【作者】 张向宇；

【导师】 周怀春； 程强；

【作者基本信息】 华中科技大学 ， 热能工程， 2011， 博士

【摘要】 工业炉是工业加热的关键设备,广泛应用于国民经济的许多行业,同时工业炉又是高能耗设备。如何节约能源,提高能源利用水平,是目前工业炉研究的当务之急。实现温度场的实时在线测量是使工业炉处于最佳燃烧状态的重要前提,提高工业炉炉膛温度场检测技术,控制炉膛温度均匀分布有助于实现工业炉的安全、优化运行。工业炉温度场检测的困难在于炉膛的大尺寸以及燃烧火焰的强脉动等特点,采用常用的热电偶、高温计、声学法、以及基于激光的光学法等都难以实时在线检测炉内三维温度分布。辐射图像处理结合辐射传递逆求解的方法已经被证实可以实现炉内温度场的可视化,并且已经在电站锅炉上得到应用。但与电站锅炉不同的是,工业炉壁面温度很高,壁面辐射不能被忽略,这使得工业炉内辐射传递过程更加复杂。同时,工业炉种类繁多,燃料类型多变,燃烧过程及中间燃烧产物更加复杂,炉内辐射特性参数更难以准确把握。因此,本文将研究如何在工业炉内利用火焰图像处理与辐射传递逆求解来实现炉内三维温度场可视化。具体工作如下：本文首先从工业炉与电站锅炉辐射过程的不同点出发,研究了高温壁面条件对辐射成像过程的影响。通过采用DRESOR法对辐射传递方程进行求解,获得了炉膛燃烧温度与火焰辐射图像间的定量传递关系,建立了适用于工业炉的辐射成像模型。燃烧火焰的辐射特性参数,是基于辐射传递逆求解的测温方法所必须的输入参数。为实现温度场与辐射特性参数的解耦,本文提出了一种正则化方法与最优化方法相结合的同时重建方法。通过在炉膛边界处安装CCD火焰探测器以获取炉内火焰在红、绿波长下的近似单色辐射强度图像,利用正则化方法从红色单色辐射强度信息中重建炉内温度分布,同时以绿色单色辐射强度为最优化目标重建炉内辐射参数。模拟研究表明,在不同的测量误差下,重建算法均能够较好地还原炉膛温度分布及辐射特性参数。进一步,利用CCD火焰图像探测器和相应的计算机图像采集处理系统,在一台热态试验炉上开展了燃烧火焰三维温度场与辐射特性参数同时重建的实验研究,验证了解耦重建算法在工业应用上的可行性。进一步对热态试验炉炉内温度分布开展了实时检测研究,通过比较三个测点处的可视化方法与热电偶测量结果,测温误差都在5%以内,风管表面最大测温误差20℃。考虑到同时重建算法非常耗时,不能满足工业现场实时测温的需要,本文提出了一种新的温度场快速重建算法--改进比色法。该方法将传统的比色法从单点测温扩展到处理多维、非均匀温度重建问题上,能够减小辐射参数不准确对温度重建的影响,能够获得比单色法更高精度的温度反演效果。最后,以一台步进式加热炉为实验对象开展了工业炉三维温度场可视化研究。通过在4个加热段的南北侧墙上各安装两支CCD火焰探测器,构建了一套加热炉三维温度场在线监测系统。以同时重建算法得到的辐射参数收敛值作为全炉膛辐射参数近似分布,利用改进比色法对炉膛三维温度场开展了可视化研究。通过与现场热电偶测量数据的对比,表明三维温度场可视化系统能够实时、准确地测量步进梁式加热炉内温度变化,给出炉内三维空间和板坯表面加热温度的不均匀性信息。同时,通过与现场计算机控制系统相互通信,可视化系统实现了加热钢坯的在线监测。综上所述,本文采用火焰图像处理技术结合辐射逆问题求解实现了工业炉内的三维温度场可视化,同时从火焰图像中反演出燃烧介质以及高温壁面的辐射特性参数。此项技术的深入研究,能够提高工业炉内加热工件质量,节省燃料量,同时减少污染排放,提高工业炉安全和经济运行水平。更多还原

