WO2002086631A1

WO2002086631A1 - Control method and control apparatus

Info

Publication number: WO2002086631A1
Application number: PCT/JP2002/003785
Authority: WO
Inventors: Shigeru Kasai; Shuji Kondo
Original assignee: Tokyo Electron Limited
Priority date: 2001-04-17
Filing date: 2002-04-16
Publication date: 2002-10-31

Abstract

A control method uses a controller for controlling a predetermined process parameter by a predetermined control parameter. The method comprises a step of determining an optimal control parameter value for each control step, a step of selecting a control parameter value corresponding to an actual setting condition change of the controller from the determined control parameter values, a step of changing the control parameter value set in the controller to the control parameter value selected in the selection step, and a step of output a control signal based on the changed control parameter from the controller.

Description

Description Control method and control device Technical field

The present invention relates to a control device and a control method for controlling a process parameter such as a chamber pressure during a film forming process. Background technology

For example, in a semiconductor manufacturing process, various vacuum processes such as a film forming process and an etching process are performed on a semiconductor wafer (hereinafter, simply referred to as a wafer) to be processed. At the time of each vacuum processing, it is necessary to introduce a predetermined gas into the chamber containing the wafer and to strictly control the pressure inside the chamber.

As a general pressure control method in such a case, the pressure in the chamber

-The measured value is fed back and the measured value is fed back to control the opening of the exhaust valve by PID.

In conventional pressure control by PID control, a combination of PID parameter overnight values is determined by trial and error, and the PID parameter overnight value of that single combination (K _P : proportional gain, Τ ：: integration time, Τ. : Derivative time).

However, the relationship between the opening of the exhaust knurl and the conductance is not a simple proportional relationship. In addition, the relationship changes depending on changes in gas flow rate and control pressure. For this reason, there is a limit to the control based on the PID parameter overnight value of a single combination as described above, that is, it is difficult to perform high-precision control. On the other hand, finding the optimal PID parameter overnight value for each predetermined condition is extremely complicated.

Such problems can occur with other process controls as well, regardless of the control scheme. Summary of the invention

The present invention has been made in view of such circumstances, and it is an object of the present invention to provide a control method and a control device capable of performing process control with high accuracy according to various process conditions without complicating the present invention. Aim.

In order to solve the above problems, a first aspect of the present invention provides a control method using a controller for controlling a predetermined process parameter by a predetermined control parameter, wherein a control method is provided for each control step in advance. A step of obtaining an optimum control parameter value; a step of selecting a control parameter value corresponding to an actual setting condition change for the controller from the plurality of obtained control parameter values; and setting the controller value. Changing the set control parameter value to the control parameter value selected in the selection step, and setting a control signal based on the changed control parameter value from the controller to a control target. And a step of outputting.

In the present invention, as described above, an optimal control parameter value for each control step is determined in advance, and a control corresponding to a change in an actual setting condition for the controller is obtained from the determined control parameter values. Select the parameter value, change the control parameter value set in the controller to the selected control parameter value, and control the control signal based on the changed control parameter value. Since the control is performed by outputting from the heater to the control object, the control can always be performed with appropriate control parameters when the setting conditions change, and extremely high-precision control can be performed. In addition, since the optimal control parameter value for each control step is determined in advance and tabulated, for example, and an appropriate control parameter value is automatically selected from the table, the complexity may be increased. Absent.

Preferably, the control by the controller is PID control or optimal regulation overnight control.

Alternatively, according to a second aspect of the present invention, there is provided a control method using a controller for controlling a predetermined process parameter by PID control based on a preset setting condition and a fed back process condition. A step of previously calculating an optimum PID parameter value for each control step; and a P-D parameter corresponding to a change in an actual setting condition for the controller from the plurality of obtained PID parameter values. Selecting the evening value; and selecting the overnight PID parameter value set in the controller. Changing the PID parameter value selected in the selecting step, and outputting a control signal based on the changed PID parameter value from the controller to a control target. Control method.

Preferably, in the step of previously determining an optimum PID parameter value for each control step, a response waveform of a process parameter when a control target is changed is obtained for each control step, and each response waveform is obtained. From this, the transfer characteristics are identified using the dead time + nth-order delay model, and the PID parameter overnight value is determined from the identified transfer characteristics by a predetermined method.

Preferably, when the controlled object is non-linear with respect to the controlled process parameter, a linear control system is formed using a non-linear compensation element for a control signal output from the controller. Is further provided.

