Transient Stability Analysis: An Engineering Guide for US Power Plants, Transmission Systems and Industrial Microgrids

Part A — Transient Stability at a Glance

Every synchronous generator connected to the US grid — from a 13.8 kV cogeneration unit at an industrial plant to a 1,300 MVA nuclear unit feeding a 500 kV switchyard — spins in lockstep with every other synchronous machine in its interconnection. In normal operation, the mechanical power delivered by each turbine equals the electrical power the generator sends to the network, and the rotor holds a steady angle relative to the rest of the system. A severe disturbance breaks that balance. During a nearby short circuit, the generator can no longer deliver its electrical power, but the turbine keeps pushing. The rotor accelerates, its angle advances, and the question becomes simple and unforgiving: when protection clears the fault, will the network be able to pull the rotor back, or will the generator slip a pole and lose synchronism?

Transient stability analysis answers that question. It is one of the core power system studies behind every generator interconnection, every transmission plan, every protection scheme for a large plant and every industrial or data center microgrid that can island from the utility.

What a transient stability study evaluates

The core relationships

• Swing equation: rotor acceleration is proportional to the difference between mechanical input power and electrical output power.
• Power-angle relationship: in the classical model, electrical power transfer equals E′V ÷ X × sin δ.
• Equal-area criterion: a single machine remains stable if the decelerating energy available after clearing is at least as large as the accelerating energy gained during the fault.
• Critical clearing time: the longest fault duration for which the system remains stable for a given operating condition and disturbance.

Keentel's worked example in one table



Keentel Engineering analyzed a 600 MVA generating unit delivering 540 MW through a step-up transformer and a 345 kV double-circuit line to a strong system. A bolted three-phase fault at the plant's 345 kV bus is cleared by tripping one circuit.

Figure 1. Top: power-angle curves before, during and after the fault, with accelerating area A1 and decelerating area A2 at the critical clearing angle. Bottom: rotor angle swing curves for clearing in 4, 7 and 8 cycles.

Multiplying by rotor speed converts torques into powers. Normalizing to the machine rating introduces the inertia constant H, defined as the stored kinetic energy at synchronous speed divided by the machine's MVA rating:

Expressing the rotor position as the electrical angle δ relative to a synchronously rotating reference gives the familiar per-unit swing equation:

where H is in seconds on the machine base, ωs is synchronous electrical speed (2π × 60 = 377 rad/s in the US), δ is in electrical radians, Pm and Pe are mechanical and electrical power in per unit, and D represents damping from damper windings, loads and controls.

The form shown without a damping term — 2H/ωs × d²δ/dt² = Pm − Pe — is the classical swing equation. It is correct for first-swing analysis but omits damping, so it cannot by itself show whether oscillations decay after the first swing.



2.2 Practical checks on units

When machines are combined or converted to a common system base, the inertia constant scales with rating:

2.3 Typical inertia constants

Higher inertia slows the rotor's response, which gives protection more time. A reciprocating engine plant with H below 1 s can lose synchronism several times faster than a large steam unit for the same per-unit disturbance one reason industrial microgrids often need much faster clearing than utility transmission.

3. The power-angle relationship

3.1 The classical model



The classical model represents the generator as a constant voltage E′ behind its direct-axis transient reactance X′d, connected through a network reactance to an infinite bus of voltage V. Neglecting resistance, electrical power is:

where X is the total transfer reactance between the internal voltage and the infinite bus. Maximum power occurs at δ = 90°:

3.2 Where the classical model is not enough

The simple sin δ curve with a 90° peak is a useful teaching tool, but real studies must go further:

• Salient-pole machines: hydro generators have different direct- and quadrature-axis reactances. A reluctance term adds a sin 2δ component, and the peak occurs below 90°.
• Flux decay and excitation: E′ is not truly constant. Fast static exciters can raise field voltage during a fault and increase post-fault power transfer; slow exciters allow E′ to decay.
• Network losses and loads: resistance and loads shift the curve and the peak angle.
• Multi-machine systems: what matters is the angle between groups of machines, not the angle to an ideal infinite bus.
• Governor response: mechanical power is not exactly constant over several seconds.

For this reason, US planning studies use detailed generator models (GENROU, GENSAL and GENTPJ in PSS®E and PSLF), excitation system models from IEEE Std 421.5, governor models and power system stabilizer models, all validated under NERC MOD-026 and MOD-027 requirements.

