• Smart Structures and Systems
• /
• v.31 no.5
• /
• pp.485-500
• /
• 2023

Machine learning-based structural health monitoring (ML-based SHM) methods are researched extensively in the recent decade due to the availability of advanced information and sensing technology. ML methods are well-known for their pattern recognition capability for complex problems. However, the main obstacle of ML-based SHM is that it often requires pre-collected historical data for model training. In most actual scenarios, damage presence can be detected using the unsupervised learning method through anomaly detection, but to further identify the damage types would require prior knowledge or historical events as references. This creates the cold-start problem, especially for new and unobserved structures. Modal-based methods identify damages based on the changes in the structural global properties but often require dense measurements for accurate results. Therefore, a two-stage hybrid modal-machine learning damage detection scheme is proposed. The first stage detects damage presence using Principal Component Analysis-Frequency Response Function (PCA-FRF) in an unsupervised manner, whereas the second stage further identifies the damage. To solve the cold-start problem, mode shape assessment using the first mode is initiated when no trained model is available yet in the second stage. The damage identified by the modal-based method would be stored for future training. This work highlights the performance of the scheme in alleviating the cold-start issue as it transitions through different phases, starting from zero damage sample available. Results showed that single and multiple damages can be identified at an acceptable accuracy level even when training samples are limited.

• Computers and Concrete
• /
• v.1 no.2
• /
• pp.131-150
• /
• 2004

We present in this work a coupled phenomenological chemo-mechanical model that represents the degradation of concrete-like materials. The chemical behaviour is described by the nowadays well known simplified calcium leaching approach. And the mechanical damage behaviour is described by a continuum damage model which involves the gradient of the damage quantity. The coupled nonlinear problem at hand is addressed within the context of the finite element method. For the equation governing the calcium dissolution-diffusion part of the problem, special care is taken to treat the highly nonlinear calcium conductivity and solid calcium functions. The algorithmic design is based on a Newton-type iterative scheme where use is made of a recently proposed relaxed linearization procedure. And for the equation governing the damage part of the problem, an augmented Lagrangian formulation is used to take into account the damage irreversibility constraint. Finally, numerical simulations are compared with experimental results on cement paste.

• Interaction and multiscale mechanics
• /
• v.1 no.2
• /
• pp.279-301
• /
• 2008

This paper develops a 3D homogenization based continuum damage mechanics (HCDM) model for fiber reinforced composites undergoing micromechanical damage under monotonic and cyclic loading. Micromechanical damage in a representative volume element (RVE) of the material occurs by fiber-matrix interfacial debonding, which is incorporated in the model through a hysteretic bilinear cohesive zone model. The proposed model expresses a damage evolution surface in the strain space in the principal damage coordinate system or PDCS. PDCS enables the model to account for the effect of non-proportional load history. The loading/unloading criterion during cyclic loading is based on the scalar product of the strain increment and the normal to the damage surface in strain space. The material constitutive law involves a fourth order orthotropic tensor with stiffness characterized as a macroscopic internal variable. Three dimensional damage in composites is accounted for through functional forms of the fourth order damage tensor in terms of components of macroscopic strain and elastic stiffness tensors. The HCDM model parameters are calibrated from homogenization of micromechanical solutions of the RVE for a few representative strain histories. The proposed model is validated by comparing results of the HCDM model with pure micromechanical analysis results followed by homogenization. Finally, the potential of HCDM model as a design tool is demonstrated through macro-micro analysis of monotonic and cyclic damage progression in composite structures.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.2
• /
• pp.387-398
• /
• 2002

The impact response and damage of CLAS panel was investigated experimentally. The facesheet material used was RO4003 woven-glass hydrocarbon/ceramic and the core material was Nomex honeycomb with a cell size of 3.2mm and a density of 96 kg/㎥. The shield plane used was RO4003 and 2024-T3 aluminum. Static indentation and impact test was conducted to characterize the type and extent of the damage observed in two CLAS panels, and the performance of antenna used in a wireless LAN system. Correlation of peak contact force, residual indentation and the delamination area shows impact damage of the panel with an aluminum shield plane is larger than that of the panel with RO4003 shield plane, although the former is more penetration resistant. The damage was observed by naked eye, ultrasonic inspection and cross sectioning. The shape and size of delamination was estimated by ultrasonic inspection, and the area of delamination linearly increases as impact energy increases. The performance of impact damaged antenna was estimated by measuring return loss and radiation pattern. It was revealed that the performance of antenna was related to the impact damage and there was a threshold that the performance of antenna fell as impact energy level changed. The threshold was between the impact energies of 1.5J and 1.75J.

• Smart Structures and Systems
• /
• v.6 no.7
• /
• pp.835-849
• /
• 2010

This study investigated propagation characteristics of piezo-activated guided waves in an aluminium rectangular-section tube for the purpose of damage identification. Changes in propagating velocity and amplitude of the first wave packet in acquired signals were observed in the frequency range from 50 to 250 kHz. The difference in guided wave propagation between rectangular- and circular-section tubes was examined using finite element simulation, demonstrating a great challenge in interpretation of guided wave signals in rectangular-section tubes. An active sensor network, consisting of nine PZT elements bonded on different surfaces of the tube, was configured to collect the wave signals scattered from through-thickness holes of different diameters. It was found that guided waves were capable of propagating across the sharp tube curvatures while retaining sensitivity to damage, even that not located on the surfaces where actuators/sensors were attached. Signal correlation between the intact and damaged structures was evaluated with the assistance of a concept of digital damage fingerprints (DDFs). The probability of the presence of damage on the unfolded tube surface was thus obtained, by which means the position of damage was identified with good accuracy.

