舰船电子工程2025，Vol.45Issue(2)：129-132,181,5.DOI:10.3969/j.issn.1672-9730.2025.02.027
基于数据驱动的遥测缓变参数快速全局K-Means聚类异常检测包络模型
Telemetry Slowly Varying Parameters Fast Global K-Means Clustering Anomaly Detection Envelope Model Based on Data Driven
胡健 1刘学1
作者信息
- 1. 中国人民解放军91550部队 大连 116023
- 折叠
摘要
Abstract
Telemetry parameters are important parameters to reflect the state and environment of aircraft.In order to realize the fast recognition and detection of telemetry slow variation parameter anomaly,and to improve the traditional method of determin-ing telemetry parameter anomaly by setting single upper and lower bounds,in this paper,a fast telemetry slow variation parameter global K-Means clustering anomaly detection envelope model based on data driven is proposed.The fast global K-Means clustering algorithm is used to calculate the clustering center of sample data,then the upper and lower bounds of the envelope are calculated using dynamic variable step size considering the noise characteristics,the envelope model of telemetry slow variation parameters anomaly detection is obtained.The simulation results show that the method proposed in this paper can effectively detect the abnor-mal of telemetry slow variation parameters.关键词
遥测缓变参数/数据驱动/K-Means聚类/包络模型/异常检测Key words
telemetry slow variation parameter/data driven/K-Means clustering/envelope model/anomaly detection分类
信息技术与安全科学引用本文复制引用
胡健,刘学..基于数据驱动的遥测缓变参数快速全局K-Means聚类异常检测包络模型[J].舰船电子工程,2025,45(2):129-132,181,5.
