Published: June 2026
Technical Level: Advanced
Category: Power Quality
The displacement of synchronous generation and linear load by inverter-based resources and electronic loads has broadened the power quality problem from the single concern of harmonic distortion to a set of three distinct phenomena — steady-state harmonics, voltage flicker, and sub-cycle transients — each governed by its own standard and each requiring its own measurement and mitigation. This paper develops the engineering basis for all three. It defines harmonic distortion and its compliance limits under IEEE 519-2022, develops the short-term and long-term flicker severity indices of IEEE 1453-2021 and their connection to the rapid voltage fluctuations produced by fluctuating and renewable sources, and characterizes the transient environment through the location categories of IEEE C62.41. For each phenomenon the paper identifies the dominant sources in a power-electronics-rich system, the measurement approach required to quantify it, and the mitigation technology — active filters, static compensators, and surge-protective devices — appropriate to it.
The power quality of a distribution system was historically dominated by a single steady-state concern: the harmonic distortion injected by a modest population of nonlinear loads against a background of synchronous generation and predominantly linear load. That landscape has changed. A large and growing fraction of load now passes through electronic power conversion — variable-frequency drives, switch-mode power supplies, light-emitting-diode drivers, and electric-vehicle chargers — and a comparably large fraction of generation is now interfaced through inverters rather than rotating machines. The consequence is that all three categories of power quality disturbance have intensified: harmonic distortion has risen as nonlinear load has grown, voltage flicker has increased as fluctuating renewable output and large cyclic loads have proliferated, and the transient environment has become more active as power-electronic switching events multiply.
These three phenomena are physically distinct and cannot be addressed by a single instrument or a single mitigation device. Harmonics are a steady-state distortion of the waveform shape, quantified over many cycles and remedied by filtering. Flicker is a fluctuation of the voltage magnitude slow enough to be perceived as a variation in lamp output, quantified by a perceptibility-weighted statistical index and remedied by dynamic reactive compensation. Transients are sub-cycle excursions of voltage, quantified by their peak and energy and remedied by surge-protective devices. This paper treats each in turn, establishing the controlling standard, the measurement method, and the mitigation appropriate to the disturbance.
Harmonics are sinusoidal components of the current or voltage at integer multiples of the fundamental frequency, arising whenever a load draws current that is not proportional to the applied voltage. A six-pulse rectifier front-end, the most common nonlinear load, draws current in pulses and produces a characteristic spectrum dominated by the fifth and seventh harmonics; single-phase electronic loads produce a spectrum rich in the third harmonic, which is of particular concern because the triplen harmonics sum in the shared neutral of a four-wire system rather than cancelling. The aggregate distortion is quantified by the total harmonic distortion, the ratio of the combined harmonic content to the fundamental:
Where:
is the root-mean-square magnitude of the voltage or current at harmonic order .
is the root-mean-square magnitude of the fundamental component.
is the highest harmonic order included, taken as the 50th in IEEE 519 practice.
Compliance is governed by IEEE 519-2022, which limits voltage distortion as a function of system voltage — eight percent total harmonic voltage distortion at and below 1 kV, tightening at higher voltages — and limits the customer's injected current distortion as a function of the short-circuit-to-load ratio at the point of common coupling. The dominant mitigation in a power-electronics-rich facility is the active harmonic filter, which measures the load's harmonic current and injects an equal and opposite current to cancel many orders simultaneously, routinely reducing a facility's current distortion from the high teens into the low single digits and bringing a non-compliant facility into conformance with IEEE 519.
Voltage flicker is a fluctuation in the magnitude of the supply voltage at a rate, typically between a fraction of a hertz and a few tens of hertz, that produces a visible variation in the light output of lamps and is perceived by building occupants as an annoyance. Its classical source is a large fluctuating load such as an arc furnace or a cyclically operating welder, but the integration of fluctuating renewable generation — photovoltaic output varying with passing cloud cover, wind output varying with gusts — has made flicker a concern on feeders that carry significant inverter-based resources, and the localized voltage fluctuation of clustered electric-vehicle charging contributes as well.
