JP2009116721A - Controller, temperature regulator, and gain regulator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To stably control a control object with strong interference, without generating hunting or the like, in temperature control or the like. <P>SOLUTION: Proportional gains of controllers C1, C2 of respective channels are brought into a proportional gain smaller than a proportional gain found based on a response of a detection temperature in only the channel with a changed control input, while taking the interference between the respective channels into consideration. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a control device that controls a physical state such as temperature and pressure of a control target, a temperature regulator that controls the temperature of the control target, and a gain adjustment device that is suitable for adjusting their gain.

  Conventionally, for example, in temperature control in which an object to be heated is placed on a hot plate and subjected to heat treatment, the temperature regulator is heated based on a temperature detected from a temperature sensor disposed on the hot plate. This is performed by controlling energization of a heater disposed on the hot plate so that the temperature of the plate becomes a set temperature (see, for example, Patent Document 1).

For example, FIG. 13 shows the temperature sensors 13 and 13 such as thermocouples that detect the temperature of the control target 2 having interference between the two heater blocks 2 ₁ and 2 ₂ constituting the hot plate. , A two-channel (ch) temperature control system that controls energization by the heaters 16 and 16 via the SSRs 15 and 15.

As shown in FIG. 14, the temperature controller 14 includes controllers C1 and C2 that individually correspond to the channels 1 and 2, and each of the controllers C1 and C2 includes detected temperatures θ ₁ and θ ₂ and a target temperature. PID control is performed based on the deviation between r ₁ and r ₂ .

  Conventionally, the gains of the controllers C1 and C2 are adjusted as follows.

That is, first, the operation amount of channel 1 is changed stepwise, and the response waveform of channel 1 as shown in FIG. 15 is measured. From this response waveform with ZN (Ziegler-Nichols) method, the proportional gain Kp is a gain of the controller C1, the integration time _{T I,} calculates the derivative time _{T D.}

The proportional gain Kp, integration time T _I , and differentiation time T _D are calculated by the following equations.

Kp = 1.2 / RL
T _I = 2.0L
T _D = 0.5L
Here, R is the maximum slope, and is calculated as R = K / T from the steady gain K and the time constant T measured from the step response waveform of FIG. L is a dead time.

  FIG. 16 shows an example of specific experimental results. In FIG. 16, the operation amount that is input to channel 1 is U1, the detected temperature that is output is PV1, and the operation amount that is input to channel 2. Is U2, and the detected temperature as output is PV2.

  In FIG. 16 and FIG. 17 to be described later, as shown in FIG. 18, when the step-like operation amount U11 is input to the channel 1, the operation amount of the channel 2 is U21, and the channels 1 and 2 with respect to the operation amount U11. As will be described later, when the step-like manipulated variable U22 is input to the channel 2, the manipulated variable of the channel 1 is U12, and the detected temperatures of the channels 1 and 2 with respect to the manipulated variable U22. These changes are PV12 and PV22.

  FIG. 16 shows that the detected temperature PV11 of the channel 1 is 5.2 with respect to the step input of the operation amount U11 = 1 of the channel 1. Further, channel 2 also shows that the detection temperature PV21 is 4.7 due to the influence of channel 1.

  From the response waveform PV11 of the channel 1, that is, the response waveform PV11 of the detected temperature PV1 with respect to the change U11 of the operation amount U1 of the channel 1, the slope R1 = 0.50 and the dead time L1 = 0. 70 seconds are obtained.

  Accordingly, the proportional gain Kp1, the integration time Ti, and the differentiation time Td1 of the controller C1 of the channel 1 are calculated as follows.

Kp1 = 1.2 / RL = 1.2 / (0.50 × 0.70) = 3.4
Ti = 2.0L = 2.0 × 0.7 = 1.4
Td1 = 0.5L = 0.5 × 0.7 = 0.35
Next, as shown in FIG. 17, the operation amount U2 of the channel 2 is similarly changed stepwise, and the response waveform is measured.

  From this channel 2 response waveform, that is, the response waveform PV22 of the detected temperature PV2 with respect to the change U22 of the operation amount U2 of the channel 2, a slope R2 = 0.50 and a dead time L2 = 0.70 seconds are obtained.

  Accordingly, the proportional gain Kp1, the integration time Ti, and the differentiation time Td1 of the controller C2 of the channel 2 are calculated as follows.

Kp1 = 1.2 / RL = 1.2 / (0.50 × 0.70) = 3.4
Ti = 2.0L = 2.0 × 0.7 = 1.4
Td1 = 0.5L = 0.5 × 0.7 = 0.35
As described above, based on the response waveform PV11 of the detected temperature PV1 of the channel 1 with respect to the change U11 of the stepped operation amount U1 of the channel 1, the proportional gain Kp, the integration time T _I , and the differential time T _D of the controller C1 of the channel 1 Further, based on the response waveform PV22 of the detected temperature PV2 of the channel 2 with respect to the change U22 of the stepped operation amount U2 of the channel 2, the proportional gain Kp, the integration time T _I , and the derivative time of the controller C2 of the channel 2 are calculated. calculates the T _D, respectively set as the gain of the controller C1, C2 of each channel.

That is, the gains of the controllers C1 and C2 for each channel are set based on the individual response waveforms for each channel.
Japanese Patent Laid-Open No. 2001-274069

  In the conventional example described above, the gain is calculated and adjusted independently for each channel.

  For this reason, the control target with strong interference between the channels cannot be adjusted to an appropriate gain. As shown in FIG. 19, when the target temperature is changed stepwise as shown by the broken line, There is a problem of temperature hunting.

  The present invention has been made in view of the above-described points, and an object of the present invention is to enable stable control that does not cause hunting or the like even for a control target with strong interference.

  (1) A control device according to the present invention is a proportional control device for a plurality of channels that calculates and outputs a plurality of operation amounts based on control amounts from a plurality of detection units that respectively detect a physical state of a control target. A proportional gain that is smaller than the proportional gain that is obtained based on the response of the control amount of only the channel for which the manipulated variable is changed when the manipulated variable of the channel for which the gain is to be obtained is used as the proportional gain of the channel. .

  The physical state refers to a state of various physical quantities such as temperature, pressure, flow rate, speed or liquid level.

  It is preferable to obtain the proportional gain of each channel by changing the operation amount of each channel in order.

  The change in the operation amount may be a change in the operation amount itself, or may be a change in the operation amount indirectly by changing the target value, for example.

 “Determined based on the response of the control amount only for the channel in which the operation amount is changed” means that it is obtained based on the response of the control amount only in the same channel as the channel in which the operation amount is changed. When the operation amount of channel 1 is changed, it is obtained based on the response of the control amount of only channel 1, and when the operation amount of channel 2 is changed, it is obtained based on the response of the control amount of only channel 2. Says what is required.

