Transient Stability: What It Is and What Decides It

Short answer

A system is transiently stable if, after a large disturbance such as a fault, its generators recover a stable operating state without losing synchronism. It is decided mostly by how fast the fault is cleared. Grid codes set the criteria and ask for the machine data needed to study it.

What transient stability is

Grid codes define transient stability along these lines (the same idea appears in IEEE/CIGRE stability definitions): a power system is transiently stable if, for a stable operating condition and for a particular disturbance, it reaches an acceptable stable operating condition after the disturbance. A typical disturbance is a fault on a line that is then cleared.

The key words are «a particular disturbance»: stability is not a property of the system in general, but of an operating condition facing a specific event. A system can be stable for a fault on one line and not for another closer to the generator.

Codes also define steady-state stability, which looks at the opposite: whether, after a small disturbance, the system reaches an operating condition similar to the previous one. This covers the slow oscillations that must be damped.

What happens when a line faults

A synchronous generator delivers electrical power according to the grid voltage and the angle between its rotor and that grid. The turbine supplies mechanical power. In normal operation, the two powers are equal and the angle holds.

When a nearby short circuit occurs, the voltage drops and the generator can deliver only part of its electrical power. The turbine keeps pushing, the rotor accelerates and its angle with respect to the grid increases. When the fault is cleared, the electrical power rises again and brakes the rotor. There are two outcomes:

The rotor brakes in time. The angle reaches a maximum and then oscillates around its operating value until it damps out. The system is stable.
The rotor passes the limit. The angle keeps growing, the generator loses synchronism with the grid and must be disconnected. This is where the out-of-step protection (78) acts.

The equal-area criterion summarizes this contest: the kinetic energy the rotor gains during the fault must be returnable after it is cleared. The critical clearing time is the longest fault duration for which this is still met. That is why protection speed is a stability requirement. Grid codes state it in their general protection requirements: the operating time must be short enough to guarantee system stability.

What happens to the generator in a nearby fault

Fault on the grid. Voltage drops and the generator delivers less electrical power
The rotor accelerates. The turbine keeps supplying the same mechanical power
Cleared before the critical time? Yes: continue; No: go to step 6.
Electrical power brakes the rotor. The angle oscillates and damps out
Stable system. The generator stays in service
Loss of synchronism. The generator disconnects

Which criteria grid codes set

In the general criteria for planning the operation of the power system, grid codes usually establish, among others:

A three-phase-to-ground fault is assumed on a transmission circuit, near the substation with the highest short-circuit level. The main protection clears it and the faulted element is permanently taken out of service. With that event, the system must remain stable.
Rotor angles of the machines must swing coherently and damped with respect to a reference.
The currents and impedances seen by neighboring relays must not cause the tripping of additional elements, because that would start a cascade of events.
On the main transmission buses, the transient voltage must not stay below a defined threshold for longer than a defined time.
In small-disturbance analysis, the eigenvalues must have a damping component. If they do not, the unit control systems are tuned and, as a last resort, transfers are limited.

The same codes usually define the transfer limit of a line as the lowest of the conductor thermal limit, the capacity of the current transformers, the voltage regulation limit and the transient and dynamic stability limit. In other words, on a long line, stability can be the governing constraint. The thresholds and times are those of the grid code that applies to the project; we adapt the study to the code and the requirements of the utility.

What data a stability study needs

A stability study uses a time-domain simulation with phasor models (RMS simulation), which is faster than an electromagnetic transient simulation. The data that planning codes ask for in these studies give an idea of what is modeled:

Unit inertia (H), in seconds and on a 100 MVA base.
Transient and subtransient reactance of each unit.
Parameters of the detailed machine model and of its governor, in both hydro and thermal plants.
The network, with the load flow and short-circuit model: lines, transformers, loads and voltage regulators.

The events must also be defined: where the fault is simulated, how long it lasts, which protection clears it and whether there is reclosing. Without that, the result answers nothing. The load flow and short-circuit studies provide the starting point. How to specify and review a software-based study is covered in Power system studies with software: what to specify; the models of transformers, lines and equipment for fast phenomena are in EMT study models: transformers, lines and equipment. On the frequency response of a grid with high renewable generation, read frequency response with high renewable penetration.

What results it delivers

Time evolution of the angle, speed and voltage of each machine for each event.
Critical clearing time of each fault studied, to compare it with the real time of the protection and the circuit breaker.
Cases that do not comply and the possible measures: speed up clearing, add teleprotection, adjust the voltage regulator or limit the transfer.
Need for special schemes, such as remedial action schemes, which act to prevent loss of stability.

Where it shows up in practice

Fault clearing. The maximum clearing times come from these studies. They set the relay speed, the type of teleprotection and the breaker time. That is why a slow relay or a teleprotection out of service is not only a selectivity problem.

Power swings. A stable swing can be seen as a fault by a distance relay. The power swing blocking (68) prevents that trip.

Load shedding. The automatic under-frequency load shedding scheme is usually defined by the system operator from dynamic stability studies. See under-frequency load shedding.

Plants with their own generation. A synchronous generator in the plant, connected to the utility grid, can lose synchronism for a grid fault if the fault is cleared slowly. For solar and wind plants, grid codes usually ask the system operator for RMS and EMT models for its stability and transient studies.

Common mistakes

Studying a single fault. Stability is verified with several events and locations; the most severe fault is not always the most obvious.
Using catalog clearing times instead of the real ones of the protection and the breaker.
Modeling machines without governor or voltage regulator. Their response changes the result.
Ignoring reclosing. A fault that repeats on re-energization is a second disturbance.
Confusing stability with short-circuit capacity. They are different studies that use the same model.

Stability studies with Tiza Group

We build the model, validate the base case and simulate the events your utility or your project needs, using the real clearing times of your protections. See our grid connection studies and power system studies, or write to us through contact. Studies are done remotely, worldwide.

Frequently asked questions

What is the difference between transient stability and small-signal stability?

Transient stability studies the response to a large disturbance, such as a three-phase fault, and checks whether the generators keep synchronism. Small-signal stability studies small disturbances through the eigenvalues of the system and checks that oscillations are damped.

Why can a longer fault make a generator lose synchronism?

During the fault, the generator delivers less electrical power than it receives from the turbine and accelerates. The longer the fault lasts, the further its rotor advances relative to the grid. If it passes a certain angle before the fault is cleared, it cannot recover synchronism.

Does an industrial plant with its own generation need to study it?

It depends on what the utility requires and on the capacity of the plant. If the plant has synchronous generators connected to the grid, the study verifies that they do not lose synchronism for grid faults and defines the maximum clearing time.