CN114484040B - Valve control device and vacuum valve - Google Patents

Abstract

The invention provides a valve control device and a vacuum valve capable of shortening pressure regulating time. The valve control device (2) controls the opening degree of a vacuum valve mounted in a vacuum chamber by opening control, and the valve control device (2) comprises: a feedforward controller (220) that outputs an opening degree setting value (theta 1) for opening control; and a pressure regulating controller (21) for determining that the pressure balance value of the fixed value does not exceed the pressure target value (Ps) of the valve control in a predetermined period within a fixed period in which the opening degree set value (theta 1) becomes the fixed value. The pressure regulating controller (21) controls the opening based on the determination result when it is determined that the pressure balance value does not exceed the pressure target value (Ps).

Description

Valve control device and vacuum valve

Technical Field

The present invention relates to a valve control device and a vacuum valve.

Background

In a vacuum processing apparatus such as a chemical vapor deposition (Chemical Vapor Deposition, CVD) apparatus, a vacuum valve for adjusting a pressure is provided between a process chamber (process chamber) and a vacuum pump (for example, refer to patent document 1), and the opening of the vacuum valve is controlled to automatically adjust a chamber pressure to a predetermined pressure. The vacuum valve described in patent document 1 is roughly adjusted to a pressure near a target value by an open control (open control), and is switched to a close control (close control), and approaches the target value by fine adjustment.

[ Prior Art literature ]

[ patent literature ]

[ patent document 1] Japanese patent laid-open publication No. 2018-106718

Disclosure of Invention

[ problem to be solved by the invention ]

In the invention described in patent document 1, when the opening degree is in a fixed state during the opening control, it is determined whether the predicted pressure Pp exceeds the pressure target value Ps (Pp > Ps) or does not exceed the pressure target value Ps (Pp Ps). However, in the case of determining that the predicted pressure Pp does not exceed the pressure target value Ps, it is difficult to recognize that the predicted pressure Pp does not exceed the pressure target value Ps in the stage where the predicted pressure Pp tends to rise. As a result, the pressure regulating time becomes unnecessarily long.

[ means of solving the problems ]

A valve control device according to a first aspect of the present invention controls an opening degree of a vacuum valve attached to a vacuum chamber by opening control, and includes: an opening setting unit for outputting an opening setting value of the opening control; a first determination unit that determines that the pressure balance value in the case of the fixed value does not exceed the valve control pressure target value in a predetermined period within a fixed period in which the opening degree setting value is the fixed value; and an opening degree control unit that controls the opening degree based on a determination result of the first determination unit.

The vacuum valve according to the second aspect of the present invention includes: a valve body; a valve body driving unit for driving the valve body; and the valve control device controls the valve body driving part to control the opening degree of the valve body.

[ Effect of the invention ]

According to the present invention, shortening of the pressure regulating time can be achieved.

Drawings

Fig. 1 is a block diagram showing a schematic configuration of a vacuum valve mounted to a vacuum processing apparatus.

Fig. 2 is a functional block diagram of a vacuum valve and a valve control device.

Fig. 3 is a diagram illustrating the voltage regulation logic under the split control.

Fig. 4 shows an example of the locus of points (θr, pr) obtained by the on control.

Fig. 5 is a diagram illustrating a relationship between the pressure measurement value Pr and the predicted pressure Pp, and shows a case where Pe (θr) > Ps.

Fig. 6 is a diagram illustrating a relationship between the pressure measurement value Pr and the predicted pressure Pp, and shows a case where Pe (θr) < Ps.

Fig. 7 is a diagram showing the conditions for determining the use of the predicted pressure Pp in the case of "exceeding" and "not exceeding at all", and the opening degree relationship and the pressure relationship at this time.

Fig. 8 is a diagram showing an example of the relationship between the opening θ and the effective exhaust speed Se.

Fig. 9 is a diagram showing the transition of the pressure (pressure measurement value Pr and predicted pressure Pp), index signal β, and opening setting θ1 when opening setting θ1 is fixed to value θ1_fix.

Fig. 10 is a diagram showing conditions for determining the utilization index signal β in the case of "exceeding" and "not exceeding at all", and the opening degree relationship and the pressure relationship at this time.

Fig. 11 is a flowchart showing an example of control processing when the predicted pressure Pp and the index signal β are used together.

Fig. 12 is a diagram showing transitions of the pressure measurement value Pr, the index signal β, the opening setting θ1, and the opening setting θ2 in the rising case (up case).

Fig. 13 is a diagram showing a comparative example.

Fig. 14 is a flowchart for explaining modification 1.

Fig. 15 is a diagram illustrating a rising case in modification 1.

Fig. 16 is a flowchart for explaining modification 2.

Fig. 17 is a flowchart for explaining modification 3.

[ description of symbols ]

1: valve body

2: valve control device

3: vacuum chamber

4: vacuum pump

12: valve body

13: motor with a motor housing

21: pressure regulating controller

22: motor driving part

23: storage unit

210: estimation calculation unit

220: feedforward controller

230: feedback controller

Pp: pressure pre-measurement

Pr: pressure measurement

Ps: target value of pressure

θ1, θ2, θset: opening setting

θr: opening degree measurement value

θse: target opening degree estimation value

Beta, beta 1, beta 2: index signal

Detailed Description

Hereinafter, embodiments of the present invention will be described with reference to the drawings. Fig. 1 is a block diagram showing a schematic configuration of a vacuum valve mounted to a vacuum processing apparatus. The vacuum valve includes a valve body 1 provided with a valve body 12, and a valve control device 2 for controlling the driving of the valve body. The valve body 1 is installed between the vacuum chamber 3 and the vacuum pump 4. A process gas or the like is introduced into the vacuum chamber 3 through a flow controller 32. The flow rate controller 32 is a device for controlling the flow rate Qin of the gas introduced into the vacuum chamber 3, and is controlled by a main controller (not shown) of the vacuum processing apparatus provided with the vacuum chamber 3. The pressure (chamber pressure) in the vacuum chamber 3 is measured by the vacuum gauge 31, and the measured value Pr is input to the valve control device 2.

The valve body 1 is provided with a motor 13 for driving the valve body 12 to open and close. The valve body 12 is driven to open and close by a motor 13. The motor 13 is provided with an encoder (encoder) 130 for detecting the opening/closing angle of the valve body 12. The detection signal of the encoder 130 is input to the valve control device 2 as an opening degree signal θr (hereinafter referred to as an opening degree measurement value θr) of the valve body 12.

The valve control device 2 for controlling the valve body 1 includes a pressure regulating controller 21, a motor driving unit 22, and a storage unit 23. The valve control device 2 is inputted with a pressure target value Ps of the vacuum chamber 3 from the main controller of the vacuum processing apparatus described above, in addition to the pressure measurement value Pr and the opening measurement value θr described above. The storage unit 23 stores parameters (for example, data related to an effective exhaust speed Se, a controlled device gain Gp, and the like, which will be described later) required for the control valve. The motor driving unit 22 includes an inverter circuit for driving a motor and a motor control unit for controlling the inverter circuit, and is input with an opening degree measurement value θr from the encoder 130. The chamber pressure Pr measured by the vacuum gauge 31 is input to the pressure regulating controller 21, and the target pressure Ps of the vacuum chamber 3 is input from the main controller of the vacuum processing apparatus described above.

The valve control device 2 includes an arithmetic processing device such as a microcomputer (micro computer) having a central processing unit (Central Processing Unit, CPU), a Memory (Read Only Memory (ROM), a random access Memory (Random Access Memory, RAM)), a peripheral circuit, and the like, and functions of the voltage regulator controller 21 and a motor control unit of the motor drive unit 22 are realized by a software program (software program) stored in the ROM. The storage unit 23 includes a memory of a microcomputer. In addition, a digital arithmetic unit such as a field programmable gate array (Field Programmable Gate Array, FPGA) and peripheral circuits thereof may be included.

Fig. 2 is a functional block diagram of the valve body 1 and the valve control device 2. The voltage regulation controller 21 includes an estimation operation unit 210, a feed-forward controller (feed-forward controller) 220, and a feedback controller (feedback controller) 230. The opening degree measurement value θr detected by the encoder 130 is input to the motor driving unit 22 and the estimation computing unit 210.

In the present embodiment, too, as in the invention of patent document 1 described above, coarse adjustment is performed to the vicinity of the pressure target value by the on control, and then the control is switched to the off control, and the pressure target value is approximated by fine adjustment. In the voltage regulation controller 21, the estimation computing unit 210 and the feedforward controller 220 correspond to an on-control unit, and the subtractor and the feedback controller 230 that generate the deviation ε (=Pr-Ps) correspond to an off-control unit.

The pressure measurement value Pr, the pressure target value Ps, and the opening degree measurement value θr are input to the estimation calculation unit 210 of the opening control unit. The estimation calculation unit 210 calculates a target opening degree estimation value θse, a predicted pressure Pp, and an index signal β indicating a tendency of the pressure measurement value Pr to change, based on the pressure measurement value Pr, the pressure target value Ps, and the opening degree measurement value θr. In general, even if the valve opening is fixed to a constant value, it takes a certain time for the chamber pressure to reach a pressure balance value corresponding to the valve opening. The predicted pressure Pp is a pressure estimated value obtained by passing t seconds from the time point of measuring the pressure measurement value Pr. The method for estimating the predicted pressure Pp is described in detail in japanese patent application laid-open No. 2018-106718, which is described above, and the description thereof is omitted here. In addition, the calculation method of the target opening degree estimation value θse and the index signal β will be described below.

The feedforward controller 220 of the opening control unit outputs the opening setting θ1 based on the target opening estimated value θse and the index signal β. The feedback controller 230 of the closed control unit outputs the opening degree setting θ2 based on the deviation ε=Pr-Ps. The output opening degree setting θ1 and the opening degree setting θ2 are added, and the addition result is input to the motor driving unit 22 as an opening degree setting θset. The motor driving unit 22 drives the motor 13 based on the opening setting θset and the opening measurement value θr input from the encoder 130.

As described later, in the present embodiment, the determination of the use prediction pressure Pp in the on control and the determination of the use of the index signal β related to the pressure balance value when the opening degree setting θ1 is fixed are used together. The determination using the predicted pressure Pp is performed according to the pressure adjustment logic shown in fig. 3. Hereinafter, a description will be given of the voltage adjustment logic using the predicted voltage Pp of fig. 3, and then a description will be given of the setting method of the index signal β and the determination method using the index signal β.

The coordinate system shown in fig. 3 is a coordinate system θ -P having points (θse, ps) as origins. The target opening degree estimation value θse is the opening degree of the valve body 12 at the time of estimating the calculation pressure target value Ps. In the coordinate system θ -P, which of the opening θ is controlled in the opening direction and the closing direction is determined based on which of the first to fourth quadrants the point (θr, pr) is located in and the magnitude relation between the predicted pressure Pp and the pressure target value Ps.

When points (thetar, pr) before changing the opening degree are located in the second quadrant and the fourth quadrant, the opening degree theta is adjusted in the direction of the opening degree target estimated value thetase. As a result, the pressure measurement value Pr changes in the direction of the pressure target value Ps. That is, the pressure decreases in the case of the second quadrant, and increases in the case of the fourth quadrant. The start point of the control is either the second quadrant or the fourth quadrant, and the control becomes the fourth quadrant in the case of an ascending case in which the pressure rises from the pressure Pstt at the start point, and becomes the second quadrant in the case of a descending case (down case) in which the pressure falls from the pressure Pstt at the start point.

On the other hand, when the points (θr, pr) before the opening change are located in the first quadrant and the third quadrant, the opening adjustment direction is set according to the magnitude relation between the predicted pressure Pp and the pressure target value Ps. In the first quadrant, when the predicted pressure Pp is greater than the pressure target value Ps as Pp > Ps, the opening degree is adjusted in the opening degree increasing direction (the direction indicated by the right arrow) or the opening degree value is maintained as it is as in the circle 50. Conversely, in the case of Pp Ps, the opening degree is adjusted in the direction of decreasing the opening degree (the direction indicated by the left arrow). In the third quadrant, when Pp > Ps, the opening degree is adjusted in the direction of increasing the opening degree (the direction indicated by the right arrow). Conversely, in the case of Pp Ps, the opening degree is adjusted in the direction of decreasing the opening degree (the direction indicated by the left arrow), or the opening degree value is maintained as it is as indicated by the circle 50.

Fig. 4 shows an example of the trajectory of the points (θr, pr) obtained by the on control in the case where the coordinates (θr, pr) of the start point of the control start are located in the fourth quadrant as the points A1, B1. In the example where the starting point is the point A1, the opening degree is changed by the opening control, and the movement is made to the point A2 in the vicinity of the target opening degree estimated value θse. When the pressure measurement value Pr becomes close to the target opening degree estimation value θse, the closing control is performed.

In the example where the starting point is point B1, the opening degree is changed by the opening control to point B2 in the third quadrant. Here, the estimated pressure Pp calculated by the estimation when moving to the point B2 is Pp Ps, and the opening degree of the valve body 12 is maintained at the opening degree of the position of the point B2. While the opening degree is maintained, the pressure measurement value Pr and the estimated pressure Pp calculated by the estimation continue to rise, and the positions of the points (θr, pr) move upward. Next, at the point of time of the movement to the point B3, the predicted pressure Pp becomes Pp > Ps, and the opening degree is changed in the increasing direction. As a result, the opening degree is changed to move to the point B4 in the vicinity of the target opening degree estimated value θse. When the pressure measurement value Pr becomes close to the target opening degree estimation value θse, the closing control is performed.

If the opening degree measured value θr finally becomes the target opening degree estimated value θse by the open control, the open control is switched to the closed control with the opening degree setting θ1 output from the feedforward controller 220 of fig. 2 fixed to the value, and if the switch is made to the closed control, the pressure target value Ps is input to the subtraction point of fig. 2, and the feedback controller 230 outputs the opening degree setting θ2 based on the deviation ε=pr—ps. Typically, the feedback controller 230 includes a proportional gain, an integral gain (so-called proportional integral (Proportional Integral, PI) gain). The motor driving unit 22 controls the opening degree based on the opening degree setting θset=θ1 (fixed) +θ2. As long as the valve body 12 is not operating at high speed, the opening degree measurement value θr becomes θr=θ1+θ2.

Before the closing control is started, a pressure measurement value Pr is input to the subtraction point instead of the pressure target value Ps. Therefore, the pressure deviation epsilon=0 is input to the feedback controller 230 at the time of the on control, and the opening degree setting θ2=0 is output from the feedback controller 230.

As described above, the predicted pressure Pp in the logic of fig. 3 is an estimated value of the chamber pressure after t seconds have elapsed from the present time (for example, a time longer than the control cycle as t=0.4 seconds). In the control example of b1→b2→b3→b4 shown in fig. 4, the predicted pressure Pp exceeds the pressure target value Ps (i.e., pp > Ps) at the point of time B3, and the valve opening is changed in the direction of the target opening estimated value θse and is moved to the point B4.

Fig. 5 is a graph showing a relationship between the pressure measurement value Pr and the predicted pressure Pp, wherein the horizontal axis shows time and the vertical axis shows pressure. Lines L1 and L2 show the time passes of the pressure measurement value Pr and the predicted pressure Pp when the valve opening is fixed to the opening measurement value θr at the point B2. The predicted pressure Pp is a predicted pressure after t seconds from the time point of measuring the pressure measurement value Pr, and thus the line L2 is shifted further upward than the line L1 of the pressure measurement value Pr. At the point in time of the point B2, the predicted pressure Pp becomes Pp < Ps with respect to the pressure target value Ps.

However, if the pressure measurement value Pr increases in the pressure direction in the equilibrium state with the passage of time, the predicted pressure Pp also increases, and the predicted pressure Pp exceeds the pressure target value Ps at the point of time B3. When the time passes while being fixed to the opening degree measurement value θr, the pressure measurement value Pr also exceeds the pressure target value Ps, and when the time passes sufficiently, the pressure measurement value Pr converges to the pressure balance value Pe (θr) when the opening degree measurement value θr is reached. In the equilibrium state, the predicted pressure Pp also converges to Pe (θr).

Accordingly, the opening degree measurement value θr is smaller than the opening degree corresponding to the pressure target value Ps, and readjustment in the direction of increasing the opening degree of the valve body is required. By using the predicted pressure Pp, it can be determined that readjustment is necessary at the time point before the actual pressure measurement value Pr exceeds the pressure target value Ps, that is, at the timing B3 when the predicted pressure Pp becomes Pp > Ps.

On the other hand, in the control example of a1→a2 shown in fig. 4, the predicted pressure Pp does not exceed the pressure target value Ps. At this time, it is determined at which point in time the predicted pressure Pp does not exceed the pressure target value Ps, and the pressure target value Ps becomes ambiguous. For example, if the pressure rise rate of the predicted pressure Pp is substantially zero, it can be determined that the predicted pressure Pp does not exceed the pressure target value Ps, but it takes a long time to recognize that the pressure rise rate is substantially zero. Therefore, the pressure regulating time is unnecessarily long. In the control example such as a1→a2, the calculation error of the target opening degree estimated value θse is small. However, the calculation error of the target opening degree estimated value θse is large, and as a result, the calculation error often deviates in a direction larger than the actual target opening degree θs. In particular, it is known that the difference between the gas types (molecular weights) and the gas flow rates is large in cases where the gas types (molecular weights) and the gas flow rates are greatly different from each other.

In general, when the calculation error of the target opening degree estimation value θse is unavoidable and the target opening degree θs is larger than the true target opening degree θs for the pressure target value Ps, as shown in fig. 6, the pressure balance value (pressure value in the balance state) Pe (θr) of the pressure measurement value Pr becomes Pe (θr) < Ps. At this time, the pressure balance value Pe (θr) deviates downward in the drawing from the pressure target value Ps, and the predicted pressure Pp does not exceed the pressure target value Ps at all.

Fig. 4 shows a case where the pressure target value Ps is higher than the pressure Pstt at the start of pressure regulation, but also a case where the predicted pressure Pp exceeds the pressure target value Ps and a case where the pressure target value Ps does not exceed the pressure Pstt at the start of pressure regulation in a case where the pressure target value Ps is lower than the pressure Pstt at the start of pressure regulation. In the case of descent, ps < Pstt, so that the predicted pressure Pp "exceeds" the pressure target value Ps means that Pp < Ps, and "does not exceed" means that Pp > Ps. In addition, when the predicted pressure Pp does not exceed the pressure target value Ps at all, the judgment that the predicted pressure Pp does not exceed the pressure target value Ps at all becomes ambiguous as described above, and thus there arises a problem that the pressure adjustment time becomes unnecessarily long, and the like.

Fig. 7 shows a summary of the conditions for determining the use of the predicted pressure Pp in the case of "exceeding" and "not exceeding at all", and the opening degree relationship and the pressure relationship at that time. In fig. 7, the opening θ1_1 is the opening setting θ1 at the timing of the prediction determination during the open control period, and pe (θ1_1) is the pressure balance value at the opening θ1_1. The prediction determination timing is set to a timing at which the opening θ is fixed. The target opening degree estimation value θse is described on the premise that there is an error from the actual target opening degree θs.

In the rising case (Pstt < Pr), when the predicted pressure Pp "exceeds (Pp > Ps)" the pressure target value Ps, the point (θr=θ1_1, pr) is located in the third quadrant as in the point B2 of fig. 4, and the opening θ1_1 set as θ1_1θse becomes θ1_1 < θs. That is, θ1_1θse < θs, and the pressure relationship is Pstt < Pr < Ps < Pe (θse) Pe (θ1_1).

In the rising case (Pstt < Pr), when the predicted pressure Pp "does not exceed (Pp < Ps)" the pressure target value Ps at all, the point (θr=θ1_1, pr) is located in the fourth quadrant as in the point A2 of fig. 4, such as θ1_1

The opening θ1_1 set as θse becomes θ1_1 > θs. That is, θ1_1θse > θs, and the pressure relationship is Pstt < Pr < Pe (θ1_1) Pe (θse) < Ps.

In the case of descent (Pstt > Pr), when the predicted pressure Pp "exceeds (Pp < Ps)" the pressure target value Ps, the opening θ1_1 set as θ1_1θse becomes θ1_1 > θs. That is, θ1_1θse > θs, and the pressure relationship is Pe (θ1_1) Pe (θse) < Ps < Pr < Pstt.

In the case of descent (Pstt > Pr), when the predicted pressure Pp "does not exceed (Pp > Ps)" the pressure target value Ps at all, the opening θ1_1 set as θ1_1θse becomes θ1_1 < θs. Namely, θ1_1

θse < θs, and the pressure relationship is Ps < Pe (θse) Pe (θ1_1) < Pr < Pstt.

(introduction of index Signal beta)

In the present embodiment, the index signal β is defined by the following formula (1). Regarding the pressure measurement value Pr of the vacuum chamber 3, the expression of the exhaust gas expressed by the following expression (2) holds, and the molecule (dPr/dt) of the index signal β is the amount of the expression based on the exhaust gas.

β＝(dPr/dt)/(Ps-Pr)···(1)

V×(dPr/dt)+Se×Pr＝Qin···(2)

In the formula (2), qin on the right side is the flow rate of the introduced gas into the vacuum chamber 3. On the left side of equation (2), V is the volume of the vacuum chamber 3, dPr/dt is the time derivative of the pressure measurement value Pr, and Se is the effective exhaust speed of the vacuum exhaust system associated with the exhaust of the vacuum chamber 3. The effective exhaust speed Se is an amount determined by: a conductance (conductance) determined by the chamber structure and the opening degree θr of the valve body 12, and an exhaust speed of the vacuum pump 4. The relationship between the opening θ of the valve body 12 and the effective exhaust speed Se is generally a monotonically increasing relationship as shown in fig. 8.

In general, when the vacuum valve is mounted in the vacuum chamber 3 and used, an initial correction operation related to the effective exhaust speed Se, that is, an initial correction operation related to the valve body control is performed. In general, the gain correction of the controller is performed based on the volume of the vacuum chamber 3, the sensitivity of the valve, and the like in a state where the representative gas or the average gas condition (gas type, gas flow rate) is set as the process condition to be applied. As the average condition, for example, the average molecular weight of the mixed gas is often obtained and replaced with a gas type that is relatively easy to handle.

The semiconductor process in the vacuum chamber 3 equipped with the vacuum valve includes a plurality of pressure adjustment events in which various conditions of the gas type, the gas flow Qin, and the pressure target value Ps introduced into the vacuum chamber 3 are changed at predetermined intervals. In each pressure adjustment event, immediately after the start of the pressure adjustment, the gas flow rate is converged to a predetermined flow rate value by the flow controller 32 (see fig. 1), and at the same time, the valve opening is adjusted to control the effective exhaust speed Se, whereby the chamber pressure (pressure measurement value Pr) is converged to the pressure target value Ps.

In general, the timing of completion of the flow control convergence by the flow controller 32 is earlier than the timing of completion of the pressure control convergence by the vacuum valve. Further, in the present embodiment, the flow rate value may be considered to be almost converged at the prediction determination timing at the time of the on control. That is, the gas flow Qin converges to a constant flow value Qin0, and qin=qin0=a constant value.

The predicted determination timing is a timing at which the opening degree is fixed, and the opening degree at the fixed timing is denoted as θ1_fix. Hereinafter, the effective exhaust speed Se (θ1_fix) corresponding to the opening θ1_fix is denoted as se_fix, and the effective exhaust speed Se (θs) corresponding to the opening θs at the time of the pressure target value Ps is denoted as Se0. In this case, the flow rate Qin0 is represented by the following formula (3). Pe (θ1_fix) is a pressure balance value at the opening θ1_fix.

Qin0＝Se0×Ps＝Se_fix×Pe(θ1_fix)···(3)

According to equation (3), the pressure balance value Pe (θ1_fix) is represented by the following equation (4). The expression of the exhaust gas when the opening degree is fixed to θ1_fix is expressed as the following expression (5).

Pe(θ1_fix)＝(Se0/Se_fix)×Ps···(4)

V×(dPr/dt)+Se_fix×Pr＝Qin0···(5)

The numerator and denominator of the above-described formula (1) providing the index signal β can be modified according to the formulas (4) and (5) as shown in the formulas (6) and (7).

Molecule= (dPr/dt)

＝(Se0×Ps-Se_fix×Pr)/V

＝(Se_fix/V)×{(Se0/Se_fix)×Ps-Pr}···(6)

Denominator= (Ps-Pr)

＝{(Se0/Se_fix)×Ps-Pr}-{(Se0/Se_fix)×Ps-Ps}···(7)

Therefore, the index signal β is expressed by the following expression (8). In the expression (8), only the pressure measurement value Pr is a time-varying amount, and varies with time like pr→ps. The other amount is an amount (a certain value) which does not change with time.

β＝(Se_fix/V)/[1-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pr}]···(8)

(determination method Using index Signal beta)

The determination method for the four cases shown in fig. 7 will be described below with reference to the index signal β supplied by the expression (8). Here, if the relation between the applied opening θ and the effective exhaust speed Se is in a monotonically increasing relation as shown in fig. 8, for example, if θ1_fix > θs, it may be called se_fix > Se0. In the following description, the relationship of fig. 8 is applied. First, a case where "no exceeding" is problematic in the case of using the predicted pressure Pp will be described, and then a case where "exceeding" is described.

[1 "no more than" in ascending case ]

As described with reference to fig. 7, the pressure measurement value Pr during the pressure rise satisfies the relationship Pstt < Pr < Pe (θ1_1) Pe (θse) < Ps. As described above, θ1_1 is collectively denoted as θ1_fix, and thus the following relationship of formula (9) can be obtained from the relationship of Pr < Pe (θ1_fix) < Ps and formula (4).

Pr＜(Se0/Se_fix)×Ps＜Ps···(9)

Here, if F (Pr) of the following formula (10) is used, the right denominator of the formula (8) is denoted as 1+F (Pr).

F(Pr)＝-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pr}···(10)

From the expression (9), it can be seen that { (Se 0/Se_fix) ×Ps-Ps } < 0, { (Se 0/Se_fix) ×Ps-Pr } > 0 holds, and F (Pr) > 0. Therefore, it is known that F (Pr) (i.e., the right denominator of equation (8)) becomes a positive value, and even if the pressure measurement value Pr changes with time, the index signal β= (dPr/dt)/(ps—pr) is always positive.

In the case of a rising case in which the pressure is rising from the pressure Pstt at the start of control, the pressure measurement value Pr is Pr > Pstt, and thus the following expression (11) holds according to the relation of { (Se 0/se_fix) ×ps-Pstt } > { (Se 0/se_fix) ×ps-Pr } > 0.

{(Se0/Se_fix)×Ps-Pr}/{(Se0/Se_fix)×Ps-Pstt}＜1···(11)

The relationship between the magnitudes of the formula (F) and F (Pr) obtained by multiplying F (Pr) in formula (10) by formula (11) is as shown in formula (12).

F(Pr)＞-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pstt}···(12)

The right denominator of formula (8) is 1+F (Pr) and positive, so that the following formula (13) holds.

(Right denominator) > 1- { (Se 0/Se_fix) ×Ps-Ps }/{ (Se 0/Se_fix) ×Ps-Pstt } 1/(right denominator) < 1/[1- { (Se 0/Se_fix) ×Ps-Ps }/{ (Se 0/Se_fix) ×Ps-Pstt } ]. Cndot.13

Since (se_fix/V) of expression (8) is a positive value, the index signal β satisfies expression (14) below according to the relation of expression (13).

β＜(Se_fix/V)/[1-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pstt}]···(14)

In the rising case, the pressure measurement value Pr in the expression (8) of the index signal β rises from the pressure Pstt of the starting point. As described above, the F (Pr) contained in the right denominator= 1+F (Pr) of the formula (8) is a positive value, and the smaller Pr is, the smaller F (Pr) is. Therefore, the right denominator= 1+F (Pr) of the expression (8) becomes smaller as Pr becomes smaller, and the index signal β represented by the expression (8) becomes maximum at the pressure Pstt at the start of the rise of the pressure measurement value Pr. Further, as is clear from the equation (14), the value of the index signal β (i.e., the maximum value of the index signal β) when pr=pstt becomes larger as the value of Pstt becomes smaller, and the value Se0/V when pstt=0 on the right side of the equation (14) becomes the upper limit value of the index signal β.

On the other hand, the pressure measurement value Pr starts to rise from the pressure Pstt at the start point, and in a state where it almost reaches (Se 0/se_fix) ×ps, F (Pr) represented by formula (10) becomes F (Pr) →++infinity. When F (Pr) is used, the index signal β is expressed as β= (se_fix/V)/[ 1+F (Pr) ], and thus the index signal β converges to +0 as β→ +0. As described above, in the case of "no exceeding at all" in the rising case, it is known that the index signal β monotonously decreases to the zero value with Se0/V as the upper limit.

Fig. 9 is a diagram showing the transition of the pressure (pressure measurement value Pr and predicted pressure Pp), index signal β, and opening setting θ1 when opening setting θ1 is fixed to value θ1_fix. After the opening degree setting θ1 is fixed to θ1_fix, the pressure measurement value Pr and the predicted pressure Pp also continuously rise, and if the pressure measurement value Pr and the predicted pressure Pp last for a sufficiently long time, both converge to the pressure balance value Pe (θ1_fix) at the opening degree θ1_fix. The calculated index signal β monotonically decreases toward a zero value with Se0/V as an upper limit. Therefore, by monotonously decreasing the detection index signal β, it can be determined that the opening setting θ1=θ1_fix is "no more than".

[2 "no more than" in case of descent ]

Next, a response procedure from the pressure Pstt of the starting point to (Se 0/se_fix) ×ps of the pressure measurement value Pr in the case of "no exceeding" in the case of descent is studied. As will be described later, it is also known that (se_fix/V) monotonically decreases toward zero value with the upper limit, as in the case of "no more than" in the rising case.

The pressure measurement Pr in the pressure drop satisfies the relationship of Ps < Pe (θ1_1) Pe (θse) < Pr < Pstt. As described above, θ1_1 is collectively denoted as θ1_fix, and thus the relationship of the following formula (15) can be obtained from Ps < Pe (θ1_1) < Pr, and Pe (θ1_fix) = (Se 0/se_fix) ×ps of formula (4) in the relationship of the pressure.

Ps＜(Se0/Se_fix)×Ps＜Pr···(15)

According to the formula (15) { (Se 0/Se_fix) ×Ps-Pr } < 0, { (Se 0/Se_fix) ×Ps-Ps } > 0, F (Pr) represented by the formula (10) becomes F (Pr) > 0. Therefore, the right denominator of the equation (8) becomes a positive value, and even if the pressure measurement value Pr changes with time, the index signal β becomes positive.

In the case of a drop case where the pressure falls from the pressure Pstt at the start of control, the pressure measurement value Pr is Pr < Pstt, and thus the expression (11) is similarly established according to the relation of- { (Se 0/se_fix) ×ps-Pstt } > - { (Se 0/se_fix) ×ps-Pr } > 0.

{(Se0/Se_fix)×Ps-Pr}/{(Se0/Se_fix)×Ps-Pstt}＜1···(11)

The equation (12) is similarly established with respect to the magnitude relation between F (Pr) and F (Pr) obtained by multiplying F (Pr) in equation (10) by the left side of equation (11).

F(Pr)＞-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pstt}···(12)

Therefore, the formulas (13) and (14) are established in the same manner as in the case of "no more than" in the above-mentioned ascending case.

1/(Right denominator) < 1/[1- { (Se 0/Se_fix). Times.Ps-Ps }/{ (Se 0/Se_fix). Times.Ps-Pstt } ]. Times.

β＜(Se_fix/V)/[1-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pstt}]···(14)

In the case of the drop, the pressure measurement value Pr in the expression (8) of the index signal β drops from the pressure Pstt at the start point to (Se 0/se_fix) ×ps (> Ps) corresponding to the opening θ1_fix. In the case of the descent, the { (Se 0/Se_fix) ×Ps-Pr } of the formula (8) becomes negative, and as for the absolute value thereof, the closer the pressure measurement Pr is to (Se 0/Se_fix) ×Ps, the smaller the absolute value becomes. That is, the right denominator of the expression (8) is that the closer the pressure measurement value Pr is (Se 0/se_fix) ×ps, the larger the index signal β becomes, whereas the smaller the index signal β becomes. In other words, the index signal β reaches the maximum at the pressure Pstt at the start of the drop in the pressure measurement value Pr. Therefore, in the case of the drop, the value Se0/V when pstt=0 is set to the right of the expression (14) becomes the upper limit value of the index signal β.

On the other hand, the pressure measurement value Pr decreases from the pressure Pstt at the start point, and in a state where it almost reaches (Se 0/se_fix) ×ps, F (Pr) represented by the formula (10) becomes F (Pr) →+. When F (Pr) is used, the index signal β is expressed as β= (se_fix/V)/[ 1+F (Pr) ], and thus the index signal β converges to +0 as β→ +0. As a result, it was found that, in the case of "no exceeding at all" in the case of the decrease, the index signal β monotonically decreases toward the zero value with Se0/V as the upper limit.

[3 "exceeding" in ascending case ]

In the case of "exceeding" in the rising case, as shown in fig. 7, the relationship of Pstt < Pr < Ps < Pe (θse) Pe (θ1_1) is satisfied. Since Pr < Ps < Pe (θ1_1), if Pe (θ1_1) =pe (θ1_fix) = (Se 0/se_fix) ×ps, in expression (8), it is expressed that { (Se 0/se_fix) ×ps-Pr } > 0 and { (Se 0/se_fix) ×ps-Ps } > 0. Since Pr < Ps, the following formula (16) holds.

0＜{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pr}＝-F(Pr)＜1···(16)

Based on the right-hand magnitude relation of equation (16), the right denominator= 1+F (Pr) in equation (8) becomes positive, and the index signal β becomes positive. In the case of the rising case, as described above, the expression (11) holds.

{(Se0/Se_fix)×Ps-Pr}/{(Se0/Se_fix)×Ps-Pstt}＜1···(11)

When the negative value of F (Pr) is multiplied by both sides of the expression (11), the following expression (17) can be obtained.

-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pstt}＞F(Pr)···(17)

The expression (17) is modified by the expression (18).

1-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pstt}＞1+F(Pr)

1/[1-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pstt}]＜1/{1+F(Pr)}···(18)

If both sides of the expression (18) are multiplied by (se_fix/V) as a positive value, the following expression (19) can be obtained.

(Se_fix/V)/[1-{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pstt}]＜β···(19)

In the rising case, the pressure measurement value Pr in the expression (8) of the index signal β rises from the pressure Pstt of the starting point. The { (Se 0/Se_fix) ×Ps-Ps }/{ (Se 0/Se_fix) ×Ps-Pr) of the right denominator of formula (8) takes a value of 0 to 1 as in formula (16), and the smaller Pr, that is, the farther Pr is from (Se 0/Se_fix) ×Ps, the smaller Pr becomes. Therefore, the smaller Pr is, the larger the right denominator of the expression (8) is, whereas the smaller the index signal β is. Therefore, the index signal β becomes minimum at the pressure Pstt at the start of the rise of the pressure measurement value Pr. As is clear from the equation (19), the value of the index signal β (i.e., the minimum value of the index signal β) when pr=pstt is smaller as Pstt is smaller, and the value Se0/V when pstt=0 on the right side of the equation (19) is the lower limit value of the index signal β.

In addition, the pressure measurement value Pr rises from the pressure Pstt at the start point, and in a state where it almost reaches Ps, the right denominator of the formula (8) is positive and near zero as (right denominator) → +0, thus indicating the signal at this time beta is beta → +++. That is, it is known that the index signal β monotonously increases with Se0/V as the lower limit when "exceeding" in the rising case.

[4 "exceeding" in drop case ]

In the case of "exceeding" in the drop case, as shown in fig. 7, the relationship of Pe (θ1_1) Pe (θse) < Ps < Pr < Pstt is satisfied. Since Pe (θ1_1) < Ps < Pr, in the formula (8), it is { (Se 0/Se_fix) ×Ps-Pr } < 0, and { (Se 0/Se_fix) ×Ps-Ps } < 0. Since Ps < Pr, the above equation (16) is satisfied, the right denominator of equation (8) is positive, and the index signal β is positive.

0＜{(Se0/Se_fix)×Ps-Ps}/{(Se0/Se_fix)×Ps-Pr}＝-F(Pr)＜1···(16)

In the case of "exceeding" in the descending case, the expression (11) is established similarly to the case of "not exceeding at all" in the descending case.

{(Se0/Se_fix)×Ps-Pr}/{(Se0/Se_fix)×Ps-Pstt}＜1···(11)

With regard to the expression (11), the expressions (17) to (19) can be obtained as in the case of "exceeding" in the ascending case.

In the case of the drop, the pressure measurement value Pr in the expression (8) of the index signal β drops from the pressure Pstt of the start point. The right denominator of formula (8) { (Se 0/Se_fix) ×Ps-Ps }/{ (Se 0/Se_fix) ×Ps-Pr) takes a value of 0 to 1 as shown in formula (16), and the smaller Pr, that is, the closer Pr is to (Se 0/Se_fix) ×Ps, the larger Pr becomes. Therefore, the closer Pr is to (Se 0/Se_fix). Times.Ps, the smaller the right denominator of the expression (8), whereas the larger the index signal β. Therefore, the index signal β is minimum when the pressure measurement value Pr is the pressure Pstt farthest from (Se 0/se_fix) ×ps. Further, as is clear from the equation (19), the value of the index signal β (i.e., the minimum value of the index signal β) when pr=pstt is smaller as Pstt is smaller, and the value Se0/V when pstt=0 on the right side of the equation (19) becomes the lower limit value of the index signal β.

In addition, the pressure measurement value Pr starts to decrease from the pressure Pstt at the start point, and in a state where it almost reaches Ps, the right denominator of the formula (8) is positive and near zero as (right denominator) → +0, thus indicating the signal at this time beta is beta → +++. That is, it is known that, in the case of "exceeding" in the case of decreasing, the index signal β monotonously increases with Se0/V as the lower limit.

Fig. 10 shows the results of the case "exceeding" and the case "not exceeding" in the ascending case and the descending case described above.

(timing of generation of index Signal beta)

An example of the generation timing of the index signal β by the estimation and calculation unit 210 in fig. 2 will be described. The pressure target value Ps, the pressure measurement value Pr, and the opening measurement value θr are input to the estimation computing unit 210. The estimation calculation unit 210 calculates Δpr/Δt from the difference Δpr between the pressure measurement values Pr in an appropriate control period Δt (for example, 10 ms), obtains the pressure differential value (dPr/dt) of the expression (1), and calculates the index signal β based on the pressure differential value, the pressure target value Ps, and the pressure measurement value Pr.

As described above, the index signal β does not use parameters other than the pressure target value Ps and the pressure measurement value Pr, and is therefore applicable even if the gas type and the gas flow rate of the gas introduced into the vacuum chamber 3 are unknown. In other words, even if the gas type and the gas flow rate are unknown, the index signal β with high reliability can be obtained.

When the opening degree setting θ1 of the on control is fixed, that is, when θ1_fix is set as the generation timing of the index signal β, it is sufficient to determine whether the index signal β increases monotonically or decreases monotonically with respect to the passage of time. For example, when the control period is set to 10ms, the index signal β is calculated 10 to 40 times in a period of 100ms to 400ms, and is calculated a plurality of times for each 10 ms. Further, when the tendency of continuously increasing (the previous value < the present value) is seen for the index signal β calculated a plurality of times, it is determined as "exceeding", and when the tendency of continuously decreasing (the previous value > the present value) is seen, it is determined as "not exceeding at all".

(control processing based on the predicted pressure Pp and the index signal β)

Fig. 11 is a flowchart showing an example of control processing by the voltage regulator controller 21 when the predicted voltage Pp and the index signal β are used together. Fig. 11 shows a determination process using the predicted pressure Pp and the index signal β, and a post-determination opening degree control process. Fig. 12 is a diagram showing transition of the pressure measurement value Pr, the index signal β, and the opening setting θ1 in the rising case. In the determination using both the predicted pressure Pp and the index signal β, the "exceeding" is determined by the predicted pressure Pp, and the "not exceeding" is determined by the index signal β, which will be described below.

When the pressure target value Ps is input to the pressure regulation controller 21, a target opening degree estimation value θse is calculated by the estimation calculation unit 210 based on the pressure target value Ps, and an opening degree setting θ1_fix is output from the feedforward controller 220 based on the target opening degree estimation value θse. The opening degree setting θ1_fix is input to the motor driving unit 22, and the valve body 12 is driven so that the valve body opening degree becomes the value of the opening degree setting θ1_fix.

In step S10 of fig. 11, it is determined whether or not the opening degree measurement value θr is a value of the opening degree setting θ1_fix output from the feedforward controller 220. If it is determined in step S10 that θr=θ1_fix, the flow advances to step S20. In step S20, the estimation calculation unit 210 calculates the index signal β. In step S30, m, which indicates the number of times β (previous) > β (current) is determined in step S70 described later, is set to m=0.

In step S40, a predicted pressure Pp is calculated. In step S50, it is determined whether or not the predicted pressure Pp exceeds the pressure target value Ps. If it is determined that Pp > Ps, the flow advances to step S100 to execute the opening degree changing process. The opening degree changing process of step S100 will be described below. On the other hand, if it is determined that Pp > Ps is not present, the process proceeds to step S60. In step S60, an index signal β is calculated.

In step S70, it is determined whether or not the predicted pressure Pp "does not exceed" the pressure target value Ps at all. Specifically, the index signal β (the previous time) calculated in step S20 is compared with the index signal β (the current time) calculated in step S60, and whether β (the previous time) > β (the current time) is determined. As shown in fig. 10, in both the case of the ascending case and the case of the descending case, the index signal β monotonously decreases when "not exceeding at all", and the index signal β satisfies the determination condition "β (previous) > β (current)". If it is determined in step S70 that β (previous) > β (current) is present, the routine proceeds to step S80, and if it is determined that β (previous) > β (current) is not present, the routine returns to step S30.

For example, in the case of the rising case shown in fig. 12, the index signal β (the previous time) and the index signal β (the present time) are calculated in a predetermined period including the time t1, which is fixed in a state of θ1=θse (=θ1_fix). The calculated index signal β (previous time) satisfies β (previous time) > β (current time). Accordingly, the determination in step S70 is yes (yes) and the process proceeds to step S80.

In step S80, the value of m is increased by +1. In step S90, it is determined whether or not the number m of times determined as β (previous) > β (current) has reached the upper limit value N, and if so (m=n), the process proceeds to step S100. If it is not (m < N), the routine returns to step S40, and the processing from step S40 to step S90 is repeatedly performed until m=n. The upper limit value N is a natural number. In the above example, n=10 to 40, but the present invention is not limited to this, and the index signal β may be generated at the minimum value of 1.

That is, when the fixed opening θ1_fix is the opening setting for the condition that the predicted pressure Pp "does not exceed" the pressure target value Ps at all, the processing of steps S40 to S90 is repeatedly executed, and the opening changing processing of step S100 is executed at the time point when m=n is determined in step S90 (time t2 in fig. 12). On the other hand, when the fixed opening θ1_fix is the opening setting for the condition that the predicted pressure Pp "exceeds" the pressure target value Ps, the processing of steps S30 to S80 is repeatedly executed, and finally, it is determined in step S50 that Pp > Ps, and the opening changing processing of step S100 is executed.

(opening degree changing process)

In the opening degree changing process, the opening degree target estimated value (here, this value is denoted as θse2) is calculated again by the estimation computing unit 210 of fig. 2, and the opening degree setting θ1 output from the feedforward controller 220 is set to θ1=θse2. For example, in fig. 12 for the case "no exceeding" in the rising case, the opening degree setting θ1 is changed to θse2 at time t2, and the pressure response of the pressure measurement value Pr changed as indicated by the line L1 up to that indicated by the line L2. When it is determined that the value exceeds in step S50 and the flow proceeds to step S100, the illustration of the pressure response and the like as shown in fig. 12 is omitted, and the calculated opening target estimated value θse2 becomes θse2 > θse.

In step S110, it is determined whether or not a skip condition from the open control to the close control is satisfied, and if the condition is satisfied, the open control is ended and the control is skipped to the close control. As the jump condition, for example, conceivable are: the case where the pressure measurement value Pr sufficiently approaches the pressure target value Ps so that the difference between Pr and Ps becomes 5% or less of the magnitude of Ps, or the case where the index signal β changes from monotonically decreasing to monotonically increasing, etc.

When the control is switched from the on control to the off control at time t3 in fig. 12, a predetermined opening degree setting θse2 is output from the feedforward controller 220, and the opening degree setting θ2 is output from the feedback controller 230. After the closing control is started, the pressure response of the pressure measurement value Pr changes as indicated by the line L3, and the pressure measurement value Pr converges almost to the pressure target value Ps at time t 4.

Further, although the opening θs is an opening at which the pressure balance value becomes the pressure target value Ps, when the opening setting θ1 (=θse) is θ1θs with respect to the opening θs, both "exceeding" and "not exceeding at all" may not be determined. In this case, for example, the time point at which the n_timeout secondary index signal β is calculated after the opening degree setting θ1 is fixed, that is, the time point at which the time n_timeout×Δt elapses when the calculation period Δt (for example, 10ms described above) is set, and the control may be changed from the on control to the off control. Where N_timeout > N.

Fig. 13 is a diagram showing a comparative example, and shows a case where the predicted pressure Pp is used as an index for determination in the case of "no exceeding at all". The determination that the fixed opening degree is ultimately the opening degree Pp < Ps using the predicted pressure Pp is not determined until the predicted pressure Pp sufficiently converges to the pressure balance value (time t11 in fig. 13). Therefore, the response of the pressure measurement value Pr stalls during this period, and the pressure adjustment time for bringing the pressure measurement value Pr closer to the pressure target value Ps after the jump to the closed control at time t11 becomes longer.

On the other hand, when the index signal β is used, it is possible to determine that "no exceeding at all" is performed at an early stage of the rising process of the pressure measurement value Pr as in time t2 of fig. 12. Therefore, the time from when the opening degree setting θ1 is fixed to θse until the pressure measurement value Pr converges to the pressure target value Ps can be shortened as compared with the case where the predicted pressure Pp is used. Although the predicted pressure Pp is not shown in fig. 12, the predicted pressure Pp is also rising at time t 2.

(method for calculating the target estimated value θse of the opening degree)

An example of a method of calculating the opening degree target estimated value θse will be described. Here, a description will be given of a calculation method for obtaining an opening target estimated value θse using a controlled device gain Gp, which is a correlation between the opening θ and the pressure P. The controlled device gain Gp is defined by the following equation (20). Further, | (Δp/Δθ) | represents the absolute value of (Δp/Δθ).

Gp＝|(ΔP/Δθ)|/P···(20)

As is clear from the equation (20), when the opening θ is set as an input and the pressure P of the vacuum chamber 3 is set as an output, the pressure change with respect to the opening change, that is, the amount ((Δp/Δθ)/P) by which (Δp/Δθ) is normalized by the pressure P, represents the static gain characteristic of the pressure response of the lean gas in the vacuum chamber 3 as the controlled apparatus. The controlled device gain Gp (θ) is stored in the storage unit 23 of fig. 1 as a data table (θ, gp) showing a relationship between the opening θ and the controlled device gain Gp, for example.

If the pressure adjustment state is represented as (θ, P), the pressure adjustment state before the change of the opening degree is represented as (θse, pr), and the target pressure adjustment state is represented as (θse2, ps). When the definition formula of the formula (20) is used, the relationship between the opening change Δθ and the pressure change Δp is expressed as in the formula (21).

ΔP＝-P×Gp(θ)×Δθ···(21)

The relationship between the voltage regulation state (θse, pr) and the voltage regulation state (θse2ps) is represented by the following formulas (22) and (23).

θse2=θse+dθ (integration range θse to θse2) · (22)

Ps=Pr+ -dP (integration range Pr to Ps) · (23)

The integration of the expression (22) and the expression (23) is actually represented by adding the increase Δθ of the opening degree and the increase Δp of the pressure in each quasi-static state from the pressure regulating state (θse, pr) to the pressure regulating state (θse2, ps), and thus, the θse2 and Ps can be represented by the following expressions (24) and (25). Further, Δθ_n (θ_n) in the expression (24) is an opening increment Δθ_n when the opening θ_n is expressed as, for example, an opening increment when Δθ_1 (θ_1) is changed from the opening θse to the opening θ_1. Similarly, Δp_n (p_n) represents the pressure increase at the pressure p_n.

θse2＝θse+ΣΔθ_n(θ_n)···(24)

Ps＝Pr+ΣΔP_n(P_n)···(25)

As for Δθ_n (θ_n) of the expression (24), the increment Δθ_n (θ_n) may be provided according to each opening θ_n as in the expression (26). Regarding Δp_n (p_n) of the expression (25), the relationship between the opening θ and the pressure P may be used to provide an increment Δp_n (p_n) at the pressure p_n corresponding to the increment Δθ_n (θ_n) at the opening θ_n. When the controlled device gain Gp is used as the correlation, the increment Δp_n=Δp_n (p_n) at the pressure p_n is expressed by the following expression (27) to which the expression (21) is applied. In equation (27), Δθ_n=Δθ_n (θ_n), gp_n=gp (θ_n).

Δθ_n＝Δθ_n(θ_n)···(26)

ΔP_n＝ΔP_n(P_n)＝-P_n×Gp_n×Δθ_n···(27)

When the expression (26) and the expression (27) are substituted into the expression (24) and the expression (25) and the integration operation is performed, the opening value θse2 and the pressure target value Ps in the pressure adjustment state (θse2, ps) can be substantially reached. Further, if the expression (24) and the expression (25) are modified as described below, the operations can be performed sequentially.

θ_n+1＝θ_n+Δθ_n···(28)

P_n+1＝P_n+ΔP_n···(29)

The opening θ is integrated by the formulas (26) and (28), and the pressure P is integrated by the formulas (27) and (29) in order. Then, if the value "p0+ΣΔp_n (p_n)" on the right of the equation (25) as the integrated calculation value of the pressure reaches the next pressure value (here, the pressure target value Ps) (or if the next pressure value is exceeded), the calculation is ended. The opening degree calculated by the integration when the pressure target value Ps is reached corresponds to the target opening degree estimated value θse2. Further, the present method is useful for obtaining a target opening degree estimation value because a practical approximation value can be obtained even in a state where the static state cannot be said.

Modification 1

Fig. 14 is a flowchart for explaining modification 1 of the embodiment. In the above embodiment, even when "no more than" is performed, the opening degree is changed only once by the opening degree changing process, and the control is switched to the closing control. However, the following occurs: the target opening degree estimation value θse is deviated from the pressure target value Ps due to the calculation error, and the pressure measurement value Pr cannot be brought into the vicinity of the pressure target value Ps by one opening degree change. Therefore, in the control process shown in the flowchart of fig. 14, step S200 is added to the flowchart shown in fig. 11, and the opening degree change based on the index signal β is performed up to the predetermined upper limit number of times.

In the rising case shown in fig. 15, determination of the index signal β is performed twice, and the opening degree setting θ1 is changed to θse2 and θse3. At this time, the target opening degree estimated value θse2 calculated after the first determination is larger than the target opening degree θs, and the index signal β also monotonously decreases after the change. The target opening degree estimated value θse3 calculated after the second determination is smaller than the target opening degree θs, and the index signal β after the change is changed to monotone increase. Therefore, it is determined in step S110 of fig. 14 that the skip condition (β is monotonically decreasing→monotonically increasing) is satisfied, and the control is skipped from the on control to the off control. For example, if the upper limit number of times of step S200 is twice, θse3 > θs in fig. 15, it is determined that yes (yes) is present in step S200 before yes (yes) is present in step S110, and the control jumps from the on control to the off control.

Modification 2

In the flowchart of fig. 11 in which the predicted pressure Pp is used in combination with the index signal β, the predicted pressure Pp is used for "exceeding" determination as in step S50. However, the determination may be made using the index signal β even when "exceeding". As is clear from fig. 10, when the "exceeding" is determined by the index signal β, the "exceeding" is determined when the index signal β monotonically increases. At this time, in modification 2, the flowchart of fig. 11 is modified as shown in fig. 16, for example.

If it is determined in step S10 of fig. 16 that the value of the opening degree measurement value θr is the opening degree setting θ1_fix and the index signal β is calculated in step S20, it is set to m1=0 and m2=0 in step S300. Next, when the index signal β is calculated in step S310, in step S320, it is determined whether or not the predicted pressure Pp "does not exceed" the pressure target value Ps at all, that is, whether or not β (previous) > β (current). If it is determined in step S320 that β (previous) > β (current), the flow advances to step S330 to increase m1 by 1 and to obtain m1=m1+1. In step S340, it is determined whether or not m1 indicating the number of times of β (previous) > β (current) has reached the upper limit value N1, and the process proceeds to step S100 when it has reached (m1=n1), and returns to step S310 when it has not reached (m 1 < N1).

On the other hand, when it is determined in step S320 that "β (previous) > β (current)" is not found, the routine proceeds to step S322, where after m1=0 is set, and then, in step S324, it is determined whether "exceeding", that is, whether "β (previous) < β (current)", is found. When it is determined in step S324 that "β (previous) < β (current)", the flow advances to step S326 to increase m2 by 1 to m2=m2+1, and then to step S328. On the other hand, when it is determined in step S324 that "β (previous) < β (current)" is not satisfied, the routine proceeds to step S325, where m2=0 is set, and thereafter, the routine returns to step S310. In step S328, it is determined whether or not m2 indicating the number of times β (previous time) < β (current time) has reached the upper limit value N2, and the process proceeds to step S100 when it has reached (m2=n2), and proceeds to step S310 when it has not reached (m 2 < N2).

As described above, when the processing of steps S310 to S340 is repeated and the number of times "β (previous) > β (current)" is determined to reach the upper limit value N1, or when the processing of steps S310 to S328 is repeated and the number of times "β (previous) < β (current)" is determined to reach the upper limit value N2, the opening degree changing process is performed in step S100. Then, in step S110, it is determined whether or not a skip condition from the open control to the close control is satisfied, and if the condition is satisfied, the open control is ended and the control is skipped to the close control.

As described above, the predicted pressure Pp is a pressure estimated value after t seconds from the time point of measuring the pressure measurement value Pr, and depends on an estimated value of the effective exhaust speed Se or the flow Qin based on the reference gas. In general, the gas type of the reference gas is often different from the gas type of the gas actually flowing, and thus the error in the predicted pressure Pp may be large.

On the other hand, as is clear from the definition formula (1) of the index signal β, the index signal β is calculated based on the pressure target value Ps and the actually measured pressure measurement value Pr, and the index signal β reflects the actual gas type or gas flow rate. Therefore, the reliability of the determination can be improved as compared with the determination of "exceeding" using the predicted pressure Pp.

Modification 3

In the above-described embodiment, or in modification examples 1 and 2, description has been made taking as an example the case where both the "exceeding" and the "not exceeding" occur. However, when the opening degree change amount under the opening control is not so large as to change from the fourth quadrant to the third quadrant in fig. 4, a flowchart corresponding to only the case of "no more than" as shown in fig. 17 may be employed. The flowchart shown in fig. 17 eliminates the processing of step S40 and step S50 from the flowchart shown in fig. 14. If the processing from step S30 to step S90 is repeated N times and it is determined that the number of times is not exceeded, the opening degree changing processing of step S100 is performed.

In addition, as in the case of the flowchart of fig. 14, a case is considered in which the pressure measurement value Pr cannot be brought into the vicinity of the pressure target value Ps by one opening degree change, and the opening degree change based on the index signal β is performed up to a predetermined upper limit number of times as in step S200. When the opening setting θ1 (=θse) is θ1θs and it is not determined that the value does not exceed the value at all, the operation is switched from the on control to the off control at a point in time when the n_timeout sub-index signal β is calculated after the opening setting θ1 is fixed.

Modification 4

In the first embodiment, the index signal β (= (dPr/dt)/(ps—pr)) defined by the formula (1) is introduced as an index signal instead of the predicted pressure Pp. However, the index signal used in place of the pre-pressure measurement Pp is not limited to the index signal defined by the formula (1), and various similar signals described below can be applied.

Index Signal β1 normalized to a dimensionless quantity

Based on a data set se_r (θr) representing the volume V of the vacuum chamber 3, the effective evacuation speed Se, and the opening measurement value θr, an index signal β1 defined by the following expression (30) is introduced. By normalizing to the dimensionless quantity, the versatility of the determination threshold value can be improved even when conditions such as the gas flow rate value are changed.

β1＝(V/Se_r(θr))×[(dPr/dt)/(Ps-Pr)]···(30)

Index signal β2 accurately positive

In principle, both the rising case and the falling case are judged in a range where "(dPr/dt)/(Ps-Pr)" becomes positive. However, by providing absolute values of both the denominator and the numerator as in the equations (31) and (32) in response to disturbance such as sudden flow rate change, an index signal β2 that is accurately positive can be obtained.

β2= | dPr/dt|/|ps-pr|· (31) or

β2＝(V/Se_r(θr))×|dPr/dt|/|Ps-Pr|···(32)

Those skilled in the art will appreciate that the various illustrative embodiments and modifications described above are specific examples of the following aspects.

[1] A valve control device according to one aspect controls an opening degree of a vacuum valve attached to a vacuum chamber by an opening control, and includes: an opening setting unit for outputting an opening setting value of the opening control; a first determination unit configured to determine that a pressure balance value of a case where the opening degree setting value is a fixed value does not exceed a pressure target value of valve control in a predetermined period within a fixed period; and an opening degree control unit that controls the opening degree based on a determination result of the first determination unit.

For example, as shown in fig. 17, in a predetermined period including time t1, in which the opening degree set value θ1 is equal to the fixed value θ1_fix, it is determined that the pressure balance value Pe (θ1_fix) does not exceed the pressure target value Ps based on the calculated index signal β (the previous time) and the calculated index signal β (the present time). Therefore, it is possible to judge that the pressure value does not exceed the pressure target value Ps as early as possible without waiting for the predicted pressure value P or the pressure measurement value Pr to converge to the pressure balance value that does not exceed the pressure target value Ps, and by performing further opening degree control based on the judgment result, it is possible to achieve shortening of the pressure adjustment time.

[2] The valve control device according to item [1], further comprising: and a second determination unit configured to determine that the pressure balance value exceeds the pressure target value or that the predicted pressure of the vacuum chamber exceeds the pressure target value after a predicted time of the predetermined value has elapsed, wherein the opening control unit controls the opening based on determination results of the first determination unit and the second determination unit.

For example, as shown in fig. 12, in a predetermined period including time t1, in which the opening degree set value θ1 is equal to the fixed value θ1_fix, it is determined that the pressure balance value Pe (θ1_fix) does not exceed the pressure target value Ps based on the calculated index signal β (the previous time) and the calculated index signal β (the present time). In addition, in the predetermined period, it is determined that the pressure balance value Pe (θ1_fix) exceeds the pressure target value Ps based on the index signal β (the previous time) and the index signal β (the present time) as in step S324 of fig. 16, or it is determined that the predicted pressure Pp exceeds the pressure target value Ps when the predicted pressure Pp becomes Pp > Ps as in step S50 of fig. 11.

Therefore, it is possible to determine whether the pressure value exceeds or does not exceed the pressure balance value of the pressure target value Ps as early as possible without waiting for the convergence of the predicted pressure Pp or the pressure measurement value Pr, and further opening degree control is performed based on the determination result, whereby the pressure adjustment time can be shortened.

[3] The valve control device according to item [2], wherein the first determination unit determines that the pressure balance value does not exceed the pressure target value based on a ratio of a time rate of change of the pressure measurement value to a difference value between the pressure measurement value and the pressure target value, and the second determination unit determines that the pressure balance value exceeds the pressure target value based on the ratio.

The index signal β defined by the expression (1) is a ratio of a time rate of change (dPr/dt) of the pressure measurement value Pr to a differential value (ps—pr) of the pressure measurement value Pr and the pressure target value Ps. By setting to such an amount, the ratio (i.e., the index signal β) monotonously increases in the case of "exceeding" and monotonously decreases in the case of "not exceeding at all". Therefore, by studying whether the ratio increases monotonically or decreases monotonically in the predetermined period within the fixed period, it can be determined that the ratio does not exceed the pressure target value Ps even if the pressure measurement value Pr does not converge to the pressure target value Ps in the predetermined period.

Further, the expression (1) is defined as (dPr/dt)/(ps—pr), but even if the index signal is an index signal in which the denominator and the numerator are inverted, the index signal is monotonically decreased when exceeded and monotonically increased when not exceeded, and therefore, it can be determined that the index signal is not exceeded by examining whether the index signal is monotonically increased or monotonically decreased.

[4] The valve control device according to item [3], wherein when the pressure measurement value is Pr, the pressure target value is Ps, and the time derivative of the pressure measurement value is (dPr/dt), the ratio is set to "(dPr/dt)/(Ps-Pr)". By setting the ratio in this way, it is possible to determine that the ratio is not exceeded by studying whether the ratio increases monotonically or decreases monotonically in a predetermined period within the fixed period.

[5] The valve control device according to any one of [2] to [4], comprising: and an estimating unit that estimates a target opening degree estimated value corresponding to a pressure target value based on the opening degree set value and a pressure measured value at the time of the opening degree set value, wherein the opening degree control unit controls the opening degree to the target opening degree estimated value based on the determination results of the first determining unit and the second determining unit.

For example, as in the process shown in fig. 11, when Pp > Ps is determined in step S50, or when "β (previous) > β (current)" is determined N times in step S70 and step S90, the opening degree setting θ1 is set to the target opening degree estimated value θse estimated by the estimation operation unit 210, and the opening degree of the valve body 12 is controlled to the target opening degree estimated value θse. As a result, the pressure measurement value Pr can be made closer to the pressure target value Ps.

[6] A vacuum valve of one form includes: a valve body; a valve body driving unit for driving the valve body; and controlling the opening degree of the valve body by controlling the valve body driving unit according to any one of [1] to [5 ]. This can further shorten the pressure adjustment time by the vacuum valve.

While various embodiments and modifications have been described above, the present invention is not limited to these. Other forms conceivable within the scope of the technical idea of the present invention are also included in the scope of the present invention.

Claims (6)

1. A valve control apparatus that controls an opening degree of a vacuum valve mounted to a vacuum chamber by opening control, the valve control apparatus comprising:

an opening setting unit for outputting an opening setting value of the opening control;

a first determination unit configured to determine that a pressure balance value of a case where the opening degree setting value is a fixed value does not exceed a pressure target value of valve control in a predetermined period within a fixed period; and

an opening degree control unit that controls the opening degree based on a determination result of the first determination unit,

the first determination unit determines that the pressure balance value does not exceed the pressure target value based on a ratio of a time rate of change of the pressure measurement value to a difference value between the pressure measurement value and the pressure target value.

2. The valve control device according to claim 1, further comprising:

a second determination unit configured to determine that the pressure balance value exceeds the pressure target value or that a predicted pressure of the vacuum chamber after a predicted time based on the predetermined value has elapsed exceeds the pressure target value,

the opening degree control unit controls the opening degree based on a determination result of the first determination unit and the second determination unit.

3. The valve control device according to claim 2, wherein,

the second determination unit determines that the pressure balance value exceeds the pressure target value based on the ratio.

4. The valve control device according to claim 3, wherein,

when the pressure measurement value is Pr, the pressure target value is Ps, and the time differential of the pressure measurement value is (dPr/dt), the ratio is set to "(dPr/dt)/(Ps-Pr)".

5. The valve control device according to any one of claims 2 to 4, wherein:

an estimating unit for estimating a target opening degree estimated value corresponding to a pressure target value based on the opening degree set value and a pressure measured value at the opening degree set value,

the opening control unit controls the opening to the target opening estimated value based on the determination results of the first determination unit and the second determination unit.

6. A vacuum valve, comprising:

a valve body;

a valve body driving unit for driving the valve body; and

the valve control device according to any one of claims 1 to 4, wherein the valve body driving portion is controlled to control the opening degree of the valve body.