首页|期刊导航|国际设备工程与管理(英文版)|Adaptive Wavelets Based on Second Generation Wavelet Transform and Their Applications to Trend Analysis and Prediction
国际设备工程与管理(英文版)2004,Vol.9Issue(3):170-176,7.
Adaptive Wavelets Based on Second Generation Wavelet Transform and Their Applications to Trend Analysis and Prediction
Adaptive Wavelets Based on Second Generation Wavelet Transform and Their Applications to Trend Analysis and Prediction
摘要
Abstract
In order to make trend analysis and prediction to acquisition data in a mechanical equipment condition monitoring system, a new method of trend feature extraction and prediction of acquisition data is proposed which constructs an adaptive wavelet on the acquisition data by means of second generation wavelet transform (SGWT). Firstly, taking the vanishing moment number of the predictor as a constraint, the linear predictor and updater are designed according to the acquisition data by using symmetrical interpolating scheme. Then the trend of the data is obtained through doing SGWT decomposition, threshold processing and SGWT reconstruction. Secondly, under the constraint of the vanishing moment number of the predictor, another predictor based on the acquisition data is devised to predict the future trend of the data using a non-symmetrical interpolating scheme. A one-step prediction algorithm is presented to predict the future evolution trend with historical data. The proposed method obtained a desirable effect in peak-to-peak value trend analysis for a machine set in an oil refinery.关键词
second generation wavelet transform (SGWT)/predictor/updater/trend analysis/trend predictionKey words
second generation wavelet transform (SGWT)/predictor/updater/trend analysis/trend prediction分类
机械制造引用本文复制引用
DUAN Chen-dong,JIANG Hong-kai,HE Zheng-jia..Adaptive Wavelets Based on Second Generation Wavelet Transform and Their Applications to Trend Analysis and Prediction[J].国际设备工程与管理(英文版),2004,9(3):170-176,7.基金项目
This work was supported by the National Natural Foundation of China under Grant No. 50335030, No. 50175087 and No. 50305012 ()