工业CT图像直接生成三角网格有限元模型

Direct Generation of Triangular Mesh Finite Element Model From Industrial CT Images

针对基于工业计算机断层成像(CT)图像的逆向设计存在定量分析难、自动化程度低及需要中间环节转化的问题,提出了一种将工业CT图像直接转换为三角网格有限元模型的方法。对CT图像进行轮廓识别后生成有限元模型的内外边界,根据图像像素点坐标信息将图像划分为若干单元,然后采用改进的三角剖分方法生成初始三角网格有限元模型,并对初始有限元模型进行了优化。结果表明,生成的三角网格最大边长比为1~2,最大扭曲角度数为0°~2°,网格质量满足实际工程应用需求的评价标准。并在样件、汽缸头模型上进行了实验,实验结果证明了该方法的可行性。

Regarding difficult quantitative analysis, low automation, and intermediate link transformation in reverse design based on industrial computerized tomography(CT) images, this paper proposes a method for direct conversion of industrial CT images into triangular mesh finite element models. After the contour recognition of a CT image, the inner and outer boundaries of the finite element model are generated. According to the coordinate information of image pixels, the image is divided into several elements. Then, the initial triangular mesh finite element model is generated with the improved triangulation method, and the initial finite element model is optimized. The results show that the maximum side ratio of the generated triangular meshes is between 1 and 2, and the maximum twist angle is between 0° and 2°, which meets the evaluation standard of practical engineering application. Experiments are carried out on the sample and cylinder head model, and the results prove the feasibility of the proposed method.