A new automated Chimera method for the prediction of store trajectory

A new automated Chimera method has been developed for the prediction of store trajectory. The method uses a highly robust CFD method and our cut-paste algorithm for the Chimera domain decomposition to enhance the efficiency of store trajectory simulation. The time-step size limitation due to the grid movement is relaxed with the interpolation of solutions in the previous time step. A treatment for orphan cells is also devised to improve the robustness of method. Computational results show that the current method is capable of simulating the store separation without user interruption and that a timestep size of 0.005 set, which is more than three times larger than ever reported, can be used for the Eglin Wing/Pylon/Store separation problem. Introduction Since the Chimera grid technique’ was introduced to CFD(Computational Fluid Dynamics) community, it is gaining its popularity not only because the grid generation over complex geometries is relatively simple compared to the multi-block grid generation technique but because the Chimera method is the most successful method in handling relative movement of multiple bodies. The Chimera grid technique is naturally suitable for moving body problems. One of moving body problems frequently encountered in aerodynamic applications is the trajectory prediction of a store separating from mother aircraft. There have been several reports that successfully predict the trajectory of the well-known Eglin Wing/Pylon/Store problem234 .5 Recently, the simulations of ripple release of multiple stores have been reported6.’ However, the trajectory prediction with CFD is not a routine engineering application. The ‘Senior Researcher tPrincipa1 Researcher, Member AIAA z Doctoral candidate 5 Associate Professor, Senior Member AIAA. copyright @ lgqg by the American Institute of Aerondcs and Astronautics, IX. AU tights reserd. major stumbling block against the goal comes from the fact that most Chimera methods rely on interactive approaches for the Chimera hole cutting8.g The computations, furthermore, are often hindered by the lack of the efficiency and the robustness of the CFD tools. In this paper, we present our efforts toward the goal. Our efforts include implementing an automated Chimera hole cutting method, cut-paste algorithm,‘O combined with two step donor cell search. Cut-paste algorithm is the advancing front technique for the Chimera hole cutting. The solid walls or non-penetrable boundaries are used as the initial hole-cutting boundary so that the user input is minimized. With this method, the Chimera hole cutting can be performed automatically for unsteady applications. Also, two step donor cell search ensures fast and reliable search. As stated by Lijewski,4 the computation often fails due to the presence of orphan cells. A proper handling of orphan cells without loss of accuracy is a “must” for full unsteady simulations. Furthermore, the time-step size is limited by the presence of the points which are initially hole points and become normal points as the grid moves without being updated. We devise methods to handle such problems to enhance the robustness of our Chimera method. In addition, a highly robust CFD method” is used for aerodynamic forces and moment calculations. The object of this paper is to present our attempt to enhance the efficiency and the robustness of the computational method and the Chimera method so that the store prediction method can be used as an engineering tool in near future. The trajectory computations of Eglin Wing/Pylon/Store and ripple release from TER (Tripple Ejector Rack) are presented showing the current method satisfies the objective of our study. . Chimera Hole Cutting The existing Chimera methods need user specified hole cutting surfaces for the Chimera hole cutting. In PEGASUS,8 for example, the user must specify the hole cutting-surfaces as well as the hole grid, while in