• Journal of Biomedical Engineering Research
• /
• v.42 no.4
• /
• pp.167-174
• /
• 2021

The purpose of this study is to classify TIFF images, PNG images, and JPEG images using deep learning, and to compare the accuracy by verifying the classification performance. The TIFF, PNG, and JPEG images converted from chest X-ray DICOM images were applied to five deep neural network models performed in image recognition and classification to compare classification performance. The data consisted of a total of 4,000 X-ray images, which were converted from DICOM images into 16-bit TIFF images and 8-bit PNG and JPEG images. The learning models are CNN models - VGG16, ResNet50, InceptionV3, DenseNet121, and EfficientNetB0. The accuracy of the five convolutional neural network models of TIFF images is 99.86%, 99.86%, 99.99%, 100%, and 99.89%. The accuracy of PNG images is 99.88%, 100%, 99.97%, 99.87%, and 100%. The accuracy of JPEG images is 100%, 100%, 99.96%, 99.89%, and 100%. Validation of classification performance using test data showed 100% in accuracy, precision, recall and F1 score. Our classification results show that when DICOM images are converted to TIFF, PNG, and JPEG images and learned through preprocessing, the learning works well in all formats. In medical imaging research using deep learning, the classification performance is not affected by converting DICOM images into any format.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.12
• /
• pp.1817-1825
• /
• 2021

Omnidirectional mobile robots can be mobile in any direction without changing the robot's direction, making them easy to apply in many applications and providing excellent maneuverability. Omnidirectional mobile robots have non-linear dynamic components such as friction, making them difficult to model accurately. In this paper, we linearize the mobile robot system using the mobile robot's inverse dynamics and integral sliding mode control method to remove these nonlinear components. And the position and velocity gains are optimized using a genetic algorithm to realize the optimal performance of the proposed system control method. As a result of the performance evaluation, the genetic algorithm's control method showed superior performance than the control method with an arbitrary gain. And the proposed inverse dynamic and integral sliding mode control method can be applied to other control methods. It can be beneficial for designing a linear control system.

• Nuclear Engineering and Technology
• /
• v.54 no.1
• /
• pp.36-48
• /
• 2022

As a valid numerical method to obtain a high-resolution result of a flow field, computational fluid dynamics (CFD) have been widely used to study coolant flow and heat transfer characteristics in fuel rod bundles. However, the time-consuming, iterative calculation of Navier-Stokes equations makes CFD unsuitable for the scenarios that require efficient simulation such as sensitivity analysis and uncertainty quantification. To solve this problem, a reduced-order model (ROM) based on proper orthogonal decomposition (POD) and machine learning (ML) is proposed to simulate the flow field efficiently. Firstly, a validated CFD model to output the flow field data set of the rod bundle is established. Secondly, based on the POD method, the modes and corresponding coefficients of the flow field were extracted. Then, an deep feed-forward neural network, due to its efficiency in approximating arbitrary functions and its ability to handle high-dimensional and strong nonlinear problems, is selected to build a model that maps the non-linear relationship between the mode coefficients and the boundary conditions. A trained surrogate model for modes coefficients prediction is obtained after a certain number of training iterations. Finally, the flow field is reconstructed by combining the product of the POD basis and coefficients. Based on the test dataset, an evaluation of the ROM is carried out. The evaluation results show that the proposed POD-ROM accurately describe the flow status of the fluid field in rod bundles with high resolution in only a few milliseconds.

