Modeling and optimization of shaped charge liner collapse and jet formation

This research project concentrates on the modeling and optimization of shaped charge liner collapse and jet formation. The research has produced an improved analytical model of shaped charge liner collapse and subsequent jet formation. The improved model has been used to investigate the direct distributed parameter optimization of a nonlinear dynamic system using variable metric nonlinear programming. This investigation consisted of optimizing the liner contour required to produce a desired jet profile in the least squares sense. Additionally, a very practical discrete optimization problem was chosen so that continuum modeling and experimentation could be used to verify the optimization results.; Analytical shaped charge mathematical models have proven very useful in the prediction of shaped charge characteristics while requiring minimal computer time and memory, as well as conservative user effort. Thus, parametric investigation using analytical shaped charge modeling is both practical and economically feasible. Unfortunately, current analytical models rely on many empirical relationships and do not incorporate constitutive relationships for the various materials. Thus, in order to predict shaped charge characteristics over a broader range of materials and geometries, a more fundamental analytical model has been developed.; When parametric investigation becomes practical, the question of parametric optimization naturally arises. The process of shaped charge liner collapse and jet formation is normally modeled as a distributed parameter nonlinear dynamic system. Optimization of distributed parameter systems has traditionally been very difficult. Nonlinear dynamic system behavior adds more complication to an already difficult optimization problem. Recent advances of numerical methods for discrete parametric optimization provide dramatic increases in efficiency and reliability. The application of these new variable metric sequential quadratic programming algorithms to distributed parameter nonlinear dynamic systems should prove to be a significant contribution to the technology base.