JP7388849B2 - Control parameter adjustment device and method - Google Patents

Description

The present invention relates to a technique for adjusting parameters of a controller such as a PID, and particularly to a control parameter adjusting device and method for adjusting control parameters of a controller using a limit cycle method.

Adjustment of PID parameters of a PID controller is normally easily performed using an auto-tuning (AT) function provided in the controller itself. A typical method for this AT function is a limit cycle method in which the manipulated variable MV is preset at two positions (OH_AT, OL_AT) and a limit cycle is thereby generated (see Patent Document 1).

The limit cycle method will be briefly explained below. FIG. 15 is a waveform diagram for explaining the conventional limit cycle method, and FIG. 16 is a flowchart for explaining the process flow of the conventional limit cycle method. However, to simplify the explanation, the setting value SP used when executing PID control is used as the limit cycle switching point, but it is not essential that the limit cycle switching point is the setting value SP.

When AT is executed, the controlled variable PV and the set value SP are compared (step S300 in FIG. 16), and if the controlled variable PV is larger than the set value SP, the lower limit value OL_AT of the manipulated variable MV is output to the controlled object (step S300 in FIG. 16). S301), when the controlled variable PV is less than or equal to the set value SP, the upper limit value OH_AT of the manipulated variable MV is output to the controlled object (step S302 in FIG. 16).

Next, vertical movement extreme value detection processing is performed to detect the extreme value of the control amount PV (step S303 in FIG. 16). The processes of steps S300 to S303 are performed for each control period, and when four extreme values of the control amount PV are detected, the detection of vertical extreme values is completed. Here, the deviation Er between the set value SP and the control amount PV is obtained by the following equation.
Er=SP-PV...(1)

As shown in FIG. 15, among the four extreme values detected, the deviation at the latest extreme value is the first extreme value deviation Er1, the deviation at the second newest extreme value is the second extreme value deviation Er2, The deviation at the third newest extreme value is defined as a third extreme value deviation Er3. Further, the time from time t5 when the polarity of the deviation Er is reversed immediately before the first extreme value deviation Er1 to time t6 when the first extreme value deviation Er1 is obtained is defined as the first manipulated variable switching elapsed time Th1, Further, the time from time t3 when the polarity of the deviation Er is reversed immediately before the second extreme value deviation Er2 to time t4 when the second extreme value deviation Er2 is obtained is defined as the second manipulated variable switching elapsed time Th2. .

Finally, a PID parameter consisting of the proportional band Pb, integral time Ti, and differential time Td is calculated as shown in the following equation, and the calculated PID parameter is set in the controller (step S304 in FIG. 16).
Pb=250.0 |Er2-Er1|/(OH_AT-OL_AT)...(2)
Ti=6.0(Th1+Th2)...(3)
Td=1.2(Th1+Th2)...(4)
With this, AT ends.

As described above, there is a controller that executes AT using two-position operation amounts (OH_AT, OL_AT) that are set internally in advance. Many commercially available controllers use the manipulated variable upper limit value OH and manipulated variable lower limit value OL used during actual control as two-position manipulated variables. Normally, OH=100% and OL=0%, so the two-position operation amounts during AT are often OH_AT=100% and OL_AT=0%.

In a controlled object with high heat retention in temperature control, the manipulated variable MV required to maintain the controlled variable PV near the set value SP is quite low, approximately MV=20% or less. In this situation, if you perform AT from MV = 0% to 100% with the setting of OH_AT = 100%, the temperature will rise quickly and fall slowly (high heat retention and difficult to cool down), so you will spend a lot of time on AT. The problem arises that it is necessary (FIG. 17). In the example shown in FIG. 17, since the controlled object has high heat retention and does not easily cool down, the time T_OL during which the manipulated variable lower limit value OL_AT is output is longer than the time T_OH during which the manipulated variable upper limit value OH_AT is output. I understand that.

Regarding this kind of problem, it is not limited to the case where the operation amount upper limit value OH and operation amount lower limit value OL used during actual control are used as the two-position operation amount during AT, but also when the two-position operation set internally in advance If the quantities (OH_AT, OL_AT) correspond to those that give an inappropriate waveform as in FIG. 17, a similar problem will occur. The way to deal with this problem is for an expert who can determine that AT is inappropriate to use a recorder or the like to observe the movement of the control amount PV during AT, and to determine the operation amount upper limit OH and operation amount lower limit OL. The only option was to change the value to an appropriate value and rerun AT. Therefore, there is a possibility that the inappropriateness of AT may be overlooked in many cases.

