基于支持向量回归的飞灰含碳量软测量

SVR-based study on soft-sensing measurement of carbon content in fly ash

针对锅炉飞灰含碳量难以准确测量的问题,提出了一种基于支持向量回归的软测量方法。以大唐潮州电厂超超临界1 000MW机组为研究对象,建立了基于支持向量回归的飞灰含碳量软测量模型,采用交叉验证法优化了模型的惩罚参数C和核函数参数g,并利用测试数据和在3种机组负荷下的随机数据验证了模型的准确性和泛化能力。仿真结果表明,飞灰含碳量软测量模型的预测精度较高,且泛化能力较强,为锅炉飞灰含碳量测量提供了一种有效的方法。

Against the problem that carbon content in boiler fly ash is difficult to measure accurately,a soft sensor method based on support vector regression was proposed.Taking the ultra supercritical unit in Da-tang Chaozhou Power Plant as the research object,a support vector regression based model predicting the carbon content in fly ash was established,and cross validation method was applied to optimize the penalty parameter C and kernel function parameter g for the model.Moreover,the model＇s accuracy and generaliza-tion capability were verified by using the test data and random data obtained under three unit load condi-tions.The simulation results show that the soft sensor model of the carbon content in fly ash has higher prediction accuracy and better generalization ability,which provides an effective way for measurement of the carbon content in boiler fly ash.