Soft Sensor Development Based on Quality-Relevant Slow Feature Analysis and Bayesian Regression with Application to Propylene Polymerization

Data-driven soft sensors are widely used to predict quality indices in propylene polymerization processes to improve the availability of measurements and efficiency. To deal with the nonlinearity and dynamics in propylene polymerization processes, a novel soft sensor based on quality-relevant slow feature analysis and Bayesian regression is proposed in this paper. The proposed method can handle the dynamics of the process better by extracting quality-relevant slow features, which present both the slowly varying characteristic and the correlations with quality indices. Meanwhile, a Bayesian inference model is developed to predict the quality indices, which takes advantages of a probability framework with iterative maximum likelihood techniques for parameter estimation and a sparse constraint for avoiding overfitting. Finally, a case study is conducted with data sampled from a practical industrial propylene polymerization process to demonstrate the effectiveness and superiority of the proposed method.

Comparative study of Metallocene catalyst propylene polymerization with different iteration rates

This paper proposed a mathematical model corresponding to metallocene catalyzed propylene polymerization that uses the Me2Si [Ind]2ZrCL2 and Et [Ind]2ZrCL2. Comprehensive kinetic models consisting of mass and population balance equations, are developed based on elementary reactions proposed in the reaction mechanism. The result from the above indicates that metallocene catalysts in the presence of ethylene zirconium dichloride and methylene zirconium dichloride shows modularity and new peaks are obtained. The temperature variation from 25 to 75 also increase the rate and reason for the same could be chain shuttling polymerization. The model is presented through simulative study. Initially genetic approach is used but convergence rate is poor. To achieve best possible result, particle swarm optimization is used. The optimization approach with particle swarm optimization is implemented. The local and global solutions are comparable entities and replace each other in case value of local variable is not optimized. From the simulative study it is discovered that Et [Ind]2ZrCL2 produce best possible polymers both at 25 and 750C.

Propylene Polymerization and Deactivation Processes with Isoselective {Cp/Flu} Zirconocene Catalysts

Industrially relevant single-site precatalysts used to produce isotactic polypropylene (iPP) include C2-symmetric {SBI} and C1-symmetric {Cp/Flu} complexes of group 4 metals. While the latter can produce iPPs with a higher degree of isotacticity, they also suffer from poor productivity compared to their {SBI} counterparts. Several causes for this trend have been suggested—2,1-Regioinsertions are frequently pointed out, as they are suspected to drive the catalyst into a dormant state. While this event does not seem to significantly impact the productivity of {SBI} systems, the influence of these regioerror is poorly documented for isoselective {Cp/Flu} precatalysts. To address this issue, new Ph2X(Cp)(Flu) (Ph2X = Ph2C, FluC, Ph2Si) proligands (2a–k) and some of the corresponding dichlorozirconocenes (3a–h,k) were synthesized. These new compounds were characterized and tested in homogeneous propylene polymerization at 60 °C and the amounts of regioerrors in the resulting polymers were examined by 13C NMR spectroscopy. A possible correlation between poor productivity and a high number of regioerrors was investigated and is discussed. Furthermore, a C-H activation process in the bulky nBu3C substituent upon activation of 4c (the dimethylated analog of 3c) by B(C6F5)3 has been evidenced by NMR; DFT calculations support this C-H activation as a deactivation mechanism.