首页> 外国专利> AREA DIVIDING METHOD FOR ANALYSIS OF VARIOUS PROBLEMS OF SOLID DYNAMICS AND METHOD FOR DISCRETELY ANALYZING VARIOUS PROBLEMS OF SOLID DYNAMICS

AREA DIVIDING METHOD FOR ANALYSIS OF VARIOUS PROBLEMS OF SOLID DYNAMICS AND METHOD FOR DISCRETELY ANALYZING VARIOUS PROBLEMS OF SOLID DYNAMICS

机译:固体动力学各种问题的区域划分方法和固体动力学各种问题的离散分析方法

摘要

PROBLEM TO BE SOLVED: To provide an area dividing method, for which a nodal point is not used in the approximate analysis of various problems of solid dynamics, and a method for discretely analyzing various problems of solid dynamics while applying this area dividing method.;SOLUTION: First of all, an analysis object area is divided into plural small areas (S1) and a state vector field is assumed by approximating the state vectors of respective small areas with the finite polynomial of a coordinate variable in an orthogonal curve coordinate system (S2). Next, the coordinate transformation of a boundary side (or boundary plan) is performed concerning the said finite polynomial (S3). The relational expression of an unfixed coefficient in the finite polynomial transformed in S3 is generated on the basis of a prescribed variation equation (S4). A square matrix is generated by adjusting the number of unfixed coefficients provided in S4 and multi-dimensional simultaneous linear equations with the unfixed coefficients as a square matrix are prepared (S5). Since respective elements (small areas) can be independently handled, the compression and solving calculation of simultaneous linear equations are performed by executing the skipped compressing processing of the matrix (S6). The accuracy of the approximate solution provided in S6 is investigated and when the accuracy is within an allowable error, the analytic calculation is finished. In the other case, processing is returned to S1 and analysis is performed by further dividing that small area or processing is returned to S2 and the element polynomial solution is reset.;COPYRIGHT: (C)2000,JPO
机译:要解决的问题:提供一种区域划分方法,其中在实体动力学的各种问题的近似分析中不使用结点,以及一种在应用该区域划分方法时离散地分析固体动力学的各种问题的方法。解决方案:首先,将分析对象区域划分为多个小区域(S1),并通过使用正交曲线坐标系中的坐标变量的有限多项式近似各个小区域的状态向量来假定状态向量场( S2)。接下来,对所述有限多项式进行边界侧(或边界平面)的坐标变换(S3)。基于规定的变化方程式(S4),生成在S3中变换的有限多项式中的不固定系数的关系式。通过调整在S4中提供的未固定系数的数量来生成方矩阵,并且准备将未固定系数作为方形矩阵的多维联立线性方程(S5)。由于可以独立地处理各个元素(小区域),因此通过执行矩阵的跳过压缩处理来执行联立线性方程的压缩和求解计算(S6)。研究S6中提供的近似解的精度,当精度在允许误差范围内时,分析计算结束。在另一种情况下,将处理返回到S1并通过进一步划分小区域来执行分析,或者将处理返回到S2并重置元素多项式解。;版权:(C)2000,JPO

著录项

  • 公开/公告号JP2000339297A

    专利类型

  • 公开/公告日2000-12-08

    原文格式PDF

  • 申请/专利权人 KAWAI TADAHIKO;

    申请/专利号JP19990145546

  • 发明设计人 KAWAI TADAHIKO;

    申请日1999-05-25

  • 分类号G06F17/13;

  • 国家 JP

  • 入库时间 2022-08-22 01:26:30

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号