=Paper=
{{Paper
|id=Vol-3309/paper16
|storemode=property
|title=Test Data Processing Use for Structural Fatigue Life Assessment
|pdfUrl=https://ceur-ws.org/Vol-3309/paper16.pdf
|volume=Vol-3309
|authors=Mykola Stashkiv,Viktor Stashkiv,Iaroslav Lytvynenko
|dblpUrl=https://dblp.org/rec/conf/ittap/StashkivSL22
}}
==Test Data Processing Use for Structural Fatigue Life Assessment==
https://ceur-ws.org/Vol-3309/paper16.pdf
Test Data Processing Use for Structural Fatigue Life Assessment
Mykola Stashkiv, Iaroslav Lytvynenko, Viktor Stashkiv
Ternopil Ivan Puluj National Technical University, 56 Ruska street, Ternopil 46001, Ukraine
Abstract
Some technique and results of the strain-life fatigue analysis use test data digital processing
of the wide-spray field sprayer's booms have been described in the paper under discussion.
The test data were obtained from four channel of universal measuring system each of these
represents a uniaxial strain gauge placed in some potentially critical locations on the test
object. Two problems have been solved by fatigue analysis for each of the four channels,
namely the direct calculation of the fatigue life and back calculation for a scale factor that
gives the target fatigue life. The back calculation provides quantifiable stress or strain
reduction targets for a redesign of the wide-spray field sprayer's booms.
Keywords 1
Test data, signal, processing, software, glyph, data correction, fatigue, strain-life
1. Introduction
The working conditions of the engineering structures are often complex and the operating
environment is not always favorable. In the long-term operation, the strength of the structural
elements keeps decreasing and it fails eventually. A perusal of the broken parts in almost any
structure’s elements will show that a high number of failures occur at stresses below the yield strength
of their materials. This complex phenomenon is known as “Fatigue”.
Fatigue failure is one of the most typical failure modes of the structural elements, especially for the
mechanical structure under loads of random character. Fatigue is responsible for up to 90% of the in
service part failure which occur in industry [1].
Therefore, an important engineering problem is to study the fatigue of materials and structural
elements and predict the fatigue life of the structure. It makes possible to prevent fatigue failure of the
material and ensure the safety and stability of engineering structure during the design life [2].
Nowadays, the research on the fatigue state of mechanical structure is primarily focused on two
aspects: life prediction and reliability evaluation. The research on life prediction mainly targets the
prediction theory and method of the remaining service life of the mechanical structure, and is
relatively systematic.
The established life prediction methods predominantly include the stress-life fatigue theory (S-N
fatigue analysis or High Cycle Fatigue – HCF), strain-life theory fatigue (E-N fatigue analysis or Low
Cycle Fatigue – LCF), fracture mechanics theory, damage mechanics theory, and the probability
statistics based on the life prediction method.
Different authors use divers methods and approach to evaluate the fatigue life of structures.
In particular, in the paper [1] a fatigue life evaluation model based on equivalent elastic modulus is
proposed for in-service mechanical structure. In the proposed model, parameters that represent the
operating conditions of the mechanical structure, such as load, vibration, and shaft torque, etc., are
used as the generalized load. To replace the fatigue stress, the statistical method is used here, which is
also used in the conventional fatigue analysis method.
ITTAP’2022: 2nd International Workshop on Information Technologies: Theoretical and Applied Problems, November 22–24, 2022,
Ternopil, Ukraine
EMAIL: stashkiv@tntu.edu.ua (M. Stashkiv); iaroslav.lytvynenko@gmail.com (I. Lytvynenko); viktor.stashkiv@gmail.com (V. Stashkiv)
ORCID: 0000-0002-7325-8016 (M. Stashkiv); 0000-0001-7311-4103 (I. Lytvynenko)
©️ 2022 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
In the paper [3] the fatigue life calculation in operational load conditions is presented. The
obtained load runs were so processed as to get a set of sinusoidal cycles by using the following
methods: full cycles counting method and rainflow counting method. On the basis of such sets of
cycles of load were prepared block load spectra of equivalent amplitudes, obtained with the use of an
original method of these authors, in which two-parameter fatigue characteristics were applied. The
work resulted in comparison of fatigue life results for load spectra determined by using the assumed
cycles counting methods and the assumed two-parameter fatigue characteristics.
