輻射光譜法電站鍋爐燃燒檢測診斷研究
發(fā)布時間：2018-07-26 10:19
【摘要】：燃煤火電機組是我國電力工業(yè)的支柱,占主導地位。因此,準確有效的監(jiān)測現(xiàn)代燃煤電站鍋爐爐內(nèi)火焰,對電站鍋爐和電廠的高效安全運行具有十分重要的意義。本文基于自行研制的以光纖光譜儀為核心的火焰檢測系統(tǒng)對燃料燃燒火焰和電站鍋爐爐內(nèi)燃燒狀況進行了研究,提出了對光譜火焰測溫法的改進方法和一種簡單方便并適合在線測量的火焰光譜發(fā)射率測量方法,自行研制了火檢探針,在一臺300MW煤粉鍋爐上開展了試驗研究,給出了利用該火檢探針在電站鍋爐實測的結果,并對結果進行了深入的分析。本文的主要內(nèi)容及創(chuàng)新點如下:(1)火焰溫度是爐內(nèi)燃燒監(jiān)測診斷的一個重要參數(shù),基于CCD光纖光譜儀的火焰溫度檢測方法因其具有獨特的優(yōu)點而成為火焰溫度測量領域的研究熱點之一。由于CCD本身的光譜響應特性,光譜儀需要用黑體爐進行標定。但當標定溫度和被測溫度相差較大時,測得的被測物體溫度與實際溫度誤差較大。另外,當光譜儀測得的響應強度與標定時對黑體爐的響應強度相差較大時,得到的溫度值也會存在較大誤差。本文就上述兩種誤差提出了解決方法,實驗證明改進之后測量誤差大大減小,使得基于光纖光譜儀的火焰輻射測溫范圍大幅度擴大。(2)以往在線測量非灰體火焰光譜發(fā)射率的方法,由于理論不完善,存在諸多缺陷。本文主要針對不同燃料的燃燒火焰,提出一種基于近紅外光譜儀的測量火焰光譜發(fā)射率的方法,并與其他火焰光譜發(fā)射率模型進行了比較和驗證。另外,作者還提出了一種基于火焰光譜發(fā)射率的混合燃料組份的檢測方法。(3)火焰的閃動頻率,作為火焰穩(wěn)定性的一個參照,是火焰檢測和燃燒診斷的一個重要判據(jù)。但是由于在工業(yè)爐膛內(nèi)的火焰信號具有較寬的頻譜,在信號的整個頻譜范圍內(nèi)都包含噪聲,并且噪聲在高頻范圍內(nèi)占支配地位,在對信號進行處理之前必須對信號作消噪處理,這樣才能高效提取信號中的有用信息。本文中利用小波閾值法對電站鍋爐中采集到的光譜信號進行消噪,然后對消噪后的信號作快速傅里葉變換得到火焰的閃動頻率。(4)在利用光譜法研究溫度、輻射率和火焰閃動頻率的測量方法之后,研制了火檢探針,本文給出了該火檢探針在電站鍋爐實測的結果,并對結果進行了深入的分析,為基于這三個參數(shù)提出一套鍋爐燃燒診斷方案奠定了基礎。這也是本課題研究的最終目的:建立火力發(fā)電站中燃煤鍋爐的火焰檢測和燃燒診斷體系,為電站鍋爐的高效穩(wěn)定安全運行和變負荷時各參數(shù)的調節(jié)提供正確高效的依據(jù)和保障。
[Abstract]:Coal-fired thermal power unit is the mainstay of our country's electric power industry, occupies the leading position. Therefore, the accurate and effective monitoring of the flame in the furnace of the modern coal-fired power plant boiler is of great significance to the efficient and safe operation of the utility boiler and the power plant. In this paper, the flame detection system based on fiber-optic spectrometer has been developed to study the combustion status of fuel combustion flame and boiler furnace. In this paper, an improved method of spectral flame temperature measurement and a simple and convenient method for measuring flame spectral emissivity are presented. The flame detection probe is developed, and the experimental study is carried out on a 300MW pulverized coal boiler. The measured results of the fire detection probe in the boiler of power station are given, and the results are analyzed in depth. The main contents and innovations of this paper are as follows: (1) Flame temperature is an important parameter for combustion monitoring and diagnosis in furnace. The flame temperature detection method based on CCD optical fiber spectrometer has become one of the research hotspots in flame temperature measurement field because of its unique advantages. Because of the spectral response of CCD, the spectrometer needs to be calibrated by blackbody furnace. However, when the difference between the calibrated temperature and the measured temperature is large, the error between the measured object temperature and the actual temperature is larger. In addition, when the response intensity measured by the spectrometer is different from the response intensity of the calibration to the blackbody furnace, there is also a large error in the obtained temperature value. In this paper, the methods of solving the above two kinds of errors are put forward. The experimental results show that the measurement error is greatly reduced after the improvement. The range of flame radiation temperature measurement based on fiber-optic spectrometer is greatly expanded. (2) the methods of on-line measurement of flame spectral emissivity of non-gray body have many defects due to the imperfect theory. In this paper, a method of measuring flame spectral emissivity based on near infrared spectrometer is proposed, and compared with other flame spectral emissivity models. In addition, the author also proposes a method for the detection of mixed fuel components based on flame spectral emissivity. (3) Flame flash frequency, as a reference for flame stability, is an important criterion for flame detection and combustion diagnosis. However, because the flame signal in the furnace has a wide spectrum, noise is included in the whole spectrum range of the signal, and the noise dominates in the high frequency range, so the signal must be de-noised before the signal is processed. In this way, useful information can be extracted efficiently. In this paper, the wavelet threshold method is used to de-noise the spectral signal collected in the boiler of power station, and then the fast Fourier transform of the de-noised signal is used to get the flash frequency of the flame. (4) the temperature is studied by using the spectral method. After measuring the emissivity and flame flashover frequency, a fire detector probe is developed. The measured results of the probe in the boiler of power station are given, and the results are analyzed deeply. It lays a foundation for a boiler combustion diagnosis scheme based on these three parameters. This is the ultimate purpose of this study: to establish the flame detection and combustion diagnosis system of coal-fired boilers in thermal power stations, to provide a correct and efficient basis and guarantee for the efficient, stable and safe operation of utility boilers and the adjustment of the parameters when the load is changed.
