本文選題：狀態(tài)監(jiān)測(cè) + 小波神經(jīng)網(wǎng)絡(luò)��； 參考：《華北電力大學(xué)》2017年碩士論文

【摘要】：風(fēng)能,作為可再生能源,無(wú)窮無(wú)盡,清潔環(huán)保,已成為許多國(guó)家可持續(xù)發(fā)展戰(zhàn)略的一個(gè)重要組成部分,因此,風(fēng)力發(fā)電得到了迅速的發(fā)展。風(fēng)電機(jī)組工作環(huán)境惡劣,長(zhǎng)期受到正常和極端溫度、降雨、積雪、沙塵、太陽(yáng)輻射等環(huán)境因素的影響,各部件也必將不可避免隨著運(yùn)行時(shí)間的變化而老化,可靠性下降,導(dǎo)致故障發(fā)生,影響風(fēng)電場(chǎng)的安全穩(wěn)定。風(fēng)力發(fā)電機(jī)作為風(fēng)電機(jī)組故障率較高的部件,對(duì)其進(jìn)行實(shí)時(shí)狀態(tài)監(jiān)測(cè),及時(shí)發(fā)現(xiàn)故障征兆,確定合理的維護(hù)方案,對(duì)降低維護(hù)成本和提高機(jī)組的可靠性具有重大意義。目前,風(fēng)電機(jī)組通過(guò)傳感器實(shí)時(shí)地采集其重要參數(shù),這將使得存儲(chǔ)數(shù)據(jù)從GB級(jí)上升到TB級(jí),甚至是PB級(jí)。在大數(shù)據(jù)背景下,如何能夠快速的處理日益增長(zhǎng)的海量狀態(tài)監(jiān)測(cè)數(shù)據(jù),并且能夠準(zhǔn)確地分析當(dāng)前情況下風(fēng)力發(fā)電機(jī)的運(yùn)行狀態(tài)成為了新的課題。在此背景下,本文采用溫度趨勢(shì)分析的方法對(duì)上述問(wèn)題展開(kāi)研究。(1)在能夠獲取風(fēng)力發(fā)電機(jī)實(shí)時(shí)監(jiān)測(cè)數(shù)據(jù)的基礎(chǔ)上,建立了用于風(fēng)力發(fā)電機(jī)溫度預(yù)測(cè)的小波神經(jīng)網(wǎng)絡(luò)模型。通過(guò)相關(guān)系數(shù)法對(duì)風(fēng)力發(fā)電機(jī)溫度的影響因素進(jìn)行分析,確定了網(wǎng)絡(luò)輸入,通過(guò)試湊法得到網(wǎng)絡(luò)隱含層神經(jīng)元個(gè)數(shù),從而確定網(wǎng)絡(luò)結(jié)構(gòu)。(2)針對(duì)在使用風(fēng)機(jī)監(jiān)測(cè)數(shù)據(jù)對(duì)小波神經(jīng)網(wǎng)絡(luò)訓(xùn)練時(shí)出現(xiàn)的收斂速度慢、易陷入局部最優(yōu)現(xiàn)象,本文采用改進(jìn)的花朵授粉算法對(duì)小波神經(jīng)網(wǎng)絡(luò)的參數(shù),包括權(quán)值、伸縮因子和平移因子進(jìn)行優(yōu)化。通過(guò)引入混沌序列和t分布變異,使花朵授粉算法具有更好的尋優(yōu)能力,加快了小波神經(jīng)網(wǎng)絡(luò)的訓(xùn)練速度,提高了精度。(3)針對(duì)海量風(fēng)電機(jī)組狀態(tài)監(jiān)測(cè)數(shù)據(jù),本文提出了改進(jìn)的并行化花朵授粉算法優(yōu)化小波神經(jīng)網(wǎng)絡(luò)(CITDMFPA-WNN)模型,并將該模型部署在Spark平臺(tái)上,利用優(yōu)化后的參數(shù)進(jìn)行溫度預(yù)測(cè)。通過(guò)引入并行化,提高計(jì)算速度,使算法具備處理海量數(shù)據(jù)的能力。(4)采用上述模型利用風(fēng)力發(fā)電機(jī)實(shí)時(shí)監(jiān)測(cè)數(shù)據(jù)進(jìn)行風(fēng)力發(fā)電機(jī)溫度預(yù)測(cè),然后采用滑動(dòng)窗口統(tǒng)計(jì)方法對(duì)溫度殘差,即預(yù)測(cè)溫度值與實(shí)際溫度值的差值,進(jìn)行分析來(lái)確定對(duì)風(fēng)力發(fā)電機(jī)工作異常監(jiān)測(cè)時(shí)所需的均值和標(biāo)準(zhǔn)差的閾值,從而確定風(fēng)力發(fā)電機(jī)的實(shí)時(shí)運(yùn)行狀態(tài),達(dá)到在線(xiàn)狀態(tài)監(jiān)測(cè)的目的。最后,進(jìn)行了對(duì)比實(shí)驗(yàn)和算例分析。選用我國(guó)內(nèi)蒙古某風(fēng)電場(chǎng)的真實(shí)運(yùn)行數(shù)據(jù),在實(shí)驗(yàn)室搭建了云計(jì)算集群,對(duì)本文提出的算法進(jìn)行性能測(cè)試和風(fēng)力發(fā)電機(jī)狀態(tài)監(jiān)測(cè)驗(yàn)證。實(shí)驗(yàn)表明本文設(shè)計(jì)的算法具有良好的準(zhǔn)確性和并行性,并且能夠應(yīng)用于風(fēng)力發(fā)電機(jī)的狀態(tài)監(jiān)測(cè)。
[Abstract]:Wind energy, as a renewable energy, is endless, clean and environmental protection, has become an important part of the sustainable development strategy of many countries. Therefore, wind power generation has been developed rapidly. The working environment of wind turbines has been affected by environmental factors such as normal and extreme temperatures, rain, snow, dust, and solar radiation for a long time. It will inevitably deteriorate with the change of running time, decrease the reliability, cause the failure and affect the safety and stability of the wind farm. As a component with high failure rate of the wind turbine, the wind turbine can monitor it in real time, find out the fault symptoms in time, determine the reasonable maintenance