JPS58168103A - Feedback controlling system - Google Patents

Abstract

PURPOSE:To attain the adaptive control even for a process where the characteristic change is unclear, by detecting the presence or the absence of hunting of an output due to a control parameter prepared in a table and switching the control parameter into a different control parameter when the hunting is detected. CONSTITUTION:The table 7 for the 1st-the n-th rank of control parameters corresponding to the range of fluctuation of the process characteristics of concern, is prepard, the strongest control parameter among them, e.g. the 1st rank, is selected and set to a feedback controller 8. Its output (u) is given to a process 2 through a final control element 1, and the presence or the absence of hunting phenomenon is detected by inputting a process variable (x) and the output (u) of the final control element to a hunting detector 4. In detecting the hunting phenomenon, a changeover switch 9 is switched and the output of the final control element is fixed to that before the hunting takes place and stored in a storage device 5. The switch 9 is changed over after a prescribed time, and the control parameter is switched into the 2nd rank parameter for the execution of control.

Description

DETAILED DESCRIPTION OF THE INVENTION This invention relates to a feedback control method for process characteristics. For details, see page 1. In the scissors control method, control parameters are adapted to the preprocess OIF# characteristics to achieve feedback control of good 1. This is intended for use in adaptive control to achieve this.

Generally, in feedback batter control of a process, it is important to appropriately set control parameters (for example, proportional gain, integral time constant, and differential time constant) according to the characteristics of the process.

However, since the percentage of the process changes depending on the operating conditions of the process, it is necessary to change the control parameters accordingly, and control that satisfies such conditions is called adaptive control.

If the above-mentioned adaptive control is not performed and the control parameters are too strong for the process characteristics at that time, a so-called hunting phenomenon occurs in which the fluctuations in the process variables and the fluctuations in the manipulated variables are related to each other and diverge. If left unattended, the process may diverge and become non-hazardous. On the other hand, if the control parameters are set weakly so that hunting does not occur under any operating conditions of the process, the control parameters will be too weak during normal operation, and the safety level against hunting is more than sufficient. The coordination of control is impaired accordingly, and sufficient control performance cannot be obtained.

When trying to perform adaptive control to solve the problems mentioned above, if the process characteristics depend on the operating conditions and are expressed by physical formulas, it is difficult to calculate the process characteristics using the physical formulas. It is possible to determine this and vary the control parameters depending on it. In fact, an example of achieving adaptive control by approximating changes in process characteristics and changing control parameters accordingly is disclosed in Japanese Patent Application No. 54-41950.
No. C JP-A-55-134120),%
The inventors of the present invention have proposed this method in, for example, No. 89950 (Patent Publication No. 17008/1989).

However, in reality, there are very few processes in which changes in characteristics can be accurately expressed by physical formulas, and therefore it is difficult to implement adaptive control for processes in general. The actual situation is 9.

The present invention has been made in view of the above-mentioned conventional technical circumstances, and an object of the present invention is to provide process feedback that enables adaptive control even for processes whose characteristics change is unknown. The objective is to provide a control method.

In order to achieve the above object, in the present invention, in a process feedback control method, a table of several levels of strength is provided as control parameters in advance,
It has the function of sampling manipulated variables and process variables at predetermined time intervals to detect hunting phenomena in the process, and stores nine manipulated variables for a predetermined period of time after sampling.
When the hunting phenomenon is detected, the feedback flj11 loop is cut, the manipulated variable is fixed for a certain period of time to the value before the hunting phenomenon occurs, and the hunting is suppressed, and then the control parameters are changed to one step looser.
Control using the feedback control loop was executed again.

Next, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing one embodiment of the present invention. In the figure, l is the operating end, °2 is the process, 3 is the process variable detection end, 4 is the process control hunting detection device, 5 is the storage device, 6 is the timer, 7 is the control parameter table, and 8 is the feedback adjustment needle. , 9 indicates a changeover switch.