In the paper [4] the fatigue life is calculated for random loading with single overloads. There is
used the spectral method of fatigue life assessment. The transformation of the power spectral density
of the loading signal with the use of the correction factor based on the information obtained from the
spectral kurtosis is used to modification of the fatigue life assessment algorithm. The proposed
procedure is verified of the calculation results comparison with the results obtained by the rain cycle
calculation method.
In the paper [5] based on the Miner rule, a new linear damage accumulation rule is proposed to
consider the strengthening and damaging of low amplitude loads with different sequences using fuzzy
sets theory.
In the paper [6] the average value of variable loads is taking into account by determining a
substitute cycle characterized by mean value Sm = 0 and amplitude Saz ≠ Sa. Fatigue life calculations
were carried out based on hybrid method. This method combines calculations in strain approach and
in stress terms.
In the paper [7] has been considered the process of progressive localized permanent structural
changes occurring in a material subjected to conditions that produce fluctuating stresses or strains at
some points and that may culminate in fracture after a sufficient number of fluctuations when
subjected to Bending, Axial, Torsion or combined equivalent stresses. Are defined the life of the
machine components to determine the number of cycles to failure and Fatigue factor of safety.
In the paper [8] an energy-based approach was used to predict fatigue life under uniaxial and
multiaxial random loads. Such a method uses time-based damage accumulation model compared to
the classical cycle-based damage accumulation model.
A fatigue model can be selected in different ways. When there is no prior knowledge on the fatigue
case, a suitable fatigue model can be choose based on a few questions regarding loading conditions
and expected fatigue failure. In the diagram below are summarized the key questions which should
ask when select method evaluating fatigue (fig. 1).
Figure 1: Selection of the fatigue model type
Fatigue life calculation of a varying load spectrum is a complex task. There are various approaches
suggested to simplify the load spectrum into simple major and minor cycle. There are many different
methods are used for definition estimated number of equivalent load cycles, such as full cycle
counting method (FCM), peak counting method (PCM), simple-range counting method (RCM),
rainflow counting method (RFM), range-pair counting method (RPM), two-parametric fatigue
characteristics (TFC), etc.
In practice, the most often used are methods that do not limit the schematic load process and allow
processing any load mode. These include the methods: of maximums, of complete cycles, of ranges
(amplitudes), of augmented ranges, and "rain flow" method.
In the paper [9] is an analysis of the influence of cycle counting methods on fatigue life
calculations. Based on set of cycles, with variable parameters block load spectrums have been
developed for substitute amplitude, designated using the author's method, which uses two-parametric
fatigue characteristics.
In the paper [10] are examines the effect of cycle counting method on the estimated number of
equivalent cycles. Authors evaluate four different a cycle counting method: peak counting, level-
crossing, mean-crossing, and rainflow counting.
In the paper [11] a universally cycle counting procedure is presented that can be used to the
simplest uniaxial experiment, to the most complex experiment, variable amplitude and frequency, and
non-proportional multiaxial fatigue loading. Despite of its simplicity, the proposed cycle counting
method has different advantages compared to the known procedures. Its completely independent of
the damage criterion since the procedure standalone definition, allows to be combined it with any
damage criterion. The method does not require periodicity of the loading cycle, and therefore can be
directly used to analyze variable frequency and amplitude, multiaxial fatigue loading.
In the paper [12] a new algorithm for the Modified Graphical Rainflow counting method is
presented. This method allows the application of Miner's rule in order to assess the fatigue life of a
structure subject to complex loading.
The rainflow algorithms are one of the most popular counting methods used in fatigue and failure
analysis in conjunction with lifetime estimation models despite the nascence of new methods. This
method is also most often implemented in specialized software to analysis fatigue life.
Modern software makes it possible to analyze the fatigue failure of a structure under conditions of
long-term action of loads that vary over time, opens up unique opportunities for designing and
optimizing structures, including requirements for durability, product life under normal and extreme
operating conditions.
