基于支持向量機模型的滾動軸承運行狀態(tài)預(yù)測研究
發(fā)布時間:2018-07-15 18:30
【摘要】:為了提高支持向量機對滾動軸承運行狀態(tài)預(yù)測的準(zhǔn)確性,采用遺傳算法對支持向量機的核參數(shù)與懲罰因子進(jìn)行優(yōu)化,選取最優(yōu)支持向量機參數(shù);研究了時間序列預(yù)測中嵌入維數(shù)對預(yù)測精度的影響,提出了嵌入維數(shù)優(yōu)選-參數(shù)優(yōu)化支持向量機預(yù)測模型。通過軸承壽命加速試驗表明,該模型可以提高滾動軸承運行狀態(tài)預(yù)測的準(zhǔn)確性,達(dá)到軸承故障預(yù)警時間預(yù)測的目的。
[Abstract]:In order to improve the accuracy of SVM prediction of rolling bearing running state, the kernel parameters and penalty factors of SVM are optimized by genetic algorithm, and the optimal SVM parameters are selected. The influence of embedding dimension on prediction accuracy in time series prediction is studied, and a prediction model of embedding dimension optimal selection and parameter optimization support vector machine is proposed. The accelerated test of bearing life shows that the model can improve the accuracy of rolling bearing operation state prediction and achieve the purpose of bearing fault warning time prediction.
【作者單位】: 中國石油大學(xué)(北京)機械與儲運工程學(xué)院;
【基金】:國家科技重大專項(2011ZX05055)
【分類號】:TH133.33;TP181
[Abstract]:In order to improve the accuracy of SVM prediction of rolling bearing running state, the kernel parameters and penalty factors of SVM are optimized by genetic algorithm, and the optimal SVM parameters are selected. The influence of embedding dimension on prediction accuracy in time series prediction is studied, and a prediction model of embedding dimension optimal selection and parameter optimization support vector machine is proposed. The accelerated test of bearing life shows that the model can improve the accuracy of rolling bearing operation state prediction and achieve the purpose of bearing fault warning time prediction.
【作者單位】: 中國石油大學(xué)(北京)機械與儲運工程學(xué)院;
【基金】:國家科技重大專項(2011ZX05055)
【分類號】:TH133.33;TP181
【參考文獻(xiàn)】
相關(guān)期刊論文 前1條
1 張穎璐;;基于遺傳算法優(yōu)化支持向量機的網(wǎng)絡(luò)流量預(yù)測[J];計算機科學(xué);2008年05期
相關(guān)碩士學(xué)位論文 前1條
1 陳麗琳;基于多嵌入維數(shù)的時用水量LSSVM組合預(yù)測[D];浙江大學(xué);2013年
【共引文獻(xiàn)】
相關(guān)期刊論文 前10條
1 劉春;;遺傳算法優(yōu)化BP神經(jīng)網(wǎng)絡(luò)的網(wǎng)絡(luò)流量預(yù)測[J];信息安全與技術(shù);2014年06期
2 陳曦;;基于包容性檢驗和神經(jīng)網(wǎng)絡(luò)的網(wǎng)絡(luò)流量預(yù)測[J];電視技術(shù);2014年11期
3 呂威;馬維e,
本文編號:2125001
本文鏈接:http://www.wukwdryxk.cn/kejilunwen/jixiegongcheng/2125001.html
最近更新
教材專著