Implementation of the QGD Algorithm Using AMR Technology and GPU Parallel Computing

The paper presents an algorithm based on the quasi-gasdynamic approach for the solution of unsteady compressible flows over a wide range of Mach numbers. It is implemented on the AMReX open platform, which uses adaptive mesh refinement technology to facilitate parallelization of computations on GPU architectures. To validate its effectiveness, the developed solver is applied to the numerical simulation of the shock-vortex interaction problem with flow parameter values of \(M_v = 0.9\) and \(M_s = 1.5\) . Cross-validation to assess its performance is conducted with OpenFOAM-based solvers, specifically rhoCentralFoam and QGDFoam. Schlieren fields are used to evaluate oscillations of the numerical schemes and algorithms, while resolution capabilities of the algorithm are assessed by comparing density fields in five cross-sections with the reference values.