摘要
针对聚丙烯双环管反应器的工艺特点,提出了一种聚丙烯熔融指数软测量机理建模方法。该方法在单环管熔融指数预报模型的基础上,根据聚合物混合原理,建立了两种适用于双环管反应器的机理模型:指数模型和幂数模型,并采用带遗忘因子的递推最小二乘法(FFRLS)和粒子群算法(PSO)辨识模型参数,增强模型的适应性。最后,利用某装置实际生产数据进行模型验证,结果表明,所建模型的预报结果与实验室化验值吻合较好,能够在很大范围内跟踪反应器内熔融指数的变化。
An improved soft-sensing method of polypropylene melt index is developed based on the unique features of polypropylene dual-loop reactor.On the basis of melt index predictive model in single-loop reactor and polymer blend principle,two mechanism models to the dual-loop reactor are given,which are called exponential model and power model.The recursive least square method with forgetting factor （FFRLS） and the particle swarm optimization （PSO） algorithm are used to identify the unknown model parameters respectively,enhancing the adaptability of the model.The validation is carried out using actual production data.The results show that the predicted values of these two models are good agreement with the values of laboratory analysis,and can track a large range of melt index changes.
出处
《控制工程》
CSCD
北大核心
2010年第2期193-196,共4页
Control Engineering of China
基金
国家"863"基金资助项目(2007AA04Z193)
山东省自然科学基金资助项目(Y2007G49)
关键词
软测量
双环管反应器
机理建模
参数辨识
熔融指数
soft sensor
dual-loop reactor
first-principle modeling
parameter identification
melt index
