JPH07334249A - Method for deciding parameter of temperature control - Google Patents

Abstract

PURPOSE:To accurately keep the temperature to be controlled at the target temperature by decreasing the influence of an integral value including an initial error. CONSTITUTION:When a process temperature is lower than a critical value (b) of lowness, a deviation coefficient and an integration coefficient are calculated (S2-S4) by assuming process temperature as (b). Then, an index value alphashowing the lowness of a velocity coefficient and the deviation coefficient is calculated (S5) and when the index value alpha is larger than 1, respective parameters are obtained by performing the calculation of corrected deviation coefficient=EXalpha, corrected velocity coefficient=DELTAEXalpha and corrected integration coefficient=I/alpha<x> by defining a value X as >=1.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention controls the temperature of a controlled object based on the inferred value of a fuzzy rule based on the deviation between the temperature of the controlled object and the target temperature and the rate of change of the deviation, and the integrated value of the deviation. The present invention relates to a system, and more particularly to a temperature control parameter determination method for determining a deviation coefficient, a speed coefficient, and an integration coefficient.

[0002]

2. Description of the Related Art In the case of controlling the temperature of a dead time + second-order lag system control object based on an inferred value of a fuzzy rule and an integrated value of a deviation, data relating to the temperature of the control object and the fuzzy rule are associated with each other. It is necessary to determine the coefficient of deviation, the coefficient of change, and the coefficient of integration. In the prior art, these coefficients are determined by the method described below.

The heating amount applied to the controlled object is 100% to 0.
The difference between the peak temperature reached by the controlled object after the time t _cut changed to%, and the tuning temperature T _tuning which is the temperature of the controlled object at the time t _cut is _calculated as
Detected as _{(os) tuning} . Further, a row passage period L _{(os) tuning} , which is a period from the time t _cut to the detection of the row passage temperature T _{(os) tuning,} is detected.

From the tuning temperature T _tuning , the target temperature SV of the controlled object, the _running temperature T _{(os) tuning} , and the _running period L _{(os) tuning} , a value a, which is based on the characteristic of the controlled object,
c and d, a = 200, c = 0.6, d = 0.0037
5, the deviation coefficient E and the integration coefficient I are

[0005]

[Equation 5] E = (a + T _tuning −SV) × T _{(os) tuning} /
(A−T _{(os) tuning} ) I = c / {[1 + d × (T _tuning −SV + E)] × E ×
L _{(os) tuning} } is calculated.

Further, the values g and h determined from the speed coefficient ΔE indicating the deviation speed at the time t _cut and the deviation coefficient E corresponding to the characteristics of the controlled object are set as g = 8 and h = 0.25.
The index value α indicating the smallness of the speed coefficient ΔE and the deviation coefficient E is

[0007]

## EQU6 ## Calculation is performed by α = max (g / E, h / ΔE).

As a result of the above calculation, when the index value α is 1 or more, at least one of the deviation coefficient E and the speed coefficient ΔE is small, and is susceptible to noise and the like. Therefore, in this case, correct the small value,
For the purpose of reducing the influence of noise, the modified deviation coefficient E
_new and modified speed coefficient ΔE _new and modified integration coefficient I _new

[0009]

[Equation 7] E _new = E × α ΔE _new = ΔE × α I _new = I / α

[0010]

In the control of the controlled object, the case of α ≧ 1 will be described. When the temperature of the controlled object becomes a temperature within the range of the modified deviation coefficient E _new , the deviation is integrated. Be started. Further, when the temperature of the controlled object changes along the straight line 33 (see FIG. 7) whose slope is shown as the corrected deviation coefficient E _new , the integrated value of the deviation shows a value without error. However, the temperature change actually occurring in the controlled object does not show the change along the straight line 33, and the curve 3
As shown by 4, the rate of increase decreases as the temperature approaches the target temperature SV.

