The Prediction Method of Frequency Spectrum Based on Full Vector Support Vector Regression
[1]LI Lingjun,CHEN Chao,HAN Jie,et al.The Prediction Method of Frequency Spectrum Based on Full Vector Support Vector Regression[J].Journal of Zhengzhou University (Engineering Science),2016,37(03):78-.[doi:10.13705/j.issn.1671-6833.2016.03.018]
Copy
Journal of Zhengzhou University (Engineering Science)[ISSN
1671-6833/CN
41-1339/T] Volume:
37
Number of periods:
2016 03
Page number:
78-
Column:
Public date:
2016-05-10
- Title:
- The Prediction Method of Frequency Spectrum Based on Full Vector Support Vector Regression
- Keywords:
- full vector spectrum; support vector regression; time series; frequency spectrum prediction
- CLC:
- TH17
- Abstract:
- In order to predict the fault type and the fault position of rotating machinery more accurately, a newprediction method of frequency spectrum based on full vector support vector regression is proposed. The newmethod uses the full vector spectrum technology to merge the homologous double channel signal information,and uses support vector regression to predict the full vector spectrum characteristic parameters,which can en-sure the comprehensiveness of the training data and the accuracy of predict result. This method can forecastthe frequency spectrum of the vibration signal accuracy and can then give the technical based for fault type andfault position predict. The experimental results of the frequency spectrum forecast of a 1000 MW steam turbineshaft show that this method can predict fully and accurately the spectrum’s structure of vibration signal.
- References:
- -
- Similar References:
Memo
-
Last Update:
