http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
서민규(Minkyu Seo),이형근(Hyeongkeun Lee) 한국자동차공학회 2011 한국자동차공학회 부문종합 학술대회 Vol.2011 No.5
A CAE simulation for the slam durability test of moving parts is proposed using explicit code. Previously other techniqes usging implicit code were proposed, but they are restricted in the description of the impact of the moving parts. To overcome the restrictions, Explicit code is adopted and proper modeling methods are proposed for the moving parts which includes side doors, hood, trunk lid. In this paper, the detail modelings of subcomponents are proposed including weather strip, locking mechanism, latch, striker, gas lifter, balance bar. Using the developed modeling technique, slam simulations and stress analyses are done and the life is predicted for the door, hood and trunk lid. Also the future works is discussed to simulate the behavior of the moving parts better with the test result.
엔진룸 음향모드 제어를 통한 부밍대역 공기기인소음개선 연구
서민규(Minkyu Seo),채기상(Ki-Sang Chae),김대운(Daewoon Kim) 한국자동차공학회 2015 한국자동차공학회 학술대회 및 전시회 Vol.2015 No.11
The reduction of vehicle booming noise was conducted normally in the area of structure borne sound or intake and exhaust sound. This paper investigated the application of the Helmholz Resonator to control the acoustic mode of the engine room. The acoustic FEM and acoustic BEM techniques were examined on the cavity mode analysis. Generally the acoustic FEM is applied to the closed volume like the passenger room and the acoustic BEM to the open volume. The engine room is connected to the outer volume through the grill in the front area and through the drive line and the exhaust pipe in the rear area. This paper used the acoustic FEM technique, took the engine room and the proper exterior open volume. Considering the acoustic boundary conditions, acoustic modes was calculated and visualized. The three dimensional visualized mode gave us good results when compared with the experiment and the acoustic BEM technique. The three dimensional visual of the cavity mode gave us the optimal position of the resonator. By taking the proper shape of the resonator, the airborne booming sound was reduced at the test vehicle of high speed.
문봉관(Bongkwan Moon),안성현(Sunghyun An),서민규(Minkyu Seo),박은주(Eunju Park),임한규(Hankyu Lim) 한국정보기술학회 2021 Proceedings of KIIT Conference Vol.2021 No.11
물가는 여러 요인들에 의하여 결정된다. 본 논문에서는 딥러닝을 이용하여 예상치 못했던 전염병의 확산이 식품의 물가에 미치는 영향을 알아보고자 한다. 이를 위해 학습 특성으로 국제유가, 달러 환율, 소비자 물가지수, 코로나 확진자 수를 열로 추가하여 사용한다. 딥러닝 모델로 시계열 예측에 주로 사용되는 LSTM 모델을 사용하여 식품의 물가를 예측하는 시스템을 설계하고 Flask를 사용하여 이를 웹서버에 연동한다. 특정 요인에 대하여 사용자의 요청이 있을 경우, 이미 학습된 모델이 있으면 학습된 모델을 기반으로 물가 예측 그래프를 바로 출력한다. 사용자 요청에 학습된 모델이 없다면 딥러닝을 실행하여 모델을 학습한 후 그래프를 출력하도록 구현하였다. Prices depend on a number of factors. In this paper, we are going to study about the effect of the unexpected spread of infectious disease using deep learning. International oil prices, dollar exchange rates, consumer price indexes, and the number of COVID-19 confirmed patients are added as columns for learning characteristics. A deep learning model LSTM, which is mainly used for time series prediction, is used to design a system for predicting food prices, and the system is linked to a web server using Flask. When there’s a user’s request for a specific factor, if there is a model that has already been learned, a price prediction graph will immediately be shown based on the learned model. If there is no model learned in advance by user request, deep learning is performed to learn the model and then show a graph.