  The control amount is not limited to the control amount itself, and for example, a differential value (change rate) of the control amount may be used.

  As a method for obtaining the gain based on the response of the control amount, a conventional method can be used, for example, a ZN (Ziegler-Nichols) method or the like can be used.

  For example, if there is a difference in the physical state such as a temperature difference, the object to be controlled will cause interference when the physical state changes so as to eliminate the difference. When calculated based on the response of the control amount of only the channel, that is, when the proportional gain is determined independently for each channel and the response of other channels is ignored, the proportional gain with neglected the influence of interference between channels is obtained. For this reason, in the case of a controlled object with a large amount of interference, a proportional gain larger than the original proportional gain is estimated, and the control becomes unstable.

  According to the control device of the present invention, when the operation amount of the channel for which the proportional gain is to be obtained is changed, a proportional gain smaller than the proportional gain obtained based on the response of the control amount of only the channel is obtained. Since the gain is used, the proportional gain of each channel becomes more appropriate and stable control is possible.

  (2) The temperature controller of the present invention is a multi-channel temperature controller that calculates and outputs a plurality of manipulated variables based on the detected temperatures from a plurality of temperature detecting means that detect the temperatures to be controlled, respectively. When the operation amount of the channel for which the proportional gain is to be obtained is changed, a proportional gain smaller than the proportional gain obtained based on the response of the detected temperature of only the channel for which the operation amount has been changed is proportional to the channel. Gain.

  If there is a temperature difference, the control target will cause interference due to the heat flow from the high temperature part to the low temperature part, but the proportional gain is obtained based on the response of the detected temperature only for the channel with the manipulated variable changed. That is, if the proportional gain is obtained independently for each channel, ignoring the responses of other channels, the proportional gain is neglected by the influence of the heat flow between the channels. In this case, a proportional gain larger than the original proportional gain is estimated, and the control becomes unstable.

  According to the temperature controller of the present invention, when the operation amount of the channel for which the proportional gain is to be obtained is changed, a proportional gain that is smaller than the proportional gain that is obtained based on the response of the detected temperature of only that channel is obtained. Since the proportional gain is used, the proportional gain of each channel becomes more appropriate, and stable control with hunting and the like suppressed is possible.

 (3) The temperature controller of the present invention is a multi-channel temperature controller that calculates and outputs a plurality of manipulated variables based on the detected temperatures from a plurality of temperature detecting means that detect the temperatures to be controlled, respectively. , The proportional gain obtained based on the response of the detected temperature of the channel and the detected temperature of the channel other than the channel when the manipulated variable of the channel for which the proportional gain is to be changed is changed. , The proportional gain of the channel.

  According to the temperature controller of the present invention, when the operation amount of the channel for which the proportional gain is to be calculated is changed, the proportional gain that is obtained based on the response of the detected temperature of the channel and the response of the detected temperature of the channel other than the channel. Is the proportional gain of the channel, that is, the proportional gain of each channel is obtained based on the response of not only the channel but also of other channels, so compared to the proportional gain obtained based on the response of only the channel. Therefore, the proportional gain is more appropriate in consideration of interference between channels, and stable control with hunting and the like suppressed is possible.

  (4) In the embodiment of (3), the proportional gain of the channel is smaller than the proportional gain obtained based on the response of the detected temperature of only the channel when the operation amount of the channel is changed. Proportional gain.

  According to this embodiment, when the operation amount of the channel for which the proportional gain is to be obtained is changed, a proportional gain that is smaller than the proportional gain that is obtained based on the response of the detected temperature of only the relevant channel is used as the proportional gain of the relevant channel. Therefore, the proportional gain of each channel becomes more appropriate, and stable control with hunting suppressed is possible.

  (5) In the embodiment of the above (2) or (4), the proportional gain of each channel may be a smaller proportional gain as the interference of the control target is stronger.

  According to this embodiment, the stronger the interference of the control object, the smaller the proportional gain. Therefore, the proportional gain of each channel becomes more appropriate according to the degree of interference of the control object, and control with strong interference is performed. Even for a target, stable control is possible.

  (6) In one embodiment of the temperature controller of the present invention, when the change in the detected temperature of the j-th channel when the manipulated variable of the i-th channel is changed is PVji, i of the channel is i. The proportional gain Kpi of the second channel satisfies the following formula.

Kpi ≦ Kpi ′ × PVii / (PV1i + PV2i + ... PVii + ... PVni)
here,
Kpi ′: Proportional gain calculated using only the detected temperature of the channel with the manipulated variable changed Kpi: Proportional gain calculated taking into account the detected temperature other than the channel with the manipulated variable changed n: Maximum number of channels According to the above, the proportional gain Kpi ′ calculated using the detection temperature of only the channel i is the proportional gain Kpi ′ of the channel i that is the channel, and the change PVii of the detection temperature of the channel i is the detection temperature of all the channels. Since the influence of interference with other channels is corrected by multiplying by the value divided by the sum of changes (PV1i + PV2i + ... PVii + ... + PVni), it becomes a more appropriate proportional gain. Stable control with hunting and the like suppressed is possible.

 (7) In another embodiment of the temperature controller of the present invention, when the gradient of the detected temperature change of the j-th channel when the manipulated variable of the i-th channel is changed is Rji, The proportional gain Kpi of an i-th channel satisfies the following expression.

Kpi ≦ Kpi ′ × Rii / (R1i + R2i + ... Rii + ... Rni)
here,
Kpi ′: Proportional gain calculated using only the detected temperature of the channel with the manipulated variable changed Kpi: Proportional gain calculated taking into account the detected temperature other than the channel with the manipulated variable changed n: Maximum number of channels According to the above, the proportional gain Kpi ′ of the channel i that is the channel is converted to the proportional gain Kpi ′ calculated using the gradient of the detected temperature change of only the channel i, and the gradient Rii of the detected temperature change of the channel i is calculated. By multiplying the value divided by the sum of the gradients of the detected temperature of all other channels (R1i + R2i + ... Rii + ... Rni), the influence of interference with other channels is corrected. Therefore, a more appropriate proportional gain is obtained, and stable control with hunting suppressed is possible.

 (8) In still another embodiment of the temperature controller of the present invention, the proportional gain of each channel is reduced as the number of channels increases.

  According to this embodiment, as the number of channels increases, the proportional gain becomes smaller. Therefore, the proportional gain of each channel becomes more appropriate according to the degree of interference according to the number of channels. Therefore, stable control is possible.