Alternatively, in the third aspect of the present invention, the opening degree of the exhaust knob is controlled by PID control based on a preset set value of the processing vessel pressure and a feedback measured value of the processing vessel pressure. A control method that uses a controller to control the pressure inside the processing vessel by using a controller to determine the optimum PID parameter values corresponding to various changes in the set pressure in advance, and a method for determining a plurality of obtained PID parameters. A step of selecting a PI parameter value corresponding to an actual set pressure change for the controller from the evening value; and a PID parameter value set in the controller, the PID selected in the selection step. Changing the parameter value to an overnight value, and outputting a control signal based on the changed PID parameter value to the exhaust valve from the controller. is there.

Preferably, in the step of previously obtaining the optimum PID parameter value corresponding to various changes in the set pressure, the response waveform of the internal pressure of the processing vessel when the degree of opening of the exhaust valve is changed is determined by the response waveform of each set pressure. The transfer characteristic is determined for each change. From each response waveform, transfer characteristics are identified using a delay time + nth-order delay model, and a PID parameter overnight value is determined from the identified transfer characteristics by a predetermined method.

Preferably, the relationship between the degree of opening of the valve and the pressure in the processing vessel is measured in advance, and the valve is opened using a non-linear compensation element with respect to a control signal output from the controller. To ensure that the relationship between the pressure and the pressure in the processing vessel is linear. Step is further provided.

Also, preferably, in the step of previously determining the optimum PID parameter overnight value corresponding to various changes in the set pressure, the optimum PID parameter overnight value is also calculated in consideration of the set flow rate.

Further, in the above description, the PID control may be a normal PID control, an I-PD control, or a differential leading PID control. In this case, preferably, the control signal based on the changed PID parameter overnight value uses an offset value as an initial integration value.

Further, according to a fourth aspect of the present invention, there is provided a control device for controlling a predetermined process parameter by a predetermined control parameter, based on a preset setting condition and a fed-back process condition. A control unit for controlling the control object by a predetermined control parameter, a calculating means for calculating an optimum control parameter value for each control step, and an optimum value for each control step obtained by the calculating means. Storage means for storing a control parameter value; a change in a setting condition for the controller is input; and a control parameter value corresponding to the change in the setting condition is obtained from information stored in the storage means. Control parameter selection means for selecting and outputting to the controller a command to change the control parameter value to the selected control parameter value. A control device that the symptoms.

ADVANTAGE OF THE INVENTION According to this invention, when a setting condition changes, it can always control with an appropriate control parameter overnight value, and can perform extremely high-precision control. In addition, since the optimal control parameter value for each control step is determined in advance and the appropriate control parameter value is selected from the table, for example, an appropriate control parameter value is selected.

Preferably, the controller controls the control target by PID control or optimal regulation control.

Alternatively, according to a fifth aspect of the present invention, there is provided a control device for controlling a predetermined process parameter by pJD control based on a preset setting condition and a fed back process condition, A controller that controls the control target based on PID parameter values based on preset setting conditions and feedback process conditions, and an operation that calculates the optimal PID parameter value for each control step Calculation means; storage means for storing an optimum PID parameter value for each control step obtained by the calculation means; and changes in setting conditions for the controller are inputted and stored in the storage means. A PID parameter value corresponding to the change of the setting condition is selected from the information, and the PID parameter overnight value is selected for the controller: a control parameter for outputting a command to change to the PID parameter overnight value. A control device comprising: evening selection means;

Preferably, the calculation means obtains a response waveform of a process parameter when the control target is changed for each control step, and calculates a transfer characteristic of the transfer characteristic from each response waveform using a dead time + n-order delay model. Identification is performed, and the PID parameter overnight value is determined from the identified transfer characteristics by a predetermined method.

Alternatively, in the sixth aspect of the present invention, the pressure in the processing vessel is controlled by PID control based on a preset set value of the pressure in the processing vessel and a measured value of the pressure in the processing vessel fed back. A control device for controlling the opening of the exhaust valve by a pID parameter based on a preset set value of the pressure in the processing vessel and a fed back process condition; and A calculating means for calculating an optimum PID parameter value corresponding to the change of the set pressure, and an optimum PID parameter value corresponding to various changes of the set pressure obtained by the calculating means. Storage means, and an actual set pressure change for the controller is input, and a PID parameter value corresponding to the set pressure change is selected from information stored in the storage means, and the controller Control parameter selection means for outputting a command to change the PID parameter value to the selected PID parameter value.