The use of software allows reducing the time of product development and eliminates the cost of
eliminating fatigue damage and destruction at an early stage of design. Also, during the operation of
the product, when fatigue damage is detected, software allows you to analyze the impact of these
damages on the structure life, preventing accidents and disasters.
The main source of input data for fatigue life analysis with the software is the previous results of
structure finite element analysis or the results of field tests. The quality of a fatigue life analysis is
thus directly dependent on the quality of the results (stress or strain) obtained from a structural
analysis or experiment.
Currently, there is a lot of different software that allows perform structural analysis and implement
the structure fatigue life analysis, in particular [13 - 17]: MSC Fatigue, AFGROW, SIMULIA fe-safe,
SolidWorks, COMSOL Multiphysics, ANSYS, nCode.
ANSYS software is especially popular among researchers. In particular, in the paper [14] the
principle of stochastic fatigue failure analysis is presented and the random vibration of a structure is
analyzed using ANSYS software.
In the paper [15] use ANSYS for fatigue life prediction rocket combustion chambers with finite
element analysis. Fatigue life prediction is an important part of design studies and loading cycle
optimization, but existing calculation methods based of a finite element method (FEM) are often
numerically inefficient. The authors propose a procedure of training an artificial neural network to
predict the number of cycles to failure based on the studying object geometry analysis. This
publication is one of the first to apply machine learning to fatigue life prediction.
Among the whole amount of software to processing experimental data, the nCode software of
HBM Prenscia [16, 17] is the most appropriate. It designed for working with large amounts of test
data, for signal processing, and performing various studies, such as, for example, fatigue life analysis.
nCode GlyphWorks software has a convenient object-oriented graphical interface and it is
optimized for the complex work with large volumes of multi-channel data. The software functional
includes a wide range of tools for temporary, frequency and statistical analysis of signals. Moreover,
some tools are available to assess both the resource and fatigue durability of the products and
synchronized reproduction of GPS signals, video and other data obtained during the tests, and also a
convenient mechanism for automated creation of reports has been implemented. nCode GlyphWorks
software has a module system which enables us to create the required feature set either by means of
large embedded libraries or by the use of language Python. Working templates in nCode GlyphWorks
can be prepared in advance and can be found in the specified library to provide reliability and high
speed of calculation [18].
2. Strain-life fatigue analysis
2.1. Test data obtaining and correcting
In this paper are implemented strain-life fatigue analysis use nCode GlyphWorks software's tools.
The goal of this analysis is to use the strain gauge measured test data to predict wide-spray field
sprayer booms fatigue life.
The test data were obtained using the methods, means and software presented in [19 - 22]. They
were obtained from four channel of universal measuring system each of these represents a uniaxial
strain gauge placed in some potentially critical locations on the test object.
In the paper [18] the test data have been corrected to remove the data drift effect by tools of the
nCode GlyphWorks. The obtained in [18] the time series data output file contains the stress, which is
the function in time. For low-cycle fatigue analyze the strain is decisive, therefore it is necessary to
convert the stress into strain.
It is known that stress within the limits of linear elastic strain (Hooke's law) is determined by the
formula [23]:
σ =ε ⋅E, (1)
where ε - is strain and
E – is modulus of elasticity (Young’s modulus), MPa. For steel E is about 2·105 MPa.
The input data must be converted into microstrain, therefore we will transform the data from the
channels according to formula (1) taking into account a factor of 10−6:
σ −6
ε = / 10 , (2)
E
Converting the stress into strain and strain-life fatigue analysis are implemented by the nCode
GlyphWorks - data processing system that contains a comprehensive set of standard and specialized
tools for performing durability analysis and other insightful tasks such as digital signal processing.
2.2. Project development and setting
Under test data digital processing conditions by specialized software tools nCode GlyphWorks
procedure of the strain-life fatigue assessment has been implemented.
GlyphWorks is a multi-channel, multi-file, multi-format environment for processing large amounts
of data. GlyphWorks provides a graphical, process–oriented environment that contains leading
analysis capabilities for research of various processes. GlyphWorks represents data analysis processes
graphically and lets drag and drop graphical representations of interactive data analysis processes that
allow create and save sophisticated working projects for later re-use [18].
