Solving nonlinear pseudoparabolic equations with nonlocal boundary conditions in reproducing kernel space (Q735066)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Solving nonlinear pseudoparabolic equations with nonlocal boundary conditions in reproducing kernel space |
scientific article |
Statements
Solving nonlinear pseudoparabolic equations with nonlocal boundary conditions in reproducing kernel space (English)
0 references
14 October 2009
0 references
Nonlinear pseudoparabolic equations with nonlocal boundary conditions arise in various physical phenomena, for example, they describe the dynamics of ground moisture. A spatial one-dimensional problem of this type is investigated in this paper on the domain \(D=\{(x,t)\in \mathbb R^2: 0<x<1\), \(0<t<T\), \(T>0\}\). The authors expand previous results concerning this type of problems from a computational point of view, more precisely, a numerical scheme for solving such a problem is proposed. First, the original problem is transformed to a problem with homogeneous boundary and initial conditions, which is necessary for constructing so called reproducing kernel spaces,on which the proposed numerical method is based. The reproducing kernel spaces for each, time and space variable functions are constructed and then the construction of the reproducing kernel space \(W(D)\) for functions with two (time and space) variables is done. Then the boundedness of the linear operator arising from the transformed problem on this function space is proved. Using the properties of such linear operator and reproducing kernel functions, a complete countable system of functions is constructed in \(W(D)\) . Then the solution of equivalent problem to the original one can be expressed as a countable linear combination of this system with known coefficients. The proposed numerical method is based on the approximation of this solution using a finite truncation of this combination. The uniform convergence of the numerical solution and its derivatives is proved. Finally a numerical example using the proposed new numerical method is presented.
0 references
nonlocal boundary conditions
0 references
reproducing kernel space
0 references
convergence
0 references
numerical example
0 references
0 references
0 references
0 references
0 references