Creating Actionable Synchrophasor Information Using Osisoft S and SISCO Technologies

Application of Synchrophasors

Creating Actionable Synchrophasor Information using OSIsoft’s and SISCO technologies

Version: 0.001Date:

Systems Integration Specialist Company, Inc.
6605 19½ Mile Road
Sterling Heights, Michigan 48314

Creating Actionable Synchrophasor Information using OSIsoft’s and SISCOPage1
Technologies

Introduction

There are many different methods in which to discuss the value of Synchrophasors. Figure 1 depicts the generally accepted information flow from: the field data acquisition, transmission of the information through a communication network using IEEE C37.118 or IEC 61850, into concentrators that provide information to business applications and other analysis tools.

Figure 1: Synchrophasor Information Flow

CIGRE and other organizations have identified several of the key applications for which synchrophasor information can be used:

Frequency oriented applications consisting of monitoring/analysis of: oscillatory stability and frequency monitoring.
Voltage oriented applications consisting of monitoring/analysis of: angles, magnitude, and stability.
Power oriented applications consisting of monitoring/analysis of: oscillatory stability and power monitoring (e.g. both active and reactive).
State Estimation
Transformer Fault Detection
Transient Monitoring
Line Thermal Monitoring

This whitepaper will concentrate on the technologies/configuration aspects required for the aforementioned set of applications.

Frequency Oriented Applications

There are several well recognized uses/analysis of power system frequency information.

Monitoring of System Frequency: There is typically a single monitored frequency that represents a utility’s/operational area’s recognized “system frequency”. There needs to be a mechanism to monitor and display this information.
Detection of Instability due to Frequency Deviations
Detection of frequency oscillations: The oscillatory frequencies of interest are typically five(5) Hz or less.
Detection of harmonic/sub-harmonic shifts: The typical sub-synchronous frequencies of interest are 15-45Hz.
Flicker: The typical sub-synchronous frequencies of interest are 5-20Hz.

Besides the ability to acquire and display the required information, there are two(2) additional algorithms that can be leveraged.

Application
Difference Calculation / FFT
Monitoring
Frequency Deviations / X
Frequency Oscillations / X
Harmonics / X
Flicker / X

Application of FFTs

Based upon the analysis of the Frequency ranges required for each application, the following table can be developed:

Application / Frequency
Range (Hz) / PMU Reporting Rate / Conclusion/Recommendation
20 / 30 / 80 / 120
Oscillations / 0-2 / √ / √ / √ / √ / FFT spectral analysis is valid at all PMU reporting rates.
Harmonics / 15-45 / √ / Only information reported at a rate of 90 samples/second or better can be utilized to create a FFT for the specified frequency range.
If the user desires to detect harmonic shifts, it is recommended that the spectral power metric be used to detect such a shift.
Additionally, some field units may be able to supply the desired harmonic information.
Flicker / 5-20 / √ / √ / Only information reported at rates of 40 samples/second or better can be utilized to create an accurate FFT for the specified frequency range.
It is recommended that a field device be utilized to detect flicker as this is more common.

FFTs made Simple

The OSIsoft FFT interface is capable of performing the appropriate FFT calculations[hf1] as well as the following:

Calculate the FFT bins (e.g. frequencies) that have the most spectral power.
Calculate the spectral power which allows the easy detection of spectral (e.g. frequency shifts).

Of primary importance, to the use of FFTs, is to insure that the spectral “bins” are accurate. If the FFT calculation can’t result in accurate spectral information, then only the spectral shift capability is feasible to use to detect changes.

Harry Nyquist developed the required sampling theory as it relates to the required sampling rates when attempting to convert from the continuous (e.g. analog) to discrete (e.g. digital). The theory basically states that in order to represent an analog signal of frequency (f), the signal must be sampled at twice the signal frequency:

Sampling Rate = 2 * f

PMUs report samples at a given rate (e.g. the inherent sampling rate), therefore, the maximum signal frequency is determined by the PMU reporting rate.

PMU Reporting Rate/2 = maximum signal frequency (fmax)

PMU Reporting Rate
(measurements/second) / Maximum Frequency that can be Represented Digitally (Hz)
20 / 10
30 / 15
80 / 40
120 / 60

Therefore, if the PMU reporting rate does not allow the frequency desired, FFT spectral analysis will not yield accurate results (e.g. frequency decomposition/harmonic detection). However, the spectral power calculations of the OSIsoft FFT interface will still yield significant indications to the shift in spectral power, although the actual frequencies causing the shift will not be able to be determined.

The next decision point is the desired frequency resolution of the FFT. This can be calculated by:

Frequency Resolution = Sample Rate / FFT Points

Frequency Resolution = Frequency range / Bins numbers

In many cases, the user knows the frequency resolution desired and the frequency range of interest, but needs to determine the number of FFT “bins” that are required:

Bins = Frequency range/Frequency Resolution

Given the number of calculated “bins”, it is important to the prevalent FFT algorithms that the number of “bins” always be a power of 2.

As an example consider desiring a FFT over the Frequency range of 0-5 Hz with a Frequency Resolution of 0.1 Hz. The number of calculated bins would be 50 (e.g. 5/.1), but in reality the FFT would need to be configured for 64 bins. Therefore, the actual Frequency Resolution would be 0.078125 Hz.

There is one additional parameter required to allow an FFT to properly calculate the spectral components, that is the time required to obtain the FFT samples (a.k.a. Filling Time Buffer):

Filling Time Buffer = 1/Frequency Resolution

If the Filling Time is greater than the signal duration that is trying to be detected, the FFT will not be able to detect such a signal.

Continuing the 0-5 Hz example, the Filling Time would be: 12.8 seconds (e.g. 1/0.078125).

Note: This FFT section/configuration does not deal with aliasing of signal/frequencies within the FFT and this issue is ignored for the sake of simplicity.

Definitions and Abbreivations

PMU / Phasor Measurement Unit
Phasor Measurment Unit / A unit that monitors/measures the electrical network and provides Synchronized Phasor measurements as its output.
SynchroPhasor / (a.k.a. Synchronized Phasor): A phasor calculated from data samples using a standard time signal as the reference for the measurement. Synchronized phasors from remote sites have a defined common phase relationship. [from IEEE Std C37.118™-2005 IEEE Standard for Synchrophasors for Power Systems]
Synchronized Phasor / A phasor calculated from data samples using a standard time signal as the reference for the measurement. Synchronized phasors from remote sites have a defined common phase relationship. Syn: synchrophasor. [from IEEE Std C37.118™-2005 IEEE Standard for Synchrophasors for Power Systems]
Phasor / A complex equivalent of a simple cosine wave quantity such that the complex modulus is the cosine wave amplitude and the complex angle (in polar form) is the cosine wave phase angle. [from IEEE Std C37.118™-2005 IEEE Standard for Synchrophasors for Power Systems]
Nyquist rate / A rate that is twice the highest frequency component in the input analog signal. The analog signal must be sampled at a rate greater than the Nyquist rate to be represented accurately in digital form. [from IEEE Std C37.118™-2005 IEEE Standard for Synchrophasors for Power Systems]
Flicker / A variation of input voltage, either magnitude or frequency, sufficient in duration to allow visual observation of a change in electric light source intensity.[From: IEEE Std IEEE Recommended Practice for Powering and Grounding Electronic Equipment]
FFT / fast fourier transform

[hf1]Is it FFT or DFT