The basic analysis building blocks used in GlyphWorks are termed glyphs. Glyphs are connected
by pipes, which contain the data that passes between glyphs and attach at the glyph’s pads (different
types of I / O pads are marked with different colors). In fact, Glyph is a calculation module (template)
with specified algorithms of certain functions performance and with possible setting of different
parameters of its properties. A set of glyphs with functional connections is the detailed design of the
research [18].
The main windows of the GlyphWorks interface are shown on fig. 2. The general view of Interface
nCode GlyphWorks can be changed by turning on the several other windows using the View menu:
1. Analysis Workspace – Where the process is created
2. Glyph Palette – Glyphs available for processing
3. Available Data – Data that can be analyzed
4. Diagnostics – List of process, error messages, etc.
Figure 2: The main window of the GlyphWorks interface
GlyphWorks glyphs are classified according to their functionality. The glyphs are available in the
glyph palette (the window on the right side on fig. 2). There are sections in the glyph palette for
glyphs that input data, perform basic digital signal processing (DSP), display results, and so on [18].
Glyphs are organized into the following categories (palettes), according to their functionality:
Input; Function; Basic DSP; Signal; Design Life; Frequency; Fatigue; Accelerated Testing; Optimized
Testing; Glyph Builder; SuperGlyph; Display and Output. Glyphs and input files, etc., can be dragged
onto the workspace from their respective palettes [18].
In GlyphWorks, a process is defined as a combination of glyphs that define a data flow. A process
typically starts with an input glyph to define the data to be processed. Additional glyphs define
subsequent steps in the process for calculation, display, or writing output [18].
The procedure of strain-life fatigue analysis is implemented according to the developed detailed
design (fig. 3) that contains the following glyphs:
1 – Time Series Input, 5 – Strain Life,
2 – RainFlow, 6 – Histogram Display,
3 – XY Display, 7 – MetaData Display and
4 – Arithmetic, 8 – XY Display.
Figure 3: The process strain-life fatigue analysis (completed)
Functional purpose of these glyphs, structural relations between glyphs and their parameters
settings are described below.
The obtained in [18] the time series data output file was uploaded to the glyph TimeSeriesInput
(fig.3, glyph 1). These time series graphs are displayed in the central window of the glyph.
The settings of calculation template Rainflow (fig. 4) is left as default.
Figure 4: Rainflow properties
Rainflow glyph (fig.3, glyph 2) counting results are a correlation between of the load cycles
number and stress. The glyph Rainflow output is connected with the XY Display glyph input (fig.3,
glyph 1). The XY Display glyph has been used to show the output data from Rainflow glyph (fig. 5).
Figure 5: Rainflow glyph counting result view in XY Display glyph
A special glyph Arithmetic (fig.3, glyph 4) makes possible to perform mathematical operations on
the data prepared in advance. This glyph input is connected with the output TimeSeriesInput glyph.
Setting of the Arithmetic1 glyph properties is shown on fig. 6. In the tab General in section
Operator must be choose the type of arithmetical operation Equation. In section EquationDefinition is
necessary to write the parameter Equation as « (Test 1/2e5)*10e6 » in accordance with equation (2).
Figure 6: Arithmetic glyph properties
We have connected a Strain Life glyph (fig.3, glyph 5) from the Fatigue palette to Time Series
Input glyph (TSInput1). The Strain Life glyph has six outputs, as detailed on fig 7.
Figure 7: Strain Life glyph inputs and outputs
Two of the outputs are for histogram data (red pads). The upper pad is for the histogram of damage
per bin, and the lower pad is for the corresponding rainflow cycle count matrix.
We will be connecting to these outputs the next glyphs (see the fig. 2):
– XY Display glyph input is connecting to the time series output (blue pads).
– Histogram Display glyph two inputs are connecting to the both Histogram outputs (red pads).
– Metadata Display glyph input is connecting to Fatigue Results output (green pads).
We connecting most of the outputs but we don't always need to do this - we can use only the pads
we want, and leave the others blank.
Before running the process, the properties of the strain life analysis need to be set; in this case, the
material properties need to be defined.
The StrainLife1 glyph property dialog has two tabs: Advanced and Materials. The Advanced tab
shows the glyph properties. The Mode property defines the type of calculation to be performed. In this
case, we will be using the Damage mode, which performs a fatigue damage calculation.