【學位授予單位】：上海理工大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TM621.2
本文編號：2145683
[Abstract]:Coal-fired thermal power unit is the mainstay of our country's electric power industry, occupies the leading position. Therefore, the accurate and effective monitoring of the flame in the furnace of the modern coal-fired power plant boiler is of great significance to the efficient and safe operation of the utility boiler and the power plant. In this paper, the flame detection system based on fiber-optic spectrometer has been developed to study the combustion status of fuel combustion flame and boiler furnace. In this paper, an improved method of spectral flame temperature measurement and a simple and convenient method for measuring flame spectral emissivity are presented. The flame detection probe is developed, and the experimental study is carried out on a 300MW pulverized coal boiler. The measured results of the fire detection probe in the boiler of power station are given, and the results are analyzed in depth. The main contents and innovations of this paper are as follows: (1) Flame temperature is an important parameter for combustion monitoring and diagnosis in furnace. The flame temperature detection method based on CCD optical fiber spectrometer has become one of the research hotspots in flame temperature measurement field because of its unique advantages. Because of the spectral response of CCD, the spectrometer needs to be calibrated by blackbody furnace. However, when the difference between the calibrated temperature and the measured temperature is large, the error between the measured object temperature and the actual temperature is larger. In addition, when the response intensity measured by the spectrometer is different from the response intensity of the calibration to the blackbody furnace, there is also a large error in the obtained temperature value. In this paper, the methods of solving the above two kinds of errors are put forward. The experimental results show that the measurement error is greatly reduced after the improvement. The range of flame radiation temperature measurement based on fiber-optic spectrometer is greatly expanded. (2) the methods of on-line measurement of flame spectral emissivity of non-gray body have many defects due to the imperfect theory. In this paper, a method of measuring flame spectral emissivity based on near infrared spectrometer is proposed, and compared with other flame spectral emissivity models. In addition, the author also proposes a method for the detection of mixed fuel components based on flame spectral emissivity. (3) Flame flash frequency, as a reference for flame stability, is an important criterion for flame detection and combustion diagnosis. However, because the flame signal in the furnace has a wide spectrum, noise is included in the whole spectrum range of the signal, and the noise dominates in the high frequency range, so the signal must be de-noised before the signal is processed. In this way, useful information can be extracted efficiently. In this paper, the wavelet threshold method is used to de-noise the spectral signal collected in the boiler of power station, and then the fast Fourier transform of the de-noised signal is used to get the flash frequency of the flame. (4) the temperature is studied by using the spectral method. After measuring the emissivity and flame flashover frequency, a fire detector probe is developed. The measured results of the probe in the boiler of power station are given, and the results are analyzed deeply. It lays a foundation for a boiler combustion diagnosis scheme based on these three parameters. This is the ultimate purpose of this study: to establish the flame detection and combustion diagnosis system of coal-fired boilers in thermal power stations, to provide a correct and efficient basis and guarantee for the efficient, stable and safe operation of utility boilers and the adjustment of the parameters when the load is changed.
【學位授予單位】：上海理工大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TM621.2
【參考文獻】
相關博士學位論文 前1條
1 宋揚;光譜發(fā)射率在線測量技術研究[D];哈爾濱工業(yè)大學;2009年
，本文編號：2145683
本文鏈接：http://sikaile.net/kejilunwen/dianlilw/2145683.html
