论文部分内容阅读
生产装置的“假液位”现象严重影响了工业生产过程的正常运行。为此,运用软测量思想建立数学模型,监测直接测量结果,防止假液位的发生,保障生产安全。通过比较机理建模法与基于数据驱动建模法的优劣,提出采用PSO-SVM法建立数学模型,并以SBR泄料槽液位的监测为例进行分析。以集管进料压力P_1、泄料槽入口温度T_A丁二烯分离系统压力p_O这3个量作为辅助变量来预测泄料槽液位L。结果表明,该模型预测值与实际值符合良好,具有较强的预测性能,能够较好地对SBR泄料槽液位进行监测,有效避免SBR泄料槽假液位的产生。
Production equipment “fake liquid level ” phenomenon has seriously affected the normal operation of industrial production process. To this end, the use of soft measurement ideas to establish mathematical models to monitor the direct measurement results to prevent the occurrence of false liquid level to ensure the safety of production. By comparing the advantages and disadvantages of mechanism modeling and data-driven modeling, a mathematical model based on PSO-SVM is proposed and the monitoring of SBR blowdown tank level is taken as an example. The spill level L is predicted with the three variables of the header feed pressure P_1 and the spout inlet temperature T_A butadiene separation system pressure p_O as auxiliary variables. The results show that the predictive value of the model is in good agreement with the actual value and has strong predictive performance. It can monitor the liquid level of SBR spill tank well and avoid false liquid level of SBR spill tank effectively.