JP2635931B2 - Mold temperature control method for glass forming machine - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[Industrial application] In mold temperature control in the glass manufacturing process, a change from a currently produced product type to a product type having a different mold temperature is performed in a short time, and the mold temperature during production is stabilized. The present invention relates to a method for fuzzy control of a mold temperature for carrying out the conversion.

[0002]

2. Description of the Related Art A general method of controlling the temperature of a mold in a glass manufacturing process is to install a pressure gauge in an air duct to measure the pressure in the duct and control the rotation speed of a suction damper or a motor of a fan to cool the mold. A method of controlling the amount of wind to be constant has been adopted (hereinafter, referred to as “prior art 1”).

Further, as disclosed in Japanese Patent Publication No. 2-1776 (hereinafter referred to as "Prior Art 2"), the flow rate of the cooling air is detected and compared with the set flow rate calculated by the following equation (1). Attempts have been made to control the amount of cooling air with an output based on the deviation.

V _SV = N (V _O + K _T T _A ) (1) where, V _SV : air flow set value V _O : reference air flow determined by the type of glass product K _T : air flow / temperature determined by the type of glass product factor T _a: cooling air temperature N: number of operating mold

Further, Japanese Patent Application Laid-Open No. Sho 62-119127 (hereinafter referred to as "prior art 3") discloses a mold temperature at which a specimen representing a mold is kept at a constant temperature with respect to air sampled from a duct. A cooling sensor is described.

The process characteristics of the mold temperature include disturbance factors (cooling air volume, cooling air temperature, air temperature, humidity, gob temperature, number of operating molds) and product factors (in the case where different products are produced on the same line). An ideal temperature control method has not been established in that it greatly changes depending on the influence of the above-mentioned factors) and equipment factors (the effects of production on different lines).

[0007] Further, there is no control method for changing to a production type having a different mold temperature in a short time, and the cooling air flow for obtaining the optimum mold temperature must rely on the judgment and operation based on the experience of a skilled operator. After setting the cooling air flow rate due to the difference in experience, there are cases where the product may wrinkle due to insufficient cooling, or the product surface may be warped due to overcooling, and it takes a considerable amount of time to find the optimal cooling air flow. Cost.

As in the prior art 1, the temperature of the mold during the production is controlled by a one-loop PI by controlling the air volume to be constant.
In the case of automatic control by D (PI) control, the influence of disturbance factors cannot be completely absorbed, and the mold temperature changes by about ± 9 ° C. in one day. This variation in mold temperature causes product defects such as wrinkles and undulations as described above. Therefore, the operator needs to adjust the air volume accordingly.

[0009] Suppose that the cooling air temperature,
Even if the air volume is program-controlled by the humidity and the number of operating molds, the mold temperature fluctuates due to other disturbance factors (gob temperature, air temperature) and product factors, resulting in poor reproducibility. In addition, in order to improve the control method, it is necessary to reconstruct from the beginning, and a considerable number of steps are required until completion.
Further, even when the equipment is different, it is necessary to perform tuning from the beginning, and the reality is poor. Even if a mold temperature cooling sensor is used as in the prior art 3, the above-mentioned problems cannot be avoided.

[0010]

The conventional mold temperature control method is performed as described above, and the present invention has been made to solve the above-mentioned problems. Using the equipment shown in the configuration, a mold temperature control method that enables accurate and easy control of the mold temperature is provided by a fuzzy control system that uses the experienced operator's empirical judgment and the driving operation process as a control law. Aim.

An object of the present invention is to provide a fuzzy control method of a mold temperature which enables accurate and easy control of a mold temperature by a multivariable fuzzy control system in which a constraint condition relating to mold temperature adjustment is added to a control law.

[0012]

According to the present invention, in order to achieve this object, glass melted in a glass melting furnace is fed from a forge to a gob forming section through a feeder to form a gob. In the method of controlling the temperature of a mold of a glass forming machine, the temperature of the mold is controlled by controlling the amount of cooling air of the mold by feeding this to a mold of the glass forming machine. A suction damper and a blower are installed in the duct to supply cooling air, and (b) a thermometer is installed in the air duct between the blower and the glass forming machine to measure the temperature (hereinafter referred to as "wind temperature"). (C) The temporal change of the difference between the current measured value and the previous measured value of the wind temperature is defined as the antecedent of the fuzzy inference rule, and the membership function and the inference rule of the fuzzy set are set in advance. Based on fuzzy inference, To estimate the mold temperature, and performs control of the mold cooling air flow.