4. The equal-area criterion

4.1 The principle

For a single machine connected to an infinite bus, multiplying the swing equation by dδ/dt and integrating gives a powerful result. The rotor returns toward equilibrium only if the energy it gains while accelerating can be removed while it decelerates. Graphically, on the power-angle diagram:

• Accelerating area A1: between the mechanical power line and the during-fault power curve, from the initial angle δ0 to the clearing angle δc.
• Decelerating area A2: between the post-fault power curve and the mechanical power line, from δc to the maximum angle the rotor reaches.

The system is stable if the available decelerating area is at least as large as the accelerating area:

The largest possible decelerating area ends at δmax, where the post-fault curve crosses the mechanical power line again on its falling side:

4.2 Critical clearing angle



Solving the equal-area condition at the limit gives the critical clearing angle. When the during-fault and post-fault maximum powers are r1 and r2 times the pre-fault maximum Pmax:

4.3 Critical clearing time



For a bolted three-phase fault at the generator's high-voltage bus, electrical power during the fault is zero (r1 = 0), and the swing equation can be integrated directly. The rotor angle grows quadratically with time:

so the critical clearing time is:

For any other fault a fault partway down a line, an unbalanced fault, or a fault with power still flowing there is no closed-form solution, and the critical clearing time is found by numerical simulation.

5. Worked example: a 600 MVA unit on a 345 kV double-circuit line

5.1 System data

5.2 Step 1: pre-fault conditions



The pre-fault external reactance, with both circuits in parallel, is 0.15 + 0.20 + 0.05 = 0.40 pu. Solving the load flow for 540 MW at 1.0 pu terminal voltage gives a generator reactive output of about 0.17 pu (100 Mvar) and a stator current of 0.92 pu. The internal voltage behind transient reactance is:

giving E′ = 1.084 pu at an angle of 35.5° relative to the infinite bus.

The pre-fault angle confirms the operating point: δ0 = sin⁻¹(0.90 ÷ 1.55) = 35.5°.

5.3 Step 2: maximum angle and critical clearing angle

With post-fault Pmax = 1.20 pu:

• δmax = 180° − sin⁻¹(0.90 ÷ 1.20) = 131.7°
• r1 = 0 and r2 = 1.20 ÷ 1.55 = 0.78

Applying the critical clearing angle equation (or equivalently setting A1 = A2 directly on the post-fault curve) gives δcr = 53.9°.

The accelerating area at the limit is Pm × (δcr − δ0) = 0.90 × 0.321 rad = 0.289 pu·rad, matched exactly by the decelerating area under the post-fault curve between 53.9° and 131.7°.



5.4 Step 3: critical clearing time

The critical clearing time is 0.123 s, or 7.4 cycles at 60 Hz. Keentel's numerical integration of the swing equation gives the same result to within 0.1 ms, confirming the analytical method.



5.5 Step 4: what the clearing time means

Modern 345 kV protection with communication-assisted tripping and 2- or 3-cycle breakers typically clears faults in about 3.5 to 5 cycles, so this unit is stable for normal clearing. But if a breaker fails and breaker failure protection must trip adjacent breakers, total clearing time often reaches 10 to 15 cycles — beyond the 7.4-cycle limit. That delayed-clearing event, required by NERC TPL-001, would drive design changes such as faster breaker failure schemes, independent pole tripping, power system stabilizer tuning or a lower operating limit.

5.6 Sensitivity to inertia and loading

Two conclusions follow. First, CCT is much more sensitive to loading than to inertia: increasing output from 90% to 100% cuts the clearing time margin by almost 30%. Second, critical clearing time is not a property of the generator alone; it belongs to a specific operating condition, network topology and disturbance.

5.7 Fault location matters

If the same three-phase fault occurs at the middle of one 345 kV circuit instead of at the plant bus, some power still flows during the fault. The during-fault transfer reactance rises to 2.35 pu, so the during-fault Pmax is about 0.46 pu instead of zero. Keentel's simulation gives a critical clearing time of about 0.185 s, or 11.1 cycles — 50% longer than for the close-in fault. That is why stability studies always include close-in faults at the plant's high-voltage bus: they are usually the most severe.