Preferably, the calculation means obtains a response waveform of the pressure in the processing container when the opening degree of the exhaust valve is changed for each change in the set pressure, and uses a dead time + n-order model from each response waveform. The transfer characteristics are identified by using the specified transfer characteristics, and a PID parameter overnight value is obtained from the identified transfer characteristics by a predetermined method. Preferably, the apparatus further includes a compensating means for compensating the relationship between the opening degree of the valve and the pressure in the processing container to be linear by using a non-linear compensation element for a control signal output from the controller. I do.

Preferably, the calculating means calculates an optimum PID parameter—evening value in consideration of a set flow rate.

Further, in the above description, the PID control by the controller may be normal PID control, I-PD control, or differential leading PID control. In this case, preferably, the control signal based on the changed PID parameter overnight value uses an offset value as an initial integration value. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a cross-sectional view showing a film forming system provided with a control device used for carrying out the present invention.

FIG. 2 is a flowchart for explaining the steps of one embodiment of the present invention. FIG. 3 is a diagram showing an example of a table for performing a step response test of a pressure value with respect to various valve opening changes.

FIG. 4 is a diagram showing an example of a fitting result in auto tuning.

FIG. 5 is a diagram showing a pulp opening degree-pressure curve.

FIG. 6 is a diagram illustrating an example of a relationship between a gas flow rate and a maximum ultimate pressure.

FIG. 7 is a block diagram of a normal PID control system.

FIG. 8 is a block diagram of the PI-D control system.

FIG. 9 is a block diagram of the IP control system. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be specifically described with reference to the accompanying drawings. FIG. 1 is a cross-sectional view showing a film forming system provided with a control device used for carrying out the present invention. As shown in FIG. 1, the film forming system 100 includes a film forming apparatus 1 and a control apparatus 2. The film forming apparatus 1 has a cylindrical chamber 11. A mounting table 12 on which the semiconductor wafer W is mounted is provided at the bottom of the chamber 11. Inside the mounting table 12, a heat sink 13 is buried. A shower head 14 is provided above the mounting table 12 so as to face the upper surface of the mounting table 12. A gas supply mechanism 16 for supplying a gas for film formation is connected to the shower head 14 via a gas line 15. The gas line 15 is provided with a valve 17 and a mass flow controller 18. Then, the deposition gas supplied from the gas supply mechanism 16 passes through the gas line 15 to the shower head 14, and a large number of gases formed on the lower surface of the shower head 14. The liquid is discharged from the discharge holes 14a. An exhaust pipe 21 is provided at the bottom of the chamber 11. An exhaust valve (variable conductance valve) 22 is attached to the exhaust pipe 21. The pressure in the chamber 11 is adjusted by adjusting the degree of the exhaust valve 22.

The control device 2 includes a pressure sensor 3 1 for detecting the pressure in the chamber 1 1 1, a pressure gauge 3 2 for calculating the pressure from a detection signal of the pressure sensor 3 1, a pressure value of the pressure gauge 3 2, and a set pressure. A PID controller 33 that controls the degree of exhaust valve 22 based on the above, and a setter 34 that outputs a set pressure to the PID controller 33.

In the present embodiment, as the constituent elements of the control device 2, an arithmetic device 35, an input device 36 for inputting data to the arithmetic device 35, and a memory 13 for storing the arithmetic result of the arithmetic device 35 are further included. 7 and. From the input device 36, response characteristics for each set pressure change value, which will be described later, are input to the arithmetic device 35. The calculating device 35 calculates an optimum PID parameter value at each set pressure change value. Then, the calculation result by the calculation device 35 is input to the memory 37. In the memory 37, these data can be stored in a tabulated state.

In actual control, the previously set pressure and the set pressure are input from the setter 34 to the arithmetic unit 35. Then, the arithmetic unit 35 first calculates a change in the set pressure. Next, the arithmetic unit 35 compares the change value with the table information stored in the memory 37 and, based on the data in the memory 37, corresponds to the change value (set pressure change). Select the PID parameter overnight value. Then, the arithmetic unit 35 outputs a command to the PID controller 33 to change the previously set control parameter value to the currently selected control parameter value. Further, in the present embodiment, there is provided a steady-state gain linearizer 38 for compensating the steady-state gain of the control signal output from the PID controller 33 so as to form a linear control system using a nonlinear compensation element. The set flow rate is also input to the stationary gain linearizer 38, and processing is performed in consideration of the information.