scheme, reduce the maintenance cost and improve the maintenance cost. The reliability of the unit is of great significance. At present, the wind turbines collect their important parameters in real time through sensors, which will increase the storage data from the GB level to the TB level, or even the PB level. In the large data background, how to quickly handle the growing mass state monitoring data and accurately analyze the current situation. The running state of the force generator has become a new topic. Under this background, this paper uses the method of temperature trend analysis to study the above problems. (1) on the basis of obtaining real-time monitoring data of wind turbines, a wavelet neural network model for wind generator temperature prediction is established. The influence factors of the motor temperature are analyzed, the network input is determined, the number of neurons in the hidden layer of the network is obtained by the trial and error method, and the network structure is determined. (2) the improved flower pollination algorithm is adopted in this paper in view of the slow convergence speed and the local optimal phenomenon when the wind turbine monitoring data is used in the training of the wavelet neural network. The parameters of the wavelet neural network, including the weight value, the expansion factor and the translation factor, are optimized. By introducing the chaos sequence and the variation of t distribution, the flower pollination algorithm has a better optimization ability, quickening the training speed of the wavelet neural network and improving the precision. (3) the paper puts forward the change of the state monitoring data of the mass wind turbines. The proposed parallel flower pollination algorithm optimizes the wavelet neural network (CITDMFPA-WNN) model, and deploys the model on the Spark platform to make use of the optimized parameters to predict the temperature. By introducing parallelization to improve the computing speed, the algorithm has the ability to deal with massive data. (4) the real-time monitoring of wind turbines is used in this model. The data is used to predict the temperature of the wind generator, and then the statistical method of sliding window is used to analyze the difference between the temperature residual and the actual temperature. The value and the threshold value of the standard deviation for the abnormal monitoring of the wind generator are analyzed to determine the real-time running state of the wind generator and reach the online shape. Finally, the comparison experiment and the example analysis are carried out. The cloud computing cluster is set up in the laboratory of a wind farm in Inner Mongolia, and the performance test and the wind generator state monitoring and verification are carried out in the laboratory. The experiment shows that the algorithm designed in this paper has good accuracy and is good. It can be applied to condition monitoring of wind turbines.
【學(xué)位授予單位】：華北電力大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP183;TM315