【Abstract】 As the pivotal heating equipment, industrial furnace was widely used in domestic industrial economy, which was also high energy consumption. The problem of how to improve energy utilization became the top priority of the research of industrial furnaces. Real-time temperature measurement is the premises for optimal combustion of furnace. Improving the temperature detection skill and achieving uniform temperature distribution maybe helpful to safe and optimize operation of industrial furnace.The difficulty of temperature measurement of industrial furnace lies in the large size of the furnace and the strong pulse of the burning flame, which make it difficult to detection the three-dimensional temperature distribution by commonly used thermocouples, pyrometers, acoustic method, and laser-based method. Radiation image processing combined with radiative inverse solving has been proved to be effect for temperature visualization, which has been applied in the power plant boiler. But different from the boiler, the wall temperature of industrial furnace is high, and the wall radiation can not be ignored, which makes the radiation process in the furnace more complicated. Meanwhile, the wide range of industrial furnaces, the variable of fuel type, the more complex of combustion process and the intermediate products of combustion, would make the radiation parameters more difficult to grasp. Therefore, this article will concentrate on how the use the flame image processing and radiative inverse solving to realize the three-dimensional temperature visualization in industrial furnace. Specifically as follows:Firstly, we researched the impact of high wall temperature on radiation imaging process in the industrial furnace, originated from the difference of the furnace and boiler. By using the DRESOR method to solve the radiative transfer equation, we obtained the transitive relation between combustion temperature and flame images, and established a single wavelength radiation imaging model for industrial furnaces.Flame radiation parameter is the prerequisite input for temperature measurement by inverse radiative transfer method. In order to decouple the temperature and radiation parameters, this paper presents a new simultaneously reconstruction method which combined by regularization method and optimization method. Firstly, the flame monochromatic radiation intensity under two wavelengths corresponding to red and green color were get by CCD cameras which were installed on the boundary of furnace, then red monochromatic intensity was used to reconstruct the temperature distribution by regularization method, while the green monochromatic intensity was used fo the optimization target to rebuild radiation parameters. Simulation showed that the reconstruction algorithms were able to reconstruct the furnace temperature distribution and radiation parameters under different measurement errors.Further, experimental study of simultaneous reconstruction of temperature and radiation parameters were carried out on a hot experimental furnace by using CCD detectors and the corresponding computer image acquisition and processing system, which verified the feasibility of the reconstruction algorithm. Further tests of real-time temperature detection were carried out in the hot experimental furnace. Comparision of three measurement points between thermocouples and visualization system showed that, the measurement error in the whole process was less than 5% and the duct surface temperature measurement error was no more than 20℃.The simultaneous reconstruction algorithm was very time consuming, which could not meet the needs of real-time temperature measurement in the furnace. This paper presented a new algorithm for fast temperature reconstruction-Improved colorimetry method. In ths method, the traditional colorimetric method was extended from single-point temperature calculation to solve the multi-dimensional, non-uniform temperature reconstruction. At the same time, the improved colorimetry method could reduce the influence of inaccuracy of radiation parameters, which could obtain a more accurate temperature reconstruction result than monochromatic method.At last, three-dimensional temperature visualization of furnace was researched on a walking beaming reheating furnace. A monitoring system was established by installing two CCD cameras on north and south walls of every four sections. The convergent values of simultaneous reconstruction algorithm were set as the approximate radiation parameters distribution in the furnace, and then the three dimensional temperature were visiualized by improved colorimetric method. Compared with thermocouples, the visiualization system could deduct the three-dimensional temperature in the walking beam reheating furnace in real time and accurately, which could also give the asymmetric heating information of furnace and the slab surface. Meanwhile, online monitoring of billet heating was realized by communicating with the on-site computer control system.In summary, the three-dimensional temperature visualization and deduction of radiation properties of media and high-temperature wall in industrial furnace was realized by the combination method of flame image processing and radiation inverse solving. Further research of this technology would improve the workpiece heating quality in industrial furnace, reduce the fuel consumption and pollution emissions, and finally, achieve the safe and economic operation of industrial furnaces.更多还原