JPS6355603A - Control system for self-tuning controller - Google Patents

Abstract

PURPOSE:To ensure the simple handling with a self-tuning controller by connecting a proportion arithmetic part and an integration arithmetic part in parallel with each other at the time of in a tuning mode and observing the limit cycle during the process control. CONSTITUTION:A self-tuning controller 10 contains changeover switches SW1-SW5 and sets selectively normal operation mode and a tuning mode according to the connection states of those switches. A proportion arithmetic part 11 is provided to the controller 10 together with an integration arithmetic part 12, a differentiation arithmetic part 13, an output part 14 which delivers a manipulated variable (m) obtained by adding each output together to a process 1, and an input part 15. Furthermore an observing unit 16 for deviation (e) is added together with a mode switch control part 17 and nonlinear elements 18A-18B. Thus it is possible to follow up the observed value and set value of the limit cycle in a tuning mode by means of the part 12 even through the values of two positions of the nonlinear elements 18 are reduced. The the influence given to a process 1 can be evenly reduced regardless of the gains of the process 1.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides a feedback control system for controlling a controlled object (
This invention relates to a control method for a controller that performs a PID calculation on the deviation between a control amount fed back from a process and a set value, and outputs the obtained manipulated variable to the process.

[Conventional technology]

Controllers such as the above are widely used in industry, but when a change in process characteristics occurs during control execution, it is important to update the PID parameters promptly to continue optimal control. This is a prerequisite for

For this purpose, conventionally a nonlinear element is added to the controller, and when determining the process characteristics, the nonlinear element* is inserted into the signal path, and a discontinuous control operation (generally two-position control) is performed for the deviation. It is proposed that it be done. If a limit cycle occurs in the process, its waveform (
It is easy to determine process characteristics and optimal parameters from observations of amplitude and period.

[Problem that the invention seeks to solve]

However, in the conventional example described above, when the value of the second place value (absolute value) M of the nonlinear element is increased, the fluctuation range of the manipulated variable becomes larger. It is not suitable unless you need it.

On the other hand, since the limit cycle uses the process equilibrium point (set value sp = manipulated variable m when controlled variable pv) as the operating reference point, if M is made too small, disturbances may occur in the process or the set value If there is a large change in the limit cycle, the limit cycle may no longer occur. Furthermore, the control amount in such a case will remain far from the set value.

As described above, the conventional controller as described above has poor practicality as a self-tuning controller having a function of automatically determining optimal parameters.

[Means for Solving the Problems] The present invention continues process control in the tuning mode with a proportional calculation section and an integral calculation section connected in parallel and a nonlinear element inserted before the proportional calculation section. However, the limit cycle is observed, and the amplitude of the limit cycle is determined from the integrated value of deviation values sampled at a constant period.

[Effect]

In tuning mode, by using the integral calculation unit, it is possible to observe the limit cycle and follow the set value even if the values at two positions of the nonlinear element are reduced, and the output of the nonlinear element is Since it acts on the process via the proportional calculation section, the influence on the process can be reduced to - regardless of the gain the process has.

Furthermore, when determining the amplitude of the limit cycle, the peak value is not directly read, so even if noise is added to the peak value, it will not be taken as the amplitude.

〔Example〕

FIG. 1 is a block diagram showing one embodiment of the present invention. In the figure, a self-tuning controller 210 of this embodiment includes changeover switches SWI to SW5.

These switches are all switched in conjunction, and A(
When it is turned on the Auto) side, it is normal operation mode.
The tuning mode is when the power is turned to the T (Tnnlng) side.

However, the controller of this embodiment has each switch SWI.
- Two modes, normal operation mode and tuning mode, are selectively set depending on the connection state of SW5.

Here, the controller 10 includes a proportional calculation section 11, an integral calculation section 12, a differential calculation section 13, and an output section 14 that outputs the operation ftm obtained by adding the outputs of each of these calculation sections to the process 1; The input section 15 calculates the deviation e between the control fLpv fed back from the control fLpv and the set value ap from 1 and supplies it to the calculation section. Outputs the given manipulated variable m.

m=mt+m2+ms K, TI, and Td are proportional gain, fractional time, and differential time, respectively, and these are called PID parameters. The first term on the right side is the output ml of the proportional calculation section 11, the second term is the output m2 of the integral calculation section 12, and the third term is the output m3 of the differential calculation section 13. Depending on the nature of the process and the purpose of control, various modifications are possible, such as eliminating the differential term (PI control) or making the proportional/derivative term act only on the p-ma (IPD control), but these are collectively known as Such control operation is called a PID controller.

In this way, the controller 10 performs general PID control in the normal operation mode, but at that time, the observation device 16 monitors the movement of the deviation e.