  (9) In a preferred embodiment of the temperature controller of the present invention, when the proportional gain, integral gain, and differential gain of the channel are Kp, Ki, Kd, and the manipulated variable of the channel is changed, the channel Kp ′> Kp and Kp ′ / Kp> Ki, where Kp ′, Ki ′, and Kd ′ are the proportional gain, integral gain, and differential gain obtained based on the response of only the detected temperature. The relationship of '/ Ki or Kp' / Kp> Kd '/ Kd is satisfied.

 (10) In the embodiment of the above (2) or (4), it has gain search means for determining the quality of the gain based on the evaluation index and searching for the gain of each channel, and the gain search means The proportional gain of the channel may be searched as a search range with a value smaller than the proportional gain obtained based on the detected temperature response of only the channel when the operation amount of the channel is changed.

  According to this embodiment, as the proportional gain of each channel, a value smaller than the proportional gain obtained based on the response of the detected temperature of only the channel whose operation amount is changed is obtained by searching, and hunting or the like is performed. Suppressed and stable control becomes possible.

 (11) In another embodiment of the temperature controller of the present invention, the temperature controller includes a non-interacting unit that eliminates or reduces the influence of the control by each channel on the control by another channel, The difference between the detected temperatures may be fed back to at least one of the input side and the output side of each channel so as to cancel the interference.

  The non-interacting means preferably feeds back the difference between the detected temperatures to an operation amount on the output side of each channel, or a deviation or a target temperature on the input side.

  According to this embodiment, since the interference is canceled by the non-interacting means, more stable control is possible.

  (12) The gain adjusting apparatus according to the present invention is a control device for a plurality of channels that calculates and outputs a plurality of operation amounts based on control amounts from a plurality of detection units that respectively detect physical states of control targets. In a device that adjusts the gain of each channel, when the manipulated variable of the channel for which the proportional gain is to be obtained is changed, the gain is smaller than the proportional gain obtained based on the response of the control variable of only the channel for which the manipulated variable is changed. The proportional gain of the channel is adjusted to the proportional gain.

  According to the gain adjusting device of the present invention, the proportional gain of each channel is adjusted to a proportional gain that is smaller than the proportional gain obtained based on the response of the control amount of only the channel with the manipulated variable changed. Gain becomes more appropriate and stable control becomes possible.

  (13) A gain adjusting apparatus according to the present invention is a multiple channel temperature controller that calculates and outputs a plurality of manipulated variables based on detected temperatures from a plurality of temperature detecting means that detect temperatures to be controlled, respectively. In the device for adjusting the gain of each of the channels, when the operation amount of the channel for which the proportional gain is to be obtained is changed, the proportional gain obtained based on the response of the detected temperature of only the channel for which the operation amount has been changed. Also, the proportional gain of the channel is adjusted to a small proportional gain.

  According to the gain adjusting apparatus of the present invention, the proportional gain of each channel is adjusted to a proportional gain smaller than the proportional gain obtained based on the response of the detected temperature of only the channel with the manipulated variable changed. Gain becomes more appropriate, and stable control with suppressed hunting and the like becomes possible.

  (14) A gain adjusting apparatus according to the present invention is a multiple channel temperature controller that calculates and outputs a plurality of manipulated variables based on detected temperatures from a plurality of temperature detecting means for detecting temperatures to be controlled, respectively. In the device for adjusting the gain of each channel, when the operation amount of the channel for which the proportional gain is to be obtained is changed, the response of the detection temperature of the channel and the detection temperature of the channel other than the channel when the operation amount is changed The proportional gain of the channel is adjusted to the proportional gain obtained based on the response.

  According to the gain adjusting device of the present invention, the proportional gain of each channel is obtained based on the response of not only the channel but also the other channel. Therefore, it is possible to adjust to a more appropriate proportional gain in consideration of the interference, and it is possible to perform stable control while suppressing hunting and the like.

 (15) In the embodiment of the above (14), the proportional gain of the channel is smaller than the proportional gain obtained based on the response of the detected temperature of only the channel when the operation amount of the channel is changed. You may adjust to a proportional gain.

  According to this embodiment, the proportional gain of each channel is adjusted to a proportional gain that is smaller than the proportional gain obtained based on the response of the detected temperature of only the channel with the manipulated variable changed. It becomes appropriate, and stable control with suppressed hunting becomes possible.

(16) In the above embodiment (13) or (15), the proportional gain of each channel may be adjusted to a smaller proportional gain as the interference of the control target is stronger.
According to this embodiment, the stronger the interference of the control object, the smaller the proportional gain is adjusted. Therefore, the proportional gain of each channel becomes more appropriate according to the degree of interference of the control object, and the interference is strong. Even if it is a control target, stable control is possible.

 (17) In one embodiment of the gain adjusting apparatus of the present invention, when the change in detected temperature of the j-th channel when the manipulated variable of the i-th channel is changed is PVji, the channel i is The proportional gain Kpi of the second channel may be adjusted so as to satisfy the following expression.

Kpi ≦ Kpi ′ × PVii / (PV1i + PV2i + ... PVii + ... PVni)
here,
Kpi ′: Proportional gain calculated using only the detected temperature of the channel with the manipulated variable changed Kpi: Proportional gain calculated taking into account the detected temperature other than the channel with the manipulated variable changed n: Maximum number of channels According to the above, the proportional gain Kpi ′ calculated using the detection temperature of only the channel i is the proportional gain Kpi ′ of the channel i that is the channel, and the change PVii of the detection temperature of the channel i is the detection temperature of all the channels. Since the influence of interference with other channels is corrected by multiplying by the value divided by the sum of changes (PV1i + PV2i + ... PVii + ... PVni), a more appropriate proportional gain is obtained. Stable control with suppressed hunting and the like becomes possible.

  (18) In another embodiment of the gain adjusting apparatus of the present invention, when the gradient of the detected temperature change of the j-th channel when the manipulated variable of the i-th channel is changed is Rji, The proportional gain Kpi of an i-th channel may be adjusted so as to satisfy the following expression.

Kpi ≦ Kpi ′ × Rii / (R1i + R2i + ... Rii + ... Rni)
here,
Kpi ′: Proportional gain calculated using only the detected temperature of the channel with the manipulated variable changed Kpi: Proportional gain calculated taking into account the detected temperature other than the channel with the manipulated variable changed n: Maximum number of channels According to the above, the proportional gain Kpi ′ of the channel i that is the channel is converted to the proportional gain Kpi ′ calculated using the gradient of the detected temperature change of only the channel i, and the gradient Rii of the detected temperature change of the channel i is calculated. Correct the influence of interference with other channels by multiplying by the value divided by the sum of the gradients of the detected temperature of all other channels (R1i + R2i + ... Rii + ... + Rni). Therefore, a more appropriate proportional gain is obtained, and stable control with hunting and the like suppressed is possible.

