• Journal of the Korean Society for Precision Engineering
• /
• v.32 no.5
• /
• pp.463-470
• /
• 2015

Solar cell is typical representative of renewable green energy. Silicon wafer contributes about 66 percent to its cost structure. In its manufacturing, micro-cracks are often occurred due to manufacturing process such as wire sawing, grinding and cleaning. Their detection and classification are important to process feedback information. In this paper, a classification method of micro-cracks is proposed, based on the fusion of principal component analysis(PCA) and neural network. The proposed method shows that it gives higher results than single application of two methods, in terms of shape and size classification of micro-cracks.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.20 no.4
• /
• pp.1309-1319
• /
• 1996

The research for the classification of the natural defects in welding zone is performd using the neuro-pattern recognition technology. The signal pattern recognition package including the user's defined function is developed to perform the digital signal processing, feature extraction, feature selection and classifier selection, The neural network classifier and the statistical classifiers such as the linear discriminant function classifier and the empirical Bayesian calssifier are compared and discussed. The neuro-pattern recognition technique is applied to the classificaiton of such natural defects as root crack, incomplete penetration, lack of fusion, slag inclusion, porosity, etc. If appropriately learned, the neural network classifier is concluded to be better than the statistical classifiers in the classification of the natural welding defects.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.27 no.5
• /
• pp.67-74
• /
• 2023

A large number of concrete structures are built and used around the world. To ensure their safe and continuous use, these structures require constant inspection and maintenance. While man-powered inspection and maintenance techniques are efficient, they can only provide intermittent status checks at the time of on-site inspection. Therefore, there is a growing need for a system that can continuously monitor the condition of the structure. A study was conducted to detect cracks and damage by installing a conductive coating on the surface of a concrete structure. A parallel resistance pattern that can monitor the occurrence and progression of cracks was developed by reflecting the structural characteristics of concrete structure. An empirical study was conducted to veryfy the application of the proposed method. The crack detection pattern was installed on the reinforced concrete beams, and the crack monitoring method was verified through applying a load on the beams.

• Geotechnical Engineering
• /
• v.14 no.2
• /
• pp.107-126
• /
• 1998

Ground settlements induced by tunnel excavation cause the foundations of the neighboring building structures to deform. An expert system called NESASS( Neural network Expert System for Adjacent Structure Safety analysis) was developed to analyze the structural safety of such building structures. NESASS predicts the trend of ground settlements resulting from tunnel excavation and carries out a safety analysis for building structures on the basis of the predicted ground settlements. Using neural network technique. the NESASS learns the database consisting of the measured ground settlements collected from numerous actual fields and infers a settlement trend at the field of interest. The NESASS calculates the magnitudes of angular distortion, deflection ratio, and differential settlement of the structure. and in turn, determines the safety of the structure. In addition, the NESASS predicts the patterns of cracks to be formed in the structure, using Dulacska model for crack evaluation. In this study, the ground settlements measured from Seoul subway construction sites were collected and classified with respect to the major factors influencing ground settlement. Subsequently, a database of ground settlement due to tunnel excavation was built. A parametric study was performed to select the optimal neural network model for the database. A comparison of the ground settlement predicted by the NESASS with the measured ones indicates that the NESASS leads to reasonable predictions. The results of confidence evaluation for safety evaluation system of the NESASS are presented in this paper.

• Journal of the Korean Ceramic Society
• /
• v.44 no.3 s.298
• /
• pp.162-168
• /
• 2007

Silica wet gels were prepared ken water glass ($29\;wt%\;SiO_{2}$) by using Amberlite as a ion exchange resin. After washing in distilled water, the wet gels were further aged in a solution of TEOS/EtOH to strengthen of 3-dimensional network structure. As increase TEOS content in aging solution, BET surface area and porosity of the ambient dried silica aerogels were significantly decreased, and average pore diameter was also decreased 30 nm to -10 nm. Also, higher density and compressive strength were obtained in case of higher TEOS content. This is due to precipitation of $SiO_{2}$ nano particles by TEOS. Hence, TEOS addition plays an important role of both strengthening and stiffness of silica wet gel network. By adding over 30 vol% TEOS, a crack-free monolithic silica aerogel tiles were obtained and its density, compressive strength, and thermal conductivity were shown $0.232g/cm^{3}$, 7.3 MPa, and 0.029 W/mk, respectivly.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.8
• /
• pp.1195-1200
• /
• 2013

