Parameter estimation for LLDPE gas-phase reactor models

Abstract

-Product development and advanced control applications require models with good predictive capability. However, in some cases it is not possible to obtain good quality phenomenological models due to the lack of data or the presence of important unmeasured effects. The use of empirical models requires less investment in modeling, but implies the need for larger amounts of experimental data to generate models with good predictive capability. In this work, nonlinear phenomenological and empirical models were compared with respect to their capability to predict the melt index and polymer yield of a low-density polyethylene production process consisting of two fluidized bed reactors connected in series. To adjust the phenomenological model, the optimization algorithms based on the flexible polyhedron method of Nelder and Mead showed the best efficiency. To adjust the empirical model, the PLS model was more appropriate for polymer yield, and the melt index needed more nonlinearity like the QPLS models. In the comparison between these two types of models better results were obtained for the empirical models.

Energy Time Series Forecasting

“…A first example for a hill climbing algorithm is called the Hooke-Jeeves algorithm [130]. It highly depends on the problem at hand whether one algorithm dominates the other one [178,243]. Thus, it is following a path to an optimum.…”

Industrial polyethylene melt index prediction using ensemble manifold learning–based local model

“…According to a collection of earlier papers in 1994–1997, methods of linear estimation, extended Kalman filters, and artificial neural networks (ANN) were adopted for MI prediction. During the last two decades, popular soft sensors for MI prediction include partial least squares, orthogonal least squares, improved ANN methods, support vector regression (SVR), Gaussian process regression (GPR), relevance vector machines, extreme learning machines, fuzzy systems, and other integrated methods . Without in‐depth understanding of the phenomenology involved, these data‐driven soft sensors can generally be constructed in a relatively simple manner.…”

Neural Network Estimator for Gas-Phase Ethylene Polymerization Process

“…The ethylene polymerization process involves many challenging problems including nonlinear dynamic behavior, multivariable interactions between each state variable and unmeasurable state variables such as concentration, melt index and reaction rate. Estimation of unknown variables in polymerization processes is important for product quality control as well as in avoiding disruption in maintaining this quality [1][2][3][4][5].…”