一种ECVT有限元自动剖分方法  

A Method of FEM Self-Organizing Mesh Generation for ECVT

作　　者：邹星星[1] 赵进创[1] 傅文利[1] 马增强[2] 

机构地区：[1]广西大学计算机与电子信息学院,广西南宁530004 [2]石家庄铁道大学电气与电子工程学院,河北石家庄050043

出　　处：《石家庄铁道大学学报（自然科学版）》2017年第3期106-110,共5页Journal of Shijiazhuang Tiedao University(Natural Science Edition)

基　　金：国家自然科学基金(61362025)

摘　　要：ECT三维传感器场域剖分是当前ECT直接三维成像研究的关键问题,提出一种三维有限元自动剖分方法并得到结果验证。采用四面体作为剖分单元,利用Delaunay四面体剖分算法反复迭代实现三维实体自动剖分;为清晰显示剖分图像,结合画家算法对网格进行消隐处理。仿真结果表明,随着迭代次数增加,剖分精度提高,其剖分时长也会相应增加。根据需求剖分精度设置迭代次数,能有效的剖分传感器场域,实现了三维场域自动剖分,为计算三维敏感场从而提高ECT三维成像精度奠定基础。As three-dimensional sensor field mesh of ECT is a key issue in direct 3D ECT currently,a method of 3D FEM self-organizing mesh generation is proposed.The method adopts tetrahedron as split unit,and uses Delaunay triangulation algorithm for auto meshing 3D solid.Finally,the painter algorithm is used to process grid blanking.Simulation results show that with the increase in the number of iterations,mesh accuracy is improved while more time is spent.Setting iteration according to the precision of mesh can effectively mesh the sensor field,achieving the three-dimensional auto meshing,which provides basis for calculating the three-dimensional sensitivity field in order to improve the accuracy of three-dimensional imaging of ECT.

关 键 词：电容层析体积成像 有限元 自动剖分 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]