• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.383-392
• /
• 2023

Post-earthquake building condition assessment is crucial for subsequent rescue and remediation and can be automated by emerging computer vision and deep learning technologies. This study is based on an endeavour for the 2nd International Competition of Structural Health Monitoring (IC-SHM 2021). The task package includes five image segmentation objectives - defects (crack/spall/rebar exposure), structural component, and damage state. The structural component and damage state tasks are identified as the priority that can form actionable decisions. A multi-task Convolutional Neural Network (CNN) is proposed to conduct the two major tasks simultaneously. The rest 3 sub-tasks (spall/crack/rebar exposure) were incorporated as auxiliary tasks. By synchronously learning defect information (spall/crack/rebar exposure), the multi-task CNN model outperforms the counterpart single-task models in recognizing structural components and estimating damage states. Particularly, the pixel-level damage state estimation witnesses a mIoU (mean intersection over union) improvement from 0.5855 to 0.6374. For the defect detection tasks, rebar exposure is omitted due to the extremely biased sample distribution. The segmentations of crack and spall are automated by single-task U-Net but with extra efforts to resample the provided data. The segmentation of small objects (spall and crack) benefits from the resampling method, with a substantial IoU increment of nearly 10%.

• Journal of Mechanical Science and Technology
• /
• v.16 no.7
• /
• pp.889-895
• /
• 2002

This paper presents the effect of external corrosion, material properties, operation condition and design thickness in pipeline on failure prediction using a failure probability model. The predicted failure assessment for the simulated corrosion defects discovered in corroded pipeline is compared with that determined by ANSI/ASME B31G code and a modified B31G method. The effects of environmental, operational, and random design variables such as defect depth, pipe diameter, defect length, fluid pressure, corrosion rate, material yield stress and pipe thickness on the failure probability are systematically studied using a failure probability model for the corrosion pipeline.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.13 no.5
• /
• pp.400-408
• /
• 2003

This paper investigates the vibration characteristics of steam generator (SG) U-tubes with defect. The operating SG shell-side flow field conditions for determining the fluidelastic instability parameters such as added mass are obtained from three-dimensional SG flow calculation. Modal analyses are performed for the U-tubes either with axial or circumferential flaw with different sizes. Special emphases are on the effects of flaw orientation and size on the modal and instability characteristics of tubes, which are expressed in terms of the natural frequency, corresponding mode shape and stability ratio. Also, addressed is the effect of the internal pressure on the vibration characteristics of the tube.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.20 no.41
• /
• pp.135-145
• /
• 1997

The statistical approach to quality control aims at alerting its user to any variations in the properties of a manufactured product. Motorola developed and pursued a quality management program called six sigma. The goal of six sigma programs is to improve customer satisfaction through reducing and eliminating defects. six sigma uses several statistical measure to characterize defect levels and process capabilities. The upper and lower specification limits are $\pm6\sigma$(sigma) from nominal, and the process mean is centered at nominal. only 0.002PPM are outside specification limits. Cp=2. this is the design target in a six sigma program. This article presents an important tool available for quality control of a production process at the occurrence of defects in manufactured products at view low levels to improve the efficiency of the manufacturing productivity and to satisfy customer through the reduction of defect rates. To understand the consequences of the level of quality on competitive position, a more technical perspective is needed.

• Journal of the Korean Society for Heat Treatment
• /
• v.12 no.1
• /
• pp.40-46
• /
• 1999

Rotary bending fatigue tests were carried out to investigate the improvement of fatigue limit in annealed and austempered spheroidal graphite cast iron. Main results obtained are as follows. (1) The tensile strength(hardness) of Series C is higher than that of Series B, and fatigue limits are 245MPa in Series C, 230MPa in Series B and 195MPa in Series A, respectively. (2) The fatigue limits of Series B and Series A are mainly governed by the resistance to fatigue crack initiation. Whereas, the fatigue limit of Series C is governed by the resistance to fatigue crack initiation and growth. The defect size and the resistance to crack initiation and growth should be considered to clarify the fatigue properties in spheroidal graphite cast iron. (3) Improvement of fatigue limit by half-austempering is more reasonable than that of full-austempering treatment in multi defective materials as spheroidal graphite cast iron.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.05a
• /
• pp.42-45
• /
• 2002

General wheel mark in mono-crystalline silicon wafer finding is able to be expected because it depends on radius ratio and angular velocity ratio of wafer and wheel. The pattern is predominantly determined by the contour of abrasive grits resulting from a relative motion. Although such a wheel mark is made uniform pattern if the process parameters are fixed, sub-surface defect is expected to be distributed non-uniformly because of characteristic of mono-crystalline silicon wafer that has diamond cubic crystal. Consequently it is considered that this phenomenon affects the following process. This paper focused on kinematic analysis of wafer grinding process and simulation program was developed to verify the effect of process variables on wheel mark. And finally, we were able to predict sub-surface defect distribution that considered characteristic of mono-crystalline silicon wafer

• Journal of the Korean Society for Precision Engineering
• /
• v.14 no.9
• /
• pp.30-36
• /
• 1997

In this study, the development of an aluminum forged piston was tried to substitute the cast piston, in which there were internal defects such as blow hole and shrink pipe. A gasoline engine piston was chosen as an example for developing the forged piston. Before aluminum forging, model, material (plasticine) test was carried out to investigate the forgeability and internal flow pattern of the forged piston at room temperature. From the result of model material test, an aluminum piston to be forged was redesigned. The aluminum pistion was forged in hot process. The quality of a forged piston was compared with that of a cast piston in the point of mechanical properties, internal defect and microstructure. It was proved that the forged piston was superior to the cast piston.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.1
• /
• pp.32-39
• /
• 2004

Traditional forging of a hollow T-shaped part has been applied to forge a solid T-shaped product from a solid billet and then to machine the hollow in that. In a case, a hollow T-shaped part can be forged by backward-extruding from a solid billet. In this study, four types of forging were suggested for manufacture of hollow T-shaped parts. Forging simulations for each of these forging methods were carried out to investigate folding defect, metal flow pattern, effective strain, and forging loads. Experimental works were carried out to be compared with the simulation results. Here, the ratio of the thickness of the hollow tube to that of the flange was selected to investigate a forging defect like folding.

• Journal of Power System Engineering
• /
• v.15 no.3
• /
• pp.18-24
• /
• 2011

In this study, the reflection of the L(0,2), axially symmetric guided elastic wave from defects in pipes above ground is examined using finite element method. Phase and group velocity dispersion curves for the pipe were presented for the selection of the excitation mode. Some simple signal processing was applied to determine the amplitude of each of the reflected waves and to calculate the reflection coefficient. The results show the reflection coefficient of this mode is very close to a linear function of the circumferential extent of the defect. The motivation for the work was the development of a technique for inspecting chemical plant pipelines, but the study addresses the nature of the reflection function and its general applicability.

• Journal of the Korean Society for Precision Engineering
• /
• v.24 no.12
• /
• pp.104-110
• /
• 2007

Internal defects are mainly caused by a corrosive action and degradation in the pipe used in a nuclear power plant or factory. The ESPI method have the many advantages when compared with conventional method. The advantage are the area measurement ability at one time and non-contact measurement ability in the real-time. In this paper, we studied on the measurement of a internal defect by using out of plane ESPI technique. Here, we compared the experimental results using out of plane ESPI with the FEM results.