Research And Application Of Optimization Technology Of Large-scale Purified Terephthalic Acid Process

Purified terephthalic acid (PTA) which is the raw material of polyester is one of the most important chemical products. China has already become the world’s largest PTA producer and consumer. Though the domestic PTA equipment has made a breakthrough, there is still a big gap in operation between the domestic and abroad plants due to the lack of in-depth understanding of mechanism and optimization technology. Based on the experimental mechanism analysis and plant operation data, the modeling and optimization study of the p-Xylene oxidation process, the azeotropic distillation solvent dehydration process, and the crude terephthalic acid hydrogenation process of PTA plant is carried out. The main contributions of this dissertation are described as follows:(1) Because of the complexity of chemical process, traditional optimization methods may easily fall into the local minimum. Particle swarm optimization (PSO) is a simple but effective swarm intelligence optimization, which has attracted more and more researchers’ attention. An improved phase angle based PSO (θ-PSO) algorithm using crossover and mutation is put forward. Then another improved θ-PSO with memory recall and varying scale randomization strategy is studied. The eligible memory particles are recalled when the landscape changes. And the vary scale randomization is introduced through the evolution to maintain the swarm diversity. The offline error in the non-trivial multimodal dynamic functions MPB indicates that this improved θ-PSO deals well with the complex dynamic tracking and optimization.(2) Based on the analysis of main and side reaction mechanism of p-Xylene oxidation, industrial macro reaction kinetics and the combustion loss models of acetate and p-Xylene are established. Then an integrated process model of p-Xylene oxidation is achieved. The operating conditions are adjusted according to the optimization results of the improved crossover and mutation θ-PSO algorithm, which obtains a significant effect.(3) In order to overcome the control difficulties, such as nonlinear and coupling characteristics, a multivariable predictive control strategy of p-Xylene oxidation process is put forward using the advanced control software FRONT-Suite. The4-carboxybenzaldehyde (4-CBA) concentration in the product of p-Xylene oxidation process together with the oxygen concentration in the reactor and first crystallizer is much more stable than before.(4) After the selection of the proper property method of the azeotropic distillation, the mechanism and a process model of an azeotropic distillation solvent dehydration process of INVISTA technology is studied. According to the model analysis and device operational experience, the energy-saving orientated optimization is carried out. including the adjustment of the solvent dehvdration column’s hack flow and the condensing temperature. After the application, the azeotropic distillation solvent dehydration process runs more economically.(5) A generalized dynamic kinetics of the catalytic hydrogenation reaction is studied, which has an engineering kinetics with a time-dependent deactivation function. It suggests that in the actual industrial hydrogenation process, the catalyst deactivation is slow and even. Then the process model of an AMOCO patented industrial terephthalic acid hydrogenation purification process is established using the actual industry data in ASPEN PLUS platform. The model predictive performance of the product quality4-CBA is good. The operating conditions are adjusted according to the optimization results of the improved crossover and mutation θ-PSO algorithm, which achieves a remarkable effect.