基于贝叶斯算法的岸边集装箱起重机有限元模型修正

Finite Element Model Updating of the container crane based on Bayesian Algorithm

依据贝叶斯统计方法结合马尔科夫蒙特卡洛模拟(MCMC),提出了一种高效的基于贝叶斯概率分布的智能有限元模型修正方法,主要针对存在不确定性因素的模型修正。结合马尔科夫蒙特卡洛抽样中的MH算法解决了贝叶斯模型修正计算效率低下的问题。建立了HIT101号岸边集装箱起重机有限元模型,将岸边集装箱起重机结构的约束考虑为大车与轨道支座的刚度系数及码头弹性系数,结合现场实际运行模态测试响应结果对模型的弹性模量、各部件密度、码头弹性系数、轨道处扭转弹簧刚度进行了贝叶斯修正,提高了有限元模型的精度,验证了基于贝叶斯模型修正方法在大型复杂结构上应用的有效性。

In this paper,a Bayesian model updating method based on design parameters is established,which is based on Bayesian statistical method and combine with MCMC simulations.Combined with the MH algorithm in the Markov Monte Carlo Sampling theory,the computational efficiency of Bayesian model updating is increased.The finite element model of HIT101 quayside container crane was established,and the constraint of crane structure was considered as the stiffness coefficient of the trolley and the rail support and the wharf elastic coefficient.The bayesian model updating was carried out in combination with the operating mode test response results of the model's elastic modulus,the density of each component,the wharf elastic coefficient and the torsion spring stiffness at the track.The accuracy of the finite element model is improved and the validity of the bayesian model correction method in large complex structures is verified.