• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.359-366
• /
• 2021

A number of construction companies in Korea invest considerable human and financial resources to construct a system for managing apartment defect data and for categorizing repair tasks. Thus, this study proposes machine learning models to automatically classify defect complaint text-data into one of the sub categories of 'finishing work' (i.e., one of the defect repair tasks). In the proposed models, we employed two word representation methods (Bag-of-words, Term Frequency-Inverse Document Frequency (TF-IDF)) and two machine learning classifiers (Support Vector Machine, Random Forest). In particular, we conducted both binary- and multi- classification tasks to classify 9 sub categories of finishing work: home appliance installation work, paperwork, painting work, plastering work, interior masonry work, plaster finishing work, indoor furniture installation work, kitchen facility installation work, and tiling work. The machine learning classifiers using the TF-IDF representation method and Random Forest classification achieved more than 90% accuracy, precision, recall, and F1 score. We shed light on the possibility of constructing automated defect classification systems based on the proposed machine learning models.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1940-1944
• /
• 2004

• Journal of Applied Reliability
• /
• v.14 no.3
• /
• pp.176-181
• /
• 2014

In these days, new and renewable energy is getting popular around globe and wind power generator is one of the renewable energy. In this study, we conducted a study on defect detection of composite material blade for wind power generator by applying active infrared thermography and produced a defect test piece by applying composite material used for blade of wind power generator. An infrared thermal camera and 2 kW halogen lamp are used for the purpose of research as equipments. Also, we analyzed temperature characteristic by using infrared thermal camera after checking a heat source on a test piece and found effectiveness of infrared thermography to blade of wind power generator by detecting defects resulting from temperature difference of a test piece, which eventually improve the safety and reliability of the composite material blade.

• Composites Research
• /
• v.24 no.2
• /
• pp.9-13
• /
• 2011

,In this paper, the symmetric axis parameter method, which was proposed to identify defects, dislocations and stacking fault, with perfect structures in the zinc-blende materials, was introduced as a way to distinguish between elastic and plastic deformation. LAMMPS, a molecular dynamics programme of Sandia National Laboratories, was used to perform nanoindentation simulation on silicon, a zinc-blende material. Defects in silicon (111) under spherical indentation showed the threefold pattern and the slip system in the form of ring crack. Also simulation results show good agreement with experimental results and existing theoretical analyses.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.25 no.1
• /
• pp.48-53
• /
• 2016

An EMAT can be used to reliably detect defects as it serves as a non-contact transducer with the ability to transmit ultrasonic waves into specimens without couplant. Moreover, an EMAT can easily generate desired waves by altering the design of the coil and magnet. This study proposes an SH-EMAT to evaluate the integrity of the TIG welding part. A stainless steel was welded using the TIG welding method. The welding current was varied to create artificial defects. Both the PA-UT and the RT were applied to verify the defect size. The experimental results generated by using the EMAT were compared with those methods. The amplitude was observed to decrease with an increase in the defect size. These results confirmed that the presence of defects can be reliably detected by attenuation of signal amplitude. The results demonstrated that the proposed method is suitable for evaluating the integrity of TIG welding.

• Tribology and Lubricants
• /
• v.33 no.4
• /
• pp.141-147
• /
• 2017

The vibration data of bearings are very useful for monitoring and determining the condition of the bearings. The defect frequencies of ball bearings have been used for monitoring there condition. However, it is not easy to verify the defect frequencies as the wear progress. Therefore there is a need for an easy method to monitor the damages of bearings in real-time and to observe the variations in vibration characteristics as the wear progress. In this study, a bearing test equipment is constructed to diagnose the damage of bearings. The friction coefficient and vibration data are measured by using a torque sensor and an acceleration sensor, and the correlation between the measured data is analyzed to diagnose the condition of the bearing. We reached the following conclusions from the results. When the ball surface, inner and outer rings of a ball bearing are damaged, the friction coefficient increases to over 0.02 with an adhesion on the surface. Moreover this damage occurs more quickly with an increase in the number of revolutions. In the vibration characteristics, the amplitude of vibration wave appears high with an increase in the friction coefficient. In the high frequency range between 1000 and 2000 Hz, a wide range of frequency components with high amplitude occurs continuously irrespective of the number of revolutions.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.10
• /
• pp.47-55
• /
• 2015

To verify the function of vehicle is becoming more and more difficult because many electronic control units have been embedded in vehicle with development of electronics industry. The reliability of vehicle should be considered above all important because malfunction of vehicle can cause damage of human life. In this paper, defect diagnosis platform based on CAN network is proposed to improve the reliability of vehicle. Reliability of vehicle is significantly increased by adopting the structural test via dedicated test path after manufacturing. Besides, the test cost is reduced because additional test pins are not required.

• Corrosion Science and Technology
• /
• v.12 no.6
• /
• pp.288-294
• /
• 2013

When the automotive body enters an electrocoating tank while applying an electric current, its steel surface is exposed to a very low induced current. Consequently, surface defects of coating may arise if the steel surface has lack of electric uniformity due to local defects such as local oxide. In this study, we investigated the preceding assessment methods to evaluate steel susceptibility of the low induced current during electrocoating before mass production. Prior to general electrocoating, we applied low constant voltage such as 3V or low constant current densities such as $0.35mA/cm^2$ and $0.50mA/cm^2$. In result, we confirmed that such methods were efficient for assessing steel susceptibility of low induce current during electrocoating.

• Korean Journal of Veterinary Research
• /
• v.50 no.3
• /
• pp.171-177
• /
• 2010

The purpose of this study was to evaluate the osteogenic capacity of water extract of danshen (Salvia miltiorrhiza Bunge). We have established in rat critical-sized calvarial defect model using the combination with collagen scaffold and danshen hydrophilic extract. All rats were extinguished at 8 weeks after bone graft surgery, and the bone regeneration ability of bone grafting sides was evaluated by plain radiography and micro-CT. These results revealed water extract of danshen had the potential to promote osteogenesis especially continuous oral administration with local treatment compared to one-shot local treatment. This compound may provide a new alternative agent for growth factors to promote bone healing and bone regeneration. In conclusion, these results suggest that danshen hydrophilic extract have the potential to promote osteogenesis in bone defects. Further studies about fusion technology with salvianolic acid B, peptides, growth factors, and scaffolds using of the combination of tissue engineering, cell engineering and mechanical engineering are needed.

• Smart Structures and Systems
• /
• v.13 no.6
• /
• pp.901-925
• /
• 2014

The reliability of a Bridge management System depends on the quality of visual inspection and the reliable estimation of bridge condition rating. However, the current practices of visual inspection have been identified with several limitations, such as: they are time-consuming, provide incomplete information, and their reliance on inspectors' experience. To overcome such limitations, this paper presents an approach of automating the prediction of condition rating for bridges based on digital image analysis. The proposed methodology encompasses image acquisition, development of 3D visualization model, image processing, and condition rating model. Under this method, scaling defect in concrete bridge components is considered as a candidate defect and the guidelines in the Ontario Structure Inspection Manual (OSIM) have been adopted for developing and testing the proposed method. The automated algorithms for scaling depth prediction and mapping of condition ratings are based on training of back propagation neural networks. The result of developed models showed better prediction capability of condition rating over the existing methods such as, Naïve Bayes Classifiers and Bagged Decision Tree.