鑒別性鄰域嵌入方法及其在水電機(jī)組異常檢測(cè)中的應(yīng)用
發(fā)布時(shí)間：2019-06-27 15:57
【摘要】：可再生能源開發(fā)戰(zhàn)略是國家十二五規(guī)劃的重要組成部分。小水電是一種資源分布廣、開發(fā)潛力大、環(huán)境影響小、可擴(kuò)展利用的可再生能源,在國家能源發(fā)展戰(zhàn)略上有著重大意義。在現(xiàn)階段,考慮到水電機(jī)組的復(fù)雜性以及小水電站位置的苛刻性,通常采用專人值守的形式進(jìn)行設(shè)備維護(hù)與異常監(jiān)測(cè)。其過程不僅效率低下,而且過分依賴于工作人員的經(jīng)驗(yàn)知識(shí),往往具有較高的誤判率,因此有必要研究機(jī)器學(xué)習(xí)理論與統(tǒng)計(jì)學(xué)理論并實(shí)現(xiàn)高性能識(shí)別算法,用于實(shí)現(xiàn)無人值守的小水電監(jiān)測(cè)系統(tǒng)。鄰域嵌入分析算法能夠有效地進(jìn)行數(shù)據(jù)分簇可視化操作,如何提升鄰域嵌入算法的鑒別性能并應(yīng)用于水電機(jī)組噪聲源識(shí)別具有非常重要的研究價(jià)值。 本文針對(duì)依據(jù)噪聲源進(jìn)行的水電機(jī)組異常檢測(cè)問題,分析了鄰域嵌入分析算法的有監(jiān)督擴(kuò)展與線性投影擴(kuò)展技術(shù),設(shè)計(jì)了相應(yīng)的鑒別性鄰域嵌入分類算法。主要工作如下： (1)提出了基于拉斯優(yōu)化的鑒別性鄰域嵌入分類算法DEE。在異常檢測(cè)應(yīng)用任務(wù)中,現(xiàn)有的鄰域嵌入分析算法由于缺少類別標(biāo)簽,識(shí)別率較為低下。DEE算法不僅能夠直觀地進(jìn)行數(shù)據(jù)分簇可視化展示,而且通過拉斯方向提升了模型構(gòu)建效率,使之得能勝任中大規(guī)模數(shù)據(jù)的鑒別任務(wù)。在3類公共數(shù)據(jù)集中驗(yàn)證了該算法的分簇能力與識(shí)別性能。 (2)通過引入核技巧,在保留其線性投影特性的同時(shí)將DEE進(jìn)行非線性核化擴(kuò)展,提出了兩種不同類型的核化鑒別性鄰域嵌入分類算法KDEE1與KDEE2。考慮到水電機(jī)組設(shè)備異常運(yùn)行時(shí)振動(dòng)噪聲特征的非線性性質(zhì)。在模型構(gòu)建過程中,依據(jù)微分對(duì)象的不同,將核化的DEE版本稱之為KDEE1和KDEE2,兩者都能應(yīng)用于非線性輸入數(shù)據(jù),且分簇能力與識(shí)別效率都較DEE有進(jìn)一步的提升。 (3)分析了水電機(jī)組異常振動(dòng)時(shí)所采集到的噪聲特性及其預(yù)處理手段,將本文所提的KDEE1、KDEE2、DEE等鑒別性算法應(yīng)用于實(shí)際的水電機(jī)機(jī)異常檢測(cè)中。通過噪聲特征子空間分簇以及識(shí)別率兩種實(shí)驗(yàn)結(jié)果,表明核鑒別鄰域嵌入分析算法具有較高的實(shí)用價(jià)值。
[Abstract]:Renewable energy development strategy is an important part of the 12th five-year Plan. Small hydropower is a kind of renewable energy with wide distribution of resources, great development potential, small environmental impact and extensible utilization, which is of great significance in the national energy development strategy. At present, considering the complexity of hydropower units and the rigour of the position of small hydropower stations, equipment maintenance and abnormal monitoring are usually carried out in the form of special personnel. The process is not only inefficient, but also too dependent on the experience and knowledge of staff, and often has a high misjudgment rate. Therefore, it is necessary to study machine learning theory and statistical theory and realize high performance recognition algorithm, which can be used to realize unattended small hydropower monitoring system. Neighborhood embedding analysis algorithm can effectively carry out data clustering visualization operation. How to improve the identification performance of neighborhood embedding algorithm and apply it to the identification of noise sources of hydropower units is of great research value. In this paper, aiming at the problem of abnormal detection of hydropower units based on noise sources, the supervised extension and linear projection expansion techniques of neighborhood embedding analysis algorithm are analyzed, and the corresponding discriminant neighborhood embedding classification algorithm is designed. The main work is as follows: (1) A discriminant neighborhood embedding classification algorithm DEE. based on Russ optimization is proposed. In the application task of anomaly detection, the existing neighborhood embedding analysis algorithms have low recognition rate due to the lack of category tags. Dee algorithm can not only intuitively display data clustering visualization, but also improve the efficiency of model construction through Russ direction, so that it can be competent for the identification task of medium and large scale data. The clustering ability and recognition performance of the algorithm are verified in three kinds of common data sets. (2) by introducing kernel technique, DEE is extended by nonlinear nucleation while retaining its linear projection characteristics, and two different types of kernel discriminant neighborhood embedding classification algorithms KDEE1 and KDEE2. are proposed. The nonlinear properties of vibration and noise characteristics of hydropower units in abnormal operation are taken into account. In the process of model construction, according to the difference of differential objects, the nucleated DEE version called KDEE1 and KDEE2, can be applied to nonlinear input data, and the clustering ability and recognition efficiency are further improved compared with DEE. (3) the noise characteristics and preprocessing methods collected during abnormal vibration of hydropower units are analyzed, and the KDEE1,KDEE2,DEE and other discriminant algorithms proposed in this paper are applied to the actual anomaly detection of hydropower machines. The experimental results of noise feature subspace clustering and recognition rate show that the kernel discriminant neighborhood embedding analysis algorithm has high practical value.