  (19) In a gain adjustment device according to another embodiment of the present invention, the proportional gain of each channel may be adjusted to be smaller as the number of channels increases.

  According to this embodiment, as the number of channels increases, the proportional gain is adjusted to be smaller. Therefore, the proportional gain of each channel becomes more appropriate according to the degree of interference according to the number of channels, and the number of channels is increased. Regardless, stable control is possible.

  (20) In a preferred embodiment of the gain adjusting device of the present invention, when the proportional gain, integral gain, and differential gain of the channel are set to Kp, Ki, Kd, and the operation amount of the channel is changed, the channel Kp ′> Kp and Kp ′ / Kp> Ki, where Kp ′, Ki ′, and Kd ′ are the proportional gain, integral gain, and differential gain obtained based on the response of only the detected temperature. The gain is adjusted so as to satisfy the relationship of '/ Ki or Kp' / Kp> Kd '/ Kd.

  (21) In another embodiment of the gain adjusting device of the present invention, the gain adjusting device includes gain search means for determining the quality of the gain based on the evaluation index and searching for the gain of each channel, The proportional gain of the channel is searched as a search range for a value smaller than the proportional gain obtained based on the detected temperature response of only the channel when the manipulated variable of the channel is changed, and the proportional gain is calculated. You may adjust.

  According to this embodiment, as the proportional gain of each channel, a value smaller than the proportional gain obtained based on the response of the detected temperature of only the channel whose operation amount is changed is obtained by the search. By adjusting to the proportional gain, stable control with hunting and the like suppressed can be achieved.

  According to the present invention, when the operation amount of the channel for which the proportional gain is to be obtained is changed, a proportional gain that is smaller than the proportional gain that is obtained based on the response of the control amount only for the channel is used as the proportional gain of the channel. Therefore, the proportional gain of each channel becomes appropriate considering the influence of interference, and stable control is possible.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1)
FIG. 1 is a configuration diagram of a temperature control system using a temperature controller according to the present invention.

Similarly to FIG. 13 described above, the temperature controller 1 of this embodiment, for example, determines the temperature of the control target 2 having two heater blocks 2 ₁ and 2 ₂ constituting the hot plate as the temperature of a thermocouple or the like. This is detected by the sensors 13 and 13, respectively, and the energization by the heaters 16 and 16 is controlled via the SSRs 15 and 15, and will be described by being applied to two-channel temperature control. It can be similarly applied to three or more channels.

The temperature controller 1 is a temperature control unit that calculates and outputs the manipulated variable based on the deviation between the detected temperatures θ ₁ and θ 2 of the two control points of the control target ₂ and the target temperatures r ₁ and r ₂ . Controllers C1 and C2 and gain adjusting means 5 for adjusting the gains of the controllers C1 and C2 as described later are provided.

  Each of the controllers C1 and C2, the gain adjusting means 5 and the like are configured by a microcomputer, for example.

  Each of the controllers C1 and C2 is a PID controller, but may be any controller that performs at least P control.

  In the temperature controller 1 of this embodiment, gain adjustment by the gain adjusting means 5 is performed as follows so that hunting or the like does not occur even for a control object with strong interference.

  First, in FIG. 1, the connection switches 7 and 8 between the controlled object 2 and the controllers C1 and C2 are turned off, and the controlled object 2 is connected to the gain adjusting means 5 side.

  As shown in FIG. 2, the MV setting means 9 of the gain adjusting means 5 sequentially adds stepwise operation amounts U1, U2 for each channel, and the PV acquisition means 10 of the gain adjusting means 5 controls the control amount PV1 at that time. , PV2 is acquired.

  FIG. 3 shows an example of a specific experimental result when a stepped operation amount is given to the channel 1, and U11, U21, PV11, and PV21 correspond to the above-described FIG.

  In FIG. 3, the detected temperature PV11 of the channel 1 becomes 5.2 with respect to the input of the operation amount U11 = 1 of the channel 1, and the detected temperature PV21 of the channel 2 becomes 4.7 due to interference. Show.

  At this time, the slope R1 = 0.50 of the channel 1 and the dead time L1 = 0.70 seconds are obtained from the response waveform of the channel 1 as shown in FIG.

  Conventionally, as described above, the gain of the controller C1 of the channel 1 is calculated from the slope R1 and the dead time L1, but in this embodiment, the slope R1 is not directly used for gain calculation, but the slope R1 is Assuming that the apparent gradient R1 ′ is not the true gradient R1, the true gradient R is obtained as follows.

  In the following description, the slope R will be described by distinguishing between the true slope R and the apparent slope R ′ as necessary.

  Here, the apparent slope R ′ is a slope obtained from the response waveform and is a slope affected by the interference between channels. Therefore, the inclination R described in the above-described conventional example is an apparent inclination R ′.

  On the other hand, the true inclination R is an inclination obtained by correcting the apparent inclination R ′ so as to eliminate or reduce the influence of interference between channels, and is an inclination used in this embodiment.

  FIG. 4 shows a step response waveform, where a solid line indicates a case where there is no interference between channels, and a broken line indicates a case where there is interference between channels.

  When there is interference, heat flows out to the interference area, so that the waveform of the temperature rise becomes gentle as shown by the broken line. Originally, the true slope R (= K / T) is to be obtained. The apparent slope R ′ (= K / T ′) is obtained. Further, the relationship between the true slope R and the apparent slope R ′ is as follows.

R = K / T> R ′ = K / T ′
Therefore, in the gain design by the ZN method, the proportional gain Kp = 1.2 / RL is estimated to be larger than an appropriate value.

  Therefore, in this embodiment, the true slope Ri and dead time Li are calculated for each channel by the following equations.

Ri = Ri ′ × (PV1i + PV2i +... + PVni) / PVii
Li = LI '
i: Channel number n: Maximum number of channels PV1i: Control amount of channel 1 when step input is given to channel i (PV1)
When applied to the experimental results of FIG.
R1 = R1 ′ × (PV11 + PV21) / PV11
= 0.5 × (5.2 + 4.7) /5.2=0.95
L1 = L1 ′ = 0.7
It becomes.

  Thus, in order to calculate the true slope R1 of the channel 1, not only the detection temperature PV11 that is the control amount of the channel 1 but also the detection temperature PV21 that is the control amount of the channel 2 is used, and interference is considered. It has become a thing.

  Here, the reason why the true slope R is obtained from the apparent slope R ′ as described above will be qualitatively explained.

  When the number of channels is n, the slope of the temperature rise in a state of being divided without interference between n channels is defined as true R. However, in actuality, since n pieces are connected, the heat flow applied to the i-th channel leaks to other channels. Therefore, the temperature rise of channel i decreases and the temperature of other channels rises. When the interference is very large, all channels rise with the same slope R. The slope at that time is a fraction of the number of channels.

