基于Smith预估补偿的聚丙烯反应器自适应控制系统研究

Research of Polypropylene Reactor Adaptive Control System with Smith Predictive Compensation

摘要:

针对含有时滞的聚丙烯多区循环反应器温度串级控制系统为复杂的高阶系统，被控对象的滞后时间较大，且受到反应器内部放热、入口循环气温度、夹套水流量等干扰的影响，系统过程控制较为困难。通过以一阶惯性环节加纯滞后来逼近高阶系统，应用模型参考自适应控制算法，使自适应机构能够在线的整定反应器串级控制副回路的Smith预估器，使副回路Smith预估器的动态特性与副被控对象的动态特性接近一致；同时，根据期望的性能指标，在串级控制主回路设置一个参考模型，再次应用自适应算法，使得主回路被控对象的输出尽可能跟踪参考模型的输出。仿真结果表明，该控制系统对于具有时滞的聚丙烯反应器温度控制是有效的。

Abstract:

For the polypropylene multi-zone circulating reactor cascade control system with time delay as a complex higher-order system, the controlled object has a large lag time, and the process control of the system is difficult due to the influence of internal heat release, reactor inlet temperature, water flow in the jacket and other disturbances. The first-order plus pure lag approach to the higher-order system is adopted, and the model reference adaptive control method is applied to make the online setting of the sub-loop Smith-predictor of the adaptive mechanism, which makes the dynamic of the Smith-predictor close to the dynamic of the sub-control object. At the same time, a reference model is set in the main loop according to the expected performance index, and adaptive algorithm is applied again to make the output of the controlled object in the winner loop track the output of the reference model as much as possible. The simulation results are employed to illustrate the effectiveness of this approach.