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。
NOx release and unburned carbon in fly ash are two aspects of optimize combustion inutility tangentially firing coal burned boiler. There are also important problems in power plant. There are affected by many factors and complicated. Building the model of unburned carbon in fly ash and NOx release is the premise to optimize the coal combustion. In paper, firstly, using cross validation algorithm to analysis on sample datas and select the model's parameter o" and C; Secondly , Building the model of unburned carbon in fly ash and NOx release with support vector machine theory; Finally, the model is checked by experimental datas. The modeling results showed that support vector machine is a good tool for building combustion models and has relatively small errors and has higher calculation ability.