Benchmarking of different Multi-pass Shake-the-Box strategies

Within this study several advanced Multi-pass Shake-the-Box processing strategies are compared to the standard single-pass approach and are evaluated based on their reconstruction quality of two time-resolved experimental flow cases. The covered processing strategies are Forward-Backward tracking, Variable Time-step Shake-the-Box (VT-STB) and the Extended Initialization approach, which is proposed for the first time within this work. All multi-pass approaches improve the trajectory reconstruction results, when compared to the single-pass processing. From the considered multi-pass strategies the Forward-Backward tracking shows the most universal improvements and it is recommended to always combine it with the other approaches. The proposed superior reconstruction results of VT-STB for flows with a large dynamic velocity range can be confirmed within this study. For the Extended Initialization strategy it is found, that it enhances the reconstruction quality during the critical convergence phase but these improvements cannot be carried over completely along the complete time series. It is therefore recommended for high seeding density cases, where the calculation time for full Forward-Backward tracking schemes might be to expensive.