JP5473298B2 - Temperature control device for injection machine heating cylinder - Google Patents

Description

  The present invention relates to a temperature control device for an injection machine heating cylinder, and more particularly to a temperature control device for an injection machine heating cylinder that can suitably update control parameters during operation of the injection machine heating cylinder.

2. Description of the Related Art Conventionally, there is known a temperature control device that reestimates a transfer function parameter of an injection machine heating cylinder during operation of the injection machine heating cylinder and updates a PID parameter of a PID controller based on the reestimated transfer function parameter. (See Patent Document 1).
JP 2000-218674 A

In the conventional temperature control device for an injection machine heating cylinder, in order to update the PID parameter, it is necessary to re-estimate the transfer function parameter of the injection machine heating cylinder.
However, always re-estimating the transfer function parameter of the injector heating cylinder even for small fluctuations in the operating condition results in an increase in the amount of calculation. To avoid this increase in the amount of calculation, the operator manually sets the transfer function parameter. However, it is difficult to determine which of the many transfer function parameters should be adjusted.
Therefore, an object of the present invention is to provide a temperature control device for an injection machine heating cylinder that can update a control parameter without re-estimating a transfer function parameter of the injection machine heating cylinder by adjusting only one parameter. It is to provide.

In a first aspect, the present invention relates to an adjustment parameter output device (50) that outputs one adjustment parameter δ that can be adjusted by an operator, a transfer function parameter of the injection machine heating cylinder (Y), and the adjustment parameter δ. Based on the control parameter calculator (31) for calculating the control parameter based on the target temperature r (t), the cylinder temperature y (t) of the injector heating cylinder (Y), and the control parameter, the heater operation amount u ( There is provided a temperature control device (100) for an injection machine heating cylinder, comprising a controller (32) for calculating t) and driving a heater (H) of the injection machine heating cylinder (Y).
In the above configuration, the controller (32) may be a PID controller or a model prediction controller.
In the temperature control device for an injection machine heating cylinder according to the first aspect, if the operator changes the adjustment parameter δ, the change is reflected in the control parameter. That is, if only one parameter is adjusted, the control parameter can be updated without re-estimating the transfer function parameter of the injector heating cylinder.

In a second aspect, the present invention provides the temperature controller (100) for an injection machine heating cylinder according to the first aspect, wherein the control parameter calculator (31) calculates the control parameter using a pole placement method. There is provided a temperature control device (100) for an injection machine heating cylinder characterized by being a pole placement method control parameter calculator (31).
In the temperature control device for the injection machine heating cylinder according to the second aspect, since the pole placement method is used, a large number of control parameters can be adjusted to one adjustment parameter δ.

In a third aspect, the present invention relates to a target temperature r (t), a cylinder temperature y (t), and a heater operation amount in the temperature controller (100) for an injection machine heating cylinder according to the first or second aspect. a control performance evaluator (80) that evaluates the control performance based on u (t), automatically adjusts the adjustment parameter δ based on the evaluation result, and passes it to the adjustment parameter output unit 50; The container (50) provides the temperature control device (100) for an injection machine heating cylinder, which outputs the adjustment parameter δ passed from the control performance evaluator (80).
In the temperature control device for the injector heating cylinder according to the third aspect, the adjustment parameter δ is automatically adjusted according to the actually measured operation state, so the adjustment parameter δ is optimized while the operation is continued, and the control parameter is always optimized. I can do it.

In a fourth aspect, the present invention relates to a transfer function parameter based on a heater operation amount u (t) and a cylinder temperature y (t) in the temperature controller (100) for an injection machine heating cylinder according to the third aspect. And a transfer function parameter estimator (40) passed to the control parameter calculator (31). The control performance evaluator (80) determines whether or not the automatically adjusted adjustment parameter δ is within the recommended adjustment range. A temperature control device (100) for an injection machine heating cylinder is provided, wherein the transfer function parameter estimator (40) newly estimates and determines a transfer function parameter if it is not within the recommended adjustment range.
In the temperature control device for an injection machine heating cylinder according to the fourth aspect, when the adjustment of the adjustment parameter δ is not enough, the transfer function parameter is estimated again, so that it can cope with a large change in the operating state. I can do it.