6. What happens during and after a fault



The sequence of events follows four stages:

Three points are often overlooked:

• The angle keeps increasing after clearing. The rotor is above synchronous speed when the fault clears, so the angle continues to advance until the decelerating energy brings speed back to synchronous. The maximum angle occurs after clearing, not at clearing.
• Unbalanced faults are less severe. Single-line-to-ground faults leave two phases transferring power, so less acceleration occurs. Three-phase faults are typically the design basis for normal clearing; single-line-to-ground faults are often the design basis for delayed clearing in planning criteria.
• Mechanical power is not instantaneously constant. Steam and gas turbine governors respond within seconds; fast valving can reduce mechanical power within a fraction of a second on some steam units, improving stability.

7. Multi-machine systems

Real interconnections contain thousands of machines. For multi-machine systems, stability depends on whether groups of machines separate from each other. Engineers examine:

• Relative rotor angles: the spread between the most advanced and least advanced machine angles, commonly referenced to a center-of-inertia angle.
• Coherent groups: machines that swing together, such as all units at one plant or in one region.
• Inter-area modes: groups of machines in different regions oscillating against each other at 0.1 to 0.8 Hz.
• Local plant modes: a plant oscillating against the rest of the system at roughly 0.8 to 2 Hz.

The center-of-inertia angle is:

Commonly used screening criteria flag instability when relative rotor angles exceed about 180° or when angles diverge without bound. Utilities and planning coordinators also apply damping criteria, such as requiring oscillations to decay to a defined percentage within a set time, and voltage recovery criteria.

8. Modeling for US transient stability studies

8.1 Generator models

8.2 Controls

• Excitation systems: static and brushless exciters represented by IEEE Std 421.5 models such as ST4C, ESST1A, AC7B and EXST1, with their limiters.
• Power system stabilizers: PSS2A, PSS2B and PSS2C models add damping to electromechanical oscillations. Many US utilities and regions require PSS on large units.
• Governors and turbines: steam, gas turbine and hydro governor models such as IEEEG1, GGOV1 and HYGOV, with deadbands and droop consistent with FERC Order No. 842 requirements for new resources.
• Inverter-based resources: WECC generic models (REGC, REEC and REPC) in positive-sequence studies, and vendor-specific models in EMT studies.

8.3 Loads

Loads strongly affect transient stability. Composite load models represent induction motors that slow down and stall during faults, drawing high reactive current and depressing voltage. Static constant-impedance load assumptions can produce optimistic results.

8.4 Protection

Relays that may operate during the simulation must be represented: distance relays that could trip on stable swings, generator out-of-step relays, under- and over-frequency and voltage relays, and IBR ride-through settings. NERC PRC-026 requires owners of certain elements to evaluate whether their relays could trip on stable power swings identified in planning studies.

8.5 Model validation

Model quality determines study quality. NERC MOD-026 (verification of excitation and voltage control models) and MOD-027 (verification of turbine-governor and active power models), and the successor standard developed to consolidate them, require generator owners to provide models validated against measured data. Event playback using PMU and DFR recordings is increasingly used to confirm model accuracy.

9. Positive-sequence RMS versus EMT simulation

Classical transient stability problems involving synchronous machines are well suited to RMS simulation. As inverter-based resources replace synchronous generators, many US transmission providers require EMT studies alongside RMS studies, especially where short-circuit ratio is low.

10. US planning criteria and standards

10.1 NERC TPL-001

NERC TPL-001, Transmission System Planning Performance Requirements, defines the contingencies that planners must study and the performance required. Key categories include:

TPL-001 requires that generators remain stable and that voltage oscillations be adequately damped for the specified events, with performance requirements that differ by category. Many planning coordinators, ISOs and utilities add specific criteria for voltage recovery, damping and relay margin.

10.2 Other NERC and FERC requirements

• FAC-002: studies of new or materially modified facilities, including stability effects of generator and transmission interconnections.
• PRC-024 and PRC-029: generator and IBR frequency and voltage ride-through, which must be reflected in stability simulations.
• PRC-026: relay performance during stable power swings.
• PRC-019: coordination of generator voltage regulating controls, limiters and protection.
• MOD-032, MOD-026 and MOD-027: model data and model verification for planning studies.
• FERC Order No. 2023: interconnection process reforms affecting study timing, cluster studies and model requirements.

