JP7079078B2 - Control device - Google Patents

Description

The present invention relates to a control device.

As a system that repeatedly inputs a signal of the target waveform to the control target and causes the control target to output a response waveform according to the target waveform, for example, a pressure for testing the fatigue durability of a test piece such as a pipe or a hose. There are testers and vibration testers.

A tester using fluid pressure is used for the pressure tester and vibration tester, and the control device that controls the tester has a target waveform in order to give pressure and vibration along the target waveform to the test piece. Is used as an input to control the servo valve of the testing machine.

Specifically, the control device is a vibration tester that generally feeds back the pressure in the pressure tester so that the response waveform output by the tester to the target waveform follows the target waveform. In this case, feedback control is performed to feed back displacement, velocity or acceleration (see, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 11-101708

However, in the conventional control device, when the target waveform contains a steep peak, the peak value of the response waveform matches the peak value of the target waveform even if the target waveform is input to the feedback loop and feedback control is executed. I can't let you.

There is always a response delay in the controlled object, and when the controlled object is a pressure tester, the pressure medium such as hydraulic oil and the test piece have elasticity, so the feedback control is executed as shown in FIG. Even so, the waveform of the peak part in the response waveform (broken line in the figure) becomes dull with respect to the target waveform (solid line in the figure). Therefore, even if the target waveform is input to the feedback loop and the feedback control is executed, the peak value of the response waveform does not match the peak value of the target waveform.

From a macroscopic point of view, if the feedback control is continued for a long time, the peak value of the response waveform approaches the target waveform. It is a problem that cannot be overlooked that the peak values of both do not match.

Therefore, an object of the present invention is to provide a control device capable of accurately matching the peak value of the response waveform to be controlled with the peak value of the target waveform including a steep peak.

In order to achieve the above object, the control device of the present invention is a control device that repeatedly outputs a response waveform to the control target based on the target waveform, and is predetermined among the target waveform and the response waveform output by the control target. A waveform correction unit that corrects the target waveform to obtain the corrected waveform based on the difference at a predetermined point, an error correction unit that corrects the corrected waveform to obtain the input waveform, and a control that controls the control target based on the input waveform. A predetermined point is set as a point where a peak in the target waveform is provided , and the waveform correction unit obtains a correction value based on the difference between the peak value at the point of the target waveform and the output value of the response waveform at the predetermined point. , The correction value is sequentially added to the value of the predetermined point of the target waveform to obtain the correction waveform, and the error correction unit corrects it based on the difference between the target waveform and the response waveform output by the control target during the previous control. Correct the waveform and obtain the input waveform to be input to the control unit at the next control . In the control device configured in this way, the corrected waveform becomes a waveform corrected so that the value at a predetermined point becomes large, so that it is caused by the control response delay of the controlled object and the elasticity of the pressure medium or the test piece. Even if the response waveform becomes dull at the peak portion, the value at the predetermined point in the response waveform can be accurately matched with the value at the predetermined point in the target waveform. Therefore, the control device configured in this way can accurately match the response waveform even in the peak portion where it is difficult to match the response waveform with the target waveform only by the feedback control.

Further, in the control device , the waveform correction unit obtains a correction value based on the difference between the peak value at the predetermined point of the target waveform and the output value of the response waveform at the predetermined point, and sets the correction value to the value of the predetermined point of the target waveform. Corrected waveforms are obtained by sequentially adding them. When the control device obtains the correction waveform in this way, the target waveform is corrected according to the difference, so that the response waveform is corrected to match the target waveform at a predetermined point each time the control device controls the control target. The response waveform matches the target waveform at a predetermined point quickly.

Further, the error correction unit is based on the difference between the target waveform and the response waveform output by the control target during the previous control, and the difference between the maximum peak value of the target waveform and the maximum peak value of the response waveform output by the control target during the previous control . Then, the corrected waveform may be corrected and the input waveform to be input to the control unit at the next control may be obtained . In the control device configured in this way, the modified waveform obtained by modifying the target waveform is corrected and used as the input waveform to the control unit. Therefore, the degree of coincidence of the response waveform with the target waveform at a predetermined point is increased, and the number of controls is increased. At the same time, error correction can be performed to bring the response waveform closer to the target waveform. Therefore, according to this control device, the time until the entire response waveform follows the target waveform and the response waveform follows the target waveform can be extremely short.

