Soft computing technique to solve second order nonlinear boundary value problems

In this study, an intelligence computational scheme is presented for solving linear and nonlinear singular models using the well-known artificial neural networks (ANNs), genetic algorithm is a global search scheme, active-set is a competent local search scheme and the hybrid of global and local search. The neural network provides convenient approaches to obtain valuable prototypes based on an unsubstantiated error for singular models. The incentive for awarding this research work originates a consistent structure combines with the influential geographies of ANNs to handle the challenges of the singular models. Broad numerical research is accomplished to indorse the convergence, robustness and accuracy of the suggested numerical scheme. The numerical outcomes are also compared with the true results to examine the perfection of the planned numerical structure.