Modeling COVID-19 Using a Modified SVIR Compartmental Model and LSTM-Estimated Parameters

This article presents a modified version of the SVIR compartmental model for predicting evolution COVID-19 pandemic, which incorporates vaccination and saturated incidence rate, as well piece-wise time-dependent parameters that enable self-regulation based on epidemic trend. We have established positivity ODE explored its local stability. Artificial neural networks are used to estimate parameters. Numerical simulations conducted using fourth-order Runge–Kutta numerical scheme, results compared validated against actual data from Autonomous Communities Spain. The also includes explicit examine potential future scenarios. In addition, is transformed into system one-dimensional PDEs with diffusive terms, solved finite volume framework fifth-order WENO reconstruction in space an RK3-TVD scheme time integration. Overall, this work demonstrates effectiveness improving our understanding pandemic supporting decision-making public health.