摘要
锅炉燃烧系统的建模是实现锅炉节能减排的基础,为实现燃煤锅炉节能减排,针对现有建模方法多步预测精度不足问题,采用经验模态分解算法把复杂的输出信号转化为多个具有周期性规律或趋势相对平稳的模态信号,降低数据复杂度,基于K近邻联合互信息法得出迟延时间,提出基于动态时间规整距离进行在线更新的最小二乘支持向量机算法,并进行锅炉燃烧系统的建模,基于锅炉实际运行数据的仿真结果表明,该方法可以有效提高模型的自适应能力和多步预测精度,为后续实现闭环燃烧优化控制打下了良好基础。
Modelling of the boiler combustion system is the basis for energy saving and emission reduction of boilers,in order to realize the energy saving and emission reduction of coal-fired boilers and solve the problem of insufficient multi-step prediction accuracy of existing modeling methods,the empirical mode decomposition algorithm wasfirst used to transform the complex output signals into multiple modal signals with periodic rules or relatively stable trends,so as to reduce the data complexity.The delayed time was obtained based on the K-nearest neighbor joint mutual information method,a least-squares support vector machine algorithm based on the dynamic time regular distance for online update was proposed,the boiler combustion system was modeled,and the simulation results based on the actual operation data of the boiler showed that the proposed method could effectively improve the adaptability and multi-step prediction accuracy of the model,and lay a good foundation for the subsequent realization of closed-loop combustion optimization control.
作者
辛超
孙成田
张效源
孙凯
孙凯进
XIN Chao;SUN Chengtian;ZHANG Xiaoyuan;SUN Kai;SUN Kaijin(CHN Energy Feixian Power Generation Co.,Ltd.,Feixian 276001,Shandong China;Southeast University,Nanjing 210096,China)
出处
《粘接》
CAS
2024年第1期117-120,共4页
Adhesion
基金
国家自然科学基金资助项目(项目编号:51476027)。
关键词
锅炉燃烧系统
经验模态分解算法
K近邻联合互信息
最小二乘支持向量机
boiler combustion system
empirical modaldecomposition algorithm
k-nearest neighbor joint mutual information
least squares support vector machine
