基于改进生成对抗网络的飞参数据异常检测方法

doi:10.3785/j.issn.1008-973X.2022.10.008

浙江大学学报(工学版)

2022, Vol. 56

Issue (10): 1967-1976 DOI: 10.3785/j.issn.1008-973X.2022.10.008

自动化技术、信息工程

基于改进生成对抗网络的飞参数据异常检测方法

张鹏1(

),田子都2(

),王浩1

1. 中国民航大学工程技术训练中心，天津 300300
2. 中国民航大学电子信息与自动化学院，天津 300300

Flight parameter data anomaly detection method based on improved generative adversarial network

Peng ZHANG1(

),Zi-du TIAN2(

),Hao WANG1

1. Engineering Training Center, Civil Aviation University of China, Tianjin 300300, China
2. College of Electronic Information and Automation, Civil Aviation University of China, Tianjin 300300, China

全文: PDF(2369 KB) HTML

摘要：

针对民用飞行器安全性、可靠性要求严苛，实际民航运营中飞行参数的异常样本稀少，整体样本不平衡且缺少标注的问题，研究深度学习与生成对抗网络技术，提出基于改进生成对抗网络的飞参数据飞行级异常检测方法. 该方法不依赖样本数量与标签，实现无监督学习的检测方法. 针对飞参数据，输入正常数据样本，应用易收敛的WGAN-GP改进型生成对抗网络模型，模拟生成正常数据样本，计算输入数据与模拟正常数据的巴氏距离，实现对异常数据的检测. 通过美国国家航空航天局模拟飞参数据的人工合成数据集以及真实运营环境下采集的快速存取记录器数据构建的飞参数据集，开展试验验证. 结果表明，与常用无监督模型相比，提出方法在部分异常检测性能指标上有显著提升.

关键词： 异常检测; 生成对抗网络; 飞参数据; 机器学习; 无监督学习

Abstract:

Deep learning and generative adversarial network technology were analyzed aiming at the stringent safety and reliability requirements of civil aircrafts, the scarcity of abnormal samples of flight parameter data in actual civil aviation operations, the imbalance of the overall sample and the lack of labeling. A flight-level anomaly detection method for flight parameter data was proposed based on improved generative adversarial network. The method does not rely on the number of samples and labels, and realizes the detection method of unsupervised learning. Normal data samples were input for flight parameter data. The easy-to-converge WGAN-GP improved generative adversarial network model was applied to simulate normal data samples. The Bhattacharyya distance between the input data and the simulated normal data was calculated, and the detection of abnormal data was realized. The test verification was conducted through the artificial synthesis data set of NASA simulated flight parameter data and the flight parameter data set constructed by the quick access recorder data collected in the real operating environment. Results show that the proposed method has a significant improvement in some anomaly detection performance indicators compared with the commonly used unsupervised model.

Key words: anomaly detection generative adversarial network flight parameter data machine learning unsupervised learning

收稿日期: 2021-10-23 出版日期: 2022-10-25

CLC:

V 328

基金资助: 国家自然科学基金-民航联合基金资助项目（U1733201）

作者简介: 张鹏（1963—），男，教授，从事航空机载设备故障诊断的研究. orcid.org/0000-0002-2210-1262. E-mail： pzhang@cauc.edu.cn

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	作者相关文章
	张鹏
	田子都
	王浩

引用本文:

张鹏,田子都,王浩. 基于改进生成对抗网络的飞参数据异常检测方法[J]. 浙江大学学报(工学版), 2022, 56(10): 1967-1976.

Peng ZHANG,Zi-du TIAN,Hao WANG. Flight parameter data anomaly detection method based on improved generative adversarial network. Journal of ZheJiang University (Engineering Science), 2022, 56(10): 1967-1976.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2022.10.008 或 https://www.zjujournals.com/eng/CN/Y2022/V56/I10/1967

