• Nuclear Engineering and Technology
• /
• v.55 no.1
• /
• pp.339-352
• /
• 2023

Prediction of the time-related traits of pebble flow inside pebble-bed HTGRs is of great significance for reactor operation and design. In this work, an image-driven approach with the aid of a convolutional neural network (CNN) is proposed to predict the remaining time of initially loaded pebbles and the time interval of paired flow images of the pebble bed. Two types of strategies are put forward: one is adding FC layers to the classic classification CNN models and using regression training, and the other is CNN-based deep expectation (DEX) by regarding the time prediction as a deep classification task followed by softmax expected value refinements. The current dataset is obtained from the discrete element method (DEM) simulations. Results show that the CNN-aided models generally make satisfactory predictions on the remaining time with the determination coefficient larger than 0.99. Among these models, the VGG19+DEX performs the best and its CumScore (proportion of test set with prediction error within 0.5s) can reach 0.939. Besides, the remaining time of additional test sets and new cases can also be well predicted, indicating good generalization ability of the model. In the task of predicting the time interval of image pairs, the VGG19+DEX model has also generated satisfactory results. Particularly, the trained model, with promising generalization ability, has demonstrated great potential in accurately and instantaneously predicting the traits of interest, without the need for additional computational intensive DEM simulations. Nevertheless, the issues of data diversity and model optimization need to be improved to achieve the full potential of the CNN-aided prediction tool.

• Proceedings of the Korean Society for Rock Mechanics Conference
• /
• 2000.09a
• /
• pp.117-128
• /
• 2000

The development of proper joint model, which can describe real phenomena exactly and still can be used easily, is one of the most important element for the analysis of the mechanical and hydraulic behavior of discontinuous rock mass. In this study, an elasto-plastic constitutive model of joint behavior considering asperity degradation was extended with the concept of first and second order asperities. The proposed model was implemented to numerical code with discrete finite joint element. The parametric study with the various asperity angles and degradation coefficients showed that the model can reproduce the shear behavior of typical rough joints well. Results of laboratory monotonic and cyclic shear tests were compared with those of numerical tests to validate the model. The hydraulic model considering the relations between gouge production and aperture was introduced to the mechanical model. In an attempt to examine the performance of the model, comparative numerical test was conducted. Permeability between joint surfaces increased rapidly at the first stage, but became nearly constant with increasing shear displacement due to gouge production and uniform variation of aperture distribution.

• Tunnel and Underground Space
• /
• v.10 no.3
• /
• pp.366-377
• /
• 2000

The development of proper joint model, which can describe real phenomena exactly and still can be used easily, is one of the most important element for the analysis of the mechanical and hydraulic behavior of discontinuous rock mass. In this study, an elasto-plastic constitutive model of joint behavior considering asperity degradation was extended with the concept of first and second order asperities. The proposed model was implemented to numerical code with discrete finite joint element. The parametric study with the various asperity angles and degradation coefficients showed that the model can reproduce the shear behavior of typical rough joints well. Results of laboratory monotonic and cyclic shear tests were compared with those of numerical tests to validate the model. The hydraulic model considering the relations between gouge production and aperture was introduced to the mechanical mode1. In an attempt to examine the performance of the model, comparative numerical test was conducted. Permeability between joint surfaces increased rapidly at the first stage, but became nearly constant with increasing shear displacement due to gouge production and uniform variation of aperture distribution.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.203-211
• /
• 2016

The genetic code is a key to bio-informatics and to a science of biological self-organizing on the whole. Modern science faces the necessity of understanding and systematically explaining mysterious features of ensembles of molecular structures of the genetic code. This paper is devoted to symmetrical analysis for genetic systems. Mathematical theories of noise-immunity coding and discrete signal processing are based on Jacket matrix methods of representation and analysis of information. Both of the RNA and Jacket Matrix property also have the Element(Block) - wise Inverse Matrices. These matrix methods, which are connected closely with relations of symmetry, are borrowed for a matrix analysis of ensembles of molecular elements of the genetic code. This method is presented for its simplicity and the clarity with which it decomposes a Jacket Matrix in terms of the genetic RNA Codon.

• Steel and Composite Structures
• /
• v.42 no.1
• /
• pp.1-22
• /
• 2022

In cold-formed steel (CFS) structures, such as trusses, transmission towers and portal frames, the use of back-to-back built-up CFS unequal angle sections are becoming increasingly popular. In such an arrangement, intermediate welds or screw fasteners are required at discrete points along the length, preventing the angle sections from buckling independently. Limited research is available in the literature on axial strength of back-to-back built-up CFS unequal angle sections. The issue is addressed herein. This paper presents an experimental investigation reported by the authors on back-to-back built-up CFS unequal angle sections with intermediate stiffeners under axial compression. The load-axial shortening behaviour along with the deformed shapes at failure are reported. A nonlinear finite element (FE) model was then developed, which includes material non-linearity, geometric imperfections and modelling of intermediate fasteners. The FE model was validated against the experimental test results, which showed good agreement, both in terms of failure loads and deformed shapes at failure. The validated finite element model was then used for the purpose of a parametric study comprising 96 models to investigate the effect of longer to shorter leg ratios, stiffener provided in the longer leg, thicknesses and lengths on axial strength of back-to-back built-up CFS unequal angle sections. Four different thicknesses and seven different lengths (stub to slender columns) with three overall widths to the overall depth (B/D) ratios were investigated in the parametric study. Axial strengths obtained from the experimental tests and FE analyses were used to assess the performance of the current design guidelines as per the Direct Strength Method (DSM); obtained comparisons show that the current DSM is conservative by only 7% and 5% on average, while predicting the axial strengths of back-to-back built-up CFS unequal angle sections with and without the stiffener, respectively.

