A mesoscopic damage model of solid propellants under thermo-mechanical coupling loads. (October 2019)

• Record Type: Journal Article
• Authors: Rui, Shi; Boyang, Sun; Xinhui, Liu; Weishang, Yao; Xiaowen, Jia; Huimin, Tan; Wen, Luo; Haijun, Xi; Yu, Chen
• Is Part Of: Polymer testing. Volume 79(2019)
• Publication Date: 2019-10
• Subjects: Stereomicroscope -- In-situ thermo-mechanical coupling load -- Solid propellant -- Quantitative analysis -- Regression analysis -- Artificial neural networks; Polymers -- Testing -- Periodicals; Polymères -- Tests -- Périodiques; 620.1920287
• DOI: 10.1016/j.polymertesting.2019.105927
• Languages: English
• ISSNs: 0142-9418

Abstract

Abstract: Solid propellants are well-studied particle reinforced polymer composite materials that are usually subject to complex stress actions. However, their destructive performances under thermo-mechanical coupling loads have been rarely studied. In the present work, the meso-damage process of nitrate ester plasticized polyether (NEPE) propellant under thermo-mechanical coupling loadings was recorded with a home-made apparatus consisting of an in-situ loading system. The stresses, strains, and crack structures of the propellant at different loading times were measured. The numerical parameters including the crack propagation rate, stress intensity factor, fractal dimension, etc., were calculated by the numerical and quantitative analysis method. Based on the quantitative relationships of these parameters, the failure mechanism of the propellant was proposed. A damage model of the NEPE propellant under different initial and loading conditions was established by regression analysis and artificial neural network modeling and validated with experimental data. Our work has provided an important strategy for studying the meso-damage properties of solid propellants in practical applications. Highlights: The ISTMC-SSM apparatus was designed. The damaging parameters were calculated by the numerical quantitative analysis. The failure mechanism was proposed based on the quantitative relationships. The regression analysis and artificial neural network was used to construct the damage model.