High Fatigue Damage Research Articles

Fatigue deformation behavior and fracture mechanism of high fatigue-resistant laser directed energy deposition fabricated titanium alloy: Effect of multi-scale microstructure

The application of the laser directed energy deposition (LDED) titanium (Ti) alloys has been severely impeded by their poor low cycle fatigue (LCF) properties in aerospace industry. Herein, we propose to improve the LCF properties of LDED-ed Ti6Al4V alloys via microstructure adjustment. A solid solution aging 850AA treatment is made to regulate the microstructure into a multi-scale composition dominated by discontinuous αGB and coarsened αP+(αS+β) phases. The LCF properties of LDED-ed Ti6Al4V with a multi-scale microstructure even outperform the wrought Ti6Al4V at high strain amplitudes. Furthermore, this multi-scale microstructure has special local plastic deformation behavior and dislocation activities under cyclic stress loading, which strongly affects the internal stress evolution and crack propagation. The fatigue softening behavior is attributed to the combined effect of back stress and friction stress. Fatigue cracks are deflected by the α/β boundary when the angle (φ) between the long axis of α lamellae and the crack direction is less than 35°. The transgranular fracture occurs when φ is close to 90° (70°–90°). Meanwhile, the fatigue crack usually propagates along the basal or prismatic planes with high Schmid factors. These findings provide novel insights into the microstructure design of LDED-ed Ti alloys with high fatigue damage tolerance.

Bond performance of freeze-thaw damaged concrete and reinforcing bars subjected to high-cycle fatigue loading

Concrete bridges are susceptible to the detrimental effects of freeze-thaw cycles (FTCs) and dynamic vehicle loading in cold regions, compromising the bond between reinforcing bars and concrete. After high-cycle fatigue loading using eccentric pull-out tests, an experiment on the bond performance of freeze-thaw damage specimens was carried out. The investigation focused on the impact of FTCs, fatigue load cycles (FLCs), and concrete strengths on the bond behaviour. The test results showed that eccentric pull-out specimens of reinforced concrete with freeze-thaw damage under high-cycle fatigue loading were more susceptible to experiencing splitting failure. As the number of FTCs increased, the peak and the residual bond strength decreased continuously. Fatigue loading had a slight improvement effect on the peak bond strength, but it significantly reduced the residual bond strength when the specimens suffered from higher fatigue damage. The peak slip initially increased and subsequently decreased with FTCs, showing minimal sensitivity to the number of FLCs. Considering the stiffness degradation of the ascending section, a bond-slip model for freeze-thaw damaged concrete and reinforcing bars under high-cycle fatigue loading was established, and verified through experimental results.

Fatigue assessment of suspended inter-array power cables for floating offshore wind turbines

The present paper aims to investigate the fatigue behavior of suspended inter-array power cables connecting two floating offshore wind turbines (FOWTs). Two spar FOWTs interconnected with a dynamic power cable are chosen as the base configuration for the present investigation. The base configuration is modeled in the dynamic analysis software OrcaFlex. The relationship between the axial tensions and curvatures of the power cable and its stresses is obtained by cross-section analysis in UFLEX (special purpose Finite Element program for nonlinear stress analysis of complex umbilical and cable cross-sections). Several different environmental conditions and load directions are applied. A fatigue life of 85 years is determined for the base configuration, with curvature-induced stresses being identified as the main contributor to the fatigue damage. Effects of marine growth on the fatigue life of the power cable are investigated. It is observed that marine growth negatively affects the fatigue life of the power cable. A parametric study on the effect of the number of buoys attached to the cable on the fatigue life is performed. Decreasing the number of buoys below the optimal value leads to an increase in fatigue life due to lower curvatures. In addition to the base configuration, an alternative configuration using semi-submersible floaters is investigated under the same metocean conditions and cable geometry. The power cable fatigue life is higher for semi-submersible floaters compared to the spar floaters. The present findings reveal that the main reason is the reduced number of large amplitude curvature cycles. The critical component prone to fatigue failure for all evaluated cases is the armor layer, and the largest fatigue damage always occurs at buoy-cable connection points. The critical direction of environmental load resulting in the highest fatigue damage is found to be the load case where waves, wind and current approach the suspended cable in a crossflow direction.

