本文選題：600MW汽輪機(jī)組 + 振動故障機(jī)理分析��； 參考：《華北電力大學(xué)》2015年碩士論文

【摘要】：為了滿足生活及生產(chǎn)用電量急劇增加的需求,人們需要不斷地對汽輪發(fā)電機(jī)組容量進(jìn)行改造及研制；但是,隨著機(jī)組容量的不斷增大,對于材料及運行環(huán)境的要求也會相應(yīng)地提高,特別是在國家實行了“上大壓小”政策之后,我國的主力發(fā)電機(jī)組的容量也進(jìn)入了600MW時代,而接踵而來的問題便是細(xì)微的機(jī)組振動故障卻帶來廣泛的影響,因此對其進(jìn)行600MW汽輪機(jī)組的振動故障診斷就變得更加重要。本文針對600MW汽輪機(jī)組的振動,闡述了目前600MW汽輪機(jī)振動診斷的過程,和最新的研究現(xiàn)狀。對于汽輪機(jī)振動的故障進(jìn)行機(jī)理分析并且根據(jù)現(xiàn)場的實際運行狀況進(jìn)行解讀。對于振動故障的特征提取分析,應(yīng)用目前較為常用的時域分析、頻域分析、時頻域分析分別進(jìn)行分析處理,研究針對與600MW大型汽輪機(jī)振動的特征提取方法,故障診斷方面采取的是目前研究較為熱門的支持向量機(jī)的分類診斷方法,并應(yīng)用交叉驗證和網(wǎng)格搜素、粒子群優(yōu)化算法、遺傳算法對支持向量機(jī)進(jìn)行參數(shù)尋優(yōu)。最終根據(jù)自己的所掌握汽輪機(jī)機(jī)組的振動知識,選取對企業(yè)最為有益的汽輪機(jī)振動故障診斷系統(tǒng)。
[Abstract]:In order to meet the demand of increasing power consumption in daily life and production, people need to transform and develop the capacity of turbine-generator set constantly, but with the increasing of capacity of turbine generator, The requirements for materials and operating environment will also be raised accordingly, especially after the implementation of the policy of "high pressure and small pressure", the capacity of the main generating units in our country has also entered the 600MW era. However, the following problems are caused by minor vibration faults, so it is more important to diagnose the vibration faults of 600MW steam turbines. Aiming at the vibration of 600MW steam turbine, this paper expounds the process of vibration diagnosis of 600MW steam turbine at present, and the latest research status. The mechanism of turbine vibration fault is analyzed and interpreted according to the actual operation condition. For the feature extraction analysis of vibration fault, this paper applies the time domain analysis, frequency domain analysis and time frequency domain analysis to analyze and deal with the vibration characteristics of 600MW large steam turbine. In the aspect of fault diagnosis, a popular classification and diagnosis method of support vector machine (SVM) is adopted, and the parameters of SVM are optimized by cross-validation, mesh search, particle swarm optimization and genetic algorithm. Finally, according to the knowledge of turbine unit vibration, the most beneficial fault diagnosis system of steam turbine is selected.
【學(xué)位授予單位】：華北電力大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：TM621

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 陳鏗;韓伯棠;;混沌時間序列分析中的相空間重構(gòu)技術(shù)綜述[J];計算機(jī)科學(xué);2005年04期

2 楊世錫,焦衛(wèi)東,吳昭同;獨立分量分析基網(wǎng)絡(luò)應(yīng)用于旋轉(zhuǎn)機(jī)械故障特征抽取與分類[J];機(jī)械工程學(xué)報;2004年03期

3 陳進(jìn);信號處理在機(jī)械設(shè)備故障診斷中的應(yīng)用(連載)[J];振動與沖擊;1999年04期

，

本文編號：2058353