Therefore, if the balance of the 2-position manipulated variables (OH_AT, OL_AT) during AT using the limit cycle method is inappropriate, the 2-position manipulated variables (OH_AT, OL_AT) that will achieve an appropriate balance are automatically calculated, and the appropriate balance is determined. An AT method has been proposed in which AT is automatically re-executed using two-position balance operation amounts (OH_AT, OL_AT) (see Patent Document 2).

The main points of the method disclosed in Patent Document 2 will be explained below. If the balance between the two-position operation amounts (OH_AT, OL_AT) during AT using the limit cycle method is inappropriate, the AT waveform will be as shown in FIG. 17.

For example, in temperature control, if the temperature rises quickly and the temperature drops slowly, if the manipulated variable MV=OL_AT and the manipulated variable MV=OH_AT is output while the temperature is decreasing, the temperature will start to rise at a steep slope, so the temperature PV will be set. The time period during which the value falls below the value SP is short. That is, since the manipulated variable MV=OL_AT is switched when the temperature PV exceeds the set value SP, the time during which the manipulated variable MV=OH_AT is output becomes relatively short. Further, as the temperature rises at a steep slope, the temperature PV greatly exceeds the set value SP. That is, the absolute value of the amount by which the temperature PV exceeds the set value SP (deviation Er1) becomes relatively large.

On the other hand, since the temperature decreases at a gentle slope, it takes time for the temperature PV to return to the set value SP, and the time for the temperature PV to exceed the set value SP becomes longer. While the temperature PV exceeds the set value SP, the manipulated variable MV=OL_AT is maintained, and the time during which the manipulated variable MV=OL_AT is output is relatively long. Furthermore, since the temperature decreases at a gentle slope, the absolute value of the amount by which the temperature PV falls below the set value SP (deviation Er2) becomes relatively small.

In the above example, the balance is inappropriate because OH_AT of the two-position operation amounts (OH_AT, OL_AT) is relatively too high compared to OL_AT. Therefore, if the ratio of absolute values of deviations RE=|Er2|/|Er1| is less than or equal to a specific threshold value, it can be determined that the manipulated variable OH_AT is relatively too high. Conversely, if the ratio RE=|Er2|/|Er1| is equal to or higher than another specific threshold value, it can be determined that the manipulated variable OL_AT is relatively too low. If the extreme value deviation Er1 is the local maximum value and the extreme value deviation Er2 is the local minimum value, RE=|Er2|/|Er1|, but if the extreme value deviation Er1 is the local minimum value and the extreme value deviation Er2 is the local maximum value, then RE=|Er2|/|Er1| In the case of a value, RE=|Er1|/|Er2|. It goes without saying that equivalent determination is possible even when RE=|Er1|/|Er2|.

It is possible to correct the 2-position manipulated variables (OH_AT, OL_AT) by assuming a linear relationship by the unbalanced ratio RE, and a more appropriate 2-position manipulated variable (OH_AT, OL_AT) can be calculated using the following formula. It can be calculated by
OL_AT_new=OL_AT+(OH_AT-OL_AT)(RE-1)
/(RE+1) ...(5)
OH_AT_new=OH_AT+(OH_AT-OL_AT)(RE-1)
/(RE+1) ...(6)

In the method disclosed in Patent Document 2, for example, when RE>1.0, the manipulated variable lower limit value OL_AT is corrected using equation (5), but the manipulated variable upper limit value OH_AT is not corrected. Conversely, when RE<1.0, the manipulated variable upper limit value OH_AT is modified according to equation (6), but the manipulated variable lower limit value OL_AT is not modified. When RE=1.0, neither the manipulated variable lower limit value OL_AT nor the manipulated variable upper limit value OH_AT is corrected.

As described above, according to the method disclosed in Patent Document 2, the two-position operation amounts (OH_AT, OL_AT) that provide an appropriate balance are automatically calculated, and the two-position operation amounts (OH_AT, OL_AT) that provide an appropriate balance are automatically calculated. ) can be automatically re-executed, but since the AT itself is essentially redone, there is a problem that the AT execution time (time required to complete) becomes longer, and improvements are needed. .

Japanese Patent Application Publication No. 2003-330504 Patent No. 4546437

The present invention was made to solve the above problems, and an object of the present invention is to provide a control parameter adjustment device and method that can shorten the AT execution time of the limit cycle method.

