关键词: 过程系统, 优化, 算法, 加氢裂化反应, 智能可视化优化方法

Abstract: The historical production data in hydrocracking contains rich operational knowledge,which is not fully mined and exploited. To solve the problem,a novel method is proposed for solving the operation optimization in hydrocracking by intelligent visualization optimization method based on rough set. The continuous operation variables in the production data from practical industrial case are discretized by adaptive discretization. Because reduction of condition attributes is an N-P hard problem,thus heuristic method based on line-up competition algorithm is proposed to carry out the reduction. Results show that the proposed heuristic method can reserve the key operation variable and reduce the redundant ones efficiently,and 4 of 12 variables are reduced in hydrocracking reaction data. On the basis of the reduction,the processed data are mapped onto the plane by intelligent visualization optimization method,thus contours are created. The optimizing region and optimal operating point of the yield of aviation kerosene can be seen intuitively. Optimization result shows that the yield of aviation kerosene is promoted from previous 33.98% to 37.58% within the constraints.

Key words: systems engineering, optimization, algorithm, hydrocracking, intelligent visualization optimization method