论文部分内容阅读
大型四角切圆电站锅炉NOx排放和飞灰含碳量是燃烧优化的两个方面,也是电厂关心的重要问题。影响两者的因素众多而且复杂,对锅炉飞灰含碳量和NOx排放特性进行建模是实现燃烧优化的一个前提。文中首先利用交叉验证算法分析样本数据,对支持向量机中的参数C和σ进行选择,再应用支持向量机理论建立了飞灰含碳量和NOx排放特性模型,最后利用试验数据对模型进行了校验。研究结果表明,采用支持向量机算法建模误差较小,达到了比较准确的预测效果。
The NOx emissions from large boiler and the carbon content of fly ash are the two aspects of combustion optimization, which are also the important concerns of power plants. The factors that affect both are numerous and complex. Modeling the carbon content and NOx emission characteristics of boiler fly ash is a prerequisite for achieving combustion optimization. In this paper, the sample data is first analyzed by cross-validation algorithm, the parameters C and σ in SVM are selected, then the model of carbon content and NOx emission of fly ash is established by using support vector machine theory. Finally, check. The results show that the SVM algorithm has a smaller modeling error and achieves a more accurate prediction result.