期刊文献+

热分析灰色有限元数学模型 被引量:4

A NEW THERMAL ANALYSIS APPROACH BASED ON GRAY FINITE ELEMENT METHOD
在线阅读 下载PDF
导出
摘要 以区间代数为工具的灰色数学与有限单元模型相结合所形成的灰色有限元数学模型 ,可以充分表达模型输入参数的灰色特性 ,通过求解区间线性方程组可以得出一个用区间数表示的灰色温度场 ,能够更好地符合工程实际需要。考虑到区间代数的扩张性 ,从最佳逼近的意义出发将区间线性方程组转换成极大极小值类型的优化数学模型 ,所推荐使用的基于李兴斯方法的二次优化策略较之通用的序列二次规划算法计算效率更高。文中叙述了温度场灰色有限元模型及其优化模型的建立过程 ,介绍了两种极大极小值优化的求解方法 ,并给出了数值算例。 The finite element method (FEM),with its flexibility in dealing with complex geometries,is an ideal approach to employ in the solution of such problems.Exact values for the parameters of FEM,such as geometrical dimensions,properties of the material,or heat transfer coefficients,should be available to achieve reliable results.In practice,due to the measurement errors and the lack of information,some specific crisp values of them could not be obtained,or considered to be representative for the whole spectrum of possible results. Stimulated by the ideal combination of interval arithmetic and finite difference model presented by the authors,a gray finite element method (gFEM) model is constructed in this paper for transient thermal analysis.Considering the unavoidable interval expanding due to intrinsic characteristics of interval arithmetic,especially when iterating large scale sparse linear equations,we convert the model in the optimal sense to a minimax programming.A twice applied Li method,which tries to substitute the minimax programming to an unconstrained optimization,is found more efficient than the general sequential quadratic programming (SQP).By numeral examples shown at the end of this paper,it is believed that the gFEM model is applicable in practice if the computing time is confirmed to be more acceptable.
出处 《中国电机工程学报》 EI CSCD 北大核心 2002年第3期113-117,共5页 Proceedings of the CSEE
关键词 汽轮机 转子 热分析 灰色理论 有限元 数学模型 tempratue field thermal stresses gray theory interval arithmetic steam turbine
  • 相关文献

参考文献3

二级参考文献2

  • 1张光,博士学位论文,1994年
  • 2张光,中国电机工程学报,1993年,13卷,1期

共引文献146

同被引文献40

引证文献4

二级引证文献95

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部