  When the interference is not very large, the inclination R of the channel to be heated is the largest, and the inclination R decreases as the distance from the periphery increases. If all the temperature rises R due to the leaked heat flow are collected to increase the temperature of one cell, the true slope R can be obtained. This means the following mathematical formula.

Ri = Ri ′ × (R1i + R2i +... + Rni) / Rii
This may be left as it is, but the above formula is obtained when PV of steady temperature that is easy to measure is used instead of R. When the time constants of the first order delays are equal and equal to T, the ratio of R is proportional to the ratio of K from the equation R = K / T, and if MV is constant, PV = K × MV, so R is PV And the above formula is obtained.

  Next, FIG. 5 shows an example of a specific experimental result of the channel 2, and corresponds to FIG.

  For channel 2, as with channel 1, true slope R and dead time L are determined from the step response waveform.

  As shown in FIG. 5, with respect to the input of the operation amount U22 = 1 of the channel 2, the detection temperature PV22 of the channel 2 becomes 5.2, and the detection temperature PV12 of the channel 1 becomes 4.7 due to interference.

  At this time, a slope R2 '= 0.50 and a dead time L1' = 0.70 seconds were obtained from the response waveform of channel 2.

Therefore, like channel 1, the true slope R2 and dead time L2 of channel 2 are
R2 = R2 ′ × (PV12 + PV22) / PV22
= 0.5 × (4.7 + 5.2) /5.2=0.95
L1 = L1 ′ = 0.7
It becomes.

  Next, from the true slope R and dead time L obtained as described above, the gains of the controllers C1 and C2 of each channel are designed by the ZN method.

The controller C1 of channel 1
R1 = 0.95
From L1 = 0.70 seconds,
Kp1 = 1.2 / RL = 1.2 / (0.50 × 0.95) = 2.5
Ti1 = 2.0L = 2.0 × 0.7 = 1.4
Td1 = 0.5L = 0.5 × 0.7 = 0.35
It becomes.

Similarly, the controller C2 of channel 2
R2 = 0.95
From L2 = 0.70 seconds,
Kp2 = 1.2 / RL = 1.2 / (0.50 × 0.95) = 2.5
Ti2 = 2.0L = 2.0 × 0.7 = 1.4
Td2 = 0.5L = 0.5 × 0.7 = 0.35
It becomes.

  The gain design unit 12 of the gain adjustment unit 5 sets the gain calculated as described above to each of the controllers C1 and C2, and then disconnects the control target 2 from the gain adjustment unit 5 to control each controller C1C2. Control is performed by connecting the object 2.

  FIG. 6 is a diagram showing step responses by the controllers C1 and C2 whose gains are adjusted as described above, and corresponds to FIG. 19 described above.

  As shown in FIG. 6, when one target temperature is changed stepwise as shown by a broken line, the waveform is hunting at first, but is finally set to the target value.

(Embodiment 2)
FIG. 7 is a configuration diagram of a temperature control system using a temperature controller according to another embodiment of the present invention.

  The temperature controller 1-1 of this embodiment uses a feedback type model structure proposed by the present applicant in, for example, Japanese Patent Application Laid-Open No. 2004-94939, as a model of the control target 2. Description will be made by applying to temperature control of two channels.

The temperature regulator 1-1 of this embodiment calculates the operation amount based on the deviation between the detected temperatures θ ₁ and θ _{2 of the} two control points of the control target ₂ and the target temperatures r _{1 and} r _2. Controllers C1 and C2 are provided, and this configuration is the same as that of the above-described embodiment.

In the temperature controller 1-1 of this embodiment, the difference between the detected temperatures θ ₁ and θ ₂ is calculated on the output side of the controllers C1 and C2 so that hunting or the like does not occur even if the object is a control object with strong interference. Are provided with non-interacting elements F1 and F2 for feedback, respectively, and gain adjusting means 5-1 for adjusting the gains of the non-interacting elements F1 and F2 and the controllers C1 and C2.

  Here, the model structure of the controlled object 2 will be briefly described.

This model structure has been previously proposed by the applicant of the present invention as described above, and the difference between the temperatures θ ₁ and θ ₂ of the respective channels is made positive and negative on the input side via the interference term 6. H ₁ and H ₂ indicate the degree of heat generated by the heater.

  This model structure is a thermal model of a thermal interference system, and when there is a temperature difference, the movement of heat occurs, and this movement of heat is equivalent to the meaning of Fourier's law that is proportional to the temperature difference. .

  The temperature difference causes a heat transfer from one channel to the other, blocking the heat interference phenomenon where one channel loses heat (negative) and the other channel adds heat (positive). The interference of the control object of the thermal system is a diagram of the Fourier law that when there are two temperatures and there is a difference in temperature, the movement of the heat quantity proportional to the temperature difference occurs. This model structure is hereinafter also referred to as a “temperature difference model”.

In this embodiment, the transfer function Gβ of the interference term 6 representing the degree of interference and the transfer functions H ₁ and H ₂ representing the degree of heat generation are expressed by the following equations, respectively.

Gβ = β / (Tβ · s + 1)
H ₁ = H ₂ = Hi = hi / (Ti · s + 1)
Here, β is the degree of interference, Tβ is the dead time of interference, hi is the amount of heat generated by the heater, and Ti is the dead time.

The transfer function of the element Hsi of the model structure is Hs ₁ = Hs ₂ = Hsi = gi / (Tsi · s + 1).

  The gain adjusting unit 5-1 obtains a step response waveform of the control target 2 at that time, and an MV setting unit 9 that gives a step-like operation amount to the control target 2 in order to obtain the characteristics of the control target 2. Means 10; interference degree estimating means 11 for determining the degree of interference between the two channels and setting the decouplers F1 and F2, and gain design means 12 for setting the gains of the controllers C1 and C2. Yes.

The transfer functions Fu ₁ and Fu ₂ of the decoupling devices F 1 and F 2 of this embodiment are expressed by the following equations, respectively.

次に、この実施形態のコントローラの設計の手順を説明する。
（１）先ず、制御対象２の波形を取得し、見かけの傾きＲｉ’を求める。

Next, a procedure for designing the controller of this embodiment will be described.
(1) First, the waveform of the controlled object 2 is acquired, and the apparent slope Ri ′ is obtained.

  That is, in FIG. 7, the connection switches 7 and 8 between the controlled object 2 and the controllers C1 and C2 are turned off, and the controlled object 2 is connected to the gain adjusting unit 5-1 side.

  As shown in FIG. 2 described above, the MV setting unit 9 adds stepwise operation amounts U1 and U2 for each channel, and the PV acquisition unit 10 acquires the control amounts PV1 and PV2 at that time.

