Title Participants Abstract "Smart S-N curve for fatigue lifetime predictions of offshore wind turbine support structures affected by corrosion" "Seyed Ahmad Elahi, Farid Mehri Sofiani, Somsubhro Chaudhuri, Wim De Waele" "The lifetime of steel support structures for offshore wind turbines is affected by fatigue and corrosion damage. Due to the time-variant uncertainties associated with environmental conditions and mechanical stresses, numerical models that can accurately predict corrosion and fatigue deterioration are needed for a reliable assessment of structural integrity. One of the main ambitions of the project MAXWind is to develop a numerical framework for corrosion-fatigue assessment, which consists of models for pitting corrosion, initiation and propagation of short cracks from pits, and long crack propagation. Integrating these models allows to develop so-called “smart” S-N curve. A microstructure-based model is used to simulate the initiation and propagation of short cracks from corrosion pits. Using data of pit shape evolution versus time of exposure to sea water as input to the short crack model, the degraded fatigue strength of S355 steel due to pitting corrosion representative for the North Sea is determined (Figure 1). Figure 2 illustrates the concept of the so-called smart S-N curve and virtual load spectra used to calculate fatigue damage accumulation. The black curve is the S-N curve in air, i.e., without exposure to sea water, of which the knee-point is arbitrarily taken at 10^7 cycles. The fatigue strength degradation shown in Figure 1 corresponds to the orange line (fatigue strength line) in Figure 2. The fatigue damage accumulation corresponding to the load spectrum for a specific year can be computed using Miner’s rule. The load bins which are below the fatigue strength line are assumed not to contribute to damage accumulation. Since pitting corrosion causes the fatigue strength line to have a decreasing behavior, more load bins with lower stress values will contribute to damage accumulation as time passes. Using representative load spectra based on monitoring campaigns and characterizing the time-dependent fatigue strength of the material, will allow to assess the lifetime of an OWT structure in the design stage. In future work, the corrosion fatigue model will be extended with algorithms for long crack growth simulation. This will allow to calculate the remaining useful life of OWT structures and can support decision-making activities, viz. definition of optimized inspection and maintenance plans, and finally have an impact on reducing energy production O&M expenses." "On the application of infrared thermography and potential drop for the accelerated determination of an S-N curve" "Nahuel Micone, Wim De Waele" "Comment on the paper entitled 'A new cumulative fatigue damage rule based on dynamic residual S-N curve and material memory concept' by Peng Z., Huang H., Zhou J. and Li Y. Published in Metals (2018; 8 (6): 456)" "Kris Hectors, Wim De Waele" "A simple finding on variable amplitude (Gassner) fatigue SN curves obtained using Miner's rule for unnotched or notched specimen" "Michele Ciavarella, Pietro D'Antuono, G. P. Demelio" "In this note, starting from the SN curve under Constant Amplitude (CA) for the fatigue life of the uncracked (plain) specimen, we obtain that Gassner curves for Variable Amplitude (VA) loading using the simple Palmgren-Miner's law are simply shifted CA curves. Further, using the Critical Distance Method in a very clean and powerful form proposed by Susmel and Taylor for VA loading, we find similar result for notched specimen, the spectrum loading results in the same multiplicative term for notched, cracked and unnotched specimen. Hence, the present proposal can be considered as a simple empirical unified approach for rapid assessment of the notch effect under random loading, which simplifies the recent proposal by Susmel and Taylor. To their extensive validations, we add some specific comparison with experimental data from the Literature on our further findings." "Generalized definition of “crack-like” notches to finite life and SN curve transition from “crack-like” to “blunt notch” behavior" "Michele Ciavarella, Pietro D'Antuono, G. P. Demelio" "We start by generalizing the concept of “crack-like” notches to finite life. We note that, according to the well established Critical Distance Method (originally proposed by Neuber and Peterson), SN curves of all notches are very close to SN curves for cracked specimen, up to a certain number of cycles N∗, above which they tend to the SN curve of the uncracked/unnotched (plain) specimen, reduced by the stress concentration factor Kt. We suggest therefore new simple estimates for rapid assessment of the notch effect under finite life: for the “crack-like” notch, substituting the notch with a crack leads to immediate use of crack solutions in databases, whereas the knowledge of the stress ahead of a notch is also well known analytically in many cases, or can be adapted. We show one example with the SAE keyhole tests, leading to fully analytical SN curves, in much better agreement with experiment than, for example, strain-life approaches using Neuber's rule." "Antimicrobial resistance in Salmonella enterica subspecies enterica serovar Dublin from dairy source calves in the central San Joaquin Valley, California (1998-2002)" "Anna Berge, Elizabeth Thornburg, John M Adaska, Robert B Moeller, Patricia C Blanchard" "A feasibility study of the Master SN curve approach for short fiber reinforced composites" "Ignace Verpoest, Stepan Lomov" "Short fiber reinforced composite (SFRC) materials have a different fiber orientation distribution (FOD) at every point. The fatigue properties of SFRC are known to depend on the FOD. The Master SN curve (MSNC) method for predicting an SN curve for a given FOD based on the known SN curve for the reference FOD is used to predict the local SN curve of a SFRC component by relating the damage at the microscopic level to the macroscopic fatigue properties. A simplified version of MSNC method, which needs even less experimental input, uses