Methodology for Application of Damage Mechanics Approach to Model High Temperature Fatigue Damage Evolution in a Turbine Disc Superalloy

Aeroengine gas turbine components operate under complex loading environments. Turbine disc is one such component which experiences stresses at high temperature and accumulates life critical cyclic damage in the material during usage. This accumulation of cyclic damage results into significant deterioration in material strength which in turn may initiate the failure in these rotating components. This necessitates the need to develop an advanced lifing approach for fatigue life assessment of turbine disc. As there are no available standards for damage mechanics application, an attempt has been made in the present study to develop a methodology for application of damage mechanics approach to model high temperature fatigue damage evolution in a turbine disc Superalloy. High temperature (650oC) stress-controlled fatigue tests on turbine disc alloy have been performed to evaluate parameters for damage mechanics based models. Using this approach, damage evolution has been simulated at specimen level. A good correlation has been observed in the damage mechanics based model’s predicted damage and experimentally determined values.

Numerical study of an ice-offshore wind turbine structure interaction with the pile-soil interaction under stochastic wind loads

In order to better understand the ice-induced collision/vibration mechanism of offshore wind turbines (OWTs) with ice-breaking cones, the Tsai-Wu yield criterion of ice for bending failure is coupled to the nonlinear collision simulation tool ANSYS/LS-DYNA in this paper, and a numerical approach is adopted to predict dynamic ice force. Through comparisons of ice force between simulations and physical tests, the accuracy of dynamic ice force simulated by the proposed approach is verified. Meanwhile, to consider the pile-soil interaction (PSI) effect, the Euler-Bernoulli Beam theory and the Timoshenko Beam theory are used, respectively, and the dynamic characteristics of the OWT are compared and analyzed. The discrepancies between the calculation results of the two methods and the influence of PSI on OWTs under combined wind-ice conditions are discovered. Further, the fully coupled interaction model between ice and flexible OWT with PSI under wind and ice loadings is established. Based on the full three-dimensional (3D) interaction model, the influence of the ice velocities, ice thicknesses, and cone angles on the dynamic ice force and structural response is investigated. The area damage rate and fatigue damage rate of the monopile foundation are proposed to further carry out damage and fatigue analyses under wind and ice loads. Higher fatigue damage is associated with smaller area damage rate. In addition, the ice thickness has a more significant effect on OWT damage and fatigue compared with the ice velocity.

Toward developing Ti alloys with high fatigue crack growth resistance by additive manufacturing

Fatigue crack growth behaviors were investigated by three-point bending tests for TA19 alloy fabricated by laser metal deposition and four kinds of heat-treated samples. The crack growth resistance of the TA19 samples in the near-threshold regime and Paris regime was evaluated through the experimental characterization and theoretical analysis of the interaction between fatigue crack and α/β phase interface, columnar prior-β grain boundary and colony boundary. The results show that in the near-threshold regime, the fatigue crack propagation threshold and resistance increase with the increase of widths of lamellar αp phases and colonies, and the decrease of the number of α laths with an angle (φ) relative to the applied stress direction ranging from 75° to 90°. In the Paris regime, the fatigue cracking path can be deflected at colony boundaries or columnar prior-β grain boundaries. The larger the deflection angle, the more tortuous the cracking path and the lower the fatigue crack growth rate. The angle (γ) of the columnar prior-β grain growth direction relative to the build direction affects not only φ of different α variants, but also the fatigue cracking path deflection angle (θij) at columnar prior-β grain boundaries. An optimal combination of γ = 0°-15°-0°-15° for several adjacent columnar prior-β grains is derived from the theoretical analysis, and that can effectively avoid φ being in the range from 75° to 90° and make θij as large as possible. Such findings provide a guide for the selection of scanning strategies and process parameters to additively manufacture Ti alloys with high fatigue damage tolerance.

Effect of PAN fiber with bionic layered surface structure generated in situ by Fenton reaction on performance of SBS/RP asphalt binder