  From the acquired step response waveform, the apparent slope R1 'is obtained as described above, and the dead time L1 is obtained.

  An example of specific experimental results is shown.

  First, when step input is given to channel 1, that is, when U11 = 1 and U21 = 0, for example, PV11 = 2.018, PV21 = 1.918 are obtained as detected temperatures PV of each channel, and apparent A slope R1 ′ = 0.45 and a dead time L1 ′ = 0.85 are obtained.

  Next, when step input is given to channel 2, that is, when U12 = 0 and U22 = 1, for example, PV12 = 1.918 and PV22 = 2.107 are obtained as the detected temperature PV of each channel, and the apparent temperature is obtained. Slope R2 ′ = 0.45 and dead time L2 ′ = 0.85.

 (2) Next, as in the above-described embodiment, the true slope Ri and dead time Li are calculated for each channel by the above-described equations.

That is, from the experimental result of (1), the true slope R1 and dead time L1 of channel 1 are
R1 = R1 ′ × (PV11 + PV21) / PV11
= 0.45 × (2.017 + 1.918) /2.017=0.87
L1 = L1 ′ = 0.85
The true slope R2 and dead time L2 of channel 2 are
R2 = R2 ′ × (PV12 + PV22) / PV22
= 0.45 × (1.918 + 2.017) /2.017=0.87
L2 = L2 ′ = 0.85
(3) Next, the gains of the controllers C1 and C2 are designed by the ZN method using the true slope Ri and the dead time Li obtained for each channel as described above.
From the result of (2) above, the gains of the controllers C1 and C2 of each channel are as follows.

The controller C1 of channel 1
R1 = 0.87
L1 = 0.85
Than,
Kp1 = 1.2 / RL = 1.2 / (0.87 × 0.85) = 1.62
Ti1 = 2.0L = 2.0 × 0.85 = 1.7
Td1 = 0.5L = 0.5 × 0.85 = 0.425
It becomes.

Similarly, the controller C2 of channel 2
R2 = 0.87
L2 = 0.85
Than,
Kp2 = 1.2 / RL = 1.2 / (0.87 × 0.85) = 1.62
Ti2 = 2.0L = 2.0 × 0.85 = 1.7
Td2 = 0.5L = 0.5 × 0.85 = 0.425
It becomes.

  As described above, the gains of the controllers C1 and C2 are determined.

(4) Next, in order to design the non-interacting devices F1 and F2, the above-described interference component β is identified.
This interference component β can be obtained from the steady-state characteristics as shown in, for example, WO 2006/088072, previously proposed and published internationally by the applicant of the present invention. This can be obtained by measuring the change in output in the steady state of each channel when an input is given.

  Here, the outline of the derivation of the interference component β will be described with reference to the contents of the internationally published application.

FIG. 8 is an application of the above-described model structure to a plurality of channels. The element Hsi in FIG. 7 corresponds to the model elements 60 _{1 to} 60 _n in FIG. 8, and the interference term 6 is a feedback element 61 _{12 to} 61. _(n-1) corresponds to _n . Moreover, in FIG. 8 and FIG. 9, (theta) has shown thermal resistance.

More model elements ₆₀ 1 to 60 _n of the (n), the constant gain _K. 1 to _K a first-order lag system with _n and time constant _T. 1 to _{T n,} feedback element ₆₁ 12 a plurality (n-1) to 61 _{(n-1) n} is a constant value based on the thermal resistances θ _{12 to} θ _{(n−1) n} . Further, the inputs are P _{1 to} P _n and the outputs are T _{1 to} T _n .

  Here, considering the steady characteristics, the parameters are obtained from the steady state.

When the transfer function G (s) of the control target of the thermal system is a first-order lag system,
G (s) = K / (1 + Ts)
It becomes. Here, K is a steady gain, T is a time constant, and s is an operator of Laplace transform.

Since s = 0 in steady state,
G (s) = K
It becomes.

The steady gain K is obtained by using a constant heat flow (input) P and steady temperature (output) T, and K = T / P
It can be expressed as.

On the other hand, the thermal resistance θ is similar to Ohm's law using the heat flow P and the temperature T: θ = T / P
It can be expressed as.

Therefore, in steady state,
K = θ
It can be expressed as.

Therefore, the model structure of Figure 8, when considering only the steady-state characteristics (s = 0), the model elements ₆₀ 1 _~60n, together with term disappears time constant _T. 1 to _{T n,} the constant gain _K. 1 to _{K n} can be replaced with the thermal resistance _{theta. 1 to} theta _n, as shown in FIG. 9, the model elements ₆₀ 1 to 60 _n of a plurality (n) with one input and one output of the thermal resistance theta ₁ through? _n, respectively, the model elements ₆₀ 1 to 60 _n heat resistance theta ₁₂ is fed back to the input side the difference between the outputs of the through? _(n-1) feedback element ₆₁ 12-61 plurality having _n respective _{(n-1) (n-} 1) n It becomes a model structure provided with.

The thermal resistances θ _{1 to} θ _n and θ _{12 to} θ _{(n−1) n} , which are parameters of each element of the model structure of FIG. 9, are obtained as follows.

That is, when the input step input _{P 11} to the first input _{P 1,} the change in the output _T 1 through T _n that is measured in the steady _state, and T 11 _{through T 1n,} to the second input _{P 2} when you enter the step input _{P 22,} the change in the output _T 1 through T _n that is measured in the steady _state, and T 21 _{through T 2n,} Similarly, the input step input _{P nn} input _{P n} of the n When the changes in the outputs T _{1 to} T _n measured in the steady state are T _{n1 to} T _nn , the thermal resistances θ _{1 to} θ _n , θ _{12 to} θ _{(n−1) n of} each element Can be calculated by the following equations (3) and (4).

If the control target is a heat treatment board, the step inputs P _{11 to} P _nn can be assumed to be heater outputs for heating the heat treatment board, and the change in each output T _{1 to} T _n measured in a steady state. as is _T 11 _{through T nn,} it can be assumed temperature change.

上記式によって、各要素の熱抵抗が算出できる理由について、上記国際公開された出願に詳述している。

The reason why the thermal resistance of each element can be calculated by the above formula is described in detail in the internationally published application.

  As described above, by measuring the change in output in the steady state of each channel when step-like input is given to each channel in turn, the feedback element, that is, the thermal resistance of the interference term 6 in FIG. Can be obtained.

  In this example, for example, β1 / h1 = 5.0 and β2 / h2 = 5.0 are obtained.

 (5) Next, the decoupling devices F1 and F2 are designed according to the transfer functions of the above-described equations (1) and (2). In this embodiment, the steady state (S = 0) is designed as described above.