According to the method of the present invention, in addition to the above, a pressure gauge is installed in a blower duct between a blower and a glass forming machine to measure a pressure (hereinafter, referred to as "source pressure"), and the blower and the glass forming machine are measured. Install a hygrometer in the ventilation duct between the two to measure the humidity (hereinafter referred to as "humidity") and measure the temporal change in the difference between the current and previous measured values of the original pressure and humidity. Fuzzy inference is performed as a prerequisite to the fuzzy inference rules, and fuzzy inference is performed based on the membership function of the fuzzy set and inference rules set in advance, the mold temperature is estimated, and the mold cooling air volume is controlled. And

According to the method of the present invention, in addition to the above, a thermometer is installed in the gob forming section to measure the temperature of the glass supplied to the mold (hereinafter, referred to as "gob temperature") and to measure the temperature of the glass forming machine. Install a thermometer in the vicinity and check the temperature (hereinafter, “temperature”
), And the temporal change in the difference between the current measurement value and the previous measurement value of the above gob temperature and temperature is regarded as the antecedent of the fuzzy inference rule, and a member of a fuzzy set that is set in advance. The fuzzy inference is performed based on the ship function and the inference rule, the mold temperature is estimated, and the mold cooling air volume is controlled.

In addition to the above, the method of the present invention detects the flow rate of glass supplied to the glass forming machine per unit time, and installs a section working number detector in the glass forming machine to detect the section working number. Then, the time variation of the difference between the current measurement value and the previous measurement value of each of the glass flow rate and the section operation number is set as the antecedent of the fuzzy inference rule, and the membership function of the fuzzy set is set in advance. And fuzzy inference based on the inference rule to estimate the mold temperature and control the mold cooling air flow.

In the method of the present invention, the glass melted in a glass melting furnace is fed from a fore hearth to a gob forming section via a feeder to form a gob, which is fed to a mold of a glass forming machine. In the mold temperature control method for a glass forming machine in which the amount of cooling air of the mold is controlled to control the temperature of the mold, (a) a suction damper and a blower are installed in a ventilation duct, and cooling is performed. Supply the wind,
(B) A thermometer is installed in the air duct between the blower and the glass forming machine to measure the temperature (hereinafter referred to as “wind temperature”),
(C) The temporal change in the difference between the current measured value and the previous measured value of the wind temperature is defined as the antecedent of the fuzzy inference rule, and is based on the membership function of the fuzzy set and the inference rule set in advance. Then, fuzzy inference is performed to estimate the mold temperature, and (d) the temporal change amount of the difference between the current estimated value and the previous estimated value of the estimated mold temperature is defined as the antecedent of the fuzzy inference rule. A fuzzy inference is performed based on a membership function of a fuzzy set and an inference rule set in advance, and (e) a deviation amount between the estimated mold temperature and the set temperature is set as a precondition of the fuzzy inference rule. Fuzzy inference is performed based on the set membership function of the fuzzy set and the inference rule, and (f) a deviation value between the changed set temperature and the set temperature is set as an antecedent of the fuzzy inference rule, and a preset fuzzy inference rule is set. Membership of the set Based on the functions and inference rules, performs a fuzzy inference, and performs control of the mold cooling air flow.

According to the method of the present invention, in addition to the above, a pressure gauge is installed in a blower duct between a blower and a glass forming machine to measure a pressure (hereinafter, referred to as "source pressure"). Install a hygrometer in the ventilation duct between the two to measure the humidity (hereinafter referred to as "humidity") and measure the temporal change in the difference between the current and previous measured values of the original pressure and humidity. Fuzzy inference is performed as a prerequisite to the fuzzy inference rules, and fuzzy inference is performed based on the membership function of the fuzzy set and inference rules set in advance, the mold temperature is estimated, and the mold cooling air volume is controlled. And

According to the method of the present invention, in addition to the above, a thermometer is installed in a gob forming section to measure the temperature of the glass supplied to the mold (hereinafter referred to as "gob temperature") and to measure the temperature of the glass forming machine. Install a thermometer in the vicinity to determine the temperature (hereinafter, “temperature”
), And the temporal change in the difference between the current measurement value and the previous measurement value of the above gob temperature and temperature is regarded as the antecedent of the fuzzy inference rule, and a member of a fuzzy set that is set in advance. The fuzzy inference is performed based on the ship function and the inference rule, the mold temperature is estimated, and the mold cooling air volume is controlled.

