| H. Wendland, Meshless Galerkin methods using radial basis functions, Preprint, Universitat Gottingen, 1997, to appear in Math. Comp. |
.... X which achieve, due to their geometry, to optimally balance against q X by maximizing their uniformity = q X =hX; We shall say that such center sets are optimally distributed in the domain Note that the results of this paper do also take direct impact on recent meshless Galerkin [32] and collocation methods [10, 12, 13] which are using radial basis functions for solving partial di erential equations (see the papers [11, 30] for an up to date overview) In these applications, each center from the set X can be understood as a method parameter. The choice of the set X is ....
....mentioned so far in this text, in particular the decay condition (2.7) are for instance satis ed by the thin plate splines and Wendland s compactly supported radial basis functions (cf. 29] Local error estimates of a form similar to (2. 2) are also available for both meshless Galerkin methods [32] and collocation methods [12] As to the stability of the interpolation process, estimates on the condition number of the collocation matrix are due to [1, 2, 19, 20] dominated by the magnitude of q X . We close this section by quoting a corresponding result to be found in [29] see also [24, ....
H. Wendland, Meshless Galerkin methods using radial basis functions, Preprint, Universitat Gottingen, 1997, to appear in Math. Comp.
....The purpose of this paper is more speci c. Note that grid free (meshless) methods are novel techniques for numerically solving partial di erential equations. In order to mention two which are related to radial basis functions, recent theoretical developments include meshless Galerkin type methods [15] and collocation methods [7, 8] Their competitiveness in practical applications, however, remains to be shown by numerical comparison with classical and well established methods, such as nite element methods. This contribution makes one step into this direction. We intend to draw the attention ....
H. Wendland, Meshless Galerkin methods using radial basis functions, Math. Comp. 68, 1999, 1521-1531.
....The purpose of this paper is more speci c. Note that grid free (meshless) methods are novel techniques for numerically solving partial di erential equations. In order to mention two which are related to radial basis functions, recent theoretical developments include meshless Galerkin type methods [15] and collocation methods [7, 8] Their competitiveness in practical applications, however, remains to be shown by numerical comparison with classical and well established methods, such as nite element methods. This contribution makes one step into this direction. We intend to draw the attention ....
H. Wendland, Meshless Galerkin methods using radial basis functions, Math. Comp. 68, 1999, 1521-1531.
....now. Liu et al. 50] proposed variants of the SPH method based on the idea of reproducing kernels of higher order and wavelets. There exist also generalizations of the nite di erence ap 2 proach to the gridless setting [49] Furthermore, Kansa [40, 41] Schaback and Franke [28, 29] and Wendland [73] used the radial basis approach from approximation theory to construct meshless methods for the discretization of PDEs. The mass packet method of Yserantant [74, 75] is somewhat di erent from the classical particle methods. Here, the particles are not considered in the sense of statistical ....
....obtain a consistency order 5 of one 3 Besides these weight functions also the thin plate splines, Gaussians and especially the so called SPH spline [11, 26, 31, 51] are used in other meshless methods. Note that circular patch shapes would allow also for radial functions as weight functions W i [28, 29, 40, 41, 73]. 4 Note that, together with collocation and area weighting of the resulting shape functions, this gives basically the smoothed particle hydrodynamics method SPH, which was rst proposed in [51] and further elaborated in [32, 55, 56, 71] 5 Consistency orders are usually given as exponents of ....
H. Wendland, Meshless Galerkin Methods using Radial Basis Functions, Math. of Comput., 68 (1999), pp. 1521-1531.
....space is then used for discretizing the di erential or integral equations. The existence, uniqueness, and convergence proofs in applying the RBFs were given by Micchelli [17] Powell [18] Madych and Nelson [19] for scattered data interpolation and Wu [20] Franke and Schaback [21] and Wendland [22] for solving PDEs respectively. The advantages of RBFs method are truly meshfree, spatial independent and easy to implement. Kansa [23] rst applied this RBFs for solving partial di erential equations (PDEs) of elliptic, parabolic, and hyperbolic types. Hon et al. further extended the use of the ....
H. Wendland, "Meshless Galerkin methods using radial basis functions", Mathematics of Computation, in press.
.... achieve, due to their geometry, to optimally balance hX; against q X by maximizing their uniformity X; q X =hX; We shall say that such center sets are optimally distributed in the domain Note that the results of this paper do also take direct impact on recent meshless Galerkin [32] and collocation methods [10, 12, 13] which are using radial basis functions for solving partial di erential equations (see the papers [11, 30] for an up to date overview) In these applications, each center from the set X can be understood as a method parameter. The choice of the set X is ....
....mentioned so far in this text, in particular the decay condition (2.7) are for instance satis ed by the thin plate splines and Wendland s compactly supported radial basis functions (cf. 29] Local error estimates of a form similar to (2. 2) are also available for both meshless Galerkin methods [32] and collocation methods [12] As to the stability of the interpolation process, estimates on the condition number of the collocation matrix are due to [1, 2, 19, 20] dominated by the magnitude of q X . We close this section by quoting a corresponding result to be found in [29] see also [24, ....
H. Wendland, Meshless Galerkin methods using radial basis functions, Preprint, Universitat Gottingen, 1997, to appear in Math. Comp.
