Guide For Recorder and Data Acquisition Requirements For Characterization of Power Quality Events

1 Overview

1.2 Scope

This guide establishes the data acquisition attributes necessary to characterize the electromagnetic phenomena.

The objective of this guide is to describe the technical measurement requirements for each type of disturbance defined in 24 categories of typical characteristics of power system electromagnetic phenomena. Each category is discussed in several standards in terms of emission limits, severity levels, planning levels or immunity levels. The initial publication of this document covers the category of RMS voltage variations referred to as sags or sag (dip)s.

Basically, when the signal to measure is supplied from a perfectly stable wave-form generator, results displayed by one instrument are very close to those obtained by others. However, when connected to the power system, several phenomena can affect the assessment in such a way that some instruments can measure disturbance levels which do not exist and several errors can be added to the remaining harmonics assessed.

2 Definitions

2.1 Background

Some definitions given in this section have not been fully harmonized with definitions made in other IEEE standards and IEC documents.. Where several definitions for the same terms exist, they are both included. Before scheduling the power quality measurement, the user must select the instrument which meets his needs. This section addresses the language and the assessment specification which need to be understood to communicate with the instrumentation manufacturer. Also, the user needs to define his need clearly.

Without the assessment specification of the instrument, or with an inappropriate specification, users are unable to assess the severity level of the disturbance on equipment; this can lead to incorrect conclusions and costly decisions. Users should ask manufacturers for the complete measurement-instrument specification, this may include the following information:

· sampling rate,

· bandwidth amplitude/frequency

· accuracy (the instrument should stay within the limits 100% of the time)

· precision (the instrument should stay within the limits 68.27% of the time)

· resolution (signal-to-noise ratio or effective bits)

· differential mode argument and amplitude accuracy,

· common mode rejection

· anti-aliasing filter (cut-off frequency, filter order, filter type)

· window width, (interval of the sampled signal used for each analysis)

· number of windows analysed per second,

· type of weighted window used (rectangular, triangular, Blackman, Vincent, …)

· synchronisation device used (electronic phase lock loop or digital synchronisation),

· accuracy of the synchronised device,

· the common mode ratio rejection,

· flag when the phase lock loop is not synchronized

· flag when hardware or software error occurs,

· flag when some frequency components present in the signal are not recorded,

· immunity of the instrument to disturbances in the supply voltage (transient, voltage sag (dip)s, voltage distortion etc…),

· operating environment (temperature, vibration, humidity)