Further, the error correction unit is the initial peak difference, which is the difference between the maximum peak value of the target waveform and the maximum peak value of the response waveform output by the control target when the target waveform is input to the control unit, and is the maximum peak value of the target waveform. The value obtained by dividing the current peak difference, which is the difference between the maximum peak values of the response waveform output by the control target by inputting the input waveform to the control unit, is calculated as the peak value ratio, and the target waveform and the control target during the previous control are calculated. The correction value may be obtained by multiplying the difference from the response waveform output by the above by the peak value ratio, and the correction value may be sequentially added to the correction waveform to obtain the input waveform to be input to the control unit . According to the control device configured in this way, it is prevented that the response waveform greatly overshoots the target waveform when it approaches the target waveform, and the response waveform follows the target waveform well, and the response waveform becomes the target waveform. When they match or are sufficiently close to each other, the correction value becomes very small, so that the effect of automatically stopping the correction of the input waveform can be obtained.

According to the control device of the present invention, the peak value of the response waveform to be controlled can be accurately matched with the peak value of the target waveform including a steep peak.

It is a system block diagram of the cylinder control device in one Embodiment. It is a control block diagram of the control device in one Embodiment. It is a figure explaining an example of a target waveform. It is a figure which showed the target waveform and the response waveform at the time of the first control. It is a figure which showed the correction waveform at the time of the second control. It is a figure which showed the difference calculated by the target waveform, the response waveform, and the error correction part at the time of the first control. It is a figure which showed the input waveform and the response waveform at the time of the second control. It is a figure which showed the target waveform and the response waveform when only the conventional feedback control was executed.

Hereinafter, the present invention will be described based on the embodiments shown in the figure. As shown in FIG. 1, in this example, the control device 1 according to the embodiment has a pressure detection unit 2 for detecting the pressure P output by the pressure tester A and an input waveform in order to control the pressure tester A. The control unit 3 that controls the servo valve V of the pressure tester A according to the above procedure, the waveform correction unit 4 that corrects the target waveform to obtain the corrected waveform, and the error that corrects the corrected waveform and obtains the input waveform to be input to the control unit 3. It is configured to include a correction unit 5. The control device 1 controls the pressure tester A so as to output the pressure of the repeated response waveform based on the repeatedly input target waveform.

On the other hand, the pressure tester A includes a servo valve V that controls the output fluid pressure, and the pressure output by the pressure tester A is adjusted by the control of the servo valve V by the control device 1. Although not shown, the pressure tester A includes, for example, a pump and a booster cylinder that receives pressure fluid from the pump, and the fluid pressure output from the booster cylinder is adjusted by the servo valve V. Pressure is applied to the inside of the test piece T such as a hose. The servo valve V is a proportional solenoid valve driven by a solenoid, and adjusts the pressure applied to the test piece T according to the amount of current supplied from the control device 1.

In this example, the pressure detection unit 2 is installed in a pipeline H that guides the pressure output by the pressure tester A to the test body T, detects the pressure P output by the pressure tester A, and controls the control unit 3. , Is input to the waveform correction unit 4 and the error correction unit 5. The pressure detection unit 2 may be provided at a position where the pressure output by the pressure tester A can be detected.

The target waveform is a pressure command in which the pressure tester A indicates the pressure to be applied to the test body T in chronological order, and the control device 1 controls the pressure tester A with respect to one input of the target waveform. When the output of the response waveform is completed, it is input to the control device 1 again. Therefore, the control device 1 controls the pressure tester A, which is the control target, to output the pressure of the repeated response waveform with respect to the repeatedly input target waveform.

As shown in FIG. 2, in this example, the control unit 3 is based on an input waveform instructing a pressure (instructed pressure) P ^* to be output by the pressure tester A and a pressure P detected by the pressure detection unit 2. A control command I ^* indicating the amount of current to be given to the servo valve V is generated, and the current is supplied to the servo valve V.

In this example, the input waveform input to the control unit 3 is a waveform obtained by correcting the target waveform by the waveform correction unit 4 and further correcting the correction waveform by the error correction unit 5. However, at the first time when the control unit 3 first controls the servo valve V, the input waveform is a target waveform showing an ideal pressure waveform to be loaded on the test body T as shown in FIG. The target waveform is set in advance as a single pressure command for instructing the pressure to be applied to the test piece T in chronological order. The target waveform indicates one indicated pressure P ^* for each control cycle, and in this example, it is composed of 5000 target pressure data. Since the control cycle is 0.2 ms in this example, the target waveform is a data set of the target pressure for 1 s. As shown in FIG. 3, the target waveform of this example includes points b, c, point d, point e, point f, and point g, which are turning points between the start point a and the end point h. The waveform is provided, and the indicated pressure is set at each point b, c, d, e, f, g in addition to the points a and h where the indicated pressure is 0. The adjacent points have a waveform connected by a straight line. Therefore, if the indicated pressure is set for each point b, c, d, e, f, g, the data between the points can be complemented according to the above conditions to automatically create a data set of the target waveform. Further, the input waveform is also composed of 5000 target pressure data similar to the target waveform. The target waveform of this example is an example, and it is natural that a waveform other than the waveform shown in the figure may be used as the target waveform.

Specifically, as shown in FIG. 2, the control unit 3 outputs the pressure P ^** indicated by the input waveform output from the input waveform calculation unit 53, which will be described later, and the pressure tester A (pressure detection unit 2). The addition unit 31 that obtains the deviation E from the pressure P to be generated, the compensation unit 32 that obtains the control command I ^* by PID compensating the deviation E obtained by the addition unit 31, and the current to the servo valve V according to the control command I ^* . It is configured to include a driver 33 for supplying the above.

The addition unit 31 obtains a deviation E between the pressure P ^** indicated by the input waveform and the pressure P detected by the pressure detection unit 2, and inputs the deviation E to the compensation unit 32. In this example, the compensation unit 32 obtains a control command I ^* that indicates the current applied to the servo valve V by proportionally integrating and differentially compensating the deviation E. That is, in this example, the control unit 3 controls the servo valve V by pressure feedback. Although the compensation unit 32 is a PID compensator in this example, it may be a PI compensator for proportional integral compensation.

Then, the obtained control command I ^* is input to the driver 33, and the driver 33 supplies a current to the servo valve V according to the control command I ^* . The driver 33 is provided with, for example, an analog circuit using an operational amplifier or the like, or a switching element, and is a circuit capable of adjusting the amount of current supplied from the power supply to the solenoid of the servo valve V by turning the switching element on and off.

As shown in FIG. 2, the waveform correction unit 4 has a predetermined point of the target waveform and a predetermined point of the response waveform which is the waveform of the pressure P output by the pressure tester A detected by the pressure detection unit 2. The difference calculation unit 41 for obtaining the difference, the correction value calculation unit 42 for obtaining the correction value to be added to the target waveform based on the difference, and the correction value calculation unit 42 for adding the obtained correction value to the target waveform to obtain the correction waveform used for this control. It includes a correction waveform calculation unit 43 and a storage unit 44 that integrates and stores the obtained correction values.

The difference calculation unit 41 obtains the difference between the target waveform at a predetermined point and the response waveform which is the waveform of the pressure P output by the pressure tester A detected by the pressure detection unit 2. The difference calculation unit 41 obtains the difference between the value at the predetermined point in the target waveform and the value (output value) of the pressure P output by the pressure tester A corresponding to the predetermined point of the target waveform in the response waveform. .. In this example, the predetermined point is set by designating specific data from the target pressure data in the target waveform, and as shown in FIG. 3, the data having the maximum peak value in the target waveform. The point b corresponding to is set as a predetermined point. Therefore, the difference calculation unit 41 is the difference between the maximum peak value in the target waveform and the value (output value) of the pressure P output by the pressure tester A corresponding to the predetermined point of the target waveform in the response waveform. Find ΔP. In FIG. 4, the input waveform is set as the target waveform at the time of initial control, the target waveform shown by the solid line in the figure, and the pressure output by the pressure tester A to be controlled with respect to the input of the target waveform to the control unit 3. The response waveform (wavy line in the figure) which is the waveform of P is shown. The difference calculation unit 41 calculates the difference ΔP between the value of the indicated pressure P ^* of the maximum peak value of the target waveform, which is a predetermined point, and the value of the pressure P (output value) at the predetermined point in the response waveform output by the pressure tester A. Ask.

The correction value calculation unit 42 obtains a correction value by multiplying the difference ΔP calculated by the difference calculation unit 41 by a correction coefficient (gain). The storage unit 44 sequentially adds the correction values sequentially obtained by the correction value calculation unit 42 to obtain and store the integrated value of the correction values. That is, the storage unit 44 adds the correction value newly obtained by the correction value calculation unit 42 to the integration value stored by itself, and sequentially updates and stores the integration value. The correction waveform calculation unit 43 adds the integrated value stored in the storage unit 44 and the correction value previously obtained by the correction value calculation unit 42, adds the correction waveform to the input target waveform, and uses the correction waveform for this control. Ask for. Specifically, the corrected waveform calculation unit 43 corrects the value of the point b, which is a predetermined point in the target waveform, by adding the integrated value and the corrected value, and between the corrected point b and the point a. A data set of the corrected waveform is generated by complementing the waveform so that the corrected point b and the point c are connected by a straight line. In this example, since the target waveform is a waveform connecting each point with a straight line, the correction waveform calculation unit 43 complements the connection between the points a and b and the points b and c with a straight line. .. When the target waveform is a waveform connecting the points with a line other than a straight line, the correction waveform calculation unit 43 may complement the waveform according to the condition of the line connecting the points.

In this way, the control device 1 obtains a correction value every time the pressure tester A is controlled by inputting a target waveform, and the correction value calculated for each control is sequentially applied to the target waveform. It is added to obtain the corrected waveform. In other words, the modified waveform is obtained by integrating the modified values and adding them to the target waveform. Therefore, if the difference ΔP between the target waveform and the response waveform is not 0 at a predetermined point, the corrected waveform is basically corrected every time the control device 1 executes control. At the first time of control, the difference calculation unit 41 does not calculate the difference ΔP and outputs 0. Therefore, the correction value calculation unit 42 also obtains the correction value as 0, and the correction waveform calculation unit 43 is input. Add 0 to the target waveform. Therefore, at the time of initial control, the correction waveform calculation unit 43 outputs the target waveform as it is to the error correction unit 5, so that the waveform correction unit 4 does not correct the target waveform. At the time of the second control, the correction value calculation unit 42 multiplies the difference ΔP between the value of the predetermined point in the target waveform and the value corresponding to the predetermined point in the response waveform output by the pressure tester A at the time of the first control by the correction coefficient. To find the correction value. At the time of this second control, the correction waveform calculation unit 43 adds the correction value to the data value of the point b, which is a predetermined point of the target waveform, obtains the correction waveform, and outputs the correction waveform to the error correction unit 5. In this way, the waveform correction unit 4 obtains a correction value according to the difference ΔP each time the control progresses, and generates a correction waveform.

If the difference ΔP from the value of point b in the target waveform to the value corresponding to point b in the response waveform (output value) has a positive value, and the correction coefficient is _KS , then positive _KS · ΔP A correction value with a value is sought. Then, as shown in FIG. 5, the corrected waveform (solid line in the figure) generated by the waveform correction unit 4 has a larger value by KS · _ΔP at the point b than the target waveform (broken line in the figure). It becomes. If the difference ΔP between the value of point b in the target waveform and the value corresponding to point b in the response waveform (output value) continues to have a positive value, the corrected waveform has a value with respect to the target waveform at point b. When the response waveform of the pressure tester A matches the target waveform at the point b as a result of being controlled by the control unit 3 based on the correction waveform, the correction of the target waveform by the waveform correction unit 4 is stopped. The correction value calculation unit 42 corrects the target waveform in the waveform correction unit 4 by setting the correction value to 0 when the absolute value of the obtained correction value is sufficiently small and becomes equal to or less than the preset stop threshold value. You may stop it.

When the storage unit 44 does not store the correction value but stores the correction waveform obtained by the correction waveform calculation unit 43, the correction waveform calculation unit 43 stores the correction value in the correction waveform previously stored by the storage unit 44. It may be configured to add the correction value obtained by the calculation unit 42 to obtain a new correction waveform. Even in this case, the correction values sequentially obtained by the correction value calculation unit 42 are sequentially added to the target waveform, so that the correction values obtained are added to the correction waveform to update the correction waveform. And, it is equivalent to obtain the corrected waveform by sequentially adding the corrected value obtained each time to the target waveform.

The error correction unit 5 corrects the correction waveform and obtains an input waveform to be input to the control unit 3. As shown in FIG. 2, the error correction unit 5 includes a difference calculation unit 51 for obtaining a difference between the target waveform and the response waveform which is the waveform of the pressure P output by the pressure tester A detected by the pressure detection unit 2. A correction value calculation unit 52 that obtains a correction value to be added to the input waveform input to the control unit 3 this time based on the difference, and a correction value calculation unit 52 that adds the obtained correction value to the input waveform input this time to the control unit 3 next time. It includes an input waveform calculation unit 53 for obtaining an input waveform to be input, and a storage unit 54 for storing the obtained correction value.

The difference calculation unit 51 obtains the difference between the target waveform and the response waveform which is the waveform of the pressure P output by the pressure tester A detected by the pressure detection unit 2. In FIG. 6, the input waveform is set as the target waveform at the time of initial control, the target waveform shown by the solid line in the figure, and the pressure output by the pressure tester A to be controlled with respect to the input of the target waveform to the control unit 3. The response waveform (wavy line in the figure) which is the waveform of P is shown. Specifically, the difference is data in which the difference from the pressure P of the response waveform output by the pressure tester A with respect to the indicated pressure P ^** as one data indicated by the input waveform is chronologically arranged. Therefore, when the response waveform changes with respect to the target waveform as shown in FIG. 6, these differences become the waveform data as shown in the lower part in FIG. Further, in this example, the difference calculation unit 51 calculates the difference only in the correction target section set in the target waveform, as shown in FIG. In this example, the target waveform has two sections in which the indicated pressure P ^* instructed on the way is constant, and in this section, the indicated pressure P ^* instructed by the target waveform is input to the control unit 3. If pressure feedback is performed, the response waveform also follows the target waveform well, so there is no need to correct the corrected waveform. Therefore, in such an interval, it is not necessary to calculate the difference, calculate the correction value, and store the correction value. Therefore, in this example, the correction target section is set as the section of the target waveform other than the section where the indicated pressure P ^* does not change, that is, the section where the indicated pressure P ^* fluctuates. Specifically, as shown in FIG. 6, the correction target section is set in the range from the first data of the target waveform to the 1000th and the range from the 3000th to the 3400th. In this way, the correction target section is set by designating the number in the order of data, and since the target waveform is the data of the target pressure in the time series, the range may be set according to the time.

The correction value calculation unit 52 obtains a correction value by multiplying the difference calculated by the difference calculation unit 51, the peak value ratio, and the correction coefficient. The input waveform calculation unit 53 adds the correction value previously obtained by the correction value calculation unit 52 stored in the storage unit 54 to the correction value obtained by the correction value calculation unit 52, and adds the correction waveform to the input correction waveform. Next time, the input waveform to be input to the control unit 3 is obtained. The storage unit 54 adds the correction value newly obtained by the correction value calculation unit 52 to the correction value stored by itself, and sequentially updates and stores the correction value. In this way, the control device 1 obtains a correction value every time the pressure tester A is controlled by inputting an input waveform to the control unit 3, and the correction value is calculated every time the control is performed. Is sequentially added to the corrected waveform to obtain the input waveform. In other words, the input waveform is obtained by integrating the correction value and adding it to the correction waveform. Therefore, if the difference between the input waveform and the corrected waveform is not 0, the input waveform is basically corrected every time the control device 1 executes control. At the first time of control, the difference calculation unit 51 does not calculate the difference and outputs 0. Therefore, the correction value calculation unit 52 also obtains the correction value as 0. At the first time of control, since the waveform correction unit 4 also outputs the target waveform as described above, the input waveform calculation unit 53 adds 0 to the input target waveform. Therefore, at the time of initial control, the input waveform calculation unit 53 inputs the target waveform to the control unit 3 as it is, so that the input target waveform is output as it is without being corrected. In the second and subsequent controls, the waveform correction unit 4 corrects the target waveform and outputs the corrected waveform, and the correction value calculation unit 52 also obtains the correction value. Therefore, the error correction unit 5 adds the correction value to the correction waveform. The input waveform is generated and output to the control unit 3.

When the storage unit 54 does not store the correction value but stores the input waveform obtained by the input waveform calculation unit 53, the input waveform calculation unit 53 corrects the correction value to the input waveform previously stored by the storage unit 54. The correction value obtained by the calculation unit 52 may be added to obtain a new input waveform. Even in this case, the correction value obtained by the correction value calculation unit 52 is still added to the correction waveform, so the correction value obtained is added to the input waveform to update the input waveform. And, it is equivalent to obtain the input waveform by sequentially adding the correction value obtained each time to the target waveform.

Here, the calculation content of the correction value calculation unit 52 will be described in detail. The above-mentioned peak value ratio is the initial peak difference, which is the difference between the maximum peak value of the target waveform and the maximum peak value of the response waveform output by the pressure tester A at the first time of control, and is the maximum peak value of the target waveform and the pressure tester. A is the value obtained by dividing the current peak difference, which is the difference between the maximum peak values of the response waveforms output during the current control. The maximum peak value is the maximum value of the waveform, and as shown in FIG. 6, the maximum peak value P1 at the point b of the target waveform shown by the solid line in FIG. 6 and the response output by the pressure tester A for the first time of control. The difference in the maximum peak value P2 of the waveform is defined as the initial peak difference δini. At the time of the first control, the target waveform is not corrected and is used as the input waveform as it is. Therefore, the difference is obtained as shown in the waveform shown at the bottom of FIG. 6, and the correction value is obtained, and the next control is continued. Occasionally, a correction value is added to the correction waveform to obtain an input waveform. As shown in FIG. 7, when the input waveform is corrected as shown by the solid line in FIG. 7 and the response waveform changes as shown by the broken line in FIG. 7 with respect to the input waveform at the time of the second control, one point in FIG. The difference between the maximum peak value P1 of the target waveform shown by the chain line and the maximum peak value P3 of the response waveform output this time by the pressure tester A is the peak difference δnow this time. The correction value calculation unit 52 obtains the initial peak difference δini and the current peak difference δnow, and calculates the peak value ratio. At the time of the second control, the peak value ratio α is calculated by α = δnow / δini. At the time of the third control, the peak difference δnow is obtained this time as the difference between the maximum peak value P1 of the target waveform and the peak value of the response waveform, and the initial peak difference δini is constant during the control when it is obtained at the first control. Is the value of.

When the difference calculated by the difference calculation unit 51 is ε, the peak value ratio is α, and the correction coefficient is K _H , the correction value calculation unit 52 calculates the correction value = K _H · ε · α. Ask.

The control device 1 is configured in this way, and the target waveform is repeatedly input. The waveform correction unit 4 corrects the target waveform to generate the corrected waveform with respect to the input of the target waveform, and the error correction unit 5 corrects the correction waveform to generate the input waveform. The control unit 3 to which the input waveform obtained in this way is input performs pressure feedback control based on the indicated pressure P ^** indicated by the input waveform and the pressure P detected by the pressure detecting unit 2, and the pressure tester A outputs the pressure tester A. Control the pressure. The control device 1 controls the pressure tester A for one input of the target waveform as the first (first time) control, and as described above, the target waveform is directly input to the control unit 3 as an input waveform at the first time. Will be done.

At the time of the first control, there is a difference between the target waveform which is an input waveform and the pressure response waveform output by the pressure tester A due to the pressure feedback control by the control unit 3. At the time of the second control, the difference between the target waveform which is the first input waveform and the response waveform which is the actually detected pressure waveform is calculated for the input of the target waveform to the control device 1, and the correction value calculation is performed. The unit 42 and the correction value calculation unit 52 perform the above-mentioned calculation to obtain a correction value and a correction value, respectively.

Therefore, in the second control, the waveform correction unit 4 corrects the target waveform to generate a correction waveform, and the error correction unit 5 adds the correction value to the correction waveform, corrects the correction waveform, and generates an input waveform. .. Since the target waveform is the input waveform in the first control, the input waveform input in the second control is corrected by adding the correction value to the correction waveform after the target waveform is corrected by the correction value. Will be done. Since the input waveform and the response waveform obtained by the second control are corrected by the correction value and the correction value is added to the target waveform, the peak value becomes large at the point b as shown in FIG. At the same time, it becomes a waveform in which the amplitude increases up and down in the correction target section. In the second control, since the input waveform is corrected in this way, the response waveform also becomes a waveform closer to the target waveform than the response waveform at the time of the first control in which the target waveform is input to the control unit 3. After that, as the number of controls is repeated, the input waveform is sequentially corrected, and the response waveform approaches the target waveform.

Then, the target waveform is corrected based on the difference between the target waveform at the predetermined point and the response waveform which is the waveform of the pressure P output by the pressure tester A detected by the pressure detection unit 2, and the correction waveform is obtained. Generate. Since the control device 1 of this example obtains the corrected waveform in this way, the corrected waveform becomes a waveform corrected so that the value at a predetermined point becomes large, and the response delay in the control of the pressure tester A, the pressure medium, and the test Even if the response waveform is blunted at the peak portion due to the elasticity of the body T, the value at the predetermined point in the response waveform can be accurately matched with the value at the predetermined point in the target waveform. Therefore, according to the control device 1 of the present invention, it is possible to improve the followability of the response waveform to be controlled with respect to the target waveform including a steep peak.

Further, in the control device 1 of this example, a predetermined point is set as a point where a peak in the target waveform is provided, and the waveform correction unit 4 is based on the difference between the peak value at the point of the target waveform and the output value of the response waveform at the predetermined point. Since it is configured to obtain the corrected waveform, it is possible to accurately match the response waveform even in the peak portion where it is difficult to match the response waveform with the target waveform only by the feedback control.

In the control device 1 of this example, since the response waveform can be matched with the target waveform at a predetermined point, the predetermined point may be set to a value other than the peak of the target waveform. Further, if the predetermined points are set to a plurality of points of the target waveform, the control device 1 can match the response waveform with the target waveform at the points selected as the predetermined points.

Further, in the control device 1 of this example, the waveform correction unit 4 obtains a correction value based on the difference between the peak value at a predetermined point of the target waveform and the output value of the response waveform at the predetermined point, and determines the correction value of the target waveform. The corrected waveform is obtained by sequentially adding to the point values. When the control device 1 obtains the correction waveform in this way, the target waveform is corrected according to the difference. Therefore, every time the control device 1 controls the pressure tester A to be controlled, the response waveform becomes the target waveform at a predetermined point. It is corrected so that it matches the target waveform at a predetermined point of the response waveform, and the match with the target waveform is quick. In addition to the correction value, the correction of the target waveform in the waveform correction unit 4 is performed by determining the gain according to the difference, multiplying the value of the predetermined point in the target waveform by the gain, and correcting the value. A method of obtaining a waveform can also be adopted.

Further, in the control device 1 of this example, the control unit 3 that controls the pressure tester (control target) A based on the input waveform, the target waveform, and the response waveform output this time by the pressure tester (control target) A. The correction waveform is corrected and input to the control unit 3 based on the difference between It is provided with an error correction unit 5 for obtaining the above. In the control device 1 configured in this way, since the corrected waveform obtained by correcting the target waveform is corrected and used as the input waveform to be input to the control unit 3, control is performed while increasing the degree of matching of the response waveform with the target waveform at a predetermined point. Error correction that brings the response waveform closer to the target waveform with the number of times can be performed at the same time. Therefore, according to the control device 1, the time until the entire response waveform follows the target waveform and the response waveform follows the target waveform can be extremely short.

Further, in the control device 1 of this example, the maximum peak of the target waveform is the difference between the maximum peak value of the target waveform and the maximum peak value of the response waveform output by the pressure tester (controlled object) A for the first time. The peak value ratio is the value obtained by dividing the current peak difference, which is the difference between the value and the maximum peak value of the response waveform output by the pressure tester (control target) A, and the pressure tester (control) with respect to the target waveform and the target waveform. Target) The correction value is obtained by multiplying the difference from the response waveform output by A by the peak value ratio, and the correction value is sequentially added to the correction waveform to obtain the input waveform.

Here, if the correction value is obtained by multiplying the difference ε by only the correction coefficient _KH without multiplying the difference ε by the peak value ratio α, the input waveform will be controlled if the number of controls increases unless the difference ε becomes 0. The larger the number, the larger the amplitude. Therefore, even if an input waveform whose response waveform matches the target waveform is obtained in the control several times, it is corrected so that the amplitude of the input waveform becomes large in the subsequent control, and the response waveform is the target. It overshoots the waveform significantly. On the other hand, in this example, the error correction unit 5 obtains the correction value by multiplying the difference ε between the input waveform and the response waveform by the peak value ratio α. Since the peak value ratio α is the value obtained by dividing the current peak difference δnow by the initial peak difference δini, the value of the peak value ratio α becomes smaller as the peak value of the response waveform approaches the peak value of the target waveform. To go. Therefore, when the correction value is obtained in this way, when the response waveform approaches the target waveform, the degree of amplitude amplification of the next input waveform becomes smaller than that of the previous input waveform. As a result, when the response waveform approaches the target waveform, it is prevented from overshooting the target waveform significantly, and the target waveform is well followed. Further, since the correction value is obtained based on the difference ε between the input waveform and the response waveform whose amplitude is larger than the target waveform, not the difference between the target waveform and the response waveform, it is better to use the difference between the target waveform and the response waveform. However, the time required for the response waveform to follow the target waveform can be shortened. When obtaining the correction value, the difference ε is multiplied by the correction coefficient _KH . This correction coefficient _KH is an adjustment element for adjusting the amount of the response waveform approaching the target waveform by one control. The value can be set arbitrarily and can be omitted if unnecessary.