....basis functions has been shown in [9] 10] by the benefits of their corresponding multilevel interpolation scheme. Just very recently, the attraction of radial basis functions has even been enhanced since they were used for the purpose of solving partial differential equations [8] 11] 12] [24]. For all applications mentioned above, the approximation to an unknown function f : R d R, d 1, is for a fixed continuous basis function OE : 0; 1) R and a fixed set X = fx 1 ; xN g ae R d containing pairwise distinct finitely many centre points given by an element s f;X from ....
....the decay condition (2.8) are for instance satisfied by the thin plate splines and Wendland s compactly supported radial basis functions (cf. 22] Local error estimates of the form (2. 2) are also available for meshless Galerkin methods using radial basis functions, as recently developed in [24]. Moreover, generalizations of (2.9) in L p , 1 p 1, are provided by [26] According to [26] the order of the 6 A. Iske Perfect Centres for RBFs above error estimate (2.9) as well as of its generalization in L p for 1 p 1 cannot be improved by further conditions on the native function ....
H. Wendland, Meshless Galerkin methods using radial basis functions, Preprint, Universitat Gottingen, 1997, to appear in Math. Comp.
....mesh free MQ RBFs for solving various initial and boundary values problems. The existence, uniqueness, and convergence proofs in applying the RBFs were given by Micchelli [27] Powell [28] Madych and Nelson [29] for scattered data interpolation and Wu [30] Franke and Schaback [31] and Wendland [32] for solving PDEs respectively. In the papers two important features of the RBFs method had been observed: 1) it is a truly mesh free algorithm; and (2) it is spatial dimension independent in the sense that the convergence order is of O(h d 1 ) one of the esti5 mates for MQ) where h is the ....
.... by assuming that the representation (28) satis es the given partial di erential equations (15) 18) From the boundary conditions (20) 23) a unique set of the undetermined coecients j is obtained by solving the resultant system of equations (refer Franke and Schaback [31] and Wendland [32] for the theoretical foundation of this method) In practice, among all the commonly used RBFs, the MQ RBF: j (x) k x x j k 2 2 j ) 1=2 (29) is proven to give the most accurate numerical approximation (refer Franke [15] and hence is used in the present computations. Madych and ....
H. Wendland, 'Meshless Galerkin methods using radial basis functions', Mathematics of Computation, in press.
....the emerging meshless techniques for solving differential equations, since it only requires the information about the location of the points Q, which need not be interconnected. Meshless methods are a challenging topic with the potential to become a feasible alternative to finite element methods [27]. In this paper we will furnish convergence order estimates for the collocation problem explicitly in terms of a mesh norm h that measures the density Collocation with Zonal Kernels on Spheres 3 of the points Q. The crux of our approach is to transform the collocation problem to a Lagrange ....
Wendland, H., Meshless Galerkin methods using radial basis functions, Math. Comp. 68 (1999), 1521--1531.
....Liu et al. 49] proposed variants of the SPH method based on the idea of reproducing kernels of higher order and wavelets. There exist also gener alizations of the finite difference approach to the gridless setting [48] Furthermore, Kansa [39, 40] Schaback and Franke [27, 28] and Wendland [71] used the radial basis approach from approximation theory to construct meshless methods for the discretization of PDEs. The mass packet method of Yserantant [72, 73] is somewhat different from the classical particle methods. Here, the particles are not considered in the sense of statistical ....
....choice of weight function, particle 3 Besides these weight functions also the thin plate splines, Gaussians and especially the so called SPH spline [11, 25, 30, 50] are used in other meshless methods. Note that circular patch shapes would allow also for radial functions as weight functions W i [27, 28, 39, 40, 71]. 4 Note that, together with collocation and area weighting of the resulting ansatz functions, this gives basically the smoothed particle hydrodynamics method SPH, which was first proposed in [50] and further elaborated in [31, 54, 55, 70] distribution and topology of the cover. Thus, we ....
H. Wendland, Meshless Galerkin Methods using Radial Basis Functions. to appear in Math. of Comput., 1997.
.... 55, 61, 66 68] especially as tools for collocation [32, 33] and the Dual Reciprocity Method (DRM) 3, 4, 5, 18, 20, 30, 31, 40, 41] Unfortunately, the mathematical theory of RBF [35 38] lagged back behind the numerical applications to PDE for quite some time, but recently there was some progress [15, 59] towards a solid underpinning of numerical algorithms using RBF s for solving PDE s. We shall provide a short account of such results. Furthermore, the construction of compactly supported radial basis functions (CSRBF) by Wendland, Wu and Schaback [46, 56, 62] based on a toolkit by Wu and ....
....support scales on data subsets of dioeerent densities. In many cases, linear convergence with increasing levels is observed, but the theoretical investigations are still rather limited [11, 22, 42] 5 RayleighRitz applications For a bounded domain Omega with C 1 boundary Omega Wendland [59] considers problems of the form Gamma d X i;j=1 x i a ij u x j (x) c(x)u(x) f(x) x 2 Omega (3) d X i;j=1 a ij (x) u(x) x j i (x) h(x)u(x) g(x) x 2 Omega (4) where a ij ; c 2 L1( Omega Gamma , i; j = 1 : n, f 2 L 2( Omega Gamma , a ij ; h 2 L1 ( ....
H. Wendland, Meshless Galerkin methods using radial basis functions, to appear in Mathematics of Computation.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC