摘要
应用支持向量机算法对燃煤锅炉结渣问题进行数学建模,并利用模拟退火算法对支持向量机模型参数进行了优化,最终获得最优参数组合。模型将煤的软化温度tSt、硅铝比w(SiO2)/w(A12O3)、碱酸比J和硅比G以及锅炉的无因次切圆直径t和无因次实际切圆直径d作为输入变量,以结渣程度作为输出,用试验数据对模型进行了校验和参数的寻优,利用优化后的模型对15台锅炉结渣特性进行预测评判,有14个正确,评判准确率为93.33%,由此表明此方法是合理有效的。同时为了配合该模型,采用高级语言编程开发出了相应的预测评判系统。
A support vector machine-based algorithm was used to establish a mathematical model for predicting the slagging characteristics of a coal-fired boiler.The simulation annealing algorithm was employed to optimize the parameters of the model in question.Finally,an optimal parameter combination was obtained.With the demineralization temperature tst,silicon/aluminum ratio w（SiO2）/w（Al2O3）,alkali/acid ratio J and silicon ratio G of the coal as well as the non-dimensional tangential circle diameter Φt and the actual non-dimensional tangential circle diameter Φd of the boiler serving as the input variables and the slagging degree as the output in the model,the test data were used to check the model and optimize its parameters.Then,the optimized model was employed to predict and judge the slagging characteristics of 15 boilers.14 boilers were correctly predicted with the judgement correctness rate being 93.33%,showing that the method is rational and effective.In the meanwhile,to coordinate with the model in question,an advanced language was used to design a program to develop a corresponding prediction and judgement system.
出处
《热能动力工程》
CAS
CSCD
北大核心
2011年第4期440-444,495,共5页
Journal of Engineering for Thermal Energy and Power
基金
国家重点基础研究发展计划基金资助项目(973)(2007CB206904)
吉林省科技发展计划基金资助项目(20070529)
关键词
支持向量机
退火算法
燃煤锅炉
结渣预测
support vector machine
annealing algorithm
coal-fired boiler
slagging
