摘要
温度是丙烯腈生产的重要条件,在研究丙烯腈生产工艺的基础上,对丙烯腈流化床反应器进料系统PID控制器参数进行了整定,进一步采用阶梯式控制策略,用渐消记忆的递推最小二乘法辨识模型实现丙烯腈流化床反应器反应温度的广义预测控制,较好地解决了由反应器进料流量波动、反应压力变化等因素引起的反应温度的大幅波动问题,提高了反应器温度的控制品质,同时也提高了丙烯腈收率。
At the basis of analyzing acrylonitfile technics, reactor feed system PID controller parameters was tuned and applied a stair-like control scheme. The fading memory recursive least-square identification model was used to realize the generalized predictive control of reaction temperature of acrylonitrile fluid bed reactor. The result demonstrates that such algorithm can well solve the sharp reaction temperature fluctuations brought by reactor feed flow fluctuations, reactor pressure changes, etc. It has greatly improved reaction temperature control quality and well enhanced selectivity to acrylonitrile.
出处
《化工自动化及仪表》
CAS
2008年第3期58-61,66,共5页
Control and Instruments in Chemical Industry
基金
国家"863"计划项目(2007AA04Z195)
关键词
丙烯腈
流化床反应器
反应温度
先进控制
阶梯式广义预测控制
acrylonitrile
fluid bed reactor
reaction temperature
advanced control
stair-like generalized predictivecontrol