The other Mode options are “Kf”, which performs a back calculation to determine for each channel
the Kf required to achieve a certain desired fatigue life determined by the property TargetLife.
Similarly, the “ScaleFactor” mode performs a back calculation on scale factor on the input to achieve
a target life.
Under the Loading category (fig. 8), the InputUnits property = Auto that will attempt to sensibly
interpret the channel Y axis unit string. We set the units of the channels as uE, which the Strain Life
glyph will interpret correctly as microstrain. If is necessary to enforce other unit of strain can select
one from the InputUnits pull down menu.
Figure 8: The Advanced tab of the Strain Life glyph properties
Under the Material category (fig. 8), the MaterialDataSource is by default = MXD_database. This
requires the strain-life material parameters to be entered under the MaterialData category.
This can be a convenient way of enforcing certain material data in the process. However, we will
be selecting a material from the provided materials database file.
The default database is provided in the GlyphWorks installation and is of the nCode MXD format
used in other nCode software such as DesignLife. There are to pick and browse different materials
databases. The available sets of strain-life data are listed in the GUI.
The Materials tab window to choose a material is show on the fig. 9.
We select Hot Rolled HSLA Steel FeE255TM in the list by clicking it, or highlight it and click the
Apply button. This material name has now been entered the Advanced tab under MaterialName: Hot
Rolled HSLA Steel FeE255TM (fig. 8).
The other properties on this glyph were left at their defaults.
We used MaterialsManager to review the material data parameters selected. For this is necessary to
select the MaterialsManager іn the Main Menu (see the fig. 2).
Figure 9: Selecting material
This will display a dialog for connecting to a materials database. In our case, we need to select the
nCode materials file «iceflow_standard.mxd». For the Database Type is necessary to select nCode
MXD Material Database (fig. 10).
Figure 10: Materials Manager
On the fig.11 is showed a list of all available materials in this database. The default view of this set
of materials is a tree view. The basic material properties of the selected Hot Rolled HSLA Steel
FeE255TM are listed on the right side of the screen. These are static material properties like elastic
modulus and tensile strength.
In addition to static material properties, this database also contains material fatigue curves. The
expanded contents under Hot Rolled HSLA Steel FeE255TM shows any fatigue properties assigned to
this material. In this case, this FeE255TM material has a strain-life curve (fig. 12).
It could have any number of different types of fatigue curves.
Figure 11: Static material properties
Figure 12: Displaying fatigue curve with elastic and plastic lines
In addition to reviewing data, MaterialsManager provides also the ability to add new materials
data, edit existing materials data, delete existing materials data and view a plot of the data.
2.3. Strain-life fatigue analysis
The strain-life fatigue computation is carried out by starting the implementation of the developed
process using the button Run on the toolbar (see the fig. 2)
The available metadata from the calculation are defined as follows:
Metadata Name Description
DamageStatus Value set to 0 for a successful calculation. The value will be set to 1 if
the calculation failed or static failure occurred.
Duration Length of input time history.
Life Calculated life value. This value will be given as either a numerical
value in the equivalent units or as “Beyond cutoff” or “Static failure”.
MaxStrainLocal Maximum local strain (uE) including effect of Kf
MaxStrainNominal Maximum nominal strain (uE).
MinStrainLocal Minimum local strain (uE) including effect of Kf.
MinStrainNominal Minimum nominal strain (uE).
NumCyclesCounted The number of cycles counted.
TotalDamage Total calculated damage.
After the startup is complete we can analyze the computation results, in particular to review the
cyclic content by looking at the rainflow and damage histograms (fig. 13).
Figure 13: Damage and rainflow histograms
At the top and bottom there are the damage and rainflow histograms, respectively. The rainflow
histogram shows the number of cycles counts on the vertical axis, plotted against those cycles' ranges
and means. The largest number of cycles is seen on the left, where range is near zero – these are small
fatigue cycles. The biggest fatigue cycles are to the right. Fortunately, there are very few of these
occurrences. This is often typical of strain gauge data, which may have many small cycles and a few
large ones.
