JPH04817B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a mold temperature control device used for controlling mold temperature during continuous injection molding operations of an injection molding machine.

[Conventional technology]

In general, when using an injection molding machine, it is necessary to control the temperature of the mold with high accuracy in order to continuously produce stable and high quality products during continuous injection molding operations.

Incidentally, an injection molding machine produces a molded product by filling a cavity of a molding die with molten resin as a molding material.

At this time, the temperature T _R [℃] of the molten resin injected by the injection device is generally the temperature θ of the molding die.
higher than [℃]. Therefore, the temperature of the molding die increases and fluctuates each time the molten resin is injected.

However, in the past, in order to suppress such fluctuations in mold temperature, operators had to adjust the supply device for heat medium such as cooling water to adjust the temperature of the mold through trial and error until the mold stabilized at a predetermined target temperature. was being manipulated.

[Problem that the invention seeks to solve]

As described above, the conventional technology relies on trial and error by the operator without estimating fluctuations in mold temperature, resulting in a low product yield.

Furthermore, it is difficult to accurately control the temperature of the mold with general automatic control that treats such fluctuations in mold temperature simply as a disturbance.

In addition, in reality, the temperature of the mold is adjusted using cooling water, etc., which has a very slow response speed, as a heat medium. On the contrary, there was a problem in that the control resulted in unstable control.

Therefore, in view of the above-mentioned drawbacks, the technical problem of the present invention is to predict mold temperature fluctuations and compensate for them sequentially, and to create an injection molding machine suitable for quick and precise control of a heat medium with a slow response speed. An object of the present invention is to provide a mold temperature control device.

[Means for solving problems]

According to the present invention, a mold having a heat medium temperature control means for controlling the temperature of a heat medium in thermal contact with the mold for each injection molding cycle in order to adjust the temperature of the mold of an injection molding machine. In the temperature control device, a first temperature detection means detects the temperature of the molten resin injected into the mold of the injection molding machine as a first detected value, and a predetermined calculation formula is calculated from the first detected value. Based on the calculated time average temperature and a predetermined mold temperature target value, the mold temperature in the next injection molding cycle is calculated as the mold temperature target value. a mold temperature estimating means for calculating a command value to the heat medium temperature control means so that the temperature becomes the same; and after each injection molding cycle, the temperature of the mold in the injection molding cycle is detected as a second detected value. a second temperature detection means for determining the parameters of the calculation formula so as to correspond to the actual measurement value of the time average temperature from the actual measurement value of the time average temperature based on the second detected value and the calculated time average temperature; a correction means for correcting as a correction parameter; and a correction means for predicting the future value of the parameter of the arithmetic expression to be corrected based on the correction parameter successively corrected by the correction means, and using the future value as a correction parameter in the correction means. A correction operation predictive mold temperature control device is obtained, which is characterized by having a correction operation means.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a correction operation prediction mold temperature control device according to an embodiment of the present invention will be described with reference to the drawings.

As shown in FIG. 1, 1 is a mold, which is composed of a fixed mold and a movable mold. Molten resin is injected from an injection cylinder 2 into a mold 1, and a molded product is formed according to a predetermined injection molding cycle.

At this time, the mold temperature fluctuates due to the heat from the high temperature molten resin. Therefore, the cooling water temperature adjustment device 8, which is a heat medium temperature control means, drives the cooling water circulation pump 7 in accordance with the example cooling water temperature value Θ ^* from the mold temperature control device, which will be described later, to control the mold temperature in the next molding cycle. 1 becomes a predetermined mold temperature target value θ _r .

The mold temperature control device includes a first calculation section (not shown) that calculates the time average temperature of the mold 1, and a transmission line 1.
a mold temperature estimating device 14 comprising a second calculation unit (not shown) that receives the mold temperature target value from 6 and calculates a command value to the cooling water temperature adjustment device 8; a correction device 15 for correcting the calculation contents of step 14; and a correction operation for predicting future correction values based on the correction trends corrected in the correction device 15 from the past to the present and guiding the future correction values to the correction device 15. It has a prediction device 17.

