• Transactions of Materials Processing
• /
• v.29 no.2
• /
• pp.89-96
• /
• 2020

The finite element method has been widely applied in the sheet metal forming process. However, the finite element method is computationally expensive and time consuming. In order to tackle this problem, surrogate modeling methods have been proposed. An artificial neural network (ANN) is one such surrogate model and has been well studied over the past decades. However, when it comes to ANN with two or more layers, so called deep neural networks (DNN), there is distinct a lack of research. We chose to use DNNs our surrogate model to predict the behavior of sheet metal in the deep drawing process. Thickness variation is selected as an output of the DNN in order to evaluate workpiece feasibility. Input variables of the DNN are radius of die, die corner and blank holder force. Finite element analysis was conducted to obtain data for surrogate model construction and testing. Sampling points were determined by full factorial, latin hyper cube and monte carlo methods. We investigated the performance of the DNN according to its structure, number of nodes and number of layers, then it was compared with a radial basis function surrogate model using various sampling methods and numbers. The results show that our DNN could be used as an efficient surrogate model for the deep drawing process.

• The Journal of Korea Robotics Society
• /
• v.13 no.1
• /
• pp.1-7
• /
• 2018

The image-to-image translation is one of the deep learning applications using image data. In this paper, we aim at improving the performance of object transfiguration which transforms a specific object in an image into another specific object. For object transfiguration, it is required to transform only the target object and maintain background images. In the existing results, however, it is observed that other parts in the image are also transformed. In this paper, we have focused on the structure of artificial neural networks that are frequently used in the existing methods and have improved the performance by adding constraints to the exiting structure. We also propose the advanced structure that combines the existing structures to maintain their advantages and complement their drawbacks. The effectiveness of the proposed methods are shown in experimental results.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.5 no.2
• /
• pp.186-190
• /
• 2004

By this time, a mold with nano size pattern was produced using a fabrication of X-ray lithography method and in a m icro size's case it was produced using fabrication of Deep UV lithography. In this paper, we produced mold with 400 $\mu{m}$depth pattern using a new technology of SLS(Selective Laser Sintering) Rapid Prototyping method. In addition to enhance strength and thermal stability, we produced Ni structure with a thickness of 300 $\mu{m}$ on a surface of mold using electro forming method.

• Journal of Korea Multimedia Society
• /
• v.22 no.5
• /
• pp.547-557
• /
• 2019

To extract the depth map from a single image, a number of CNN-based deep learning methods have been performed in recent research. In this study, the GAN structure of Pix2Pix is maintained. this model allows to converge well, because it has the structure of the generator and the discriminator. But the convolution in this model takes a long time to compute. So we change the convolution form in the generator to a depthwise convolution to improve the speed while preserving the result. Thus, the seven down-sizing convolutional hidden layers in the generator U-Net are changed to depthwise convolution. This type of convolution decreases the number of parameters, and also speeds up computation time. The proposed model shows similar depth map prediction results as in the case of the existing structure, and the computation time in case of a inference is decreased by 64%.

• Journal of Korean Association for Spatial Structures
• /
• v.20 no.2
• /
• pp.51-58
• /
• 2020

Recently, deep learning that is the most popular and effective class of machine learning algorithms is widely applied to various industrial areas. A number of research on various topics about structural engineering was performed by using artificial neural networks, such as structural design optimization, vibration control and system identification etc. When nonlinear semi-active structural control devices are applied to building structure, a lot of computational effort is required to predict dynamic structural responses of finite element method (FEM) model for development of control algorithm. To solve this problem, an artificial neural network model was developed in this study. Among various deep learning algorithms, a recurrent neural network (RNN) was used to make the time history response prediction model. An RNN can retain state from one iteration to the next by using its own output as input for the next step. An eleven-story building structure with semi-active tuned mass damper (TMD) was used as an example structure. The semi-active TMD was composed of magnetorheological damper. Five historical earthquakes and five artificial ground motions were used as ground excitations for training of an RNN model. Another artificial ground motion that was not used for training was used for verification of the developed RNN model. Parametric studies on various hyper-parameters including number of hidden layers, sequence length, number of LSTM cells, etc. After appropriate training iteration of the RNN model with proper hyper-parameters, the RNN model for prediction of seismic responses of the building structure with semi-active TMD was developed. The developed RNN model can effectively provide very accurate seismic responses compared to the FEM model.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.466-467
• /
• 2018

Recently, artificial intelligence has been applied to various fields such as image recognition, image recognition speech recognition, and natural language processing, and interest in Deep Learning technology is increasing. Many researches on Convolutional Neural Network(CNN), which is one of the most representative algorithms among Deep Learning, have strong advantages in image recognition and classification and are widely used in various fields. In this paper, we propose a new network structure that transforms the general CNN structure. A typical CNN structure consists of a convolution layer, ReLU layer, and a pooling layer. Therefore in this paper, We intend to construct a new network by adding fully connected layer inside a general CNN structure. This modification is intended to increase the learning and accuracy of the convoluted image by including the generalization which is an advantage of the neural network.

