發(fā)布時(shí)間：2018-10-18 10:13

【摘要】：隨著大容量、高參數(shù)的風(fēng)力發(fā)電機(jī)組投入商業(yè)運(yùn)行,對(duì)機(jī)組設(shè)備故障診斷的實(shí)時(shí)性、準(zhǔn)確性及有效性的要求也越來越高,而故障診斷是保證機(jī)組安全可靠運(yùn)行的重要方法之一。風(fēng)速頻繁變化、沖擊大、變載荷的運(yùn)行特點(diǎn),導(dǎo)致風(fēng)電機(jī)組故障類型多且頻率高。而齒輪箱是風(fēng)電機(jī)組最重要傳動(dòng)部件之一,也是故障高發(fā)部件,且造成風(fēng)電機(jī)組的停機(jī)時(shí)間也最長。本論文分析了傳統(tǒng)風(fēng)電機(jī)組齒輪箱故障診斷方法的特點(diǎn)及存在的問題,嘗試將可有效解決類別不平衡問題的代價(jià)敏感學(xué)習(xí)應(yīng)用于風(fēng)電機(jī)組齒輪箱故障診斷,探索齒輪箱故障診斷的新方法。主要研究成果如下:針對(duì)代價(jià)敏感支持向量機(jī)(Cost-sensitive Support Vector Machine,CSVM)在樣本數(shù)據(jù)量較大時(shí)訓(xùn)練速度過慢的問題,提出增量代價(jià)敏感支持向量機(jī)(Incremental Cost-sensitive Support Vector Machine,ICSVM)。該算法有效利用KKT條件,對(duì)增量樣本集中的樣本進(jìn)行有效的選取,剔除對(duì)下一步訓(xùn)練無效的樣本,得到邊界支持向量集。在UCI標(biāo)準(zhǔn)數(shù)據(jù)集上進(jìn)行仿真實(shí)驗(yàn),驗(yàn)證的ICSVM的有效性。給出了基于ICSVM風(fēng)電機(jī)組齒輪箱軸承故障診斷方法的具體實(shí)現(xiàn)過程,該試驗(yàn)結(jié)果表明,該方法平均誤分類代價(jià)最低;故障類識(shí)別率更高;該方法具有訓(xùn)練速度快,非常適合風(fēng)電機(jī)組的在線故障診斷。針對(duì)最小二乘支持向量機(jī)(Least Squares Support Vector Machine,LSSVM)不具有代價(jià)敏感性的問題,提出代價(jià)敏感最小二乘支持向量機(jī)(Cost-sensitive Least Square Support Vector Machine,CLSSVM)。在LSSVM原始的優(yōu)化問題上嵌入不同的誤分類代價(jià)參數(shù),以平均誤分類代價(jià)最小為優(yōu)化目標(biāo),詳細(xì)推導(dǎo)了CLSSVM算法。最后將其應(yīng)用于UCI標(biāo)準(zhǔn)數(shù)據(jù)集和風(fēng)電機(jī)組齒輪箱軸承故障診斷。試驗(yàn)結(jié)果表明,該方法的平均誤分類代價(jià)最低,克服LSSVM不具有代價(jià)敏感性的問題,能夠提高故障類樣本的正確率;CLSSVM的訓(xùn)練時(shí)間短,非常適合風(fēng)電機(jī)組的在線診斷。
[Abstract]:With the large capacity and high parameter wind turbine being put into commercial operation, the requirements of real-time, accuracy and effectiveness of fault diagnosis are becoming higher and higher, and fault diagnosis is one of the important methods to ensure the safe and reliable operation of the unit. The frequent variation of wind speed, high impact and variable load result in many fault types and high frequency of wind turbine. The gearbox is one of the most important transmission parts of wind turbine, and is also the part with high fault rate, which results in the longest downtime of wind turbine. In this paper, the characteristics and problems of the traditional gearbox fault diagnosis method of wind turbine are analyzed, and the cost sensitive learning, which can effectively solve the problem of class imbalance, is tried to be applied to the fault diagnosis of the gearbox of wind turbine. To explore a new method of gearbox fault diagnosis. The main research results are as follows: aiming at the problem of slow training speed of cost sensitive support vector machine (Cost-sensitive Support Vector Machine,CSVM) when the sample data is large, an incremental cost sensitive support vector machine (Incremental Cost-sensitive Support Vector Machine,ICSVM) is proposed. Using the KKT condition effectively, the algorithm selects the samples in the incremental sample set effectively, removes the samples that are not valid for the next training, and obtains the boundary support vector set. The effectiveness of the ICSVM is verified by simulation experiments on the UCI standard data set. The realization process of bearing fault diagnosis method for wind turbine gearbox based on ICSVM is presented. The experimental results show that the method has the lowest average misclassification cost, higher fault class recognition rate and faster training speed. It is very suitable for on-line fault diagnosis of wind turbine. To solve the problem that least squares support vector machine (Least Squares Support Vector Machine,LSSVM) is not cost sensitive, a cost sensitive least squares support vector machine (Cost-sensitive Least Square Support Vector Machine,CLSSVM) is proposed. Different misclassification cost parameters are embedded in the original optimization problem of LSSVM. The CLSSVM algorithm is deduced in detail with the minimum average misclassification cost as the optimization objective. Finally, it is applied to UCI standard data set and wind turbine gearbox bearing fault diagnosis. The experimental results show that this method has the lowest average misclassification cost and can improve the accuracy of fault samples by overcoming the problem that LSSVM is not sensitive to cost, and the training time of CLSSVM is short, so it is very suitable for on-line diagnosis of wind turbines.
【學(xué)位授予單位】：長沙理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TM315