2.2 IEEE & IEC Definitions

Term	Definition	Source
Accuracy	The quantity of freedom from mistake or error which the IEC dictionary and IEEE 100 characterize as the closeness of an indicated value of a measuring instrument to the corresponding true value.	IEC dictionary and IEEE 100
	Accuracy is the degree of agreement between measured values and the true values (i.e. closeness to the reality).
Aliasing	The distortion caused by sampling a signal at an inappropriate rate and which results in the overlapping of the sidebands around the harmonics of the sampling frequency in the spectrum of the sample signal. Note.: A high frequency which exceeds the operating range of the instrument affects the frequency component in the frequency range of the analysis. IEC has requested use of an anti-aliasing filter which should attenuate to below 50 dB all frequencies above the operating range of the instrument. Observation windows may be flagged during the presence of frequency components which exceed the operating range of the instrument. If these high frequency components occur too often, the user may need a wider bandwidth instrument.	IEC dictionary
Analogue signal	A signal in which the characteristic quantity representing information may, at any instant, assume any value within a continuous interval. Note. For example, an analogue signal may follow continuously the values of another physical quantity representing information	IEC dictionary
Anti-aliasing	correction intended to reduce the aliasing.	IEC dictionary
Bandwidth	The width of a frequency band over which a given characteristic of an equipment or transmission channel does not differ from a reference value by more than a specified amount or ratio. Note.: the given characteristic may, for example, be the amplitude/frequency, the phase/frequency or the delay/frequency characteristic. The quantitative difference between the limiting frequencies of a frequency band. For most power quality users should specify the bandwidth as the frequency for which the maximum sinusoidal input signal has decreased below the accuracy specified
Channel	individual measurement path through an instrument Note: “Channel” and “phase” are not the same. A voltage channel is by definition the difference in potential between two conductors. Phase refers to a single conductor. A channel may be between two phases on a poly-phase system, or between a phase and neutral, or between a phase and earth.	IEC 1000-4-30
declared input voltage, U_din	value obtained from the declared supply voltage by a transducer ratio	IEC 1000-4-30
declared supply voltage	value of a voltage different from the nominal voltage obtained by agreement between the electricity supplier and a consumer	IEC 1000-4-30
Dynamic accuracy	Accuracy determined for a time-varying output	IEEE 100
Dynamic range	The ratio, usually expressed in decibels, of the maximum to the minimum signal input amplitude over which an amplifier can operate within some specified range of performance.	IEEE 100
flagged data	for any measurement time interval in which interruptions, sag (dip)s or swells occur, the measurement results of all other parameters made during this time interval are flagged	IEC 1000-4-30
flicker	impression of unsteadiness of visual sensation induced by a light stimulus whose luminance or spectral distribution fluctuates with time	(IEV 161-08-13)
Frequency bins	The minimum bandwidth at which two signals close in frequency can still be analysed.
fundamental component	component whose frequency is the fundamental frequency	(IEV 101-14-49 modified)
fundamental frequency	frequency in the spectrum obtained from a Fourier transform of a time function, to which all the frequencies of the spectrum are referred NOTE - In case of any remaining risk of ambiguity, the power supply frequency should be referred to the polarity and speed of rotation of the synchronous generator(s) feeding the system.	(IEV 101-14-50 modified)
harmonic component	any of the components having a harmonic frequency NOTE Its value is normally expressed as an r.m.s. value. For brevity, such component may be referred to simply as a harmonic	(IEC 61000-2-2, definition 3.2.4).
harmonic frequency	frequency which is an integer multiple of the fundamental frequency NOTE - The ratio of the harmonic frequency to the fundamental frequency is the harmonic order	(IEC 61000‑2-2, definition 3.2.3).
hysteresis	phenomenon represented by a characteristic curve which has a branch, called ascending branch, for increasing values of the input variable, and a different branch, called descending branch, for decreasing values of the input variable. NOTE – Sometimes a dead band effect can be superposed on a hysteresis phenomenon.	(IEV 351-14-22)
influence quantity	any quantity which may affect the working performance of a measuring equipment NOTE - this quantity is generally external to the measurement equipment	(IEV 311-06-01 modified).
interharmonic component	component having an interharmonic frequency NOTE - Its value is normally expressed as an r.m.s. value. For brevity, such a component may be referred to simply as an interharmonic	(IEC 61000-2-2, definition 3.2.6).
interharmonic frequency	any frequency which is not an integer multiple of the fundamental frequency NOTE 1 By extension from harmonic order, the interharmonic order is the ratio of an interharmonic frequency to the fundamental frequency. This ratio is not an integer (Recommended notation m). NOTE 2 In the case where m < 1 the term subharmonic frequency may be used.	(IEC 61000-2-2, definition 3.2.5)
interruption	reduction of the voltage at a point in the electrical system below a threshold	IEC 1000-4-30
interruption threshold	voltage magnitude specified for the purpose of detecting the start and the end of a voltage interruption	IEC 1000-4-30
Mean ():	The mean of a number of measurements performed during a period or an interval of a given quantity is defined by
measurement uncertainty	maximum expected deviation of a measured value from its actual value	IEC 1000-4-30
nominal voltage, U_N	voltage by which a system is designated or identified	IEC 1000-4-30