• Structural Engineering and Mechanics
• /
• v.75 no.3
• /
• pp.327-337
• /
• 2020

The stability and dynamic performance of a flapper-nozzle servo valve depend on several factors, such as the motion of the armature component and the deformation of the spring tube. As the only connection between the armature component and the fixed end, the spring tube plays a decisive role in the dynamic response of the entire system. Aiming at predicting the vibration characteristics of the servo valves to combine them with the control algorithm, an innovative dynamic stiffness based on a distributed parameter model (DPM) is proposed that can reflect the dynamic deformation of the spring tube and a suitable discrete method is applied according to the working condition of the spring tube. With the motion equation derived by DPM, which includes the impact of inertia, damping, and stiffness force, the mathematical model of the spring tube dynamic stiffness is established. Subsequently, a suitable program for this model is confirmed that guarantees the simulation accuracy while controlling the time consumption. Ultimately, the transient response of the spring tube is also evaluated by a finite element method (FEM). The agreement between the simulation results of the two methods shows that dynamic stiffness based on DPM is suitable for predicting the transient response of the spring tube.

• Structural Engineering and Mechanics
• /
• v.72 no.5
• /
• pp.585-595
• /
• 2019

Foundations are a vital part of structures. Over time, the foundations can deteriorate due to unforeseen overloads and/or settlements, resulting in the appearance of cracks in the concrete. These cracks produce changes in the static and dynamic behavior of the affected foundation, which alter its load carrying capacity. In this work, non-destructive techniques of relative simplicity of application are presented for the detection, location, and quantification of damage, using numerical models, solved with the finite element method and Power Series. For this, two types of parameters are used: static (displacement and elastic curvature) and dynamics (natural frequencies). In the static analysis, the damage detection is done by means of a finite elements model representing a beam supported on an elastic foundation with a discrete crack that varies in length and location. With regard to dynamic analysis, the governing equations of the model are presented and a method based on Power Series is used to obtain the solution for a data set, which could be the Winkler coefficient, the location of the crack or the frequency. In order to validate the proposed methodologies, these techniques are applied to data obtained from laboratory tests.

• Proceedings of the KIEE Conference
• /
• 1994.07a
• /
• pp.150-152
• /
• 1994

In this paper, the optimum shape design of stator slot of induction motors for iron loss reduction is proposed. To obtain the flux distribution in induction motors, 2-D finite element method with voltage source is employed. The iron loss is calculated from the iron loss data given by the iron manufacturer. To calculate the sensitivity of iron loss to shape variation, the sensitivity analysis of discrete approach is used. The proposed algorithm is applied to a 3-phase squirrel cage induction motor. The nodes at stator slot boundary of the induction motor are defined as design parameters. By controlling these parameters under the constant volume of iron, we can minimize the iron loss. Furthermore, the stator copper loss is reduced by increasing the slot area. So the stator slot area is determined at the point that the summation of iron loss and copper loss of stator is minimized. Since the constraint of constant volume of iron is nonlinear to the design parameters, the Gradient Projection method is used as an optimization algorithm.

• Geomechanics and Engineering
• /
• v.30 no.1
• /
• pp.27-44
• /
• 2022

Gravelly soil is a kind of special geotechnical material, which is widely used in the subgrade engineering of railway, highway and airport. Its mechanical properties are very complex, and will greatly influence the stability of subgrade engineering. To investigate the mechanical properties and failure mechanism of gravelly soils, this paper introduced and verified a new discrete element method (DEM) of gravelly soils in large scale direct shear test, which considers the actual shape and broken characteristics of gravels. Then, the stress and strain characteristics, particle interaction, particle contact force, crack development and energy conversion in gravelly soils during the shear process were analyzed using this method. Moreover, the effects of gravel content (GC) on the mechanical properties and failure characteristics were discussed. The results reveal that as GC increases, the shear stress becomes more fluctuating, the peak shear stress increases, the volumetric strain tends to dilate, the average particle contact force increases, the cumulative number of cracks increases, and the shear failure plane becomes coarser. Higher GC will change the friction angle with a trend of "stability", "increase", and "stability". Differently, it affects the cohesion with a law of "increase", "stability" and "increase".

• Current Optics and Photonics
• /
• v.7 no.5
• /
• pp.529-536
• /
• 2023

The Morlet wavelet transform method is proposed to analyze a single interferogram with spatial carrier frequency that is captured by an optical interferometer. The method can retain low frequency components that contain the phase information of a measured optical surface, and remove high frequency disturbances by wavelet decomposition and reconstruction. The key to retrieving the phases from the low-frequency wavelet components is to extract wavelet ridges by calculating the maximum value of the wavelet transform amplitude. Afterwards, the wrapped phases can be accurately solved by multiple iterative calculations on wavelet ridges. Finally, we can reconstruct the wave-front of the measured optical element by applying two-dimensional discrete cosine transform to those wrapped phases. Morlet wavelet transform does not need to remove the spatial carrier frequency components manually in the processing of interferogram analysis, but the step is necessary in the Fourier transform algorithm. So, the Morlet wavelet simplifies the process of the analysis of interference fringe patterns compared to Fourier transform. Consequently, wavelet transform is more suitable for automated programming analysis of interference fringes and avoiding the introduction of additional errors compared with Fourier transform.