• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.4
• /
• pp.221-232
• /
• 2023

This paper presents a novel convolutional neural network (CNN)-based approach for predicting bolt clamping force in the early bolt loosening state of bolted structures. The approach entails tightening eight bolts with different clamping forces and generating frequency responses, which are then used to create a similarity map. This map quantifies the magnitude and shape similarity between the frequency responses and the initial model in a fully fastened state. Krylov subspace-based model order reduction is employed to efficiently handle the large amount of frequency response data. The CNN model incorporates a regression output layer to predict the clamping forces of the bolts. Its performance is evaluated by training the network by using various amounts of training data and convolutional layers. The input data for the model are derived from the magnitude and shape similarity map obtained from the frequency responses. The results demonstrate the diagnostic potential and effectiveness of the proposed approach in detecting early bolt loosening. Accurate bolt clamping force predictions in the early loosening state can thus be achieved by utilizing the frequency response data and CNN model. The findings afford valuable insights into the application of CNNs for assessing the integrity of bolted structures.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.27 no.1
• /
• pp.25-35
• /
• 2023

Considering the non-linear behavior of structure and soil when evaluating a nuclear power plant's seismic safety under a beyond-design basis earthquake is essential. In order to obtain the nonlinear response of a nuclear power plant structure, a time-domain SSI analysis method that considers the nonlinearity of soil and structure and the nonlinear Soil-Structure Interaction (SSI) effect is necessary. The Boundary Reaction Method (BRM) is a time-domain SSI analysis method. The BRM can be applied effectively with a Perfectly Matched Layer (PML), which is an effective energy absorbing boundary condition. The BRM has a characteristic that the magnitude of the response in far-field soil increases as the boundary interface of the effective seismic load moves outward. In addition, the PML has poor absorption performance of low-frequency waves. For this reason, the accuracy of the low-frequency response may be degraded when analyzing the combination of the BRM and the PML. In this study, the accuracy of the analysis response was improved by adjusting the PML input parameters to improve this problem. The accuracy of the response was evaluated by using the analysis response using KIESSI-3D, a frequency domain SSI analysis program, as a reference solution. As a result of the analysis applying the optimal PML parameter, the average error rate of the acceleration response spectrum for 9 degrees of freedom of the structure was 3.40%, which was highly similar to the reference result. In addition, time-domain nonlinear SSI analysis was performed with the soil's nonlinearity to show this study's applicability. As a result of nonlinear SSI analysis, plastic deformation was concentrated in the soil around the foundation. The analysis results found that the analysis method combining BRM and PML can be effectively applied to the seismic response analysis of nuclear power plant structures.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.1-7
• /
• 2023

An artificial intelligence (AI) method based on image deep learning is proposed to predict the entire displacement shape of a structure using the feature of partial displacements. The performance of the method was investigated through a structural test of a steel frame. An image-to-image regression (I2IR) training method was developed based on the U-Net layer for image recognition. In the I2IR method, the U-Net is modified to generate images of entire displacement shapes when images of partial displacement shapes of structures are input to the AI network. Furthermore, the training of displacements combined with the location feature was developed so that nodal displacement values with corresponding nodal coordinates could be used in AI training. The proposed training methods can consider correlations between nodal displacements in 3D space, and the accuracy of displacement predictions is improved compared with artificial neural network training methods. Displacements of the steel frame were predicted during the structural tests using the proposed methods and compared with 3D scanning data of displacement shapes. The results show that the proposed AI prediction properly follows the measured displacements using 3D scanning.

• Journal of the Korean Geotechnical Society
• /
• v.24 no.10
• /
• pp.57-70
• /
• 2008

The advantage of piezocone penetration test is a guarantee of continuous data, which is a source of reliable interpretation of target soil layer. Many researches have been carried out f3r several decades and several classification charts have been developed to classify in-situ soil from the cone penetration test result. Since most present classification charts or methods were developed based on the data which were compiled over the world except Korea, they should be verified to be feasible for Korean soil. Furthermore, sometimes their charts provide different soil classification results according to the different input parameters. However, unfortunately, revision of those charts is quite difficult or almost impossible. In this research a new soil classification model is proposed by using fuzzy C-mean clustering and neuro-fuzzy theory based on the 5371 CPT results and soil logging results compiled from 17 local sites around Korea. Proposed neuro-fuzzy soil classification model was verified by comparing the classification results f3r new data, which were not used during learning process of neuro-fuzzy model, with real soil log. Efficiency of proposed neuro-fuzzy model was compared with other soft computing classification models and Robertson method for new data.

