity—‘Horizon-2’;
- ultimately with less than 10% synchronous capacity, and relying virtually completely on IBR—‘Horizon-3’.
In days of yore, frequency control comprised of two main stages; governor control and automatic generator control. The former would open or close steam valves to allow turbine-generator sets to speed up with imposition of a sudden load increase; the latter would make the final correction. Tie lines between generators made synchronisation between generators easier but were abolished when the energy market came into force, Frequency control is no longer the province of governors; reliance is placed on inertia and frequency control ancillary system (FCAS) pumping energy into the system. Replacing synchronous machines with inverters can only widen the gap unless a control system (not yet in existence) is implemented in which voltage and frequency control are the top priorities—and would have to be paid for (see fig 2). AEMO will be challenged to contain the frequency spread as the NEM increasingly has IBR as primary energy sources. It is more than likely that FCAS will not do the job in 100%, or close to, inverter-based generator grids.
A brief explanation:
| Synchronous generator | Inverter (IBR) |
| Detects change in revs/min (frequency) instantaneously | Detects change in frequency with time delays from 10 (grid forming) to 100 milliseconds (grid following) |
| Consequence: ideal for preventing frequency nadirs (low point) being reached with the result of load shedding | Consequence: whole of system analysis (overriding market design) necessary including distribution network underfrequency load shedding relays and necessitating complicated, new control schemes of rooftop solar. |
For the technically inclined, the diagram illustrating power flow in a small, theoretical grid explains the essence of maintaining power flow into distribution networks, if life is not to become very, very hard for AEMO as the systems operator. See the quote from the AEMO report, mentioned above.
Transition points may take the form of:
- Major changes in the asset mix or configuration of the NEM, such as large synchronous generators retiring or regularly decommitting from the market
- Threshold events, for example operational demand falling below regional and NEM-wide minimum secure levels for maintenance of voltage and frequency control, driven by distributed photovoltaic (DPV) generation peak output growth in conjunction with low underlying consumption, and
- Operational shifts as changes in available security mechanisms allow relaxation of constraints such as minimum numbers of online synchronous generators.
As the NEM progresses from the Horizon-1 scenario, it is evident that control beyond energy market operation will be necessary in AEMO’s control room. Tim Jordan, a commissioner of the Australian Energy Market Commission (AEMC), will be examining new market structures—although it appears without the benefit of engineering input. However, Jordan’s remarks were rather carefully calibrated on the question of the growth of consumer energy resources (CER) and the effect on large scale generation. He deferred to the ‘magic’ of smart meter deployment and virtual power plant (VPP) as somehow providing a solution in the future. Maybe the AEMC has proposals in waiting—but just as likely it’s a fuzzball statement.
Related article: We have the bull by the tail
Modifying the national electricity grid of the NEM and WEM, ‘in flight’, will require control systems separate to the energy dispatch engine of AEMO. This might sound satisfactory but as a result frequency bandwidth has widened.
Lost in the day-to-day chatter about the effect of the renewable transition on electrical energy prices is the influence of over-capacity of essentially free sunshine and wind, and the disruption household and business rooftop solar is causing because of large daytime periods of energy independence.
For those of an academic frame of mind, Tom Brown’s (TU Berlin) paper, ‘Price formation without fuel costs: the interaction of elastic demand with storage bidding’ might be of interest but the real question begged is whether potential investors will turn into real ones based on this highly theoretical paper.
Other than a reference to ‘price formation theory’, there was no bankable response from the Commissioner—unsurprisingly. The successful bidders in the Capacity Insurance Scheme (CIS), who provide base levels of megawatts with the Commonwealth in contracts for difference are betting that taxpayers will be funding energy price difference to the contracted base levels, i.e. curtailment will see to ‘low price formation’.
In all of these discussions, the takeaway is that technology will mould itself around market designs. It is an unrealistic expectation.
Modifying the airplane in flight safely will not be done by consulting the passengers when they would like to reach their destination (renewable energy targets), rather the flight crew will have to be in total control.
