共查询到17条相似文献,搜索用时 187 毫秒
1.
2.
燃煤锅炉是复杂的多变量系统,其飞灰的含碳量形成机理复杂,不能用简单的数学公式估算。现场实炉测试这些数据具有工作量大,测试工况有限等缺点;燃煤锅炉运行参数及燃料特性等因素影响着飞灰的含碳量,其相互耦合,导致分析数据过程困难。神经网络建模将燃煤锅炉视为黑箱,应用该方法可以良好的描述其输入输出之间的黑箱特性,因此,人工神经网络应用广泛。利用燃煤锅炉试验数据,采用3层BP(back propagation)神经网络构建了锅炉飞灰的含碳量排放特性模型。通过锅炉的实测数据验证,该BP神经网络对飞灰含碳量相对预测误差在0.19%~0.50%,预测效果良好。测试结果表明,建立的神经网络预测模型可以准确逼近验证样本数据,也能够较好的逼近非验证样本数据,具有良好的泛化能力。 相似文献
3.
飞灰含碳量是反映燃煤锅炉机组燃烧效率的重要技术指标和运行经济指标,同时也影响锅炉的安全运行。超临界对冲火焰锅炉由于掺烧劣质煤,经常出现飞灰含碳量偏高的现象。本文以660MW超临界对冲火焰锅炉为研究对象,将影响飞灰含碳量的负荷、煤粉细度等十个运行参数作为输入量,应用BP神经网络的非线性动力学特性和自学习能力,建立了飞灰含碳量预测模型。经网络预测,与实际值的误差小于5.48%。在预测模型的基础上,对飞灰含碳量影响因素进行单因素影响规律分析。预测和分析结果表明,本模型方法能有效提取各参数对飞灰含碳量的影响规律,可用于锅炉飞灰含碳量的分析、预测和优化调节。 相似文献
4.
5.
6.
提出了采用BP神经网络模型与改进热力计算相结合的方法确定锅炉运行参数基准值。计算中采用BP神经网络模型预测飞灰含碳量的基准值,并根据锅炉运行负荷选取炉膛出口烟气温度计算公式,采用登山原理确定过量空气系数的方法确定关键运行参数基准值。最后,以一台HG1025/18.2-M锅炉为例,计算70%、50%负荷下该锅炉运行参数的基准值,得到随着锅炉负荷的降低炉膛出口过量空气系数明显增加,飞灰含碳量和机械未完全燃烧热损失显著降低。证明该方法能够很好地反映锅炉负荷、煤质特性参数改变对运行参数基准值的影响。 相似文献
7.
8.
电站锅炉飞灰含碳量的优化控制 总被引:11,自引:1,他引:11
利用人工神经网络对锅炉飞灰含碳量进行建模,并采用混合遗传算法与复合形法进行运行工况寻优,获得当前最佳的锅炉燃烧调整方式,这种方法同时解决了锅炉变工况下运行参数基准值的问题。应用该模型对某台300MW四角切圆燃煤电站锅炉的飞灰含碳量进行优化控制研究,其结果可指导运行人员进行参数优化调整,降低锅炉飞灰含碳量,提高燃烧经济性。图3表4参7 相似文献
9.
利用人工神经网络对锅炉飞灰含碳量进行建模,并采用混合遗传算法与复合形法进行运行工况寻优,获得当前最佳的锅炉燃烧调整方式,这种方法同时解决了锅炉变工况下运行参数基准值的问题。应用该模型对某台300MW四角切圆燃煤电站锅炉的飞灰含碳量进行优化控制研究,其结果可指导运行人员进行参数优化调整,降低锅炉飞灰含碳量,提高燃烧经济性。 相似文献
10.
11.
飞灰含碳量是反映电站煤粉锅炉燃烧效率的一个重要指标。基于误差反向传播(BP)神经网络方法,建立了11-23-1型BP神经网络模型。根据某电站四角切圆煤粉锅炉特点选取了煤粉细度、燃烧器摆角、烟气含氧量、5个煤种参数、燃烧器喷口运行组合等11个影响燃烧的参数作为神经网络的输入因子,对建立的模型进行训练,得到模型参数。以此进行预测,与实际值的误差不超过6%。在此基础上,又提出了单参数影响飞灰含碳量的简化分析方法,使神经网络包含的多维非线性规律在一定条件下简洁、直观地反映出来。计算和分析结果表明,本模型方法能有效提取各参数对飞友含碳量的影响规律,可用于锅炉飞灰含碳量的分析、预测和优化调节。 相似文献
12.
13.
This paper introduces a novel on-line monitoring performance method of coal-fired power unit. Support vector machine (SVM) is used to predict the unburned carbon content of fly ash in the boiler and the exhaust steam enthalpy in turbine, which are two difficulties in the real time economic performance calculation model in coal-fired power plant. Comparison between the output of SVM modeling and the experimental data shows a good agreement, and compared with conventional artificial neural network techniques, SVM can achieve better accuracy and generalization. This presented monitoring method is proven by the results of application cases in a practical coal-fired power plant. 相似文献
14.
15.
16.
Li JIA Baoguo FAN Xianrong ZHENG Xiaolei QIAO Yuxing YAO Rui ZHAO Jinrong GUO Yan JIN 《Frontiers in Energy》2021,15(1):112-123
The mercury emission was obtained by measuring the mercury contents in flue gas and solid samples in pulverized coal (PC) and circulating fluidized bed (CFB) utility boilers. The relationship was obtained between the mercury emission and adsorption characteristics of fly ash. The parameters included unburned carbon content, particle size, and pore structure of fly ash. The results showed that the majority of mercury released to the atmosphere with the flue gas in PC boiler, while the mercury was enriched in fly ash and captured by the precipitator in CFB boiler. The coal factor was proposed to characterize the impact of coal property on mercury emissions in this paper. As the coal factor increased, the mercury emission to the atmosphere decreased. It was also found that the mercury content of fly ash in the CFB boiler was ten times higher than that in the PC boiler. As the unburned carbon content increased, the mercury adsorbed increased. The capacity of adsorbing mercury by fly ash was directly related to the particle size. The particle size corresponding to the highest content of mercury, which was about 560 ng/g, appeared in the range from 77.5 to 106 µm. The content of mesoporous (4–6 nm) of the fly ash in the particle size of 77.5–106 µm was the highest, which was beneficial to adsorbing the mercury. The specific surface area played a more significant role than specific pore volume in the mercury adsorption process. 相似文献