Research On The Densification Behavior And Density Evolution Law Of PBX Modeling Powder

Polymer bonded explosives(PBXs)are widely used in the field of weapons and ammunition.Densification of molding powders is a key step in the manufacturing process of explosives.The final molding density is a key parameter for the quality of molding because it directly affects PBX mechanics,detonation and safety performance.With the emergence of new formulas,the determination of pressing parameters based on tests has been unable to meet the needs of engineering practice.How to determine parameters in a cost-effective manner based on the characteristics of modeling powders has become an important manufacturing technology topic.The densification behavior of modeling powder and its density evolution during press forming are studied in this work.It is of great significance for evaluating the compressibility of modeling powders and achieving bidirectional prediction between process parameters and expected density.The proposal of this topic is based on extensive research on the status quo at home and abroad.It mainly addresses the following two issues.First,the current research on densification behavior mainly uses the physical/mechanical information of off-line collection and analysis samples after press-forming,and lacks understanding of the behavioral features in the press-molding process;Second,existing work has not yet satisfactorily described the density evolution of PBX molding powder densification,and more focused on the phenomenological effects of molding parameters on final density.There is no effective method to achieve bidirectional prediction of pressing process parameters and final molding density.This study proposes that the densification process of PBX modeling powder is regarded as the density evolution process induced by pressing process parameters.The combination of numerical simulation,experimental testing,and theoretical analysis was used to study the densification behavior and density evolution.The main conclusions and innovations are as follows.Strength parameters,elastic moduli,loading speed,and number of particles all have significant effects on the numerical simulation results of the discrete element modeling process for densification of modeling powders.The engineering-oriented calculations need to be modified based on the macroscopic experimental results.Based on the load-displacement curve measured experimentally,a discrete element numerical simulation parameter system suitable for the densification process was determined.The deformation characteristics of modeling powder densification,such as the evolution of force chain,the evolution of contact pair,and the evolution of rupture number,have been obtained under the premise of matching the numerical simulation output with the macroscopic experimental results.This work enhances the understanding of mechanical behavior in densification,and the mastered numerical simulation method can be used for the study of mechanical behavior of similar materials and working conditions.According to the series of experimental results under different compression molding process parameters,two methods of function fitting and machine learning algorithm(SVM)were separately used to describe the entire process of densification including the three stages of loading,unloading and recovery.In principle,this method can realize the bidirectional prediction calculation between the suppression parameter and the desired density,and provide methods and data for the compressibility evaluation of PBX modeling powder and the determination of pressing process parameters.The mastereing of machine learning algorithm can also be applied to other data mining including but not limited to the mechanics experimental curve data rule.It should be pointed out that there are still deficiencies in this study.For example,in the discrete element numerical simulation of molding powder densification,the phase transition of high-molecular materials such as binders is not taken,into account;And,because of the inconvenience of the actual process parameters and the inherent differences in molding/isostatic pressure,when the model parameters obtained in this study are directly applied to engineering practice,there may be unforeseen errors.