摘要
基于OPC技术实现过程数据的实时采集,并对所需变量进行数据滤波与异常检测,再利用RBF神经网络建立乙烯裂解炉过程多输入多输出(MIMO)裂解产物收率在线软测量模型以及模型校正方法,以乙烯和丙烯收率之和最大为目标,基于遗传算法对RBF神经网络模型进行操作优化,得到裂解过程的最优操作条件以指导生产。实际的工业应用表明,该方法提高了乙烯和丙烯的收率,具有良好的适应性和稳定性,对实际生产有重要的指导意义。
The real-time data collection based on Object Linking and Embedding (OLE) for Process Control (OPC) technology is achieved, and data filtering and anomaly detection methods for necessary variables are given. An online soft measurement model based on Radical Basis Functions (RBF) neural network for ethylene cracking furnace's Multi-In-Multi-Out (MIMO) process is built. At the same time, a method to adjust the online model is also studied. The model is optimized by Genetic Algorithm (GA) based on maximizing the sum of the yields of ethylene and propylene to find the optimal operation conditions to guide production. The actual industrial applications show that the yields of ethylene and propylene are increased by using this method, and the method has good adaptability and stability and has important operational guiding significance to the actual production process.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2009年第8期1003-1007,共5页
Computers and Applied Chemistry
基金
国家高技术研究发展计划(863)(2007AA04Z170)
作者简介
作者简介:尚田丰(1985-),男,河南商丘人,硕士研究生,从事计算机应用技术研究.
通讯作者:耿志强(1973-),男,河南开封人,副教授,博士,从事过程智能控制研究与开发.电话:010-64426960:E-mail:gengzhiqiang@mail.buct.edu.cn