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李狀;馬志勇;姜銳;柳亦兵;;風(fēng)電機(jī)組齒輪箱故障分類方法研究[J];機(jī)械設(shè)計(jì)與制造;2015年02期

2 周進(jìn);房寧;郭鵬;;基于相對(duì)主元分析的風(fēng)電機(jī)組塔架振動(dòng)狀態(tài)監(jiān)測與故障診斷[J];電力建設(shè);2014年08期

3 Nassim Laouti;Sami Othman;Mazen Alamir;Nida Sheibat-Othman;;Combination of Model-based Observer and Support Vector Machines for Fault Detection of Wind Turbines[J];International Journal of Automation & Computing;2014年03期

4 丁碩;常曉恒;巫慶輝;魏洪峰;楊友林;;基于LVQ神經(jīng)網(wǎng)絡(luò)風(fēng)電機(jī)組齒輪箱故障診斷研究[J];現(xiàn)代電子技術(shù);2014年10期

5 尹金良;劉玲玲;;代價(jià)敏感相關(guān)向量機(jī)的研究及其在變壓器故障診斷中的應(yīng)用[J];電力自動(dòng)化設(shè)備;2014年05期

6 尹金良;朱永利;鄭曉雨;王國強(qiáng);;代價(jià)敏感VBGP在變壓器故障診斷中的應(yīng)用[J];電工技術(shù)學(xué)報(bào);2014年03期

7 芮曉明;張穆勇;霍娟;;試運(yùn)行期間風(fēng)電機(jī)組平均故障間隔時(shí)間的估計(jì)[J];中國電機(jī)工程學(xué)報(bào);2014年21期

8 周真;周浩;馬德仲;張茹;蔣永清;;風(fēng)電機(jī)組故障診斷中不確定性信息處理的貝葉斯網(wǎng)絡(luò)方法[J];哈爾濱理工大學(xué)學(xué)報(bào);2014年01期

9 王彤;;基于最小二乘支持向量機(jī)的軌道電路故障診斷方法[J];鐵道標(biāo)準(zhǔn)設(shè)計(jì);2014年02期

10 尹玉萍;劉萬軍;;基于AC-DE算法的風(fēng)電機(jī)組齒輪箱故障診斷方法[J];計(jì)算機(jī)工程與應(yīng)用;2014年13期

相關(guān)博士學(xué)位論文 前1條

1 孫鮮明;復(fù)雜工況下風(fēng)力發(fā)電機(jī)組關(guān)鍵部件故障分析與診斷研究[D];沈陽工業(yè)大學(xué);2014年

相關(guān)碩士學(xué)位論文 前1條

1 趙立超;大型風(fēng)力發(fā)電機(jī)的齒輪箱故障診斷[D];沈陽工業(yè)大學(xué);2011年

，

本文編號(hào)：2278815