By Phil Kreveld
A public lecture at the University of Melbourne by Dr Janusz Bialek of the Imperial College, London brings to mind Odysseus’ challenge of navigating his sailors safely through the perilous waters populated by the monster-populated islands of Scylla and Charybdis.
On one are the threats of harm posed by their inhabitants of only inverters; on the other, the danger of the ‘golden fleece’ of renewables being snatched out of crew’s hands. The churning waters will drive the vessel to one or the other island. A dilemma presents; try and avoid either in the hope for a safe, congenial solution beyond the two challenging choices—or set the rudder towards one, in the expectation that it will be the lesser of the grave consequences to be encountered.
Related article: A transmission line is not an extension cord
Australian energy planners are the master and helmsman; the engineers are the hapless crew anxiously peering over the gunwales. Their cries are lost in the howling storms set upon them by Canberra gods and the minor gods of the states. Wishful thinking on reaching whatever challenging target will see a stranding, either on experimental, mercurial grids given to inexplicable failures; or one extracting horrendous financial sacrifices. The master and helmsman need to seek the counsel of the crew and to take them into their confidence.
Enough hyperbole; the essence of Bialek’s presentation is that we face an unavoidable choice—head for the inverters or avoid meeting their challenge—in short, clearing their small signal and transient stability hurdles. The other path would see a recourse to keeping the technology of old and patching renewable inverters on to this comfortable matrix with well-rehearsed stability conserving procedures. Both involve unknowable capital investment in addition to generators, be they wind, solar and batteries.
Beyond generation is the very hazy issue of grid augmentation and overall control of a futuristic electricity system. It is the right adjective: there are theoretical studies aplenty but not the solace conferred by practical experience. The patchwork solution—knitting inverters onto a comforting matrix of synchronous machines—is much less experimental and horrendously capital intensive. AEMO, a year and a half ago, launched a number of 40 or so being necessary. Lower numbers have also been mentioned but it is likely to be some billions of dollars—and long, long supply waiting lines.
The UNIFI consortium of inverter manufacturers is involved in voltage forming inverters for grid connection. Worthy effort, but to be borne in mind is that overall control schemes of entirely inverter-based resources grids, including the proportions of grid forming to grid following IBR is nebulous at best. Bialek commenced his talk with the victory for alternating current Nikola Tesla gained over direct current of Thomas Alva Edison. We live with that victory today—and its peculiar requirement for very tightly constrained frequency variation.
There is a fundamental relationship between power variation and frequency.
The above formula, which we can take to apply for a complete electricity system, indicates that any time, generated power Pg and consumed power Pc differ, there is a rate of change of frequency, RoCoF. The symbol K, includes a weighted summation of inertia in the grid system. Inertia, something this writer suffers from, is simply a resistance to change.
However—electrically there is a crucial difference between the old-fashioned synchronous machines and IBR, wind generators are a mix between the old and the new.
The essence of control, Bialek pointed out, is in nimbleness of response to frequency change because that will remain essential. This gets us back to our renewable odyssey. The physical inertia of heavy rotating masses (turbine-generators) constrain RoCoF initially, allowing the turbine governor to change steam valve settings. Thus, for synchronous machines the response is immediate, RoCoF, being turbine-generator deceleration or acceleration. The detection of speed change is immediate!
Not so for IBR, for example in grid supporting ones, some 100 milliseconds or more are necessary to establish that a frequency change has happened before a response takes place. For grid forming, the time is less but it is never immediate. In practice, this can mean that with big swings between generated and consumed power (generally sudden spurts in consumed power), frequency may have dropped to a low value (nadir) where under-frequency load shedding preserves the stability of the remainder of the electricity system, i.e., a lack of nimbleness with unfortunate consequences.
If we retain synchronous machines, whether as generators or synchronous condensers, we will slow the renewables transition and chose what we hope is the less hazardous landing, paying an enormous insurance premium. If we head to the other side of the straits, there are the perils of insufficient knowledge. For example, IBR have secret control circuitry (degrees of nimbleness) because that is proprietary information for the OEM. Instead, a black box is provided for testing in grid models! The black box is meant to indicate a realistic performance but it has already been seen many times that often it not so! Not only that, the grid offered for the test is a digital ‘twin’ of the NEM grid—but who says it is?
A digital twin of 500 generator bus bars and 3,000 load busbars has a mechanical alter ego, a set of inter connected springs with changing weights and fluctuating forces. In other words, there is no limit to the number of test scenarios an individual generator, seeking connection, should or could be exposed to because the test results depend on all the other generators and loads in the system. It’s not the way connection approvals are given—nor would it be practicable.
Related article: Why is it so?
Returning for a moment to the challenges of Odysseus: If we head for IBR shores new control paradigms are needed because of increased speed of response in the absence of physical inertia. It would be in pursuit of an essential recognition that the world of electrical engineering is changing fundamentally, that electrotechnology is going from cumbersome electromagnetic machines to ones comprising of power electronics. If we set course for the other shore so as to avoid the challenges of the new, we may well stumble badly by virtue of having wagered out future on dying technology.
The engineering crew should appeal to the warring gods to rest their battles—and to let sweet reason and rational thought prevail.
Even then, the transition journey will be difficult enough!