• Journal of the Korean Geotechnical Society
• /
• v.37 no.9
• /
• pp.37-50
• /
• 2021

Recently, due to the expansion of urban infrastructure using underground spaces in urban areas, many adjacent constructions and excavations have been made carried out between existing facilities, and complaints related to the stability of existing facilities due to close construction have become significant issues. In this study, it was closely reviewed for the existing metro tunnel structure in the new Dongbuk urban metro railway to determine the behavioral characteristics of tunnel structure according to adjacent tunnel construction. Also, it was analysed the evaluation of the safety zone and excavation method for metro tunnel structure. And after a detailed damage assessment, track irregularities and structural calculation by using a numerical analysis, stability of the metro tunnel structure according to nearby tunnel excavation was evaluated to be secured for safety. This study is expected to be applied as practical reference to review the evaluation of safety effects of existing tunnel structure and buildings according to adjacent construction in complex deep urban tunnelling.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.49-57
• /
• 2022

Recently, various uses of artificial intelligence have been made possible through the deep artificial neural network structure of machine learning, demonstrating human-like capabilities. Unfortunately, the deep structure of the artificial neural network has not yet been accurately interpreted. This part is acting as anxiety and rejection of artificial intelligence. Among these problems, we solve the capability part of artificial neural networks. Calculate the size of the artificial neural network structure and calculate the size of data that the artificial neural network can process. The calculation method uses the group method used in mathematics to calculate the size of data and artificial neural networks using an order that can know the structure and size of the group. Through this, it is possible to know the capabilities of artificial neural networks, and to relieve anxiety about artificial intelligence. The size of the data and the deep artificial neural network are calculated and verified through numerical experiments.

• Geomechanics and Engineering
• /
• v.36 no.3
• /
• pp.303-316
• /
• 2024

Seismic records are composed of mainshock and a series of aftershocks which often result in the incremental damage to underground structures and bring great challenges to the rescue of post-disaster and the repair of post-earthquake. In this paper, the repetition method was used to construct the mainshock-aftershocks sequence which was used as the input ground motion for the analysis of dynamic time history. Based on the Daikai station, the two-dimensional finite element model of soil-station was established to explore the failure process of station under different seismic precautionary intensities, and the concept of incremental damage of station was introduced to quantitatively analyze the damage condition of structure under the action of mainshock and two aftershocks. An arc rubber bearing was proposed for the shock absorption. With the arc rubber bearing, the mode of the traditional column end connection was changed from "fixed connection" to "hinged joint", and the ductility of the structure was significantly improved. The results show that the damage condition of the subway station is closely related to the magnitude of the mainshock. When the magnitude of the mainshock is low, the incremental damage to the structure caused by the subsequent aftershocks is little. When the magnitude of the mainshock is high, the subsequent aftershocks will cause serious incremental damage to the structure, and may even lead to the collapse of the station. The arc rubber bearing can reduce the damage to the station. The results can offer a reference for the seismic design of subway stations under the action of mainshock-aftershocks.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.61 no.3
• /
• pp.55-65
• /
• 2019

Although crack on concrete exists from its early formation, crack requires attention as it affects stiffness of structure and can lead demolition of structure as it grows. Detecting cracks on concrete is needed to take action prior to performance degradation of structure, and deep learning can be utilized for it. In this study, transfer learning, one of the deep learning techniques, was used to detect the crack, as the amount of crack's image data was limited. Pre-trained Inception-v3 was applied as a base model for the transfer learning. Web scrapping was utilized to fetch images of concrete wall with or without crack from web. In the recognition of crack, image post-process including changing size or removing color were applied. In the visualization of crack, source images divided into 30px, 50px or 100px size were used as input data, and different numbers of input data per category were applied for each case. With the results of visualized crack image, false positive and false negative errors were examined. Highest accuracy for the recognizing crack was achieved when the source images were adjusted into 224px size under gray-scale. In visualization, the result using 50 data per category under 100px interval size showed the smallest error. With regard to the false positive error, the best result was obtained using 400 data per category, and regarding to the false negative error, the case using 50 data per category showed the best result.