The damage histogram shows a similar plot, but the vertical axis shows fatigue damage. The
relationship between the number of cycles and fatigue life (and hence damage) is defined by the
material's EN curve. It is worth notice that all of those numerous small cycles seen in the rainflow
histogram don't contribute any fatigue damage, and that most of the fatigue content comes from those
few big cycles.
These plots are useful as diagnostic tools when trying to interpret fatigue results. These histograms
can help us understand the types of fatigue cycles seen in use and answered such questions as: how
big are the biggest cycles? How often do they occur? Are these cycles statistically representative of
actual usage, or are they outliers? How damaging are these big cycles? How sensitive are the fatigue
answers to the distribution of cycles?
These plots can also view as a top-down color map by selecting the TopView button in the toolbar.
We can also look at how damage accumulates over time (fig. 14). We can do this because the
strain gauge data was measured against time, and the fatigue damage calculations track cycles.
Figure 14: Time series of damage (XYDisplay1)
The XYDisplay1 glyph displays time histories of when damage occurred in the time domain for
each channel. As we saw earlier, the strain time history is rainflow cycle counted to understand cyclic
fatigue content. Each counted cycle's starting and ending times are recorded, and the amount of
fatigue damage caused by that cycle is calculated.
Half of the damage calculated for this cycle is assigned to the cycle's starting time and half the
damage to the ending time. This creates a time series with the same number of points as the original
channel, but the Y-axis value is now damage. This can be used to visualize damage accumulation in
the time domain – anywhere there's a damage spike, there must have been an important fatigue cycle.
We added the original strain gauge time histories to this plot by connecting TSInput1's blue output
pad (its only pad) to the second blue input pad on XYDisplay1. Then the glyph automatically changed
from showing four channels' damage to showing damage and strain for a single channel. The rest of
the strain gauge channels can be browse using the blue arrow NextChannel button in the toolbar. We
can also view these damage channels with a logarithmic Y axis, since damage values tend to vary
over orders of magnitude (fig. 15 - fig. 18).
Figure 15: Plotting damage and strain gauge data in the time domain for channel 1
Figure 16: Plotting damage and strain gauge data in the time domain for channel 2
Figure 17: Plotting damage and strain gauge data in the time domain for channel 3
Figure 18: Plotting damage and strain gauge data in the time domain for channel 4
The damage and strain gauge versus time plots we can see that the large strain magnitudes
contribute to fatigue damage. This can be helpful in diagnosing fatigue problems and can help to
answer the question: What causes the damage? What's happening when damaging fatigue cycles
occur? How is the part being used at that time?
The channel in this display could be anything that was recorded synchronously with the strain
gauge data – speed, temperature, acceleration, etc. – so it's easy to cross-reference fatigue
accumulation with other usage parameters. We can even link these damage time histories to GPS data
and see where the test article was when it was damaged.
А more thorough browse all the fatigue analysis results allow Metadata Display glyph. In this
glyph under StrainLife1_Results section we can see fatigue results for the each channel.
The Metadata Display glyph allows setting the Display Type to Custom the creation of a custom
table with different column content. The each channel now has its own row (fig. 19). This tabular
display is easier to view than the tree views that been earlier. Also there are Export and Copy buttons
that allow easy export results to other software programs such as Excel.
Figure 19: Simplified table of fatigue results
The longest fatigue life is 92670 time history repeats (for channel number 1). These repeats mean
repetitions of the entire input signal, with all the cycles that it contained.
The shortest fatigue life is 3981 time history repeats (for channel number 4). The fatigue life in
hours in this case is only about 484 hours.
The number of cycles per hour can be defined as 3981 / 484.3 ≈ 8.2.
Standard operating time of sprayers is 350 hours per season. For a service life of 7 years, this
amounts to 2450 hours of operation or about 2450*8.2 ≈ 20000 load cycles.
Thus, fatigue life is need to last about 20000 repeats in order to meet exploitation requirements.
In the current design, among the investigated hotspots are one over and two under-designed
relative to this life target.
The under-designed areas need some attention to improve their fatigue performance.
We can use the Strain Life glyph's back calculation capabilities to assess how the stress should be
increased or decreased to meet this target life. This is called a back calculation because we know the
fatigue life and want to calculate what would normally be an input parameter – the stress level.
