Abstract: A model of grade transition is presented for a commercialized fluidized bed gas-phase
polyethyleneproduction process. The quantity of off-specification product and the time of
grade transition can be minimizedby the optimization of operating variables, such as
polymerization temperature, the ratio of hydrogen to ethylene,the ratio of co-monomer to
ethylene, feed rate of catalyst, and bed level. A new performance index, the ratio ofmelt
flow (MFR), is included in the objective function, for restraining the sharp adjustment of
operation variablesand narrowing the distribution of molecular weight of the resin. It is
recommended that catalyst feed rate andbed level are decreased in order to reduce the grade
transition time and the quantity of off-specification product.This optimization problem is
solved by an algorithm of sequential quadratic programming (SQP) in MATLAB.There is
considerable difference between the forward transition and reverse transition of grade with
regard to theoperating variables due to the non-linearity of the system. The grade
transition model is extended to a high spacetime yield (STY) process with the so-called
condensed model operation. In the end, an optimization strategy formulti-product transition
is proposed with two-level optimization of the objective function J(x, u) on the basis
ofthe optimal grade transition model. A sequential transition of six commercial
polyethylene grades is illustrated foran optimal multi-product operation.

Key words: polymer, continuous polymerization, grade transition, multi-product operation, two-level optimization

摘要： A model of grade transition is presented for a commercialized fluidized bed gas-phase
polyethyleneproduction process. The quantity of off-specification product and the time of
grade transition can be minimizedby the optimization of operating variables, such as
polymerization temperature, the ratio of hydrogen to ethylene,the ratio of co-monomer to
ethylene, feed rate of catalyst, and bed level. A new performance index, the ratio ofmelt
flow (MFR), is included in the objective function, for restraining the sharp adjustment of
operation variablesand narrowing the distribution of molecular weight of the resin. It is
recommended that catalyst feed rate andbed level are decreased in order to reduce the grade
transition time and the quantity of off-specification product.This optimization problem is
solved by an algorithm of sequential quadratic programming (SQP) in MATLAB.There is
considerable difference between the forward transition and reverse transition of grade with
regard to theoperating variables due to the non-linearity of the system. The grade
transition model is extended to a high spacetime yield (STY) process with the so-called
condensed model operation. In the end, an optimization strategy formulti-product transition
is proposed with two-level optimization of the objective function J(x, u) on the basis
ofthe optimal grade transition model. A sequential transition of six commercial
polyethylene grades is illustrated foran optimal multi-product operation.

关键词: polymer;continuous polymerization;grade transition;multi-product operation;two-level optimization