基于多尺度动态卷积的风切变识别与危害分级算法

doi:10.7527/S1000-6893.2026.33499

本期目录 | 过刊浏览 | 高级检索

前一篇 | 后一篇

基于多尺度动态卷积的风切变识别与危害分级算法

张千千¹,沈淳²,高航²,³,李健兵⁴

1. 国防科技大学电子科学学院复杂电磁环境效应国家重点实验室
2. 国防科技大学电子科学学院
3. 中南大学电子信息学院
4. 国防科技大学

收稿日期:2026-02-10 修回日期:2026-04-23 出版日期:2026-04-27 发布日期:2026-04-27
通讯作者: 李健兵
基金资助:
国家自然科学基金

A Wind Shear Recognition and Hazard Grading Algorithm Based on Multi-scale Dynamic Convolution

Received:2026-02-10 Revised:2026-04-23 Online:2026-04-27 Published:2026-04-27
Contact: Jian-Bing LI
Supported by:
National Natural Science Foundation of China

摘要/Abstract

摘要： 针对传统风切变检测算法因单一阈值和单一匹配模板导致的虚警/漏警问题，本文提出一种基于多尺度动态卷积与自适应权重协同的风切变识别方法。通过设计脉冲响应、渐变响应和长程梯度三种卷积核，构建数据驱动的动态权重融合模型，依据实时特征与历史统计量自适应分配权重，实现风切变位置与危害指数的精准提取。实验结果表明，该算法能有效量化风切变危害，其自适应协同机制显著提升了检测一致性。不同年份、不同机场的实测数据验证了方法的稳定性和可靠性。

关键词: 多尺度动态卷积, 风切变识别, 风切变危害分级, 航空安全, 激光雷达

Abstract: To address the false alarm and missed alarm issues caused by single thresholds and single matching templates in traditional wind shear detection algorithms, this paper proposes a wind shear identification method based on multi-scale dynamic convolution and adaptive weight collaboration. By designing three types of convolution kernels, namely impulse response, gradient response, and long-range gradient, a data-driven dynamic weight fusion model is constructed. The model adaptively allocates weights based on real-time features and historical statistics, achieving precise extraction of wind shear location and hazard index. Experimental results show that this algorithm can effectively quantify wind shear hazards, and its adaptive collaborative mechanism significantly improves detection consistency. Measured data from different years and airports verify the stability and reliability of the method.

Key words: Multi-scale dynamic convolution, Wind shear recognition, Wind shear hazard classification, Aviation safety, Lidar

中图分类号:

张千千沈淳高航李健兵. 基于多尺度动态卷积的风切变识别与危害分级算法[J]. 航空学报, doi: 10.7527/S1000-6893.2026.33499.

