How inverter-based power systems are turning grid reliability into a tradable service
For most of the history of electricity systems, stability was not something engineers needed to explicitly procure. It came for free, embedded in the physics of large spinning machines.
Coal, gas, hydro and nuclear plants were not just sources of energy. Their rotating turbines provided inertia that resisted sudden changes in frequency, their electromagnetic behaviour helped stabilise voltage, and their high fault currents made protection systems straightforward. The grid was stable because it was built around machines that were, by their nature, stabilising.
That assumption is now breaking down.
As countries decarbonise, synchronous generators are being displaced by inverter-based resources: wind, solar photovoltaic plants, battery storage systems, and increasingly electrified industrial loads. These technologies are highly efficient and flexible, but they are electrically decoupled from the rotating mass that once held the system together.
The result is not a gradual adjustment, but a regime change in how power systems behave.
Across regions as diverse as Chile, Australia, Spain, and parts of Northern Europe, system operators are encountering the same underlying issue: stability is no longer an automatic consequence of having enough generation online. It is becoming something that must be actively created, managed, and increasingly, purchased.
From machines to software-defined grids
At the centre of this transition is a shift in inverter behaviour.
Most existing renewable generation uses grid-following inverters. These devices act like controlled current sources. They measure the grid voltage, lock onto it using a phase-locked loop, and inject power accordingly. This works well when the grid is strong and predictable.
But when system strength declines, this approach becomes fragile. In weak grids — typically where short-circuit levels fall below certain thresholds — the very act of measurement begins to interact with the system being measured. The inverter is no longer following a stable reference; it is influencing it.
Engineers describe this in terms of short-circuit ratio, or SCR. Below a level of roughly 3, grid-following behaviour can become increasingly sensitive to small disturbances. Voltage and phase estimation begin to interact with network impedance, and oscillations can emerge not from a single fault, but from the interaction of many devices behaving correctly in isolation.
Grid-forming inverters are designed to address this problem directly.
Rather than tracking the grid, they create it. They operate as controlled voltage sources, establishing their own internal frequency and voltage trajectory, and allowing synchronisation to emerge through interaction with other devices.
In effect, the system shifts from one in which devices follow a shared electrical reference, to one in which they collectively construct that reference.
Where the shift is already visible
This transition is not theoretical. It is already visible in how different power systems are operating under stress.
Chile offers one of the clearest examples. Its northern grid, anchored by the solar-rich Atacama Desert, is characterised by long transmission distances and relatively weak local demand. Electricity must travel hundreds of kilometres before reaching major consumption centres. In such conditions, grid strength is structurally low, and stability constraints increasingly influence how much renewable generation can actually be dispatched.
Australia presents a similar but distinct case. In South Australia, high renewable penetration combined with limited synchronous generation has created periods in which system stability depends heavily on inverter behaviour rather than conventional inertia. System operators have already had to intervene to ensure minimum levels of stabilising services are maintained.
Spain illustrates a different form of stress. While part of a highly interconnected European system, certain regions experience weak grid conditions during high renewable output and constrained interconnection with France. Stability challenges emerge not from isolation, but from bottlenecks within an otherwise strong system.
Northern Europe adds another layer. Offshore wind expansion and the growing use of HVDC links are introducing converter-dominated interfaces into what was previously a largely synchronous environment. Stability issues are increasingly local, appearing at the edges of transmission infrastructure rather than across entire countries.
The common thread is that electrical “strength” is becoming spatially uneven, and increasingly dependent on control behaviour rather than purely on physical infrastructure.
A metric that is starting to lose its meaning
For decades, engineers have relied on short-circuit ratio as a simple measure of grid strength. It compares the available fault level in a network to the size of connected converters. High values indicate strong grids; low values indicate weak ones.
But in inverter-heavy systems, this metric is beginning to lose its clarity.
SCR was developed for a world in which generators injected current into a voltage set by something else. It does not fully capture a system in which multiple devices are simultaneously shaping that voltage through software-defined control.
