http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
김성현(Sung-hyun Kim),방용식(Yong-sik Bang),강대오(Dae-Oh Kang),이용훈(Yong-Hoon Lee),허승진(Seung-Jin Heo),임홍재(Hong-Jae Yim) 한국자동차공학회 2010 한국자동차공학회 학술대회 및 전시회 Vol.2010 No.11
This study was conducted to apply for integrated robust optimal design considered multi-disciplinary constrained conditions. Multi-disciplinary constrained conditions mean that lightweight design, structure stiffness and safety. Because such as these design parameters are conflictive relation, it is hard to design optimal model through conventional parameter study. Also robust design process is necessary to consider manufacturing tolerance because of performance deviation. Process was constructed with 4-phase. First phase is formatting the base model. In this study, base model was simplified for efficiency of analysis and also verified with test data simultaneously. In second phase, effect analysis was performed to select design variables largely effecting on performance such as bending and torsion stiffness. Form second phase, totally 11 design variables were selected finally among the 105 variables. Third phase is formatting RBF type Meta-model established by using selected design variables and LHC design of experiment (DOE). In the last phase, optimum result was calculated by using the EasyDesign program. Bending, Torsion stiffness and roof crush reaction force were used for constrained conditions and mass minimization was object function for optimization.
이정현(Junghuyng Lee),김형일(Huyngil Kim),신경호(Kuyngho Shin),김일환(Ilhwan Kim),허승진(Seung Jin Heo),임홍재(Hong Jae Yim) 한국자동차공학회 2011 한국자동차공학회 학술대회 및 전시회 Vol.2011 No.11
This study was conducted to apply for optimal design considered multi-disciplinary constrained conditions. Multidisciplinary constrained conditions mean that lightweight design, structure stiffness and safety. Because such as these design parameters are conflictive relation, it is hard to design optimal model through conventional parameter study. First, effect analysis was performed to select design variables largely effecting on performance such as dynamic stiffness, static stiffness, full frontal crash, ODB offset crash and RCAR offset crash. From first phase, totally 12 design variables were selected. Second phase is RBF type Meta-model established by using selected design variables and LHC design of experiment (DOE). In the last phase is to find the optimal solution that meets constraints and objectives by using a sequential approximate optimization. Dynamic torsional stiffness, static torsional stiffness, some displacement and crash box section force were used for constrained conditions and mass minimization was object function for optimization.
조인트 DB와 다분야 구속 조건 최적설계를 이용한 차체 설계 프로세스 개발
박현정(Hyun-Jung Park),최연욱(Yeon-wook Choi),허승진(Seung-Jin Heo),이경원(Kyungwon Lee),김정호(Jung-Ho Kim),이창건(Changkun Lee),김용석(Yongsuk Kim) 한국자동차공학회 2014 한국자동차공학회 학술대회 및 전시회 Vol.2014 No.11
To achieve effectiveness and optimized design for light weight vehicle body, many concept design methodologies have been studied. However, study using database would be not conducted yet. Therefore, this paper presents body structure development process using joint database and multi-disciplinary design optimization. The suggested process was constructed with 4-phase. First phase is knowledge-based template construction. Knowledge-based template is consisted of geometry database and performance database. To construct joint database, it is necessary to evaluate joint’s cross section. In second phase, body structure conceptual design using joint database is progressed. For joint’s section selection, what-if design method is used. In addition, implicitly parametric geometry model is used to realize easily shape or topology changes. Parametric geometry model was constructed using SFE CONCEPT. In the last phase, body structure detail design using robust design optimization method is fulfilled. Considering the deviation of thickness design variables, six-sigma terms added into the inequality constraints for bending stiffness and torsional stiffness and equality constraints for 1st bending mode and 1st torsion mode.
유윤기(Yun-Ki You),김정호(Jung-Ho Kim),김동석(Dong-Seok Kim),이경원(Kyung-Won Lee),허승진(Seung-Jin Heo),임홍재(Hong-Jae Yim),강대오(Dae-Oh Kang) 한국자동차공학회 2009 한국자동차공학회 학술대회 및 전시회 Vol.2009 No.11
This study was conducted to establish initial process for integrated optimal design considered multi-disciplinary constrained conditions. Multi-disciplinary constrained conditions mean that lightweight design, structure stiffness and safety. Process was constructed with 4-phase. First phase is formatting the base model. In this study, base model was simplified for efficiency of analysis and also verified with test data simultaneously. In second phase, effect analysis was performed to select design variables largely effecting on performance such as bending and torsion stiffness. Form second phase, totally 11 design variables were selected finally among the 105 variables. Third phase is formatting RBF type Meta-model established by using selected design variables and LHC design of experiment(DOE). In the last phase, optimum result was calculated by using the LS-OPT program. Bending, Torsion stiffness and roof crush reaction force were used for constrained conditions and mass minimization was object function for optimization. This study is first-year result among the three years plan. Therefore more design variables and detail load case for safety such as frontal, side, rear and damageability will be applied to next step study.
김성현(Sunghyun Kim),김용석(Yongsuk Kim),김형일(Hyungil Kim),기원용(Won yong Ki),유지수(Ji Sue Yoo),허승진(Seung Jin Heo),임홍재(Hong Jae Yim) 한국자동차공학회 2012 한국자동차공학회 부문종합 학술대회 Vol.2012 No.5
This study was conducted to apply for optimal design considered multi-disciplinary constrained conditions. Multidisciplinary constrained conditions mean that lightweight design, structure stiffness and safety. Because such as these design parameters are conflictive relation, it is hard to design optimal model through conventional parameter study. First, effect analysis was performed to select design variables largely effecting on performance such as dynamic stiffness, static stiffness, full frontal crash, ODB crash and RCAR offset crash. From first phase, totally 12 design variables were selected. Second phase is Radial Basis Function (RBF) type Meta-model established by using selected design variables and Latin Hypercube (LHC) Design of Experiment (DOE). In the last phase is to find the optimal solution that meets constraints and objectives by using a Genetic Algorithm (GA) which consists of three genetic operations named selection, cross-over and mutation. To find optimum, this study uses LS-OPT and LS-DYNA.