The present invention is a control parameter adjusting device equipped with a limit cycle auto-tuning function for setting control parameters of a controller, which alternately outputs a preset upper limit manipulated variable and a lower limit manipulated variable to a controlled object. A limit cycle auto-tuning calculation unit configured as shown in FIG. a ratio calculation unit configured to calculate a ratio of absolute values; a correction coefficient setting unit configured to set a correction coefficient for a correction amount of the upper limit value or the lower limit value; and a correction coefficient setting unit configured to calculate the ratio and the correction coefficient. a correction unit configured to calculate a value estimated to be appropriate for the upper limit value or the lower limit value based on the above, and correct the upper limit value or the lower limit value to the calculated value , The correction coefficient is set to a value smaller than 1, and the limit cycle autotuning calculation unit adjusts the control parameters of the controller based on the response of the control amount after the upper limit value or the lower limit value is corrected. The method is characterized in that the calculated control parameters are calculated and set in the controller .

Further, the present invention provides a control parameter adjustment method for setting control parameters of a controller using a limit cycle method, in which a first control method that alternately outputs a preset upper limit manipulated variable and a lower limit manipulated variable to a controlled object is provided. step, and a second step that calculates the ratio of the absolute value of the deviation at each of the first two extreme values of the controlled variable caused by the output of the manipulated variable among the deviations between the control set value and the controlled variable. step, the ratio , and a correction coefficient for the amount of correction of the upper limit value or the lower limit value, calculate a value that is estimated to be appropriate for the upper limit value or the lower limit value, and a third step of correcting the value to the calculated value; calculating a control parameter of the controller based on the response of the control amount after the upper limit value or the lower limit value is corrected; and a fourth step of setting the correction coefficient in the controller, and the correction coefficient is set to a value smaller than 1 .

According to the present invention, by calculating the ratio of the absolute values of the two extreme value deviations detected in the first vertical movement of the limit cycle and correcting the two-position manipulated variable based on this ratio, the corrected Since an appropriate vertical movement of the controlled variable can be quickly obtained with the two-position manipulated variable, the AT execution time can be shortened.

Further, in the present invention, by correcting the two-position manipulated variable based on the ratio of the absolute values of two extreme value deviations and the correction coefficient, it is possible to correct the correction error of the two-position manipulated variable, and Even when the order of is high, it is possible to prevent the accuracy of AT from decreasing.

FIG. 1 is a waveform diagram showing an example of changes in the manipulated variable and the controlled variable during AT execution. FIG. 2 is a block diagram showing the configuration of a control parameter adjustment device according to a reference example of the present invention. FIG. 3 is a block diagram showing the configuration of a controller to be adjusted by a control parameter adjustment device according to a reference example of the present invention. FIG. 4 is a flowchart illustrating the operation of the control parameter adjustment device according to the reference example of the present invention when performing AT. FIG. 5 is a waveform diagram showing an example of changes in the manipulated variable and the controlled variable during conventional AT execution. FIG. 6 is a waveform diagram illustrating an example of changes in the operation amount and control amount during conventional AT execution. FIG. 7 is a waveform diagram showing an example of changes in the manipulated variable and the controlled variable during AT execution according to the reference example of the present invention. FIG. 8 is a waveform diagram showing an example of changes in the manipulated variable and the controlled variable during AT execution according to the reference example of the present invention. FIG. 9 is a block diagram showing the configuration of a control parameter adjustment device according to an embodiment of the present invention . FIG. 10 is a flowchart illustrating the operation of the control parameter adjustment device according to the embodiment of the present invention when performing AT. FIG. 11 is a diagram illustrating a correction error of the two-position manipulated variable that occurs in the case of a high-order controlled object. FIG. 12 is a waveform diagram showing an example of changes in the manipulated variable and the controlled variable during AT execution according to the embodiment of the present invention. FIG. 13 is a waveform diagram showing an example of changes in the manipulated variable and the controlled variable during AT execution according to the embodiment of the present invention. FIG. 14 is a block diagram showing an example of the configuration of a computer that implements the control parameter adjustment device according to the reference example and the embodiment of the present invention. FIG. 15 is a waveform diagram for explaining the conventional limit cycle method. FIG. 16 is a flowchart illustrating the process flow of the conventional limit cycle method. FIG. 17 is a waveform diagram for explaining the problems of the conventional limit cycle method.