• Nuclear Engineering and Technology
• /
• v.54 no.3
• /
• pp.860-866
• /
• 2022

An accurate prediction of the pressure drop along the flow paths is crucial in the design of advanced passive systems cooled by heavy liquid metal coolants. To date, a generic pressure drop correlation over spacer grids by Rehme has been applied extensively, which was obtained from substantial experimental data with multiple types of components. However, a few experimental studies have reported that the correlation may give large discrepancies. To provide a more reliable correlation for ring-type spacer grids, the current numerical study aims at figuring out the most critical factor among four hypothetical parameters, namely the flow area blockage ratio, number of fuel rods, type of fluid, and thickness of the spacer grid in the flow direction. Through a set of computational fluid dynamics simulations, we observed that the flow area blockage ratio dominantly influences the pressure loss characteristics, and thus its dependence should be more emphasized, whereas the other parameters have little impact. Hence, we suggest a new correlation for the drag coefficient as C_B = C_ν,m/ε^2.7, where C_ν,m is formulated by a nonlinear fit of simulation data such that C_ν,m = -11.33 ln(0.02 ln(Re_b)).

• Steel and Composite Structures
• /
• v.42 no.6
• /
• pp.779-789
• /
• 2022

This work presents a non-linear cylindrical bending analysis of functionally graded plate reinforced by single-walled carbon nanotubes (SWCNTs) in thermal environment using a simple integral higher-order shear deformation theory (HSDT). This theory does not require shear correction factors and the transverse shear stresses vary parabolically through the thickness. The material properties of SWCNTs are assumed to be temperature-dependent and are obtained from molecular dynamics simulations. The material properties of functionally graded carbon nanotube-reinforced composites (FG-CNTCRs) are considered to be graded in the thickness direction, and are estimated through a micromechanical model. The non-linear strain-displacement relations in the Von Karman sense are used to study the effect of geometric non-linearity and the solution is obtained by minimization of the total potential energy. The numerical illustrations concern the nonlinear bending response of FG-CNTRC plates under different sets of thermal environmental conditions, from which results for uniformly distributed CNTRC plates are obtained as benchmarks.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.11
• /
• pp.791-800
• /
• 2022

The number and locations of force generators and their force directions of Active Vibration Control System(AVCS) are important to maximize the airframe vibration reduction performance of helicopters. The present AVCS simulation using an exhaustive test method attempts to determine the best number and locations of force generators and their force directions for maximization of the airframe vibration reduction performance of UH-60A helicopter at 158 knots. The 4P hub vibratory loads of the UH-60A helicopter are calculated using DYMORE II, a nonlinear multibody dynamics analysis code, and MSC.NASTRAN is used to predict the vibration responses of the UH-60A airframe. The AVCS framework with an exhaustive test method is constructed using MATLAB Simulink. As a result, when applying AVCS with the optimal combination of the force generators, the 4P airframe vibration responses of UH-60A helicopter are reduced by from 19.35% to 98.07% compared to the baseline results without AVCS.

• Structural Engineering and Mechanics
• /
• v.88 no.4
• /
• pp.389-403
• /
• 2023

In order to explore the influence of tooth root cracks on the dynamic characteristics of multi-stage planetary gear transmission systems, a concentrated parameter method was used to construct a nonlinear dynamic model of the system with 30-DOF in bending and torsion, taking into account factors such as crack depth, length, angle, error, time-varying meshing stiffness (TVMS), and damping. In the model, the energy method was used to establish a TVMS model with cracks, and the influence of cracks on the TVMS of the system was studied. By using the Runge- Kutta method to calculate the differential equations of system dynamics, a series of system vibration diagrams containing cracks were obtained, and the influence of different crack parameters on the vibration of the system was analyzed. And vibration testing experiments were conducted on the system with planetary gear cracks. The results show that when the gear contains cracks, the TVMS of the system will decrease, and as the cracks intensify, the TVMS will decrease. When cracks appear on the II-stage planetary gear, the system will experience impact effects with intervals of rotation cycles of the II-stage planetary gear. There will be obvious sidebands near the meshing frequency doubling, and the vibration trajectory of the gear will also become disordered. These situations will become more and more obvious as the degree of cracks intensifies. Through experiments, the theoretical results are in good agreement with experimental results, verifying the correctness of the theoretical model. This provides a theoretical basis for fault diagnosis and reliability research of the system.