overdeviation	difference between the measured value and the nominal value of a parameter, only when the measured value of the parameter is greater than the nominal value	IEC 1000-4-30
power quality	characteristics of the electricity at a given point on an electrical system, evaluated against a set of reference technical parameters. NOTE These parameters might, in some cases, relate to the compatibility between electricity supplied on a network and the loads connected to that network.	IEC 1000-4-30
Observation Window	Sampled signal which occurs during a predefined period.	IEC 1000-4-30
Precision	The quantity of coherence or repeatability of measurement data, customarily expressed in terms of the standard deviation of the extended set of measurement results from a well-defined (adequately specified) measurement process in a state of statistical control. The standard deviation of the conceptual population is approximated by the standard deviation of an extended set of actual measurements.
	Precision is an indication of the likelihood that a measurement will differ from the same measurement at a different, but not correlated, time. The assessment interval specification is also needed to assess precision values. For example, the precision could be assessed using measurements every 3-s interval.
Resolution	The smallest change in the measured or supplied quantity to which a numerical value can be assigned without interpolation. Note.: the resolution of most digital instruments is specified in terms of the number of bits that the analogue to digital converter can provide. Manufacturers have invented several wordings such as effective bits.	IEC dictionary
rms ()	The rms of a number of measurements performed during a period or an interval of a given quantity is defined by
	square root of the arithmetic mean of the squares of the instantaneous values of a quantity taken over a specified time interval and a specified bandwidth	(IEV 101-14-16 modified).
r.m.s. voltage refreshed each half cycle, U_rms(1/2)	value of the r.m.s. voltage measured over one cycle, commencing at a fundamental zero-crossing, and refreshed each half cycle NOTE1: This technique is independent for each channel, and will produce r.m.s. values at successive times on different channels for polyphase systems. NOTE2: This value is used only for voltage sag (dip), voltage swell, and interruption detection	IEC 1000-4-30
Range of influence quantities	range of values of a single influence quantity	IEC 1000-4-30
reference channel	one of the voltage measurement channels designated as the reference channel for poly-phase measurements	IEC 1000-4-30
retained voltage, U_ret	minimum value of U_rms(1/2)recorded during a voltage sag (dip) or interruption. NOTE The retained voltage is expressed as a value in volts, or as a percentage or per unit value of the declared input voltage.	IEC 1000-4-30
sliding reference voltage, U_sr	voltage magnitude averaged over a specified time interval, representing the voltage preceding a voltage sag (dip) or swell NOTE - The sliding reference voltage is used to determine the voltage change during a sag (dip) or a swell	IEC 1000-4-30
sag (dip) threshold	voltage magnitude specified for the purpose of detecting the start and the end of a voltage sag (dip)	IEC 1000-4-30
Sampled signal	Signal representing a variable which is only intermittently observed and represented. The sequence of values of a signal taken at discrete instants.
Sideband	The spectral components resulting from the modulation of a sinusoidal carrier and lying above or below the carrier frequency. Note: Sideband designates a frequency band lying above or below a sinusoidal carrier frequency and containing spectral components of significance produced by modulation. Flicker originated from a lamp is produced by the modulation of the power-system voltage operated at the fundamental frequency. Thus, the fundamental frequency is the carrier frequency and the modulation produces the sidebands.	IEC dictionary
signal	A measurable variable, one or more parameters of which carry information about one or more variables which the signal represents
swell threshold	voltage magnitude specified for the purpose of detecting the start and the end of a swell	IEC 1000-4-30
time aggregation	combination of several sequential values of a given parameter (each determined over identical time intervals) to provide a value for a longer time interval Note Aggregation in this document always refers to time aggregation.	IEC 1000-4-30
Traceability	Traceability is an attribute of some measurements. Measurements have traceability to the designated standards if an only if scientifically rigorous evidence is produced on a continuing basis to show that the measurement process is producing measurement results (data) for which the total measurement uncertainty, relative to national or other designated standards is quantified
underdeviation	absolute value of the difference between the measured value and the nominal value of a parameter, only when the value of the parameter is lower than the nominal value
Voltage sag (dip)	temporary reduction of the voltage at a point in the electrical system below a threshold NOTE - 1: Interruptions are a special case of sag (dip). Post-processing may be used to distinguish between sag (dip)s and interruptions. NOTE - 2; In some areas of the world a voltage sag (dip) is referred to as sag. The two terms are considered interchangeable, however this standard will only use the term voltage sag (dip).
voltage swell	temporary increase of the voltage at a point in the electrical system above a threshold
voltage unbalance	condition in a poly-phase system in which the r.m.s. values of the line voltages (fundamental component), or the phase angles between consecutive line voltages, are not all equal NOTE - 1: The degree of the inequality is usually expressed as the ratios of the negative and zero sequence components to the positive sequence component NOTE - 2: In this standard voltage unbalance is considered in relation to three-phase systems.	(VEI 161-08-09 modified)

