Process for preparing polyethylene

• US 20050020784A1
• Filed: 07/22/2003
• Published: 01/27/2005
• Est. Priority Date: 07/22/2003
• Status: Active Grant

• Current Assignee: Fina Technology, Inc. (Totalenergies Se)
• Original Assignee: Fina Technology, Inc. (Totalenergies Se)
• Inventors: Noll, Patrick
• Granted Patent: US 7,402,635 B2
• Time in Patent Office: Days
• US Class Current: 526/64
• CPC Class Codes: B01J 19/0006 Controlling or regulating processes controlling or regulating in general G05
• CPC Class Codes: B01J 19/2435 Loop-type reactors
• CPC Class Codes: B01J 2219/00191 Control algorithm
• CPC Class Codes: B01J 2219/00243 Mathematical modelling
• CPC Class Codes: C08F 10/02 Ethene
• CPC Class Codes: C08F 2/00 Processes of polymerisation
• CPC Class Codes: C08F 2/14 Organic medium
• CPC Class Codes: C08F 210/14 Monomers containing five or more carbon atoms
• CPC Class Codes: C08F 210/16 Copolymers of ethene with alpha-alkenes, e.g. EP rubbers
• CPC Class Codes: C08F 2400/02 Control or adjustment of polymerization parameters
• CPC Class Codes: G05B 13/048 using a predictor
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: Y10S 706/906 Process plant

Abstract

Disclosed is process for producing polyethylene using a slurry loop reactor. The process includes using a mathematical model to predict a plurality of process control parameters based on the desired product properties and reactor characteristics and controlling the process using the predicted process control parameters. Also disclosed is a process controller programmed with the model and a method for optimizing the configuration of a loop reactor using the model.

Claims

1. A process for producing polyethylene using a slurry loop reactor comprising using a mathematical model to predict a plurality of process control parameters based on the desired product properties and reactor characteristics and controlling the process using the predicted process control parameters.

2. The process of claim 1 wherein the variables used to prepare the mathematical model include a plurality of variables selected from the group consisting of:

ethylene flow into reactor, ethylene flow out of reactor, isobutane flow into reactor, isobutane flow out of reactor, hexene flow into reactor, hexene flow out of reactor, hexene conversion in reactor, hydrogen flow into reactor, hydrogen flow out of reactor, hydrogen conversion in reactor, polymer flow out of reactor, liquid flow out of reactor, total mass flow into reactor, total mass flow out of reactor, total volume flow out of reactor, catalyst flow into reactor, catalyst flow out of reactor, concentration of ethylene in the reactor liquid, concentration of hexene in the reactor liquid, concentration of hydrogen in the reactor liquid, temperature of reactor, pressure of reactor, weight concentration of solids in the reactor slurry, volume concentration of solids in the reactor slurry, weight concentration of solids in the settling leg solids bed, number of settling legs, reactor volume, settling leg diameter, settling leg height, bulk density of reactor polymer, density of reactor polymer, density of reactor, liquid, density of reactor slurry, residence time of reactor solids, catalyst activity, catalyst productivity, catalyst diameter, catalyst feed factor, catalyst activity factor, terminal velocity of settling polymer, polymer settling rate, rate of polymer leaving reactor that is not part of the settling leg solids bed, rate of slurry leaving reactor that is part of the settling leg solids bed, rate of slurry leaving reactor that is not part of the settling leg solids bed, viscosity of reactor liquid, density of catalyst, polymer diameter, Archimedes number for polymer settling in settling leg, Reynolds number for polymer settling in settling leg, acceleration due to gravity, cross sectional area of a settling leg occupied by polymer, and cross sectional area of a settling leg.

3. The process of claim 2 wherein the variables used to prepare the mathematical model are:

polymer flow out of reactor, liquid flow out of reactor, concentration of ethylene in the reactor liquid, temperature of reactor, pressure of reactor, weight concentration of solids in the reactor slurry, reactor volume, settling leg diameter, settling leg height, bulk density of reactor polymer, density of reactor polymer, density of catalyst.

4. The process of claim 2 wherein the variables used to prepare the mathematical model are:

polymer flow out of reactor, liquid flow out of reactor, concentration of ethylene in the reactor liquid, temperature of reactor, pressure of reactor, weight concentration of solids in the reactor slurry, reactor volume, settling leg diameter, settling leg height, bulk density of reactor polymer, density of reactor polymer, density of catalyst, concentration of hexene in the reactor liquid, concentration of hydrogen in the reactor liquid, hexene conversion in reactor, hydrogen conversion in reactor.

5. The process of claim 1 wherein the process is controlled using the mathematical model which has been incorporated into a computer spreadsheet.

6. The process of claim 1 wherein the process is controlled using the mathematical model which has been incorporated into a controller.

7. The process of claim 6 wherein the controller is a neural network model based controller.

8. A controller for an industrial slurry loop reactor used to produce high-density polyethylene comprising a controller programmed with an algorithm to control slurry loop reactor using the process of claim 1.

10. A process for optimizing the configuration of a polyethylene loop reactor comprising using the model of claim 1 to design the reactor.

11. The process of claim 10 wherein the reactor is designed such that settler size, cycle time, reactor volume and energy consumption are optimized

9. The controller of claim 9 wherein the controller is a neural network model based controller.