Signal features

Different features that can be calculated with the vibration signal

pyvib.features.LOG(x)

Exponential of the mean absolute logarithm

Parameters

x (float 1D array) – Signal

pyvib.features.SDofIHC(x)

Standard deviation of inverse hyperbolic cosine

Parameters

  • x (float 1D array) – Signal

  • paper (Taken from) –

  • Machinery (A Model-Based Method for Remaining Useful Life Prediction of) –

  • al. (Yaguo Lei et) –

pyvib.features.SDofIHS(x)

Standard deviation of inverse hyperbolic sine

Parameters

  • x (float 1D array) – Signal

  • paper (Taken from) –

  • Machinery (A Model-Based Method for Remaining Useful Life Prediction of) –

  • al. (Yaguo Lei et) –

pyvib.features.absoluteMean(x)

Get absolute mean of signal

Parameters

x (float 1D array) – Signal

pyvib.features.approximateEntropy(U, N=1000, m=2, r=None)

Approximate the entropy of a signal

Parameters

  • U (float 1D array) – Signal

  • N (int, optional) –

  • m (int, optional) –

  • r (float, optional) –

  • https (//en.wikipedia.org/wiki/Approximate_entropy) –

  • from (Default values taken) –

  • Yan –

  • Ruqiang –

  • Gao. (and Robert X.) –

  • monitoring." ("Approximate entropy as a diagnostic tool for machine health) –

  • (2007) (Mechanical Systems and Signal Processing 21.2) –

pyvib.features.bearingEnergy(Y, df, X, bearing)

Energy within band of typical characteristic frequencies

Parameters

  • Y (float 1D array) – Spectrum aplitude

  • df (float) – Frequncy spacing in Hz

  • X (float) – Shaft speed in Hz

  • bearing (float 1D array) – Bearing characteristic frequencies in orders (i.e. per revolution) bearing[0] - Inner race bearing[1] - 2x roller spin frequency bearing[2] - Cage frequency bearing[3] - Outer race frequency

pyvib.features.clearanceFactor(x)

Clearance factor

Parameters

x (float 1D array) – Signal

pyvib.features.crestFactor(x)

Crest factor

Parameters

x (float 1D array) – Signal

pyvib.features.frequencyCenter(f, Y)

Frequency center of spectrum

Parameters

  • f (float 1D array) – Frequency of FFT

  • Y (float 1D array) – Amplitude of FFT

pyvib.features.impulseFactor(x)

Impulse factor

Parameters

x (float 1D array) – Signal

pyvib.features.kurtosis(x)

Get kurtosis value

Parameters

x (float 1D array) – Signal

Returns

K – Kurtosis

Return type

float

pyvib.features.kurtosisFactor(x)

Kurtosis factor

Parameters

x (float 1D array) – Signal

pyvib.features.maxToMinPowerDensityDrop(Y, df, X, bearing)

Maximum to minimum power density drop

Parameters

  • Y (float 1D array) – Spectrum aplitude

  • df (float) – Frequncy spacing in Hz

  • X (float) – Shaft speed in Hz

  • bearing (float 1D array) – Bearing characteristic frequencies in orders (i.e. per revolution) bearing[0] - Inner race bearing[1] - 2x roller spin frequency bearing[2] - Cage frequency bearing[3] - Outer race frequency

pyvib.features.medianFrequency(Y, df)

Median frequency of a spectrum

Parameters

  • Y (float 1D array) – Spectrum aplitude

  • df (float) – Frequency spacing between bins

pyvib.features.myoPulsePercentage(x, eps=5.0)

Myo pulse percentage Sum of all impulses greater than a threshold eps

Parameters

  • x (float 1D array) – Signal

  • eps (float, opional) – Threshold

pyvib.features.peakToPeak(x)

Peak-to-peak of signal

Parameters

x (float 1D array) – Signal

pyvib.features.rms(y)

Get RMS vlaue

Parameters

y (float 1D array) – Signal

Returns

RMS – RMS

Return type

float

pyvib.features.rootMeanSquareFrequency(f, Y)

Root mean square frequency

Parameters

  • f (float 1D array) – Frequency of FFT

  • Y (float 1D array) – Amplitude of FFT

pyvib.features.shapeFactor(x)

Shape factor

Parameters

x (float 1D array) – Signal

pyvib.features.skewnessFactor(x)

Skewness factor

Parameters

x (float 1D array) – Signal

pyvib.features.slopeSignChange(x, epsilon=0.5)

Slope sign change

Parameters

  • x (float 1D array) – Signal

  • epsilon (float, optional) –

  • paper (Taken from) – Nayana, B. R., and P. Geethanjali. “Analysis of Statistical Time-Domain Features Effectiveness in Identification of Bearing Faults From Vibration Signal.” IEEE Sensors Journal 17.17 (2017): 5618-5625.

pyvib.features.snr(r, Fs, ma=0.05, mb=0.5, cb=3, mc=0.05, md=0.6, c=2, c_ech=0.05, J_min=3, toler=0.1)

Estimates the Signal-to-noise ratio.

Based on “About periodicity and signal to noise ratio - The strength of the autocorrelation function.” by Nadine Martin and Corinne Mailhes

Parameters

  • r (float 1D array) – The signal to estimate SNR of

  • Fs (float) – Sampling rate

  • ma (float) – Lag support start, percentage of r.size

  • mb (float) – Lag support end, percentage of r.size

  • cb (int) – Tolerance factor

  • mc (float) – Lag support start, percentage of r.size

  • md (float) – Lag support end, percentage of r.size

  • c (int) – Tolerance factor

  • c_ech (float) – Tolerance factor

  • J_min (int) – Minimum number of detected maxima

  • toler (float) – Tolerance factor applied to median

Returns

  • SNR_hat (float) – Estimated SNR ratio in dB

  • flags (list, size=4) – Extra information about the calculation

    Element[i]: 0 : boolean

    0 if r is not aperiodic (Positive) 1 if r is random noise

    1boolean

    0 if card(ksi) > J_min (Positive) 1 else

    2float

    Confidence value ratio

    3float

    Estimated fundamental frequency

pyvib.features.squareMeanRoot(x)

Square mean root of signal

Parameters

x (float 1D array) – Signal

pyvib.features.standardmoment(x, k)

Get standard moment of choice

Parameters

  • x (float 1D array) – Signal

  • k (int) – Desired moment

  • Returns –

  • SM (float) – Standard moment of choice

pyvib.features.waveformLength(x)

Waveform length of signal

Parameters

  • x (float 1D array) – Signal

  • paper (Taken from) – Nayana, B. R., and P. Geethanjali. “Analysis of Statistical Time-Domain Features Effectiveness in Identification of Bearing Faults From Vibration Signal.” IEEE Sensors Journal 17.17 (2017): 5618-5625.

pyvib.features.willsonAmplitude(x, epsilon=0.5)

Willson amplitude

Parameters

  • x (float 1D array) – Signal

  • paper (Taken from) – Nayana, B. R., and P. Geethanjali. “Analysis of Statistical Time-Domain Features Effectiveness in Identification of Bearing Faults From Vibration Signal.” IEEE Sensors Journal 17.17 (2017): 5618-5625.

pyvib.features.zeroCrossing(x, epsilon=0.5)

Zero crossing og signal

Parameters

  • x (float 1D array) – Signal

  • epsilon (float, optional) –

  • paper (Taken from) – Nayana, B. R., and P. Geethanjali. “Analysis of Statistical Time-Domain Features Effectiveness in Identification of Bearing Faults From Vibration Signal.” IEEE Sensors Journal 17.17 (2017): 5618-5625.