
@Article{cmes.2012.087.355,
AUTHOR = {S.Yu. Reutskiy},
TITLE = {A Novel Method for Solving One-, Two- and Three-Dimensional Problems with Nonlinear Equation of the Poisson Type},
JOURNAL = {Computer Modeling in Engineering \& Sciences},
VOLUME = {87},
YEAR = {2012},
NUMBER = {4},
PAGES = {355--386},
URL = {http://www.techscience.com/CMES/v87n4/26838},
ISSN = {1526-1506},
ABSTRACT = {The paper presents a new meshless numerical technique for solving nonlinear Poisson-type equation <i>∇<sup>2</sup>u = f (x) + F(u,x)</i> for <i>x ∈ R<sup>d</sup>, d =1,2,3</i>. We assume that the nonlinear term can be represented as a linear combination of basis functions <i>F(u,x) = ∑<sub>m</sub><sup style="margin-left:-5px">M</sup>q<sub>m</sub>φ<sub>m</sub></i>. We use the basis functions φ<sub>m</sub> of three types: the  the monomials, the trigonometric functions and the multiquadric radial basis functions. For basis functions φ<sub>m</sub> of each kind there exist particular solutions of the equation <i>∇<sup>2</sup>ϕ<sub>m</sub> = φ<sub>m</sub></i> in an analytic form. This permits to write the approximate solution in the form <i>u<sub>M</sub> = u<sub>f</sub> +∑<sub>m</sub><sup style="margin-left:-5px">M</sup>q<sub>m</sub>Φ<sub>m</sub></i>, where <i>Φ<sub>m</sub> = ϕ<sub>m</sub> + ω<sub>m</sub></i>. The term <i>ω<sub>m</sub></i> provides that <i>Φ<sub>m</sub></i> satisfies the homogeneous conditions on the boundary of the domain. Substituting <i>u<sub>M</sub></i> into the equation for <i>F</i>, we transform it to the system of nonlinear equations <i>F(u<sub>M</sub>,x<sub>n</sub>) = ∑<sub>m</sub><sup style="margin-left:-5px">M</sup>q<sub>m</sub>φ<sub>m</sub></i>(x<sub>n</sub>), n = 1,...,M</i> for the unknown coefﬁcients <i>q<sub>m</sub></i>. Then the  nonlinear system is solved numerically. Numerical experiments are carried out for accuracy and convergence investigations. A comparison of the numerical results obtained in the paper with the exact solutions or other numerical methods indicates that the proposed method is accurate and efficient in dealing with complicated geometry and strong nonlinearity.},
DOI = {10.3970/cmes.2012.087.355}
}



