Research Of Soft Sensor And Optimization On Catalytic Reforming Process

Catalytic reforming is one of the key processes in petroleum reining,converting gasoline boiling range low-octane hydrocarbons to high octane compounds which can blended into gasoline.Other valuable byproducts include aromatic feedstock and hydrogen.With the high demands for process security,stability and economy,process modeling,control and optimization technology has gradually become one of the hot areas of applied research in petrochemical industry.Due to the uncertainty and complexity of industrial catalytic reforming processes,process modeling and optimization of catalytic reforming process is a challenge task.The objective of this dissertation is to develop the soft sensor and optimization approaches with good features of easy using and cost-effectiveness ant to apply to commercial naphtha catalytic reforming processes.In this dissertation,the main contributions are as follows.The soft sensor model of research octane number(RON)and aromatic yield based on kernel principal component analysis(KPCA)and weighted least square support vector machine(WLSSVM)is proposed.The kernel principal component anallysis(KPCA)method could not only solve the linear correlation of the input and compress data but also simply the model structure.A novel hybrid chaos particle swarm optimization simulated annealing(CPSO-SA)algorithm is applied to optimize WLSSVM parameters to improve learning performance and generalization ability of the model.Furthermore,The KPCA-WLSSVM is applied to develop the soft sensor model for research octane number(RON)and aromatic yield about reforming production in the process of continuous catalytic reforming,and the satisfactory result is obtained.A cloud adaptive differential evolution(CADE)is proposed based on the characteristics of the cloud model and differential evolution algorithm.The benchmark test functions simulation results show that the proposed algorithm has fine capability of finding global optimum.Finally,the industrial simulation result proved that the performance of hybrid algorithm CADE was better than those of GA,PSO and DE algorithm in obtaining optimal solutions.The algorithm has good optimization performance and at the same time has high convergence speed in operation optimization of the catalytic reforming unit.