柴油调合在线优化内模控制

On-line Optimal Internal Model Control for Diesel Blending Process

根据某炼油厂加工不同外来原油时的柴油调合生产实测数据，建立了以柴油组分油流量和倾点为输入参数、柴油倾点为输出参数的柴油调合ＲＢＦ神经元网络模型，该模型能较好预测柴油产品倾点值；提出了闭环优化控制策略。

The are neuroid model of diesel blending is established based on the experimental data of diesel blending production line in certain oil refinery for different raw oil supplied externally. The input parameters of the model are the flows and pour points of the compositions of the diesel, while the output parameter is the pour point of the diesel. The model is capable to estimate the pour point of the diesel product. The closed-loop optimal control strategy is stated.