• Nuclear Engineering and Technology
• /
• v.55 no.6
• /
• pp.2096-2106
• /
• 2023

Rotating machinery is widely applied in important equipment of nuclear power plants (NPPs), such as pumps and valves. The research on intelligent fault diagnosis of rotating machinery is crucial to ensure the safe operation of related equipment in NPPs. However, in practical applications, data-driven fault diagnosis faces the problem of small and imbalanced samples, resulting in low model training efficiency and poor generalization performance. Therefore, a deep convolutional conditional generative adversarial network (DCCGAN) is constructed to mitigate the impact of imbalanced samples on fault diagnosis. First, a conditional generative adversarial model is designed based on convolutional neural networks to effectively augment imbalanced samples. The original sample features can be effectively extracted by the model based on conditional generative adversarial strategy and appropriate number of filters. In addition, high-quality generated samples are ensured through the visualization of model training process and samples features. Then, a deep convolutional neural network (DCNN) is designed to extract features of mixed samples and implement intelligent fault diagnosis. Finally, based on multi-fault experimental data of motor and bearing, the performance of DCCGAN model for data augmentation and intelligent fault diagnosis is verified. The proposed method effectively alleviates the problem of imbalanced samples, and shows its application value in intelligent fault diagnosis of actual NPPs.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.10
• /
• pp.4968-4986
• /
• 2017

In this paper, we primarily address the difficulty of automatic generation of a plausible depth map from a single image in an unstructured environment. The aim is to extrapolate a depth map with a more correct, rich, and distinct depth order, which is both quantitatively accurate as well as visually pleasing. Our technique, which is fundamentally based on a preexisting DepthTransfer algorithm, transfers depth information at the level of superpixels. This occurs within a framework that replaces a pixel basis with one of instance-based learning. A vital superpixels feature enhancing matching precision is posterior incorporation of predictive semantic labels into the depth extraction procedure. Finally, a modified Cross Bilateral Filter is leveraged to augment the final depth field. For training and evaluation, experiments were conducted using the Make3D Range Image Dataset and vividly demonstrate that this depth estimation method outperforms state-of-the-art methods for the correlation coefficient metric, mean log10 error and root mean squared error, and achieves comparable performance for the average relative error metric in both efficacy and computational efficiency. This approach can be utilized to automatically convert 2D images into stereo for 3D visualization, producing anaglyph images that are visually superior in realism and simultaneously more immersive.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.512-515
• /
• 2009

In this paper, we propose layout based multimodal contents authoring tool for Digilog Book. In authoring step, users create a virtual area using mouse or pen-type device and select property of the area repetitively. After finishing authoring step, system recognizes printed page number and generate page layout including areas and property information. Page layout is represented as a scene graph and stored as XML format. Digilog Book viewer loads stored page layout and analyze properties then augment virtual contents or execute functions based on area. Users can author visual and auditory contents easily by using hybrid interface. In AR environment, system provides area templates in order to help creating area. In addition, proposed authoring tool separates page recognition module from page tracking module. So, it is possible to author many pages using only single marker. As a result of experiment, we showed proposed authoring tool has reasonable performance time in AR environment. We expect that proposed authoring tool would be applicable to many fields such as education and publication.

• Journal of the Korean Society for Advanced Composite Structures
• /
• v.6 no.4
• /
• pp.8-15
• /
• 2015

The demand for environmental consideration is on the increase in civil engineering. This study focuses on the development of technology to reduce the use of carbonate cement and improve its performance by using a silicate mineral and hardening agents, and presents the test results for the demonstrative evaluation of the properties of the raw material. Highly active feldspar was used as a binder to augment the bonding of the carbonate cement, and their change in strength was observed after test piece construction with the addition of soluble hardening agent. The uniaxial compression strength of the test piece of the general Portland cement with the addition of 0.5% soluble hardening agent, showed an increase by 33% and that of the test piece of cement with the addition of 70% substituted with feldspar increased by 28%. The strength of viscous soil; classified as soft ground, showed an increase of a maximum of 1.7 times when it was mixed with cement and solidifier depending on the curing period. These tests confirmed that a soluble solidifier is effective for improving the strength of a cement binder and that the highly active feldspar can be used as a binder.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.8
• /
• pp.81-88
• /
• 2017

This paper proposes a method of detecting rain regions from a single image. More specifically, a way of training the deep convolutional neural network based on the collected rain and non-rain patches is presented in a supervised manner. It is also shown that the proposed rain detection method based on deep convolutional neural network can provide better performance than the conventional rain detection method based on dictionary learning. Moreover, it is confirmed that the application of the proposed rain detection for rain removal can lead to some improvement in detail representation on the low-frequency regions of the rain-removed images. Additionally, this paper introduces the rain transfer method that inserts rain patterns into original images, thereby producing rain effects on the resulting images. The proposed rain transfer method could be used to augment rain patterns while constructing rain database.

• 한국전산유체공학회:학술대회논문집
• /
• 2011.05a
• /
• pp.302-310
• /
• 2011

A fluid in an enclosure can be heated by electric heating, chemical reaction, or fission heat. In order to remove the volumetric heat of the fluid, the walls surrounding the enclosure must be cooled. In this case, a natural convection occurs in the pool of the fluid, and it has a dominant role in heat transfer to the surrounding walls. It can augment the heat transfer rates tens to hundreds times larger than conductive heat transfer. The heat transfer by a natural convection in a regular shape such as a square cavity or semi-circular pool has been studied experimentally and numerically for many years. A pool of an inverted triangular shape with 10 degree inclined bottom walls has a good cooling performance because of enhanced boiling critical heat flux (CHF) compared to horizontal downward surface. The coolability of the pool is determined by comparing the thermal load from the pool and the maximum heat flux removable by cooling mechanism such as radiative or boiling heat transfer on the pool boundaries. In order to evaluate the pool coolability, it is important to correctly expect the thermal load by a natural convection heat transfer of the pool. In this study, turbulence models with modifications for buoyancy effect were validated for unsteady natural convections by volumetric heating. And natural convection in the triangular pool was evaluated by using the models.