When a predetermined tuning start condition is satisfied, such as when the absolute value of the deviation monitored by this observation device 16 exceeds a predetermined value or when its responsiveness becomes favorable, the mode switching control unit 17 is switch S'w1
~SW5 is switched to the T side, and the controller 10 shifts to tuning mode.

In the tuning mode, a second
Nonlinear element 1 with two-position nonlinear characteristics as shown in the figure
8A is inserted, and a nonlinear element 18B having characteristics as shown in FIG. 3 is inserted before the integral calculation section 12.
These are connected in parallel. That is, K,
With Ti unchanged and Td = 0, the differential operation disappears, and the manipulated variable m is basically the following (2) in the range -M≦e≦M,
(This is the sum of the outputs from equation 31.

ml=± to -M (sign depends on the sign of e)...
(2) Even in this case, it still works to eliminate the deviation e. That is, the process feedback control system is still maintained.

The observation device 1 monitors whether the vibration of m2 is sufficiently small compared to ml and whether the vibration of e is sufficiently small compared to m, and if both conditions are not satisfied, the Decrease K and increase TI. This is K, Tl
This is to keep the process state on the safe side when the Actions matter.

If the monitored movement satisfies each of the above conditions, wait until the vibration of the deviation e becomes a stable continuous vibration, that is, a limit cycle, then observe the amplitude X and period Tc of the vibration, and record the data. The signal is sent to the regulator 19.

Here, to find the amplitude of the limit cycle,
In contrast to the conventional method of simply detecting the peak value, a method is used to obtain the peak value from the integrated value of the output (instantaneous value) detected at a constant sampling period Ts' (for example, 03 seconds).

The signal to be observed is, for example, as shown in Figure 4 (4
) It is theoretically known that if the area of a half wave (1/2 period) is determined by integration, the following is obtained.

y=lx Convert 2 more Yn-Yn-, =T, @Xn Therefore, for half a cycle, Y, -yo=T,X.

Yn - Y6 = TelΣX1 1 = 1 Assuming the initial value Yo-0, the area of the half period is Yn = T,
i ・・・・・・・・・・・・・・・・・・
(6) 1w1. Therefore, (5), (0'a=Y from formula 61
n) In this way, by simply integrating each sample value X1t, the amplitude can be easily obtained by calculating the equation (7) at the timing when half a cycle has passed. In the conventional method of directly detecting the peak value, for example, if even one noise is added to the peak as shown in Fig. 5, an abnormal peak value may be taken as it is, but with this method, Such problems can be solved. Moreover, it is an extremely simple algorithm and is suitable for controlling processes and deriving power from power. Further, as shown in FIG. 6, for example, even if the waveform is slightly distorted, the peak value can be calculated accurately.

peak -to -p@ak
) is determined from data for one period (two half waves) using the following equation.

From the vibration @X and the period X obtained in this way, the regulator 19 calculates the limit sensitivity Kc and the vibration period T'co' at that time by the conventional limit sensitivity method.

If the hysteresis characteristics of the nonlinear elements 18A and 18B are ignored, this can be obtained from the following equation (9), (1(l1).

Kc = 4KM/παX ・Old・・・・・・・・・
・ (9) TeO'=αTc ・・・・・
・・・・・・・・・Orj where M is 2 of nonlinear element 18
It is a position value.

The reason why Ke and Tco' are expressed by the above equation will be explained next.

Figure 7 is a diagram showing the limit cycle generation point, in which P) is the process transfer function Gp(jω), and (B) is T.
-1/H Nyquis)11 obtained from the descriptive function N(X, ω) of the controller 10 when l is set to a significant value.
It is a diagram. The intersection point of B) and (B) indicated by circle in the figure is the point where the limit cycle occurs, while the point at -180° of phase of cp indicated by C in the figure is the nonlinear element 18. Limits in a control system using only proportional control 4 parts 11
This is the point at which a cycle (this is called an ideal limit cycle) occurs. If the amplitude Xo and period Teo of the ideal limit cycle are found, Kc and T,. ′ is determined by the following formula.

Kc =4KM/πX, ・・・・・・・・・・・・
...Q3Tea'= T'co...
・・・・・・・・・・・・・・・・・・(13 However, it is difficult to obtain these data directly from the actual process, that is, while continuing process control.) Therefore, in the present invention, the data of the ideal limit cycle generation point is determined from the data of the limit cycle generation point in the above-mentioned tuning state.

First, the transfer function Gp(a) indicating the characteristics of the process
(II is the Laplace operator) focuses on the movement from the equilibrium point and is approximated as follows.