• Smart Structures and Systems
• /
• v.17 no.2
• /
• pp.167-194
• /
• 2016

This paper investigates the Synergetics based Damage Detection Method (SDDM) for frame structures by using surface-bonded PZT (Lead Zirconate Titanate) patches. After analyzing the mechanism of pattern recognition from Synergetics, the operating framework with cooperation-competition-update process of SDDM was proposed. First, the dynamic identification equation of structural conditions was established and the adjoint vector (AV) set of original vector (OV) set was obtained by Generalized Inverse Matrix (GIM).Then, the order parameter equation and its evolution process were deduced through the strict mathematics ratiocination. Moreover, in order to complete online structural condition update feature, the iterative update algorithm was presented. Subsequently, the pathway in which SDDM was realized through the modified Synergetic Neural Network (SNN) was introduced and its assessment indices were confirmed. Finally, the experimental platform with a two-story frame structure was set up. The performances of the proposed methodology were tested for damage identifications by loosening various screw nuts group scenarios. The experiments were conducted in different damage degrees, the disturbance environment and the noisy environment, respectively. The results show the feasibility of SDDM using piezoceramic sensors and actuators, and demonstrate a strong ability of anti-disturbance and anti-noise in frame structure applications. This proposed approach can be extended to the similar structures for damage identification.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.4
• /
• pp.787-793
• /
• 2002

In this study, a bar impact test of low velocity was carried out to gain an insight into the damage mechanism and sequence induced in alumina plates(AD 85 and AD 90) under impact conditions. An experimental setup utilizing an instrumented long bar impact was devised, that can measure directly the impact force applied to the specimen and supply a compressive contact pressure to the specimen. During the bar impact testing, the influences of the contact pressure applied along the impact direction to the specimen on the fracture behavior were investigated. The measured impact force profiles explained well the damage behavior induced in alumina plates. The higher contact pressure to the specimen led to the less damage due to the suppression of radial cracks due to the increase in the apparent flexural stiffness of plate. It had produced the change of damage pattern developed in the specimen; from the radial cracks to the local contact stress dominant damage. It would contribute to the improvement of the ballistic property in ceramic plates. The observed results showed the following sequence in damage developed: The development of cone crack at impact region, the formation of radial cracks from the rear surface of plate depending on the plate thickness, the occurrence of crushing within the cone envelope and the fragmentation.

• Fashion & Textile Research Journal
• /
• v.8 no.4
• /
• pp.441-448
• /
• 2006

This research is the hair damage as treating a permanent wave before and after that is compared and analyzed the change of physical and mechanical properties. This is the survey of women's hair in 20 years old. On the basis of this we would like to analyze a extend of hair damage. Also, we would to show a basic data for hair damage prevention and hair improvement to keep the beautiful and healthy hair. The conclusion is as follow. : The swelling degree after the treatment was found to be greater than before permanent wave treatment. For the formational characteristics wave, untreated hair certainly had more elastic S curl wave than damaged hair in all the permanent wave treatments, and damaged hair and extremely damaged hair had less elasticity and had saggy S curl wave. The protein permanent and soft permanent wave had thicker, gorgeous, and better elastic wave than the regular permanent wave and direct heating permanent wave in all the hair condition. As the degree of damage on hair got greater, the tensile strength dramatically decreased and as the degree of damage got greater, the elongation was great as well. For treatment method, direct heating permanent wave showed the greatest effect, causing the most damage.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.20 no.4
• /
• pp.1241-1250
• /
• 1996

The characteristics of fatigue cumulative damage and fatigue life of 8-harness satin woven CFRP composites with a circular hole under constant amplitude and 2-level block loading are estimated by Stochastic Makov chain model. It is found in this study that the fatigue damage accumulation behavior is very random and the fatigue damage is accumulated as two regions under constant amplitude fatigue loading. In constant amplitude fatigue loading the predicted mean number of cycles to a specified damage state by Markov chain model shows a good agreement with the test result. The predicted distribution of the fatigue cumulative damage by Markov chain model is similar to the test result. The fatigue life predictions under 2-level block loading by Markov chain model revised are good fitted to the test result more than by 2-parameter Weibull distribution function using percent failure rule.

• Structural Engineering and Mechanics
• /
• v.70 no.1
• /
• pp.33-41
• /
• 2019

The influence of two parameters on fatigue damage predictions of a variably loaded cantilever beam has been examined. The first parameter is the geometry of the cantilever beam and the weld connecting it to a rear panel. Variables of the geometry examined here include the cantilever length, the weld width on the critical cross-section and the angle of the critical cross-section. The second parameter is the safety factor, as set out by the Eurocode 3 standard. An analytical approach has been used to calculate the stresses at the critical cross-section and standard rainflow counting has been used for the extraction of the load cycles from the load history. The results here suggest that a change in the width and angle of the critical cross-section has a non-linear impact on the fatigue damage. The results also show that the angle of the critical cross-section has the biggest influence on the fatigue damage and can cause the weld to withstand fatigue better. The second parameter, the safety factor, is shown to have a significant effect on the fatigue damage calculation, whereby a slight increase in the endurance safety factor can cause the calculated fatigue damage to increase considerably.