This type of back calculation provides quantifiable stress or strain reduction targets for a redesign
or countermeasures. For this we set the StrainLife glyph's properties (fig. 8) as follows:
Mode = ScaleFactor and
TargetLife = 20000.
After running the process, can noticed that the calculation time increases because it has to
iteratively solve for a scale factor that gives the target fatigue life for each channel.
The Metadata Display hasn't showed yet any new answers because its column content doesn't
include them. To see the back calculated scale factors, need to edit Metadata Display properties,
adding StrainLife1_Results.ScaleFactor. The results table looks like in fig. 20.
Each channel's fatigue life is 20000 repeats +/- 1%. The last column is the back calculated scale
factor that results in the target life.
This scale factor answer is very important. It says that if channel 4 strain time history were
multiplied by 0.753, the fatigue life would increase to 20000 repeats.
Figure 20: Scale factor results from back calculation
In other words, a strain reduction of about 25% for channel 4 is required to meet the 20000 repeat
durability targets. Likewise, a 20% strain reduction is required for channel 3.
For the channel 1 show scale factors > 1, indicating that the element's area is currently over
designed and strain levels can be increase to meet the target life.
3. Conclusion
The developed procedure of the strain-life fatigue analysis under test data digital processing by
nCode GlyphWorks software's tools made it possible to calculate of the fatigue life for each of the
four channels and to study about what causes fatigue cycles and damage by viewing them in the
histogram and time domains.
Analysis of the time histories that display when damage occurred in the time domain for each
channel, made it possible to establish that the small fatigue cycles have the largest number but make
small damage. While the biggest fatigue cycles have small number but they create damage spike that
is the large strain magnitudes contribute to fatigue damage.
The results of the strain - life fatigue calculation show that load cycles number of the channels 3
and 4 do not provide required meeting the 20000 repeat durability targets (or 2450 hours).
The channels 4 and 3 are needed a strain reduction of about 25% and 20% respectively are
required to meet the needed durability targets.
The channel 2 strain level completely provide required to meet the durability targets.
The channel 1 strain level it is advisable to increase of about 23% to meet the target life.
The results of the strain-life fatigue analysis are preliminary and need to be clarified taking into
account structural, technological and operational features, such as, for example, stress concentrators,
technological defects, the presence of an aggressive environment, etc.
4. Acknowledgements
I am very grateful to the company HBM Prenscia and the team nCode for the possibility to use
their software and for the information support. My special thanks to Lukasz Pieniak – Account
Manager Prenscia.
5. References
[1] Xiqiang Ma, Fang Yang, Jishun Li, Yujun Xue and Zhiqiang Guan, Fatigue life assessment
method of in-service mechanical structure, Advances in Mechanical Engineering 2021,
Vol. 13(2), 1–9.
[2] M. M. Pedersen, Introduction to Metal Fatigue, 2018, Department of Engineering, Aarhus
University. Denmark, 91 pp.: Technical report ME-TR-11
[3] B. Ligaj, R. Sołtysiak, Problems of Equivalent Load Amplitude in Fatigue Life Calculations,
Polish Maritime Research, 1(89), 2016, Vol. 23, 85 – 92.: 10.1515/pomr-2016-0012
[4] M. Böhm, M. Kowalski, Fatigue life assessment algorithm modification in terms of taking into
account the effect of overloads in the frequency domain, Fatigue Failure and Fracture Mechanics
XXVII. AIP Conference Proceedings 2028, 020003 (2018). https://doi.org/10.1063/1.5066393
[5] Shun-Peng Zhu, Hong-Zhong Huang and Zhong-Lai Wang, Fatigue life estimation considering
damaging and strengthening of low amplitude loads under different load sequences using fuzzy
sets approach, International Journal of Damage Mechanics, Vol. 20, 2011, 876 – 899.
[6] B. Ligaj, K. Karolewska, Fatigue life calculations of structural elements by means of equivalent
load spectrum and material properties for LCF and HCF, Fatigue Failure and Fracture Mechanics
XXVII. AIP Conference Proceedings 2028, 020007 (2018). https://doi.org/10.1063/1.5066397
[7] M. Karthik Kumar, N. Rakshith, N. Yathisha, S. Rohith, Fatigue Life Estimation of Machine
Components, International Research Journal of Engineering and Technology (IRJET), Volume
05, Issue 06, 2018, 1328- 1333.
