基于贝叶斯原理的梁结构模型修正方法

Beam Structure Model Modification Method Based On Bayesian Principle

在利用正则化方法修正钢筋混凝土梁有限元模型时,通常会出现的病态解及测点信息与模态信息不完整的情况,本文基于此种情况提出了一个基于贝叶斯定理的钢筋混凝土梁的有限元模型修正方法,针对测点有限的情况,建立了适用于此种情况的目标函数和似然函数。数值计算结果表明,本方法在仅提供低阶的测量模态信息和不完整的振型信息的情况下,依然能够准确得到修正结构参数。

When using the regularization method to modify the finite element model of reinforced concrete beams, there may be wrong solutions and incomplete information of measuring points modal information. Based on this situation, this paper proposes a finite element model correction method for reinforced concrete beams based on Bayes’ principle. An objective function and a likelihood function suitable for this situation are established for a limited number of measurement points. Numerical calculation results show that the method can accurately obtain the modified structural parameters even when only the low-order measurement modal information and incomplete mode information are provided.