基于最小二乘支持向量回归机的燃煤锅炉结渣特性预测

State Prediction of Slagging on Coal-fired Boilers Based on Least Squares-support Vector Machine for Regression

对燃煤锅炉结渣特性建模预测并结合优化算法实现燃烧优化是降低锅炉结渣几率有效的方法。文中将煤的软化温度tST、硅铝比w(SiO2)/w(Al2O3)、碱酸比J、硅比G以及锅炉的无因次炉膛平均温度φt、无因次切圆直径φd等作为输入变量,以结渣程度作为输出,建立最小二乘支持向量回归机燃煤锅炉结渣预测模型。同时采用显微镜原理对惩罚参数γ和核参数σ进行寻优,快速有效地获得二者的最优组合。通过对5台锅炉结渣特性进行预测评判,结果表明此方法是合理可行的。同时依据本方法及面向对象的高级语言,开发了相应的预测评判系统。

Building a model to predict the state of slag on coal-fired boilers is a good way to optimize the coal combustion and reduce the risk of boiler slag. This paper built the least squares-support vector machine for regression （LS- SVMR） to predict the state of slag on coal-fired boilers, in which there were six input vectors, which were softening temperature （tST）, SiO2-Al2O3 ratio（w（SiO2）lw（Al2O3））, alkaliacid ratio（J）, percentage of silicon content（G）, the dimensionless average temperature furnace（φ） and the dimensionless inscribed circle diameter furnace（φ）, and one output vectors, which was slagging degree. At the same time, to obtain the optimal combination of penalty parameter 7and nuclear parameter a, the principle of microscope was used effectively. The feasibility of this method was proved by the result of predicting the state of slag on the five coal-fired boilers. Besides, the prediction system has been developed by object- oriented high-level language accordingly.