航空发动机

傅莉1 ，崔哲1 ，邓洪伟 2.航空发动机后向RCS统计特性分析方法[J].航空发动机,2024,50(1):72-78

航空发动机后向RCS统计特性分析方法

Statistical Characteristics Analysis Method of Aeroengine Backward RCS

DOI：

英文关键词:Radar Cross Section K-nearest neighbor method kernel density estimation statistical characteristics aeroengine

基金项目:国家自然科学基金（61602321）资助

作者	单位
傅莉1 ，崔哲1 ，邓洪伟 2	1.沈阳航空航天大学自动化学院，沈阳 110136； 2.中国航发沈阳发动机研究所，沈阳 110015

摘要点击次数: 318

全文下载次数: 245

中文摘要:

为解决采用传统固定带宽核密度估计方法分析雷达散射截面（RCS）统计特性时精度低的问题，设计了K最近邻法计算 Epanechnikov核密度估计的动态窗宽。以每个相邻样本的欧氏距离判断样本局部密度，通过样本点与最近邻的距离来调整核函数的窗宽以完成核密度估计，并将其用于发动机后向RCS的统计特性分析。采用改进的Epanechnikov核密度估计与传统核密度估计，对服从固定分布的4种RCS随机样本点的累积概率密度函数进行拟合，以验证算法的精度。结果表明：改进的Epanechnikov 核密度估计的均方根误差比传统核密度估计的分别减小31.2%、38.8%、38.1%、31.9%。结合第2代RCS统计特性分析模型，以 Kolmogorov-Smirnov拟合精度检验为拟合指标，应用改进的Epanechnikov核密度估计计算发动机后向RCS的统计特性并对其规律进行分析可知，对数正态分布更符合C波段和X波段的HH和VV极化的统计特性分布；卡方分布更符合C波段以及Ku波段的HV 和VH极化；威布尔分布更符合X波段的HV、VH极化以及Ku波段的HH、VV极化。

英文摘要:

In order to solve the problem of low accuracy of traditional fixed bandwidth kernel density estimation to analyze the statisti? cal characteristics of Radar Cross Section （RCS）， the K-nearest neighbor method was designed to calculate the dynamic window width of Epanechnikov kernel density estimation. The Euclidean distance of each adjacent sample was used to judge the local density of the sample， and the window width of the kernel function was adjusted by the distance between the sample point and the nearest neighbor to complete the kernel density estimation， which was used to analyze the statistical characteristics of the backward RCS of the engine. The improved Epanechnikov kernel density estimation and the traditional kernel density estimation were used to fit the cumulative probability density function of four kinds of RCS random sample points following a fixed distribution to verify the accuracy of the algorithm. The results show that compared with the traditional kernel density estimation， the root mean square error of the improved Epanechnikov kernel density estimation is reduced by 31.2%， 38.8%， 38.1%， and 31.9%， respectively. Combined with the second-generation statistical characteristic analysis model of RCS， by using the Kolmogorov-Smirnov goodness-of-fit test as the fitting index， and applying the improved Epanechnikov kernel density estimation to calculate the statistical characteristics of engine backward RCS and analyze their regularities， it can be concluded that the lognormal distribution is more consistent with the statistical characteristics of HH and VV polarization in C-band and X-band； the chi-square distribution is more consistent with the HV and VH polarization of C-band and Ku band； the Weibull distribution is more consistent with HV and VH polarization in X-band and HH and VV polarization in Ku band.

查看全文查看/发表评论下载PDF阅读器