As core components of precision-guided projectiles, projectile-borne components are highly susceptible to failure or even damage in complex high-overload environments, thereby significantly compromising launch reliability and safety. However, accurately characterizing the mechanical behavior of propellants remains challenging due to the limitations in the current internal ballistic theory and the constraints of large-scale artillery firing experiments. This complicates the high-precision numerical modeling of projectile launch, and obstructs investigations into the failure mechanisms of projectile-borne components. Therefore, this paper identifies propellant parameters using the computational inverse method under uncertainty, further establishes high-precision numerical models of projectile launch, and explores the failure mechanisms of projectile-borne components in complex high-overload environments. First, a projectile launching experiment is meticulously designed and executed to obtain the breech pressure and muzzle velocity. Then, a general simulation model is built, and the powder burn model is used to simulate the ignition and combustion. Subsequently, the propellant parameters are effectively identified with the computational inverse method by the combination of the experiments and simulations. A high-precision numerical model of projectile launch is modified with the parameters validated by another experiment, and the high-overload characteristics during projectile launch are thoroughly analyzed based on this model. Finally, the high-overload characteristics of projectile-borne components are analyzed to elucidate the stress variation laws and to reveal the failure mechanisms influenced by time and spatial locations. This research provides an effective method for perfectly identifying propellant parameters and building high-precision numerical models of projectile launch. Additionally, it provides significant guidance for the anti-high overload design and analysis of projectile-borne components.