In addition to the above, the method of the present invention detects the flow rate of glass supplied to the glass forming machine per unit time and installs a section operating number detector in the glass forming machine to detect the operating number of sections. Then, the time variation of the difference between the current measurement value and the previous measurement value of each of the glass flow rate and the number of operating sections is set as an antecedent of the fuzzy inference rule, and the membership of the fuzzy set is set in advance. It is characterized by performing fuzzy inference based on a function and an inference rule, estimating a mold temperature, and controlling a mold cooling air volume.

Further, according to the method of the present invention, the glass melted in the glass melting furnace is fed from a fore hearth to a gob forming section via a feeder to form a gob, which is fed to a mold of a glass forming machine. In the mold temperature control method for a glass forming machine in which the amount of cooling air of the mold is controlled to control the temperature of the mold, (a) a suction damper and a blower are installed in a ventilation duct, and cooling is performed. Supply the wind,
(B) A pressure gauge is installed in a duct between the blower and the glass forming machine to measure a pressure (hereinafter, referred to as “source pressure”), and (c)
Install a hygrometer in the duct between the blower and the glass forming machine,
Measure humidity (hereinafter referred to as "humidity"), (d) Install a thermometer near the glass forming machine, measure the temperature (hereinafter referred to as "temperature"), and (e) Last production of the same product The above-mentioned source pressure, humidity, the deviation amount between the current source pressure, humidity, and temperature and the current source pressure, humidity, and temperature are defined as the antecedent of the fuzzy inference rule, and based on a membership function and an inference rule of a fuzzy set that is set in advance. A fuzzy inference is performed to estimate a mold temperature, and a temporal change amount of a difference between the current estimated value and the previous estimated value of the estimated mold temperature is set as a precondition of the fuzzy inference rule, and a preset fuzzy inference rule is set. It is characterized by performing fuzzy inference based on the membership function of the set and inference rules, and controlling the die cooling air volume.

In the method of the present invention, in addition to the above, a thermometer is installed in a duct between a blower and a glass forming machine to measure a temperature (hereinafter referred to as "wind temperature"), and to measure the temperature of the same product during the previous production. The deviation between the above wind temperature and the current wind temperature is used as the antecedent of the fuzzy inference rule, and fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance to estimate the mold temperature. Then, the temporal change in the difference between the current estimated value and the previous estimated value of the estimated mold temperature is used as the antecedent of the fuzzy inference rule, and based on the membership function of the fuzzy set and the inference rule set in advance. And it is characterized by performing fuzzy inference and controlling the die cooling air volume.

In the method of the present invention, in addition to the above, a thermometer is installed in a gob forming section to measure the temperature of the glass supplied to the mold (hereinafter referred to as "gob temperature"), and to produce the same product in the previous production. The amount of deviation between the above-mentioned gob temperature and the current gob temperature is taken as the antecedent part of the fuzzy inference rule, and fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance, and the mold temperature is calculated. And the temporal change of the difference between the current estimated value and the previous estimated value of the estimated mold temperature as the antecedent part of the fuzzy inference rule, and the membership function and the inference rule of the fuzzy set set in advance. Based on the above, a fuzzy inference is performed to control a mold cooling air volume.

In addition to the above, the method of the present invention detects the flow rate of glass supplied to the glass forming machine per unit time and installs a section operating number detector in the glass forming machine to detect the operating number of sections. Then, the time variation of the difference between the current measurement value and the previous measurement value of each of the glass flow rate and the number of operating sections is set as an antecedent of the fuzzy inference rule, and the membership of the fuzzy set is set in advance. Fuzzy inference is performed based on the function and the inference rule, and the above glass flow rate and the number of sections operated at the previous production of the same product, the current glass flow rate, and the deviation from the section operation table are calculated using the antecedent part of the fuzzy inference rule. age,
Fuzzy inference is performed based on a preset fuzzy set membership function and inference rules, mold temperature is estimated, and the time difference between the current estimated value and the previous estimated value of the estimated mold temperature is estimated. The change is the antecedent of the fuzzy inference rule,
It is characterized in that fuzzy inference is performed based on a preset fuzzy set membership function and inference rules to control the amount of mold cooling air.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An overall configuration of a mold cooling facility for carrying out the method of the present invention will be described. Referring to FIG. 1, FIG. 2 and FIG. 3, the glass melted in the glass melting furnace
Is fed to the feeder 2 to form a gob 3. The formed gob 3 is delivered to a molding machine 4.
Is supplied to the mold provided in each section 5 of the above to form a glass product. During this process, about 1100
The gob at a temperature between 1200 ° C and 1200 ° C is
30 ° C.) (about 450 ° C. to 600 ° C.)
° C), the heat is removed, and the glass product becomes about 550 ° C to 650 ° C and is taken out of the molding machine.