As a result, SCR is no longer purely a property of the network. It is becoming a function of both network structure and inverter behaviour — a moving boundary rather than a fixed characteristic.
Below certain levels, typically between 1.5 and 3, system behaviour becomes highly sensitive not just to physical configuration, but to control settings inside inverters: their response speeds, their damping characteristics, and the way they share power with one another.
Stability as an interaction problem
In the synchronous machine era, stability was largely a matter of physical response. Turbines resisted frequency changes; system inertia smoothed disturbances; protection systems reacted to predictable fault currents.
In inverter-dominated systems, stability becomes an interaction problem.
Each grid-forming inverter behaves like a controlled oscillator, with its own internal notion of frequency and voltage. When many such devices operate together, stability depends on how these oscillators interact through the electrical network.
If well coordinated, the result can be remarkably robust: fast frequency response, strong voltage control, and the ability to restart systems without external assistance. If poorly coordinated, the same interaction can produce low-frequency oscillations, circulating power flows, or subtle instabilities that are difficult to detect using traditional planning tools.
The difference lies not in individual devices, but in their collective behaviour.
The hidden layer: impedance
Beneath this interaction lies a less intuitive but increasingly important concept: impedance.
Every inverter presents an effective electrical “signature” to the grid, shaped not only by hardware but by control software. This includes how quickly it responds to changes, how it limits current, and how it behaves under disturbance.
When many inverters are connected in weak or long electrical networks, these signatures interact. In some cases they reinforce stability. In others, they create resonances that were not present in the physical network itself.
This is one of the reasons system operators are increasingly relying on more detailed modelling techniques in weak-grid conditions. The behaviour of the system is no longer defined solely by wires and transformers, but also by the collective dynamics of distributed control systems.
When fault current is no longer a signal
Perhaps the most immediate operational consequence of this shift lies in protection systems.
Traditional grids rely heavily on fault current magnitude to detect and isolate faults. Large synchronous machines naturally produce high fault currents, making this approach reliable and intuitive.
Grid-forming inverters do not.
They are typically limited to near-rated current levels even during faults, which is necessary to protect semiconductor devices. But this also means that fault current is no longer a reliable indicator of fault location or severity.
As a result, system operators in countries such as Australia and Chile, and increasingly across Europe, are moving toward alternative protection philosophies based on voltage behaviour, differential measurements, and communication-assisted schemes.
This is not an incremental change. It is a redesign of one of the most fundamental layers of power system operation.
Stability as a service
Taken together, these changes point to a broader shift in how electricity systems are designed and operated.
Stability, once an implicit by-product of generation technology, is becoming an explicit service.
System operators are increasingly required to ensure not only that there is enough energy available, but that there is enough “system strength” — a combination of inertia-like response, voltage stiffness, and fast frequency support.
In practice, this is already leading to procurement of non-traditional assets. Battery systems are being contracted not just for energy shifting, but for fast frequency response and stability support. Synchronous condensers are being installed not for generation, but for fault level reinforcement. In some markets, grid-forming capability is beginning to be specified explicitly in grid connection requirements.
What was once invisible is becoming contractual.
A different kind of electricity system
The deeper change, however, is not in markets or equipment, but in system architecture.
Electricity networks are becoming less like mechanical systems and more like software-defined systems. Stability is no longer an emergent property of physics alone, but the outcome of many interacting control systems operating across a shared network.
Grid-forming inverters sit at the centre of this transition. They do not simply replace traditional generators. They redefine what it means for a power system to be stable.
The challenge ahead is not whether these technologies work in isolation — they do — but whether thousands of them, built by different manufacturers and deployed across different regulatory environments, can collectively behave as a coherent system under stress.
In that sense, the energy transition is doing more than decarbonising electricity supply.
It is changing the nature of stability itself.
Author: Derek Michalski, Editor
Read this article on Substack