Therefore, the integrated value is the sum of the area of the area 31 indicated by the diagonal lines and the area of the area 32 indicated by the vertical lines. That is, the integrated value includes an error indicated as the area of the area 32 at the start of integration. Moreover, this error is not corrected in the subsequent control.

As described above, the integrated value includes an error at the start of the control, and this error affects the control without being corrected thereafter. As a result, the accuracy of control is reduced, and there is a problem in that the width of hunting that occurs at the temperature of the controlled object increases.

The present invention was devised to solve the above problems, and its purpose is to maintain the temperature of a controlled object at a target temperature with high accuracy by reducing the influence of an integral value including an error. It is to provide a method of determining a parameter of temperature control that can be performed.

[0014]

In order to solve the above-mentioned problems, the method for determining the parameters of the temperature control according to the present invention has a dead time +
The operation amount given to the controlled object of the secondary delay system is 100% to 0
% Time was changed to t _cu and peak temperature controlled object has reached the after _t, the time t _cut differences rows overtemperature of tuning the temperature T _tuning is the temperature of the controlled object in T
_{(os) tuning} is detected, and the _transition temperature T from the time t _cut
Passing period L, which is the period until _{(os) tuning} is detected
_{(os) tuning} is detected, and when the overpass temperature T _{(os) tuning} is smaller than the value b set as the limit of the smallness of the overpass temperature T _{(os) tuni ng.}

[0015]

And Equation 8] T _{(os) tuning} = b, and a tuning temperature T _tuning the target temperature of the controlled object SV and line over-temperature T _{(os) tu ning} a row over the period L _{(os) tuning,} the value a, c , D are values based on the characteristics of the controlled object, the deviation coefficient E and the integration coefficient I are

[0016]

[Equation 9] E = (a + T _tuning −SV) × T _{(os) tuning} /
(A−T _{(os) tuning} ) I = c / {[1 + d × (T _tuning −SV + E)] × E ×
L _{(os) tuning} }, and from the speed coefficient ΔE and the deviation coefficient E indicating the deviation speed at time t _cut , the values g and h are determined as values corresponding to the characteristics of the controlled object, and the speed coefficient ΔE and the deviation The index value α that indicates the smallness with the coefficient E is

[0017]

[Mathematical formula-see original document] When α = max (g / E, h / ΔE) and the index value α is 1 or more, the modified deviation coefficient E _new , the modified speed coefficient ΔE _new, and the modified integral coefficient I _new are calculated as follows. Value X is 1 or more

[0018]

[Equation 11] E _new = E × α ΔE _new = ΔE × α I _new = I / α ^x

[0019]

Operation: The overpass temperature T _{(os) tuning} is
When it is smaller than the value b set as the limit of the smallness of _{(os) tuning}

[0020]

[ _Equation 12] Since T _{(os) tuning} = b is set, even when the _passing temperature T _{(os) tuning} is small,

[0021]

[Equation 13] E = (a + T _tuning −SV) × T _{(os) tuning}
The deviation coefficient E shown as / (a−T _{(os) tuning} ) does not become small. Therefore, the deviation coefficient E does not become small even when the _passing temperature T _{(os) tuning} is small.

[0022]

(14) I = c / {[1 + d × (T _tuning −SV +
E)] × E × L _{(os) tuning} } is prevented from becoming a large value even when the row temperature T _{(os) tuning} is small.

The value X is set to a value of 1 or more, and the modified integration coefficient I _{new is set} to

[0024]

## EQU15 ## To obtain I _new = I / α ^x

[0025]

[Equation 16] E _new = E × α ΔE _new = ΔE × α I _new = I / α ^{x In} the modification of each value, the order of the index value α multiplied by the integral coefficient increases as an absolute value. To do.
That is, in the correction, the correction of the integration coefficient I becomes a higher-order correction in contrast to the correction of the deviation coefficient E and the speed coefficient ΔE. Therefore, as a relative relationship, the integration coefficient I will be compressed more strongly. Therefore, the influence of the error given to the integrated value at the initial stage of integration is reduced.

[0026]

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 3 shows an electrical configuration of a system to which an embodiment of the temperature control parameter determining method according to the present invention is applied, and this system performs pipe forming.

The extruder die 14, which needs to be heated and cooled by the heat generated by the resin when the pipe is molded, is a control target of the dead time + secondary delay system. Then, if necessary, it is heated by the heater 15 or the solenoid valve 1
It is cooled by water whose inflow is controlled by 6. The temperature of the extruder die 14 is detected by the thermocouple 11 and amplified by the amplifier 12, and then the A / D converter 13
Is converted into a digital signal at.

The control unit 17 to which this digital signal is guided
Is a block that controls the heater 15 or the solenoid valve 16 in accordance with the output of the A / D converter 13 to maintain the temperature of the extruder die 14 at the target temperature. The present embodiment is executed by this control unit 17.

The control unit 17 includes the extruder die 1
A data table in which a fuzzy rule configured by the deviation between the temperature of 4 and the target temperature and the changing speed of the deviation is developed in advance on the deviation-changing speed plane is stored. And
This data table is referred to via the deviation coefficient and the coefficient of change speed, and the temperature of the extruder die 14 is controlled based on the referred value and the integrated value of the deviation.

FIG. 1 is a flow chart showing an embodiment of the present invention, and FIG. 2 is a extruder die 1 during auto tuning.
It is explanatory drawing which shows the temperature change of No. 4. An embodiment of the present invention will be described below with reference to these drawings as necessary.

After setting the tuning temperature T _tuning for auto-tuning, the extruder die 14 is heated with an operation amount (heating amount) of 100%. Then, when the temperature of the extruder die 14 becomes equal to the tuning temperature T _tuning (time t _cut ), the operation amount is set to 0%.

The temperature of the extruder die 14 is such that the manipulated variable is 0%.
Shows a rise even after reaching the peak temperature T_{(peak) tuning}To
After reaching, it begins to descend. Peak temperature obtained at this time
Degree T _{(peak) tuning}And tuning temperature T_tuningDifference from
Over temperature T_{(os) tuning}To detect as. Also at time t
_cutFrom temperature T_{(os) tuning}Until is detected
Passing period L_{(os) tuning}To detect as. Also at time t
_cutThe rate of change of the temperature of the extruder die 14 in
It is detected as the number ΔE (step S1).

Then, the detected overpass temperature T
_{(os) tuning} is compared with the value b set as the limit of the smallness of the overpass temperature T _{(os) tuning} , and the overpass temperature T
_{When (os) tuning} is less than the value b (set to 1)

[0035]

## _EQU17 ## _Let T _{(os) tuning} = b. Further, if the line over-temperature T _{(os) tuning} is larger than the value b does not change the value of the row over the temperature T _{(os) tuning.} That is, the overpass temperature T _{(os) tuning} is set to a value of 1 or more (steps S2 and S3).

From the tuning temperature T _tuning , the target temperature SV of the extruder mold 14, the _running temperature T _{(os) tuning} , and the _running period L _{(os) tuning} , the values a, c, and d are obtained from the extruder mold 14.
As a value based on the characteristic of, the deviation coefficient E and the integration coefficient I are

[0037]

[Equation 18] E = (a + T _tuning −SV) × T _{(os) tuning}
/ (A−T _{(os) tuning} ) I = c / {[1 + d × (T _tuning −SV + E)] × E ×
L _{(os) tuning} }. In addition, in this example, from the experimental results

[0038]

The best value is obtained as a = 200 c = 0.6 d = 0.00375 (step S4).

From the speed coefficient ΔE and the deviation coefficient E,
The values g and h are determined in accordance with the characteristics of the extruder die 14, and the index value α indicating the smallness of the speed coefficient ΔE and the deviation coefficient E is set.

