The invention belongs to the field of time sequence anomaly detection, and discloses an aircraft brake system anomaly detection method based on a generative adversarial network and an auto-encoder. The method comprises the following specific steps: (1) preparing a training data set and a test data set of an aircraft brake system anomaly detection model; (2) inputting normal data in the training data set of the aircraft brake system into a generative adversarial network, (3) inputting the normal data, the abnormal data and unmarked data in a training data set of the aircraft brake system into an auto-encoder frame, and obtaining a discriminator capable of distinguishing the normal data and the abnormal data in the aircraft brake system through training; and (4) carrying out anomaly detection on related data by utilizing an aircraft brake system anomaly detection model based on the generative adversarial network and the auto-encoder, and judging whether an anomaly occurs or not according to an anomaly score.
本发明属于时间序列异常检测领域,公开了一种基于生成对抗网络和自编码器的飞机刹车系统异常检测方法。其具体步骤包括:(1)准备飞机刹车系统异常检测模型的训练数据集与测试数据集(2)将飞机刹车系统训练数据集中的正常数据输入生成对抗网络中,通过训练得到能够区分飞机刹车系统中正常数据和异常数据的鉴别器(3)将飞机刹车系统训练数据集中的正常数据、异常数据以及未标记的数据输入自编码器框架中,进一步训练编码器与解码器(4)利用基于生成对抗网络和自编码器的飞机刹车系统异常检测模型对相关数据进行异常检测,根据异常分数判断是否出现异常。
Aircraft brake system anomaly detection method based on generative adversarial network and auto-encoder
基于生成对抗网络和自编码器的飞机刹车系统异常检测方法
2023-10-20
Patent
Electronic Resource
Chinese
IPC: | G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / B64F GROUND OR AIRCRAFT-CARRIER-DECK INSTALLATIONS SPECIALLY ADAPTED FOR USE IN CONNECTION WITH AIRCRAFT , Boden- oder Flugzeugträgerdeckeinrichtungen besonders ausgebildet für die Verwendung in Verbindung mit Luftfahrzeugen / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Anomaly Detection Using Convolutional Neural Network and Generative Adversarial Network
SAE Technical Papers | 2023
|Traffic anomaly detection method based on graph convolutional neural network auto-encoder
European Patent Office | 2023
|Fishing boat trajectory anomaly analysis method based on generative adversarial network
European Patent Office | 2023
|