an assumption of constant SN curve slope is also presented in this paper. The paper validates both variants of the MSNC method on three sets of experimental data on fatigue of short fiber composites and analyses their accuracy. It is demonstrated that the MSNC approach needs only one SN curve as input with no specific requirements to the fiber orientation of the test coupon. Test coupons could have either uniform fiber orientation in the thickness or a “skin core” orientation variation." "The Master SN curve approach - A hybrid multi-scale fatigue simulation of short fiber reinforced composites" "Ignace Verpoest, Stepan Lomov" "Typical short fiber reinforced composites (SFRCs) components have a different statistical distribution of orientation of fibers at different points leading to different static and fatigue behavior at different locations across the component. To link component-scale calculations with this variability of fiber orientations, each element in the FE model is modeled as a Representative Volume Element (RVE); the static and fatigue properties must be calculated for each of these elements. While there are established methods to estimate the static properties, there are none for the fatigue properties. A hybrid (combination of micromechanics and tests) and multi-scale (damage in micro-scale linked to macroscale fatigue properties) method of predicting the SN curve for every point in a short fiber composite has been developed. This proposed method is based not only on tests but on a combination of manufacturing simulation, tests and multi-scale mechanics. An extensive test program was undertaken to study the fatigue behavior of short fiber composites and validate the concept of the Master SN curve (MSNC) approach. The MSNC approach is compared with two prevalent approaches – strength based scaling and test based interpolation. The MSNC approach was found to be in a good agreement with the experimental results and was confirmed to be more accurate than the prevalent methods." "Master SN curve approach : a hybrid multiscale approach to fatigue simulation of short fiber composites" "Atul Jain, Yasmine Abdin, Stefan Straesser, Wim Van Paepegem, Verpoest Ignaas, Stepan Lomov" "Typical short fiber reinforeced composites (SFRC) components have a different local statistical distribution of fiber orientation leading to different mechanical and fatigue behavior at different points. For a component-sized calculation, each element in the FE model is modelled as a Representative Volume Element (RVE). Unlike for static properties, there are no analytical methods to estimate the SN curve of a RVE. As a result one usually has to rely on expensive, time consuming and difficult to perform test-based interpolation methods. To overcome this problem we developed a hybrid multi-scale method to predict the local SN curves of SFRC; this proposed method involves both multiscale-mechanics and tests. The key assumption behind this scheme is that for SFRC the damage development and consequent stiffness degradation during cyclic loading with different orientation distributions are similar. Damage operator which represents the loss of stiffness due to different damage events is calculated for the first cycle of loading. Validated models were developed within the framework of the Mori-Tanaka formulation to model fiber matrix debonding and matrix non-linearity; the damage parameter is calculated based on loss of stiffness due to fiber matrix debonding. The proposed models are validated with experiments. Next for the validation of the assumptions, the loss of stiffness curves were extracted for each of the data points and a statistical comparison of the curves is performed for the data points which have similar lifetime to failure. A method to compare the loss of stiffness by characterizing the curves and then performing an ANOVA analysis using Fisher criteria is developed. It was proven that for similar life to failure, the variation in loss of stiffness curves for SFRC with different orientation is not statistically significant." "Master SN curve approach- A hybrid multiscale approach to fatigue simulation of short fiber composites" "Ignace Verpoest, Stepan Lomov" "© 2015 International Committee on Composite Materials. All rights reserved. Typical short fiber reinforeced composites (SFRC) components have a different local statistical distribution of fiber orientation leading to different mechanical and fatigue behavior at different points. For a component-sized calculation, each element in the FE model is modelled as a Representative Volume Element (RVE). Unlike for static properties, there are no analytical methods to estimate the SN curve of a RVE. As a result one usually has to rely on expensive, time consuming and difficult to perform test-based interpolation methods. To overcome this problem we developed a hybrid multi-scale method to predict the local SN curves of SFRC; this proposed method involves both multiscale-mechanics and tests. The key assumption behind this scheme is that for SFRC the damage development and consequent stiffness degradation during cyclic loading with different orientation distributions are similar. Damage operator which represents the loss of stiffness due to different damage events is calculated for the first cycle of loading. Validated models were developed within the framework of the Mori-Tanaka formulation to model fiber matrix debonding and matrix non-linearity; the damage parameter is calculated based on loss of stiffness due to fiber matrix debonding. The proposed models are validated with experiments. Next for the validation of the assumptions, the loss of stiffness curves were extracted for each of the data points and a statistical comparison of the curves is performed for the data points which have similar lifetime to failure. A method to compare the loss of stiffness by characterizing the curves and then performing an ANOVA analysis using Fisher criteria is developed. It was proven that for similar life to failure, the variation in loss of stiffness curves for SFRC with different orientation is not statistically significant."