参考文献

[1]LINDEN P F, SIMPSON J E.Microbursts: a hazard for aircraft[J].Nature, 1985, 317(17):601-602 [2]HALLOWELL R G, CHO J Y N.Wind shear system cost benefit analysis[J].Lincoln Lab. J., 2010, 18(2):47-68 [3]OCONNOR A, KEARNEY D.Low level turbulence detection for airports[J].International Journal of Aviation Aeronaut and Aerospace, 2019, 6(1):1-24 [4]赵燕勤，陈仁良.低空风切变系统建模及其对直升机飞行安全威胁定性分析[J].航空学报, 2020, 41(07):123413-1 [5]ZHAO Y Q, CHEN R L.Systemically modelling of low-altitude windshear and its qualitative threat analysis to helicopter safety[J].Acta Aeronautica Sinica, 2020, 41(07):123413-1 [6]HUANG T J, TIAN Y H, LI J, et al.Salient Region Detection and Segmentation for General Object Recognition and Image Understanding[J].Science China Information Sciences, 2011, 54(12):2461-2470 [7]沈淳，李健兵，高航等.低空复杂风场全天候雷达精细探测技术[J].电子学报, 2024, 52(04):1189-1204 [8]SHEN C, LI J B, GAO H, et al.Precision Detection of Low-Level Complex Wind Field by Radar Under All-Weather Conditions[J].ACTA ELECTRONICA SINICA, 2024, 52(04):1189-1204 [9]朱上翔.大气扰动及其对飞行的影响[J].航空学报, 1985, 6(2):148-156 [10]ZHU S X.A review of progress in the study of atmospheric disturbance and its effect of flight[J].Acta Aeronautica Sinica, 1985, 6(2):148-156 [11]李健兵，王雪松.分布式软目标雷达特性与感知技术述评[J].雷达学报：2021, 10(1):86-99.[J].雷达学报, 2021, 10(1):86-99 [12]LI J B，WANG X S.Review of Radar Characteristics and Sensing Technologies of Distributed Soft Target[J].Journal of Radars, 2021, 10(1):86-99 [13]ICAO.Meteorological Service for International Air Navigation[R]. Quebec: International Civil Aviation Organization, 2010:1-206. [14]HANG G, CHUN S, XUESONG W, et al.Interpretable semi-supervised clustering enables universal detection and intensity assessment of diverse aviation hazardous winds[J].Nat Commun, 2024, 15(7347):1-12 [15]JAMES W, WILSON N, ANDREW C, et al.Nowcasting Thunderstorms: A Status Report[J].Meteorological Society, 1998, 79(10):2079-2100 [16]WOODFIELD A A, WOODS J F.Worldwide Experience of Wind Shear during 1981–1982[C]. In AGARD Flight Mechanics Panel Conf. on Flight Mechanics and System Design Lessons from Operational Experience, Royal Aircraft Establishment, 1983:28. [17]YUAN J, et al.Microburst,windshear,gust front,and vortex detection in mega airport using a single coherent Doppler wind lidar[J].Remote Sens, 2022, 14(7):1626-1 [18]CHAN P K, HON K K, SHIN D K.Combined Use of Headwind Ramps and Gradients Based on LIDAR Data in the Alerting of Low-level Windshear Turbulence[J].Meteorologische Zeitschrift, 2011, 20(6):661-670 [19]蒋立辉，闫妍，熊兴隆，等.基于斜坡检测的多普勒激光雷达低空风切变预警算法[J].红外与激光工程, 2016, 45(1):0106001-1 [20]JIANG L H, YAN Y, XIONG X L, et al.Doppler lidar alerting algorithm of low-level wind shear based on ramps detection[J].Infrared and Laser Engineering, 2016, 45(1):0106001-1 [21]CHAN P K, KUO M L, LI C M.An Update of the LIDAR Windshear Alerting System (LIWAS) in Hong Kong[C]. 13th Conference on Aviation, Range and Aerospace Meteorology, New Orleans, Louisiana, USA, 2008. [22]LEE Y F, CHAN P W.2014 (21): 86–93.[J].Meteorol Appl, 2014, 21(1):86-93 [23]HON K K, CHAN P K.Application of LIDAR-derived Eddy Dissipation Rate Profiles in Low-level Wind Shear and Turbulence Alters at Hong Kong International Airport[J].Meteorological Applications, 2014, 21(1):74-85 [24]CHAN P W.Latest aviation applications of LIDAR at the Hong Kong International Airport[J].In 15th Conference on Aviation, Range, and Aerospace Meteorology 1–4（American Meteorological Society）, 2011, 1(1):3-12 [25]马愈昭，陈楠，熊兴隆.基于和相位差校正法的风切变预警算法[J].系统工程与电子技术, 2020, 42(1):53-60 [26]MA Y Z, CHEN N, XIONG X L.Wind shear warning algorithm based on PCA and phase difference correction[J].Systems Engineering and Electronics, 2020, 42(1):53-60 [27]陈星，李贞，庄子波，等.侧风激光雷达修正因子的小尺度风切变检测算法[J].光学精密工程, 2018, 26(4):927-935 [28]CHEN X, LI Z, ZHUANG Z B, et al.A small scale windshear detection algorithm of modified F-factor for wind-profiling lidar[J].Optics and Precision Engineering, 2018, 26(4):927-935 [29]蒋立辉, 高志光, 熊兴隆, 等.基于激光雷达图像处理的低空风切变类型识别研究[J].红外与激光工程, 2012, 41(12):3410-3415 [30]JIANG L H, GAO Z G, XIONG X L, et al.Study on type recognition of low attitude wind shear based on lidar image processing[J].Infrared and Laser Engineering, 2012, 41(12):3410-3415 [31]蒋立辉, 陈红, 庄子波, 等.小波不变矩的低空风切变识别[J].红外与激光工程, 2014, 43(11):3783-3787 [32]JIANG L H, CHEN H, ZHUANG Z B, et al.Recognition on low-level wind shear of wavelet invariant moments[J].Infrared and Laser Engineering, 2014, 43(11):3783-3787 [33]蒋立辉, 张春庆, 熊兴隆, 等.基于仿真雷达图像的低空风切变类型识别研究[J].激光与红外, 2013, 43(3):334-338 [34]JIANG L H, ZHANG C Q, XIONG X L, et al.Study on type recognition of low-level wind shear based on radar image simulation[J].Laser and Infrared, 2013, 43(3):334-338 [35]蒋立辉, 陈红, 庄子波, 等.基于小波矩的改进遗传算法风切变识别[J].计算机应用, 2014, 34(3):898-901 [36]JIANG L H, CHEN H, ZHUANG Z B, et al.Wind shear recognition based on improved genetic algorithm and wavelet moment[J].Journal of Computer Applications, 2014, 34(3):898-901 [37]庄子波, 陈星, 高浩, 等.基于概率支持向量机的激光雷达风切变图像的识别[J].北京理工大学学报, 2014, 34(4):412-416 [38]ZHUANG Z B, CHEN X, GAO H, et al.The Recognition of Lidar Wind Shear Images Based on Probability Support Vector Machines[J].Transactions of Beijing Institute of Technology, 2014, 34(4):412-416 [39]熊兴隆, 陈楠, 李永东等, 等.基于卷积神经网络的低空风切变类型识别[J].系统工程与电子技术, 2019, 41(4):772-779 [40]XIONG X L, CHEN N, LI Y D, et al.Type recognition of low-level wind shear based on convolutional neural network[J].Systems Engineering and Electronics, 2019, 41(4):772-779 [41]蒋立辉, 田百全, 熊兴隆, 等.基于多普勒激光雷达低空风切变的数值仿真[J].红外与激光工程, 2012, 41(7):1761-1766 [42]JIANG L H, TIAN B H, XIONG X L, et al.Numerical simulations of low attitude wind shear based on Doppler lidar[J].Infrared and Laser Engineering, 2012, 41(7):1761-1766 [43]姜楠，杨宇.湍流中的多尺度结构及其相对运动[J].科学技术与工程, 2006, 6(20):3254-3258 [44]JIANG N, YANG Y.Multi-scaling Structures and Their Relative Motions in Turbulent Flow[J].Science Technology and Engineering, 2006, 6(20):3254-3258 [45]张千千，沈淳，高航，等.基于激光雷达的斜坡改进风切变预警算法[J].激光与红外, 2023, 53(8):1186-1195 [46]ZHANG Q Q, SHEN C, GAO H, et al.Ramp-improved algorithm in low-level wind shear alerting based on LIDAR[J].Laser and Infrared, 2023, 53(8):1186-1195 [47]HON K K，CHAN P K.Historical analysis (2001–2019) of low-level wind shear at the Hong Kong International Airport[J].Meteorological Applications, 2022, 29(2):1-13

