JPH10131902A - Cylinder positioning control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent shock at the time of braking by imputting the empirical value as the initial value of back pressure deceleration stroke and time lag, outputting a back pressure command signal when piston displacement satisfies the condition of a position where back pressure acts, and outputting a braking signal when the condition of subsequent elapsed time and time lag is satisfied. SOLUTION: The empirical value close to the convergent value with a target position and load taken into account is used as the initial value of back pressure deceleration stroke Spb and time lag Tv. A piston 11 is advanced by a solenoid valve A 'on' signal, and piston displacement X is detected by a position detector 26 and differentiated to input speed V to a computer 28. An output position Xp(k) of a back pressure command signal is determined, and when the piston displacement X satisfies the condition X>=Xp(k) of a position where back pressure acts, a back pressure command signal to switch a solenoid valve B on is outputted. When elapsed time T and time lag Tv after the output of the back pressure command signal satisfy the condition of T>=Tv, a braking signal is outputted. Piston motion is therefore stabilized, and shock associated with stop can be prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a cylinder positioning control device for stopping a repeatedly operating cylinder at a target position.

[0002]

2. Description of the Related Art Conventionally, a brake mechanism for braking a piston rod and a position sensor for detecting the position of the piston rod are provided, and the cylinder is positioned by learning control that captures not only the overrun distance but also the cylinder speed. 2. Description of the Related Art A cylinder positioning control device that controls movement of a piston rod to stop a cylinder at a target position is known.
(See JP-A-8-210302). When performing positioning using a conventional cylinder positioning control device, when the piston approaches the target position, the brake mechanism is operated to perform direct braking without deceleration due to back pressure (direct braking method). Had been stopped. In this direct braking system, the piston is decelerated and stopped by sliding friction between the brake piece and the piston rod. Therefore, the braking force is large, and a large impact is generated in the case of a high speed and large load. This impact not only degrades the operating stability of the piston, but also shortens the life of the brake piece and the piston rod and causes loud noise.

[0003]

SUMMARY OF THE INVENTION The present invention relates to a cylinder positioning control device, in which when a piston is stopped, a back pressure is applied to decelerate the piston, and thereafter, braking is performed. It is an object to prevent occurrence.

[0004]

According to a first aspect of the present invention, a cylinder positioning control device includes the following means A to F to achieve the above object. A. Means for inputting an empirical value as an initial value of the back pressure deceleration stroke Spb and the time delay Tv. B. Means for outputting a piston forward or backward command signal. C. Means for inputting displacement X and velocity V of the piston. D. Formula [Xp (k) = Target position Xd-V (Tvr + Ttr) -S
means for determining the position Xp (k) at which the back pressure command signal is to be output by using the predicted value of pb (k)]. Here, Tvr is the response time of the solenoid valve, and Ttr is the pressure propagation time from the solenoid valve to the cylinder. E. FIG. Condition X for position where piston displacement X exerts back pressure
Means for judging whether or not ≧ Xp (k) is satisfied, and outputting a back pressure command signal when the above condition is satisfied. F. Elapsed time T and time delay Tv after output of back pressure command signal
Means for judging whether a condition of T ≧ Tv is satisfied, and outputting a braking signal when the condition is satisfied. According to a second aspect of the present invention, in the cylinder positioning control device, the following means A to K are provided to achieve the above object. A. Means for outputting a piston forward or backward command signal. B. Means for inputting displacement X and velocity V of the piston. C. Means for determining the position Xp (k) at which the back pressure command signal is issued. D. Condition X for position where piston displacement X exerts back pressure
Means for judging whether or not ≧ Xp (k) is satisfied, and outputting a back pressure command signal when the above condition is satisfied. E. FIG. Speed Vp (k) detected when condition X ≧ Xp (k) is satisfied
Means to enter. F. Elapsed time T and time delay Tv after output of back pressure command signal
Means for judging whether a condition of T ≧ Tv is satisfied, and outputting a braking signal when the condition is satisfied. G. FIG. Means for inputting the maximum displacement Xm and the stop position Xs of the piston after the piston stops. H. It is determined whether or not the cyclic operation has ended after the return of the piston. If the cyclic operation has not ended, the braking error Se is determined.
Means to determine. I. Means for determining an accurate value ΔTv (k) by fuzzy inference. J. Using ΔTv (k) obtained by means I, Tv (k + 1) = T
Means for correcting the next time delay Tv by v (k) + ΔTv (k). K. Means for calculating the predicted value of the next back pressure deceleration stroke Spb. The present invention relates to the cylinder positioning control device,
The braking error Se is determined using the following equations (a) and (b), and the following equation (c) is determined.
The position Xp (k) for issuing the back pressure command signal is determined using FIG.
Determining the accurate value ΔTv (k) using the equation (6) shown in the following is a third configuration. Se = Xv-XmXs = XmSe = Xm-XsXs <Xm (a) Xv = Xp + Vp (Tvr + Ttr) + Vp (Tv-Tvr-Ttr + Tr) / 2 (b) Xp (k) = Target position Xd-V ( Tvr + Ttr) -Predicted value of Spb (k) (c) where Tr is a dead time of braking, and Tvr, Ttr
And Tbr, Tvr is the response time of the solenoid valve, Ttr is the pressure propagation time from the solenoid valve to the cylinder, Tbr is the dead time of the braking device, and in equation (6), L (*) and F
(*) Are the base length and the abscissa of the center of gravity of the triangle in the output area of ΔTv, respectively.

[0005]

FIG. 1 shows an embodiment in which the present invention is applied to an air cylinder 10. As shown in FIG. In FIG. 1, a braking device (braking device) 13 is provided at the tip of an air cylinder 10,
A piston 11 is slidably fitted to the air cylinder 10. A piston rod 12 connected to the piston 11 is inserted into a braking device 13, and an inertial load 14 is connected to a tip of the piston rod 12. Inside the air cylinder 10, a head 11 and a back pressure chamber (rod side chamber) 17 are moved by a piston 11.
The head side chamber 16 is connected to the port A of the solenoid valve A by a pipe 50, and the pipe 50 is provided with a first speed controller 23. Similarly, the back pressure chamber 17 is connected to the port A of the solenoid valve B by a pipe 51, and the pipe 51 is provided with a second speed controller 24. P of solenoid valve A
The port is connected to the outlet port of the drive pressure reducing valve 19, and the P port of the solenoid valve B and the inlet port of the drive pressure reducing valve 19 are connected to the air pressure source 22.

The brake device 13 has an operating side chamber (cylinder chamber) and a non-operating side chamber (not shown). The operating side chamber is connected to the A port of the electromagnetic valve C for brake via a pipe 52, and the non-operating side chamber is It is connected to the B port of the solenoid valve C via a pipe 53. The P port of the solenoid valve C is connected to the outflow port of the brake pressure reducing valve 20, and the inflow port of the brake pressure reducing valve 20 is connected to the air pressure source 22. The position of the piston 11 is continuously detected by the position detector 26, and the position detector
The output of 26 is input to the signal processing means 27 through the wiring, and the output of the signal processing means 27 is input to the computer 28. The output of the computer 28 is amplified by the amplifier 29 and transmitted to the solenoid valves A to C.

In the position shown in FIG. 1, the working side chamber of the braking device 13 is exhausted to the atmosphere through the A port and the R port of the solenoid valve C, and the non-operating side chamber of the braking device 13 has compressed air from the air pressure source 22. Flows through the brake pressure reducing valve 20 and the P port / B port of the solenoid valve C. At this time, since the brake is in the non-braking state and the head side chamber 16 and the back pressure chamber 17 of the air cylinder 10 are communicated with the atmosphere, the piston 11 is stopped, and the force acting on both sides of the piston 11 is approximately Atmospheric pressure.

The operation of the air cylinder 10 will be described with reference to FIGS. When a switch control signal (electromagnetic valve ON signal) is sent to the operation unit of the electromagnetic valve A, the electromagnetic valve A is switched from the position I to the position II, and the compressed air depressurized by the drive pressure reducing valve 19 is compressed. The air flows into the head side chamber 16 of the air cylinder 10 through the P port / A port and the pipe 50. The air in the back pressure chamber 17 of the air cylinder 10 is discharged to the atmosphere through the pipe 51, the speed controller 14, and the ports A and R of the solenoid valve B, and the piston 11 advances. The displacement X of the piston 11 is detected by the position detector 26, and a piston displacement Xp for outputting a back pressure command signal is set as a control signal of the solenoid valve B as described later. Therefore, when the computer 28 determines that X≥Xp, an ON signal is sent from the computer 28 to the operation section of the solenoid valve B, and the solenoid valve B is switched from the position I to the position II. By switching the solenoid valve B, the compressed air from the air pressure source 22 that has not been depressurized flows into the back pressure chamber 17 of the air cylinder 10 through the P port / A port of the solenoid valve B and the pipe 51. At this time, since the solenoid valve A maintains the position II, the compressed air flowing into the back pressure chamber 17 acts as a back pressure on the piston 11, and the forward movement of the piston 11 is reduced.

When the speed of the piston 11 is reduced to substantially zero after a lapse of time Tv from the operation of the solenoid valve B, an ON signal is sent from the computer 28 to the operation section of the solenoid valve C,
The solenoid valve C is switched from position I to position II. When the solenoid valve C is switched to the position II, the compressed air decompressed by the pressure reducing valve 20 flows into the working side chamber of the braking device 13 through the P port / A port of the solenoid valve C and the pipe 52, and at an extremely low speed. The moving piston rod is tightened by the brake piece, and stable braking is realized. In the present invention, the piston displacement X
p is corrected at any time by a self-learning method so that the positioning error of the piston 11 converges. In addition, time delay T
v is corrected at any time by fuzzy inference so that the brake is applied when the speed of the piston 11 is near zero.

FIG. 3 shows direct braking without the application of a back pressure and back pressure deceleration braking in which the vehicle is decelerated and then braked by the action of the back pressure. As mentioned above, piston 11
When the solenoid valve B is switched from the position I to the position II after the stable forward movement is started, the back pressure chamber 17 changes from the exhaust state to the air supply state, the force balance of the piston 11 is lost,
The speed of 11 drops to almost zero. As can be seen from FIG.
Compared to the direct braking process, the back pressure deceleration braking process has smoother changes in speed and acceleration, the back pressure deceleration braking greatly improves the motion stability of the piston 11, and the impact between the brake piece and the piston rod surface is reduced. Have been avoided.

The characteristics of the back pressure deceleration process required for realizing positioning by using back pressure deceleration braking will be described. As parameters representing characteristics, the stroke (distance) of the back pressure deceleration process is set to Spb, and the time is set to Tpr. And
In the process from the time when the back pressure starts to rise to the time when the piston 11 decelerates to zero speed, Spb is the distance traveled by the piston 11, and Tpr is the time of this process. Spb
And Tpr are mainly affected by inertial load M and stable speed V
∞ and the back pressure start position Xvs (the position of the piston 11 when the back pressure is applied) can be considered. Xvs is the head side chamber 16 and the back pressure chamber 17 of the air cylinder 10 when the back pressure is applied.
Represents the volume relationship. Spb and Tpr are the above 3
As a function of the two influencing factors, it can be expressed as: Spb = F _S (M, V∞, Xvs) (1) Tpr = F _T (M, V∞, Xvs) (2) Since these two functions are complicated and difficult to grasp, we use fuzzy inference and self-learning. The only way to control this unknown process is leverage.

Fuzzy braking using fuzzy inference will be described. If the actual deceleration time Tpr can be accurately predicted from the consideration of the back pressure deceleration process, the ideal value of the time delay Tv can be obtained as Tv = Tpr-Tbr (3). Here, Tbr is a dead time (constant value) of the braking device. For the sake of simplicity, the response times of the solenoid valves B and C are equally set to Tvr, the pressure propagation time in the pipe 51 from the solenoid valve B to the back pressure chamber 17 of the air cylinder 10, and the braking time from the solenoid valve C The pressure propagation time in the pipe 52 to the working side chamber (cylinder chamber) of the device 13 is equal,
And Since Tpr is unknown, its value is obtained here by fuzzy inference. Fuzzy inference constructs the relationship between input and output by establishing fuzzy inference rules using empirical values for controlled objects whose relationship between input and output is inaccurate and unknown. It is an intelligent control method.

The result of deceleration braking when the value of the non-ideal time delay Tv is considered will be examined, and then input / output variables for fuzzy inference will be established according to the control target. If Tv is smaller than the ideal value, that is, if the predicted value of Tv is small, braking occurs during deceleration, and a constant braking distance occurs. Conversely, when Tv is larger than the ideal value, that is, when the predicted value of Tv is large, the braking is later than the stop of the piston 11, so that
Not only is it disadvantageous for maintaining the position, but also, as the target position is closer to the stroke end of the piston 11, the bouncing phenomenon of the piston 11 due to the instantaneous rise of the back pressure is more likely to occur. FIG. 4 shows a conceptual diagram of these two types of braking results. Here, the braking point is represented by Xv, the maximum displacement point in the deceleration braking process is represented by Xm, and the stop point is represented by Xs. In case 2,
After reaching the maximum displacement point Xm, it returned and stopped.

Thus, the input variables of the fuzzy inference are determined as Se = Xv-XmXs = XmSe = Xm-XsXs <Xm (4) The purpose of fuzzy inference is to converge the braking error Se. Xv in equation (4) is obtained as follows by detecting the velocity Vp at the point of the piston displacement Xp. Xv = Xp + Vp (Tvr + Ttr) + Vp (Tv-Tvr-Ttr + Tr) / 2 (5) where Tr is a dead time of braking, and Tvr, Ttr
And Tbr.

The time Tpr of the back pressure deceleration process is a target value of fuzzy inference. Since ΔTv is equal to ΔTpr, the output variable of the fuzzy inference is determined as ΔTv. Considering the relationship between Se and ΔTv, in the k-th operation, if Se is negative, it means that Tv is relatively small.
In the +1 operation, Tv must be increased. That is, ΔTv must be positive. The opposite is also true. Fuzzy variables and fuzzy rules are defined as follows. NB: Negative R1 = If Se is NB, ΔTv is PB. NS: negative small R2 = If Se is NS, .DELTA.Tv is PS. ZR: zero R3 = If Se is ZR, then ΔTv is ZR. PS: small or small R4 = If Se is PS, ΔTv is NS. PB: positive R5 = If Se is PB, ΔTv is NB.

As shown in FIG. 5, the membership function of each fuzzy variable employs a triangular waveform function.
Fuzzy rules and membership functions are determined by expert experience. This is also a feature of fuzzy control. The expert experience used in the present invention was obtained by repeatedly correcting parameters from control effects based on preliminary experimental data. The fuzzy inference in the present invention is determined by the mass centroid method. This method belongs to a kind of continuous fuzzy control, and the specific operation is as follows. That is, for the Se obtained after the k-th operation, the fitness f (*) of one or two fuzzy variables is determined by the membership function of the Se input area.
(That is, the vertical axis coordinate, the range of values is 0 to 1), and the degree of conformity is reflected in the output area by the fuzzy rule to obtain one or two shadow triangles having the same height (see FIG. 5). The area of the shadow triangle is the so-called “mass”. By obtaining the abscissa of the total center of gravity of these shadow triangles, an accurate value ΔTv of the output variable of the fuzzy inference is obtained. Further, using this value, Tv is corrected and the (k + 1) th operation is performed. This method is easy to calculate and the input / output relationship is clear. The calculation formula is as shown in formula (6) in FIG. Where L
(*) And F (*) are the base length and the abscissa of the center of gravity of the triangle of the output area of ΔTv, respectively. Next, one example is shown. When Se = 1.0, f (ZR) is about 0.33 and f (PS) is 0.67 due to the membership function. The center of gravity of the corresponding shadow triangle is obtained by equation (6-1) in FIG.

The learning positioning by the learning control will be described. To consider learning positioning of the back pressure deceleration process,
The back pressure deceleration process is divided into two stages, a back pressure deceleration reaction stage and a back pressure deceleration stage. The strokes of these two stages of the piston are referred to as Spa and Spb, respectively. The displacement overrun amount in the back pressure deceleration process is as follows: Sp = Spa + Spb (7) Spa is obtained by the following equation (8). Spa = (Tvr + Ttr) .Vp (8) Since only Spb is unknown in the equation of Sp, Spb is a target object of the back pressure deceleration learning control. The learning controller is designed as shown in equations (8) to (12). Equations (11) and (1
2) is shown as (11) and (12) in FIG. S′pb (k) = Xm (k) −Xp (k) −Spa (k) (9) Spb (k) = S′pb (k) −λ · SeSe <0 Spb (k) = S′pb (k) Se ≧ 0 (10)

In order to consider the effect of braking during deceleration on Spb prediction, a compensation factor λ is introduced to compensate for a shortened portion of the stroke due to braking. In general, a large value of λ is selected so as not to affect the convergence speed of the positioning error. In particular, when the initial value of Spb is far from the actual value, λ is usually set to 2 to 5. Compared to direct braking positioning, back pressure deceleration braking positioning has a slower convergence, and usually requires 3-5 learnings before convergence. In FIG.
The convergence process by fuzzy braking and learning positioning in the periodic operation of a single target point is shown. From FIG. 7, it can be seen that the braking error Se and the positioning error e converge after several learnings and are kept constant. This indicates that it is possible and effective to perform positioning control by back pressure deceleration braking using fuzzy inference and self-learning.

Next, control of the cylinder positioning control device will be described with reference to the flowchart of FIG. When the control program starts, it is determined in step S1 whether the piston 11 of the air cylinder 10 has been stopped for a long time. When the piston 11 is stopped for a long time, the braking device 13 is used to reduce the static friction force and stabilize the response time Tvr of the solenoid valves B and C and the dead time Tbr of the braking device 13.
Is actuated twice beforehand.
Prior measurement values are used for Tvr and Tbr and the pressure propagation time Ttr from the solenoid valves B and C to the cylinder. In step S1, when it is determined that the piston 11 has not stopped for a long time, the process proceeds to step S3, and when it is determined that the piston 11 has stopped for a long time, the on / off operation of the brake device 13 is performed twice in step S2, Thereafter, the process proceeds to step S3.

In step S3, the back pressure deceleration stroke S
Determine the initial values of pb and time delay Tv. Spb and Tv
As an initial value of に, an empirical value close to the convergence value in consideration of the target position and load is used. In step S4, the solenoid valve A
Output an ON signal to move the piston forward. When the solenoid valve B is on in the previous return state, the solenoid valve B is turned off. In the state shown in FIG. 1, the solenoid valve A is turned on and switched to the position II, and the solenoid valves B and C are kept in the off state and at the position I, so that the piston 11 starts to move forward. In step S5, from the input signal from the position detector 26, 0.
The piston displacement X is detected at a sampling period of 5 ms,
This is differentiated to obtain the velocity V, which is input to the computer 28.

In step S6, the position Xp (k) for outputting the back pressure command signal is determined. Under the condition that the last term on the right side of the equation (13) in FIG. 8 is known and the velocity V is continuously detected, the equation (13) [Xp in FIG. (k) =
Target position Xd-V (Tvr + Ttr) -Predicted value of Spb (k)]
Is required. Since the equation (14) in FIG. 8 represents the stroke of the piston 11 from when the solenoid valve B is turned on until the piston 11 stops, the last term (predicted value) on the right side of the equation (13) in FIG. If the value is close to the true value, the piston 11 can be stopped at the target position Xd by turning on the solenoid valve B at the position Xp (k). In step S7, it is determined whether or not the displacement X of the piston satisfies the condition X ≧ Xp (k) of the position for applying the back pressure. When this condition is not satisfied, the process returns to step S5, and when it is satisfied, the process proceeds to step S8. In step S8, the back pressure command signal (the solenoid valve B
ON) is output. At this time, the solenoid valve B is turned on to switch from the position I to the position II, the compressed air flows into the back pressure chamber 17 of the air cylinder 10, and the forward movement of the piston 11 is reduced by the action of the back pressure.

In step S9, the condition X ≧ Xp (k)
Is satisfied, the speed V detected is Vp (k), which is read and input into the computer. Vp (k) is changed to Spb (k +
It is used for the prediction of 1) (see equations (8) and (9)). In step S10, it is determined whether or not the elapsed time T and the time delay Tv after the output of the back pressure command signal satisfy the condition of T ≧ Tv. When this condition is not satisfied, the process returns to step S10, and when it is satisfied (when the speed of the piston 11 is reduced to near zero), the process proceeds to step S11. In step S11, a braking signal is output, and the solenoid valve C is moved to the position I.
To the position II, the compressed air flows into the working side chamber of the braking device 13, and the piston rod is clamped by the brake piece.

In step S12, it is determined whether or not the piston has stopped and whether or not the displacement sampling values are equal within one second. If not stopped (the sampling values are not equal), the process returns to step S12,
If the operation has stopped (the sampling values are equal), the process proceeds to step S13. In step S13, the maximum value Xm of the sampling value of the displacement and the stop position Xs of the piston
Is detected and read into a computer. Using these values and equation (4), the braking error Se is calculated.
In step S14, a return signal for the piston is issued.
The return signal is a signal for turning off the solenoid valves A and C, and this signal causes both the solenoid valves A and C to move from the position II to the position I.
, The solenoid valve B maintains the position II (ON), and the piston 11 retreats.

In step S15, it is determined whether or not the piston 11 has returned, that is, whether or not the piston displacement X = 0. When this condition is not satisfied, the process returns to step S15, and when this condition is satisfied, the process proceeds to step S16.
Move to In step S16, it is determined whether or not the cyclic operation has ended, that is, whether or not the stop position of the piston 11 has been the target position. If the periodic operation has been completed, the process proceeds to the end. If the periodic operation has not been completed, the process proceeds to step S17. In the present invention, the periodic operation is a prerequisite for fuzzy inference and self-learning control, and reaches the target value through several periodic operations. In step S17, the expression
The braking error Se is determined by (4) and (5). This is an important point in the fuzzy reasoning of the present invention. Since Se is greatly affected by the delay time Tv, if Tv is not ideal, it immediately appears in Se. Therefore, the purpose of modifying the value of Tv is to make Se converge to zero. Converging Se to zero guarantees braking stability. When Xm = Xs, Xv calculated by equation (5) is not an actual braking point, so
In this case, the influence on Se by the inappropriate value of Tv is further increased.

In step S18, a precise value ΔTv is determined by fuzzy inference. That is, ΔTv is determined from the value of Se using Expression (6) in FIG. The fact that Se does not become zero means that the value of Tv is inappropriate. That is, the predicted value of Tpr is inaccurate. The purpose of fuzzy inference is to make Se converge to zero,
This is to obtain the true value of Tpr. In step S19, the time delay of the next k + 1 is corrected by the equation [Tv (k + 1) = Tv (k) + ΔTv (k)] using the ΔTv obtained from the previous k operation. In step S20, equations (9), (10), (11)
Then, the next process of back pressure deceleration Spb is calculated.

[0026]

According to the first aspect of the present invention, the piston is advanced, the back pressure is applied to the piston at the optimum position to reduce the speed, and the brake is stopped when the piston speed approaches zero. Let it. Therefore, the movement of the piston was extremely stable, and the impact accompanying the stop could be completely prevented, and stable braking of the piston could be realized. According to the second and third aspects of the present invention, the time delay Tv for determining the output timing of the braking signal is corrected as needed by fuzzy inference, and the piston is decelerated to nearly zero speed before braking. At the same time, the piston displacement Xp for outputting the back pressure command signal is corrected as needed by a self-learning method, and the positioning error of the piston is gradually reduced to a minimum. Thus, highly accurate positioning was realized.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a positioning system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a relationship between piston displacement and operation of a solenoid valve according to the embodiment of the present invention.

FIG. 3 is a comparison diagram of motion characteristics between direct braking and back pressure deceleration braking according to the embodiment of the present invention.

FIG. 4 is a diagram illustrating two types of braking effects when a non-ideal Tv value is used in the embodiment of the present invention.

FIG. 5 shows a membership function of input and output variables of fuzzy inference.

FIG. 6 shows a flowchart of the system according to the embodiment of the present invention.

FIG. 7 is a diagram showing a change in an error due to repetition of learning in the embodiment of the present invention.

FIG. 8 shows a part of an equation according to the embodiment of the present invention.

[Explanation of symbols]

 11 piston

Claims (3)

[Claims] 1. A cylinder positioning control device comprising the following means A to F: A. Means for inputting an empirical value as an initial value of the back pressure deceleration stroke Spb and the time delay Tv. B. Means for outputting a piston forward or backward command signal. C. Means for inputting displacement X and velocity V of the piston. D. Formula [Xp (k) = Target position Xd-V (Tvr + Ttr) -S
means for determining the position Xp (k) at which the back pressure command signal is to be output by using the predicted value of pb (k)]. Here, Tvr is the response time of the solenoid valve, and Ttr is the pressure propagation time from the solenoid valve to the cylinder. E. FIG. Condition X for position where piston displacement X exerts back pressure
Means for judging whether or not ≧ Xp (k) is satisfied, and outputting a back pressure command signal when the above condition is satisfied. F. Elapsed time T and time delay Tv after output of back pressure command signal
Means for judging whether a condition of T ≧ Tv is satisfied, and outputting a braking signal when the condition is satisfied. 2. A cylinder positioning control device comprising the following means A to K: A. Means for outputting a piston forward or backward command signal. B. Means for inputting displacement X and velocity V of the piston. C. Means for determining the position Xp (k) at which the back pressure command signal is issued. D. Condition X for position where piston displacement X exerts back pressure
Means for judging whether or not ≧ Xp (k) is satisfied, and outputting a back pressure command signal when the above condition is satisfied. E. FIG. Speed Vp (k) detected when condition X ≧ Xp (k) is satisfied
Means to enter. F. Elapsed time T and time delay Tv after output of back pressure command signal
Means for judging whether a condition of T ≧ Tv is satisfied, and outputting a braking signal when the condition is satisfied. G. FIG. Means for inputting the maximum displacement Xm and the stop position Xs of the piston after the piston stops. H. It is determined whether or not the cyclic operation has ended after the return of the piston. If the cyclic operation has not ended, the braking error Se is determined.
Means to determine. I. Means for determining an accurate value ΔTv (k) by fuzzy inference. J. Using ΔTv (k) obtained by means I, Tv (k + 1) = T
Means for correcting the next time delay Tv by v (k) + ΔTv (k). K. Means for calculating the predicted value of the next back pressure deceleration stroke Spb. 3. A braking error Se is determined by using the following equations (a) and (b), and a position Xp for issuing a back pressure command signal is determined by using the following equation (c).
(k) is determined, and the accurate value ΔTv is calculated using the equation (6) shown in FIG.
3. The cylinder positioning control device according to claim 2, wherein (k) is determined. Se = Xv-XmXs = XmSe = Xm-XsXs <Xm (a) Xv = Xp + Vp (Tvr + Ttr) + Vp (Tv-Tvr-Ttr + Tr) / 2 (b) Xp (k) = Target position Xd-V ( Tvr + Ttr) -Predicted value of Spb (k) (c) where Tr is a dead time of braking, and Tvr, Ttr
And Tbr, Tvr is the response time of the solenoid valve, Ttr is the pressure propagation time from the solenoid valve to the cylinder, Tbr is the dead time of the braking device, and in equation (6), L (*) and F
(*) Are the base length and the abscissa of the center of gravity of the triangle in the output area of ΔTv, respectively.