Note that, as described later, the mold temperature estimation device 14 and the correction device 15 are realized by adaptive observation devices.

To explain specifically, first, mold temperature estimation device 1
4 is the future mold temperature for each injection molding based on the molten resin temperature T _R detected by the molten resin temperature sensor 10, which is the first detection means for detecting the temperature of the molten resin, and the set temperature Θ of the cooling water. Calculate and estimate the time average temperature θ _{i (av)} of the mold for each cycle, and find the cooling water temperature Θ ^* that can realize a predetermined mold temperature based on the time average temperature θ _{i (av).} , this cooling water temperature Θ ^* is output to the cooling water temperature adjusting device 8 as a command value Θ ^* . The cooling water temperature adjustment device 8 adjusts the cooling water temperature so that it follows the command value Θ ^* .

Here, the mold is uniquely developed using the calculation formula (Tanimura, Akashi; 30th Automatic Control Association Conference, preprint manuscript) that leads to the time-average temperature θ _{i (av)} and the calculation formula as a model. The arithmetic expressions based on the state space representation in a discrete time system with the injection molding cycle time as the sampling period will be explained below.

First, when modeling the thermal characteristics of the mold and approximating the mold temperature to the unsteady thermoelectricity within the calculation sphere, the following assumptions are made using FIG.

Fixed and movable molds 110 and 112 and fixed and movable platens 114 and 116
A heating device (not shown) is installed near the joint surface with the fixed mold and the movable mold 11 on the joint surface.
In order to compensate for heat dissipation from 0 and 112, this joint surface is treated as a heat insulating surface.

Heat transfer from the molten resin (not shown) to the fixed mold and movable molds 110 and 112 occurs only at the start of resin injection, so that the temperature plane of the molten resin has a uniform thickness in the vicinity of the molding cavity. , is assumed to be formed instantaneously on the parting surface of the mold.

The fixed mold and the movable mold 110 and 112 are
Assuming a symmetrical temperature distribution regarding the dividing surface, the dividing surface is treated as an adiabatic surface.

Heat radiation from the fixed mold and movable molds 110 and 112 to the surrounding outside air is ignored.

Assume a plane of cooling water temperature parallel to the dividing plane.

Based on the above main assumptions, as shown in FIG. 2, the areas of the isothermal surface of the molten resin temperature on the heat insulation surface 150 and the isothermal surface of the cooling water temperature near the in-mold cooling water pipe 132 are calculated as follows: A ₀ and A ₁ [m ² ], and the distance between the two surfaces is r ₁ [m].

Next, as shown in FIG. 3, when the fixed mold 110 is approximated to a sphere, the molding surface 130 of the fixed mold 110
The heat conduction between and the in-mold cooling water pipe 132 is treated as unsteady heat conduction, and the temperature distribution is expressed as A ₀ , A ₁ ,
Using r ₁ , it is defined as in the first equation. Furthermore, r ₀
[m] is a virtual quantity, and the temperature distribution is treated only in the radial direction.

r ₀ = r ₁ (A ₀ + √ ₀・₁ ) / (A ₁ − A ₀ ), A ₁ > A ₀
...(1) Next, the unsteady heat conduction equation of equation (1) is expressed by partial differential equations of equations (2) to (5).

∂(rθ)／∂t ₁ =k∂ ² (rθ)／∂r ² ...(2) θ _t1=0 =f ₁ (r) ...(3) θ _r=a =Θ, a=r ₀ +r ₁ ...(4) 0≦t ₁ ≦T ...(5) However, θ [℃] is the temperature of the mold, k [m ² /s] is the temperature conductivity, and Θ [℃] is the cooling water temperature. , f ₁ (r) [℃]
is the initial condition in the i-th cycle, T [s] is the injection period, and ΔV [m] is the thickness of the plane of the molten resin temperature.

When the partial differential equations of equations (2) to (5) are solved analytically for each cycle, the temperature θ at time t ₁ is the 6th
Formula (Kensuke Kawashita: Thermal Conduction Theory, p334, Kawade Shobo 1941)
become that way.

Note that f ₁ (r) is defined here using the final state θ _t-1=T in the i-1th cycle and equations (7) to (10).

f ₁ (r)=θ _t1-1=T , 0<r<r ₀ −Δr ……(7) f ₁ (r)=T _R , r ₀ −Δr≦r≦r ₀ ……(8) f ₁ (r)= _θt1-1=T , _r0 <r≦a...(9) _θtp=T =Θ...(10) However, T _R [°C] is the temperature of the molten resin.

Furthermore, regarding f ₁ (r), equations (11) to (13)
Perform interpolation as in Eq.

0=S ₀ <S ₁ <S ₂ ...S _N-1 <S _N ...(11) f ₁ (r)=f ₁ (S ₁ ), 0<r<S ₁ ...(12) f ₁ ( r)=( _f1 ( _SJ ) _−f1 ( _SJ+1 ))/( _SJ − _SJ+1 )・
r+(S _J・f ₁ (S _J+1 ) −S _J+1・f ₁ (S _J ))/(S _J −S _J+1 ), S _J < r < S _J+1 …
...(13) Using this, solve the integral of the second term of equation (7) analytically. Here, as an example of the calculation, the solution for the 10th cycle is shown in FIG.

Next, regarding the solution θ of equation (6), the time average temperature for each cycle θ _1(av) [°C] is determined by equation (14) below.
is required.

An example of this calculation is shown in FIG.

Note that the infinite series in Equations (6) and (14) above can be truncated at a finite number of terms in practice.
Furthermore, the degree of compatibility between the actually measured time-average temperature and the calculated value of equation (14) can be improved by adjusting kΔr.

Next, we will introduce arithmetic expressions using state space representations (Digital control of dynamic systems; GF Franklin, JD Powell, translated by Hiromasa Haneda, Morikita Publishing,
(1985), pp128-133).

First, the arithmetic expression using the state space expression in this example is Equation (14), which is an expansion of Equation (6) above, and
Furthermore, it converts from infinite dimension to finite dimension. This is shown in equation (15) below.

X _(i+1) =AX _(i) +bΘ _(i) +eT _R Y _(i)av =CX _(i) +dΘ _(i) +fT _R ...(15) Here, X _(i) , X _{(i+ 1)} are both estimated values,
Θ _(i) is the cooling water temperature in the i-th injection molding cycle, and T _R is the melting water temperature. The temperature of the resin, Y _(i)av , is the estimated value of the time-average temperature of the mold in the i-th injection molding cycle. That is, equation (15) is Y _(i)av . , X _(i+t) , X _(i) , Θ _(i) , _T Based on a simple calculation formula related to conduction, the time average temperature of the mold for each injection molding cycle is given.

Next, after each injection molding cycle, the correction device 15 detects the mold temperature sensor 1, which is a second temperature detection means.
Actual value of time-average temperature y _(i)av based on the mold temperature detected by 1. Time average temperature calculated as
Y _(i)av is input, and the parameters in the calculation contents of the mold temperature estimating device 14 are set as the actual measured value y _(i)av . Correct to correspond to. Furthermore, the mold temperature estimating device 14 recalculates the cooling water temperature Θ ^* that can realize a predetermined mold temperature every time the parameters are corrected.

In addition, if the fluctuation of the resin temperature T _R is small,
The resin temperature T _R can also be treated as a constant value, and the molten resin temperature sensor 10 can be omitted.

Next, the details of the parameter correction calculation performed by the correction device 15 will be specifically explained.

First, the mold temperature estimation device 14 (first calculation section)
The time average temperature of the mold Y _(i)av . is No. (15)
(16) using the state space expression below as well as Eq.
Expressed by the formula.

X _(i+t) = A(k)X _(i) +b(k)Θ _(i) +e(k)T _R Y _(i)av . _{= C (} ^k )
(k), b(k), C(k), d(k), e(k), and f(k) are matrices with k as a parameter.

On the other hand, the correction device 15 first measures the time-average temperature of the mold in the i-th injection molding cycle from the actual measurement value of the mold temperature from the mold temperature sensor 11 after the end of the i-th injection molding cycle. value
Y _(i)av ． seek. Next, the actually detected time-averaged temperature y _(i)av . and the time average temperature Y _(i)av calculated by the mold temperature estimation device 14. A correction amount Δk of the parameter k is calculated based on the difference between the parameters k and Equation (17) below.

Δk=−k(Y _(i)av .−y _(i)av .)/(Xm ₍₁₊₁₎ −Y _(i)av
．． )...(17) However, Xm ₍₁₊₁₎ is the m-th row element of the vector X ₍₁₊₁₎ , which is an arbitrary position where the mold temperature sensor 11 detects the temperature of the mold 1. is the estimated temperature of .

Here, Equation (17) is analytically obtained from Equation (14) for determining the time-averaged temperature θ _{i (av)} of the mold. That is, in the i-th injection molding cycle, the correction amount Δk of the parameter k is Δk=−(Y _(i)av .−y _(i)av .)/(∂Y _(i)av ./∂k)
...(18) is obtained as. At this time, according to equation (14), ∂Y _(i)av . Since /∂k can be obtained analytically, by substituting it into equation (18), equation (17) is obtained.

In addition, in the correction rule of equation (17), the correction device 15
If the sensitivity of the correction operation is too high, 0<α<1
Using an appropriate number α, Δk=−αk・(Y _(i)av .−y _(i)av .)/(Xm ₍₁₊₁₎ −Y _(i
)av ． )...(19) may also be used.

As a result, the correction device 15 first corrects the parameter k in the calculation content (formula (17)) of the mold temperature estimating device 14 to k+Δk using Δk obtained by equation (17). At the same time, the mold temperature estimating device 14 recalculates the cooling water temperature Θ ^* (command value) that can realize the target value θr of the predetermined mold temperature every time the parameter k is updated, and the cooling water temperature adjusting device 8 is output.

The command value Θ ^* is calculated based on the following equation (20).

X ^* =A(k)X ^* +b(k)Θ ^* +e(k)T _R θ _r =C(k)X ^* +d(k)Θ ^* +f(k)T ^R ...(20) However, ^* is a vector of the same size as X _(j+1) .

In this way, when the correction device 15 corrects the parameter k of the mold temperature estimating device 14, the mechanism for estimating the state variable (X _(1+t) ) while identifying the unknown parameter (k+Δk) is adapted to the observed value. (Adaptive control; Kunihiko Ichikawa et al., Shokodo, (1984) pp.54).

Here, after correction of the calculation parameter k several times, if the tendency of correction of the unknown parameter of the adaptive observer is like the slope shown in Fig. 6, then the future value of the unknown parameter is determined based on the known value18. We get k ^* 19. As a result, the estimation process in the sample period of the shaded area 20 in FIG. 6 is no longer necessary, and the parameter k can be directly replaced with k ^* , which will be estimated in the future, in equation (16) using the state space expression mentioned above. I can do it. This is expressed in equation (21).

X _(j+1) = A (k ^* )・X _(j) + b (k ^* ) Θ _(j) + e (k ^* ) T _R Y _(j) = C (k ^* )・X _(j) + d ( k ^* ) Θ _(j) + f (k ^* ) T _R …
...(21) However, X _(j) and X _(j+1) are the jth and j+
A state vector expressing the temperature distribution in the mold just before injection in the first injection molding cycle, Θ _(j) is the cooling water temperature in the j-th injection molding cycle, and Y _(j) is the j-th injection molding cycle. is the estimated value of the time-average temperature of the mold at .

In the correction operation prediction device 17, the correction device 15 uses k ^* as a correction calculation parameter in the mold temperature estimation device 1.
4 to be sent. Mold temperature estimation device 1
Step 4 recalculates the cooling water temperature Θ ^* that can realize the mold temperature target value θr. Θ ^* is calculated by replacing the parameter k with k ^* in the above equation (20).

