Integrated Radial Basis Function Networks Research Articles

Integrated High Accuracy Multiquadric Quasi-interpolation Scheme For Solving The Nonlinear Klein-gordon Equation

A collocation scheme based on the use of the multiquadric quasi-interpolation operator 2 W L , integrated radial basis function networks (IRBFNs) method and three order finite difference method is applied to the nonlinear Klein-Gordon equation. In the present scheme, the three order finite difference method is used to discretize the temporal derivative and the integrated form of the multiquadric quasi-interpolation scheme is used to approximate the unknown function and its spatial derivatives. Several numerical experiments are provided to show the efficiency and the accuracy of the given method.

Local moving least square-one-dimensional integrated radial basis function networks technique for incompressible viscous flows

This paper presents a local moving least square-one-dimensional integrated radial basis function networks method for solving incompressible viscous flow problems using stream function-vorticity formulation. In this method, the partition of unity method is employed as a framework to incorporate the moving least square and one-dimensional integrated radial basis function networks techniques. The major advantages of the proposed method include the following: (i) a banded sparse system matrix which helps reduce the computational cost; (ii) the Kronecker-i property of the constructed shape function which helps impose the essential boundary condition in an exact manner; and (iii) high accuracy and fast convergence rate owing to the use of integration instead of conventional differentiation to construct the local radial basis function approximations. Several examples including two-dimensional (2D) Poisson problems, lid-driven cavity flow and flow past a circular cylinder are considered, and the present results are compared with the exact solutions and numerical results from other methods in the literature to demonstrate the attractiveness of the proposed method.

Free vibration analysis of laminated composite plates based on FSDT using one-dimensional IRBFN method

This paper presents a new effective radial basis function (RBF) collocation technique for the free vibration analysis of laminated composite plates using the first order shear deformation theory (FSDT). The plates, which can be rectangular or non-rectangular, are simply discretised by means of Cartesian grids. Instead of using conventional differentiated RBF networks, one-dimensional integrated RBF networks (1D-IRBFN) are employed on grid lines to approximate the field variables. A number of examples concerning various thickness-to-span ratios, material properties and boundary conditions are considered. Results obtained are compared with the exact solutions and numerical results by other techniques in the literature to investigate the performance of the proposed method.

Integrated-RBF network method for free vibration analysis of laminated composite plates

This paper presents a new effective radial basis function (RBF) collocation technique for the free vibration analysis of laminated composite plates using the first order shear deformation theory (FSDT). The plates, which can be rectangular and non-rectangular, are simply discretised by means of Cartesian grids. Instead of using conventional differentiated RBFs, integrated RBF networks [1,2,3] are employed on grid lines to approximate the field variables. Several test problems with different geometries, boundary conditions, thickness to length ratios and material properties are considered. The obtained numerical results are in good agreement with existing published results and exact solutions. Numerical results show that the proposed technique is very stable and converges fast with grid refinement.

A Cartesian-grid integrated-RBF method for viscoelastic flows

This paper is concerned with the use of integrated radial-basis-function networks (IRBFNs), Cartesian-grids and point collocation for numerically solving 2D flows of viscoelastic fluids. In the proposed method, RBFNs, which are constructed through integration, are employed to approximate the field variables including stresses, and the governing equations are discretised by means of point collocation. Advantages gained from the integration process over differentiated RBFNs process are (i) better accuracy (ii) more straightforward implementation of boundary conditions and (iii) more stable solutions. The method is verified through several test problems, e.g. planar Poiseuille flow and fully-developed flow in a square duct, with different types of viscoelastic fluids, namely Oldroyd-B and CEF. The obtained results show that (i) The proposed method is easy to implement; (ii) Its pre-processing is simple and (iii) Accurate results are obtained using relatively-coarse grids.

A moving IRBFN-based integration-free meshless method

A novel approximation method using integrated radial basis function networks (IRBFN) coupled with moving least square (MLS) approximants, namely moving integrated radial basis function networks (MIRBFN), is proposed in this work. In this method, the computational domain w is divided into finite sub-domains w which satisfy point-wise overlap condition. The local function interpolation is constructed by using IRBFN supported by all nodes in subdomain w. The global function is then constructed by using Partition of Unity Method (PUM), where MLS functions play the role of partition of unity. As a result, the proposed method is locally supported and yields sparse and banded interpolation matrices. The computational efficiency are excellently improved in comparison with that of the original global IRBFN method. In addition, the present method possesses the Kronecker-d property, which makes it easy to impose the essential boundary conditions. The proposed method is applicable to randomly distributed datasets and arbitrary domains. In this work, the MIRBFN method is implemented in the collocation of a first-order system formulation to solve PDEs governing various problems including heat transfer, elasticity of both compressible and incompressible materials, and linear static crack problems. The numerical results show that the present method offers high order of convergence and accuracy.