The molding machine is generally composed of 1 to 12 sections, and each section is provided with a duct 8 for cooling a mold. Cooling air is taken in by outside air by a blower and has a high pressure (about 5000 to 12000P).
Send the wind of a). A damper 9 is provided in the outside air intake of the blower, and the opening degree of the damper 9 is adjusted to change the mold cooling air flow.

The basic structure of the method of the present invention will be described. As one of the goals is to stabilize the mold temperature,
As shown in (1), a thermometer 11 is installed in the duct 8, the wind temperature is measured, the signal is measured as a signal which is a fluctuation factor of the mold temperature, and the measured signal is input to the control method. The membership function of the fuzzy set, which is set in advance as the antecedent of the fuzzy inference rule, is calculated from the temporal change amount of the difference between the previous measurement value and the current measurement value of the measured data or the deviation amount between the current measurement value and the set value. Fuzzy inference is performed based on the inference rules to estimate how the mold temperature will change in the future. (Hereafter, the estimated value of the mold temperature is referred to as the estimated mold temperature.)

In the method of the present invention, when more detailed control is performed, as shown in FIGS.
In the duct, in addition to the thermometer 11, a pressure gauge 10 and a hygrometer 12 are installed, and in addition to the wind temperature, pressure, humidity, and the like are measured. A thermometer 14 is installed, and a molding machine timing device 15
, The number of operating sections is detected, and a signal that is a cause of mold temperature fluctuation is measured from these measuring instruments and used as an input to the control method. The molding machine timing device 15 catches the situation where the gob is supplied to the mold by the sensor,
The number of operating sections is detected. Further, a cooling time adjusting valve at the timing of the molding machine is provided at the cooling air inlet of the window nozzle to adjust the amount of cooling air.

According to the present invention, accurate results can be obtained without directly measuring the mold temperature. Generally, to measure the mold temperature, a method of directly attaching a thermometer such as a thermocouple or a sensor to the mold or a method of measuring the temperature of the inner surface of the mold when the mold is opened by contact. There is. In the direct measurement method, it is necessary to replace the thermocouple or the sensor each time the mold is replaced. Further, in the method of measuring by contact, it is necessary to change the set value of the measurement when the opening timing of the mold is changed. Therefore, in measuring the mold temperature, not only the installation of the measuring instrument but also the operation of the operator is increased.

Fuzzy inference is performed from the deviation between the estimated mold temperature and the set temperature or the temporal change in the estimated mold temperature.
The opening degree of a damper provided in the outside air intake port of the blower (hereinafter referred to as “damper opening degree”) is determined, and control is performed so that the mold temperature is stabilized.

Next, a control block of the fuzzy control will be described. FIG. 4 shows each control block constituting the fuzzy control. It is composed of a process data calculation unit, a fuzzy inference unit, and the like. The process data calculation unit
Input the temperature, source pressure, humidity, wind temperature, glass flow rate, gob temperature, blower damper opening from the process, calculate the deviation from each set temperature in the database section, and the current and previous measured values Calculate the amount of temporal change from the above. The fuzzy inference unit performs fuzzy inference from data input from the process data operation unit, a membership function corresponding to the data, and an inference rule. The operation amount calculation unit calculates a qualitative operation amount obtained by fuzzy inference into a quantitative value. If there are multiple inference rules, the weighted average value is calculated using the membership value, and the
U). The operation amount evaluation unit adds the obtained operation amount change amount (ΔU) and the current operation amount, and evaluates whether or not the value is within a predetermined range. If the value is correct, it is sent to the control unit and output to the process.