In this paper, we proposed an enhanced Max-Min neural network by auto-tuning of learning rate using fuzzy control method. For the reduction of training time required in the competition stage, the method was proposed that arbitrates dynamically the learning rate by applying the numbers of the accuracy and the inaccuracy to the input of the fuzzy control system. The experiments using real concrete crack images showed that the enhanced Max-Min neural network was effective in the recognition of direction of the extracted cracks.

• Smart Structures and Systems
• /
• v.20 no.6
• /
• pp.669-682
• /
• 2017

The normalised version of bispectrum, the so-called bicoherence, has often proved a reliable method of damage detection on engineering applications. Indeed, higher-order spectral analysis (HOSA) has the advantage of being able to detect non-linearity in the structural dynamic response while being insensitive to ambient vibrations. Skewness in the response may be easily spotted and related to damage conditions, as the majority of common faults and cracks shows bilinear effects. The present study tries to extend the application of HOSA to damage localisation, resorting to a neural network based classification algorithm. In order to validate the approach, a non-linear finite element model of a 4-meters-long cantilever beam has been built. This model could be seen as a first generic concept of more complex structural systems, such as aircraft wings, wind turbine blades, etc. The main aim of the study is to train a Neural Network (NN) able to classify different damage locations, when fed with bispectra. These are computed using the dynamic response of the FE nonlinear model to random noise excitation.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2003.10a
• /
• pp.111-114
• /
• 2003

Thermosets are highly cross-linked polymers with a three-dimensional molecular structure. The network structure gives rise to mechanical properties, however, one major drawback of thermosets, which also results from their network structure, is their poor resistance to impact and to crack initiation. In this study, to solve this problem, the reactive thermoplastics such as amine terminated polyetherimide (ATPEI), ATPEI-CTBN-ATPEI(ABA) triblock copolymer, CTBN-ATPEI(AB) diblock copolymer, and carboxyl group modified ATPEI was synthesized, after that blended with epoxy resin, and the carbon fiber reinforced composites were fabricated. The impact load, energy, and delamination were investigated by using drop weight impact test and C-scan test. As a results, the ABA/epoxy blend system showed good impact properties.

• Computers and Concrete
• /
• v.9 no.3
• /
• pp.215-226
• /
• 2012

This article employs Support Vector Machine (SVM) for determination of fracture parameters critical stress intensity factor ($K^s_{Ic}$) and the critical crack tip opening displacement ($CTOD_c$) of concrete. SVM that is firmly based on the theory of statistical learning theory, uses regression technique by introducing ${\varepsilon}$-insensitive loss function has been adopted. The results are compared with a widely used Artificial Neural Network (ANN) model. Equations have been also developed for prediction of $K^s_{Ic}$ and $CTOD_c$. A sensitivity analysis has been also performed to investigate the importance of the input parameters. The results of this study show that the developed SVM is a robust model for determination of $K^s_{Ic}$ and $CTOD_c$ of concrete.

• International Journal of Concrete Structures and Materials
• /
• v.5 no.1
• /
• pp.29-33
• /
• 2011

This article employs Least Square Support Vector Machine (LSSVM) for determination of fracture parameters of concrete: critical stress intensity factor ($K_{Ic}^s$) and the critical crack tip opening displacement ($CTOD_c$). LSSVM that is firmly based on the theory of statistical learning theory uses regression technique. The results are compared with a widely used Artificial Neural Network (ANN) Models of LSSVM have been developed for prediction of $K_{Ic}^s$ and $CTOD_c$, and then a sensitivity analysis has been performed to investigate the importance of the input parameters. Equations have been also developed for determination of $K_{Ic}^s$ and $CTOD_c$. The developed LSSVM also gives error bar. The results show that the developed model of LSSVM is very predictable in order to determine fracture parameters of concrete.