ICCM Conferences, The 12th International Conference on Computational Methods (ICCM2021)

Font Size: 
Sensitivity analysis of artillery external ballistic based on the approximate high dimensional model representation
Nichen Tong, Qiming Liu, Xu Han

Last modified: 2021-06-10

Abstract


Abstract

In the design of artillery external ballistic, sensitivity analysis can effectively quantify the influence of multi-source uncertain parameters on the dispersion of projectile landing point to improve the precise attack ability of artillery. However, the variance- based sensitivity analysis method is difficult to quantify the artillery muzzle and projectile parameters due to the high-dimensional integral calculation. Therefore, a global sensitivity analysis method based on the approximate high dimensional model representation (AHDMR) is proposed in this paper to measure the influence of these multi-source uncertain parameters. First, the numerical simulation model is constructed based on the external ballistic theory to obtain the relationship between the parameters of artillery muzzle and the dispersion of projectile landing points. Then, with the use of the Hermite orthogonal polynomial, we get the AHDMR of a response function including some Gaussian distributed parameters, and the sensitivity index can be approximately calculated based on the Monte Carlo sampling. The calculation accuracy of the proposed sensitivity analysis method is validated by two numerical examples. Finally, the importance of multi-source parameters is effectively evaluated using the approximate sensitivity analysis method. Sensitivity analysis results show that the ranking of influential parameters is muzzle velocity, projectile diameter, projectile mass, projectile length, horizontal wind speed, launching angle and longitudinal wind speed in turn, which have guiding significance on the optimization design of artillery external ballistic and the improvement of artillery precise attack ability.


An account with this site is required in order to view papers. Click here to create an account.