2.3 Classes of measurement performance

For each parameter measured, two classes of measurement performance are defined:

· Class A performance

· This class of performance is used where precise measurements are necessary, e.g. for contractual applications, verifying compliance with standards, resolving disputes, etc. Any measurements of a parameter carried out with two different instruments complying with the requirements of class A, when measuring the same signals, will produce matching results within the specified uncertainty.

To ensure that matching results are produced, class A performance instrument requires a bandwidth characteristic and a sampling rate sufficient for the specified uncertainty of each parameter.

· Class B performance

This class of performance may be used for statistical surveys, troubleshooting applications, and other applications where low uncertainty is not required.

For each performance class the range of influencing factors that shall be complied with is specified in Clause 6. Users shall select the class of measurement performance taking account of the situation of each application case.

NOTE - 1: A measurement instrument may have different performance classes for different parameters.

NOTE - 2: The instrument manufacturer shall declare influence quantities which are not expressely given and which may degrade performance of the instrument.

3 Organisation of the measurements

3.2 Basic Blocks

The electrical quantity to be measured may be either directly accessible, as is generally the case in low-voltage supply systems, or accessible via measurement transducers.

The whole measurement chain is shown in Figure 1.

Figure 1 – Measurement chain

An "instrument" usually includes the whole measurement chain (see Figure 1). In this standard, the normative part does not consider the measurement transducers including uncertainty, but the Annex A.2 gives guidance.

3.3 Electrical values to be measured

Measurements can be performed on single-phase or poly-phase supply systems. Depending on the context, it may be necessary to measure voltages between phase conductors and neutral (line-to-neutral) or between phase conductors (line-to-line) or between neutral and earth. It is not the purpose of this standard to impose the choice of the electrical values to be measured. Moreover, except for the measurement of voltage unbalance, which is intrinsically poly-phase, the measurement methods specified in this document are such that independent results can be produced on each measurement channel.

Current measurements can be performed on each conductor of supply systems, including the neutral conductor and the protective earth conductor.

NOTE: It is often useful to measure current simultaneously with voltage, and to associate the current measurements in one conductor with voltage measurements between that conductor and a reference conductor, such as an earth conductor or a neutral conductor.

3.4 Measurement aggregation over time intervals

· For class A performance

A measurement time interval of magnitudes (supply voltage, harmonics, interharmonics and unbalance) shall be over a 10-cycle time interval for 50 Hz power system or 12-cycle time interval for 60 Hz power system.

NOTE The uncertainty of this measurement is included in the uncertainty measurement protocol of each parameter.

Measurement time intervals are aggregated over three different time intervals. Clauses A.6 and A.7 discuss some applications of these aggregation time intervals. The aggregation time intervals are:

– 3-s interval (150 cycles for 50 Hz nominal or 180 cycles for 60 Hz nominal),

– 10-min interval,

– 2-h interval.

· For class B performance

The manufacturer shall indicate the method, number and duration of aggregation time intervals.

3.4.1 Measurement aggregation algorithm

Aggregations are performed using the square root of the arithmetic mean of the squared input values.

NOTE: For flicker measurements, the aggregation algorithm is different (see IEC 61000‑4-15).

Three categories of aggregation are necessary:

– Cycle aggregation

The data for the 150/180-cycle time interval shall be aggregated from fifteen 10/12-cycle time intervals.

NOTE: This time interval is not a "time clock" interval; it is based on the frequency characteristic.

– From cycle to time clock aggregation

The 10-minute value shall be tagged with the absolute time (e.g. 01H10.00). The time tag is that at the end of the 10-minute aggregation. If the last 10/12 cycle value in a 10-minute aggregation period overlaps in time with the absolute 10-minute clock boundary, that 10/12 cycle value is included in the aggregation for this 10-minute interval.

On commencement of the measurement, the 10/12 cycle measurement shall be started at the boundary of the absolute 10-minute clock, and shall be re-synchronised at every subsequent 10-minute boundary.

Note: This technique implies that a very small amount of data may overlap and appear in two adjacent 10-minute aggregations.

– Time clock aggregation

The data for the “2-h interval” shall be aggregated from twelve 10-min intervals.

3.4.2 Time clock uncertainty

The time clock uncertainty is:

· For class A performance

± 20 ms for 50 Hz or ± 16,7 ms for 60 Hz.