• Journal of the Korean Geotechnical Society
• /
• v.25 no.8
• /
• pp.33-45
• /
• 2009

The conventional single-core PBDs have been widely used in order to accelerate consolidation settlement of soft grounds. When using the single-core PBD in a thick clay deposit, a delay of consolidation may occur due to high confining pressure in the thick deposit and necking of drains. This study is to compare the performances of soil improvement by the single-core and double-core PBD installed at a site in Busan New Port which exhibits approximately a 40m-thick clay layer. An in-situ test program was performed at the test site where a set of the double-core PBDs and single-core PBDs were installed to compare the efficiency of each drain. In addition, the discharge capacity of each PBD has been measured using the modified Delft Test. A series of laboratory tests for estimating in-situ soil properties have also been performed in order to obtain input parameters for a numerical program ILLICON. The discharge capacity of the double-core PBD is higher than that of the single-core PBD in the modified Delft Test. However it is observed from the comparative in-situ test and numerical analysis that there is no difference in the performance of ground improvement between the two drain systems. This discrepancy comes from the fact that the amount of water released during consolidation in most common field conditions is much smaller than the capacity of even the single core PBD. And thus, considering actual field conditions, it can be concluded that the single-core PBD has enough discharge capacity even in the thick clay deposit such as this test site.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.6B
• /
• pp.697-707
• /
• 2008

The grid-based KIneMatic wave STOrm Runoff Model (KIMSTORM) by Kim (1998) predicts the temporal variation and spatial distribution of overland flow, subsurface flow and stream flow in a watershed. The model programmed with C++ language on Unix operating system adopts single flowpath algorithm for water balance simulation of flow at each grid element. In this study, we attempted to improve the model by converting the code into FORTRAN 90 on MS Windows operating system and named as ModKIMSTORM. The improved functions are the addition of GAML (Green-Ampt & Mein-Larson) infiltration model, control of paddy runoff rate by flow depth and Manning's roughness coefficient, addition of baseflow layer, treatment of both spatial and point rainfall data, development of the pre- and post-processor, and development of automatic model evaluation function using five evaluation criteria (Pearson's coefficient of determination, Nash and Sutcliffe model efficiency, the deviation of runoff volume, relative error of the peak runoff rate, and absolute error of the time to peak runoff). The modified model adopts Shell Sort algorithm to enhance the computational performance. Input data formats are accepted as raster and MS Excel, and model outputs viz. soil moisture, discharge, flow depth and velocity are generated as BSQ, ASCII grid, binary grid and raster formats.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.12 no.6
• /
• pp.429-441
• /
• 2010

In this study, fire resistance of a fireproof material sprayed upon an immersed tunnel was experimentally evaluated under $HC_{inc}$ and IS0834(duration of 4 hours) fire scenarios. Under $HC_{inc}$ fire scenario, the maximum inner temperatures of a concrete specimen at the depth of 0, 25 and 50 mm from the interface between the structure and the fire-proofing layer were $311^{\circ}C$, $194^{\circ}C$ and $142^{\circ}C$ respectively. Similarly, the corresponding maximum temperatures under IS0834 fire scenario were $332^{\circ}C$, $222^{\circ}C$ and $179^{\circ}C$ respectively. From the results, it was revealed that the two different fire scenarios assumed in this study have almost the same fire capacity as each other in the maximum temperature concept. In addition, a structural analysis of the immersed tunnel under $HC_{inc}$ fire scenario was carried out to verify the effects of the fireproof material on its structural stability. Material loss and deterioration of a concrete specimen without any fire-proofing measure was also experimentally evaluated to obtain input parameters for the structural analysis under such a severe fire scenario. From the results, it was confirmed that the application of fireproof measures to the immersed tunnel is essential for its structural stability even under a severe fire scenario.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.519-524
• /
• 2023

