基于改进区间逆响应面法的塔机有限元模型修正

Finite Element Model Updating of Tower Crane Based on Improved Interval Inverse Response Surface Method

摘要:

有限元模型修正技术广泛应用于机械等领域。在工程实际中，由于多种因素的影响，实际结构（如塔式起重机，简称塔机）与有限元模型之间普遍存在不确定性误差，造成有限元分析结果失真，因此研究结构的不确定性有限元模型修正具有重要意义。由于塔机应用的广泛性和事故的高发性，在考虑参数不确定性的情况下，对塔机的有限元模型进行了修正。为提高模型修正效率，引入响应面模型来代替塔机的有限元模型，利用RBF神经网络具有对复杂问题高精度拟合的优点，提出了一种改进的区间逆响应面方法对塔机进行了不确定性修正。通过三自由度弹簧-质量系统证明了所提出方法的可行性，并对实际塔机结构进行了区间修正，改善了传统区间逆响应面方法的不足，结果具有很高的计算精度和计算效率。

Abstract:

The finite element model updating technology is widely used in the mechanical and other fields. Because of the influence of many factors， there are many uncertain errors between the actual structure （such as tower crane） and the finite element model， resulting in the distortion of the finite element analysis results. Therefore， it is of great significance to study the finite element model updating with uncertain parameters. In this paper， the finite element model of a tower crane is updated considering the uncertainty of parameters. In order to improve the efficiency of model updating， the response surface model is introduced to replace the finite element model of tower crane. Considering the fact that the RBF neural network has the advantage of high precision fitting for complex problems， an improved interval inverse response surface method is proposed to update the tower crane with uncertain parameters. The feasibility of this method is proved by an example of spring-mass calculation， and the actual tower crane structure is modified. This method improves the deficiency of the interval inverse response surface method， whose result has a good calculation accuracy and efficiency.