10.3 IEEE guidance

• IEEE Std 1110: guide for synchronous generator modeling practices in stability analyses.
• IEEE Std 421.5: excitation system models for power system stability studies.
• IEEE Std C37.102: guide for AC generator protection, including out-of-step (78) protection.
• IEEE Std 3002 series (successor to IEEE Std 399, the Brown Book): recommended practices for industrial and commercial power system analysis, including stability studies.
• IEEE PES Power System Relaying and Control Committee reports: power swing and out-of-step considerations for transmission protection.

11. Critical clearing time and protection design

11.1 Components of total clearing time

11.2 Stability margin



A study should not stop at a yes/no answer for normal clearing. Engineers compare the actual clearing time with the critical clearing time:

For the worked example, normal clearing at 4 cycles gives about 46% margin, while a 12-cycle stuck-breaker case has negative margin and is unstable. Many utilities set minimum margins for critical plants, often expressed in cycles above the expected clearing time.

11.3 Out-of-step and power swing protection

When a generator loses synchronism, it slips poles and produces large current and torque pulsations that can damage the machine and shaft. Generator out-of-step protection (device 78) detects the apparent impedance trajectory crossing the generator and transformer and trips the unit, ideally at a favorable angle to limit breaker stress. On transmission lines, power swing blocking prevents distance relays from tripping on stable swings, while out-of-step tripping separates the system at planned locations if instability does occur. Settings must come from the same stability simulations, using the apparent impedance trajectories of stable and unstable swings.



12. Methods for improving transient stability

13. Grid-connected versus islanded operation

Industrial plants, refineries, data centers, universities and hospitals often own synchronous generators that run in parallel with the utility and may need to island during a disturbance. Their stability behaves very differently in the two modes:

13.1 Islanding studies

A study for a facility that can island typically evaluates:

1. Separation detection: how the facility detects the grid event, using under-frequency, under-voltage, rate-of-change-of-frequency or transfer trip signals.
2. Separation timing: how quickly the facility breaker opens, and whether local machines are still stable at that moment.
3. Power balance after separation: if the facility was importing power, local generation must pick up the deficit or load must be shed.
4. Frequency and voltage recovery: governor and excitation response, motor reacceleration and load shedding stages.
5. Protection in island mode: reduced fault current may affect feeder and generator protection.
6. Resynchronization: synchronizing checks (device 25) for reconnection to the utility.

13.2 Example: island separation from a 69 kV utility tie



A 40 MW industrial plant with two 25 MVA gas turbine generators at 13.8 kV imports 8 MW from a 69 kV utility tie. When a utility fault causes separation, the generators must instantly pick up the 8 MW import. With a combined H of about 4.5 s on 50 MVA, the initial rate of frequency decline is roughly:

Without fast load shedding, the plant frequency would reach typical generator under-frequency trip settings within a few seconds, depending on governor response. Studies define load shedding blocks and timing so frequency recovers before generator protection operates.

14. Inverter-based resources and transient stability

Solar, wind and battery storage connected through inverters are changing transient stability in four ways:

• Less synchronous inertia: fewer synchronous machines online means larger angle and frequency deviations for the same disturbance.
• Different fault response: IBRs supply limited, controlled fault current, roughly 1.0 to 1.5 times rated, and their active power during and after faults depends on control settings.
• Ride-through behavior: if IBRs reduce active power or trip during faults, synchronous machines nearby experience different accelerating power than in traditional studies.
• New stability forms: converter-driven stability and weak-grid interactions, which may require EMT studies.

IEEE 2800-2022 and NERC PRC-029 define ride-through and recovery requirements for transmission-connected IBRs. Transient stability studies for plants near large IBR concentrations must model these resources accurately, including active power recovery ramp rates after faults, which directly affect neighboring generators' rotor angle swings.



15. Practical study workflow

15.1 Flat run and initialization checks

Every dynamic study should begin with a flat run of at least 10 to 20 seconds with no disturbance. All machine angles, speeds, voltages and controller outputs should remain constant. Drift usually indicates initialization errors, inconsistent data or controllers operating at limits.



