为了确定催化重整的优化操作区域,以方便催化重整过程中操作参数的调节,基于数据驱动的智能可视化优化方法,提出了优化催化重整过程的操作方法。基于该过程的操作数据,以脱戊烷油量和辛烷值为目标,应用智能可视化优化方法建立了降维映射模型,将操作数据降维到二维平面,并生成目标等值线。由此,在映射平面上找到了协调脱戊烷油量和辛烷值两目标的优化操作区域。该优化区域为催化重整过程中操作参数的调节带来了便利,当某个或几个操作参数发生变化时,可通过调节其他参数,使操作点仍然落在优化区域内。应用所提方法对一个季度的操作数据进行统计,结果表明,有96%的操作点在优化区域内,且脱戊烷油量和辛烷值得到了同步提高,验证了所确定优化区域的有效性。
In order to determine the optimal operating area for catalytic reforming to facilitate the adjustment of operating parameters in catalytic reforming process, a data-driven intelligent visualization optimization method was proposed to optimize the operation of the catalytic reforming process. Based on the operation data of the plant, the method of intelligent visualization optimization was used to establish a dimension-reduction mapping model in order to increase the yield of depentanized oil and octane value. The operating data from a multidimensional space are reduced to a two-dimensional plane, and then the target contour is generated. In the mapping plane, the optimal operating area for the yield of depentanized oil and octane value can be found easily. The discovery of the optimized region can facilitate the adjustment of the operating parameters in the catalytic reforming process. When one or several operating parameters are changed, the operating point can still fall within the optimized region by adjusting other parameters. This knowledge of area optimization has been applied in the refinery.Through the operating data statistics in a quarter, 96% of the operating point is in the optimization area, and the yield of the depentanized oil and octane value are increased synchronously. The results show that the optimized region is effective.