Fault detection method based on support vector data description
The invention discloses a fault detection method based on support vector data description. The fault detection method comprises the following steps: step 1, acquiring normal training data on a nuclear source platform; 2, obtaining projection data and a projection matrix of the normal training data t...
Gespeichert in:
Hauptverfasser: | , , , , , , , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention discloses a fault detection method based on support vector data description. The fault detection method comprises the following steps: step 1, acquiring normal training data on a nuclear source platform; 2, obtaining projection data and a projection matrix of the normal training data through a locality preserving projection method; 3, pre-processing the projected data; 4, training a support vector data description model; step 5, collecting data on line; 6, calculating online data after projection according to the projection matrix; step 7, preprocessing the projected online data; 8, inputting the processed data into a support vector data model to calculate real-time statistics; and 9, judging the sizes of D and R. The method has the beneficial effects that the indexes such as the detection rate and the real-time online performance are well improved, and the method has greater advantages in the aspect of fault detection performance.
本发明公开了一种基于支持向量数据描述的故障检测方法,包括如下步骤:步骤1:在核源平台上采集正常训练数据;步骤2:局部保持投影方法 |
---|