[0040]

[Mathematical formula-see original document] It is calculated as α = max (g / E, h / ΔE). In addition, in this example, from the experimental results

[0041]

The value g = 8 h = 0.25 is used as the best value (step S5).

When the index value α is 1 or less,
The deviation coefficient E, the speed coefficient ΔE, and the integration coefficient I are used as parameters. When the index value α is 1 or more, the correction deviation coefficient E _new , the correction speed coefficient ΔE _new, and the correction integration coefficient I _new are set to a value X of 1 or more.

[0043]

[Equation 22] E _new = E × α ΔE _new = ΔE × α I _new = I / α ^x , and correction deviation coefficient E _new and correction speed coefficient ΔE
_new and the modified integral coefficient I _new are used as parameters (steps S6 and S7).

As described above, the overpass temperature T _{(os) tuning} is
When it is smaller than the value b set as the limit of the smallness of this overpass temperature T _{(os) tuning}

[0045]

[ _Equation 23] Since T _{(os) tuning} = b, even when the _passing temperature T _{(os) tuning} is small,

[0046]

[Equation 24] E = (a + T _tuning −SV) × T _{(os) tuning}
The deviation coefficient E shown as / (a−T _{(os) tuning} ) does not become small. Therefore

[0047]

[Expression 25] I = c / {[1 + d × (T _tuning −SV +
E)] × E × L _{(os) tuning} }, the integration coefficient I has a large value while maintaining a predetermined relationship with the deviation coefficient E even when the _running temperature T _{(os) tuning} is small. Is prevented.

Further, the value X is set to a value of 1 or more, and the modified integration coefficient I _{new is set} to

[0049]

## EQU26 ## To obtain I _new = I / α ^x

[0050]

[Equation 27] E _new = E × α ΔE _new = ΔE × α I _new = I / α ^{x In} the modification of each value, the absolute value of the order of the index value α multiplied by the integration coefficient increases. It will be.

That is, in the correction, the correction of the deviation coefficient E and the speed coefficient ΔE is higher-order correction of the integration coefficient I. That is, as a relative relationship, the integral coefficient I is compressed more strongly. Therefore, the influence of the error given to the integrated value at the initial stage of integration is reduced.

In FIG. 4, the vertical axis represents the modified integral coefficient I _old obtained in the prior art and the horizontal axis represents the modified integral coefficient I _new obtained in the present embodiment. The relationship between these two types of modified integral coefficients is shown. , And the curve of 51 to 54 shows that the value X is 1.2,
The case where the values are 1.5, 1.8 and 2.0 is shown. Then, in the present embodiment, the value X is set to 1.5 from the experimental result.

FIG. 5 shows each value obtained in this example,
It shows a list with each value obtained in the prior art,
The characteristics of the extruder die 14 to be controlled are that the time required to reach 250 degrees is 5 when heating at 100%.
000 seconds, the quadratic coefficient of the temperature rise during heating is 0.
01.

Further, the target temperature SV and the tuning temperature T
Both _tuning is 150 degrees

[0055]

[Expression 28] Row passing temperature T _{(os) tuning} = 0.24 Row passing period L _{(os) tuning} = 26 Speed coefficient ΔE = 0.04 is obtained. Then, from these values, each value before correction is obtained, and each corrected value is obtained.

The values obtained in this example are

[0057]

[Equation 29] Modified deviation coefficient E _new = 8.00 Modified speed coefficient ΔE _new = 0.32 Modified integration coefficient I _new = 0.00102. When temperature control is performed using these values as parameters, extrusion is performed. The temperature of the machine die 14 changes as indicated by 41 in FIG. This temperature change, when compared to the temperature change in the prior art indicated by 42,
The temperature change is maintained at the target temperature with extremely high accuracy.

The present invention is not limited to the above embodiment,
The values a, b, c, d, g and h can be set to other optimum values according to the controlled object.