15.2 Interpreting results

16. Common study pitfalls

1. Using classical models for detailed studies: classical models omit excitation, saturation and damper effects and may give misleading results.
2. Unvalidated or default data: typical parameters can differ significantly from the actual machine and control settings.
3. Ignoring delayed clearing: normal clearing may pass, while stuck-breaker cases fail.
4. Modeling loads as constant impedance: this can hide voltage recovery and motor stall problems.
5. Leaving out protection: relays that trip during the event change the outcome.
6. Studying only peak load: light-load, high-renewable or high-transfer cases may be more severe.
7. Inadequate simulation length: some instabilities appear on the second or later swings.
8. Ignoring IBR behavior: active power recovery and ride-through settings change neighboring generator swings.

17. Worked example 2: how fault type changes the critical clearing time



Planning criteria use different fault types for different events, so engineers need to know how much each type shortens or lengthens the clearing time limit. In positive-sequence stability programs, an unbalanced fault is represented by connecting a fault shunt at the faulted bus in the positive-sequence network. Its value comes from the negative- and zero-sequence Thevenin impedances at that bus:

With a fault shunt Xf at the plant's 345 kV bus, the during-fault transfer reactance between the generator internal voltage and the system becomes:

where Xa is the reactance from the generator internal voltage to the faulted bus (X′d plus the step-up transformer, 0.45 pu) and Xb is the reactance from the bus to the system (two circuits in parallel plus the system equivalent, 0.25 pu). Taking Z2 equal to the positive-sequence Thevenin reactance at the bus (0.161 pu) and a zero-sequence Thevenin reactance of 0.10 pu, typical for a bus fed by a delta-wye grounded step-up transformer, Keentel's simulations give:

The single-line-to-ground fault allows more than four times the clearing time of the three-phase fault. This is why many planning criteria apply three-phase faults with normal clearing but single-line-to-ground faults for delayed-clearing events: a stuck breaker on a single-phase fault is far less severe. It is also why independent pole operation of breakers, which limits a stuck breaker to one phase, is such an effective stability measure at large generating stations.

18. Numerical integration in stability programs



Commercial stability programs solve thousands of nonlinear differential equations for machines, controls and loads, together with the algebraic network equations, at every time step. Two broad approaches are used:

A typical positive-sequence time step is a quarter to half cycle (about 4 to 8 ms). Engineers should check that results do not change when the time step is halved, that fast controllers with very small time constants are not causing numerical instability, and that network switching events occur exactly at the intended times. Results that change with time step, or controllers that oscillate at the integration frequency, indicate model or data problems rather than physical behavior.

For the single-machine worked examples in this article, Keentel integrated the swing equation with a 0.1 ms step and confirmed the analytical critical clearing time to within 0.1 ms.

19. Direct methods and screening

Running every contingency at every operating point is time-consuming, so engineers use screening techniques to focus detailed simulation where it matters:

• Transient energy function (Lyapunov) methods: extend the equal-area concept to multi-machine systems by comparing the system's kinetic and potential energy at fault clearing with a critical energy value. They provide a stability margin directly but depend on simplified models.
• Extended equal-area criterion (SIME and related methods): reduce the multi-machine system to an equivalent one-machine system for the critical group of machines, combining time-domain results with equal-area insight.
• Contingency ranking: severity indices based on early rotor angle acceleration, kinetic energy at clearing or voltage dips rank contingencies before full simulation.
• Dynamic security assessment (DSA): online tools used by many US reliability coordinators and transmission operators run stability simulations on near-real-time cases to determine operating limits.

Screening methods guide the study, but final conclusions in US planning work rest on detailed time-domain simulation with validated models.

20. Voltage and frequency performance in stability studies

A transient stability study is judged by more than rotor angles. Planning coordinators and utilities apply additional performance criteria to the same simulations:

These criteria often reveal problems that rotor angle plots alone would miss. A plant that remains in synchronism but causes prolonged low voltage at a nearby load center, or causes neighboring inverter-based plants to enter momentary cessation, still requires mitigation.

21. Large loads, data centers and transient stability

The rapid growth of large loads in the US — hyperscale data centers, AI training campuses, cryptocurrency mining facilities, hydrogen electrolyzers and large industrial electrification projects — is introducing new transient stability considerations at 115 kV to 345 kV.

• Load loss as a disturbance: large power-electronic loads can disconnect or transfer to backup power during voltage dips. Loss of hundreds of megawatts of load in a fraction of a second causes over-frequency and can accelerate nearby generators. NERC has documented events in which large data center loads dropped offline during transmission faults.
• Fault ride-through of loads: unlike generators, large loads have historically had no ride-through requirements. Transmission providers increasingly ask large-load customers to specify and verify voltage ride-through behavior.
• On-site generation and islanding: campuses with on-site gas turbines, reciprocating engines or fuel cells must be studied for both grid-connected and islanded stability, including the transition between them.
• Rapid load fluctuations: AI training workloads can change power by large fractions of the facility rating within seconds, which can excite electromechanical modes and stress on-site generators' governors.
• Modeling: large loads require dynamic models that represent their power electronics, protection and ride-through behavior, rather than static load models.

Keentel Engineering studies these facilities with the same methods used for generators: time-domain simulation of faults, ride-through and islanding, with protection represented explicitly.

22. What a complete transient stability study report contains

A study is only as useful as its documentation. A complete report allows the transmission provider, owner and future engineers to understand, reproduce and rely on the results:

1. Purpose and scope: the project, the questions the study answers and the applicable criteria.
2. System representation: base cases, study area, operating conditions and future projects included.
3. Dynamic data: generator, excitation, PSS, governor, IBR, load and protection models, with sources and validation status.
4. Contingency list: each event with fault type, location, clearing times and switching sequence.
5. Results summary: pass or fail for every event, with critical clearing times and margins.
6. Plots: rotor angles, speeds, voltages, active and reactive power, frequency and relay impedance trajectories for critical cases.
7. Sensitivities: loading, dispatch, outage and model parameter sensitivities.
8. Mitigation: options evaluated, recommended solution and its performance.
9. Assumptions and limitations: what was not modeled and why.
10. Recommendations: protection settings, equipment requirements, operating limits and model updates.

23. Questions transmission providers commonly ask

Developers and owners can shorten interconnection timelines by anticipating the questions planners usually raise during review:

24. How Keentel Engineering helps



Keentel Engineering performs transient stability and dynamic studies for generator owners, developers, utilities, transmission owners, industrial facilities and data centers, from 4 kV to 765 kV.

Why clients choose Keentel Engineering

• Depth in dynamics: experienced engineers in rotor angle, voltage, frequency and converter-driven stability.
• Study and design integration: the same team that runs the studies designs protection, substations and interconnection facilities.
• NERC and ISO experience: practical knowledge of TPL-001, PRC and MOD standards and the study requirements of US ISOs and utilities.
• Clear reporting: results presented with margins, sensitivities and recommendations that decision-makers can act on.
• Licensed professional engineering: a Florida-registered firm with offices in Tampa, Austin, Sacramento and Baltimore, supporting projects nationwide.

Part C — Case Studies

The following are illustrative scenarios based on conditions typical of US generation, transmission and industrial projects. They show how Keentel Engineering approaches transient stability problems; site details are generalised and figures are rounded.

Case Study 1 — 345 kV combined-cycle plant: stuck-breaker instability

Situation

A developer adds a 750 MW combined-cycle plant connected to a 345 kV ring bus with two outgoing lines. Normal-clearing studies under NERC TPL-001 P1 show stable performance, but the transmission provider requires P4 stuck-breaker cases.

Analysis

Keentel's study shows that a three-phase fault on one line with a stuck ring-bus breaker leaves the plant connected through a single remaining line after about 12 cycles. The critical clearing time for this condition at full output is about 9 cycles. The steam turbine and both combustion turbines lose synchronism in the simulation.

Engineered solution

Keentel evaluates three options: independent pole operation of the critical breakers, a faster breaker failure scheme, and a remedial action scheme to trip one combustion turbine for the specific contingency. Independent pole operation converts the stuck-breaker three-phase fault into a much less severe single-phase stuck pole, and combined with optimized breaker failure timing restores stability with margin.

Outcome

All P1 to P7 events are stable with the recommended design, without a remedial action scheme or an output limit, and protection requirements are documented in the interconnection facilities study.

Case Study 2 — 138 kV industrial cogeneration plant: islanding and load shedding

Situation

A chemical plant with two 30 MVA steam turbine generators at 13.8 kV normally imports 10 MW from a 138 kV utility tie. During a regional disturbance, the plant separated from the utility but both generators tripped on under-frequency within seconds, shutting down the process.