E-mail：hkxb@buaa.edu.cn

关于我们

期刊社服务

专业学科

封面文章

友情链接

主管单位：中国科学技术协会主办单位：中国航空学会北京航空航天大学

基于多尺度动态卷积的风切变识别与危害分级算法

A Wind Shear Recognition and Hazard Grading Algorithm Based on Multi-scale Dynamic Convolution

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

本文评价

[1]	牛国臣, 栾向宇. 基于特征点提取的多传感器融合机坪建图方法[J]. 航空学报, 2024, 45(S1): 730806-730806.
[2]	庄子波, 张春辉, 陈星, 邵靖媛, 陈柏纬. 基于DCGAN的终端区激光雷达晴空湍流识别[J]. 航空学报, 2024, 45(16): 329748-329748.
[3]	刘鹏, 吕曦, 李志忠. 任务复杂度对自动化意识的影响[J]. 航空学报, 2015, 36(11): 3678-3686.
[4]	杭义军, 刘建业, 李荣冰, 孙永荣. 基于混合特征匹配的微惯性/激光雷达组合导航方法[J]. 航空学报, 2014, 35(9): 2583-2592.
[5]	杜小平, 宋一铄, 曾朝阳. 调频连续波激光雷达目标相对距离及径向速度信息提取方法[J]. 航空学报, 2014, 35(2): 523-531.
[6]	李爽;彭玉明;刘宇飞. 星际着陆自主障碍检测与规避技术[J]. 航空学报, 2010, 31(8): 1584-1592.