입 · 출력 변수간 불확실성 상관성을 고려한 차량 구조 유한요소 모델의 확률론적 보정 및 검증
김원곤, 이규석, 손혜정 et al. · 대한기계학회 춘추학술대회 · 2016
This paper presents calibration and validation of automotive structural finite element model based on the experiments considering various sources of uncertainty. In experiments, the plain principal strains on automotive structure surface at four points are measured. Also, displacement at the loading point are measured under various sources of uncertainty. In calibration procedures, genetic algorithm is used to calibrate the statistical moments of input uncertain variables and correlation coefficients between random variables to minimize a statistical difference between the simulation responses and experimental responses. To measure the statistical difference, calibration metrics including Probability Residual (PR) and Correlation Coefficient Residual (CCR) are implemented. For the sake of considering uncertainties, the Monte Carlo Simulation (MCS). At the end, the calibrated model is validated by the displacement at loading point.
Cited by 0 works.