Method of optimal controller design for multivariable predictive control utilizing range control

• US 5,572,420 A
• Filed: 04/03/1995
• Issued: 11/05/1996
• Est. Priority Date: 04/03/1995
• Status: Expired due to Term

• Current Assignee: Honeywell Incorporated (Honeywell International Inc.)
• Original Assignee: Honeywell Incorporated (Honeywell International Inc.)
• Inventors: Lu, Zhuxin J.
• Primary Examiner(s): RUGGIERO, JOSEPH F
• Application Number:
  US08/415,881
• Time in Patent Office: 582 Days
• Field of Search: 364/148, 364/149-151, 364/152-156, 364/157, 364/164, 364/165, 364/176, 318/561
• US Class Current: 700/33
• CPC Class Codes: G05B 13/048 using a predictor

Abstract

A process control system which includes at least one manipulated variable and at least one controlled variable, provides a method for robust control of a process. Predetermined constraints of the manipulated variables and the controlled variables, and the present values of the manipulated variables are obtained. The controller is loaded with parameters which define an optimal controller, the parameters being calculated off-line. To determine the parameters a single min-max statement is defined for a worst case model of the process which operates in conjunction with a best case controller. The single min-max statement is converted to a corresponding canonical expression in the form of a minimization problem, the resultant solution of the minimization problem being the parameter. New values are calculated for the controlled variables for a predetermined number of points in the future, such that the values of the controlled variables are within the predetermined range thereby obtaining an optimal robustness of the resultant controller. The manipulated variables are also calculated to be within predetermined constraints, and the controlled variables to fall within a predetermined range when controllable. From a plurality of solutions, a most robust solution is selected. Then the manipulated variables are adjusted to cause the process control system to drive the values of the controlled variables to the calculated values.

Claims

1. In a process control system having a controller for providing robust control to a process, the process further having at least one manipulated variable and at least one process variable, a method for providing the robust control of a process, comprising the steps of:

a) loading the controller with parameters which define an optimal controller, the parameters calculated off-line including the steps of;

i) defining a single min-max statement for a worst case model of the process to operate in conjunction with the controller having a best case;

ii) converting said single min-max statement to a corresponding minimization expression, the corresponding minimization expression being a canonical of the single min-max statement; and

iii) solving the minimization problem yielding a resultant solution, the resultant solution being the parameters;

b) initializing the robust control to have predetermined constraints of the manipulated variables and the controlled variables;

c) obtaining present values of the manipulated variables and the process variables said process variables corresponding to measurement parameters of the process;

d) calculating new values of the process variables for a predetermined number of points in the future in order to have the values of the process variables within the predetermined range to obtain an optimal robustness of the resultant controller, the manipulated variables being within predetermined constraints, and the process variables falling within a predetermined range when controllable;

otherwise, keeping process variable constraint violations to a minimum;

e) from a plurality of solutions, selecting a most robust solution; and

f) controlling the process in accordance with the most robust solution.