针对飞灰含碳量测量的研究现状和不足,采用基于粒子群优化的支持向量回归法对飞灰含碳量软测量展开建模研究,该方法利用粒子群算法的寻优功能,实现支持向量机模型的参数优化,使模型具有良好的预测能力.以大唐潮州电厂1000 MW超临界机组为研究对象,将现场采集数据分为训练数据和测试数据,分别用来辨识飞灰含碳量软测量模型和检验模型的泛化能力.仿真结果表明,飞灰含碳量软测量模型仿真输出与实际输出基本吻合,验证了模型的有效性和泛化能力.%Contraposing researches and deficiencies of measurement of carbon content in fly ash,this paper uses support vector regression (SVR) based on particle swarm optimization (PSO) to study the modeling of carbon content in fly ash. This method uses optimization function of particle swarm algorithm to achieve parameter optimization of support vector machine,which makes the model a good predictive ability. Data collected from the Datang Chaozhou Power Plant 1000MW ultra-supercritical unit will be divided into training data and test data,which were separately used to identify the parameters of carbon content in fly ash soft sensor model and test the model's generalization capability. The simulation results show that the carbon content in fly ash soft sensor model simulation and actual outputs are consistent,which verifies the validity and generalization ability of the model.
展开▼