As described above, in this embodiment, the mold temperature estimating device takes the time average of the mold temperature that changes from moment to moment in real time. A stable temperature value (time average temperature) can be used as a future value without considering each temperature change of the mold as a disturbance. In addition, a correction device is used to correct the calculation parameters used to determine the future value for each injection molding cycle based on the actual measured value of the mold temperature. Can be a value. Furthermore, with the correction motion prediction device,
Since the future correction value is predicted based on the past correction tendency of the correction device, it is possible to set the future value to be a value that will make the predetermined mold temperature reach the target value more quickly. Based on this future value, a precise command value is given to the control means that controls the temperature of the heat medium that is in thermal contact with the mold so that the mold reaches a predetermined target temperature in the next injection molding cycle. be able to. That is, it is possible to perform precise control that matches the characteristics of a heat medium such as cooling water, which is an actual control target that has a slow response speed.

〔Effect of the invention〕

As described above, according to the present invention, mold temperature control of an injection molding machine is suitable for rapid and precise control of a heat medium with a slow response speed by predicting mold temperature fluctuations and sequentially compensating for them. equipment can be provided.

[Brief explanation of drawings]

Fig. 1 is a block diagram of an injection molding machine equipped with a mold temperature control device that is an embodiment of the present invention, Fig. 2 is a simplified perspective view of the mold, and Fig. 3 is a fixed mold with a sphere. Figure 4 is a diagram showing the change in temperature distribution inside the mold in the 10th cycle, Figure 5 is a model diagram showing the change in temperature distribution in the mold in the 10th cycle, and Figure 5 is the model diagram showing the temperature distribution in the mold in the 1st to 6th injection molding cycles according to equation (14). FIG. 6 is a diagram showing a calculation example of the time-average temperature θ _1(av) of the mold, and a diagram showing the tendency of correction of calculation parameters based on the actual measured value of the mold temperature. In the figure, 1: Mold, 2... Injection cylinder, 7...
Cooling water circulation pump, 8...Cooling water temperature adjustment device, 9
... Cooling water flow, 10 ... Molten resin temperature sensor, 11 ... Mold temperature sensor, 14 ... Mold temperature estimation device, 15 ... Correction device, 16 ... Mold temperature target value, 17 ... Correction operation prediction device, 18... Known calculation parameters, 19... Estimated values of calculation parameters, 20... Unnecessary time for estimation processing.

Claims (1)

[Scope of Claims] 1. A mold having a heat medium temperature control means for controlling the temperature of a heat medium in thermal contact with the mold for each injection molding cycle in order to adjust the temperature of the mold of an injection molding machine. The mold temperature control device includes a first temperature detection means for detecting the temperature of the molten resin injected into the mold of the injection molding machine as a first detected value, and a predetermined calculation based on the first detected value. The time average temperature of the mold for each injection molding cycle is calculated based on the formula, and the mold temperature in the next injection molding cycle is determined from the calculated time average temperature and a predetermined mold temperature target value. To achieve the mold temperature target value,
a mold temperature estimating means for calculating a command value to the heat medium temperature control means; and a second temperature detection means for detecting the temperature of the mold in the injection molding cycle as a second detected value after each injection molding cycle is completed. and correcting the parameters of the calculation formula as correction parameters so as to correspond to the actual measurement value of the time average temperature based on the actual measurement value of the time average temperature based on the second detected value and the calculated time average temperature. a correction means, and a correction operation prediction means for predicting future values of parameters of the arithmetic expression to be corrected based on correction parameters sequentially corrected by the correction means, and for making the future values as correction parameters in the correction means. A correction operation predictive mold temperature control device, comprising: 2. In the correction operation predictive mold temperature control device according to claim 1, the calculation formula uses a state space expression in a discrete time system with an injection molding cycle time as a sampling period. Predictive mold temperature control device. 3. The corrective operation predictive mold temperature control device according to claim 1, wherein the mold temperature estimation means is realized by an adaptive observation device.