In the film forming system 100 configured as described above, the semiconductor wafer W is mounted on the mounting table 12 in the chamber 11 of the film forming apparatus 1. The semiconductor wafer W mounted on the mounting table 12 is heated to a predetermined temperature by the heater 13. At the same time, a predetermined deposition gas is supplied from the shower head 14 at a predetermined flow rate. As a result, a film forming process is performed on the semiconductor wafer W. In this case, the pressure in the chamber 11 is also controlled by the control device 2.

When controlling the pressure, if the pressure setting changes, the PID parameter overnight value of the PID controller 33 is optimized based on the previously set pressure and the set pressure. Then, the PID controller 33 outputs a control signal based on the optimized PID parameter overnight value. Hereinafter, an example of the pressure control flow of the present embodiment will be described with reference to the flowchart of FIG. This control flow consists of the optimal PID parameter overnight value stapling process (ST1), the PID parameter overnight value selection process (ST2), the PID parameter overnight value changing process (ST3), and the control signal output process (ST4 ), A control signal compensation step (ST5), and a control target control step (ST6).

First, in the optimal PID parameter overnight value table forming step (ST1), predetermined information is input to the arithmetic unit 35 from the input unit 36 in advance. Then, the arithmetic unit 35 performs fitting of the response characteristics for each set pressure change value, and obtains an optimum PID parameter value for each set pressure change value. The operation results are input to the memory 37. In the memory 37, these data are stored in a table.

In this step (ST1), a step response test of a pressure change value with respect to various valve opening changes is performed based on, for example, a template as shown in FIG. Also, When the flow rate changes, the characteristics of the pressure change value can change even in the step response between the same valve openings. For this reason, the step response test is performed by changing the flow rate. For example, a step response test is performed for two typical flow rates used in actual control. In the step response test, for example, a sampling cycle is set to 0.2 sec, and a time series data of 100 steps is taken. The data obtained in this way is input to the arithmetic unit 35 by the input unit 36. As a result, the arithmetic unit 35 automatically calculates the optimal PID parameter overnight value by the following method (auto tuning). The optimal PID parameter overnight value is stored in the memory 37.

First, the arithmetic unit 35 identifies the transfer characteristics using the model shown in equation (1). C

Multiple first-order lags: G (s) = ^e , (1)

(1 + Ts) ⁿ

Here, G (s) is the transfer function of the controlled object, K is the steady-state gain, L is the dead time, and T is the time constant. The step response to this equation (1) is given by the following equation (2).

With respect to the response waveform, using the above equation (2) as a model, the arithmetic unit 35 performs fitting using the nonlinear least squares method, and determines a dead time L and a time constant Τ. . Here, the order η of the transfer characteristic is determined so as to perform optimal fitting, and is given in advance. As a result of the experiment, the value of η was different for each response waveform.

Next, using the dead time L and the time constant Τ obtained by the fitting, the critical period T _c = 2π / ωο of the transfer characteristic is calculated by the non-linear equation shown in the following equation (3) by the bisection method. It is determined by solving. The range of the solution is 0 <w _c TT / L. Soshi Then, using the limit period _Tc , the limit sensitivity Kc is obtained by the equation (4). One-n tan ^-1 (Τω „) =-(3)

+

τ / 2

ω / κ (4)

As can be seen from Eq. (4), the critical sensitivity Kc is a value that depends on the steady-state gain K. Since the steady-state gain K is linearized by the steady-state gain linearizer 38 (described later), the one corresponding to the step response of 09 Odeg is used here. That is, when control is performed using the capacitance value of 10 Torr (1333 Pa), the value is given by the following equation (5).

