摘要:

AbstractThe Petrov‐Galerkin method is applied to transient convective diffusion problems with convection dominance. Weighting functions are used which differ from the shape functions by both symmetric and non‐symmetric modifications. It is proved that the symmetric modifications largely influence the accuracy. By a one‐dimensional and a two‐dimensional example it is shown how to choose this modification to maximize a

ISSN:0029-5981    

DOI:10.1002/nme.1620191002

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractRegardless of their simplicity, all structures have an infinite number of degrees‐of‐freedom (d.o.f.) when subjected to dynamic loading. The usual finite element method reduces the infinite d.o.f. system to a model with a limited d.o.f. while capturing the significant physical behaviour. The modal analysis reduces the number of d.o.f. further to a limited number of modal co‐ordinates. However, accurate results comparable to the original finite element model may not be possible unless higher modes are included. The present paper is to recommend a response analysis which makes use of both the natural modes and the mass and stiffness matrices of the system to improve the convergence with respect to the number of modes. While the effects of lower modes are analysed similar to the modal analysis, the effects of higher modes are included in the system matrices and the information for higher modes is not n

ISSN:0029-5981    

DOI:10.1002/nme.1620191003

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractThis paper addresses the problem of thermo‐elastic deflection in beams using a two‐dimensional finite elemement idealization of the cross‐section. Bending effects are introduced in a gerneralized plane strain formulation assuming that constant curvature dominates the flexural deformation. The method leads to a plane strain calculation which is very interesting from the computational point of view; thermal deendence of the physical properties can be taken into account, which would be inconceivable in a three‐dimensional

ISSN:0029-5981    

DOI:10.1002/nme.1620191004

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractA numerical technique is developed using an implicit time‐marching scheme to study the deformation history of steel building frames in fire, by considering the change in the material properties due to rise of temperature, effect of creep and large deformations. A computer program has been developed for the analysis of plane frames based on the above algorithm using incremental and iterative procedures. Based on such a study, the length of time for which the structure can perform its function without either local or overall failure can be predicte

ISSN:0029-5981    

DOI:10.1002/nme.1620191005

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractThis paper describes a simple procedure to derive a static infinite element from a linear isoparametric finite element.

ISSN:0029-5981    

DOI:10.1002/nme.1620191006

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractIn this paper the use of strongly stretched finite difference grids is examined in detail for a simple model equation. Numerical solutions of this equation demonstrate the overall second‐order accuracy of the difference approximations. The behaviour of two simple iterative methods, S.O.R. and stationary A.D.I. is discussed for several cases with strong grid stretching and with variable diffusion coefficients. The results show that grid compression near to boundaries can greatly enhance the rates of convergence of these iterative methods, whereas with grid compression in the centre the rate of convergence can be extremely slow. The case of a diffusion coefficient which increases away from the boundaries of a region is analogous to grid compression in the centre and again the convergence rates can degrade considerabl

ISSN:0029-5981    

DOI:10.1002/nme.1620191007

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractA special matrix is introduced, the elements of which are zero or first‐order differential operators. This matrix is used to define a boundary value problem which covers a wide class of engineering applications. An equivalent variational formulation is found using an extension of the Gauss theorem. From this variational principle the equations of the elements are derived. The unified formulae presented here can be useful for educational purposes and for the design of a finite element system for total analysi

ISSN:0029-5981    

DOI:10.1002/nme.1620191008

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractThe Gauss‐Jordan inversion algorithm requires 0(n3) arithmetic operations. In some practical applications, like state estimation or short‐circuit calculation in power systems, the given matrix is sparse and only part of the inverse is needed. Most frequently in the diagonal dominant case this is the diagonal. There are two ways to exploit sparsity to determine elements of the inverse:1Columnwise inversion via the solution of sparse linear systems with columns of the unit matrix as right‐hand sides.2Application of the algorithm of Takahashiet al6.The latter algorithm arises very naturally from multiplying the left‐ and right‐hand factors of the Zollenkopf bifactorization in reverse order. We indicate how computing time can be gained in the important symmetric diagonal dominant case if only part of the diagonal is needed and compare computing times of the method of Takahasiet al.6with several variants of columnwise inversion. Whereas most of the theory holds for general matrices, experiments are performed on the symmetric diagonal dominant case. For band matrices the operation count is 0(n) both for computing a column of the inverse by columnwise inversion and the diagonal by the algorithm of Takaha

ISSN:0029-5981    

DOI:10.1002/nme.1620191009

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractIn this paper we present orsder three, four and five backward product integration methods for the generalized Abel equationWe illustrate their use by silving the small angle deflection problem, and show computationally that these methods are convergent for all ϵ ϵ(0,1). The coefficients of order three and four methods are given in Applenidx

ISSN:0029-5981    

DOI:10.1002/nme.1620191010

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY

摘要:

AbstractCorrelation is a measure of similarity between two functions. When two functions are exactly similar and equal, the correlation between them is maximum. The correlation concept can be used for system design, e.g. the four‐bar mechanism, the slider crank mechanism, etc. The motion generated by four‐bar linkage can be expressed as a function of link dimensions, input and output angles. The product of generated function and desired function gives correlation. The best possible design would be when the deviation of actual correlation from its maximum value (i.e. 100 per cent correlation) is least. This concept, in fact, leads to a modified least squares approach and based on this an objective function is formulated and minimized to give link dimensions. The results obtained are better than the least squares technique, and this is illustrated through different numerical examp

ISSN:0029-5981    

DOI:10.1002/nme.1620191011

出版商:John Wiley&Sons, Ltd    

年代:1983

数据来源: WILEY