• Advances in nano research
• /
• v.16 no.4
• /
• pp.341-352
• /
• 2024

This study investigates the heat and mass transfer characteristics of a MoS₂ nanoparticle suspension in ethylene glycol over a porous stretching sheet. MoS₂ nanoparticles are known for their exceptional thermal and chemical stability which makes it convenient for enhancing the energy and mass transport properties of base fluids. Ethylene glycol, a common coolant in various industrial applications is utilized as the suspending medium due to its superior heat transfer properties. The effects of variable thermal conductivity, variable mass diffusivity, thermal radiation and thermophoresis which are crucial parameters in affecting the transport phenomena of nanofluids are taken into consideration. The governing partial differential equations representing the conservation of momentum, energy, and concentration are reduced to a set of nonlinear ordinary differential equations using appropriate similarity transformations. R software and MATLAB-bvp5c are used to compute the solutions. The impact of key parameters, including the nanoparticle volume fraction, magnetic field, Prandtl number, and thermophoresis parameter on the flow, heat and mass transfer rates is systematically examined. The study reveals that the presence of MoS₂ nanoparticles curbs the friction between the fluid and the solid boundary. Moreover, the variable thermal conductivity controls the rate of heat transfer and variable mass diffusivity regulates the rate of mass transfer. The numerical and statistical results computed are mutually justified via tables. The results obtained from this investigation provide valuable insights into the design and optimization of systems involving nanofluid-based heat and mass transfer processes, such as solar collectors, chemical reactors, and heat exchangers. Furthermore, the findings contribute to a deeper understanding of stretching sheet systems, such as in manufacturing processes involving continuous casting or polymer film production. The incorporation of MoS₂-C₂H₆O₂ nanofluids can potentially optimize temperature distribution and fluid dynamics.

• Korean Chemical Engineering Research
• /
• v.52 no.5
• /
• pp.581-586
• /
• 2014

Main objects for wastewater treatment operation are to maintain effluent water quality and minimize operation cost. However, the optimal operation is difficult because of the change of influent flow rate and concentrations, the nonlinear dynamics of microbiology growth rate and other environmental factors. Therefore, many wastewater treatment plants are operated for much more redundant oxygen or chemical dosing than the necessary. In this study, the optimal control scheme for dissolved oxygen (DO) is suggested to prevent over-aeration and the reduction of the electric cost in plant operation while maintaining the dissolved oxygen (DO) concentration for the metabolism of microorganisms in oxic reactor. The oxygen uptake rate (OUR) is real-time measured for the identification of influent characterization and the identification of microorganisms' oxygen requirement in oxic reactor. Optimal DO set-point needed for the micro-organism is suggested based on real-time measurement of oxygen uptake of micro-organism and the control of air blower. Therefore, both stable effluent quality and minimization of electric cost are satisfied with a suggested optimal set-point decision system by providing the necessary oxygen supply requirement to the micro-organisms coping with the variations of influent loading.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.6
• /
• pp.443-453
• /
• 2020

This study using the active vibration control technique attempts to alleviate numerically the airframe vibration of a UH-60A helicopter. The AVCS(Active Vibration Control System) is applied to reduce the 4/rev vibration responses at the specified locations of the UH-60A airframe. The 4/rev hub vibratory loads of the UH-60A rotor is predicted using the nonlinear flexible dynamics analysis code, DYMORE II. Various tools such as NDARC, MSC.NASTRAN, and MATLAB Simulink are used for the AVCS simulation with five CRFGs and seven accelerometers. At a flight speed of 158knots, the predicted 4/rev hub vibratory loads of UH-60A rotor excite the airframe, and then the 4/rev vibration responses at the specified airframe positions such as the pilot seat, rotor-fuselage joint, mid-cabin, and aft-cabin are calculated without and with AVCS. The 4/rev vibration responses at all the locations and directions are reduced by from 25.14 to 96.05% when AVCS is used, as compared to the baseline results without AVCS.