2047x (p (Torr) / 10)

K P n..P max (5)

2047x (u (deg) / 90) u 90

Based on the above, for example, the improved limit sensitivity method (Kuwata: "Improved limit sensitivity method and PID: Characteristics of I-PD control", Transactions of the Society of Instrument and Control Engineers, Vol.20, pp.232-239, 19 87), the PID parameter overnight value (Kp: proportional gain,: integration time, Τ: differentiation time) for the transfer characteristic of each response waveform is obtained. As a result, a table of PID parameter overnight values can be created. In this method, KP,, TD are obtained by the following equations (6), (7), (8), respectively. ·

K _n = f _K (L _N ) xK _c (K, n, L, T) (6)

Ί = f _I (L _N ) xT _c (n, L, T) (7)

T _D = f _D (L _N ) xT _c (n, L, T) (8) where Kc and _Tc are the above equations (3) and (4) and T, respectively. = 2 TT / C _C , where f κ, f:, and f D are dimensionless in dead time As a function of the normalized dead time L _N ≡L / Tc, In _c This method is uniquely determined by a predetermined technique, f K, f I sf D values normalized dead time of L _N and the control object properties since the variable depending on the type, f _K, the value of f I sf D is the control of a conventional method by remote precision fixed is realized. f K f I sf D

As a method of uniquely determining the function as a function of N ≡ L / Tc, the closed loop transfer function H (s) is matched with a reference model M (s) whose characteristics are known, and the PID parameter overnight value (K _P : Proportional gain,: integration time, T _D : derivative time).

By the above method, the optimum PID parameter value is calculated for all the response conditions in FIG. This will create a table night. Furthermore, similar table data is created for different gas flow rates.

As an example of such auto-tuning, a description will be given of a step response from fitting to calculation of a PID parameter overnight value in a step response from 0 de g to 75 de g in FIG. As the film forming apparatus, a TiN film forming apparatus having the structure shown in FIG. 1 was used, and the gas flow rate was 0.75 LZmin. Figure 4 shows the fitting results. Here, the fitting order n in the above equations (1) and (2) was set to 6 (n = 6). In FIG. 4, it can be seen that the curve of equation (2) is well fitted to a plurality of measurement values indicated by dots. At this time, the dead time L and the time constant T, which are the fitting parameters, are

L = 0.359874,

T = 0.243904

Converged.

Next, these values are substituted into the above equation (3), and the equation (3) is solved by the dichotomy, and the limit period Tc {= 2π / ωο) = 3.476521 is obtained.

For the steady-state gain K, when the value corresponding to the valve opening of 9 O deg is calculated by the above equation (5), K = 0.200318. When this is substituted into the above equation (4), the critical sensitivity Kc is Kc = 8.504282. Next, from the value of the dead time L obtained as described above and the value of the limit sensitivity Tc, the standardized dead time L _N is L _N = 0.103516. Using this value, the above equations (6), (7), (8) F K,: f 、 and f D are obtained by Kitamori's partial model matching method as follows.

f _K = 0. 510081,

f I = 0. 364807,

f D = 0. 183519

By substituting these values and the values of Kc and _Tc into the above equations (6), (7) and (8), the final PID parameter value is obtained as follows.

Τι = 1.268260.

After the above-described optimal PID parameter overnight value tabulation process (ST1), in the PID parameter overnight value selection process (ST2), during actual control, the arithmetic unit 35 changes the previously set pressure from the previously set pressure to the set pressure. Then, the change value to is compared with the value stored in the memory 37, and the PID parameter value corresponding to the change value is selected. Specifically, for example, a PID parameter value corresponding to the combination that is the same as or closest to the current change value is selected from the table data corresponding to each change value shown in FIG.

As a specific example, when the target pressure value changes under a certain gas flow rate, the change in the knob opening degree with respect to this is estimated. _Assuming that the fully opened state of the valve is 0 deg, the fully closed state of the valve is 90 deg, and the maximum ultimate pressure under a certain gas flow rate is _P∞ax ,

For the pressures Pi, P2, and _{P3 for U2} and _U3 , the values normalized as _pn / _max are as shown in the graph of FIG.

Even if the above graphs are created for different gas flow rates, the same graphs will be obtained by normalization.

The valve opening degree-pressure curve obtained as described above can be curve-fitted by a ninth-order polynomial expressed by the following equation (9). By using the equation (9), it is possible to know the change in the rep opening degree with respect to the normalized pressure change.

Pn (9)

On the other hand, the change of Pmax for different gas flow rates is as shown in Fig. 6. This shows that Pmax changes almost linearly with the gas flow rate.

From the above, it is possible to estimate the change in valve opening degree → 2 when the target pressure value changes from Pi to P 2 in the entire gas flow rate range used in the processing equipment. Their to, thus the same as or opening variation → U ₂ as estimated by or its corresponding opening change in FIG close most (obtained from step response test) PID parameters Ichita The value is selected.

The above tuning is performed for two types of flow rates. Assuming that the two flow rates are F and F ₂ (Fi <F ₂ ), the table of flow rate F! Is used for the flow rate below these average values (Ft + F ₂ ) / 2, and the average value (F! + F ₂₎ Z2 or more flow in the flow rate can be as table F ₂ is used. Each flow, maximum ultimate pressure p _max in the valves fully closed in F _2, when tuning is completed, P ID parameter - is saved with evening value. If the flow rate at the time of control is one of the flow rates F i and F ₂ at the time of tuning, one of the stored values is used as p _max , but the flow rate at the time of control is If the flow rate is different from F ₂ , the value estimated from the relationship between the gas flow rate and the maximum reached pressure shown in FIG. 6 is used as p _max from the relation between FIG. 5 and FIG.

After the PID parameter overnight value is selected as described above, in the PID parameter overnight value changing step (PT3), the selected PID parameter overnight value is output to the PID controller 33. . Then, the PID parameter overnight set in the PID controller 33 so far is changed to the selected PID parameter overnight.

In the control signal output step (ST4), the PID controller 33 outputs a control signal based on the measured pressure and the set pressure using the changed PID parameter overnight value. Examples of the control method of the PID controller 33 include ordinary PID control, PI-D control (differential leading type PID control), and I_PD control.

The control law of normal PID control is given by the following equation (10), and its transfer function is given by the following equation (11). The block diagram of this control system is as shown in Fig. 7. In the equation (10), u: manipulated variable, e = (ry): deviation, r: target value, y: observation, and the first term on the right side is proportional action, the second term is integral action, The third term is Indicates differential operation. u (t) = K _P | e (t) + ije (t) dt + T _D ^, (10)

The control law of PI-D control is given by the following equation (1 2). This control differs from ordinary PID control in that the derivative action only affects the observable y. If a differential operation is performed on the target value when the target value changes in the form of a step function, the Derby function will be included in the operation amount, and a sharp change in the input u called the set point kick will occur. Can live. In PI-D control, to avoid this, the derivative action is performed only on the observable. The block diagram of this control system is as shown in FIG. u (t) (12)

The control law of I-PD control is given by the following equation (13). In I-PD control, in addition to differential motion, proportional motion affects only the observed quantity y. If the target value changes in the form of a step function and the proportional operation is performed on the target value, the step function is included in the manipulated variable, which may not be practically preferable. In I-PD control, in order to avoid this, the proportional action is also performed only on the observed quantity.

Generally, if the PID parameter is adjusted so that the controllability when the target value changes is good, a steady-state error tends to occur when a steady-state disturbance is applied. Conversely, if the adjustment is made so that the controllability against disturbance becomes good, the overshoot will increase when the target value changes. It is reported that the use of I-PD control can reduce the trade-off between the overshoot amount and disturbance suppression when the target value changes. The block diagram of this control system is as shown in FIG. u (t) = K _P f- y (t) + ife (t) dt-T _n (13)

1 丄) J

Also, in the actual pressure control using the above PID control, PI_D control or I-PD control, the time to reach the target pressure value can be further reduced by the following method It is.

That can be estimated from the relationship shown in FIGS. 5 and 6, when you change the target pressure value is p ₂ the valve opening change → 2. Therefore, when the target pressure value is changed to p _2, the value ιΐ2 of valve opening (10) below, (12) Shikima other are give as the initial integral value of (13).

That is, the integral of these equations is

: E (t) is the indefinite integral of j¾ (t) dt

: C is the integration constant where:

K p C = u ₂

It is Ti.

For example, in a digitally controlled system, when the manipulated variables are output in a time-sharing manner, the integral manipulated variables at t = tl, t2, t3, ... are as follows. Integral manipulated variable at manipulated variable 0 t = tl 'E (tl) + u ₂ t E (tl)} + u ₂

Integral manipulated variable at t = t3 'fe (t3)-E (t2)} + u ₂

I

(The same applies hereinafter) As a result, the valve operation amount at time t = t1 has a certain large value (offset value) regardless of the integral value within a short time.

U

In this way, the valve opening at the start of control is very close to the valve opening at the time when the target pressure value is achieved, and the time to reach the target pressure value is greatly reduced. obtain.

Now, in the control signal compensation step (ST5), the control signal output from the PID controller 33 is compensated by the stationary gain linearizer 38 to the linear control system as described above. That is, as shown in FIG. 5, the steady-state gain K = p / u changes according to the valve opening, but such a steady-state gain variation is caused by the steady-state gain linearizer 38 at 90 ° in FIG. Compensation can be made with an inverse map from the value to the valve opening.