In carrying out the present control method, a system is configured with the equipment configuration shown in FIG. 5, and input / output signals from the process are transmitted to a personal computer via a programming logic controller (hereinafter referred to as "PLC"). (PC).

The main functions of the PC are: 1. Input / output data management and filtering function 2. Fuzzy inference function 3. Type change setting function and type change processing function It is roughly divided into man-machine interface functions.

As shown in FIG. 6, the fuzzy inference unit performs fuzzy inference in the stabilization control mode during normal production based on data and fuzzy inference in the start-up settling control mode in type change. Will be This 2
Switching between the two modes is performed automatically according to the operation status of the process.

Control target values, parameters, measured values (gob temperature, temperature, humidity, source pressure set value, glass flow rate, number of sections operated, etc.) used for fuzzy control, alarm set values (upper limit value, lower limit value), etc. Since data differs for each product type, the data is automatically stored as a database during good production (hereinafter, data stored in the database is referred to as product type data). In the type change setting, it is possible to set the type change time of the type to be produced next, and to call out the type data of the type. The set value, the original pressure set value, and each control target value are set as initial values used for controlling the product type after the type change.

The PC has the following functions in addition to the above functions. 1. Operation monitoring operation using graphic screen and overview screen 2. Alarming function 3. 3. DDC (Direct Digital Control) function Trend screen display 5. Data logging function

Hereinafter, the steady control during normal production and the start-up settling control during type change will be described in detail in separate sections. (1) Steady-state control during normal production The steady-state control during normal production includes the following control loop. (A) Estimation of mold temperature (B) Stabilization control of mold temperature (C) Control corresponding to change in setting of control target value (D) Control corresponding to fluctuation of the number of operating sections

(A) Estimation of mold temperature This is a control loop for estimating mold temperature fluctuations based on past control operations and changes in the production environment in order to stabilize mold temperatures.

In the basic control loop of the present control, the difference between the presently measured wind temperature and the previously measured wind temperature is calculated, and the temporal change is obtained. Then, using the variable of the temporal change of the wind temperature as a precondition of the fuzzy inference rule, a preset fuzzy inference is performed to estimate the mold temperature.

In the present control loop, when finer control is to be performed, in addition to the above-mentioned air temperature, the difference between the humidity measured this time and the humidity measured last time, the number of operating sections measured this time and the number measured last time are used. Section operating number difference,
The difference between the original pressure measured this time and the previously measured original pressure is calculated, and the temporal change amount is obtained. The fuzzy inference is performed by using the three variables of the time change of the wind temperature, the humidity, and the number of operating sections as well as the two variables of the temporal change of the source pressure and the number of operating sections as preconditions of the fuzzy inference rule. Estimate mold temperature.

The estimated mold temperature (hereinafter referred to as "mold estimated temperature") is used as an index of mold temperature indicating (B) mold stability control.

(1) Input Data and Membership Function Input data used for fuzzy inference and its membership function are shown below.

A) Temporal change in wind temperature (ΔE) (current measured wind temperature) − (previous measured wind temperature) (see FIG. 7)

B) Temporal change in humidity (ΔE) (current measured humidity) − (previous measured humidity) (see FIG. 8)

C) Time-dependent change of the source pressure (ΔE) (current source pressure) − (previous source pressure) (see FIG. 9)

D) Number of operating sections (see FIG. 10) PB, Z and NB in the figure are names given to the membership functions. PB: positively large Z: unchanged NB: negatively large

(2) Inference Rule (Control Rule) The three variables of the temporal change of the wind temperature, the temporal change of the temperature and the number of operating sections, and the two variables of the temporal change of the set source pressure value and the operating number of sections. The inference rules for determining the estimated mold temperature from variables are shown. (See Fig. 11)

A) When the number of operating sections is PB (Table 1)

[Table 1]

In the inference rule in the above table, in the upper left case, if the number of operating sections is large, the temporal change in the wind temperature is negative and the temporal change in the humidity is positive, then the mold estimated temperature is , Indicating that it is just normal.

B) When the number of operating sections is NB (Table 2)

[Table 2]

C) Time-dependent change in source pressure (ΔE) (Table 3)

[Table 3]

[0051]

[Table 4]

When the estimated mold temperature is defined by a plurality of inference rules, the change amount (ΔU) of the estimated mold temperature is obtained by a weighted average value corresponding to each membership value. The obtained change amount (ΔU) of the estimated mold temperature and the current estimated mold temperature are added to obtain a new estimated mold temperature.