• Innovative application of Fenton reaction in the modification of polyacrylonitrile fibers. • Biomimetic multi-layered structure modified polyacrylonitrile fibers. • KF-PAN/SBS/RP modified asphalt had better viscoelastic performance. • KF-PAN/SBS/RP modified asphalt had better low temperature anti-cracking performance, high temperature stability and fatigue damage resistance. • KF-PAN/SBS/RP modified asphalt had better rutting resistance and high temperature adhesion performance. In this study, a surface modified polyacrylonitrile fiber (KF-PAN) with a fish-scale whisker-like biomimetic hierarchical structure was successfully prepared via in situ competition effect of Fenton reaction. The modified fibers were used as secondary modifiers to prepare fiber-reinforced styrene–butadienestyrene (SBS) block copolymers and rubber powder (RP) modified asphalt binders (KF-PAN/SBS/RP). Energy dispersive spectroscopy, scanning electron microscopy, and atomic force microscopy analyses revealed the successful construction of bionic hierarchy on the fiber surface, and confirmed rough surface of the KF-PAN fibers. The hierarchical structure of the fiber surface acted as a buffer in the asphalt and strengthened the mutual interaction of the fiber and asphalt. Dynamic rheometer shear and strip-tensile tests showed that the KF-PAN/SBS/RP modified asphalt exhibited better viscoelasticity, rutting resistance, and tensile properties than the PAN/SBS/RP modified asphalt. The linear amplitude sweep test showed that KF-PAN/SBS/RP modified asphalt had better fatigue resistance. Thermogravimetric analysis tests indicated that the surface bionic structure of KF-PAN fibers boosted thermal stability of the fibers, which in turn enhanced the thermal stability of KF-PAN/SBS/RP modified asphalt. Fluorescence microscopy analysis showed that incorporating KF-PAN fibers improved the SBS and RP microstructure inside the asphalt, forming a good mesh structure, and alleviating the stress concentration phenomenon in SBS/RP asphalt under external traffic load.

Understanding the higher harmonics of vortex-induced vibration response using a trend-constrained, machine learning approach

The spectra from cross-flow VIV signals contain peaks at the dominant vortex shedding frequency but also at several other frequencies, notably at three times and five times that frequency. These higher harmonic contributions are important because they are associated with high fatigue damage rates. The understanding of what controls higher harmonic response is far from complete. This paper presents a trend-constrained, data-driven model to discover important features (parameters) affecting the higher harmonic response of flexible cylinders subjected to vortex-induced vibrations. The predicted dependent parameter is the ratio of stress at the 3rd harmonic divided by the stress at the dominant VIV frequency. The known effects of damping and bending stiffness are introduced as physical constraints to improve the DNN predictions and aid in important parameter identification. The machine learning predictions with and without prior physical constraints are compared. The comparison suggests that the machine learning model with prior physical constraints better handles independent experimental datasets. It is confirmed that the higher stress ratios are associated with smaller damping parameter values and smaller bending stiffness ratios. The larger stress ratio is also found to be associated with traveling waves and single-mode-dominated responses.

Multiaxial fatigue assessment of welded joints using a principal component-based measure for non-proportionality

Non-proportional multiaxial stresses in welded structures cause higher fatigue damage than proportional stresses. In this paper, a new approach based on principal component analysis for quantifying the level of non-proportionality in stress signals is developed. The quantifier is developed for large-scale structures (civil and offshore structures), where the influence of mean stresses is negligible. The proposed quantifier is compared and validated with other approaches from literature with the relevant fatigue criteria and experimental data. The results show that the proposed principal component-based quantifier can capture the level of non-proportionality accurately, which can be used for accurate fatigue life estimations.

Study of dual grain microstructure heat treatment in steel specimen representing gas turbine rotor disc

Gas turbine discs which support the rotor blades operate at relatively high speeds and high rim temperatures and are therefore subjected high creep damage at the rim area and to very high low-cycle fatigue damage at the bore. Any heat treatment which results in homogenous grain structure and one small band of grain size will limit the life of the disc either by creep at the rim or by the fatigue at the bore. If the heat treatment parameters are optimized to increase the grain size at the rim area closer to the ASTM 6.5 and to finer grain size at the bore area closer to ASTM 11.5, the heat treated disc is likely to have much superior operational life as compared to that of heat treated which results in homogenous grain structure and size throughout the disc. There is therefore the need for exploring possibilities of obtaining variable grain structure during heat treatment operation.This study is aimed at evaluating two possible heat treatment processes that are likely to result in distinctly different grain size in a single test specimen such as a circular rod. These test rods represent rotating gas turbine discs and the objective of achieving differential grain size is to optimize and balance the creep life at one half portion of the test rod (representing the disc rim or Outer Diameter) and fatigue life at another half portion of the test rod(representing the bore or Inner Diameter).