NOTE - 1: This performance can be achieved, for example, through a synchronisation procedure applied periodically during a measurement campaign, or through a GPS receiver, or through reception of transmitted radio timing signals.

NOTE - 2: When synchronisation by an external signal becomes unavailable, the time tagging tolerance must be better than 1 s/24 h.

NOTE - 3: This performance is necessary to ensure that two class A instruments produce the same 10-min aggregation results when connected to the same signal.

NOTE 4: When a threshold is crossed, it may be useful to record the date and time.

· For class B performance

The manufacturer shall specify the method to determine 10-min intervals.

4 Supply voltage sag (dip)s and swells

4.2 Basic measurement

The basic measurement of a voltage sag (dip) and swell shall be the U_rms(1/2) on each measurement channel.

NOTE - 1: For class A, the cycle duration for U_rms(1/2) depends on the frequency. The frequency might be determined by the last non-flagged power frequency measurement (see 4.7 and 5.1), or by any other method that yields the uncertainty requirements of Clause 6.

NOTE - 2: The U_rms(1/2)value includes, by definition, harmonics, interharmonics, ripple control signals, etc.

4.2.1 Detection and evaluation of a voltage sag (dip)

4.2.1.1 Voltage sag (dip) detection

The sag (dip) threshold is a percentage of either U_din or the sliding voltage reference U_sr(see 5.4.4). The user shall declare the reference voltage in use.

NOTE: Sliding voltage reference U_sr is generally not used in LV systems. See 61000-2-8 for further information and advice.

– On single-phase system a voltage sag (dip) begins when the U_rms(1/2)voltage falls below the sag (dip) threshold, and ends when the U_rms(1/2)voltage is equal to or above the sag (dip) threshold plus the hysteresis voltage.

– On poly-phase system a sag (dip) begins when the U_rms(1/2)voltage of one channel, at least, is below the sag (dip) threshold and ends when the U_rms(1/2)voltage on all measured channels is equal to or above the sag (dip) threshold plus the hysteresis voltage.

The sag (dip) threshold and the hysteresis voltage are both set by the user according to the use.

4.2.1.2 Voltage sag (dip) evaluation

A voltage sag (dip) is characterised by a pair of data, either retained voltage (u_ret) or depth and duration.

– The retained voltage is the smallest U_rms(1/2) value measured on any channel during the sag (dip).

– The depth is the difference between the reference voltage (either U_din or U_sr) and the retained voltage. It is generally expressed in % of the reference voltage.

– The duration of a voltage sag (dip) is the time difference between the beginning and the end of the voltage sag (dip).

NOTE - 1: The sag (dip) duration measurement can be started on one channel and terminated on a different channel.

NOTE - 2: Voltage sag (dip) envelopes are not necessarily rectangular. As a consequence, for a given voltage sag (dip), the measured duration is dependent on the selected sag (dip) threshold value. The shape of the envelope may be assessed using several sag (dip) thresholds set within the range of voltage sag (dip) and voltage interruption thresholds detection.

NOTE - 3: Typically, the hysteresis is equal to 2 % of U_din.

NOTE - 4: Sag (dip) thresholds are e. g. in the range 85 % to 90 % of the fixed voltage reference for troubleshooting or statistical applications, and 70 % for contractual applications.

NOTE - 5: Retained voltage is often useful to end-users, and may be preferred because it is referenced to zero volts. In contrast, depth is often useful to electric suppliers, especially on HV systems or in cases when a sliding reference voltage is used.

NOTE - 6: Phase shift may occur during voltage sag (dip)s. See A9.4.

NOTE - 7: When a threshold is crossed, it may be useful to record the date and time.

4.2.2 Detection and evaluation of a voltage swell

4.2.2.1 Voltage swell detection

The swell detection threshold is a percentage of either U_din. or the sliding voltage reference U_sr(see 5.4.4). The user shall declare the reference voltage in use.

NOTE Sliding voltage reference U_sr is generally not used in LV systems. See 61000-2-8 for further information and advice.

– On single-phase system a swell begins when the U_rms(1/2) voltage is above the swell threshold, and ends when the U_rms(1/2) voltage is equal to or below the swell threshold minus the hysteresis voltage.

– On poly-phase system a swell begins when the U_rms(1/2) voltage of one channel, at least, is above the swell threshold and ends when the U_rms(1/2) voltage on all measured channels is equal to or below the swell threshold minus the hysteresis voltage.

