• Structural Engineering and Mechanics
• /
• v.26 no.3
• /
• pp.251-262
• /
• 2007

This paper examines the application of artificial neural networks (ANN) to the response prediction of geometrically nonlinear truss structures. Two types of analysis (deterministic and probabilistic analyses) are considered. A three-layer feed-forward backpropagation network with three input nodes, five hidden layer nodes and two output nodes is firstly developed for the deterministic response analysis. Then a back propagation training algorithm with Bayesian regularization is used to train the network. The trained network is then successfully combined with a direct Monte Carlo Simulation (MCS) to perform a probabilistic response analysis of geometrically nonlinear truss structures. Finally, the proposed ANN is applied to predict the response of a geometrically nonlinear truss structure. It is found that the proposed ANN is very efficient and reasonable in predicting the response of geometrically nonlinear truss structures.

• Proceeding of EDISON Challenge
• /
• 2016.11a
• /
• pp.19-22
• /
• 2016

Leading edge thrust is generally caused by passing air flow from lower to upper surface and it is required to have sufficient angle of attack for notable leading edge thrust. To produce leading edge thrust at low angle of attack, utilizing expansion wave accompanying low pressure is able to be a solution. Fore structure changes the direction of flow, and this flow passes the projected edge. As a result, from a perspective of the edge, it is able to have high angle of attack, and artificial expansion wave is generated. This concept shows 9.48% increase of L/D in inviscid flow, at Mach number 1.3 and angle of attack $1^{\circ}$ in maximum, and this model shows the 3.98% of increasement at angle of attack $2^{\circ}$. Although advantage of the artificial expansion wave decreased as angle of attack increase, it shows the possibility of aerodynamical improvement with artificial expansion wave.

• The Korean Journal of Applied Statistics
• /
• v.20 no.3
• /
• pp.441-447
• /
• 2007

The conjoint analysis is widely accepted in the field of marketing as a way to understand and incorporate the structure of customer preferences into the new product design process. We apply the conjoint analysis for understanding preferences about after school computer courses in elementary schools. We show that the artificial neural network analysis in addition to the conjoint analysis is very useful to understand the needs of elementary school students about after school computer courses.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.159-159
• /
• 2019

In this study, the simulation of sediment deposition at Sangju weir reservoir, South Korea, was carried out using artificial neural networks. The ANNs have typically been used in water resources engineering problems for their robustness and high degree of accuracy. Three basic variables namely turbid water inflow, outflow, and water stage have been used as input variables. It was found that ANNs were able to establish valid relationship between input variables and target variable of sedimentation. The R value was 0.9806, 0.9091, and 0.8758 for training, validation, and testing phase respectively. Comparative analysis was also performed to find optimum structure of ANN for sediment deposition prediction. 3-14-1 network architecture using BR algorithm outperformed all other combinations. It was concluded that ANN possess mapping capabilities for complex, non-linear phenomenon of reservoir sedimentation.

• Structural Engineering and Mechanics
• /
• v.55 no.6
• /
• pp.1223-1239
• /
• 2015

A method for structural damage identification based on Chaotic Artificial Bee Colony (CABC) algorithm is presented. ABC is a heuristic algorithm with simple structure, ease of implementation, good robustness but with slow convergence rate. To overcome the shortcoming, the tournament selection mechanism is chosen instead of the roulette mechanism and chaotic search mechanism is also introduced. Residuals of natural frequencies and modal assurance criteria (MAC) are used to establish the objective function, ABC and CABC are utilized to solve the optimization problem. Two numerical examples are studied to investigate the efficiency and correctness of the proposed method. The simulation results show that the CABC algorithm can identify the local damage better compared with ABC and other evolutionary algorithms, even with noise corruption.

• Journal of the Semiconductor & Display Technology
• /
• v.18 no.2
• /
• pp.78-81
• /
• 2019

In this study, we will analyze the design, development and application of these small drones using open source. These drones are used in flight exercises, aerial photography, and coding education. In the era of the fourth industrial revolution, such as the development of sensor technology, expansion of open source sharing, and application of artificial intelligence, Is expected to be able to demonstrate convergence. In this paper, we have studied the design and fabrication of small drones using open source. In the case of drones, various functions and differentiated materials are required depending on the application, and the future development of the unmanned mobile object, namely the drone, in which the creativity and the technology are combined with each other continues to be enhanced by the improvement of autonomy and artificial intelligence. Software-based architecture-based technologies have been developed in collaboration with embedded SWs that combine sensors, motors, and control systems. In hardware, it is customary to use a combination of materials and design to increase the freedom of design. It will be made in a free structure.

• Journal of the Korean Society of Visualization
• /
• v.19 no.2
• /
• pp.32-40
• /
• 2021

Vapor-driven solutal Marangoni flow is governed by the concentration distribution of solutes on a liquid-gas interface. Typically, the flow structure is investigated by particle image velocimetry (PIV). However, to develop a theoretical model or to explain the working mechanism, the concentration distribution of solutes at the interface should be known. However, it is difficult to achieve the concentration profile theoretically and experimentally. In this paper, to find the concentration distribution of solutes around 2D droplet, the reverse tracking method with an artificial neural network based on PIV data was performed. Using the method, the concentration distribution of solutes around a 2D droplet was estimated for actual flow data from PIV experiment.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.354-359
• /
• 2022

에세이 자동 평가는 주어진 에세이를 읽고 자동으로 평가하는 작업이다. 본 논문에서는 효과적인 에세이 자동 평가 모델을 위해 Argument Mining 작업을 사용하여 에세이의 논증 구조가 반영된 에세이 표현을 만들고, 에세이의 평가 항목별 표현을 학습하는 방법을 제안한다. 실험을 통해 제안하는 에세이 표현이 사전 학습 언어 모델로 얻은 표현보다 우수함을 입증했으며, 에세이 평가를 위해 평가 항목별로 다른 표현을 학습하는 것이 보다 효과적임을 보였다. 최종 제안 모델의 성능은 QWK 기준으로 0.543에서 0.627까지 향상되어 사람의 평가와 상당히 일치한다.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2022.04a
• /
• pp.135-136
• /
• 2022

As a recent technology, it is possible to partially grasp the occurrence of displacement of the entire building through artificial intelligence technology for big data through scanning. However, scanning and data processing take a lot of time, so there is a limit to constant monitoring, so constant monitoring technology of building behavior that combines wireless remote sensors and 3D shape scanning is required. Therefore, in this study, artificial intelligence program coding technology is linked. In addition, a technology capable of real-time wireless remote measurement of structure displacement will be developed through technology development in response to safety management that combines existing building technologies such as sensors. Through this, it is possible to establish an integrated management system for safety inspection and diagnosis.

• Clinical Endoscopy
• /
• v.55 no.5
• /
• pp.594-604
• /
• 2022

Over the past decade, technological advances in deep learning have led to the introduction of artificial intelligence (AI) in medical imaging. The most commonly used structure in image recognition is the convolutional neural network, which mimics the action of the human visual cortex. The applications of AI in gastrointestinal endoscopy are diverse. Computer-aided diagnosis has achieved remarkable outcomes with recent improvements in machine-learning techniques and advances in computer performance. Despite some hurdles, the implementation of AI-assisted clinical practice is expected to aid endoscopists in real-time decision-making. In this summary, we reviewed state-of-the-art AI in the field of gastrointestinal endoscopy and offered a practical guide for building a learning image dataset for algorithm development.