A comparison of two dynamic power cable configurations for a floating offshore wind turbine in shallow water

For the study of the double wave configuration for the dynamic power cable of a Floating Offshore Wind Turbine (FOWT) in shallow water, this paper presented a comparison of the hydrostatic and hydrodynamic performance between a lazy wave and a double wave configuration. The double wave shape’s parametric presentation was first developed based on an extension of a previous parametric lazy wave shape. An aero-hydro-servo-elastic integrated model fully coupling the FOWT and the dynamic power cable was then proposed in the SIMO/RIFLEX code for numerical simulation. Subsequently, the strain-cycle curve and the rain flow counting method were employed to estimate the copper conductor’s fatigue damage. The bending moment was proved to provide the major fatigue contribution. The results showed that the double wave offered superior performance than the lazy wave. The second arc of the double wave configuration was proved critical for suffering high curvature, tension, and fatigue damage. A lower position and less hogged wave arc were suggested for the determination of the double wave shape.

Probabilistic Life Models for Steel Structures Subject to Creep- Fatigue Damage

When metal structures are subjected to long-term cyclic loading at high temperature, simultaneous creep and fatigue damage may occur. In this paper probabilistic life models, described by hold times in tension and total strain range at elevated temperature have been derived based on the creeprupture behavior of 316FR austenitic stainless steel, which is one of the candidate structural materials for fast reactors and future Generation IV nuclear power plants operating at high temperatures. The parameters of the proposed creepfatigue model were estimated using a standard Bayesian regression approach. This approach has been performed using the WinBUGS software tool, which is an open source Bayesian analysis software tool that uses the Markov Chain Monte Carlo sampling method. The results have shown a reasonable fit between the experimental data and the proposed probabilistic creep-fatigue life assessment models. The models are useful for predicting expended life of the critical structures in advanced high temperature reactors when performing structural health management.

Power density based fatigue load spectrum editing for accelerated durability testing for tractor front axles

The fatigue load spectrum is one of the most important parts in accelerated durability testing of agricultural machinery; however, the traditional editing approach just focuses on influence of load amplitude on mechanical components, which is not suitable for the agricultural machinery, the loads of which are broad in frequency bandwidth and time-varying in stress amplitude. Therefore, considering the comprehensive influence of load amplitude and frequency on the fatigue life of materials, a power-density-based fatigue load spectrum editing (PD-LSD) approach for accelerated durability testing is proposed ensuring that fatigue load spectrum for accelerated durability testing of tractor components can reproduce operation load characteristics of components. The concept of power density is introduced, and a fatigue analysis method based on power density and short-time Fourier transform (STFT) is discussed. For the approach of this study, STFT and stress-life (S–N) curve were used to derive accumulative power density (AccPD) of the load signal. The AccPD identified and extracted high fatigue damage segments of original load signal and obtained reduced accelerated load spectrum of components. Based on the load signal of the 88-kW tractor front axle, this approach and the time damage editing approach were applied to edit the load signal. The results show that, the accelerated load Signal Ⅰ based on the PD-LSD approach with a shorter length of time contained almost same damage of the original load signal, and the statistical parameters (mean value, root mean square and kurtosis coefficient) and the signal amplitude domain of the accelerated load Signal Ⅰ were more consistent with the original load signal, which verifies the acceleration, accuracy and effectiveness of the approach. This research supports fatigue testing and durability analysis for a key component of agricultural machinery.

Deflection and stresses in solar central receivers

The aim of the design of central solar receivers is to withstand the high non-uniform solar-heat-flux and temperature during the solar-power-plant lifetime. This high non-uniform tube temperature causes high thermal stress, producing creep and fatigue damage. Therefore, is necessary to obtain an accurate estimation of the tube stresses during the receiver operation. In the same way, to ensure the panel integrity, the frontal and lateral tube deflections must be obtained to avoid excessive panel bowing and warpage, respectively.The huge number of simulations needed to perform the creep-fatigue analysis precludes the use of high time-consuming CFD-FEM simulations. To resolve this drawback, a reliable, accurate and fast procedure to obtain the tube stresses, using analytical stress estimation, is proposed. The procedure considers the temperature dependence of the thermo-mechanical properties.The temperature-dependent hoop stress is estimated using the solution for constant mechanical properties whereas the radial stress is estimated taking constant the Young modulus only. The temperature-dependent axial-bending stress is obtained using the non-homogeneous beam equation subjected to the movement restriction produced by tube clips. When the tube displacement is restricted by tube clips, the equivalent stress difference is less than 2% taking temperature-dependent properties and slightly higher than 10% for constant properties. The proposed stress estimation is enough accurate to perform a reliable fatigue-creep analysis and two order of magnitude faster than the CFD-FEM simulations.Finally, the tube deflection and displacement, restricted by tube clips, are derived straightforward using the temperature-dependent tube curvature and the beam theory.