  When the gain design unit 12 finishes setting the gains of the controllers C1 and C2 and the non-interactors F1 and F2 as described above, the gain adjustment unit 5-1 includes the control object 2 and the gain adjustment unit 5- 1 is disconnected, and the controllers C1 and C2 and the control object 2 are connected to perform control.

  FIG. 10 is a diagram showing step responses when the gains of the controllers C1 and C2 are adjusted as described above and the non-interactors F1 and F2 are designed. In FIG. In order to avoid obscure the waveforms of FIG. 8, the step response of each channel is shown separately in (a) and (b).

  As shown in FIG. 10, the waveform is hunting at first, but is finally set to the target value.

  For comparison with FIG. 10, FIG. 11 shows step responses when the decoupling devices F1 and F2 are not provided and the gains of the controllers C1 and C2 are designed in the same manner as in the above embodiment. Yes.

  In FIG. 10 in which the non-interfering devices F1 and F2 are provided, it can be confirmed that hunting is also eliminated and good control is performed as compared with FIG.

(Embodiment 3)
FIG. 12 is a block diagram corresponding to FIG. 1 of another embodiment of the present invention. In this embodiment, the gains of the controllers C1 and C2 are adjusted by searching.

  That is, the gain adjusting means 5-2 of this embodiment has the gain searching means 17, and adjusts the gains of the controllers C1 and C2 by the gain searching.

  As described above, the proportional gain Kp ′ calculated from the apparent slope R ′ of the step response waveform is larger than the proportional gain Kp calculated from the true slope R excluding the influence of interference.

  Therefore, in this embodiment, the proportional gain Kpi of the controller of channel i is searched below the proportional gain Kpi ′ calculated from the apparent slope R ′.

  First, the procedure from acquiring the step response waveform of the controlled object 2 to obtaining the apparent slope Ri ′ and the dead time is the same as in the above-described embodiment.

  That is, the connection switches 7 and 8 between the controlled object 2 and the controllers C1 and C2 are turned off, and the controlled object 2 is connected to the gain adjusting means 5-2 side.

  The MV setting means 9 adds stepwise operation amounts U1 and U2 for each channel, and the PV acquisition means 10 acquires the control amounts PV1 and PV2 at that time.

  From the acquired step response waveform, the apparent slope R ′ is obtained as described above, and the dead time L is obtained.

  Next, a gain calculated from the apparent slope R ′ and the dead time L is set as an initial value. This gain setting is performed using a conventional method such as the ZN method, as in the above-described embodiment.

  Next, the gain searching means sets the channel 1 gain candidate in the controller C1.

The gain search means 17 inputs a step-like target input as the target value r ₁ for channel 1. The gain search means 17 calculates the deviation sum of squares (Σ (r ₁ −θ ₁ ) ² ) of the channel 1 as an evaluation index, and stores the result.

  As a proportional gain candidate for channel 1, the proportional gain is changed little by little until it reaches zero. The integral gain and differential gain of channel 1 are not changed.

  When the series of searches is completed, the proportional gain with the smallest deviation square sum is set as Kp in the controller C1 of the channel 1.

  Next, for channel 2, the same search is performed, and the proportional gain with the smallest deviation square sum is set as Kp in the controller C2 of channel 2.

  This completes the gain search.

  Here, the method of gradually decreasing the proportional gain is shown, but an optimal proportional gain may be searched in the range of 0 to Kpi ′ by other search methods such as GA and PSO.

  In particular, a more efficient gain search can be performed by performing a gain search around the proportional gain calculated using true R as an optimal proportional gain candidate.

  In each of the above-described embodiments, the gain adjusting means is incorporated in the temperature regulator. However, as another embodiment of the present invention, the gain adjusting device is configured as a gain adjusting device and as a device independent of the temperature regulator, and by communication or the like. A gain may be set for the temperature controller.

  In each of the above-described embodiments, description has been made by applying to control of the temperature of a control target. However, the present invention is not limited to temperature, and can be applied to control of other physical quantities such as pressure, flow rate, and liquid level. .

  Of course, the present invention can also be applied to a multi-loop type control device.

  The present invention is useful for a control device such as a temperature controller for controlling a state of a controlled object, for example, a temperature state.

It is a block diagram of the temperature control system which concerns on one embodiment of this invention. It is a figure for demonstrating a step response. It is a figure which shows an output when step input is given to one channel. It is a step response waveform showing an apparent slope R 'and a true slope R. It is a figure which shows an output when the step input of the other channel is given. It is a figure which shows the step response waveform of embodiment of FIG. It is a block diagram of the temperature control system which concerns on other embodiment of this invention. It is a block diagram of the model structure of FIG. FIG. 9 is a block diagram of FIG. 8 corresponding to a steady state. It is a figure which shows the control waveform of embodiment of FIG. It is a figure which shows the control waveform corresponding to FIG. 10 of embodiment of FIG. It is a block diagram of the temperature control system which concerns on other embodiment of this invention. It is a block diagram of a temperature control system. It is a block diagram of a prior art example. It is a figure which shows a step response waveform. It is a figure which shows an output when step input is given to one channel. It is a figure which shows an output when the step input of the other channel is given. It is a time chart which shows a response | compatibility with an input and an output. It is a figure which shows the step response waveform of a prior art example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Temperature controller 2 Control object 5,5-1,5-2 Gain adjustment means C1, C2 controller F1, F2 Decoupling device

Claims (21)