(B) Stabilization control of mold temperature (A) This is a control loop for stabilizing the mold estimated temperature determined in the control of mold temperature estimation to the set temperature. The damper opening is determined by performing fuzzy inference from the temporal change amount of the difference between the current value and the previous value of the mold estimated temperature or the difference between the mold estimated temperature and the set temperature. The determined damper opening is sent to the motor of the damper at the outside air intake of the blower.

That is, as shown in FIG. 12, the data such as the measured value and the original pressure set value from each measuring instrument are used as input data to the fuzzy inference section for estimating the mold temperature (A), and the fuzzy inference result is obtained. The fuzzy inference result is input to the fuzzy inference unit, and the fuzzy inference result is sent to a motor of a damper at an outside air intake port of the blower to control a mold temperature cooling air flow.

(1) Input Data and Membership Function Input data used for fuzzy inference and its membership function are shown below.

A) Temporal change amount (ΔE) of estimated mold temperature (see FIG. 13)

B) Deviation (E) between the estimated mold temperature and the set temperature (see FIG. 14)

(2) Inference Rule (Control Rule) (Table 5) (Table 6) The amount of change in the damper opening degree based on the temporal change amount of the estimated mold temperature or the deviation amount between the estimated mold temperature and its set value ( ΔU).

[0059]

[Table 5]

[0060]

[Table 6]

(C) Control for setting change of control target value When a set value (target) of a mold temperature is changed by an operator or the like, fuzzy inference is performed based on two variables of the change amount of the set temperature and the number of sections operated. To determine the change amount (ΔU) of the damper opening.

(1) Input Data and Membership Function Input data and its membership function used for fuzzy inference are shown below.

A) Amount of change in set value of mold temperature (set value of mold temperature after change) − (set value of mold temperature before change) (see FIG. 15)

B) Number of operating sections The number of operating sections (see FIG. 16)

(2) Inference Rules (Control Rules) (Table 7) Fuzzy inference is performed by using two variables of the set value of the mold temperature and the number of sections to change the damper opening (ΔU).
To determine.

[0066]

[Table 7]

(D) Variation control of the number of operating sections If there is a variation in the number of operating sections, fuzzy inference is performed from the variation in the number of operating sections to determine the amount of change (ΔU) in the damper opening. (1) Input data and membership functions Input data and its membership functions used for fuzzy inference are shown below.

(A) Amount of change in the number of operating sections (number of operating sections before change)-(number of operating sections after change) (see FIG. 15)

(2) Inference Rules (Control Rules) (Table 8) Fuzzy inference is performed based on the change amount of the number of operating sections to determine the change amount (ΔU) of the damper opening.

[Table 8]

In the present invention, the glass melted in the glass melting furnace is fed from a fore hearth to a gob forming section via a feeder to form a gob, which is then fed to a mold of a glass forming machine. In the mold temperature control method for a glass forming machine in which the amount of cooling air of the mold is controlled to control the temperature of the mold, (a) a suction damper and a blower are installed in a ventilation duct, and cooling is performed. Supply the wind,
(B) A thermometer is installed in the air duct between the blower and the glass forming machine to measure the temperature (hereinafter referred to as “wind temperature”),
(C) The temporal change in the difference between the current measured value and the previous measured value of the wind temperature is defined as the antecedent of the fuzzy inference rule, and is based on the membership function of the fuzzy set and the inference rule set in advance. Then, fuzzy inference is performed, the mold temperature is estimated, and the mold cooling air volume is controlled.

Further, in addition to the above, a pressure gauge is installed in a blow duct between the blower and the glass forming machine, and a pressure (hereinafter, referred to as a pressure gauge) is set.
In addition to measuring the “source pressure”, a hygrometer was installed in the air duct between the blower and the glass forming machine to measure the humidity (hereinafter referred to as “humidity”), and the measured values of the source pressure and humidity were measured. The amount of temporal change in the difference between the current measurement value and the previous measurement value is used as the antecedent of the fuzzy inference rule. Is estimated, and the mold cooling air volume is controlled.

