AEROSPACE SHANGHAI, Volume. 42, Issue 2, 157(2025)
Online Anomaly Detection for Servo Systems with Generative Recurrent Networks
Figures & Tables(9)
Fig. 4. Visualization of the control command and telemetry variable data for a servo system
Fig. 5. Comparison of the prediction performance of different online learning methods
Fig. 6. Comparison of the prediction performance of different methods
Table 1.
Comparison of the prediction performance of offline training and online learning View tableView in ArticleTable 1.
Comparison of the prediction performance of offline training and online learning 项目 数据集
2
数据集3 数据集4 数据集5 数据集6 数据集7 数据集8 离线
训练
0.025 7 0.013 8 0.009 5 0.013 4 0.021 2 0.018 6 0.012 5 在线
学习
0.011 8 0.005 7 0.008 7 0.013 2 0.012 2 0.007 5 0.005 9
Table 2.
Comparison of the details of different online learning methods View tableView in ArticleTable 2.
Comparison of the details of different online learning methods 项目 模型参数初始化方式 在线学习训练样本 方 式1 上一个模型参数 流数据+回顾数据 方 式2 上一个模型参数 流数据 方 式3 随机初始化 流数据
Table 3.
Comparison of the anomaly detection accuracy of different algorithms View tableView in ArticleTable 3.
Comparison of the anomaly detection accuracy of different algorithms 项目 数据集1 数据集2 数据集3 数据集4 数据集5 数据集6 数据集7 数据集8 本文算法 0.804 9 0.750 5 0.786 4 0.763 8 0.804 1 0.774 2 0.806 7 0.756 1 对比算法 0.798 2 0.772 1 0.688 0 0.657 3 0.856 8 0.800 1 0.813 4 0.748 0 性能差异 +0.006 7 -0.021 6 +0.098 4 +0.106 5 -0.052 7 -0.025 9 -0.006 7 +0.008 1
Tools
Get Citation
Copy Citation Text
Xiao CHEN, Zan WANG, Hui LU. Online Anomaly Detection for Servo Systems with Generative Recurrent Networks[J]. AEROSPACE SHANGHAI, 2025, 42(2): 157
Paper Information
Category: Simulation and Analysis
Received: Dec. 9, 2024
Accepted: --
Published Online: May. 26, 2025
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20