[8] Shih-Chuan Tien, Haoyang Wei, Jie Chen, Yongming Liu, Energy-based time derivative damage
accumulation model under uniaxial and multiaxial random loadings, Fatigue & Fracture of
Engineering Materials & Structures (FFEMS), 2022, Volume 45, Issue 1, 159-173.
[9] B. Ligaj, An analysis of the influence of cycle counting methods on fatigue life calculations of
steel, Scientific problems of machines operation and maintenance, 4 (168), 2011, 25-43.
[10] R. Stelzer, B. Carlton, S. Mazzoni, Comparison of cycle counting methods for potential
liquefaction and structural fatigue assessment, 17th World Conference on Earthquake
Engineering, 17WCEE, 2020, 11 p.
[11] K. G. F. Janssens, A universal, damage-criterion-independent cycle counting method for
multiaxial and variable amplitude fatigue, MATEC Web of Conferences 300, 17004 (2019), 8 p.
https://doi.org/10.1051/matecconf/201930017004
[12] V. Shinde, J. Jha, A. Tewari, S. Miashra, Modified rainflow counting algorithm for fatigue life
calculation, Proceedings of Fatigue, Durability and Fracture Mechanics, 2018, pp. 381–387.
DOI: 10.1007/978-981-10-6002-1_30
[13] Fatigue Module User’s Guide. COMSOL Multiphysics, 1998–2017 COMSOL, 94 p.
[14] Song Tao, Bin Chen, Xing-Jun Fan, Structural fatigue life prediction based on ANSYS random
vibration analysis, The Annual International Conference on Material Science and Engineering
(ICMSE 2016), 235 – 237.
[15] Kai Dresiay, Günther Waxenegger-Wilfing, Jörg Riccius, Jan Deeken and Michael Oschwald,
Numerically efficient fatigue life prediction of rocket combustion chambers using artificial
neural networks, 8th European Conference for Aeronautics and Aerospace Sciences (EUCASS),
2019, 15 p. DOI: 10.13009/EUCASS2019-264.
[16] https://www.hbmprenscia.com/
[17] https://www.ncode.com/
[18] M. Stashkiv, O. Matsiuk, nCode GlyphWorks software use for test data processing, The 1st
International Workshop on Information Technologies: Theoretical and Applied Problems 2021
(ITTAP 2021), Vol-3039, 192-205.
[19] R. Hevko, M. Stashkiv, O. Lyashuk, Y. Vovk, V. Oleksyuk, O. Tson, I. Bortnyk, Investigation of
internal efforts in the components of the crop sprayer boom section, Journal of Achievements in
Materials and Manufacturing Engineering, Volume 105, Issue 1 (2021), pp. 33 – 41.
DOI: 10.5604/01.3001.0014.8743.
[20] M. Stashkiv, P. Havrilenko, M. Havrilenko, P. Palamarchuk, Ih. Bortnyk, Stand for dynamic
loads testing of sprayer’s boom, 2017. Utility model patent № 118740 (u 2017 01793), Filed
October 20st., 2016, Issued March 11th., 2017. [in Ukrainian].
[21] P. Palamarchuk, M. Havrylenko, M. Stashkiv and Ih. Bortnyk, Stand tests of crop sprayer booms,
National interdepartmental scientific and technical collection "Design, manufacture and
operation of agricultural machinery", KNTU, 2015, Issue 45, Part II, 98 - 102. [in Ukrainian].
[22] T. Rybak, Yu. Gladyo, M. Stashkiv, P. Palamarchuk, Software "File converter for experimental
data RGSP-v 1.0", 2015, Certificate of copyright registration for the work № 63257, Filed Apr.
20st., 2015, Issued Dec. 29th., 2015.[in Ukrainian].
[23] S. Timoshenko, Strength of Materials, Part I: Elementary Theory and Problems, Krieger Pub Co;
3rd edition, 1983, 1010 p.