• Smart Structures and Systems
• /
• v.33 no.1
• /
• pp.55-64
• /
• 2024

This paper proposes a fusion imaging-based coating-defect classification method for steel structures that uses zero-shot learning. In the proposed method, a halogen lamp generates heat energy on the coating surface of a steel structure, and the resulting heat responses are measured by an infrared (IR) camera, while photos of the coating surface are captured by a charge-coupled device (CCD) camera. The measured heat responses and visual images are then analyzed using zero-shot learning to classify the coating defects, and the estimated coating defects are visualized throughout the inspection surface of the steel structure. In contrast to older approaches to coating-defect classification that relied on visual inspection and were limited to surface defects, and older artificial neural network (ANN)-based methods that required large amounts of data for training and validation, the proposed method accurately classifies both internal and external defects and can classify coating defects for unobserved classes that are not included in the training. Additionally, the proposed model easily learns about additional classifying conditions, making it simple to add classes for problems of interest and field application. Based on the results of validation via field testing, the defect-type classification performance is improved 22.7% of accuracy by fusing visual and thermal imaging compared to using only a visual dataset. Furthermore, the classification accuracy of the proposed method on a test dataset with only trained classes is validated to be 100%. With word-embedding vectors for the labels of untrained classes, the classification accuracy of the proposed method is 86.4%.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.20 no.2
• /
• pp.181-190
• /
• 2007

The linkage framework of surface geometric modeling based on the NURBS and shell finite element analysis is developed in this study. In the geometrically exact shell finite element analysis, the accuracy of the analysis strongly depends upon the accurate computation of the surface geometric quantities. Therefore if we obtain the necessary geometric quantities from the NVRBS surface equation, it's possible to construct the effective linkage framework of surface modeling in the CAD systems and shell finite element analysis using geometrically exact shell finite element. Besides, the linkage framework can be applied to the analysis of general and complex surfaces as well as simple surfaces. In this study, the shell surfaces are generated by interpolating given set of data points based on the NURBS surfaces. These data points usually can be obtained from surface scanning. But the representations of the generated NURBS surface are not same to one another. The accuracy depends on the chosen parameterization methods used in NURBS. Therefore, it is needed to select the suitable parameterization method according to the geometry of the surfaces. To verify the performance and accuracy of our developed linkage framework, we solve several well-known benchmark problems and assess the performance of the developed method.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.1
• /
• pp.42-48
• /
• 2006

In this study, a repetitive response surface enhancement technique (RRSET) is proposed as a new system approximation method for the efficient multidisciplinary design and optimization (MDO). In order to represent the highly nonlinear behavior of the response with second order polynomials, RRSET introduces a design space transformation using stretching functions and repetitive response surface improvement. The tentative optimal point is repetitively included to the set of experimental points to better approximate the response surface of the system especially near the optimal point, hence a response surface with significantly improved accuracy can be generated with very small experimental points and system iterations. As a system optimizer, the simulated annealing, which generates a global design solution is utilized. The proposed technique is applied to several numerical examples, and demonstrates the validity and efficiency of the method. With its improved approximation accuracy, the RRSET can contribute to resolve large and complex system design problems under MDO environment.

• Journal of the Korean Society for Precision Engineering
• /
• v.8 no.4
• /
• pp.64-75
• /
• 1991

The design of structures of modern industry has developed to satisfy stringent service, realiability and safety. Up to now, geometric profile which means surface foughness and dimension accuracy is mainly treated in manufacturing process of work surface. But it is inevitable to evaluate changes of surface geometry as well as the nature of alterations in surface layers because surface of workpiece changes as a result of phase transformation, chemical changes, plastic deformation and stress changes. This paper is to present principal data for safety design by verifying the effect of grinding conditions and method in grinding layers and to explain the method of measuring surface integrity. In this paper, structural steel(SM20C) is used as a workpiece. Of integrity, surface roughness in view of surface texture is analyzed by frequency domain and residual stress, structures and defect of ground layers in view of surface metallurgy are investigated.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.20 no.43
• /
• pp.47-57
• /
• 1997

As a general index in surface quality of machined products, surface roughness is to measure worker's skill level, a ground product quality and machining accuracy, etc. The surface roughness is defined by a function of rotational speed and radius of a grinding wheel, distances of active grains composed of the wheel, and feed of a grinder's worktable. To predict surface roughness in horizontal surface grinding operations, probability distributions were used. Probability distribution functions(p.d.f.) of surface roughness were found as results when the size of active grains(=the radius of a grinding wheel) is given as uniform, exponential distribution, and the distance between active grains follows the distributions of uniform, exponential. For each pdf case, probabilistic features of surface roughness were also analyzed and presented. This study is a substantial step for determining mathematically the surface roughness instead of using empirical approaches. More works should be presented to develop a general model by which an accurate roughness value can be obtained in horizontal surface grinding operations.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.3
• /
• pp.83-90
• /
• 2004