【學(xué)位授予單位】：浙江工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TV734;TV738
本文編號(hào)：2506936
[Abstract]:Renewable energy development strategy is an important part of the 12th five-year Plan. Small hydropower is a kind of renewable energy with wide distribution of resources, great development potential, small environmental impact and extensible utilization, which is of great significance in the national energy development strategy. At present, considering the complexity of hydropower units and the rigour of the position of small hydropower stations, equipment maintenance and abnormal monitoring are usually carried out in the form of special personnel. The process is not only inefficient, but also too dependent on the experience and knowledge of staff, and often has a high misjudgment rate. Therefore, it is necessary to study machine learning theory and statistical theory and realize high performance recognition algorithm, which can be used to realize unattended small hydropower monitoring system. Neighborhood embedding analysis algorithm can effectively carry out data clustering visualization operation. How to improve the identification performance of neighborhood embedding algorithm and apply it to the identification of noise sources of hydropower units is of great research value. In this paper, aiming at the problem of abnormal detection of hydropower units based on noise sources, the supervised extension and linear projection expansion techniques of neighborhood embedding analysis algorithm are analyzed, and the corresponding discriminant neighborhood embedding classification algorithm is designed. The main work is as follows: (1) A discriminant neighborhood embedding classification algorithm DEE. based on Russ optimization is proposed. In the application task of anomaly detection, the existing neighborhood embedding analysis algorithms have low recognition rate due to the lack of category tags. Dee algorithm can not only intuitively display data clustering visualization, but also improve the efficiency of model construction through Russ direction, so that it can be competent for the identification task of medium and large scale data. The clustering ability and recognition performance of the algorithm are verified in three kinds of common data sets. (2) by introducing kernel technique, DEE is extended by nonlinear nucleation while retaining its linear projection characteristics, and two different types of kernel discriminant neighborhood embedding classification algorithms KDEE1 and KDEE2. are proposed. The nonlinear properties of vibration and noise characteristics of hydropower units in abnormal operation are taken into account. In the process of model construction, according to the difference of differential objects, the nucleated DEE version called KDEE1 and KDEE2, can be applied to nonlinear input data, and the clustering ability and recognition efficiency are further improved compared with DEE. (3) the noise characteristics and preprocessing methods collected during abnormal vibration of hydropower units are analyzed, and the KDEE1,KDEE2,DEE and other discriminant algorithms proposed in this paper are applied to the actual anomaly detection of hydropower machines. The experimental results of noise feature subspace clustering and recognition rate show that the kernel discriminant neighborhood embedding analysis algorithm has high practical value.
【學(xué)位授予單位】：浙江工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TV734;TV738
【二級(jí)參考文獻(xiàn)】
相關(guān)期刊論文 前1條
1 王金水;張東站;施秀升;賴興瑞;;基于免疫系統(tǒng)和模糊邏輯的自適應(yīng)網(wǎng)絡(luò)入侵檢測(cè)(英文)[J];心智與計(jì)算;2009年01期
，本文編號(hào)：2506936
本文鏈接：http://sikaile.net/kejilunwen/shuiwenshuili/2506936.html
最近更新
教材專著
