摘要
焦炉是具有大时滞、强非线性、多变量的复杂系统。直行温度受多种因素的影响,采用常规的控制方法难以将温度控制在要求的精度范围内。以某钢铁公司新2号焦炉为控制对象,在充分总结焦炉操作人员控制经验的基础上,考虑焦炉燃烧过程的特点,提出了融合多元线性回归和神经网络来建模,多变量模糊控制和专家控制相结合的温度反馈控制算法, 开发了焦炉燃烧过程温度优化控制系统。系统投入实际生产运行后,控制效果良好,实现了焦炉燃烧过程温度的优化控制。
Coke oven is a complex multivariable system with the characteristics of large time-delay and strong non-linearity. The mean flue temperature is affected by ninny factors and it is difficult to control the temperature to require precision by the normal control methods. To Steel Iron Corporation coke plant New 2 coke, a modeling method combining multiple linear regression with neural network is presented, and a control method with the temperature feedback multivarivable fuzzy control and expert control is proposed based on the features of the burning process and the control experience. The coke oven combustion process optimal control system is developed. The good performance of control can be obtained in its application, and coke oven combustion process temperature optimal control can also be realized.
出处
《控制工程》
CSCD
2006年第3期205-207,211,共4页
Control Engineering of China
基金
国家杰出青年基金资助项目(60425310)教育部青年教师奖资助项目(教人[2002]5号)
关键词
焦炉
火道温度
优化控制
软测量
coke oven
flue temperature
optimal control
soft sensor