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 楊天晴;王津;楊旭濤;張學(xué)杰;;一種Spark環(huán)境下的高效率大規(guī)模圖數(shù)據(jù)處理機(jī)制[J];計(jì)算機(jī)應(yīng)用研究;2016年12期

2 劉永強(qiáng);楊紹普;廖英英;王翠艷;;一種自適應(yīng)共振解調(diào)方法及其在滾動(dòng)軸承早期故障診斷中的應(yīng)用[J];振動(dòng)工程學(xué)報(bào);2016年02期

3 方瑞明;江順輝;尚榮艷;王黎;;采用趨勢(shì)狀態(tài)分析的風(fēng)機(jī)齒輪箱狀態(tài)在線(xiàn)評(píng)估云模型[J];華僑大學(xué)學(xué)報(bào)(自然科學(xué)版);2016年01期

4 王保義;王冬陽(yáng);張少敏;;基于Spark和IPPSO_LSSVM的短期分布式電力負(fù)荷預(yù)測(cè)算法[J];電力自動(dòng)化設(shè)備;2016年01期

5 曹莉;唐玲;吳浩;高祥;樂(lè)英高;;基于改進(jìn)小波神經(jīng)網(wǎng)絡(luò)的短時(shí)交通流量預(yù)測(cè)研究[J];四川理工學(xué)院學(xué)報(bào)(自然科學(xué)版);2015年06期

6 舒堅(jiān);郭凱;劉群;劉琳嵐;;機(jī)會(huì)傳感網(wǎng)絡(luò)連通性參數(shù)研究[J];計(jì)算機(jī)學(xué)報(bào);2016年05期

7 劉峰波;;大數(shù)據(jù)Spark技術(shù)研究[J];數(shù)字技術(shù)與應(yīng)用;2015年09期

8 李旭芳;段春林;張冬波;韓迎春;茍茹君;;遙測(cè)數(shù)據(jù)時(shí)間序列滑動(dòng)窗口動(dòng)態(tài)分割技術(shù)[J];飛行器測(cè)控學(xué)報(bào);2015年04期

9 劉國(guó)奇;毛海宇;蒲寶明;朱永峰;黃金;;基于小波神經(jīng)網(wǎng)絡(luò)的風(fēng)機(jī)故障診斷[J];小型微型計(jì)算機(jī)系統(tǒng);2015年07期

10 肖輝輝;萬(wàn)常選;段艷明;;一種基于復(fù)合形法的花朵授粉算法[J];小型微型計(jì)算機(jī)系統(tǒng);2015年06期

相關(guān)博士學(xué)位論文 前1條

1 程興國(guó);仿生算法的動(dòng)態(tài)反饋機(jī)制及其并行化實(shí)現(xiàn)方法研究[D];華南理工大學(xué);2013年

相關(guān)碩士學(xué)位論文 前10條

1 王騰;基于VxWorks的風(fēng)電機(jī)組數(shù)據(jù)采集處理及控制系統(tǒng)研究[D];北京交通大學(xué);2015年

2 李文棟;基于Spark的大數(shù)據(jù)挖掘技術(shù)的研究與實(shí)現(xiàn)[D];山東大學(xué);2015年

3 李偉;風(fēng)電機(jī)組狀態(tài)監(jiān)測(cè)與故障診斷系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[D];華南理工大學(xué);2014年

4 李華;基于信息融合的風(fēng)電機(jī)組狀態(tài)監(jiān)測(cè)研究[D];內(nèi)蒙古科技大學(xué);2014年

5 唐振坤;基于Spark的機(jī)器學(xué)習(xí)平臺(tái)設(shè)計(jì)與實(shí)現(xiàn)[D];廈門(mén)大學(xué);2014年

6 齊佳;基于LMD的風(fēng)力發(fā)電機(jī)組振動(dòng)信號(hào)分析[D];哈爾濱理工大學(xué);2014年

7 童超;基于數(shù)據(jù)挖掘方法的風(fēng)電機(jī)組狀態(tài)監(jiān)測(cè)研究[D];華北電力大學(xué);2014年

8 鄭思莉;基于小波變換的遙感圖像降噪及質(zhì)量評(píng)價(jià)研究[D];武漢理工大學(xué);2013年

9 王鳳霞;基于小波神經(jīng)網(wǎng)絡(luò)的風(fēng)力發(fā)電機(jī)組故障診斷方法的研究[D];華北電力大學(xué);2013年

10 王鵬;群智能算法的并行化研究及其在圖像配準(zhǔn)中的應(yīng)用[D];江南大學(xué);2008年

，

本文編號(hào)：1899913