  According to the temperature control device for an injection machine heating cylinder of the present invention, it is possible to update the control parameter by adjusting only one adjustment parameter δ, and a simple and easy-to-understand adjustment can be realized. In addition, since the pole placement method is used, adjustment of a large number of control parameters can be performed in one adjustment parameter δ. Further, since the adjustment parameter δ is automatically adjusted according to the actually measured operating state, the control parameter can always be optimized while the operation is continued, and the stability of the control is improved. In addition, when the transfer function parameter largely fluctuates due to a change in outside air temperature or the like, the transfer function parameter is estimated again, so that it is possible to cope with a large change in the operating state. Since the transfer function parameters are not reestimated except when necessary, an increase in the amount of calculation can be avoided.

  Hereinafter, the present invention will be described in more detail with reference to embodiments shown in the drawings. Note that the present invention is not limited thereby.

FIG. 1 is a block diagram showing a temperature control device 100 for an injection machine heating cylinder according to an embodiment of the present invention.
This temperature control device 100 for the injection machine heating cylinder measures the cylinder temperature y (t) of the injection machine heating cylinder Y by the thermocouple 10 and calculates a deviation e (t) from the given target temperature r (t). The heater operation amount u (t) is calculated based on the deviation e (t) by the PID temperature controller 32, and the heater H of the injection machine heating cylinder Y is driven.

  The pole placement method control parameter calculator 31 is based on transfer function parameters such as the system gain K, dead time L, and time constant T passed from the transfer function parameter estimator 40 and the adjustment parameter δ passed from the adjustment parameter output unit 50. Then, a PID parameter which is a control parameter of the PID controller 32 is calculated using the pole placement method, and the PID parameter is passed to the PID controller 32.

  The transfer function parameter estimator 40 estimates transfer function parameters such as the system gain K, dead time L, and time constant T based on the heater operation amount u (t) and the cylinder temperature y (t). Methods for estimating transfer function parameters are well known. For example, it is disclosed by patent document 1 (Unexamined-Japanese-Patent No. 2000-218674). Then, the estimated transfer function parameter is passed to the pole placement method control parameter calculator 31.

  The estimation command output unit 70 gives an estimation command to the transfer function parameter estimator 40 so as to estimate the transfer function parameter according to an instruction from the operator.

  The adjustment parameter output unit 50 outputs the adjustment parameter δ that is initially set or the adjustment parameter δ passed from the control performance evaluator 80.

  The control performance evaluator 80 evaluates the control performance based on the deviation e (t) and the heater operation amount u (t), adjusts the adjustment parameter δ based on the evaluation result, and passes it to the adjustment parameter output unit 50. Also, a recommended adjustment range of the adjustment parameter δ is determined, it is determined whether or not the adjusted adjustment parameter δ is within the recommended adjustment range. If the adjusted adjustment parameter δ is not within the recommended adjustment range, the transfer function parameter estimator 40 is determined. Is given a re-estimation command so that transfer function parameter estimation is performed again.

FIG. 2 is a flowchart showing an operation of a preparation process performed by the temperature control device 100 for an injection machine heating cylinder to prepare for an operation for manufacturing a product.
In step S1, the transfer function parameter estimator 40 is a generally known system identification based on the heater operation amount u (t) and the cylinder temperature y (t) obtained by using offline data or simulation. The transfer function parameters such as the system gain K, dead time L, and time constant T are estimated using a method (such as the least square method), and the estimated transfer function parameters are passed to the pole placement method control parameter calculator 31.

…（７）
Ｋ_Pは比例定数、Ｔ_Iは積分定数、Ｔ_Dは微分定数である。
ｆ０＝ｐ１−（ａ１＋１）
ｆ１＝ｐ２−（ａ２−ａ１）
ｆ２＝ａ２
ｐ１＝−２・exp（−２・ρ）
ｐ２＝exp（−４・ρ）
ρ＝τ／σ
σ＝（Ｔ＋Ｌ）／δ
ただし、τ：サンプリング時間、Ｔ：時定数、Ｌ：むだ時間である。
ａ１，ａ２は、時定数Ｔやむだ時間Ｌを基に算出する。 In step S 2, the pole placement method control parameter calculator 31 includes an adjustment parameter δ that is preset in advance from the adjustment parameter output unit 50, a transfer function parameter passed from the transfer function parameter estimator 40, Based on the design polynomial based on the placement method, the PID parameter is calculated by the equation (7).