Please refer to FIG. The process 2 is controlled by the operating end l, and as a result, the changed process variable X is measured by the detecting end 3. Table 7 of the control parameter is prepared in advance with control parameters ranging from 1 rank to n rank corresponding to the possible variation range of process characteristics (1 rank is the strongest control parameter, and n rank is the weakest control parameter). ).

Hiro in normal operation, 1 rank 10 from the table 7
The control parameters are selected and set in the feedback controller 8, and the control output of the eight cough controllers is calculated and applied to the operating end IK via the switch 9, which is used as the operating end output. At this time, using the process variable value X and the operating end output U, the hunting detection device 4 detects whether or not a hunting phenomenon has occurred. When the hunting phenomenon is detected, the changeover switch 9 is switched to the dotted line position to turn off the feedback loop and instead, the timer 6 is used to change the operating end output for a certain period of time to the operating end output before the hunting occurred. Fix it to u□.

u6 can be obtained by storing the O operating end output in the storage device 5, going back from the current time and up to the time required for hunting detection. After a certain period of time has elapsed, the changeover switch 9 returns to the solid line position and feedback loop control is restarted, but during this time the second rank control parameter is taken out from the table 7 and set in the controller 8. Feedback control based on the new parameters is performed.

Thereafter, when a hunting phenomenon is detected, the same switching procedure is performed, and then feedback control is restarted using control parameters lowered by one rank.

Note that in FIG. 1, the timer 5 may be replaced with a tracking control section (not shown). The tracking control unit tracks the output U of the operating end l via the storage device 5, and calculates the temporal integration of the deviation between the operating end output uJ before hunting occurs and the operating end output U after hunting occurs. When hunting is detected by the detection device 4 and the C-off JIX switch 9 is switched to the dotted line side (that is, the output side of the tracking control section), the time of the deviation determined so far is determined. The tracking control section outputs an operation output that changes at a constant gradient over time in a direction that cancels out the deviation over time and leads to zero, and when the time-integral value of the deviation cancels out and becomes zero, the changeover switch 9 is turned on. may be returned to its original solid line position to restart feedback loop control.

Of course, during this time, the second rank control parameters are set in the controller 8.

Figure 2 is a time chart of the signals of each part in the circuit of Figure 1, where light) is the process variable X% (b) is the manipulated variable U.
, (c) show temporal changes in the disturbance W, respectively.

See Figure 2. It is now assumed that feedback control is being stably implemented in time interval A up to time t1. Assume that at time t1, the disturbance W changes stepwise and is added to the process. Process control hunting occurs during the control operation for this disturbance, and the hunting phenomenon is detected at time t2, that is, at the third peak of the vibration of the process variable 9 is switched to the dotted line position 9, the manipulated variable U is kept at the value uQ at time tl, and this state persists over the continuing time interval C, and hunting in the process control also subsides. Time t3
At , feedback control is restarted with the control parameter being lowered by one rank, and thereafter stable control is performed until time t4.

At time t4, the disturbance W further changed stepwise and was added to the process, but this time, hunting did not occur due to this, so stable feedback control was maintained.

According to the present invention, in process feed control, the unstable state of the process due to hunting can be kept to a minimum (minimum time) over a period of time, and feedback control can be performed using control parameters that always correspond to the characteristics of the process. Therefore, control results with good coordination can be obtained.

Note that the present invention is applicable to general process control that performs feedback control.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG. 2 is a time chart of various signals in the circuit of FIG. 1. Symbol explanation 1...Operation end, 2...Process, 3.
...Process variable detection end, 4...Nching detection device, 5...Storage device, 6...
...Timer, 7...Control parameter table, 8...Feedback adjustment needle, 9...
・Transfer Switch Representative Patent Attorney Akio Namiki Patent Attorney Kiyoshi Matsuzaki Figure 1

Claims (1)

[Claims] 1) In the process feedback control method,
means for detecting a hunting phenomenon in the dragon speed control; means for disconnecting a feedback control loop when a p hunting phenomenon is detected by the detecting means; means for changing a control parameter after disconnection; and feedback after the change. and means for restarting the control loop.