Prediction And Regulation Of Microstructural Quality Indices For Metallocene-catalyzed Olefin Gas-phase Copolymerization Process

Polyolefin is an indispensable material in various fields of the national economy,and product quality has always been a bottleneck restricting the development of its industry.At present,the low-end polyolefin products in China are in surplus.The basic research of high-end products is weak,the self-sufficiency rates of these products are insufficient and they are heavily dependent on imports.The polyolefin industry is facing structural shortages.The reason is that the polyolefin production process adopts macroscopic quality indices(such as melt flow index and density),resulting in poor product quality control.In fact,the variation in product performance is mostly determined by the microscopic molecular chain structures(such as polymer molecular weight and its distribution,monomer sequence distribution,etc.).Therefore,the prediction and regulation of polymer microstructures have become the key to resolving issues with product quality.Molecular weight distribution(MWD)and short-chain branching distribution(SCBD)are important microstructures of polyolefins.The development of the mechanism model describing these two types of microstructures for polymerization processes is of great significance for the precise regulation of the olefin copolymer properties and the production of high-end products.However,process models involving microstructure prediction are usually large in scale and difficult to solve.The influence of complex polymerization mechanisms on microstructures remains unclear.Reliable estimation of kinetic parameters is extremely challenging.How to innovate the modeling method,establish the relationship between process operating conditions,kinetic parameters and microstructures,and then implement the prediction and regulation of microstructures.The answers to these questions have strong theoretical significance and industrial application benefits for the quality control of polyolefin products.Focusing on a series of key issues concerning the prediction and regulation of microstructural quality indices in the olefin copolymerization process,this thesis studies the determination method of catalyst active site types,the kinetic parameter estimation method,and the modeling and optimization approaches based on microstructures.The simulation and optimization of the pilot process of metallocene-catalyzed olefin gas-phase polymerization based on MWD and SCBD are considered.The following innovative research results are obtained:(1)Based on the accurately characterized microstructures,a method for determining the active site types of catalysts is proposed.The MWD and SCBD of olefin copolymers are obtained by gel permeation chromatography with an infrared detector(GPC-IR).The simultaneous deconvolution of MWD and SCBD is proposed to obtain the minimum number of active site types and the characteristic parameters of each active site type.Compared with the traditional sequential deconvolution method,the simultaneous deconvolution of the simulated distribution curves greatly reduces the relative deviation of the characteristic parameters and can restore the characteristic parameters more accurately.In the case of the deconvolution of characterization results of samples obtained in different catalytic systems,simultaneous deconvolution method allows obtaining more physically meaningful and regularly changing reliable characteristic parameters.(2)There are too many kinetic parameters in the polymerization process,and it is difficult to estimate all parameters accurately.Thus,a method for estimating kinetic parameters is proposed.A parameter estimable subset selection technique based on the hierarchical clustering algorithm is innovatively proposed.Compared with the traditional orthogonalization method,hierarchical clustering more comprehensively takes into account the correlation between paired parameters,and additionally provides a dendrogram reflecting the correlation of parameters.The proposed method is applied to the Dow chemical batch reactor mode.Based on the estimated parameter values,the calculated results by model are in good agreement with the experimental data.A systematic parameter estimation framework based on MWD and SCBD is further developed by combing the determination methods of the output variables and kinetic parameter nominal values.It is applied to the metallocene-catalyzed pilot-scale gas-phase polymerization process model.Compared with the random nominal values,the robustness index of the parameter estimability ranking results is increased by 104%;Based on good nominal values and the estimated parameter values,the deviation value of the model prediction is reduced by 87%.(3)A mechanism model capable of predicting MWD and SCBD of linear low-density polyethylene is developed for a metallocene-catalyzed pilot-scale gas-phase polymerization process.According to the kinetic phenomenon-the"comonomer effect" during the reaction,the "β-H effect" is considered.Then,the kinetic model,the reactor model and the prediction model of MWD and SCBD are developed.The proposed kinetic parameter estimation framework is applied to the copolymerization model,and the selected parameters are estimated.The model based on the estimated parameters can accurately predict MWD and SCBD.Compared with the experimental data,the deviation of MWD is less than 0.05,and the relative deviation of SCBD is less than 6%,and it depicts well the "comonomer effect".(4)The prediction model of dynamic MWD and SCBD of olefin copolymer is developed.According to the characteristics of MWD and SCBD,a reduction strategy combining the pseudo-steady-state assumption and the segmental method is determined.Then,the orthogonal collocation method is adopted to discretize the time dimension of the dynamic model.And the response of the first-order linear system is used to provide initial values for the state variables,which makes the model easier to converge.The reduced order strategy and solution method are applied to the dynamic model of the pilot-scale gas-phase polymerization process.The dynamic MWD and SCBD calculated by the reduced order model are in good agreement with those calculated by the cumulative Flory distribution method.The calculation time is reduced to 26.0 s.The calculation efficiency is significantly improved.(5)A dynamic optimization strategy based on MWD and SCBD is developed to realize the dynamic optimization of grade transition for the gas phase polymerization process.The stability relationship between the moment method and the dynamic SCBD model is given.It is proved that the key state variables and the dynamic SCBD reach the steady state at the same time.Based on this conclusion,the moment method is combined with the steady-state MWD and SCBD calculations to develop a dynamic optimization solution strategy based on MWD and SCBD.It can avoid the complicated calculation of the dynamic model.Based on the characteristics of the orthogonal collocation method,the optimization proposition form is reconstructed.Specifically,the integral term of the residual error of the quality index with respect to time is set as the objective function,so that it can simultaneously minimize the transition time and the waste production.The dynamic optimization strategy is applied to the metallocene-catalyzed pilot-scale gas-phase polymerization process.Under the premise of ensuring the microstructural quality indices meet the requirements of the target grade product,the grade transition time is shortened by 15.8%,and the waste production is reduced by 14.9%.