公差分析
几何尺寸和公差
计算机辅助设计
有限元法
方向(向量空间)
计算机辅助设计
插值(计算机图形学)
计算机科学
工程制图
计算机辅助
结构工程
工程类
机械工程
几何学
数学
帧(网络)
作者
Mehdi Tlija,Anis Korbi,Borhen Louhichi,Antoine Tahan
出处
期刊:Journal of Computing and Information Science in Engineering
[ASME International]
日期:2022-04-01
卷期号:23 (2)
被引量:4
摘要
Abstract Product assemblability and functional behavior are affected by geometric deviations. These deviations consist of manufacturing errors and part deformation defects caused by external loads. Taking the sources of deviations into account in tolerance analysis yields not only to more precise and reliable results but also to more complex tasks. In this context, the modeling of assembly parts with defects in a Digital MockUp (DMU) is quite promising. In this article, an integrated decision support tool is proposed to consider the causes of multiple defects, such as tolerances and external mechanical loads, in the tolerancing process. The worst-case concept and the small displacement torsor (SDT) are used to model rigid parts with orientation and positional defects. To model the part with form defects, random positions of each toleranced face points are determined using the Gaussian perturbation method (GPM) and considering the tolerance zone limits. A computer-aided design (CAD) method based on the B-spline interpolation allows the reconstruction of realistic surfaces of parts with form defects. Realistic configurations of nonrigid components subjected to external mechanical loads are determined using the finite element analysis (FEA). The realistic assembly configurations are performed by updating the mating constraints between planar and cylindrical parts. The proposed method considers all tolerance types on CAD models (positional, orientation, and form defects) and part deformations. The tolerance analysis is performed to check the compliance with the functional requirement (FR) and to correct the initial tolerance values. An industrial case study is used to validate the steps of the proposed tolerancing method.
科研通智能强力驱动
Strongly Powered by AbleSci AI