共查询到20条相似文献,搜索用时 421 毫秒
1.
2.
3.
借助锅炉燃烧特性试验数据,建立了基于支持向量回归的四角切圆燃煤锅炉NOx排放模型.经过训练和校验,并与神经网络模型进行对比,结果表明:SVR模型更加适合于实炉热态测试工况较少的小样本学习,而且其精度能够满足工程的实际要求,能够较为准确的对不同工况下的电站锅炉NOx排放特性进行预测.在获得该模型的基础上,结合全局寻优的遗传算法,以锅炉的运行调节参数为优化目标函数的自变量,对NOx排放进行寻优,并获得了具体工况下的最佳操作参数. 相似文献
4.
提出了一种锅炉运行优化的系统框架,并着重对建模方法和优化算法进行了讨论。采用一种改进的在线支持向量机方法对学习样本进行处理,并与常规支持向量机模型相结合,形成了一种自适应建模方法,以适应煤质与锅炉实际运行工况的变化。对多目标优化算法进行讨论,并介绍了NSGA II遗传算法在锅炉优化中的应用。以某大型电站锅炉为对象,对本文方法进行了应用研究,所建模型的趋势分析和测试结果均表明了本文算法的正确性,所提出的优化模型以NOx为目标函数,综合考虑了锅炉运行经济性和安全性等约束条件,优化结果表明通过运行参数的优化调整可有效降低锅炉污染物的排放。 相似文献
5.
6.
电站锅炉燃烧系统是一个复杂的多输入多输出系统,为了在同一个模型中实现高效率、低污染物排放的优化目标,对标准最小二乘支持向量机回归方法进行了扩展.借助某电厂1000MW超超临界锅炉的现场燃烧调整试验数据,建立了以锅炉热效率和NOx排放质量浓度为输出的共享最小二乘支持向量机(LSSVM)模型,采用一种改进的粒子群算法对共享模型中的锅炉运行工况进行了寻优.结果表明:在共享LSSVM模型中,锅炉热效率和NOx排放质量浓度的平均预测误差分别可达到0.028%和2.16%,搜索得到的高效率和低NOx排放的参数组合可为电站锅炉优化运行提供指导. 相似文献
7.
针对电厂循环流化床锅炉NOx排放问题进行了研究,并对人工蜂群算法进行了改进,结合最小二乘支持向量机建立了锅炉燃烧NOx排放模型,对锅炉可调参量进行了优化,降低了NOx排放浓度。将改进的人工蜂群算法与基本的人工蜂群算法和粒子群算法进行比较,说明基于改进人工蜂群算法所建立的模型能够很好的预测NOx的排放浓度,具有很强的辨识能力和泛化能力,同时也表明了改进人工蜂群算法计算速度快的优点及优化数据上的优势,通过仿真试验,优化后NOx排放浓度明显降低,体现了其工程实用价值。 相似文献
8.
在热态试验数据的基础上,分别应用BP(神经网络)和SVM(支持向量机)回归算法建立了燃煤机锅炉NOx排放特性模型,并验证了模型的准确性。结果表明,BP网络模型对检验样本的最大预测误差、最小预测误差和均方差分别为4.263%、0.556%和2.2133%,支持向量机模型对检验样本的最大预测误差、最小预测误差和均方差分别为2.121%、0.091%和0.4549%。两种智能技术都能对锅炉在不同工况下的NOx排放做出较为准确的预报,但支持向量机在泛化能力、收敛速度、最优性等方面明显优于神经网络。 相似文献
9.
10.
由于锅炉设备庞大,运行条件复杂,煤种多变等因素,很难建立锅炉NOx排放与效率的函数模型。利用最小二乘支持向量机(LS-SVM)建立了以锅炉NOx排放与热效率为输出的混合模型,并对此模型进行了校验。结果表明,该模型具有调节参数少、运算速度快、结果稳定、预测精度高等优点,可以根据燃煤特性以及各操作参数准确预报锅炉在不同工况下的NOx排放和效率。针对模型的多目标优化问题,采用多目标粒子群优化算法MOPSO(multiple objective particle swarm optimization)对某工况进行优化仿真,在提高效率的同时降低了NOx排放。 相似文献
11.
12.
13.
14.
15.
为了既提高锅炉热效率,又确保NOx排放浓度低于650 mg/m3,采用正交试验的方法进行了锅炉的燃烧优化调整.结果表明:上层磨煤机分离器折向门开度、省煤器出口氧量、炉膛与大风箱差压以及OFA风门开度是影响锅炉热效率和NOx排放的4个关键因素.为此确定了一套最优组合方案:上层磨煤机分离器折向门开度为"6",省煤器出口氧量为3.7%,炉膛与大风箱差压为800 Pa,OFA风门开度为25%.按照这一最优组合方案进行了验证试验,结果锅炉热效率达到87.12%,NOx排放浓度为640 mg/m3. 相似文献
16.
17.
《热科学学报(英文版)》2021,(4)
The utilization of powdery semi-coke as a power fuel in pulverized coal-fired power plants has become a new and potential technique to consume the excess powdery semi-coke.The characteristic of low volatile results in poor combustion performance and high NO_x emission,and to co-fire with bituminous coal is a practical strategy to address this problem.However,the co-combustion characteristics and the inherent interaction between semi-coke and coal remain insufficiently understood.In addition,the influences of secondary air arrangement,the boiler operation load,and the fuel type on co-combustion process are still unclear,which is urgent to be further explored.In the present study,experiments and numerical simulations were jointly utilized to inquire into the co-combustion behaviors and NO_x emission features of semi-coke and coal.The results demonstrated that the"out-furnace method"was a suitable choice for small-capacity boiler when the proportion of semi-coke was 33%,due to the limited combinations of the semi-coke injection position.It was recommended that semi-coke was preferred to be injected from the middle layers of the furnace under the"in-furnace method"to improve the overall co-combustion performance.The critical value of the separated over fire air ratio in this study was 27.5%,over which a slight drop of carbon content in fly ash could come about.Moreover,the elevation in the proportion of separated over fire air gave rise to the significant decline of NO_x concentration.The constricted secondary air arrangement was preferred to be employed due to the high boiler efficiency.The separated over fire air and the surrounding air needed to maintain a wide-open degree to prevent the increase of NO_x emissions and the coking of nozzles.For the load reduction regulation method adopted in this study,the NO_x concentration first rose and then dropped,while the burnout ratio decreased obviously as the operation load was reduced.Different combinations of coal and semi-coke generated significant influences on co-combustion behaviors within the furnace.The NO_(x )generated by high-volatile fuel (bituminous coal) combustion was mainly affected by volatile-N,while the NO_(x )generated by low-volatile fuel (semi-coke) was mainly impacted by char-N.This study is of guiding significance for the efficient and clean utilization and beneficial to the large-scale application of powder semi-coke in power plants. 相似文献
18.
19.
20.