In addition to the above, a thermometer is installed in the gob forming section to measure the temperature of the glass supplied to the mold (hereinafter referred to as “gob temperature”) and to measure the temperature near the glass forming machine. Install a meter and measure the temperature (hereinafter referred to as “temperature”)
Is measured, and the temporal change amount of the difference between the current measurement value and the previous measurement value of each of the gob temperature and the temperature is set as a prerequisite part of the fuzzy inference rule, and a membership function of a preset fuzzy set is used. Fuzzy inference is performed based on the inference rules to estimate the mold temperature and control the mold cooling air flow.

In addition to the above, in addition to detecting the flow rate of glass per unit time supplied to the glass forming machine, a section operating number detector is installed in the glass forming machine to detect the number of operating sections. The time variation of the difference between the current measurement value and the previous measurement value of each of the glass flow rate and the number of operating sections is assumed to be the antecedent part of the fuzzy inference rule, and the membership function and the inference rule of the fuzzy set are set in advance. Fuzzy inference based on
The mold temperature is estimated, and the amount of mold cooling air is controlled.

Further, in the method of the present invention, the glass melted in the glass melting furnace is fed to the gob forming section from the forehath via the feeder to form the gob, and the gob is formed in the mold of the glass forming machine. In a mold temperature control method for a glass forming machine, wherein a cooling air flow rate of a mold is controlled to control a mold temperature, (a) a suction damper and a blower are installed in a blow duct. (B) installing a thermometer in a ventilation duct between the blower and the glass forming machine to measure a temperature (hereinafter referred to as "wind temperature"); and (c) measuring the above-mentioned wind temperature. The amount of temporal change in the difference between the current measured value and the previous measured value is the antecedent of the fuzzy inference rule, and fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance. Estimate the mold temperature and (d) The time variation of the difference between the current estimated value and the previous estimated value of the mold temperature is used as the antecedent of the fuzzy inference rule, and the fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance. Do
(E) The amount of deviation between the estimated mold temperature and the set temperature is used as the antecedent of the fuzzy inference rule, and fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance. f) The deviation between the changed set temperature and the set temperature is used as the antecedent of the fuzzy inference rule, fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance, and the mold cooling air volume is calculated. Perform control.

In addition, in addition to the above, a pressure gauge is installed in a blower duct between the blower and the glass forming machine, and a pressure (hereinafter, referred to as a pressure gauge) is set.
In addition to measuring the “source pressure”, a hygrometer was installed in the air duct between the blower and the glass forming machine to measure the humidity (hereinafter referred to as “humidity”), and the measured values of the source pressure and humidity were measured. The amount of temporal change in the difference between the current measurement value and the previous measurement value is used as the antecedent of the fuzzy inference rule. Is estimated, and the mold cooling air volume is controlled.

In addition to the above, a thermometer is installed at the gob forming section to measure the temperature of the glass supplied to the mold (hereinafter referred to as “gob temperature”) and to measure the temperature near the glass forming machine. Install a meter and measure the temperature (hereinafter referred to as “temperature”)
Is measured, and the temporal change amount of the difference between the current measurement value and the previous measurement value of each of the gob temperature and the temperature is set as a prerequisite part of the fuzzy inference rule, and a membership function of a preset fuzzy set is used. And fuzzy inference based on the inference rule to estimate the mold temperature and control the mold cooling air flow.

In addition to the above, in addition to detecting the flow rate of glass per unit time supplied to the glass forming machine, a section operating number detector is installed in the glass forming machine to detect the number of operating sections. The amount of temporal change in the difference between the current measurement value and the previous measurement value of each of the glass flow rate and section operation number is used as the antecedent of the fuzzy inference rule, and the membership function and inference of the fuzzy set are set in advance. Based on the rules, fuzzy inference is performed, the mold temperature is estimated, and the mold cooling air volume is controlled.

(2) Start-up setting control in type change The start-up setting control in type change is constituted by the following control loop. (A) Estimation of mold temperature due to mold change (B) Setting of optimal mold temperature control air volume due to mold change

(A) Estimation of mold temperature due to mold change At the time of mold change, gob temperature, glass flow rate, number of operating sections, temperature, humidity and gob temperature at the time of mold change Fuzzy inference from the measured data of the flow rate of the glass, the number of operating sections, the temperature and the humidity, and the production environment of the production type after the mold change is performed with the damper opening of the production type data. Estimate the change in mold temperature.