[0059]

According to the temperature control parameter determining method of the present invention, the _passing temperature T _{(os) tuning} is _equal to the _passing temperature T _{(os) tuning.}
When it is smaller than the value b set as the limit of the smallness of _{(os) tuning}

[0060]

[ _Mathematical formula-see original _document ] T _{(os) tuning} = b, and the values a, c, and d are values based on the characteristics of the controlled object, and the deviation coefficient E and the integration coefficient I are

[0061]

[Equation 31] E = (a + T _tuning −SV) × T _{(os) tuning}
/ (A−T _{(os) tuning} ) I = c / {[1 + d × (T _tuning −SV + E)] × E ×
L _{(os) tuning} }, and the values g and h are determined according to the characteristics of the controlled object, and the index value α indicating the smallness of the speed coefficient ΔE and the deviation coefficient E is set.

[0062]

[Expression 32] α = max (g / E, h / ΔE) is calculated, and the value X is set to a value of 1 or more.

[0063]

[Equation 33] Each parameter is obtained with E _new = E × α ΔE _new = ΔE × α I _new = I / α ^x . Therefore, the influence of the integrated value including the initial error is reduced, so that the temperature of the controlled object can be accurately maintained at the target temperature.

[Brief description of drawings]

FIG. 1 is a flow chart showing an embodiment of a temperature control parameter determination method according to the present invention.

FIG. 2 is an explanatory diagram showing a temperature change of a control target at the time of automatic tuning.

FIG. 3 is a block diagram showing an electrical configuration of a system to which an embodiment of a temperature control parameter determination method according to the present invention is applied.

FIG. 4 is an explanatory diagram showing a relationship between a modified integration coefficient I _{old in} the related art and a modified integration coefficient I _new in the present embodiment.

FIG. 5 is an explanatory diagram showing a list of various values obtained in each of the embodiment of the present invention and the conventional technique.

FIG. 6 is an explanatory diagram showing a temperature change of a controlled object according to an embodiment of the present invention and a temperature change of a controlled object according to a conventional technique.

FIG. 7 is an explanatory diagram showing an error in integration.

[Explanation of symbols]

14 Extruder mold to be controlled 17 Control unit for executing the present invention L _{(os) tuning} overpass period SV Target temperature T _tuning Tuning temperature T _{(os) tuning} overpass temperature

Claims (1)

[Claims] 1. A peak temperature reached by the control target after a time _tcut when the operation amount given to the control target of the dead time + secondary delay system is changed from 100% to 0%, and the control at the time _tcut . detecting a difference between the tuning temperature T _tuning is the temperature of the object as a line over-temperature T _{(os) tuning,} over the line is the period from the time t _cut to said line over-temperature T _{(os) tuning} is detected The period L _{(os) tuning} is detected, and the overpass temperature T _{(os) tuning} is the overpass temperature T.
When the value is smaller than the value b set as the limit of the smallness of _{(os) tuning} , T _{(os) tuning} = b, and the tuning temperature T _tuning , the target temperature SV of the controlled object, and the overpass temperature T _{From (os) tuning} and the line-passing period L _{(os) tuning} , the deviation coefficient E and the integration coefficient I are expressed by the following equations, where the values a, c, and d are values based on the characteristics of the controlled object. a + T _tuning −SV) × T _{(os) tuning} /
(A−T _{(os) tuning} ) I = c / {[1 + d × (T _tuning −SV + E)] × E ×
L _{(os) tuning} }, and the values g and h are determined as values corresponding to the characteristics of the controlled object from the speed coefficient ΔE and the deviation coefficient E indicating the deviation speed at the time t _cut . An index value α indicating the smallness of the coefficient ΔE and the deviation coefficient E is calculated as α = max (g / E, h / ΔE), and when the index value α is 1 or more,
The correction deviation coefficient E _new , the correction speed coefficient ΔE _new, and the correction integration coefficient I _new are set to a value X of 1 or more. E _new = E × α ΔE _new = ΔE × α I _new = I / α ^A method for determining a parameter for temperature control, which is characterized by obtaining as ^x .