基于快速最近邻搜索的固体火箭发动机装药燃面计算

Grain Burnback Analysis of Solid Rocket Motor Using Fast Nearest-Neighbor Search

关键词:  固体火箭发动机  装药  燃面  最小距离函数  计算方法

Key words:  Solid rocket motor  Grain  Burning surface  Minimum distance function  Calculation method

摘要:

燃面计算是固体发动机性能仿真的重要环节，为快速求解固体发动机复杂三维装药燃面推移规律，在柱坐标下进行基于快速最近邻搜索的燃面计算方法研究。利用固体发动机装药对称性，构建柱坐标系下燃面计算框架，实现计算域缩减；采用点云数据代替面元的初始燃面表征和提取，结合k-d树完成散乱点云的有序化组织。在此基础上，提出基于快速最近邻搜索的最小距离函数快速计算方法，通过建立柱坐标下燃去体积通用积分公式，完成了任意燃去厚度时的燃面位置、面积及装药质量特性计算。使用该方法对不同几何构型的装药进行了计算，结果表明，在1s以内的计算耗时下，计算误差均小于1%，说明本方法对于复杂三维装药燃面具有较好通用性、较高精度和计算效率。

Abstract:

The burning surface calculation of complex 3-d grain is important for solid rocket motor（SRM） simulation. Aiming at the problem of fast calculation of SRM burning surface， a calculation method based on fast nearest neighbor search was established in this study. Based on the symmetry of SRM grain， the calculation domain was reduced in cylindrical coordinate system. The initial burning surface was characterized and extracted based on the point cloud data instead of surface element， and the initial point cloud was organized efficiently by k-d tree. Thus， the minimum distance function（MDF） could be fast solved based on the nearest neighbor search method. A general volume integration method in cylindrical coordinates was established to calculate the burning surface position， the burning surface area， and the mass characteristics of grain with arbitrary burning thickness. The method was used to calculate two kinds of geometrical configurations. The results show that the calculation error is less than 1% for a calculation time of less than 1s， indicating that the method has good adaptability for any shaped grain， and has high accuracy and efficiency in calculating the law of burning surface moving.