【作者】 李钧；

【导师】 阎维平；

【作者基本信息】 华北电力大学（河北） ， 热能工程， 2010， 博士

【摘要】 煤粉锅炉燃烧监测和优化,是在确保锅炉安全性的前提下,通过燃烧调整,达到提高燃烧效率和降低污染物排放的目标。本文以某电站300MW四角切圆燃烧煤粉锅炉为具体研究对象,深入分析了控制氮氧化物排放与降低飞灰可燃物含量之间的矛盾,采用预数值计算方法,研究煤粉锅炉燃烧监测和优化。探讨了把计算结果精确但耗时较长的数值计算方法,应用于锅炉燃烧监测和优化。为四角切圆燃烧煤粉锅炉燃烧监测和优化提供了一种有效的途径。具体研究内容及主要研究成果如下:(1)应用预数值计算方法,对煤粉锅炉的燃烧特性进行研究。利用四角切圆燃烧煤粉锅炉专用数值计算软件COALFIRE,对锅炉28个运行工况进行了精确预数值计算,得到了锅炉燃用煤质、煤粉细度、锅炉负荷和二次风配风方式等参数变化对炉内温度、氧浓度、一氧化碳浓度、辐射受热面热负荷,特别是飞灰可燃物含量、氮氧化物释放的影响规律。计算结果与实际运行数据吻合良好,为实现锅炉燃烧监测和优化提供了比较精确的数据支持。(2)以预数值计算结果作为训练样本,建立了基于支持向量机的锅炉燃烧工况预测模型。支持向量机算法的特点是能根据有限的样本信息,在模型的复杂性和学习能力之间寻求折衷,获得较好推广能力。利用该算法解决锅炉燃烧工况影响因素复杂,获得的工况样本数量有限的问题,实现锅炉燃烧工况的监测,并与实际运行数据对比,对模型进行了验证。结果表明所建模型对燃烧工况的变化响应灵敏,结果准确,能够满足燃烧监测的要求。(3)采用因子分析法,对锅炉燃烧工况的评价指标进行降维,定义评价燃烧工况优劣的安全性、经济性和环保性因子,推导了各因子的计算式。在提取燃烧工况的公共因子后,通过计算燃烧工况的综合评价得分,进而建立燃烧工况评判模型,并结合具体工况进行验证,结果表明,基于因子分析法的燃烧工况评价模型,评价结果合理,可用于指导锅炉燃烧调整与优化。(4)采用Visual Basic语言开发了基于预数值计算的四角切圆煤粉锅炉燃烧监测与优化系统,并进行了离线运行与测试。结果表明通过该系统能够实现对燃烧工况的监测和优化,为煤粉锅炉达到高效燃烧和低污染物排放的优化运行,提供了新的途径。更多还原

【Abstract】 The combustion monitoring and optimization of pulverized coal utility boiler aims to improve the combustion efficiency and reduce the emission of pollutions on the premise of guaranteeing the safe operation of the boiler by the methods of adjusting the boiler operating parameters. A 300MW tangential pulverized coal-fired boiler was taken as a study case. The contradiction between the unburned carbon in fly ash and emission of nitrogen oxide was analyzed. The pre-numerical calculation method was used to study the combustion monitoring and optimization. The application of numerical calculation, which was accurate but time consuming on the boiler real-time combustion monitoring and optimization, was discussed. The method offered an effective way for the combustion monitoring and optimization. Concrete research contents and main results were as follows:(1) The pre-numerical calculation was applied in the research of combustion characteristics in furnace. COALFIRE was used as an effective tool for numerical calculation of combustion and heat transfer of the tangentially pulverized coal-fired boiler. Total 28 operating conditions were calculated using COALFIRE. The influences of coal properties, pulverized coal size, boiler load and the distribution of secondary air to temperature, oxygen concentration and thermal load of radiation surfaces were discussed, with the emphasis on the variation of unburned carbon in fly ash and the released rule of nitrogen oxide. Compared with the actual operating data, the calculation result was accurate. Pre-numerical calculation offered the data base to the further work.(2) Taking the pre-numerical calculation results as the training samples, the predict model of combustion monitoring was established based on the Support Vector Machine. The Support Vector Machine has the characteristics of splitting the difference between the complexities and learning ability of model based on the limited sample information. Therefore, the model could obtain better popularizing ability. It adapted to solving the combustion problems because getting the operating samples was difficult. Compared with the calculated data and the actual operating data, the results indicated that the model responded the changes sensitively and the calculation results were reasonable, which met the request of the combustion monitoring. (3) The evaluation indexes of the combustion conditions were reduced using the factor analysis method. The security, economic and environmental factors were defined. The formula of each factors were deduced. After drawing the public factors, the combustion evaluation model was established by calculating the scores of different operating mode. The model was tested using the actual operating condition. The result showed that the model was reasonable. It can be used as a reference to direct the boiler combustion adjustment and optimization.(4) The combustion monitoring and optimization system based on pre-numerical calculation was developed using Visual Basic. The result of off-line running test indicated that it could realize the combustion monitoring and optimization. It offered a new way for the combustion monitoring and optimization on tangentially pulverized coal-fired boiler.更多还原