中国燃煤电厂二氧化碳排放量计算方法研究

发布时间：2018-02-26 22:21

本文关键词： 二氧化碳 BP神经网络计算方法燃煤电厂　出处：《北京交通大学》2014年硕士论文　论文类型：学位论文

【摘要】：气候变化已成为各国进行政治、经济和文化博弈的重要议题。因温室效应引起的环境问题逐渐引起了人们的关注。为全面控制二氧化碳等温室气体的排放,以缓解气候变暖给人类经济和社会带来的不利影响,国际各国开始纷纷采取行动——约束排放和减少排放,共同为应对气候变化做出努力。控制减排的首要环节是了解当下二氧化碳排放情况。电力行业是二氧化碳主要排放行业之一,国际上虽然已经有大量关于燃煤电厂二氧化碳排放量计算的方法研究,但大多是依据各自国家的煤炭统计数据、电力设备运行状况等设计的。我国煤炭分布不均,质量参差不齐,其质量对于发电设备影响极大。而且,电力相关统计资料并不完善,电力运行工况与国外相比存在较大差异,若直接套用国际现有方法,必然会与真实值之间存在很大误差。由于国外的方法学理念较为完善,因此借鉴国外方法建立符合中国国情的燃煤电厂二氧化碳排放计算方法是一种省时省力又较准确的计算方法。本文深入探讨了我国煤炭质量情况及其对发电性能的影响。首先,我国煤炭分布情况及煤炭质量特征显示我国煤炭资源分布不均,地区间煤炭质量差异较大,且煤炭指标如灰分、硫分、挥发分等指标数据与国外指标存在显著差异。而且,煤炭资源分布与消费分布极不协调,江苏、浙江、山东、广东等需求量较高的地区煤炭资源却较为贫瘠,致使电煤供应成为制约电煤质量的一大因素；这些地区实际用煤质量波动较大,多不符合设计煤质要求。本文选择10家典型电厂作为主要研究对象,分别从煤炭发热量、灰分、硫分、水分等煤质指标入手,分析其对发电设备的影响,结果显示《IPCC指南》中的缺省系数无法直接应用于我国电厂二氧化碳排放计算中为了更准确的建立我国燃煤电厂二氧化碳排放量的计算方法,本文结合电厂设备运行理论,通过工业分析数据(全水分Mar、收到基灰分Aar、收到基挥发分Var、固定碳FCar四个数据)预测收到基含碳量Car,继而通过锅炉燃烧理论,得出燃煤发电过程和脱硫过程的计算公式。在工业分析数据预测收到基含碳量Car时,采用BP神经网络的非线性映射特征,利用Matlab建立可通过工业分析数据(全水分Mar、收到基灰分Aar、收到基挥发分Var、固定碳FCar)预测Car的神经网络模型。通过网络学习与优化,最终使得学习后的数据预测值的相对误差绝对值为0.602%,新数据预测结果的相对误差绝对值平均可降低至2.827%。为了验证上述计算方法的准确性,以江苏某发电厂为例,利用BP神经网络模型预测,收到基平均值Car的相对误差可降至0.24%,通过计算,该燃煤电厂固定源二氧化碳排放量为4.923×106t/n。利用《2006IPCC指南》中提供的缺省因子计算所得二氧化碳排放量为5.244×106t/n,高于电厂实际二氧化碳排放量6.5个百分点。
[Abstract]:Climate change has become an important issue in the political, economic and cultural game between countries. The environmental problems caused by Greenhouse Effect have gradually attracted people's attention. In order to comprehensively control greenhouse gas emissions such as carbon dioxide, In order to mitigate the adverse effects of global warming on human economy and society, international countries have begun to take actions to curb emissions and reduce emissions, and make joint efforts to deal with climate change. The first step in controlling emission reduction is to understand the current situation of carbon dioxide emissions. The power industry is one of the major carbon dioxide emission industries. Although there has been a lot of international research on the calculation methods of carbon dioxide emissions from coal-fired power plants, However, most of them are designed on the basis of the coal statistics of their respective countries and the operation status of power equipment. The distribution of coal in China is uneven, the quality of coal is uneven, and the quality of coal has a great impact on the power generation equipment. Moreover, the statistical data related to electricity are not perfect. There is a great difference between the operating conditions of electric power and foreign countries. If the existing international methods are applied directly, there is bound to be a great error between the actual value and the actual value. Therefore, it is a time-saving and labor-saving and accurate calculation method to establish the calculation method of carbon dioxide emissions of coal-fired power plants in accordance with China's national conditions by using foreign methods for reference. In this paper, the coal quality in China and its influence on power generation performance are discussed. Firstly, the coal distribution and coal quality characteristics in China show that the distribution of coal resources in China is uneven, and the coal quality varies greatly among regions. And the coal index such as ash, sulfur, volatile matter and so on index data have the remarkable difference with the foreign index, moreover, the coal resources distribution and the consumption distribution are extremely inharmonious, Jiangsu, Zhejiang, Shandong, The coal resources in high demand areas such as Guangdong are relatively poor, resulting in the supply of thermal coal becoming a major factor restricting the quality of thermal coal, and the actual quality of coal used in these areas fluctuates greatly. In this paper, 10 typical power plants are selected as the main research objects, starting with coal calorific value, ash content, sulphur content, moisture content and so on, the influence of coal quality on power generation equipment is analyzed. The results show that the default coefficient in IPCC Guide can not be directly applied to the calculation of carbon dioxide emissions from power plants in China. In order to establish a more accurate calculation method of carbon dioxide emissions from coal-fired power plants in China, this paper combines the operation theory of power plant equipment, Based on the data of industrial analysis (all moisture, received base ash, base ash, base volatile, fixed carbon FCar), the basic carbon content Carr is predicted, and the calculation formula of coal-fired power generation process and desulfurization process is obtained through boiler combustion theory. When the base carbon content (Car) is predicted by the industrial analysis data, the nonlinear mapping feature of BP neural network is used. A neural network model for predicting Car by industrial analysis data (total moisture Marr, received base ash Aarus, received base volatile matter Vara, fixed carbon FCars) was established by using Matlab. Finally, the absolute value of the relative error of the predicted data after learning is 0.602, and the absolute value of the relative error of the new data can be reduced to 2.827 on average. In order to verify the accuracy of the above calculation methods, taking a power plant in Jiangsu province as an example, the relative error of the base average Car can be reduced to 0.24 by using BP neural network model. The fixed source carbon dioxide emission of the coal-fired power plant is 4.923 脳 10 ~ (6) t / n. using the default factors provided in the 2006 IPCC guidelines, the calculated carbon dioxide emissions are 5.244 脳 10 ~ (6) t / n, which is 6.5 percentage points higher than the actual carbon dioxide emissions of the power plant.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：X773

【参考文献】