... (7)
K _P is a proportional constant, T _I is an integral constant, and T _D is a differential constant.
f0 = p1- (a1 + 1)
f1 = p2- (a2-a1)
f2 = a2
p1 = −2 · exp (−2 · ρ)
p2 = exp (−4 · ρ)
ρ = τ / σ
σ = (T + L) / δ
However, τ: sampling time, T: time constant, L: dead time.
a1 and a2 are calculated based on the time constant T and the dead time L.

…（８）
ただし、

λ：評価パラメータ
λ＝Ｋ² …（９）
である。
Δｕ（ｔ）は、調整パラメータδに基づい導出された（７）式に示すＰＩＤパラメータによって算出されており、ｅ²（ｔ）もｕ（ｔ）を用いて制御対象を制御した結果であるため、（８）式の評価関数Ｊ１は調整パラメータδに基づいた評価関数となっている。
評価パラメータλの値は（９）式以外の方法により決定してもよい。
（９）式のようにシステムゲインＫを考慮した評価パラメータλを用いると、（８）式にはシステムゲインＫが考慮されることになる。
（８）式の評価関数Ｊ１が最小となる調整パラメータδの値は、伝達関数パラメータによらず、概ね同じ値となる。
なお、調整パラメータδは、後述するように最適値に自動調整されるため、ここでは厳密に最小値になっている必要は無い。 In step S3, the control performance evaluator 80 obtains the evaluation function J1 of the equation (8) based on the heater operation amount u (t) and the cylinder temperature y (t) obtained using control offline data and simulation. The minimum value of the adjustment parameter δ is obtained and passed to the adjustment parameter output unit 50.

... (8)
However,

λ: Evaluation parameter λ = K ² (9)
It is.
Δu (t) is calculated by the PID parameter shown in the equation (7) derived based on the adjustment parameter δ, and e ² (t) is also a result of controlling the controlled object using u (t). (8) is an evaluation function based on the adjustment parameter δ.
The value of the evaluation parameter λ may be determined by a method other than the equation (9).
When the evaluation parameter λ in consideration of the system gain K is used as in the equation (9), the system gain K is considered in the equation (8).
The value of the adjustment parameter δ that minimizes the evaluation function J1 in the equation (8) is substantially the same value regardless of the transfer function parameter.
Note that the adjustment parameter δ is automatically adjusted to an optimum value as described later, and therefore does not need to be strictly minimum here.

In step S4, the control performance evaluator 80 determines recommended adjustment ranges δ1 and δ2 shown in equation (12).
δ1 ≦ δ ≦ δ2 (12)
For example, when the system gain K fluctuates from (1-w) · K to (1 + w) · K, the uncertainty w of the system gain K is considered,
λ1 = {(1-w) · K} ²
Is substituted for λ in equation (8), and the value of the adjustment parameter δ that minimizes the evaluation function J1 is δ1. Also,
λ2 = {(1 + w) · K} ²
Is substituted for λ in equation (8), and the value of the adjustment parameter δ that minimizes the evaluation function J1 is δ2.
Note that the recommended adjustment ranges δ1 and δ2 may be set by trial and error.

FIG. 3 is a flowchart showing the operation of the temperature control process executed during the operation for manufacturing the product by the temperature control device 100 for the injection machine heating cylinder.
In step S5, the PID controller 32 calculates the heater operation amount u (t) based on the deviation e (t) between the cylinder temperature y (t) and the target temperature r (t), and the heater operation amount u (t). Is output to the heater H of the injection machine heating cylinder Y.

…（１０）
そして、不感帯係数をγとするとき、λ／γ≦Ｊ２≦γ・λであれば、調整パラメータδが適正であると判定し、現在の調整パラメータδをそのまま調整パラメータ出力器５０へ出力する。そして、ステップＳ５に戻る。
λ／γ≦Ｊ２≦γ・λでなければ、調整パラメータδが不適正であると判定し、ステップＳ７へ進む。
なお、不感帯係数γは、不感帯を設ける場合はγ＞１とし、不感帯を設けない場合はγ＝１とする。 In step S6, the control performance evaluator 80 obtains an evaluation function J2 of equation (10).

(10)
When the dead zone coefficient is γ, if λ / γ ≦ J2 ≦ γ · λ, it is determined that the adjustment parameter δ is appropriate, and the current adjustment parameter δ is output to the adjustment parameter output device 50 as it is. Then, the process returns to step S5.
If λ / γ ≦ J2 ≦ γ · λ is not satisfied, it is determined that the adjustment parameter δ is inappropriate, and the process proceeds to step S7.
Note that the dead zone coefficient γ is γ> 1 when a dead zone is provided, and γ = 1 when a dead zone is not provided.

In step S7, when the adjustment width is Δδ, the control performance evaluator 80 sets a value obtained by reducing the current adjustment parameter δ by the adjustment width Δδ as a new adjustment parameter δ if λ / γ> J2. On the other hand, if J2> γ · λ, a value obtained by increasing the current adjustment parameter δ by the adjustment width Δδ is set as a new adjustment parameter δ. Then, the process proceeds to step S8.
For example, the adjustment width Δδ is set to 1% to 10% of the initially set adjustment parameter δ.

  In step S8, the control performance evaluator 80 determines whether or not the new adjustment parameter δ satisfies the expression (12). Proceed to On the other hand, if not satisfied, it means that the current transfer function parameter is incompatible with the current state. Therefore, a re-estimation command is given to the transfer function parameter estimator 40, and the process proceeds to step S10.

  In step S9, the pole placement method control parameter calculator 31 applies the adjusted adjustment parameter δ passed from the adjustment parameter output device 50, the transfer function parameter passed from the transfer function parameter estimator 40, and the pole placement method. Based on the design polynomial based thereon, the PID parameter is recalculated by the equation (7) as in step S2. Then, the process returns to step S5.

  In step S10, the transfer function parameter estimator 40 performs a generally known system identification method (minimum 2) based on the heater operation amount u (t) and the cylinder temperature y (t) obtained during operation. Transfer function parameters such as system gain K, dead time L, and time constant T are estimated using a multiplicative method and the like, and the estimated transfer function parameters are passed to the pole placement method control parameter calculator 31.

  In step S11, the pole placement method control parameter calculator 31 receives the preset initialization parameter δ passed from the adjustment parameter output unit 50 and the transfer function parameter estimator 40, as in step S2. Based on the transfer function parameter and the design polynomial based on the pole placement method, the PID parameter is calculated by Equation (7). Then, the process returns to step S5.

According to the temperature control device 100 for the injection machine heating cylinder of the first embodiment, the following effects can be obtained.
(1) The PID parameter can be corrected during operation by adjusting only one parameter, the adjustment parameter δ.
(2) If adjustment of the adjustment parameter δ within the recommended adjustment range is sufficient, that is, small fluctuations in the operating state can be handled by adjusting only the adjustment parameter δ without re-estimating the transfer function parameter. Become.
(3) When adjustment of the adjustment parameter δ within the recommended adjustment range is not sufficient, that is, a large fluctuation in the operating state can be dealt with by re-estimating the transfer function parameter.

-Example 2-
If there is no significant fluctuation in the operating state, the transfer function parameter estimator 40 in FIG. 1 may be omitted. In this case, the transfer function parameter estimator 40 is added only at step S1 in FIG. Further, the processing in steps S8, S10, and S11 in FIG. 3 is omitted, and the process directly proceeds from step S7 to step S9.

-Supplementary explanation-
[Estimation of transfer function parameters]
Let the model of the injection machine heating cylinder Y be a “first order delay + dead time” model of the equation (1).
G (s) = K.exp (-L.s) / (1 + T.s) (1)
In Equation (1), K represents a system gain, L represents a dead time, and T represents a time constant.
Equation (1) is discretized to obtain equation (2).
A (z ⁻¹ ) · y (t) = z ⁻¹ · B (z ⁻¹ ) · u (t) + ξ (t) / Δ (2)
However,
A (z ⁻¹ ) = 1 + a1 · z ⁻¹ + a2 · z ^{−2
B (z -1) = b0 +} b1 · z -1
Δ = 1−z ⁻¹
y (t): cylinder temperature at time t u (t): heater operation amount at time t ξ (t): Gaussian white noise Transfer function parameters are estimated based on the above equation.

[Calculation of PID parameters]
The pole placement method uses Equation (3), which can set a pole based on responsiveness with one adjustment parameter δ.
P (z ⁻¹ ) = 1 + p1 · z ⁻¹ + p2 · z ⁻¹ (3)
p1 = −2 · exp (−2 · ρ)
p2 = exp (−4 · ρ)
ρ = τ / σ
σ = (T + L) / δ
However, τ: sampling time, T: time constant, L: dead time.