The swell threshold and the hysteresis voltage are both set by the user according to the use.

4.2.2.2 Voltage swell evaluation

A voltage swell is characterised by a pair of data, maximum swell voltage magnitude and duration.

– The maximum swell magnitude voltage is the largest U_rms(1/2) value measured on any channel during the swell.

– The duration of a voltage swell is the time difference between the beginning and the end of the swell.

NOTE - 1: The swell duration measurement can be started on one channel and terminated on a different channel.

NOTE - 2: Swell envelopes may not be rectangular. As a consequence, for a given swell, the measured duration is dependent on the swell threshold value.

NOTE - 3 Typically, the hysteresis is equal to 2 % of U_din.

NOTE - 4: Typically, the swell threshold is greater than 110 % of U_din

NOTE - 5: Phase shift may also occur during voltage swells.

NOTE - 6: When a threshold is crossed, it may be useful to record the date and time.

4.2.3 Calculation of a sliding reference voltage

If a sliding reference is chosen for voltage sag (dip) or swell detection, this shall be calculated using a first-order filter with a 1-min time constant. This filter is given by:

U_sr(n) = 0,9967 x U_sr(n–1) + 0,0033 x U_(10/12)rms

where

U_sr(n)is the present value of the sliding reference voltage,

U_sr(n–1)is the previous value of the sliding reference voltage, and

U_(10/12)rmsis the most recent 10/12-cycle r.m.s. value.

When the measurement is started, the initial value of the sliding reference voltage is set to the declared input voltage. The sliding reference voltage is updated every 10/12-cycles. If a 10/12-cycle value is flagged, the sliding reference voltage is not updated and the previous value is used.

4.2.4 Measurement uncertainty

4.2.4.1 Retained voltage and maximum swell voltage magnitude measurement uncertainty

· For class A performance

The uncertainty shall be DU = ±0,2 % of U_din.

· For class B performance

The manufacturer shall specify the uncertainty. In all cases the uncertainty DU shall be ±1,0 % of U_din.

4.2.5 Duration measurement uncertainty

· For class A and class B performances

The uncertainty of a sag (dip) or swell duration is equal to the sag (dip) or swell commencement uncertainty (half a cycle) plus the sag (dip) or swell conclusion uncertainty (half a cycle).

Annex A (informative)

5 ADDITIONAL CHARACTERISTICS

5.2 Characteristics Duplicated Above

5.3 Waveform Envelope Based Duration

5.3.1 Overview of Envelope Based

A trained human observer could look at a disturbance and tell when it began and ends by inspection. Unfortunately, the rms method does not determine these values properly. As mentioned previously, the total duration of an rms sag could be as much as a cycle in error (too long) due to the averaging of the moving window. The rms moving window cannot give accurate point in wave of sag initiation or recovery either.

5.3.2 Calculation of Envelope Based Duration

These algorithms attempt to replicate the human observer and determine when the actual waveform deviates from the ideal waveform and when it recovers. We will call these values the “apparent” values. This technique uses two envelopes which surround the pre-event (ideal) voltage waveform by ± 5% and ± 10%. The algorithm proceeds as follows:

When the disturbance waveform deviates from the ideal by 10% (i.e., falls outside of the 10% envelope), the event is “detected.”

The algorithm “backs-up” to see where the waveform fell outside of the 5% envelope for the last time, and that is used as the beginning (start) of the event. (This helps pick-up on the initiation of the event more accurately, without having false reporting of minor variations in voltages that were not of significance.)

· The start of the sag is the point where the last excursion outside of the ±5% band began.

· When the disturbed waveform returns inside the 10% band for at least 1/2 cycle, the sag is deemed recovered.

· The initial point at which the 1/2 cycle of recovery occurs is the point of recovery. (The 10% band was chosen since most sags do not recover to within 5% in the data set intervals.

5.3.3 Reported Data of the Waveforms based on Envelope Technique

· Apparent Sag Start: The point at which the voltage falls outside of the 5% envelope as described previously.

· Apparent Sag End: The point at which the voltage returns to the 10% envelope as described previously.

· Point-of-initiation: The angle at which the “apparent sag start” occurs using the last positive-going zero crossing as the reference angle.

