基于机器视觉的橡胶材料有限元快速建模方法研究

Research on Rapid Finite Element Modeling Method of Rubber Materials Based on Machine Vision

摘要: 针对橡胶材料几何形状复杂、传统建模效率低、参数获取困难等问题，提出了一种基于机器视觉的橡胶材料有限元快速建模方法。该方法以橡胶哑铃试件为研究对象，通过图像采集、去除背景、OpenCV轮廓识别、比例尺自动标定、提取试件的轮廓像素坐标并转换为实际尺寸，最后利用Python脚本驱动Abaqus实现自动建模与有限元分析。结果表明，该方法在几秒内就可以完成建模，效率较传统的软件建模提升上十倍；与传统软件建立的模型相比，基于机器视觉建立的模型几何尺寸误差小于2.75%；相同工况下应力分析结果与传统模型最大误差为3.37%。该方法效率高且可靠，可推广应用于各类复杂橡胶制品的快速建模与仿真分析。

Abstract: Due to the problems of complex geometry, low efficiency of traditional modeling and difficulty in parameter acquisition for rubber materials, a rapid finite element modeling method for rubber materials based on machine vision is proposed. Taking rubber dumbbell specimens as the research object, the method adopts image acquisition, background removal, OpenCV contour recognition, automatic scale calibration, extracts the contour pixel coordinates of specimens and converts them into actual dimensions. Finally, Python script is driven to realize automatic modeling and finite element analysis. The results show that the proposed method can complete the modeling within a few seconds, and the efficiency is ten times higher than that of the traditional software modeling method. Compared with the model established by traditional software, the geometric dimension error of the model established based on machine vision is less than 2.75%. Under the same working conditions, the maximum error of stress analysis results compared with the traditional model is 3.37%. The method is efficient and reliable, and can be widely applied to the rapid modeling and simulation analysis of various complex rubber products.