In hashing-based image retrieval, the hash code of a manipulated image is different from the original image, making it difficult to search for the same image. This paper proposes and evaluates a self-supervised deephashing model that generates perceptual hash codes from feature information such as texture, shape, and color of images. The comparison models are autoencoder-based variational inference models, but the encoder is designed with a fully connected layer, convolutional neural network, and transformer modules. The proposed model is a variational inference model that includes a SimAM module of extracting geometric patterns and positional relationships within images. The SimAM module can learn latent vectors highlighting objects or local regions through an energy function using the activation values of neurons and surrounding neurons. The proposed method is a representation learning model that can generate low-dimensional latent vectors from high-dimensional input images, and the latent vectors are binarized into distinguishable hash code. From the experimental results on public datasets such as CIFAR-10, ImageNet, and NUS-WIDE, the proposed model is superior to the comparative model and analyzed to have equivalent performance to the supervised learning-based deephashing model. The proposed model can be used in application systems that require low-dimensional representation of images, such as image search or copyright image determination.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.29 no.1
• /
• pp.1-13
• /
• 2024

The southwestern part of the East Sea is known to have a high primary productivity compared to those in the northern and eastern parts, which is attributed to nutrients supplies either by Tsushima Warm Current or by coastal upwelling. However, research on the biological pump in this area is limited. We developed machine learning models to estimate net community production (NCP), a measure of biological pump, with high spatial and time scales of 4 km and 8 days, respectively. The models were fed with the input parameters of sea surface temperature, chlorophyll-a, mixed layer depths, and photosynthetically active radiation and trained with observed NCP derived from high resolution measurements of surface O₂/Ar. The root mean square error between the predicted values by the best performing machine model and the observed NCP was 6 mmol O₂ m^-2 d^-1, corresponding to 15% of the average of observed NCP. The NCP in the central part of the Ulleung Basin was highest in March at 49 mmol O₂ m^-2 d^-1 and lowest in June and July at 18 mmol O₂ m^-2 d^-1. These seasonal variations were similar to the vertical nitrate flux based on the ³He gas exchange rate and to the particulate organic carbon flux estimated by the ²³⁴Th disequilibrium method. To expand this method, which produces NCP estimate for spring and summer, to autumn and winter, it is necessary to devise a way to correct bias in NCP by the entrainment of subsurface waters during the seasons.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.44 no.6
• /
• pp.177-187
• /
• 2016

This study started with the question, "Is the cost of landscape construction work in residential areas measured by public enterprises, 'in response to the needs of consumers?" The study analyzed whether the landscape construction expenditure is being introduced at an appropriate ratio according to the value the residents have regarding landscape elements. Following this, research was conducted for the purpose of providing basic data for improving the efficiency of formulating apartment landscape construction costs in the future. This research proceeded based on a questionnaire survey of residents of apartments, and the content of the questionnaire used frequency analysis and descriptive statistics research methods. To take a look at a comparative analysis of value recognition, in particular, a comparative analysis was performed based on the actual input cost based on the ratio of landscape elements by layer. Conclusions were found as follows: First, the degree of interest in the apartment landscape of the tenants was high, and the value of the landscape was high but realistic satisfaction appeared comparatively low. Second, the awareness of residents' values regarding landscape elements appeared to give "plantings" more value than "facilities". Thirdly, as a result of a mutual comparison between the values recognized by the resident regarding landscape elements and the construction input fee, depending on the landscape elements, it appeared that there is a difference in the ratio of up to 52 times from 1.25. Fourth, the fact that there is a difference in the relative proportion of value recognition and inputting construction cost indicates that it is not possible to respond to the needs of tenants during the construction cost development process. It also shows that the utility of inputting construction costs is low. Therefore, a macro-level examination such as reflecting the existing inflation rate is necessary to develop the efficient landscape construction cost of apartment such as the awareness of the value of the residents regarding landscape elements, out of the customary construction cost formulation method based on the microscopic dimensions of the consumer side.