This paper proposes a tool path generation method for machining impellers with 5-axis machining center. The shape of impeller is complex, being composed of pressure surface, suction surface and leading edge, and so on. The compound surface which is made of ruled surface such as pressure surface and suction surface and leading edge such as fillet surface, makes the tool path generation much complicated. To achieve efficient roughing, cutting area is divided into two region and then tool radius of maximum size that do not cause tool intereference is selected for shortening machining time. In finishing, accuracy is improved using side cutting for blade surface and point milling for leading edge.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.29 no.2
• /
• pp.123-130
• /
• 2016

With the development of fast construction mode in shipbuilding market, the demand on accuracy management of hull is becoming higher and higher in shipbuilding industry. In order to enhance production efficiency and reduce manufacturing cycle time in shipbuilding industry, it is important for shipyards to have the accuracy of ship components evaluated efficiently during the whole manufacturing cycle time. In accurate shipbuilding process, block accuracy is the key part, which has significant meaning in shortening the period of shipbuilding process, decreasing cost and improving the quality of ship. The key of block accuracy control is to create a integrate block accuracy controlling system, which makes great sense in implementing comprehensive accuracy controlling, increasing block accuracy, standardization of proceeding of accuracy controlling, realizing "zero-defect transferring" and advancing non-allowance shipbuilding. Generally, managers of accuracy control measure the vital points at section surface of block by using the heavy total station, which is inconvenient and time-consuming for measurement of vital points. In this paper, a new measurement method based on point clouds technique has been proposed. This method is to measure the 3D coordinates values of vital points at section surface of block by using 3D scanner, and then compare the measured point with design point based on ICP algorithm which has an allowable error check process that makes sure that whether or not the error between design point and measured point is within the margin of error.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.7 no.3
• /
• pp.81-86
• /
• 2008

I present a 3-D profiler specially devised for the profile measurement of specular surfaces that requires precision shape accuracy up to a few nanometer. A profile is reconstructed from the curvature of a test part of the surface at several locations along a line. The local curvature data are acquired with White-light Scanning Interferometry. Test measurement proves that the proposed profiler is well suited for the specular surface inspection like precision mirror.

• 한국전산유체공학회:학술대회논문집
• /
• 2009.11a
• /
• pp.228-230
• /
• 2009

The moment-of-fluid (MOF) method is a new volume-tracking method that accurately treats evolving material interfaces. Based on the moment data (volume and centroid) for each material, the material interfaces are reconstructed with second-order spatial accuracy in a strictly conservative manner. The MOF method is coupled with a stabilized finite element incompressible Navier-Stokes solver for two fluids, namely water and air. The effectiveness of the MOF method is demonstrated with a free-surface dam-break problem.

• Atmosphere
• /
• v.27 no.4
• /
• pp.385-398
• /
• 2017

Numerical weather prediction experiments were carried out by applying topographic effects to reduce or enhance the solar radiation by terrain. In this study, x and ${\kappa}({\phi}_o,\;{\theta}_o)$ are precalculated for topographic effect on high resolution numerical weather prediction (NWP) with 1 km spatial resolution, and meteorological variables are analyzed through the numerical experiments. For the numerical simulations, cases were selected in winter (CASE 1) and summer (CASE 2). In the CASE 2, topographic effect was observed on the southward surface to enhance the solar energy reaching the surface, and enhance surface temperature and temperature at 2 m. Especially, the surface temperature is changed sensitively due to the change of the solar energy on the surface, but the change of the precipitation is difficult to match of topographic effect. As a result of the verification using Korea Meteorological Administration (KMA) Automated Weather System (AWS) data on Seoul metropolitan area, the topographic effect is very weak in the winter case. In the CASE 1, the improvement of accuracy was numerically confirmed by decreasing the bias and RMSE (Root mean square error) of temperature at 2 m, wind speed at 10 m and relative humidity. However, the accuracy of rainfall prediction (Threat score (TS), BIAS, equitable threat score (ETS)) with topographic effect is decreased compared to without topographic effect. It is analyzed that the topographic effect improves the solar radiation on surface and affect the enhancements of surface temperature, 2 meter temperature, wind speed, and PBL height.