http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
유전자 알고리즘을 이용한 로커암 축의 최적설계에 관한 연구
안용수(Y. S. An),이수진(S. J. Lee),이동우(D. W. Lee),홍순혁(S. H. Hong),조석수(S. S. Cho),주원식(W. S. Joo) 한국정밀공학회 2004 한국정밀공학회 학술발표대회 논문집 Vol.2004 No.10월
This study proposes a new optimization algorithm which is combined with genetic algorithm and ANOM. This improved genetic algorithm is not only faster than the simple genetic algorithm, but also gives a more accurate solution. The optimizing ability and convergence rate of a new optimization algorithm is identified by using a test function which have several local optimum and an optimum design of rocker arm shaft. The calculation results are compared with the simple genetic algorithm.
역전파 신경회로망을 이용한 피로손상 모델링에 관한 연구
김민철(M.C.Kim),주원식(W.S.Joo),장득열(D.Y.Jang),조석수(S.S.Cho),김순호(S.H.Kim) 한국자동차공학회 1998 한국자동차공학회 춘 추계 학술대회 논문집 Vol.1998 No.11_2
Back-propagation neural networks performs computer simulations that have the potential to find the same patterns that fatigue practitioners recognize to relate experimental results to fatigue life prediction. This potential was used to construct neural networks to recognize the relation between da/dN, N/Nf, X-ray diffraction half-value breadth ratio B/Bo, fractal dimension D_f and fracture mechanical parameters for Al 2024-T3 alloy. Learning and generalization of neural networks was optimized by floating rate method. This study shows that neural networks has ability to predict fatigue crack growth rate and life on data of unlearned experimental condition.<br/>
조석수,주원식 동아대학교 생산기술연구소 1999 生産技術硏究所硏究論文集 Vol.4 No.1
Quantitative nondestructive evaluation(QNDE) gets hold of position and shape of defects in mechanical components and estimates safety of structure. The ultrasonic defect detection out of QNDE is based on refraction of ultrasonic wave by defects in solid material. Neural network is information processing system modeling human brain and has an application to defect identification with ultrasonic defect detection. The present method consists of three process. The first process gets sample data on defect and ultrasonic wave parameters according to various defect patterns. In the second process, Neural networks is learned using sample data. In third process, neural networks predicts defect patterns by unlearned data This processes are applied to the identification of size and location of defects hidden in SM45C.
프랙탈 차원을 이용한 재료손상의 자기 상사성에 관한 연구
조석수,주원식 동아대학교 생산기술연구소 1999 生産技術硏究所硏究論文集 Vol.4 No.1
The postion and the length of surface micro-crack has random properties. It is difficult to detect the progressive changes in surface micro-crack distribution that occur during fatigue stressing. In this paper to overcome this problem, fractal dimension was introduced to characterized the change in two-dimensional surface micro-crack distribution during fatigue process. Box counting method was adopted to measure the fractal dimension. The features of surface micro-crack growth and coalescence have fractal property. Fractal dimension increases with the number of cyclic load. Remaining fatigue life of mechanical components can be predicted by the relation between fractal dimension and cycle ratio to fracture. Therefore, we proposed complex system design that uses fractal dimension in material design and structure design and safety estimation.
인공지능형 네트워크를 이용한 재료거동모델링에 관한 연구
조석수,김민철,주원식 동아대학교 생산기술연구소 1999 生産技術硏究所硏究論文集 Vol.4 No.1
Backpropagation neural networks performs computer simulations that have the potential to find the same patterns that fatigue practitioners recognize to relate experimental results to fatigue life prediction. This potential was used to construct neural networks to recognize the relation between X-ray diffraction half-value breadth ratio B/Bo, fractal dimension D_f, stress amplitude Δσ, main crack length α, (Δσ/σ_ys)^ma^n and da/dN, N/N_f for Al 2024-T3 alloy. Learning and generalization of neural networks was optimized by floating rate method. This study shows that neural networks has ability to predict fatigue crack growth rate and life on data of unlearned experimental condition.