Controllers that determine optimal tuning parameters for use...

Details Controllers that determine optimal tuning parameters for use... Controllers that determine optimal tuning parameters for use...

1998-08-20

2001-06-26

Beausoleil, Robert (Department: 2181)

Data processing: generic control systems or specific application

Generic control system, apparatus or process

Optimization or adaptive control

C700S029000, C700S044000

Reexamination Certificate

active

06253113

ABSTRACT:

TECHNICAL FIELD OF THE INVENTION
The present invention is directed, in general, to control systems for process facilities and, more specifically, to controllers that determine optimal proportional integral and derivative (“PID”) tuning parameters for use in process control systems to globally optimize process facilities.
BACKGROUND OF THE INVENTION
Presently, process facilities (e.g., a manufacturing plant, a mineral or crude oil refinery, etc.) are managed using distributed control systems. Contemporary control systems include numerous modules tailored to control or monitor various associated processes of the facility. Conventional means link these modules together to produce the distributed nature of the control system. This affords increased performance and a capability to expand or reduce the control system to satisfy changing facility needs.
Process facility management providers, such as
HONEYWELL
,
INC
., develop control systems that can be tailored to satisfy wide ranges of process requirements (e.g., global, local or otherwise) and facility types (e.g., manufacturing, refining, etc.). A primary objective of such providers is to centralize control of as many processes as possible to improve an overall efficiency of the facility. Each process, or group of associated processes, has certain input (e.g., flow, feed, power, etc.) and output (e.g., temperature, pressure, etc.) characteristics associated with it.
In recent years, model predictive control (“MPC”) techniques have been used to optimize certain processes as a function of such characteristics. One technique uses algorithmic representations to estimate characteristic values (represented as parameters, variables, etc.) associated with them that can be used to better control such processes. In recent years, physical, economic and other factors have been incorporated into control systems for these associated processes. Examples of such techniques are described in U.S. Pat. No. 5,351,184 entitled “M
ETHOD OF
M
ULTIVARIABLE
P
REDICTIVE
C
ONTROL
U
TILIZING
R
ANGE
C
ONTROL
;” U.S. Pat. No. 5,561,599 entitled “M
ETHOD OF
I
NCORPORATING
I
NDEPENDENT
F
EEDFORWARD
C
ONTROL IN A
M
ULTIVARIABLE
P
REDICTIVE
C
ONTROLLER
;” U.S. Pat. No. 5,574,638 entitled “M
ETHOD OF
O
PTIMAL
S
CALING OF
V
ARIABLES IN A
M
ULTIVARIABLE
P
REDICTIVE
C
ONTROLLER
U
TILIZING
R
ANGE
C
ONTROL;” and U.S. Pat. No.
5,572,420 entitled “M
ETHOD OF
O
PTIMAL
C
ONTROLLER
D
ESIGN OF
M
ULTIVARIABLE
P
REDICTIVE
C
ONTROL
U
TILIZING
R
ANGE
C
ONTROL
” (the “'420 Patent”), all of which are commonly owned by the assignee of the present invention and incorporated herein by reference for all purposes.
Generally speaking, one problem is that conventional efforts, when applied to specific processes, tend to be non-cooperative (e.g., non-global, non-facility wide, etc.) and may, and all too often do, detrimentally impact the efficiency of the process facility as a whole. For instance, many MPC techniques control process variables to predetermined set points. Oftentimes the set points are a best estimate of a value of the set point or set points. When a process is being controlled to a set point, the controller may not be able to achieve the best control performances, especially under process/model mismatch.
To further enhance the overall performance of a control system, it is desirable to design a controller that deals explicitly with plant or model uncertainty. The '420 Patent, for example, teaches methods of designing a controller utilizing range control. The controller is designed to control a “worst case” process. An optimal controller for the process is achieved and, if the actual process is not a “worst case process,” the performance of the controller is better than anticipated.
There are a number of well known PID “tuning” formulas, or techniques, and the most common, or basic, PID algorithm includes three known user specified tuning parameters (K, &tgr;
1
, &tgr;
2
) whose values determine how the controller will behave. These parameters are determined either by trial and error or through approaches that require knowledge of the process. Although many of these approaches, which are commonly algorithms, have provided improved control, PID controller performance tuned by such algorithms usually degrades as process conditions change, requiring a process engineer to monitor controller performance. If controller performance deteriorates, the process engineer is required to “re-tune” the controller.
Controller performance deteriorates for many reasons, although the most common cause is changing dynamics of the process. Since PID controller performance has been related to the accuracy of the process model chosen, a need exists for a PID controller that allows for such uncertainty by accounting for changing system dynamics and, desirably, by incorporating the same before any tuning parameters are calculated. A further need exists for a means to extend the above-described MPC techniques into PID controller design techniques.
SUMMARY OF THE INVENTION
To address the above-discussed deficiencies of the prior art, it is a primary object of the present invention to provide a controller that determines a “best” controller to achieve optimal control performance within a process facility on a “worst case” process system and, more precisely, the controller allows for model uncertainty by accounting for changing system dynamics by incorporating the same before the tuning parameters are calculated.
In the attainment of this primary object, the present invention provides a process control system and a method of operating the same for controlling associated processes within a process facility. The control system includes at least one RPID controller that is associated with a processing system. The controller includes a storage device and a processor. The storage device is operable to represent (i) at least one of a plurality of associated processes mathematically to define the various relationships among different inputs and outputs of the one or more represented associated processes, and (ii) uncertainty factors that are associated with these defined relationships. The uncertainty factors define a range of dynamics across which the one or more represented associated processes operate, an error in the mathematical representation, or, alternatively, some combination of the same.
Responsive to the mathematical representation and the uncertainty factors, the processor is capable of determining (or is operable to) tuning parameters for use by the control system to control the one or more represented associated processes and thereby cooperate to optimize said process facility.
According to an advantageous embodiment of the present invention, a robust PID processing system determines the optimal controller tuning for the specified range of process dynamics. It is well known that process dynamics move within a range for a variety of reasons. For instance, the following list provides common plant occurrences that may change the way a process will respond to PID control: (i) process throughput is increased or decreased, (ii) feed stock quality is changed, (iii) seasonal temperature changes, and (iv) equipment becomes fouled. In each case, an RPID controller according to the present invention outperforms a conventionally tuned PID controller as the process dynamics shift, thus yielding increased economic benefits.
As introduced above, other controller tuning techniques commonly base PID tuning parameter results on a single process dynamic model. If uncertainty factors are specified for the gain, settling time, or dead-time, the calculated solution is simply “de-tuned” to compensate for the uncertainty. Unlike the robust PID of the present invention, these packages do not find the most responsive controller parameters for all models within an uncertainty range.
The foregoing has been tested in process plant applications such as within a stripper temperature controller wherein various PID tuning methods were applied, yielding the following results:
Tuning M

Affiliated with

Also associated with

Rating

Controllers that determine optimal tuning parameters for use... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Controllers that determine optimal tuning parameters for use..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Controllers that determine optimal tuning parameters for use... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2547220