The Research Of Transonic Euler Equations Parallel Algorithms And It's Applications

The present thesis solves Euler equations with quick parallel Computing methods on grids. Based on Schwarz parallel algorithms, Collectivity flow field numerical value solver can be gained via evaluating interface boundary cells by passing information. Because Euler equations have disconnected solutions named shock wave in numerical simulation flow field, and to make sure the disconnected solutions not expanding or vanishing among parallel computation sub-domains, there must be certain connections in the numerical value variable among the sub-domains. The present thesis utilizes the fundamental theorem of conservation numerical difference schemes converging to weak solutions, constructing Euler equations numerical parallel algorithms of overlap sub-domains and connected sub-domains, and making sure the numerical parallel value converging to weak solutions of Euler equations, in other words, parallel numerical schemes are conservative. Through derivation the present thesis proves numerical value flux conservation of sub-domains interface boundary can make parallel numerical schemes conservative.After domain decomposition utilizing graph domain decompositions methods, the present thesis carries on sub-domain mesh information reorder, and make it spent little time. Domain decompositions program and sub-domain mesh information reorder programs may be added to Euler equations parallel computing program, so the overall grids information data is only needed to input in the course of Euler equations parallel computation. This text makes the flux of numerical value of interface boundary among sub-domains reach the conservation with a kind of simpler method. Through the fourth chapter numerical computations, it shows that the method can carry out parallel computations without restrictions of the way of butt joint or overlapping decompositions among sub-domains and reaches high parallel efficiency.