Offshore Wind Turbine Loads and Motions in Unstable Atmospheric Conditions

Even though it is widely known that unstable atmospheric stability conditions can lead to higher turbulence, the use of proper turbulent wind models considering unstable conditions are not often used in the simulation of loads and motions of offshore wind turbines. For this reason, the Højstrup model, which was specifically developed for unstable conditions, is used to simulate a spar-buoy offshore wind turbine (OWT) and investigate the importance of unstable conditions in the design of floating offshore wind turbines. It is found that fatigue damage of a spar-buoy OWT is strongly influenced by unstable conditions, where very unstable condition gives 65% higher fatigue damage than neutral conditions for the tower top torsion, followed by 37% higher for tower base side-side bending and 24% higher for blade root flap-wise mode.

Fatigue Damage Assessment to a Rigid Planar Jumper on Model Scale

A rigid jumper is an important part of a subsea production system, and it may experience significant vortex-induced vibrations (VIVs) if subjected to current. It has normally a non-straight geometry shape in three-dimensional space. Consequently, the response of a rigid jumper under VIVs is much more complicated compared with straight pipeline structures. Currently, there are very limited studies and design guidelines including methods on how to assess the fatigue damage of rigid jumpers under VIVs. The methodology used for straight pipelines is often applied by ignoring the non-straight geometry characteristics and the multiaxial stress states. However, both experimental and numerical results show that the torsional stress does exist besides the flexural stress for rigid jumpers under VIVs. The objective of this study is to do a fatigue assessment practice based on the state-of-the-art calculation methods to a rigid jumper on model scale. The VIV response is obtained from experimental tests and numerical calculations by either force or response model methods. The influence of torsional stress on fatigue assessment is studied. Two approaches have been investigated. In the first approach, the flexural and torsional stresses are evaluated separately. The second approach uses the first principal stress to calculate the fatigue damage; thus, the flexural and torsional stresses are evaluated together. It appears that the use of the first principal stress gives higher fatigue damage if the torsional stress contribution is significant. Furthermore, the principal stress method is also less time-consuming in processing the results.

Low temperature self-healing character of asphalt mixtures under different fatigue damage degrees

The primary objective of this study is to advance the understanding of the low temperature self-healing character of asphalt mixtures under different damage degrees, thus to determine the effective strategy of asphalt pavement maintenance. Firstly, three kinds of asphalt mixtures are selected to conduct the indirect tensile (IDT) fatigue test to a certain fatigue damage degree at low temperatures, and then the resilient modulus (Mr) at different rest time is measured to quantify the healing potential. Next, the fatigue loading with different intermittent time (0 s, 1 s and 3 s) is applied to determine the impact of intermittent time on healing potential. The results indicate that the descending order of healing potential of asphalt mixtures is: SMA-11 > AC-8 > AC-11 at 5 °C and −5°C. The loading intermittent time has an obvious effect on the fatigue damage state of asphalt mixtures, while the longer the intermittent time, the less the effect on fatigue damage healing. Besides, the fatigue damage state has great influence on its healing potential of asphalt mixture. Under the low damage conditions, the initial healing rate is greater than the long term healing rate. However, the low temperature (−5°C) dramatically reduces the healing rate of asphalt mixtures, and causes their long-term healing rate to stabilize gradually to a very low level. Especially under the high fatigue damage conditions, the healing potential of asphalt mixtures will almost disappear at −5°C. Furthermore, together with meso-scale Computed Tomography (CT) scanning technique, it is found that the internal crack distribution characteristics of different graded asphalt mixtures are different even under the same damage degree, which may explain the differences in the healing potential of asphalt mixtures. The use of a fast two-dimensional (2D) scanning technology further confirms that the crack zones inside the asphalt mixture are gradually shrinking after a period of high temperature healing. Finally, the Grey relational analysis reveals that the healing time has the most significant influence on the healing potential of asphalt mixtures. The gradation type and temperature have the similar influence level on the healing potential. The correlation degree between the fatigue damage degree and healing potential is the smallest compared with the other three factors.

