Integrated neuro-swarm heuristic with interior-point for nonlinear SITR model for dynamics of novel COVID-19

The present study is related to present a novel design of intelligent solvers with a neuroswarm heuristic integrated with interior-point algorithm (IPA) for the numerical investigations of the nonlinear SITR fractal system based on the dynamics of a novel coronavirus (COVID-19). The mathematical form of the SITR system using fractal considerations defined in four groups, 'susceptible (S)', 'infected (I)', 'treatment (T)' and 'recovered (R)'. The inclusive detail of each group along with the clarification to formulate the manipulative form of the SITR nonlinear model of novel COVID-19 dynamics is presented. The solution of the SITR model is presented using the artificial neural networks (ANNs) models trained with particle swarm optimization (PSO), i.e., global search scheme and prompt fine-tuning by IPA, i.e., ANN-PSOIPA. In the ANN-PSOIPA, the merit function is expressed for the impression of mean squared error applying the continuous ANNs form for the dynamics of SITR system and training of these networks are competently accompanied with the integrated competence of PSOIPA. The exactness, stability, reliability and prospective of the considered ANN-PSOIPA for four different forms is established via the comparative valuation from of Runge-Kutta numerical solutions for the single and multiple executions. The obtained outcomes through statistical assessments verify the convergence, stability and viability of proposed ANN-PSOIPA. (C) 2021 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Alexandria University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Automated summary

This study presents a novel design of intelligent solvers integrating neuroswarm heuristic and interior-point algorithm for numerical investigations of a nonlinear SITR fractal system based on COVID-19 dynamics. Using artificial neural networks and particle swarm optimization, the study establishes solutions for the SITR model and confirms its accuracy, stability, and reliability in different forms through comparative evaluations. The proposed ANN-PSOIPA shows promising outcomes for modeling the dynamics of the SITR system and training networks efficiently.