Algorithmically Obtaining πSURF Initial Burn Rate Table for Novel High Explosives with Minimal Material

We traditionally calibrate hydrodynamic reactive flow models for high explosives by a blend of heuristic and systematic variation of the constants in the analytic equations that describe the reaction rate until the computed response matches experiments. This work shows how we accelerate this process for new explosives through automation of model development using the πSURF reaction model framework. The key feature in this context is the tabular form of the rate. In this work, we leverage this feature and employ an optimization algorithm and automated hydrocode script submission to calibrate the πSURF model for the new explosive formulation based on difurazanopyrazine (DFP). The optimization initiates with a reference rate table, in this case PBX 9501. The automated algorithm proceeds by first comparing the reference Pop plot to the experimentally measured Pop plot of DFP and scaling the reference rate to generate a trial rate; submitting a set of shock initiation problems to the hydrocode using the trial rate; post-processing the resulting reactive growth wave profiles to generate a trial Pop plot; comparing the trial results to the actual and evaluation of a cost function; this process is continued until the cost function criterion is met. This process resulted in a rate table that accurately computes the DFP Pop plot and took less than 12 hours. Here we report the optimization against the Pop plot, we are working on optimizations that include corner turning and curvature data.