Numerical analyses of rigid and flexible pavements responses under heavy vehicles’ loading

In an attempt to better understand the interaction between heavily loaded trucks and the pavement structure; rigid and flexible pavements’ responses (stresses, strains and deflections) were evaluated under simulated loading of 11-axle trucks (MI-20, MI-18, MI-14, MI-13) at gross vehicle weights (GVW) of 164,000 lb with different axle configurations, which are allowed in the state of Michigan. These responses were compared against results of a standard 5-axle semi-trailer at GVW nearing the common 80,000 lb standard in most states. Using these heavy truck types, the combined effects of truck loading, pavement thickness, joint spacing, material properties and thermal conditions on the pavement damages were evaluated. The major fatigue and faulting damages for rigid pavements as well as fatigue damage and subgrade rutting for flexible (asphalt) pavements are analysed. The finite element method (FEM) based program ISLAB2000 and multilayer elastic theory (MLET) based program JULEA were used to compute rigid and flexible pavements responses, respectively. Results show that the standard 5-axle truck yields up to 50% more fatigue damage potential under positive temperature gradient across slab (during daytime) even at a much lower gross vehicle weight than that of Michigan (MI) trucks. However, the 11-axle MI trucks provide up to 45% higher fatigue damage potential under negative temperature gradient (during nighttime). In comparison with the standard truck, the 11-axle MI trucks yield up to 17% less faulting damage for the typical rigid pavement with a thickness of 8 mm and up to 18% more faulting for larger pavement thickness. In flexible pavements, the standard truck exhibits up to 39% higher asphalt concrete fatigue damage than MI trucks. However, up to 80% more subgrade rutting damage for typical flexible pavements is observed for MI trucks with the multi-axles group configuration and larger GVW than the standard truck.

Fatigue resistance curves for single and double shear riveted joints from old portuguese metallic bridges

Abstract The maintenance and safety of ancient bridges is a major concern of governmental authorities. In particular, the safety of old riveted bridges fabricated and placed into service at the end of the 19th century deserves particular attention. These structures are susceptible to exhibit high fatigue damage levels due to their long operational period with increasing traffic intensity associated to an original design not covering the fatigue phenomenon. This paper reviews recent fatigue behaviour investigations on single and double shear riveted joints performed by Universities of Porto (Portugal), Tras-os-Montes e Alto Douro (Portugal), and Wroclaw (Poland), in particular concerning the fatigue characterization of riveted joints extracted from representative Portuguese riveted bridges, namely the Eiffel, Luiz I, Fao, Pinhao and Trezoi bridges. In order to overcome the influence of scatter and establish a reliable assessment for the obtained experimental data, two statistical approaches were used: implement linearized boundaries following the recommendation in ASTM E739 standard and defining probabilistic S N fields using the Castillo & Fernandez-Canteli model. This statistical analysis allows to propose design S N curves for single and double riveted joints and evaluate the applicability (safety) of using the design curves suggested in Eurocode 3 as well as design curves proposed by Taras and Greiner.

Artificial Neural Network Classification for Fatigue Feature Extraction Parameters Based on Road Surface Response

The aim of this paper is for classification for fatigue feature extraction parameters based on road surface response using the artificial neural network (ANN) technique. It is important for classification of the fatigue damage of automotive suspension as it is considers the random strain loading from the road surface contributed from complex variable amplitude loadings. In this study, the proposed method captured that strain signal collected from the car coil spring during the road test. Hence, the prediction of fatigue life need to assess based on actual loading to ensure the prediction results are accurate. The high amplitude segments were extracted from the strain signals using the discrete wavelet transform. This approach provides an advantage in assessing the signals containing random loads for both discontinuities and smooth components for the areas that contain high fatigue damage in the strain signal. From this, three significant fatigue features extraction parameters such as kurtosis, wavelet energy coefficients and fatigue damage were classified based on the similarities using ANN classification technique. This is important in analysing the clustering and classification were used in detection of fatigue damage according to the response of the various road surface condition. The results show the classifications using ANN give the accuracy based on the coefficient of determinant, R= 0.85 for all data. From on the accuracy of the ANN, it can be concluded that the discrete wavelet transform as a pre-processing method to extract the features from the signal for classification level of fatigue damage for the coil spring according to the response of loading based on road surface.