图 1 改进生成对抗网络的结构图

表 1 改进生成对抗网络的结构表

图 2 基于生成对抗网络的飞参数据异常检测方法

表 2 典型异常样本

图 3 各模型生成样本的可视化

图 4 异常检测结果

表 3 提出方法的异常检测性能指标

表 4 异常检测性能指标的对比

图 5 各数据集试验的接受者操作特性曲线与降维分布可视化

图 6 襟翼位置传感器参数的异常检测

图 7 扰流板偏转角度参数的异常检测

图 8 水平安定面状态参数的异常检测

1	CHANDOLA V, BANERJEE A, KUMAR V Anomaly detection: a survey[J]. ACM Computing Surveys, 2009, 41 (3): 1- 58
2	ZHAO W, LI L, ALAM S, et al An incremental clustering method for anomaly detection in flight data[J]. Transportation Research Part C: Emerging Technologies, 2021, 132: 103406 doi: 10.1016/j.trc.2021.103406
3	彭宇, 何永福, 王少军, 等飞行数据异常检测技术综述[J]. 仪器仪表学报, 2019, 40 (3): 1- 13 PENG Yu, HE Yong-fu, WANG Shao-jun, et al Flight data anomaly detection: a survey[J]. Chinese Journal of Scientific Instrument, 2019, 40 (3): 1- 13
4	孙文柱, 曲建岭, 袁涛, 等基于改进SVDD的飞参数据新异检测方法[J]. 仪器仪表学报, 2014, 35 (4): 932- 939 SUN Wen-zhu, QU Jian-ling, YUAN Tao, et al Flight data novelty detection method based on improved SVDD[J]. Chinese Journal of Scientific Instrument, 2014, 35 (4): 932- 939
5	LI L, DAS S, JOHN HANSMAN R, et al Analysis of flight data using clustering techniques for detecting abnormal operations[J]. Journal of Aerospace Information Systems, 2015, 12 (9): 587- 598 doi: 10.2514/1.I010329
6	MELNYK I, YADAV P, STEINBACH M, et al. Detection of precursors to aviation safety incidents due to human factors [C]// 13th IEEE International Conference on Data Mining Workshops. Dallas: IEEE, 2013: 407-412.
7	MELNYK I, BANERJEE A, MATTHEWS B, et al. Semi-Markov switching vector autoregressive model-based anomaly detection in aviation systems [C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco: ACM, 2016: 1065-1074.
8	BREUNIG M M, KRIEGEL H P, NG R T, et al. LOF: identifying density-based local outliers [C]// Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data. New York: ACM, 2000: 93-104.
9	GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al Generative adversarial nets[J]. Communications of the ACM, 2020, 63 (11): 139- 144 doi: 10.1145/3422622
10	SCHLEGL T, SEEBÖCK P, WALDSTEIN S M, et al. Unsupervised anomaly detection with generative adversarial networks to guide Marker discovery [C]// International Conference on Information Processing in Medical Imaging. Cham: Springer, 2017: 146-157.
11	ZENATI H, ROMAIN M, FOO C S, et al. Adversarially learned anomaly detection[C]// 2018 IEEE 18th International Conference on Data Mining Workshops. Singapore: IEEE, 2018: 727-736.
12	YU J, LIN Z, YANG J, et al. Free-form image inpainting with gated convolution [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 4471-4480.
13	陈雪云, 黄小巧, 谢丽基于多尺度条件生成对抗网络血细胞图像分类检测方法[J]. 浙江大学学报: 工学版, 2021, 55 (9): 1772- 1781 CHEN Xue-yun, HUANG Xiao-qiao, XIE Li Classification and detection method of blood cells images based on multi-scale conditional generative adversarial network[J]. Journal of Zhejiang University: Engineering Science, 2021, 55 (9): 1772- 1781
14	XIA B, BAI Y, YIN J, et al LogGAN: a log-level generative adversarial network for anomaly detection using permutation event modeling[J]. Information Systems Frontiers, 2021, 23 (2): 285- 298 doi: 10.1007/s10796-020-10026-3
15	张克明, 蔡远文, 任元基于生成对抗网络的航天异常事件检测方法[J]. 北京航空航天大学学报, 2019, 45 (7): 1329- 1336 ZHANG Ke-ming, CAI Yuan-wen, REN Yuan Space anomaly events detection approach based on generative adversarial nets[J]. Journal of Beijing University of Aeronautics and Astronautics, 2019, 45 (7): 1329- 1336
16	ARJOVSKY M, BOTTOU L. Towards principled methods for training generative adversarial networks [EB/OL]. (2017-01-17)[2022-09-29]. https://arxiv.org/pdf/1701.04862.pdf.
17	ARJOVSKY M, CHINTALA S, BOTTOU L. Wasserstein generative adversarial networks [C]// Proceedings of the 34th International Conference on Machine Learning. Sydney: PMLR, 2017: 214-223.
18	RADFORD A, METZ L, CHINTALA S. Unsupervised representation learning with deep convolutional generative adversarial networks [C]// 4th International Conference on Learning Representations. San Juan: [s. n. ], 2016.
19	GULRAJANI I, AHMED F, ARJOVSKY M, et al. Improved training of Wasserstein GANS [C]// Advances in Neural Information Processing Systems. Long Beach: Curran Associates, 2017: 30.
20	DONAHUE J, KRÄHENBÜHL P, DARRELL T. Adversarial feature learning [C]// 5th International Conference on Learning Representations. Toulon: [s. n.], 2017.
21	DAS S, MATTHEWS B L, SRIVASTAVA A N, et al. Multiple kernel learning for heterogeneous anomaly detection: algorithm and aviation safety case study [C]// Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2010: 47-56.
22	ZHAO Y, NASRULLAH Z, LI Z Pyod: a python toolbox for scalable outlier detection[J]. Journal of Machine Learning Research, 2019, 20 (5): 1- 7
23	RASMUSSEN C E. Gaussian processes in machine learning [C]// Summer School on Machine Learning. Berlin: Springer, 2003: 63-71.
24	VAN DER MAATEN L, HINTON G Visualizing data using t-SNE[J]. Journal of Machine Learning Research, 2008, 9 (11): 2579- 2605

