首页> 外文会议>Applied simulation and modelling >ESTIMATION OF DRAG COEFFICIENT FROM FREE FLIGHT DATA OF AN ARTILLERY ROCKET USING NEURAL NETWORKS
【24h】

ESTIMATION OF DRAG COEFFICIENT FROM FREE FLIGHT DATA OF AN ARTILLERY ROCKET USING NEURAL NETWORKS

机译:基于神经网络的炮弹自由飞行数据拖曳系数估算

获取原文
获取原文并翻译 | 示例

摘要

The proposed method to estimate drag coefficient fromrnflight data of artillery rocket draws its inspiration from thernresearch work done in the area of aerodynamic modelingrnand estimation of aerodynamic coefficients using FeedrnForward Neural Networks (FFNNs). The universalrnmapping capability of FFNNs has been used to buildrnflight dynamic model to predict acceleration for a givenrnset of input variables of an artillery rocket in motion. Therndrag coefficient CD was predicted by suitablerninterpolation and manipulation of the predictedrnacceleration for a given set of motion and atmosphericrnvariables. The proposed method is first validated usingrnsimulated flight data of an artillery rocket. The finalrnvalidation has been carried out using five sets of realrnflight data of a typical artillery rocket in motion. Further,rnthe procedure using Maximum Likelihood (ML) methodrnhas also been applied on real flight data of these artilleryrnrockets to estimate the values of drag coefficient atrndifferent Mach numbers. The estimated values of CDrnobtained through the proposed method and the procedurernbased on ML method have been compared to evaluate thernsuitability of the proposed method. The results show thatrnthe proposed method can advantageously be applied onrntypical flight data of an artillery rocket in motion tornestimate the drag coefficient CD.
机译:从炮弹飞行数据估算阻力系数的方法是从空气动力学建模领域的研究工作以及使用前馈神经网络(FFNN)估算空气动力学系数中获得启发的。 FFNN的通用映射能力已被用于建立飞行动力学模型,以预测运动中的火炮火箭的给定输入变量集的加速度。对于给定的运动变量和大气变量,通过适当的内插和预测的加速度的操纵来预测阻力系数CD。首先使用火炮的模拟飞行数据验证了该方法的有效性。最终验证是通过使用五套正在运行的典型火炮火箭的真实飞行数据进行的。此外,在这些大炮火箭的真实飞行数据上还采用了使用最大似然(ML)方法的程序,以估计在不同马赫数下的阻力系数值。比较了通过该方法和基于ML方法的过程获得的CDrno估计值,以评价该方法的适用性。结果表明,所提出的方法可以有利地应用于炮弹运动中的典型飞行数据,以提高阻力系数CD。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号