· Point-of-recovery: The angle at which the “apparent sag end” occurs using the last positive-going zero crossing as the reference angle.

· Apparent Sag Duration: The difference of Apparent Sag End and Apparent Sag Start.

· Difference in Duration: The difference in the sag duration given by the apparent method and the rms method (in seconds).

5.4 Phase Angle Shift

5.4.1 Overview of Phase Angle Shift

If the voltage sag were purely sinusoidal in waveshape, then this computation would be straightforward. Unfortunately, the waveforms are often distorted and the computation of a phase angle is not easy since the harmonic components will have phase angles as well. Two acceptable methods:

· The Zero Crossing Method

· The Fundamental Phase Angle

5.4.2 Calculation of Zero Crossing Method

Ideally, a sinusoid crosses the zero axis twice per cycle, 180 degrees apart. Thus, if the phase angle of the waveform shifts and the basic waveshape does not change, the phase shift could be detected by examining the zero crossings. Of course, any noise or distortion which causes the zero crossings to occur multiple times per half cycle, or severe waveform distortion will render this technique of limited value. To miminize the effect of noise, an averaging technique is employed which uses a ¼ cycle moving window. The zero crossing is said to take place when half the samples in the window are greater than zero and half are less than zero.

The zero crossings of the actual waveform are compared to those of the reference waveform. The difference in time (phase angle) of the two waveforms are then reported as the phase shift. The phase shift is inherently “jumpy” when observed on a continuous time graph since the phase can only be computed two times per cycle (at the zero crossings). The phase angles are then connected with straight lines.

5.4.3 Calculation of Fundamental Phase Angle

The Discrete Fourier Transform (using an FFT) is used to compute the dc, fundamental (60 Hz component), and harmonics of the waveform using a one-cycle window. The fundamental component is the 60 Hz component which is normally desired. Many electronic devices incorporate filters which strip away all unwanted (non-60Hz) components. Therefore, the phase angle reported by the DFT will be similar to the one seen by many types of equipment.

Since the DFT needs a periodic signal, a one-cycle window is used. This “averages” a signal which is varying in much the same way the rms calculations will.

The DFT returns the real and imaginary parts of the waveform at each frequency component, A+jB. The magnitude and phase angle is computed from these. The phase angle is a single value, so it is outputted at the center of the window.

The phase angle needs to be a continuous function. Since the phase angle is computed as the difference in the phase angle of the fundamental component of the disturbed waveform and the phase angle of the ideal waveform, angular difference can exceed ±360^o. The arctangent function, from which the phase angle is computed, will only report an angle between –180^o and +180^o. Therefore, as the phase angles are subtracted from each other over a discontinuity, a discontinuity in the phase angle plot will occur. While not incorrect (e.g., 180^o = 540^o), the phase angle plot will be difficult to interpret. An alternative method to determine the phase angle without resulting in a discontinuity was developed where the functions are subtracted from each other before the arctangent is taken.

If the fundamental component has a rectangular form of A+jB and the ideal fundamental component is C+jD, then the phase shift is computed from:

5.4.4 Reported data of the waveforms from Phase Shift Technique

· Phase Shift: The phase angle between the ideal waveform and the disturbed waveform using either the zero crossing technique or the DFT fundamental method.

· Phase Shift at Sag Start (Df): This is computed in two different ways. One is using zero crossings and the other is using the DFT values. When using zero crossings, the phase shift computed at the first zero crossing after the “disturbed sag start.” When using the DFT method, the value is computed ½ cycle after the sag start.

· Maximum Phase Shift: The maximum phase angle recorded between the “apparent sag start” and “actual sag end” times.

· Minimum Phase Shift: The minimum phase angle recorded between the “apparent sag start” and “actual sag end” times.

· Average Phase Shift: The average phase angle recorded between the “apparent sag start” and “apparent sag end” times.

· Maximum ± df/dt (rate of change or “slew rate” of phase angle): This is the rate of change of the phase angle during the interval. The initial change at the “sag start” and then final change at the recovery is not included in this calculation since the sharp discontinuity which usually accompanies the beginning and end of the event will cause a very high and meaningless value.

