The FourPointDetector class

class pylife.stress.rainflow.FourPointDetector(recorder)[source]

Implements four point rainflow counting algorithm.

from pylife.stress.timesignal import TimeSignalGenerator
import pylife.stress.rainflow as RF

ts = TimeSignalGenerator(10, {
    'number': 50,
    'amplitude_median': 1.0, 'amplitude_std_dev': 0.5,
    'frequency_median': 4, 'frequency_std_dev': 3,
    'offset_median': 0, 'offset_std_dev': 0.4}, None, None).query(10000)

rfc = RF.FourPointDetector(recorder=RF.LoopValueRecorder())
rfc.process(ts)

rfc.recorder.collective
from to
0 1.779876 2.723641
1 -1.359676 -0.778804
2 -2.125195 6.859020
3 -3.348581 4.376982
4 -3.897363 7.454785
... ... ...
1151 -2.651181 -10.562509
1152 5.215703 -11.053213
1153 4.178518 4.147719
1154 7.240966 -11.056957
1155 10.316823 -13.913987

1156 rows × 2 columns

Alternatively you can ask the recorder for a histogram matrix:

rfc.recorder.histogram(bins=16)

from                                        to                                       
(-18.419472620225868, -16.351550689938875]  (-19.090203631591, -16.99318139628545]       0.0
                                            (-16.99318139628545, -14.8961591609799]      0.0
                                            (-14.8961591609799, -12.79913692567435]      0.0
                                            (-12.79913692567435, -10.702114690368802]    0.0
                                            (-10.702114690368802, -8.605092455063254]    0.0
                                                                                        ... 
(12.599356334079033, 14.667278264366024]    (3.9770409567700433, 6.074063192075592]      0.0
                                            (6.074063192075592, 8.17108542738114]        0.0
                                            (8.17108542738114, 10.268107662686692]       0.0
                                            (10.268107662686692, 12.36512989799224]      0.0
                                            (12.36512989799224, 14.462152133297785]      0.0
Length: 256, dtype: float64

We take four turning points into account to detect closed hysteresis loops.

Consider four consecutive peak/valley points say, A, B, C, and D If B and C are contained within A and B, then a cycle is counted from B to C; otherwise no cycle is counted.

i.e, If X ≥ Y AND Z ≥ Y then a cycle exist FROM = B and TO = C where, ranges X = |D–C|, Y = |C–B|, and Z = |B–A|

Load -----------------------------
|        x B               F x
--------/-\-----------------/-----
|      /   \   x D         /
------/-----\-/-\---------/-------
|    /     C x   \       /
--\-/-------------\-----/---------
|  x A             \   /
--------------------\-/-----------
|                    x E
----------------------------------
|              Time

So, if a cycle exsist from B to C then delete these peaks from the turns array and perform next iteration by joining A&D else if no cylce exsists, then B would be the next strarting point.

__init__(recorder)[source]

Instantiate a FourPointDetector.

Parameters:

recorder (subclass of AbstractRecorder) – The recorder that the detector will report to.

process(samples)[source]

Process a sample chunk.

Parameters:

samples (array_like, shape (N, )) – The samples to be processed

Returns:

self – The self object so that processing can be chained

Return type:

FourPointDetector