Thus, the controlled object can be modeled as having a constant gain (n-fold first-order lag + dead time system). That is, when the actual control, P ID controller 33 regards the control Target origin (0, 0) in FIG. 5 and (90deg, p _max) the straight line connecting the (P = Ku). That is, the optimum PID parameter value was given during the control as described above: The PID controller 33 calculates the output probability u using a constant proportional gain. Deca opening 11 of the P 10 is the steady-state gain linearizer 38 (conversion table), it is converted into the opening degree u _d actually passed to the exhaust valve. Opening convert tapes Le of u → u _d may be prepared by solving the following (14) 9-order polynomial expression shown in FIG.

(14)

Conversion to opening u _d due to linearization of the steady-state gain is equivalent to sequentially vary the proportional gain in realtime ((4) see formula), it can further improve the controllability. If the steady-state gain is not linearized, the non-linearity of the controlled object will not be considered.

In the control target control step (ST6), the control signal compensated by the steady-state gain linearizer 38 as described above is output to the exhaust valve 22 to be controlled. to this Accordingly, the opening degree of the exhaust valve 22 is controlled, and the pressure in the chamber 11 is controlled to a predetermined value.

In the present embodiment, as described above, the optimal PID parameter for each control step—the evening value is determined in advance and formed into a template, and from this table, the actual setting condition change for the PID controller 33 is determined. Select the corresponding PID parameter overnight value, change the PID parameter overnight value previously set in the PID controller 3 3 to the selected PID parameter—evening value, and change the changed PID parameter overnight value. The control signal is output from the PID controller 3 3 to the exhaust valve 22 to be controlled, and control is performed.Therefore, when the setting conditions change, the control can always be performed with appropriate PID parameter values. Extremely accurate control can be performed. In addition, since the optimal PID parameter value for each control step is obtained in advance and tabulated, and an appropriate PID parameter value is automatically selected from the table, no complicated operation is required. Further, as described above, the control signal output from the PID controller 33 is compensated by the linearizer 38 of the constant gain to the linear control system, so that more accurate control is realized.

Note that the present invention is not limited to the above embodiment, and can be variously modified. For example, in the above-described embodiment, an example in which the PID control is performed has been described. However, the present invention is not limited to this, and another control, for example, an optimal regulation control in which the dynamic characteristic of the control target is expressed by a state equation can be adopted. Further, in the above-described embodiment, an example is described in which the opening degree of the exhaust valve of the chamber 1 is controlled and the pressure in the chamber 1 of the film forming apparatus is controlled. However, the present invention is not limited to this. The present invention can be applied to other various controls such as, for example, temperature control and flow rate control. Further, the method of obtaining the optimal control parameter values for each control step and tabulating the values is not limited to the method of the above embodiment, and various methods can be adopted. Furthermore, in the above embodiment, the control signal is compensated using the stationary gain linearizer, but such compensation of the control signal is not essential.

Claims

The scope of the claims

1. A control method using a controller for controlling a predetermined process parameter by a predetermined control parameter,

A step of obtaining an optimal control parameter overnight value for each control step in advance;

Selecting a control parameter value corresponding to an actual setting condition change for the controller from the determined plurality of control parameter value;

Changing the control parameter value set in the controller to the control parameter value selected in the selecting step;

Outputting a control signal from the controller based on the changed control parameter to a control target;

A control method comprising:

2. The control by the controller is PID control or optimal regulation control.

2. The control method according to claim 1, wherein:

3. A step of compensating the control signal output from the controller to be a linear control system using a non-linear compensation element when the control target is non-linear with respect to the controlled process parameter. Further comprising

2. The control method according to claim 1, wherein:

4. A control method using a controller that controls a predetermined process parameter by PID control based on a preset setting condition and a fed back process condition,

A step of previously determining an optimum PID parameter value for each control step, and a step of selecting a PID parameter value corresponding to a change in an actual setting condition for the controller from the obtained plurality of PID parameter values. When,

Is set in the controller; the PID parameter overnight value is selected in the selection step Changing the PID parameter to an overnight value,

Outputting from the controller a control signal based on the changed PID parameter overnight value to a control target;

A control method comprising:

5. A step of compensating the control signal output from the controller to be a linear control system using a non-linear compensation element when the control target is non-linear with respect to the controlled process parameter. Further comprising

5. The control method according to claim 4, wherein:

6. In the above-mentioned process of obtaining the optimum PID parameter value for each control step in advance,

The response waveforms of the process parameters when the control target is changed are obtained for each control step.

