Exponential quadrature rules without order reduction
For researchers in numerical analysis, this provides a method to maintain high-order time integration for PDEs with time-dependent boundary conditions, addressing a known bottleneck of order reduction.
The paper proposes a technique for integrating linear initial boundary value problems with exponential quadrature rules to achieve high-order temporal accuracy, demonstrating order 2s with s Gaussian nodes and avoiding order reduction.
In this paper a technique is suggested to integrate linear initial boundary value problems with exponential quadrature rules in such a way that the order in time is as high as possible. A thorough error analysis is given for both the classical approach of integrating the problem firstly in space and then in time and of doing it in the reverse order in a suitable manner. Time-dependent boundary conditions are considered with both approaches and full discretization formulas are given to implement the methods once the quadrature nodes have been chosen for the time integration and a particular (although very general) scheme is selected for the space discretization. Numerical experiments are shown which corroborate that, for example, with the suggested technique, order $2s$ is obtained when choosing the $s$ nodes of Gaussian quadrature rule.