5.5 Missing Voltage

5.5.1 Overview of Missing Voltage

The missing voltage is the instantaneous different between the actual waveform and the ideal waveform. In other words, this is the voltage that would be required to be injected to make the actual voltage waveform become ideal again. Since the missing voltage alone is not meaningful, the duration of the missing voltage is provided in two ways. Thus, an energy calculation could be made with this information, or statistics could get generated.

5.5.2 Calculation of Missing Voltage and its analyses

· The actual waveform is subtracted from the nominal waveform.

· The result is rectified to give all positive numbers.

· The voltage is subdivided into 5% bands.

· The missing voltage is reported as the peak value that falls into one of the bands at a specific instant of time.

5.5.3 Reported Data using Missing Voltage technique

· The amount of time (total number of sample points) that the missing voltage falls into a particular band (e.g., 30% to 35%) is tabulated onto a bar graph.

· The cumulative time that the missing voltage is greater than or equal to a particular value (i.e., ³ 30%) is shown on the same bar graph. This report will be particularly useful in determining how a series voltage compensation device will perform.

5.6 Unbalance

5.6.1 Overview of Unbalance

Many disturbances only affect one or two phases, or affect individual phases in different ways. A balanced sag is one in which all three phases sags to the same depth at the same time. The unbalance a measure of the difference in voltage among the phases. The residual value (sum of all three phases) can be of some value in determining the balance on an instantaneous basis. However, since this value oscillates with the voltage waveforms, the residual does not yield meaningful information. The unbalance ratio can be used to help give a meaningful value to the unbalance.

5.6.2 Calculation of Unbalance

The unbalance ratio is defined as the difference in the rms values between the highest and lowest phases, divided by the average of the rms values of the three phases.

5.6.3 Data Reported from Unbalance technique

This ratio is plotted on a continuous time basis using a moving window. The maximum unbalance during the interval is reported.

5.7 Harmonic Root Mean Square

5.7.1 Overview of Harmonic RMS

Most voltage disturbances are not sinusoidal. Instead, the event will often have a significant amount of distortion. A good method of gauging the distortion iss needed. Distortion is often measured using a value called the total harmonic distortion (THD). The THD reports the percent of the harmonic components relative to the fundamental. In most systems that use THD, the fundamental component is fixed in amplitude. Our waveforms do not have this property. For example, the fundamental component of the voltage will drop nearly to zero during a very deep voltage sag. As a result, scaling the harmonics relative to the fundamental is not meaningful. The harmonic rms value (hrms) is defined as the square root of the sum of the squares of all harmonic components at or above the second component. Thus it provides the same information as the THD without scaling it by the fundamental.

5.7.2 Calculations of Harmonic RMS

Two methods can be used to calculate the hrms value. The most direct way is to determine the value of each harmonic component at or above the second component, sum the squares of each component’s magnitude and then take the square root. However, a more computationally efficient method is to first find the total rms value (which is being calculated anyway), square it, subtract off the squares of the fundamental and dc components, and take the square root. This method is equivalent to the direct method since the total rms values is equal to the square root of the sum of the squares of the Fourier coefficients. The formula for hrms is given below:

5.7.3 Data Reported from Harmonic RMS Technique

HRMS

6 Bibliography

__________________________________________________________________________________

7 References

__________________________________________________________________________________

IEEE Std 100-1996, The IEEE Standard Dictionary of Electrical and Electronics Terms.

IEEE Std 1159-2001R, IEEE Recommended Practice for Power Quality Monitoring

IEC 1000-4-30, Testing and Measurement Techniques - Power Quality Measurement Methods

IEEE P1159.2 Draft December 1998

__________________________________________________________________________________

Charles Perry, Chair

Richard Bingham	Greg Olson
Math Bollen	Scott Peele
Randy Collins	Dan Sabin
Andrew Dettloff	Georges Simard
Erich Gunther	Timothy Unruh
Mark Kempker	David Vannoy
David Kreiss	Marek Waclawiak
Christopher Melhorn	James Wikston
R. Larry Morgan

Term

Accuracy

4.2.1.1 Voltage sag (dip) detection

4.2.1.2 Voltage sag (dip) evaluation

4.2.2.1 Voltage swell detection

4.2.2.2 Voltage swell evaluation

4.2.4 Measurement uncertainty

4.2.4.1 Retained voltage and maximum swell voltage magnitude measurement uncertainty