Flicker is not adequately characterized by a simple voltage variation, because human perception of light flicker depends on both the magnitude and the frequency of the fluctuation. IEEE 1453-2021, which adopts the IEC flickermeter methodology, instead quantifies flicker by a perceptibility-weighted statistical index. The short-term flicker severity, evaluated over a ten-minute interval, is denoted , with a value of unity defined as the threshold of perceptibility for the average observer; the long-term severity aggregates twelve consecutive short-term values over two hours to capture sustained exposure:
Where:
is the long-term flicker severity index.
is the -th of twelve consecutive short-term flicker severity values.
The cubic aggregation weights the highest short-term values most heavily, reflecting that occasional severe fluctuations dominate the perceived annoyance. IEEE 1453 sets compatibility levels near a short-term severity of unity and a long-term severity of 0.65 at the point of common coupling.
As a worked example, suppose a flickermeter at the point of common coupling of a feeder with clustered EV charging records twelve consecutive ten-minute short-term severity values over a two-hour window: 0.62, 0.58, 0.71, 0.55, 0.60, 0.88, 0.52, 0.49, 0.57, 0.63, 0.95, and 0.54. The long-term severity aggregates these by the cubic mean:
Where are the twelve short-term values. The result of 0.667 just exceeds the IEEE 1453 long-term compatibility level of 0.65, so this feeder is marginally non-compliant — and the cause is visible in the data: two excursions (0.88 and 0.95) dominate the cubic sum even though the arithmetic mean of the twelve values is only 0.64. This is the practical lesson of the cubic aggregation: a feeder can fail the long-term flicker limit on the strength of a few severe events while its average behavior appears acceptable, which is why mitigation targets the peak fluctuations rather than the mean. Where these are exceeded, the mitigation is dynamic reactive compensation: a static synchronous compensator injects or absorbs reactive power within milliseconds to oppose the voltage fluctuation, and a correctly sized unit reduces the short-term flicker severity from well above the perceptibility threshold to a fraction of it, eliminating occupant complaints.
The waveform disturbances addressed in this paper are illustrated in Figure 1, which shows a voltage sag followed by a transient overshoot superimposed on the nominal sinusoidal waveform.

Figure 1. Voltage disturbance waveform. The horizontal axis is time in milliseconds and the vertical axis is instantaneous voltage in per unit. The waveform shows a steady fundamental, a multi-cycle voltage sag to approximately 0.7 p.u., and a single-cycle transient overshoot before recovery. The engineer should observe that the sag and the transient are distinct phenomena with different mitigation strategies — the sag is addressed by ride-through capability and dynamic voltage support, while the transient is addressed by surge protective devices — so power quality monitoring must distinguish and characterize each independently.
A transient is a sub-cycle disturbance of the voltage waveform, lasting from a fraction of a microsecond to a few milliseconds, superimposed on the normal power-frequency voltage. Impulsive transients are unidirectional spikes produced by lightning and by certain switching events; oscillatory transients are decaying ring-downs produced by capacitor-bank energization and by the switching of inductive circuits. Although brief, transients carry enough energy to break down insulation, to disrupt or damage sensitive electronic equipment, and to cause the misoperation of control systems, and their frequency has risen with the proliferation of power-electronic switching.
The transient environment is characterized for surge-protection purposes by IEEE C62.41, which defines location categories reflecting the attenuation of a transient as it propagates from the service entrance toward interior loads. The service-entrance category is exposed to the most severe surges — open-circuit voltages on the order of several kilovolts and short-circuit currents in the thousands of amperes — while interior outlet locations, protected by the impedance of the intervening wiring, see lower current and energy for the same voltage. This categorization governs the selection and coordination of surge-protective devices: a higher-energy device is applied at the service entrance to clamp the incoming surge, and lower-energy devices are applied at panels and at sensitive equipment to suppress the residual, with the devices coordinated so that the upstream device absorbs the bulk of the surge energy. Where the consequence of a voltage sag or interruption rather than an overvoltage transient is the concern — as in a semiconductor fabrication line, where a brief sag on an adjacent feeder can scrap an entire batch of wafers — the appropriate mitigation is a dynamic voltage restorer or an uninterruptible supply that rides through the disturbance, sized against the depth and duration of the credible sag rather than against a surge energy.