In a multi-channel control device that calculates and outputs a plurality of operation amounts based on control amounts from a plurality of detection units that respectively detect physical states of control targets,
When the manipulated variable of the channel for which the proportional gain is to be calculated is changed, a proportional gain smaller than the proportional gain obtained based on the response of the controlled variable only for the channel with the manipulated variable is changed to the proportional gain of the channel. A control device. In a multi-channel temperature controller that calculates and outputs a plurality of manipulated variables based on each detected temperature from a plurality of temperature detecting means that respectively detect the temperature of the controlled object,
When the manipulated variable of the channel for which the proportional gain is to be calculated is changed, a proportional gain smaller than the proportional gain obtained based on the response of the detected temperature of only the relevant channel whose manipulated variable is changed is the proportional gain of the relevant channel. A temperature controller characterized by: In a multi-channel temperature controller that calculates and outputs a plurality of manipulated variables based on each detected temperature from a plurality of temperature detecting means that respectively detect the temperature of the controlled object,
When the manipulated variable of the channel for which the proportional gain is to be obtained is changed, the proportional gain obtained based on the response of the detected temperature of the channel that has changed the manipulated variable and the response of the detected temperature of a channel other than the channel is A temperature controller characterized by the proportional gain of the channel.   The temperature adjustment according to claim 3, wherein the proportional gain of the channel is a proportional gain smaller than the proportional gain obtained based on the response of the detected temperature of only the channel when the operation amount of the channel is changed. vessel.   The temperature controller according to claim 2 or 4, wherein the proportional gain of each channel is set to a smaller proportional gain as the interference of the control target is stronger. When the change in the detected temperature of the j-th channel when the manipulated variable of the i-th channel is changed is PVji, the proportional gain Kpi of the i-th channel, which is the channel, satisfies the following expression. Item 6. The temperature controller according to Item 2, 4 or 5.
Kpi ≦ Kpi ′ × PVii / (PV1i + PV2i + ... PVii + ... PVni)
here,
Kpi ': Proportional gain calculated using only the detected temperature of the channel with the manipulated variable changed Kpi: Proportional gain n calculated with the detected temperature other than the channel with the manipulated variable changed n: Maximum number of channels When the gradient of the change in the detected temperature of the j-th channel when the manipulated variable of the i-th channel is changed is Rji, the proportional gain Kpi of the i-th channel that is the channel satisfies the following equation: The temperature regulator according to claim 2, 4 or 5.
Kpi ≦ Kpi ′ × Rii / (R1i + R2i + ... Rii + ... Rni)
here,
Kpi ′: Proportional gain calculated using only the detected temperature of the channel with the manipulated variable changed Kpi: Proportional gain n calculated with the detected temperature other than the channel with the manipulated variable changed n: Maximum number of channels   The temperature controller according to any one of claims 2, 4 to 7, wherein a proportional gain of each channel is reduced in accordance with an increase in the number of the channels.   The proportional gain, integral gain, and differential gain of the channel are set to Kp, Ki, Kd, and the proportional gain, integral obtained based on the response of the detected temperature of the channel only when the operation amount of the channel is changed. When Kp ′, Ki ′, and Kd ′ are gains and differential gains, Kp ′> Kp and Kp ′ / Kp> Ki ′ / Ki or Kp ′ / Kp> Kd ′ / Kd The temperature controller as described in any one of Claim 2, 4-8 satisfying these.   Gain search means for determining the quality of the gain based on the evaluation index and searching for the gain of each channel, wherein the gain search means changes the proportional gain of the channel and the manipulated variable of the channel. 5. The temperature controller according to claim 2, wherein a search is made for a value smaller than a proportional gain obtained based on a detected temperature response of only the channel as a search range. Decoupling means for eliminating or reducing the influence of the control by each channel on the control by other channels,
The temperature according to any one of claims 2 to 10, wherein the non-interference means feeds back a difference between the detection temperatures to at least one of the input side and the output side of each channel so as to cancel the interference. Regulator. In a device for adjusting the gain of each channel of a control device of a plurality of channels that calculates and outputs a plurality of operation amounts based on control amounts from a plurality of detection units that respectively detect physical states of control targets,
When the manipulated variable of the channel for which the proportional gain is to be calculated is changed, the proportional gain of the channel is set to a proportional gain that is smaller than the proportional gain that is obtained based on the response of the controlled variable of the relevant channel that has changed the manipulated variable. A gain adjusting device for adjusting. In an apparatus for adjusting the gain of each channel of a multi-channel temperature controller that calculates and outputs a plurality of manipulated variables based on each detected temperature from a plurality of temperature detecting means that respectively detect the temperatures to be controlled ,
When the operation amount of the channel for which the proportional gain is to be calculated is changed, the proportional gain of the channel is set to a proportional gain that is smaller than the proportional gain that is obtained based on the response of the detected temperature of only the channel that has changed the operation amount. A gain adjusting device for adjusting. In an apparatus for adjusting the gain of each channel of a multi-channel temperature controller that calculates and outputs a plurality of manipulated variables based on each detected temperature from a plurality of temperature detecting means that respectively detect the temperatures to be controlled ,
When the manipulated variable of the channel for which the proportional gain is to be calculated is changed, the proportional gain obtained based on the response of the detected temperature of the channel that has changed the manipulated variable and the response of the detected temperature of the channel other than the channel is A gain adjusting device for adjusting a proportional gain of a channel.   The gain according to claim 14, wherein the proportional gain of the channel is adjusted to a proportional gain smaller than the proportional gain obtained based on the response of the detected temperature of only the channel when the operation amount of the channel is changed. Adjustment device.   The gain adjusting device according to claim 13 or 15, wherein the proportional gain of each channel is adjusted to a smaller proportional gain as the interference of the control target is stronger. When the change in the detected temperature of the j-th channel when the manipulated variable of the i-th channel is changed is PVji, the proportional gain Kpi of the i-th channel that is the channel satisfies the following expression. The gain adjusting device according to claim 13, 15 or 16, wherein
Kpi ≦ Kpi ′ × PVii / (PV1i + PV2i + ... PVii + ... PVni)
here,
Kpi ': Proportional gain calculated using only the detected temperature of the channel with the manipulated variable changed Kpi: Proportional gain n calculated with the detected temperature other than the channel with the manipulated variable changed n: Maximum number of channels When the gradient of change in the detected temperature of the j-th channel when the manipulated variable of the i-th channel is changed is Rji, the proportional gain Kpi of the i-th channel that is the channel satisfies the following expression: The gain adjusting device according to claim 13, 15 or 16, wherein the gain adjusting device is adjusted so as to.
Kpi ≦ Kpi ′ × Rii / (R1i + R2i + ... Rii + ... Rni)
here,
Kpi ': Proportional gain calculated using only the detected temperature of the channel with the manipulated variable changed Kpi: Proportional gain n calculated with the detected temperature other than the channel with the manipulated variable changed n: Maximum number of channels   The gain adjusting device according to any one of claims 13, 15 to 18, wherein the proportional gain of each channel is adjusted to be small in accordance with an increase in the number of the channels.   The proportional gain, integral gain, and differential gain of the channel are set to Kp, Ki, Kd, and the proportional gain, integral obtained based on the response of the detected temperature of the channel only when the operation amount of the channel is changed. When Kp ′, Ki ′, and Kd ′ are gains and differential gains, Kp ′> Kp and Kp ′ / Kp> Ki ′ / Ki or Kp ′ / Kp> Kd ′ / Kd The gain adjusting device according to any one of 13, 15 to 19, wherein the gain is adjusted so as to satisfy the above.   Gain search means for determining the quality of the gain based on the evaluation index and searching for the gain of each channel, wherein the gain search means changes the proportional gain of the channel and the manipulated variable of the channel. The gain adjusting apparatus according to claim 13 or 15, wherein a value smaller than a proportional gain obtained based on a detected temperature response of only the channel is searched as a search range to adjust the proportional gain.

Images