From each response waveform, transfer characteristics are identified using a dead time + nth-order delay model, and the PID parameter overnight value is determined from the identified transfer characteristics by a predetermined method.

5. The control method according to claim 4, wherein:

7. Based on the preset value of the pressure inside the processing vessel and the feedback measured value of the pressure inside the processing vessel, the degree of exhaust valve control is controlled by PID control to control the pressure inside the processing vessel. A control method using a controller,

A step of previously determining an optimum PID parameter overnight value corresponding to a change in various set pressures;

Selecting a PID parameter value corresponding to an actual set pressure change for the controller from the determined plurality of PID parameter values;

Changing the PID parameter overnight value set in the controller to the PID parameter overnight value selected in the selection step;

The controller outputs a control signal based on the changed PID parameter value. A step of outputting to the mind,

A control method comprising:

8. In the above-mentioned step of previously finding the optimum PID parameter overnight value corresponding to various set pressure changes,

The response waveform of the pressure inside the processing vessel when the opening of the exhaust valve is changed is obtained for each change in the set pressure.

From each response waveform, transfer characteristics are identified using a delay time + η-order delay model, and the PID parameter overnight value is determined from the identified transfer characteristics by a predetermined method.

8. The control method according to claim 7, wherein:

9. Measure the relationship between the valve degree and the pressure in the processing vessel in advance, and use a non-linear compensation element for the control signal output from the controller to determine the relationship between the valve opening and the processing vessel pressure. Step of compensating for a linear relationship with pressure

9. The control method according to claim 8, further comprising:

10. Finding the optimal PID parameter value corresponding to the change of various set pressures in advance In the above-mentioned process, the optimal PID parameter value is also calculated taking into account the set flow rate

8. The control method according to claim 7, wherein:

11. The PID control is normal PID control, I-PD control, or differential leading PID control

5. The control method according to claim 4, wherein:

12. Above: PID control is normal PID control, I-PD control, or derivative PID control

8. The control method according to claim 7, wherein:

1 3. An offset value is used for the control signal based on the changed PID parameter value.

8. The control method according to claim 7, wherein:

1 4. A control device for controlling a predetermined process parameter by a predetermined control parameter,

A controller that controls a control target by a predetermined control parameter based on a preset setting condition and a fed back process condition;

Calculating means for calculating an optimum control parameter value for each control step; storage means for storing an optimum control parameter value for each control step obtained by the calculating means;

A change in the setting condition for the controller is input, and a control parameter value corresponding to the change in the setting condition is selected from the information stored in the storage means, and the control parameter value is provided to the controller. Control parameter selection means for outputting a command to change the value to the selected control parameter value;

A control device comprising:

15. The controller controls the object to be controlled by PID control or optimal regulation overnight control.

15. The control device according to claim 14, wherein:

16. A control device that controls a predetermined process parameter by PID control based on a preset setting condition and a fed back process condition,

A controller for controlling a control target by a PID parameter overnight value based on a preset setting condition and a fed back process condition;

A calculating means for calculating an optimum PID parameter value for each control step; and an optimum PID parameter value for each control step obtained by the calculating means. Memory means to remember,

A change in the setting condition for the controller is input, and a PID parameter value corresponding to the change in the setting condition is selected from the information stored in the storage means, and the PID parameter value is transmitted to the controller. Control parameter overnight selecting means for outputting a command to change the evening value to the selected PID parameter value;

A control device comprising:

17. The calculation means obtains a response waveform of the process parameter when the control target is changed for each control step, and calculates a transfer characteristic of the transfer characteristic from each response waveform using a dead time + n-order delay model. Identification is performed, and the PID parameter overnight value is obtained from the identified transfer characteristics by a predetermined method.

17. The control device according to claim 16, wherein:

18. Compensation for compensating the control signal output from the controller to be a linear control system using a non-linear compensation element when the control target is nonlinear with respect to the controlled process parameter. Means

The control device according to claim 14, further comprising:

1 9. Compensation for compensating a control signal output from the controller to be a linear control system using a non-linear compensation element when the control target is nonlinear with respect to the controlled process parameter. Means

The control device according to claim 16, further comprising: