Data processing: generic control systems or specific application – Generic control system – apparatus or process – Optimization or adaptive control
Reexamination Certificate
1998-07-31
2001-02-20
Butler, Dennis M. (Department: 2787)
Data processing: generic control systems or specific application
Generic control system, apparatus or process
Optimization or adaptive control
C700S034000, C700S047000, C700S200000
Reexamination Certificate
active
06192283
ABSTRACT:
TECHNICAL FIELD
This invention relates to methods and devices for controlling a system parameter for machines, such as temperature, position, or pressure, and, more particularly, for adaptively controlling such a parameter by “learning” from previous system conditions and previous outputs, which have been recorded such as in a look-up table or the like, and adjusting the output to the system according to the results obtained from previous outputs.
BACKGROUND OF THE INVENTION
Over the years, a variety of control systems have been developed for controlling the output of machines or for the control of a particular variable in a process or device. For example, it has been known to use computer numerical controls to control the temperature and pressure in an injection molding machine. Such a conventional control system can utilize feedforward and feedback control loops for controlling the variable, the feedforward control being operative for canceling an incoming disturbance, while feedback control is operative for responding to the result of the action which has been taken. The two types of controllers may be combined whereby the feedback controller can compensate for any error remaining after the feedforward action has been taken.
An example of a feedback controller is a PID (proportion, integral, and derivative) controller, a device which creates a control function which is the sum of three terms, one which is proportional to the error signal, one of which is proportional to the integral of the error signal, and one which is proportional to the derivative of the error signal. Implementation of PID control is usually accomplished with analog circuits using RC (resistor and capacitor) networks and active filters. Alternatively, software may be utilized to simulate the functioning of the PID controller.
While conventional control systems are adequate for controlling certain systems and devices, such control systems have several disadvantages. For example, conventional control systems have been inadequate for providing extremely precise or “tight” regulation of output and, thus, have produced unacceptable levels of overshoot or undershoot of the controlled variable. Moreover, conventional control systems need to be tuned at setup to properly control the system or device and, because the response of a system or device may change over time, such control systems need to be periodically tuned to account for such changes in response. For example, once the gains of a PID controller are set, they will generally not adapt to changes in the controlled system as parts break in or as parts are replaced, and the user must periodically “tweak” the gains to account for such changes. In addition, the cost of conventional controllers, such as PID controllers, can be high.
Control systems have been developed which adjust to changes in the system or device over time. For example, U.S. Pat. No. 5,130,920, issued to Gebo, discloses a control system for the control of temperature, such as the temperature of a liquid in a process, through the use of a feedforward processor, a feedback processor, and an adaptive (or tuning) processor. The adaptive processor receives the outputs of the feedback and feedforward processors and adjusts them, if the actual temperature is outside of a certain range, by multiplying the outputs by a tuning constant, which is calculated based upon the outputs of the feedback and feedforward processors. The adaptation process continues on each iteration until the output from the feedback controller is reduced to zero, in order to maximize feedforward controller tuning.
It has also been known to utilize fuzzy logic to control complicated devices that include non-linear elements, and to alter the fuzzy membership functions utilized when the results of the fuzzy logic differ from desired values. U.S. Pat. No. 5,602,966, issued to Kinoshita, discloses a method and device for modeling fuzzy membership functions for use in control of a device. The invention permits automatic incremental alteration of the membership function either by shifting the function or changing its shape until the results of the inference rules obtain a desired value. The inference rules take the form of “If . . . Then” statements, and the membership functions are variables that may comprise a number of levels, such as high, medium, or low, based upon the precise numerical value of the variable. If the inference result obtained through the inference rules does not satisfy a permissible range of predetermined target inference results, the membership functions are altered by a predetermined amount. The altered membership functions and rules are then used to obtain a control variable that is used to control the controlled object.
U.S. Pat. No. 5,135,688, issued to Nakamura et al., also discloses the use of fuzzy logic for controlling a device including a fuzzy inference control method for an injection molding machine. The method contemplates use of the fuzzy inference to define a control value by inferring the status of the machine based upon the difference between the object temperature and the current temperature, as well as the temperatures of parts located between thermocontrolled components. An actual control value is calculated based upon the fuzzy inference, and the temperature of the machine is controlled using the actual control value.
While the use of the methods and systems disclosed by these references can have certain advantages in controlling a system or device, such methods and systems are not without disadvantages. Generally, such processes can be time-consuming, relatively complex to produce, and result in significant overshoot and undershoot of the desired result. Thus, a more simple and efficient method would be desirable, such as a method in which prior system outputs are recorded in a database or buffer, along with the response obtained from these prior outputs, and this recorded data is used to adapt or adjust future outputs to the system. None of these references disclose or suggest such a process.
Thus, there is a continued need for a more efficient adaptive control system that eliminates the need for conventional control devices, as well as the periodic tuning of such devices, and that quickly and accurately controls the system or device with minimal overshoot and undershoot.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a control method and apparatus that eliminates the need for periodic tuning of control devices.
It is another object of the present invention to provide a control method and apparatus that is simpler and less expensive to implement than conventional control systems.
It is yet another object of the present invention to provide a control method and apparatus that virtually eliminates overshoot and undershoot of the desired system parameter setting, and, thereby, provides tight control over the parameter.
Another object of the present invention is to provide a control method and apparatus that stores prior control values and the results obtained from such values, and adjusts future control value based upon the past results, thereby “learning” from or “adapting” to prior attempts or cycles.
A further object of the present invention is to provide a control method and apparatus that can be utilized to control a variety of system parameters and variables.
To achieve the foregoing and other objects, and in accordance with the purposes of the present invention as described above, a method for controlling a system parameter is provided comprising the steps of monitoring a current system status, accessing a data structure having data locations for at least one prior system status, and determining whether the current system status matches a prior system status in the data structure. The current system status corresponds to the status of a system to be controlled by the method, and each prior system status corresponds to a previously assessed status of the system. In addition, each prior system status has a corresponding prior output and a corresponding prio
Butler Dennis M.
Dinsmore & Shohl LLP
Siemens Energy & Automation Inc.
LandOfFree
Method and apparatus for adaptive control of a system or device does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for adaptive control of a system or device, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for adaptive control of a system or device will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2606887