From the above, in the control device 1 of this example, when the response waveform approaches the target waveform, it is prevented from overshooting the target waveform significantly, and the target waveform is well followed, and the response waveform matches the target waveform. When it is done or is sufficiently close, the correction value becomes very small, so that the effect of automatically stopping the correction of the input waveform can be obtained.

When the correction value is obtained only by multiplying the difference ε by the correction coefficient _KH without using the peak value ratio α, a threshold value β is set for the difference between the peak value of the target waveform and the peak value of the response waveform. When the absolute value of the difference between the peak value of the target waveform and the peak value of the response waveform is equal to or less than the threshold value β, the input waveform may not be corrected or the correction value may be set to 0. The threshold value β sets a range in which the response waveform is recognized to sufficiently follow the target waveform, and can be set arbitrarily. In this example, since the correction value is obtained using the peak value ratio α, the correction value becomes very small when the response waveform matches or sufficiently approaches the target waveform. However, when repeated tests are performed on the order of millions of times (number of digits), even if the correction value becomes sufficiently small, the correction value is added and the input waveform may diverge and oscillate in control. be. Therefore, even when the correction value is obtained using the peak value ratio α, it is desirable to compare and judge the difference between the threshold value β, the peak value of the target waveform, and the peak value of the response waveform.

Further, since the control device 1 of this example sets the correction target section for the target waveform and does not perform the correction outside the correction target section, the storage capacity in the storage unit 54 for storing the correction value can be small. Therefore, the cost of the control device 1 can be reduced.

From the above, the control device 1 is optimal for controlling the pressure tester A in which the output of the same waveform is repeatedly required and the response waveform is difficult to match the target waveform due to the influence of the elasticity of the pressure medium and the test body T.

This concludes the description of the embodiments of the present invention, but the scope of the present invention is not limited to the details themselves shown or described.

1 ... Control device, 3 ... Control unit, 4 ... Waveform correction unit, 5 ... Error correction unit, A ... Pressure tester

Claims (3)

A control device that repeatedly outputs a response waveform to a controlled object based on a target waveform.
Based on the difference between the target waveform and the response waveform output by the control target at a predetermined predetermined point, the waveform correction unit that corrects the target waveform to obtain the correction waveform and the correction waveform are corrected. It is provided with an error correction unit for obtaining an input waveform and a control unit for controlling the control target based on the input waveform.
The predetermined point is a point at which the peak of the target waveform is provided.
The waveform correction unit obtains a correction value based on the difference between the peak value of the target waveform at the predetermined point and the output value of the response waveform at the predetermined point, and obtains the correction value at the predetermined point of the target waveform. The corrected waveform is obtained by sequentially adding to the value of
The error correction unit corrects the correction waveform based on the difference between the target waveform and the response waveform output by the control target at the time of the previous control, and obtains an input waveform to be input to the control unit at the next control.
A control device characterized by that. The error correction unit includes the difference between the target waveform and the response waveform output by the control target during the previous control, the maximum peak value of the target waveform, and the maximum peak value of the response waveform output by the control target during the previous control . The control device according to claim 1, wherein the corrected waveform is corrected and an input waveform to be input to the control unit at the next control is obtained based on the difference between the two. The error correction unit is
The initial peak difference, which is the difference between the maximum peak value of the target waveform and the maximum peak value of the response waveform output by the control target when the target waveform is input to the control unit, is the maximum peak value of the target waveform and the above. The peak value ratio is calculated by dividing the current peak difference, which is the difference between the maximum peak values of the response waveform output by the controlled object by the input of the input waveform to the control unit.
Further, the difference between the target waveform and the response waveform output by the control target at the time of the previous control is multiplied by the peak value ratio to obtain a correction value, and the correction value is sequentially added to the correction waveform and input to the control unit. The control device according to claim 1 , wherein the input waveform is obtained.

Images