The finite-volume time-domain algorithm using least square method in solving Maxwell’s equations

A finite-volume time-domain algorithm using least square method with a well-posed perfectly matched layer (PML) has been developed for the time-domain solution of Maxwell’s equations. This algorithm uses the unstructured grids to obtain good computational efficiency and geometric flexibility. A novelty cell-wise data reconstruction scheme based on least square method is derived to achieve second-order spatial accuracy. A well-posed PML is applied to truncate computational domain by absorbing outgoing electromagnetic waves. The explicit Runge–Kutta scheme is employed to solve the semi-discrete Maxwell’s equations. Several numerical results are presented to illustrate the efficiency and accuracy of the algorithm.