[Principle of the invention]
The reason behind the need for careful processing of re-executing the limit cycle is the recognition that the control amount PV needs to move up and down several times before the limit cycle becomes stable. The initial vertical movement of the control amount PV may be a larger vertical movement as shown in FIG. 1, for example, or a smaller vertical movement, depending on the conditions at the start of AT. In particular, if the balance between the two-position manipulated variables (OH_AT, OL_AT) is inappropriate, the first vertical movement will almost certainly be an invalid vertical movement. That is, the first extreme value (maximum value, minimum value) of the control amount PV at the start of AT has low reliability.

The inventor has focused on the fact that the first extreme value of a limit cycle has low reliability, but the extreme value ratio RE (ratio RE between maximum value and minimum value) appears with high reliability. Then, the inventor can correct the manipulated variable upper limit value OH_AT or the manipulated variable lower limit value OL_AT using the extreme value ratio RE obtained at an early stage (for example, the first vertical movement) at the start of AT. We have come up with the idea that the AT execution time can be shortened because appropriate vertical movement can be quickly obtained with the manipulated variables (OH_AT, OL_AT).

[ Reference example ]
Reference examples of the present invention will be described below with reference to the drawings. FIG. 2 is a block diagram showing the configuration of a control parameter adjustment device according to a reference example of the present invention. In this reference example , the denominator of the extreme value ratio RE is the absolute value of the deviation on the maximum side of the controlled variable PV (the extreme value caused by temperature rise), and the numerator of the extreme value ratio RE is the absolute value of the deviation on the local maximum side of the controlled variable PV (the extreme value caused by temperature decrease). Adopt a formula that is the absolute value of the deviation of the value).

The control parameter adjustment device includes a set value SP acquisition unit 1 that acquires a switching point SP (set value SP) for AT execution, a control amount PV acquisition unit 2 that acquires a control amount PV during AT execution, A manipulated variable MV output section 3 that outputs the manipulated variable MV for , a two-position manipulated variable lower setting section 4 for setting the manipulated variable lower limit value OL_AT of the limit cycle AT, and a manipulated variable upper limit value of the limit cycle AT. The two-position manipulated variable upper setting section 5 for setting OH_AT alternately outputs the manipulated variable MV of the manipulated variable upper limit value OH_AT and the manipulated variable MV of the manipulated variable lower limit value OL_AT to set the manipulated variable upper limit value OH_AT or the manipulated variable upper limit value OH_AT. A limit cycle AT calculation unit 6 that calculates the PID parameter of the controller based on the response of the controlled variable PV after the quantity lower limit value OL_AT has been corrected, and a limit cycle AT calculation unit that inputs a start instruction command for the limit cycle AT from the outside. Among the deviations between the control set value SP and the controlled variable PV, the absolute value of the deviation at each of the first two extreme values of the controlled variable PV caused by the output of the manipulated variable MV is A ratio RE calculation unit 8 calculates the ratio RE, and calculates a value estimated to be appropriate for the operation amount upper limit OH_AT or the operation amount lower limit OL_AT based on the ratio RE, and calculates the operation amount upper limit OH_AT or the operation amount lower limit. A two-position operation amount correction unit 9 is provided for correcting OL_AT to the calculated value.

FIG. 3 is a block diagram showing the configuration of a controller to be adjusted by the control parameter adjustment device of FIG. 2. As shown in FIG. The controller uses a set value SP input section 20, a controlled variable PV input section 21, and a manipulated variable MV output section 22 to control the deviation between the set value SP and the controlled variable PV during normal control operation after AT. It also includes a PID control calculation section 23 that calculates the manipulated variable MV by performing PID control calculation based on the parameters.

FIG. 4 is a flowchart illustrating the operation of the control parameter adjustment device when performing AT. In the 2-position manipulated variable lower setting section 4, an initial value of the manipulated variable lower limit value OL_AT of the limit cycle AT is set in advance, and in the 2-position manipulated variable upper setting section 5, the manipulated variable upper limit value OH_AT of the limit cycle AT is preset. An initial value is set in advance.
When the start instruction section 7 receives a start instruction command for the limit cycle AT from, for example, an operator, it sends the start instruction command to the limit cycle AT calculation section 6 . As a result, the limit cycle AT calculation unit 6 is activated and the limit cycle AT is started (step S100 in FIG. 4).

The activated limit cycle AT calculation unit 6 compares the control amount PV and the set value SP (step S101 in FIG. 4). The set value SP is set by an operator and inputted to the limit cycle AT calculation unit 6 via the set value SP acquisition unit 1. The control amount PV is detected by a sensor (not shown), and is input to the limit cycle AT calculation section 6 via the control amount PV acquisition section 2.

When the controlled variable PV is larger than the set value SP, the limit cycle AT calculation section 6 outputs the manipulated variable lower limit value OL_AT set by the 2-position manipulated variable lower setting section 4 to the manipulated variable MV output section 3 as the manipulated variable MV. (Step S102 in FIG. 4). In addition, when the controlled variable PV is less than or equal to the set value SP, the limit cycle AT calculation section 6 outputs the manipulated variable upper limit value OH_AT set by the 2-position manipulated variable upper setting section 5 as the manipulated variable MV to the manipulated variable MV output section 3. Output (step S103 in FIG. 4). The manipulated variable MV output section 3 outputs the manipulated variable MV output from the limit cycle AT calculation section 6 to the controlled object.

Next, the ratio RE calculation unit 8 detects two extreme value deviations of the control amount PV obtained from the first vertical movement of the control amount PV according to the output of the manipulated variable MV during AT execution (step S104 in FIG. 4). ). The extreme value deviation Er, which is the deviation between the set value SP and the extreme value of the control amount PV, is obtained by equation (1). In this reference example , of the two extreme value deviations, the deviation at the newer extreme value is defined as the first extreme value deviation Er1x, and the deviation at the previous extreme value is defined as the second extreme value deviation Er2x.

The processing of steps S101 to S103 is repeated every control cycle, and when the extreme value deviations Er1x and Er2x can be detected (YES in step S104), the process proceeds to step S105.

Then, the ratio RE calculation unit 8 calculates the ratio RE between the absolute value of the first extreme value deviation Er1x and the absolute value of the second extreme value deviation Er2x (step S105 in FIG. 4). At this time, if the newer extreme value of the first two extreme values of the controlled variable PV is the local maximum value and the previous extreme value is the local minimum value, the ratio RE calculation unit 8 calculates the following equation. The ratio RE is calculated such that the first extreme value deviation Er1x on the local maximum side becomes the denominator and the second extreme value deviation Er2x on the local minimum value side becomes the numerator.
RE=|Er2x|/|Er1x| ...(7)

In addition, when the newer extreme value of the first two extreme values of the control amount PV is the local minimum value and the previous extreme value is the maximum value, the ratio RE calculation unit 8 calculates The ratio RE is calculated such that the first extreme value deviation Er1x on the minimum value side becomes the numerator, and the second extreme value deviation Er2x on the local maximum side becomes the denominator.
RE=|Er1x|/|Er2x| ...(8)

Next, the 2-position operation amount correction unit 9 determines whether the settings of the 2-position operation amounts (OH_AT, OL_AT) are appropriate (step S106 in FIG. 4). If the ratio RE is larger than 1 (YES in step S107 in FIG. 4), the two-position manipulated variable correction unit 9 determines that the manipulated variable lower limit value OL_AT is inappropriate, and sets a modified value OL_AT_new of the manipulated variable lower limit value as shown in the following formula. is calculated, and this correction value OL_AT_new is output to the two-position manipulated variable lower side setting section 4. The 2-position manipulated variable lower setting section 4 sets the correction value OL_AT_new as a new value of the manipulated variable lower limit value OL_AT in the limit cycle AT calculation section 6 (step S108 in FIG. 4).
OL_AT_new=OL_AT+(OH_AT-OL_AT)(RE-1)
/(RE+1) ...(9)

Further, when the ratio RE is smaller than 1 (YES in step S109 in FIG. 4), the two-position manipulated variable correction unit 9 determines that the manipulated variable upper limit value OH_AT is inappropriate, and corrects the manipulated variable upper limit value as shown in the following equation. A value OH_AT_new is calculated, and this corrected value OH_AT_new is output to the two-position manipulated variable upper setting section 5. The two-position manipulated variable upper setting section 5 sets the correction value OH_AT_new as a new value of the manipulated variable upper limit value OH_AT in the limit cycle AT calculation section 6 (step S110 in FIG. 4).
OH_AT_new=OH_AT+(OH_AT-OL_AT)(RE-1)
/(RE+1) ...(10)

When the ratio RE is 1, the 2-position manipulated variable correction unit 9 determines that the setting of the 2-position manipulated variables (OH_AT, OL_AT) is appropriate, and does not modify the 2-position manipulated variables (OH_AT, OL_AT).
The processing of steps S105 to S110 is executed only once when the extreme value deviations Er1x and Er2x can be detected.

Even after the two-position manipulated variables (OH_AT, OL_AT) are corrected, the processes of steps S101 to S103 are repeatedly executed every control cycle, and AT continues.
Next, the limit cycle AT calculation unit 6 performs a vertical movement extreme value detection process to detect three extreme values of the control amount PV after correcting the two-position manipulated variables (OH_AT, OL_AT) (step S111 in FIG. 4). When three extreme values of the control amount PV are detected, the detection of the vertical extreme values is completed. As described above, the limit cycle AT calculation unit 6 calculates the deviation at the latest extreme value as extreme value deviation Er1, the deviation at the second latest extreme value as extreme value deviation Er2, and the deviation at the third latest extreme value as extreme value deviation Er1. Let the deviation be Er3. In addition, the limit cycle AT calculation unit 6 sets the time from the time when the polarity of the deviation Er is reversed immediately before the extreme value deviation Er1 to the time when the extreme value deviation Er1 is obtained as the manipulated variable switching elapsed time Th1, and sets the extreme value deviation The time from the time when the polarity of the deviation Er is reversed immediately before Er2 to the time when the extreme value deviation Er2 is obtained is defined as the manipulated variable switching elapsed time Th2.

Then, the limit cycle AT calculation unit 6 calculates PID parameters consisting of the proportional band Pb, integral time Ti, and differential time Td using equations (2) to (4), and sets the calculated PID parameters in the controller (Fig. 4 step S112).

The processing of the limit cycle AT calculation unit 6 in steps S101 to S103, S111, and S112 is disclosed in, for example, Patent Document 1 and Patent Document 2. The limit cycle AT ends upon completion of the calculation and setting of the PID parameters.

In the normal control operation after the AT is completed, the PID control calculation section 23 of the controller receives the set value SP input from the set value SP input section 20 and the control amount PV input from the control amount PV input section 21. For example, a PID control calculation such as the following transfer function formula is performed to calculate the manipulated variable MV for each control cycle.
MV=(100/Pb) {1+(1/Tis)+Tds}(SP-PV)
...(11)
In equation (11), s is a Laplace operator. The manipulated variable MV calculated by the PID control calculation unit 23 is output to the controlled object via the manipulated variable MV output unit 22.

FIG. 5 is a waveform diagram showing an example of changes in the manipulated variable MV and the controlled variable PV during conventional AT execution, and FIG. 6 is an enlarged view of FIG. 5. In the examples of FIGS. 5 and 6, the time 20 sec. AT is started at the same time as the first setting value SP is changed, and at time 520 sec. AT ends at the point in time. Then, at a time of 800 seconds after the end of AT. At the point in time, the set value SP is changed again, and the temperature increase control is performed by the controller.

FIG. 7 is a waveform diagram showing an example of changes in the manipulated variable MV and the controlled variable PV during AT execution in this reference example , and FIG. 8 is an enlarged view of FIG. 7. In the examples of FIGS. 7 and 8, the time 20 sec. AT is started at the same time as the first setting value SP is changed, and at time 470 sec. AT ends at the point in time. Then, at a time of 800 seconds after the end of AT. At the point in time, the set value SP is changed again, and the temperature increase control is performed by the controller.

In the examples shown in FIGS. 7 and 8, the extreme value deviations Er1x and Er2x are detected in the first vertical movement of the limit cycle, and the ratio RE is calculated, so that the manipulated variable upper limit value OH_AT increases from the initial 100% to 16.9 Corrected to %. As a result, compared to the cases shown in FIGS. 5 and 6, the vertical movement time of the controlled variable PV after the two-position manipulated variable (OH_AT, OL_AT) is corrected is shortened, and the AT execution time (time required to complete) is shortened. has been done.

As described above, in this reference example , the ratio RE is calculated from the extreme value deviations Er1x and Er2x detected in the first vertical movement of the limit cycle, and the two-position operation amounts (OH_AT, OL_AT) are calculated based on this ratio RE. By correcting , it becomes possible to quickly obtain an appropriate vertical movement of the control amount PV with the corrected two-position operation amounts (OH_AT, OL_AT), so the AT execution time can be shortened.

[ Example ]
Next, examples of the present invention will be described. FIG. 9 is a block diagram showing the configuration of a control parameter adjustment device according to an embodiment of the present invention, and the same components as in FIG. 2 are denoted by the same reference numerals. The control parameter adjustment device of this embodiment includes a set value SP acquisition section 1, a controlled variable PV acquisition section 2, a manipulated variable MV output section 3, a 2-position manipulated variable lower side setting section 4, and a 2-position manipulated variable upper side. The setting section 5, the limit cycle AT calculation section 6, the start instruction section 7, the ratio RE calculation section 8, and the correction coefficient setting section 10 that sets the correction coefficient α of the correction amount of the two-position operation amount (OH_AT, OL_AT). The operation amount upper limit value OH_AT or the operation amount lower limit value OL_AT is calculated by calculating a value that is estimated to be appropriate for the operation amount upper limit value OH_AT or the operation amount lower limit value OL_AT based on the ratio RE and the correction coefficient α. A correction unit 11 with two-position operation amount correction is provided.

The controller to be adjusted by the control parameter adjustment device is as described in the reference example .
FIG. 10 is a flowchart illustrating the operation of the control parameter adjustment device of this embodiment when performing AT.
Set value SP acquisition section 1, control amount PV acquisition section 2, manipulated variable MV output section 3, 2-position manipulated variable lower side setting section 4, 2-position manipulated variable upper side setting section 5, limit cycle AT calculation section 6, and start instruction section 7 and the operations of the ratio RE calculation unit 8 (steps S100 to S105 in FIG. 10) are the same as in the reference example .

A correction coefficient α for appropriately correcting the correction amount of the two-position operation amounts (OH_AT, OL_AT) is set in the correction coefficient setting unit 10 in advance. As a result of intensive research by the inventor, it was found that when modifying the two-position manipulated variables (OH_AT, OL_AT) using the ratio RE, the amount of modification becomes large for higher-order control targets, and errors occur in the modification. There was found.

The vertical movement of the controlled variable PV relative to the manipulated variable MV output from the control parameter adjustment device or the controller is such that the higher the order of the controlled object, the more the behavior near the extreme value of the vertical movement becomes as shown at 110 in FIG. The linear behavior changes to curved behavior as shown at 111 and 112. When the curve-like behavior near the extreme value of the controlled variable PV becomes large, the absolute values of the deviations on the maximum side and minimum side become smaller by different magnifications, as shown by ΔEr2x and ΔEr1x in Fig. 11, so the ratio RE becomes biased. This affects the side where the ratio RE is large (that is, the side where the ratio RE deviates from 1.0). Therefore, the amount of correction of the two-position operation amounts (OH_AT, OL_AT) becomes large.

Therefore, in this embodiment, a value smaller than 1.0 is preset as the correction coefficient α. The 2-position operation amount correction unit 11 determines whether the settings of the 2-position operation amounts (OH_AT, OL_AT) are appropriate, as in the reference example (step S106 in FIG. 10).

If the ratio RE is larger than 1 (YES in step S107 in FIG. 10), the two-position manipulated variable correction unit 11 determines that the manipulated variable lower limit value OL_AT is inappropriate, and corrects the manipulated variable lower limit value as shown in the following equation. A value OL_AT_new is calculated, and this corrected value OL_AT_new is output to the two-position manipulated variable lower side setting section 4. The 2-position manipulated variable lower setting unit 4 sets the correction value OL_AT_new as a new value of the manipulated variable lower limit value OL_AT in the limit cycle AT calculation unit 6 (step S108a in FIG. 10).
OL_AT_new=OL_AT+α(OH_AT-OL_AT)(RE-1)
/(RE+1) ...(12)

Further, when the ratio RE is smaller than 1 (YES in step S109 in FIG. 10), the two-position operation amount correction correction unit 11 determines that the operation amount upper limit value OH_AT is inappropriate, and calculates the operation amount upper limit value as shown in the following equation. A correction value OH_AT_new is calculated, and this correction value OH_AT_new is output to the two-position manipulated variable upper setting section 5. The two-position manipulated variable upper setting section 5 sets the correction value OH_AT_new as a new value of the manipulated variable upper limit value OH_AT in the limit cycle AT calculation section 6 (step S110a in FIG. 10).
OH_AT_new=OH_AT+α(OH_AT-OL_AT)(RE-1)
/(RE+1) ...(13)

The operation of the limit cycle AT calculation unit 6 after the correction of the two-position operation amounts (OH_AT, OL_AT) (steps S111 and S112 in FIG. 10) is the same as in the reference example . The limit cycle AT ends upon completion of the calculation and setting of the PID parameters.

FIG. 12 is a waveform diagram showing an example of changes in the manipulated variable MV and the controlled variable PV during AT execution in this embodiment, and FIG. 13 is an enlarged view of FIG. 12. In the examples of FIGS. 12 and 13, the time 20 sec. AT is started at the same time as the first setting value SP is changed, and at time 464 sec. AT ends at the point in time. Then, at a time of 800 seconds after the end of AT. At the point in time, the set value SP is changed again, and the temperature increase control is performed by the controller.

In this embodiment, by employing the correction coefficient α=0.95, the manipulated variable upper limit value OH_AT is corrected from the initial 100% to 21.0%. As a result, the behavior of the control amount PV is improved to a well-balanced vertical movement (vertical movement around time 400 sec.) compared to the reference example (FIGS. 7 and 8).
In this manner, in this embodiment, it is possible to correct correction errors in the two-position manipulated variables (OH_AT, OL_AT) that occur in the case of a high-order controlled object.

The control parameter adjustment device described in the reference examples and examples can be realized by a computer including a CPU (Central Processing Unit), a storage device, and an interface, and a program that controls these hardware resources. An example of the configuration of this computer is shown in FIG.

The computer includes a CPU 200, a storage device 201, and an interface device (hereinafter abbreviated as I/F) 202. For example, a temperature sensor, a controller, etc. are connected to the I/F 202. In such a computer, a program for realizing the control parameter adjustment method of the present invention is stored in the storage device 201. The CPU 200 executes the processes described in the reference example and the example according to the program stored in the storage device 201.

The present invention can be applied to a technique for adjusting control parameters of a controller using a limit cycle method.

DESCRIPTION OF SYMBOLS 1... Set value SP acquisition part, 2... Controlled amount PV acquisition part, 3, 22... Manipulated amount MV output part, 4... 2-position manipulated variable lower side setting part, 5... 2-position manipulated variable upper side setting part, 6... Limit Cycle AT calculation unit, 7... Start instruction unit, 8... Ratio RE calculation unit, 9... 2 position operation amount correction unit, 10... Correction coefficient setting unit, 11... Correction unit with 2 position operation amount correction, 20... Set value SP Input section, 21... Controlled amount PV input section, 23... PID control calculation section.

Claims (2)

In a control parameter adjustment device equipped with a limit cycle auto-tuning function that sets the control parameters of the controller,
a limit cycle autotuning calculation unit configured to alternately output a preset upper limit manipulated variable and a lower limit manipulated variable to a controlled object;
A ratio configured to calculate the ratio of the absolute value of the deviation between the control set value and the controlled variable at each of the first two extreme values of the controlled variable caused by the output of the manipulated variable. A calculation section,
a correction coefficient setting unit configured to set a correction coefficient for a correction amount of the upper limit value or the lower limit value;
A correction configured to calculate a value estimated to be appropriate for the upper limit value or the lower limit value based on the ratio and the correction coefficient, and correct the upper limit value or the lower limit value to the calculated value. Equipped with a
The correction coefficient is set to a value smaller than 1,
The limit cycle autotuning calculation unit calculates a control parameter of the controller based on a response of the control amount after the upper limit value or the lower limit value is corrected, and sets the calculated control parameter in the controller. A control parameter adjustment device characterized by: In the control parameter adjustment method that sets the control parameters of the controller using the limit cycle method,
a first step of alternately outputting a preset upper limit manipulated variable and a lower limit manipulated variable to a controlled object;
a second step of calculating the ratio of the absolute value of the deviation between the control set value and the controlled variable at each of the first two extreme values of the controlled variable caused by the output of the manipulated variable;
Calculate the upper limit value or the lower limit value by calculating a value that is estimated to be appropriate for the upper limit value or the lower limit value based on the ratio and a correction coefficient for the amount of correction of the upper limit value or the lower limit value. The third step is to correct the value to
a fourth step of calculating a control parameter of the controller based on the response of the control amount after the upper limit value or the lower limit value is corrected, and setting the calculated control parameter in the controller ,
A method for adjusting control parameters , wherein the correction coefficient is set to a value smaller than 1 .

Images