…（４）
ただし、Δｕ（ｔ）：ヒータ操作量の差分、Ｋ_Pは比例定数、Ｔ_Iは積分定数、Ｔ_Dは微分定数である。 When the adjustment parameter δ is increased, the rising speed becomes faster. When the adjustment parameter δ is decreased, the rise is delayed.
In addition to the formula (3), any pole placement method that can be adjusted with one parameter can be used.
The control system structure based on the pole placement method may be any system such as PID control or model predictive control. However, when PID control is used as in the present embodiment, the control system structure has a desirable pole of equation (3). PID parameters are obtained so that Formula (4) shows a general PID control law.

... (4)
Here, Δu (t): difference in heater operation amount, K _P is a proportional constant, T _I is an integral constant, and T _D is a differential constant.

…（５）
（５）式のＦ（ｚ^-1）は、（６）式のＤｉｏｐｈａｎｔｉｎｅ方程式を解くことにより求められる。

…（６） In order for the pole of the equation (3) to coincide with the PID parameter of the equation (4), an equation obtained by replacing B (z ⁻¹ ) of the equation (5) obtained by the generalized minimum variance control method with a steady gain ( 4) It is obtained by comparing the equations.

... (5)
F (z ⁻¹ ) in the equation (5) can be obtained by solving the Diophantine equation in the equation (6).

(6)

…（７） Since F (z ⁻¹ ) in the equation (5) is determined based on P (z ⁻¹ ) in the equation (3) determined by the adjustment parameter δ, the equations (4) and (5) are By comparing, the PID parameter having the pole of the formula (3) can be calculated.
Therefore, the PID parameter can be calculated by the equation (7) from the equations (3) to (6).

... (7)

  The temperature control device for an injection machine heating cylinder according to the present invention is useful for temperature control of the injection machine heating cylinder.

1 is a configuration block diagram illustrating a temperature control device for an injection machine heating cylinder according to Embodiment 1. FIG. It is a flowchart which shows the operation | movement of the preparation process in the temperature control apparatus of the injection machine heating cylinder which concerns on Example 1. FIG. It is a flowchart which shows the operation | movement of the temperature control process at the time of the driving | operation in the temperature control apparatus of the injection machine heating cylinder which concerns on Example 1. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Thermocouple 20 Differentiator 31 Pole arrangement method Control parameter calculator 32 PID controller 40 Transfer function parameter estimator 50 Adjustment parameter output device 70 Estimation command output device 80 Control performance evaluator 100 Temperature control device of injection machine heating cylinder H Heater Y Injection machine heating cylinder

Claims (3)

An adjustment parameter output device (50) for outputting an initially set adjustment parameter δ, and a control parameter calculator (31) for calculating a control parameter based on the transfer function parameter of the injection machine heating cylinder (Y) and the adjustment parameter δ. ), The target temperature r (t), the cylinder temperature y (t) of the injection machine heating cylinder (Y), and the control parameter, the heater operation amount u (t) is calculated, and the injection machine heating cylinder (Y) Control performance is evaluated based on the controller (32) that drives the heater (H), the target temperature r (t), the cylinder temperature y (t), and the heater operation amount u (t), and based on the evaluation result. A control performance evaluator (80) that automatically adjusts the adjustment parameter δ and passes the adjustment parameter δ to the adjustment parameter output unit (50), and the adjustment parameter output unit (50) Temperature controller of the injection machine heated cylinder and outputs a parameter adjustment parameters passed from the control performance evaluator (80) in place of [delta] [delta] (100). The temperature control device (100) for an injection machine heating cylinder according to claim 1, wherein the control parameter calculator (31) is a pole placement method control parameter calculator (31) that calculates the control parameter using a pole placement method. A temperature control device (100) for an injection machine heating cylinder, characterized in that The temperature control device (100) for an injection machine heating cylinder according to claim 1 or 2 , wherein a transfer function parameter is estimated based on a heater operation amount u (t) and a cylinder temperature y (t), and the control parameter is set. A transfer function parameter estimator (40) to be passed to the calculator (31) is provided, and the control performance evaluator (80) determines whether or not the automatically adjusted adjustment parameter δ is within the recommended adjustment range, and the recommended adjustment range. If not, the transfer function parameter estimator (40) causes the transfer function parameter to be newly estimated .

Images