Gp(g)=e −LB/T@・・・・・・・・・
...... α4) L: Delay time T: Response slope On the other hand, the generated limit cycle is approximated by a sine wave,
It is expressed by the following formula.

e=Xsinωt・・・・・・・・・・・・・・・
・a9 Since the descriptive function of the nonlinear element having the above-mentioned two-position nonlinear characteristic is 4M/πX, the output f
fi 1 + n12 and operation im are each shown as follows.

m=m1+m2………………−・αmo control fitp
The vibration component of v is e. Then, α(, according to formula 08, here the deviation e = -e6, so (Is, from formula a9, the angular frequency ω is changed to ω = 2π/T, multi-period Tc, and then tuning is performed. If we also eliminate the integral action in the state configuration (i.e. 71 = oo), then the limit cycle at that time is the ideal limit cycle, e = X6 sinωat...
...Tc with Q4 period as Tco (=2π/ω0). =4L ・・・・・・・・・・・・
...It turns out that (a) is true.

Here, in order to find the process glue and T itself, xo
When calculating l 'rc0, the formula of CD and (a), ■
According to the formula,
...(to), XOIT co can be easily found. However, in the process of deriving 匈~(2), t3n −” 0 = θ
(θ is a sufficiently small angle). This approximation is valid enough when the system π X stem is stable (when the zero value is sufficiently small).

These ideal limit cycle data Xo.

By substituting T'co into a'a, Equation (13), the above-mentioned Equation (9), 0 can be obtained.

Here, α is expressed by Equation 09, but this is the case where the hysteresis of the nonlinear elements 18A and 18B as described above is ignored, and in reality, a correction for the hysteresis is added to this. ), deviation ef, and CIs expression, ”=Xsin(wt-φh) −・・−−
−−−−−−・−(24, and a phase delay of φh occurs.

Therefore, the calculation formula for α is corrected as follows.

Here, since the value of φh in the 3p equation is small, 5111-'
An approximation is made by setting h Y to Y.

If the nonlinear element 18A does not have a hysteresis component, for example, as shown in FIG. 8, when the multiwaveform is disturbed by the influence of noise especially when the deviation e is close to zero, the value of ml will fluctuate sharply. For example, a situation may occur where the opening degree of the pulp suddenly changes. As a result, it becomes difficult to understand the period Tc.

On the other hand, if hysteresis is provided, for example in this embodiment M = -2% and h = 0.2%, but ±
If the fluctuation is within the range of 0.2% to 0.4%, it can be absorbed by the hysteresis component as shown in FIG. In reality, the noise generally falls within the above-mentioned range, so sudden changes in the manipulated variable can be avoided, and the well-known T. Easy to take.

Note that the deviation e is set to be equal to the normal PI calculation operation for the range of -M,
This is because when the deviation (absolute value) is so large, there is an urgent need to bring the deviation close to zero as soon as possible, even if a limit cycle or the like occurs.

Furthermore, in this embodiment, the nonlinear element 18B at the front stage of the integral source word section 12 is also provided with hysteresis. This is to ensure that this occurs.

As described above (9), the adjuster 19 further determines the optimum PID parameter from Kc + Tco' determined by Ql2, and sends it to each calculation section. This can be done by known methods such as the Ziegler-Nichols method as shown in the following table.

, Incidentally, as described above, the selection of PI and PID is performed manually or automatically depending on the nature of the process.

When K and Kl are already approximately appropriate values, the value of α is often approximately 1, such as during tuning when stable normal operation continues, and in that case, (9 ), (The calculation speed of Equation 11 is further increased.

After determining the above PID parameters, the state switching control circuit 17
restores the switches SWI to SW5, and thereafter normal operation is performed using the updated PID parameters. P.I.
The D calculation is naturally performed using a velocity type calculation.

By the way, even during tuning, the controller 10 operates by maintaining the closed loop control system using the υ feedback so as to cancel the deviation e, but at that time, the apportionment calculating section 12 is not connected. For example, if a disturbance occurs in the process, 3p changes significantly, and the value of M is small, the nonlinear element 18A
Even in a case where it is impossible to track the vibration by using only the above-mentioned integral calculation unit 12, it is possible to cancel the deviation e by effectively changing the base point of the vibration caused by the nonlinear element 18A. In other words, the value of M of the nonlinear element 18A is small, for example, 1 to 1 of the entire movement range of the manipulated variable.
Since about 10% is sufficient, there is little variation in the process during tuning, and fast response can be ensured.

Specifically, such a controller 10 can be realized by using a microcomputer, for example, and executing a program stored in a memory in advance. Of course, each proportional, integral, and differential calculation unit, nonlinear element, input/output unit, each switch SWI to SW5, control unit, observation device, regulator, etc. can be constructed by individual components each having such a single function. It goes without saying that this can be achieved.

FIG. 10 shows an example using a microcomputer 20.
21 is a processor unit such as a microprocessor (C
PU), 22 is fixed memory rROM), 23 is variable memory! J (RAM), 24 is the input/output port, 25
is a setting/indicator equipped with controls for setting various constants, etc., and a display.

The CPU 21 functions as the controller 10 by chronologically executing programs stored in the fixed memory 22 in advance. This is shown in Figure 11 and 1.
This will be explained using Figure 2.

FIG. 11 is a flowchart showing an example of a main program processed by the CPU 21.

In the same figure, after initialization processing (step 101), in input processing (step 102), the CPU 21 receives feedback from process 1 via a communication path or a set value sp given from the setting/display unit 25. The control amount pv is digitally converted and taken in via the input/output port 24. Next, the CPU 21 performs control arithmetic processing to determine the operation -Kr-111 (step 103), and in output processing (step 104) converts the operation amount m into analog via the input/output port 24 and outputs it to the process 1. The above input processing step 102 to output processing step 104 are repeatedly executed at a constant sampling period.

FIG. 12 is a flowchart showing an example of the above-mentioned control calculation processing program. When the execution of the program shifts to this control calculation processing, the CPU 21 uses the variable memory 23
Stew 2/ by the flag set up using the designated area of
Check to see if it is in log mode (step 201)
, if it is not the tuning mode, the set value 3p and the control amount p
(step 202), and determines whether it satisfies a predetermined tuning start condition (step 203). If not, the PID currently stored in a predetermined area of the variable memory 23 is Using the parameters, the manipulated variable m is calculated by normal PID calculation (step 204). That is, this case corresponds to the case where the switches SWI to SW5 are turned to the A side in FIG.

On the other hand, if the tuning start condition is satisfied (step 203), after setting the tuning mode (step 205), the nonlinear element 18 is inserted before the proportional calculation section 11, and an integral calculation section is inserted in parallel with them. 12
The limit cycle is determined assuming a configuration in which the two are connected (step 206). At this time, the manipulated variable m is naturally also determined. Then, when Qυ, X and Te in equation (c) are found from the limit cycle (step 207) (this corresponds to the limit cycle being stabilized), from these data, for example, Ziegler
The optimal PID parameter is determined according to the Nichols method, and the previous PID parameter value stored in a predetermined area of the variable memory 23 is updated (step 208).
. Then cancel tuning mode (step 2)
09). Therefore, in executing the next PID calculation step 204, the calculation is performed using the new P'ID parameter.

In this way, optimal control is maintained while updating the PID parameters as appropriate.

Although we have described the case where the deviation e is monitored and the state is automatically shifted to the tuning state depending on the state, other conditions, such as the passage of a predetermined time, may be used as the tuning start condition, and the PID parameters are updated periodically. You can do it like this. Further, for example, a function may be added to the operation input from the setting/display section 25 so that the tuning state can be entered at any time.

〔Effect of the invention〕

As explained above, according to the invention, the proportional calculation section and the integral calculation section are connected in parallel during the tuning mode, and the limit calculation is performed while continuing process control with the nonlinear element inserted in the front stage of the proportional calculation section. Since the cycle is observed, a self-tuning controller that is easy to handle and can be used in a process that requires quick response can be realized at low cost.

In other words, there is no need for conventional expensive automatic adjustment devices, etc., and the configuration is simpler than conventional automatic adaptive controllers, and even when implemented on a computer, it is easy to configure and requires less memory capacity. It does not cause an increase in

Further, the change in the amount of operation given to the process for tuning is small, and the sampling time is equivalent to a normal DDC (direct digital controller).

In addition, by determining the amplitude of the limit writing cycle from the integrated value of the output values sampled at -regular intervals, the tuning operation can be made robust against noise. can.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, and FIGS. 2 and 3 each show a nonlinear element 18A. Figures 4 to 6 are waveform diagrams showing limit cycles, Figure 7 is a diagram showing limit cycle generation points, and Figures 8 and 9 are diagrams showing the nonlinear elements. A waveform diagram for explaining the effect of providing hysteresis, FIG. 10 is a block diagram showing a specific example in which the controller is configured with a microcomputer, and FIG.
The figure and FIG. 12 are flowcharts showing an example of a processing program in CPH. 1...process, 10...-controller, 11
...Proportional calculation section, 12..... Integral calculation section, 13.
... Differential operation section, 14 ... Output section, 15 ...
Input unit, 16...Observer, 17...State switching control unit, 18A, 18B...Nonlinear element, 19.
...Adjuster, sw1 to SW5...φ switch.

Claims (1)

[Claims] It has a normal operation mode and a tuning mode.In the tuning mode, the characteristics of the process are obtained from observing the limit cycles that occur in the process, and the optimal PID parameters are determined based on the results.After returning to the normal operation mode, is a self-tuning controller that performs PID control using updated PID parameters. In tuning mode, the proportional calculation section and the integral calculation section are connected in parallel, and a nonlinear element is inserted before the proportional calculation section, and process control is performed. A control method for a self-tuning controller, characterized in that the limit cycle is observed while continuing, and the amplitude of the limit cycle is determined from a value obtained by integrating deviation values sampled at a constant period.