Radial Basis Function Artificial Neural-Network-Inspired Numerical Solver

A framework for a mesh-free numerical solver of differential equations is presented in this paper. Development of the solver is derived from machine learning techniques using artificial neural networks with Gaussian radial basis functions for their neurons. The proposed method incrementally develops an approximation through the optimization of a scalar condensed form of the differential equations. Unlike traditional solvers that require grids, volumes, or meshes, along with corresponding connectivity data, the proposed framework requires only a list of independent variable values to approximate the solution. Because of this, there is no need for the derivation or inversion of system matrices. Results are presented demonstrating the stability and accuracy of the proposed method and it is demonstrated that the spatial error estimate can exceed that of traditional methods.

[1]  Eugenio Oñate,et al.  Derivation of stabilized equations for numerical solution of advective-diffusive transport and fluid flow problems , 1998 .

[2]  Xiong Zhang,et al.  Meshless methods based on collocation with radial basis functions , 2000 .

[3]  L. Lucy.  A numerical approach to the testing of the fission hypothesis. , 1977 .

[4]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[5]  I. Babuska,et al.  The partition of unity finite element method: Basic theory and applications , 1996 .

[6]  T. Chung.  Computational Fluid Dynamics: FOUR. AUTOMATIC GRID GENERATION, ADAPTIVE METHODS, AND COMPUTING TECHNIQUES , 2002 .

[7]  Robert Vertnik,et al.  Meshfree explicit local radial basis function collocation method for diffusion problems , 2006, Comput. Math. Appl..

[8]  Jooyoung Park,et al.  Universal Approximation Using Radial-Basis-Function Networks , 1991, Neural Computation.

[9]  V. Girault,et al.  Theory of a Finite Difference Method on Irregular Networks , 1974 .

[10]  Antony Jameson,et al.  Meshless Scheme Based on Alignment Constraints , 2010 .

[11]  S. Atluri,et al.  A new Meshless Local Petrov-Galerkin (MLPG) approach in computational mechanics , 1998 .

[12]  T. Belytschko,et al.  Element‐free Galerkin methods , 1994 .

[13]  Darrell Pepper.  Meshless methods for PDEs , 2010, Scholarpedia.

[14]  Andrew J. Meade,et al.  Approximation properties of local bases assembled from neural network transfer functions , 1998 .

[15]  Quan Shen.  Local RBF-based differential quadrature collocation method for the boundary layer problems , 2010 .

[16]  A. Jameson.  ANALYSIS AND DESIGN OF NUMERICAL SCHEMES FOR GAS DYNAMICS, 1: ARTIFICIAL DIFFUSION, UPWIND BIASING, LIMITERS AND THEIR EFFECT ON ACCURACY AND MULTIGRID CONVERGENCE , 1995 .

[17]  H. N. Mhaskar,et al.  Neural Networks for Optimal Approximation of Smooth and Analytic Functions , 1996, Neural Computation.

[18]  Ting-Zhu Huang,et al.  The inverses of block tridiagonal matrices , 2006, Appl. Math. Comput..

[19]  Mehdi Dehghan,et al.  Some implicit methods for the numerical solution of Burgers' equation , 2007, Appl. Math. Comput..

[20]  Kurt Hornik,et al.  Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks , 1990, Neural Networks.

[21]  John C. Slater,et al.  Electronic Energy Bands in Metals , 1934 .

[22]  B. Nayroles,et al.  Generalizing the finite element method: Diffuse approximation and diffuse elements , 1992 .

[23]  Guirong Liu.  Meshfree Methods: Moving Beyond the Finite Element Method, Second Edition , 2009 .

[24]  E. Kansa.  Multiquadrics—A scattered data approximation scheme with applications to computational fluid-dynamics—I surface approximations and partial derivative estimates , 1990 .

[25]  C. Shu,et al.  Computation of Incompressible Navier-Stokes Equations by Local RBF-based Differential Quadrature Method , 2005 .