基于优化动态PLS的常压塔重整料终馏点软测量研究

The final boiling point soft-sensing of reforming feedstock on crude column based on optimized dynamic PLS

针对某炼油厂常压塔重整料终馏点软测量模型估计精度不高的问题,利用网格搜索(grid search)法优化动态软测量建模中最大滞后时间和再采样间隔,建立了一种基于动态PLS的软测量模型。结果表明对动态软测量再采样间隔和最大滞后时间的优化,可以显著降低模型的预测误差和计算量。动态软测量的预测误差明显小于静态软测量模型,能取得较好的预测效果。

The Final Boiling Point soft-sensing of reforming feedstock on crude column is not easy to be meansured online, the traditional soft-sensing technique, as an alternative, has some problems of low prediction accuracy and poor robustness. To cope with such changes, grid search method is used to optimize maximum lag time and resampling interval of the dynamic soft measurement modeling. The simulation results show that the prediction error and computation of the soft-sensors is decreased obviously. The proposed model of optimized dynamic PLS is better than a general static soft sensor model.