Numerical solution of PDEs via integrated radial basis function networks with adaptive training algorithm

This paper develops a mesh-free numerical method for solving PDEs, based on integrated radial basis function networks (IRBFNs) with adaptive residual subsampling training scheme. The multiquadratic function is chosen as the transfer function of the neurons. The nonlinear algebraic equation systems for weights training are solved by Levenberg–Marquardt algorithm. The performance of the proposed method is demonstrated in numerical examples by approximating several functions and solving nonlinear PDEs. The result of numerical experiments shows that the IRBFNs with the adaptive procedure requires less neurons to attain the desired accuracy than conventional radial basis function networks.

A Galerkin approach incorporating integrated radial basis function networks for the solution of 2D biharmonic equations

This paper is concerned with the use of integrated radial basis function networks (IRBFNs) for the discretisation of Galerkin approximations for Dirichlet biharmonic problems in two dimensions. The field variable is approximated by global high-order IRBFNs on uniform grids without suffering from Runge's phenomenon. Double boundary conditions, which can be of complicated shapes, are both satisfied identically. The proposed technique is verified through the solution of linear and nonlinear problems, including a benchmark buoyancy-driven flow in a square slot. Good accuracy and fast convergence are obtained.

A Cartesian-Grid Discretisation Scheme Based on Local Integrated RBFNs for Two-Dimensional Elliptic Problems

This paper reports a new numerical scheme based on Cartesian grids and local integrated radial-basis-function networks (IRBFNs) for the solution of second-order elliptic differential problems defined on two-dimensional regular and irregular domains. At each grid point, only neighbouring nodes are activated to construct the IRBFN approximations. Local IRBFNs are introduced into two different schemes for discretisation of partial differential equations, namely point collocation and control-volume (CV)/subregion-collocation. Linear (e.g. heat flow) and nonlinear (e.g. lid-driven triangular-cavity fluid flow) problems are considered. Numerical results indicate that the local IRBFN CV scheme outperforms the local IRBFN point-collocation scheme regarding accuracy. Moreover, the former shows a similar level of the matrix condition number and a significant improvement in accuracy over a linear CV method.

A multidomain integrated‐radial‐basis‐function collocation method for elliptic problems

AbstractThis article is concerned with the use of integrated radial‐basis‐function networks (IRBFNs) and nonoverlapping domain decompositions (DDs) for numerically solving one‐ and two‐dimensional elliptic problems. A substructuring technique is adopted, where subproblems are discretized by means of one‐dimensional IRBFNs. A distinguishing feature of the present DD technique is that the continuity of the RBF solution across the interfaces is enforced with one order higher than with conventional DD techniques. Several test problems governed by second‐ and fourth‐order differential equations are considered to investigate the accuracy of the proposed technique. © 2008 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq 2008

A Cartesian grid technique based on one‐dimensional integrated radial basis function networks for natural convection in concentric annuli

AbstractThis paper reports a radial basis function (RBF)‐based Cartesian grid technique for the simulation of two‐dimensional buoyancy‐driven flow in concentric annuli. The continuity and momentum equations are represented in the equivalent stream function formulation that reduces the number of equations from three to one, but involves higher‐order derivatives. The present technique uses a Cartesian grid to discretize the problem domain. Along a grid line, one‐dimensional integrated RBF networks (1D‐IRBFNs) are employed to represent the field variables. The capability of 1D‐IRBFNs to handle unstructured points with accuracy is exploited to describe non‐rectangular boundaries in a Cartesian grid, while the method's ability to avoid the reduction of convergence rate caused by differentiation is instrumental in improving the quality of the approximation of higher‐order derivatives. The method is applied to simulate thermally driven flows in annuli between two circular cylinders and between an outer square cylinder and an inner circular cylinder. High Rayleigh number solutions are achieved and they are in good agreement with previously published numerical data. Copyright © 2007 John Wiley & Sons, Ltd.

Computation of Laminated Composite Plates using Integrated Radial Basis Function Networks

This paper reports a meshless method, which is based on radial-basis-function networks (RBFNs), for the static analysis of moderately-thick laminated composite plates using the first-order shear deformation theory. Integrated RBFNs are employed to represent the field variables, and the governing equations are discretized by means of point collocation. The use of integration rather than conventional differentiation to construct the RBF approximations significantly stabilizes the solution and enhances the quality of approximation. The proposed method is verified through the solution of rectangular and non-rectangular composite plates. Numerical results obtained show that the method achieves a very high degree of accuracy and a fast convergence rate.

A New High-order Time-kernel BIEM for the Burgers Equation

This paper presents a new high-order time-kernel boundary-integral-equation method (BIEM) for numerically solving transient problems governed by the Burgers equation. Instead of using high-order Lagrange polynomials such as quadratic and quartic interpolation functions, the proposed method employs integrated radial-basis-function networks (IRBFNs) to represent the unknown functions in boundary and volume integrals. Numerical implementations of ordinary and double integrals involving time in the presence of IRBFNs are discussed in detail. The proposed method is verified through the solution of diffusion and convection-diffusion problems. A comparison of the present results and those obtained by low-order BIEMs and other methods is also given.