Analysis

Keentel's dynamic study, validated against the plant's event records, shows that the separation relay operated too slowly and that the existing load shedding scheme shed only after frequency had fallen below 58.5 Hz. The generators' low inertia and slow governor response could not arrest the decline in time. Motor reacceleration after the voltage dip added further load.

Engineered solution

Keentel redesigns the islanding scheme: transfer trip from the utility substation for fast separation, a two-stage fast load shedding scheme triggered by the separation signal and import level, revised governor settings coordinated with the turbine manufacturer, and staggered motor restart. Generator protection settings are confirmed to coordinate with the new scheme.

Outcome

Simulations show the island stabilizes above 59.3 Hz for all credible import levels, and the plant can maintain critical processes during utility disturbances.

Case Study 3 — 230 kV hydro plant: post-fault oscillations and PSS retuning

Situation

A 230 kV hydro plant with four 80 MVA salient-pole units shows poorly damped power oscillations of about 1.2 Hz after nearby line faults, recorded by PMUs at the plant. Transmission planners are concerned that new solar plants nearby will make the oscillations worse.

Analysis

Keentel reproduces the events with validated GENSAL, exciter and PSS models. The units remain transiently stable, but the PSS settings, originally tuned decades earlier for a stronger system, provide little damping at the current local mode frequency. Eigenvalue analysis and time-domain simulation confirm damping below the planning coordinator's criteria.

Engineered solution

Keentel retunes the power system stabilizers using IEEE 421.5 PSS2B models, coordinates the tuning with the exciter manufacturer, and verifies performance for a full range of operating conditions, including future solar additions. Updated models are submitted under NERC model verification requirements.

Outcome



Post-fault oscillations decay well within criteria in simulation and in subsequent field events, and the plant's models are updated for future planning studies.

Part D — Technical FAQ: Transient Stability Analysis

References and Further Reading

Links were current at publication (October 2026).



• NERC Reliability Standard TPL-001, Transmission System Planning Performance Requirements. North American Electric Reliability Corporation.
• NERC Reliability Standards FAC-002, PRC-012, PRC-019, PRC-024, PRC-026, PRC-029, MOD-026, MOD-027 and MOD-032. North American Electric Reliability Corporation.
• IEEE Std 1110, Guide for Synchronous Generator Modeling Practices and Parameter Verification with Applications in Power System Stability Analyses. IEEE.
• IEEE Std 421.5, Recommended Practice for Excitation System Models for Power System Stability Studies. IEEE.
• IEEE Std C37.102, Guide for AC Generator Protection. IEEE.
• IEEE Std 2800-2022, Standard for Interconnection and Interoperability of Inverter-Based Resources Interconnecting with Associated Transmission Electric Power Systems. IEEE.
• IEEE/CIGRE Joint Task Force, Definition and Classification of Power System Stability (2004) and Definition and Classification of Power System Stability — Revisited and Extended (2021). IEEE Transactions on Power Systems.
• P. Kundur, Power System Stability and Control. McGraw-Hill.
• P. M. Anderson and A. A. Fouad, Power System Control and Stability. IEEE Press / Wiley.
  

Disclaimer

This article is general technical information for educational purposes. It is not engineering advice for any specific project and does not create a professional relationship. The worked examples use Keentel Engineering's simplified single-machine models with illustrative data; actual stability limits depend on detailed generator, control, network, load and protection models and system operating conditions. Planning criteria differ by planning coordinator, ISO/RTO and utility. Project designs must follow current standards, transmission provider requirements and project-specific engineering analysis, sealed by a licensed professional engineer where required.



Case studies are illustrative scenarios based on typical project conditions, not records of specific client projects. Figures are rounded and indicative.

NERC, FERC, IEEE, PSS®E, PSLF, PowerWorld, PowerFactory, PSCAD™, EMTP and other names are the property of their respective owners. Keentel Engineering is not affiliated with or endorsed by these organisations or any software or equipment vendor.

contact@keentelengineering.com  ·  keentelengineering.com  ·  calendly.com/keentel-engineering/15min

State of Florida — Registry No. 36853, KEENTEL LLC, DBA: KEENTEL ENGINEERING

Copyright 1995–2026 Keentel Engineering. All Rights Reserved.