[1]	孙菲,厉小润,赵辽英,余绍奇. 基于FrFT变换和全变分正则化的异常检测算法[J]. 浙江大学学报(工学版), 2022, 56(7): 1276-1284.
[2]	白文超,韩希先,王金宝. 基于条件生成模型的高效近似查询处理框架[J]. 浙江大学学报(工学版), 2022, 56(5): 995-1005.
[3]	何立,庞善民. 结合年龄监督和人脸先验的语音-人脸图像重建[J]. 浙江大学学报(工学版), 2022, 56(5): 1006-1016.
[4]	高一聪,王彦坤,费少梅,林琼. 基于迁移学习的机械制图智能评阅方法[J]. 浙江大学学报(工学版), 2022, 56(5): 856-863, 889.
[5]	屠杭垚,王万良,陈嘉诚,李国庆,吴菲. 结合大气散射模型的生成对抗网络去雾算法[J]. 浙江大学学报(工学版), 2022, 56(2): 225-235.
[6]	杨军,李金泰,高志明. 无监督的三维模型簇对应关系协同计算[J]. 浙江大学学报(工学版), 2022, 56(10): 1935-1947.
[7]	陈彤,郭剑锋,韩心中,谢学立,席建祥. 基于生成对抗模型的可见光-红外图像匹配方法[J]. 浙江大学学报(工学版), 2022, 56(1): 63-74.
[8]	黄发明,潘李含,姚池,周创兵,姜清辉,常志璐. 基于半监督机器学习的滑坡易发性预测建模[J]. 浙江大学学报(工学版), 2021, 55(9): 1705-1713.
[9]	陈雪云,黄小巧,谢丽. 基于多尺度条件生成对抗网络血细胞图像分类检测方法[J]. 浙江大学学报(工学版), 2021, 55(9): 1772-1781.
[10]	任嘉豪,王海鸥,邢江宽,罗坤,樊建人. 湍流火焰切向应变率的低维近似模型[J]. 浙江大学学报(工学版), 2021, 55(6): 1128-1134.
[11]	战友,李强,马啸天,王郴平,邱延峻. 基于宏微观纹理特征融合的路面摩擦性能预测[J]. 浙江大学学报(工学版), 2021, 55(4): 684-694.
[12]	于勇,薛静远,戴晟,鲍强伟,赵罡. 机加零件质量预测与工艺参数优化方法[J]. 浙江大学学报(工学版), 2021, 55(3): 441-447.
[13]	胡惠雅,盖绍彦,达飞鹏. 基于生成对抗网络的偏转人脸转正[J]. 浙江大学学报(工学版), 2021, 55(1): 116-123.
[14]	陈巧红,陈翊,李文书,贾宇波. 多尺度SE-Xception服装图像分类[J]. 浙江大学学报(工学版), 2020, 54(9): 1727-1735.
[15]	刘坤,文熙,黄闽茗,杨欣欣,毛经坤. 基于生成对抗网络的太阳能电池缺陷增强方法[J]. 浙江大学学报(工学版), 2020, 54(4): 684-693.

Viewed

Full text

Abstract

Cited

Shared

Discussed