Because the three phenomena differ in time scale, their measurement requires an instrument capable of resolving each. A power quality analyzer applied to a modern facility must capture steady-state harmonic spectra over many cycles, compute the flicker severity indices over ten-minute and two-hour windows, and trigger high-resolution waveform capture on sub-cycle transients and voltage sags. Continuous monitoring at the service entrance and at critical buses, rather than monitoring initiated only in response to a complaint, identifies developing problems substantially earlier and allows mitigation to be designed against measured data rather than against an after-the-fact reconstruction of an event. The measurement record also establishes the baseline against which compliance with IEEE 519, IEEE 1453, and the facility's transient-protection design is demonstrated, and it provides the evidence needed to attribute a disturbance to its source — an internal nonlinear load, a fluctuating renewable resource, or an external event on the utility system.
The most consequential finding is that power quality in an inverter-dominated system is no longer one problem but three, each with its own controlling standard and its own mitigation: harmonics under IEEE 519-2022, flicker under IEEE 1453-2021, and transients under IEEE C62.41. A facility that manages only steady-state distortion has addressed a third of its exposure, and the phenomenon that damages or disrupts its equipment is as likely to be the flicker or the transient it never measured.
The most common implementation failure is specifying mitigation for one phenomenon with hardware that aggravates another — a power-factor capacitor that corrects voltage but creates a harmonic resonance, or an SPD that clamps a transient but is uncoordinated with the downstream device it is meant to protect.
The engineer should next establish continuous, multi-parameter power-quality monitoring at the point of common coupling, because each of the three phenomena varies with the facility's load mix and grid conditions, and a program that measures all three continuously is the only basis for designing mitigation against data rather than assumption.
The analysis in this paper connects to several companion studies in this library. Readers concerned with the upstream and downstream engineering will find Power Quality Analysis develops a closely related aspect of the same problem, while Harmonic Distortion in Commercial and Industrial Power Systems extends the treatment into an adjacent domain. For the broader methodological context, Lightning Protection Systems provides complementary depth.
[1] IEEE Standard 519-2022, IEEE Standard for Harmonic Control in Electric Power Systems, IEEE, 2022.
[2] IEEE Standard 1453-2021, IEEE Standard for Measurement and Limits of Voltage Fluctuations and Associated Light Flicker on AC Power Systems, IEEE, 2021.
[3] IEEE Standard 1159-2019, IEEE Recommended Practice for Monitoring Electric Power Quality, IEEE, 2019.
[4] IEEE Standard C62.41.1-2002, IEEE Guide on the Surge Environment in Low-Voltage (1000 V and Less) AC Power Circuits, IEEE, 2002.
[5] IEEE Standard C62.41.2-2002, IEEE Recommended Practice on Characterization of Surges in Low-Voltage AC Power Circuits, IEEE, 2002.
[6] IEC Standard 61000-4-15:2010, Electromagnetic Compatibility — Testing and Measurement Techniques — Flickermeter, International Electrotechnical Commission, 2010.
[7] R. C. Dugan, M. F. McGranaghan, S. Santoso, and H. W. Beaty, Electrical Power Systems Quality, 3rd ed., McGraw-Hill, 2012.
[8] M. H. J. Bollen, Understanding Power Quality Problems: Voltage Sags and Interruptions, IEEE Press, 2000.
[9] IEEE Standard 1547-2018, IEEE Standard for Interconnection and Interoperability of Distributed Energy Resources, IEEE, 2018.