Termoelasticity-based fatigue life identification
Project description
Thermoelasticity-based fatigue life estimation
Estimation of vibration fatigue life from thermal images acquisition. The class is based on the modal damage via thermoelasticity [1] and on the FLife package [2].
Install packages using pip
.. code-block:: python
pip install IR_FLife
Import packages
.. code-block:: python
import numpy as np
import matplotlib.pyplot as plt
import pysfmov as sfmov
import IR_FLife as irfl
Material Parameters
.. code-block:: python
k = 6.51 # Slope endurance curve
B = 800.26 # Endurance curve limit [MPa]
C = 7.94 * 10**18 # Fatigue strenght [MPa**k]
km = 1.2 * 10**(-8) # Thermoelastic coefficient [Pa**(-1)]
Input thermal video
.. code-block:: python
filename = './data/rec.sfmov' # Filename of thermal acquisition
data = sfmov.get_data(filename) # Using pysfmov to open it as numpy array [°C]
fs = 400 # Smapling frequency [Hz]
dt = 1 / fs # Time step [s]
stress = 10 * (data / km ) * 10**-6 # Stress [MPa]
Class initialization
.. code-block:: python
td = irfl.IR_FLife(stress, dt) # Class initialization
Natural frequency identification from thermal video
If the location is picked with the mouse click
.. code-block:: python
td.loc_selection() # Mouse selection of central pixel of the roi
f = td.nf_identification() # Natural frequency identification
If the location id settled with roi cooirdinates
.. code-block:: python
band_pass = [5,100] # Band pass filter applied during the natural frequency identification
roi_size = 5 # ROI size [pixel]
location = (39, 79, 5, 5) # Location of interest in the field of view [pixel]
f = td.nf_identification(location = location, roi_size = roi_size, band_pass = band_pass)
Fatigue life estimation
If the fatigue life is wanted at a particular location
.. code-block:: python
location = (39, 79, 5, 5) # Location of interest in the field of view [pixel]
md = td.get_life(C, k, 'Modal', f = f, location = location)
tb = td.get_life(C, k, 'TovoBenasciutti', location = location)
dk = td.get_life(C, k, 'Dirlik', location = location)
rf = td.get_life(C, k, 'Rainflow', location = location)
print(f' Rainflow: {rf:4.0f} s')
print(f' Dirlik: {dk:4.0f} s')
print(f' Tovo-Benasciutti: {tb:4.0f} s')
print(f' Modal: {md:4.0f} s')
If the fatigue life is wanted in the spatial domain
.. code-block:: python
md = td.get_life(C, k, 'Modal', f = f)
tb = td.get_life(C, k, 'TovoBenasciutti')
dk = td.get_life(C, k, 'Dirlik')
rf = td.get_life(C, k, 'Rainflow')
Then, the spatial domain results are shown:
.. code-block:: python
plt.figure()
plt.subplot(2,2,1)
plt.imshow(dk)
plt.colorbar()
plt.title('Dirlik')
plt.subplot(2,2,2)
plt.imshow(md)
plt.colorbar()
plt.title('Modal approach')
plt.subplot(2,2,3)
plt.imshow(rf)
plt.colorbar()
plt.title('Rainflow')
plt.subplot(2,2,4)
plt.imshow(tb)
plt.colorbar()
plt.title('Tovo-Benasciutti')
Reference:
[1] Thermoelasticity-based modal damage identification. Lorenzo Capponi, Janko Slavič, Gianluca Rossi, Miha Boltežar International Journal of Fatigue (2020)
[2] Vibration Fatigue by Spectral Methods, From Structural Dynamics to Fatigue Damage – Theory and Experiments. Janko Slavič, Matjaž Mršnik, Martin Česnik, Jaka Javh, Miha Boltežar. ISBN: 9780128221907, Elsevier, 1st September 2020.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
File details
Details for the file IR_FLife-0.1.tar.gz
.
File metadata
- Download URL: IR_FLife-0.1.tar.gz
- Upload date:
- Size: 5.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/49.2.1 requests-toolbelt/0.9.1 tqdm/4.42.1 CPython/3.7.6
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 1a698135f5c76ac7798ed1d27802126be9ca16294c467e56663e6399457eefb2 |
|
MD5 | 46d73ef14f0f35d3fe6516ea14d1317d |
|
BLAKE2b-256 | ea1f2a5eda67a60d71e087d0a5e506599ee5158371eb78f37660a2b98bd95541 |