Data processing: generic control systems or specific application – Generic control system – apparatus or process – Optimization or adaptive control
Reexamination Certificate
2001-11-08
2004-06-15
Patel, Ramesh (Department: 2186)
Data processing: generic control systems or specific application
Generic control system, apparatus or process
Optimization or adaptive control
C700S028000, C700S031000, C700S037000, C700S042000, C700S045000, C318S609000, C318S610000
Reexamination Certificate
active
06751510
ABSTRACT:
FIELD OF THE INVENTION
This invention generally relates to the field of process control, and more particularly to a predictive and self-tuning PI (Proportional plus Integral) control design which enhances the performance achievable with conventional PI control while capitalizing on the strong user-support established and available for a simple PI control structure.
BACKGROUND OF THE INVENTION
PI control has existed for more than fifty years. The current methodology in tuning a PI controller has been essentially to deduce and infer a set of fixed control gains from the characteristics of the process. Over the years, the control gains tuning process has evolved from simple and heuristic procedures to more elaborate and exact computations based on mathematical models of the process, resulting in increasing performance yields.
However, whichever inferencing procedure is used, the resultant control gains derived thereby are mostly fixed. As specifications on control performance are tightened by requirements for increased precision with greater complexity of processes being controlled, the fixed nature of the control gains has direct implications on the range of processes for which the known PI control structure continues to be applicable and useful. For example, the conventional fixed-gains PI controller is well-known to be inadequate for a number of classes of linear systems, including: time-delay processes, unstable processes, and processes with time-varying parameters.
Advanced and more complex control methods may be recommended for these systems. However, such methods require more structurally-elaborate devices which, in themselves, act as a dis-incentive to adoption and utilization. Moreover, there remains the decades-old dilemma that has been facing hordes of control engineers, in that under real working conditions (e.g., including model imperfections, modest process operators, limited availability of technical support, short breakdown recovery tolerance, etc), the benefits of advanced control over PI control have never been quite obtainable. This is one main reason as to why the traditional approach to PI control remains widely used despite the clear limitations associated therewith.
OBJECTS OF THE INVENTION
It is accordingly an object of the present invention to overcome the limitations of prior art PI controls.
It is a more specific object of the invention to provide improvements over traditional forms of PI control in a novel control apparatus which retains the conventional PI control structure, and thereby to capitalize on the massive user-support established for PI controllers over the decades.
It is still a more particular object of the invention to provide a novel method of and apparatus for control gains-tuning and continuous self re-tuning of control parameters of a process based on a GPC (Generalized Predictive Control) approach to optimize performance indices.
It is yet a more detailed object of the invention to provide a predictive and self-tuning PI controller.
It is still a more specific object of the invention to provide a predictive and self-tuning PI control apparatus which expands the domain of PI control applications to include time-delay processes, unstable processes, and time-varying processes.
It is thus an object of the invention to provide an improved PI controller which, by utilizing a predictive and self-tuning approach to control, is unified in design for different classes of processes, is applicable to a broader range of linear processes than the prior art, and yields performance improvements beyond traditional fixed-gains PI control.
SUMMARY OF THE INVENTION
In accordance with the foregoing objects, the present invention provides a novel predictive and self-tuning PI controller, which optimizes performance indices derived from simple and classical user specifications.
In accordance with another feature of the invention, there is provided a method for designing a predictive and self-tuning PI controller which is based on a GPC design approach.
In accordance with the invention, there is thus provided a control apparatus having a design which is naturally amenable to optimal control applications.
In accordance with another aspect of the invention, there is provided a control apparatus having inherent predictive and self-tuning capabilities which expand the PI control application domain to include time-delay processes, unstable processes, and time-varying processes.
In accordance with a more specific feature of the invention, there is provided a control apparatus comprising a novel combination of components, including a control structure similar to a conventional PI control, and a plurality of novel elements, including a gains design means, a gains scheduling means, and a gains self-tuning means.
In accordance with another aspect of the invention, there are provided novel design, scheduling and self-tuning algorithms for the gains of a PI controller, and apparatus for implementing the same.
In accordance with yet another feature of the invention, there is provided a control apparatus including a PI process controller and a gains design means for designing gains of the process controller, the gains design means being responsive to a user specified parameter set including parameters representing a first-order mathematical model of the process, a parameter representing an estimate of a time-delay of the process, parameters representing a desired closed-loop damping factor and natural frequency, and a sampling interval, and which, in response to the input parameters specified by the user, generates a plurality of time functions associating the PI control gains to the input parameters.
It is still another feature of the invention that the user specified parameter set which is input to the gains design means may include an optional prediction horizon and that, when the optional prediction horizon is unspecified, the gains design means recommends a default prediction horizon.
In accordance with a more detailed feature of the invention, the control apparatus also includes a gains scheduling means for assigning appropriate gains to the PI controller based on a schedule, the schedule depending on elapsed time following a set-point change according to the time functions generated by the gains design means. The gains scheduling means is further responsive to a set-point signal defining a desired process variable, and outputs are the actual time-scheduled gains to be assigned to and updated on the PI controller.
In accordance with still a more detailed feature of the invention, the control apparatus further includes a gains self-tuning means for continuously adjusting weighting parameters of the time functions generated by the gains design means to optimize performance of the PI process controller. The gains self-tuning means is further responsive to a set-point error which is the difference between the desired and actual process variables. The outputs from the self-tuning component are a set of replacement weighting parameters for the time functions generated by the gains design means.
REFERENCES:
patent: 4768143 (1988-08-01), Lane et al.
patent: 5185800 (1993-02-01), Mahieux
patent: 5347446 (1994-09-01), Iino et al.
patent: 5394322 (1995-02-01), Hansen
patent: 5404289 (1995-04-01), Hang et al.
patent: 5847952 (1998-12-01), Samad
patent: 6154499 (2000-11-01), Bhaskar et al.
Huang Sunan
Lee Tong Heng
Tan Kok Kiong
Wang Qing Guo
Patel Ramesh
The National University of Singapore
LandOfFree
Predictive and self-tuning PI control apparatus for expanded... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Predictive and self-tuning PI control apparatus for expanded..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Predictive and self-tuning PI control apparatus for expanded... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3350673