• The KSFM Journal of Fluid Machinery
• /
• v.13 no.5
• /
• pp.43-53
• /
• 2010

This study investigates a design optimization of a rotating two-pass rectangular cooling channel with staggered arrays of pin-fins. The radial basis neural network method is used as an optimization technique with Reynolds-averaged Navier-Stokes analysis of fluid flow and heat transfer with shear stress transport turbulent model. The ratio of the diameter to height of the pin-fins and the ratio of the streamwise spacing between the pin-fins to height of the pin-fin are selected as design variables. The optimization problem has been defined as a minimization of the objective function, which is defined as a linear combination of heat transfer related term and friction loss related term with a weighting factor. Results are presented for streamlines, velocity vector fields, and contours of Nusselt numbers, friction coefficients, and turbulent kinetic energy. These results show how fluid flow in a two-pass square cooling channel evolves a converted secondary flows due to Coriolis force, staggered arrays of pin-fins, and a $180^{\circ}$ turn region. These results describe how the fluid flow affects surface heat transfer. The Coriolis force induces heat transfer discrepancy between leading and trailing surfaces, having higher Nusselt number on the leading surface in the second pass while having lower Nusselt number on the trailing surface. Dean vortices generated in $180^{\circ}$ turn region augment heat transfer in the turning region and in the upstream region of the second pass. As the result of optimization, in comparison with the reference geometry, thermal performance of the optimum geometry shows the improvement by 30.5%. Through the optimization, the diameter of pin-fin increased by 14.9% and the streamwise distance between pin-fins increased by 32.1%. And, the value of objective function decreased by 18.1%.

• Journal of computational fluids engineering
• /
• v.16 no.3
• /
• pp.66-74
• /
• 2011

A fluid in an enclosure can be heated by electric heating, chemical reaction, or fission heat. In order to remove the volumetric heat of the fluid, the walls surrounding the enclosure must be cooled. In this case, a natural convection occurs in the pool of the fluid, and it has a dominant role in heat transfer to the surrounding walls. It can augment the heat transfer rates tens to hundreds times larger than conductive heat transfer. The heat transfer by a natural convection in a regular shape such as a square cavity or semi-circular pool has been studied experimentally and numerically for many years. A pool of an inverted triangular shape with 10 degree inclined bottom walls has a good cooling performance because of enhanced boiling critical heat flux (CHF) compared to horizontal downward surface. The coolability of the pool is determined by comparing the thermal load from the pool and the maximum heat flux removable by cooling mechanism such as radiative or boiling heat transfer on the pool boundaries. In order to evaluate the pool coolability, it is important to correctly expect the thermal load by a natural convection heat transfer of the pool. In this study, turbulence models with modifications for buoyancy effect were validated for unsteady natural convections by volumetric heating. And natural convection in the triangular pool was evaluated by using the models.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.8
• /
• pp.1619-1625
• /
• 2017

Civil global navigation satellite system (GNSS) does not meet user performance requirements for specific PNT (Positioning, Navigation, and Time) applications. Therefore, various differential systems are used to augment GNSS for improving positioning accuracy and integrity. The MTSAT satellite augmentation system (MSAS) is the Japanese satellite based augmentation system. This paper is for analyzing the characteristics of Japanese MSAS in Korean peninsula. First of all, it was done for analyzing not only DGNSS navigation signal but also the navigation parameter through simulation and experimental tests. As a result of data analyses, the sufficient navigation satellites to determine 3-D position based on DGNSS are simultaneously available at MSAS monitering station and the southern region of Korean peninsula. It was verified that the carrier to noise signals are stable to maintain the reliable 3-D position and that the level of 2m (2drms) accuracy is very similar to the ordinary worldwide DGNSS as well.

• Annals of Rehabilitation Medicine
• /
• v.42 no.6
• /
• pp.777-787
• /
• 2018

Objective To examine the long-term effects of the low-frequency repetitive transcranial magnetic stimulation (LF-rTMS) combined with task-specific training on paretic hand function following subacute stroke. Methods Sixteen participants were randomly selected and grouped into two: the experimental group (real LF-rTMS) and the control group (sham LF-rTMS). All the 16 participants were then taken through a 1-hour task-specific training of the paretic hand. The corticospinal excitability (motor evoke potential [MEP] amplitude) of the non-lesioned hemisphere, and the paretic hand performance (Wolf Motor Function Test total movement time [WMFT-TMT]) were evaluated at baseline, after the LF-rTMS, immediately after task-specific training, 1 and 2 weeks after the training. Results Groups comparisons showed a significant difference in the MEP after LF-rTMS and after the training. Compared to the baseline, the MEP of the experimental group significantly decreased after LF-rTMS and after the training and that effect was maintained for 2 weeks. Group comparisons showed significant difference in WMFT-TMT after the training. Only in the experimental group, the WMFT-TMT of the can lifting item significantly reduced compared to the baseline and the effect was sustained for 2 weeks. Conclusion The results of this study established that the improvement in paretic hand after task-specific training was enhanced by LF-rTMS and it persisted for at least 2 weeks.