(1) Input Data and Membership Input data and its membership function used for fuzzy inference are shown below.

A) Difference between the temperature of production type data and the temperature at the time of type change (temperature at the time of type change)-(temperature of production type data) (see FIG. 18)

B) Difference between humidity of production type data and humidity at the time of type change (humidity at the time of type change)-(humidity of production type data) (see FIG. 19)

C) Difference between gob temperature of production type data and gob temperature at the time of type change (gob temperature at type change) − (gob temperature of production type data) (see FIG. 20) d) Glass flow rate of production type data And the difference between the glass flow rate at the time of mold change (glass flow rate at the time of mold change)-(glass flow rate of production type data) (see Fig. 21)

E) Difference between the number of operating sections in the product type data and the number of operating sections during model change (number of operating sections during type change)-(number of operating sections in product type data) (see FIG. 22)

(2) Inference rules (control rules) (Table 9)
(Table 10) (Table 11) (Table 12) Temperature, humidity, gob temperature, glass flow rate, number of sections operated and temperature, humidity, gob temperature at the time of model change Then, fuzzy inference is performed from the difference with the number of operating sections, and when the production type after the mold change is produced by the damper opening of the production type data, the change in the mold temperature is estimated from the difference in the production environment.

[0086]

[Table 9]

[0087]

[Table 10]

[0088]

[Table 11]

[0089]

[Table 12]

(B) Setting of the optimal amount of mold temperature control air accompanying mold change (A) From the deviation between the mold estimated temperature obtained by estimating mold temperature accompanying mold change and the set temperature Fuzzy inference is performed to determine a damper opening for obtaining an optimal amount of mold temperature control air so that the mold temperature can be quickly stabilized at the set temperature at the time of mold change.

(1) Input Data and Membership Function Input data and its membership function used for fuzzy inference are shown below.

A) Deviation of mold temperature (E) (Estimated mold temperature)-(Set temperature of mold temperature) (see FIG. 23)

(2) Inference Rule (Control Rule) (Table 13) (A) Fuzzy inference is performed from the difference between the estimated mold temperature obtained by estimating the mold temperature due to the mold change and the set temperature, and the damper is opened. The change amount (ΔU) of the degree is determined.

[0094]

[Table 13]

In the method of the present invention, the glass melted in the glass melting furnace is fed from a forge through a feeder to a gob forming section, where a gob is formed and the gob is sent to a mold of a glass forming machine. In a mold temperature control method for a glass forming machine, wherein a cooling air flow rate of a mold is controlled so as to control a mold temperature, (a) a suction damper and a blower are installed in a blower duct; Supply cooling air,
(B) A pressure gauge is installed in a duct between the blower and the glass forming machine to measure a pressure (hereinafter, referred to as “source pressure”), and (c)
Install a hygrometer in the duct between the blower and the glass forming machine,
Measure humidity (hereinafter referred to as "humidity"), (d) Install a thermometer near the glass forming machine, measure the temperature (hereinafter referred to as "temperature"), and (e) Last production of the same product The above-mentioned source pressure, humidity, the deviation amount between the current source pressure, humidity, and temperature and the current source pressure, humidity, and temperature are defined as the antecedent of the fuzzy inference rule, and based on a membership function and an inference rule of a fuzzy set that is set in advance. A fuzzy inference is performed to estimate a mold temperature, and a temporal change amount of a difference between the current estimated value and the previous estimated value of the estimated mold temperature is set as a precondition of the fuzzy inference rule, and a preset fuzzy inference rule is set. Fuzzy inference is performed based on the membership function of the set and the inference rules to control the die cooling air volume.

In addition to the above, a thermometer is installed in a duct between the blower and the glass forming machine to measure a temperature (hereinafter referred to as “wind temperature”), and to measure the above-mentioned wind temperature during the previous production of the same product. And the amount of deviation from the current wind temperature as the antecedent of the fuzzy inference rule, perform fuzzy inference based on the membership function of the fuzzy set and the inference rule set in advance, estimate the mold temperature, The temporal change of the difference between the current estimated value and the previous estimated value of the estimated mold temperature is defined as the antecedent part of the fuzzy inference rule, and the fuzzy set membership function and the inference rule are used for the fuzzy inference rule. Inference is performed to control the mold cooling air volume.

In addition to the above, a thermometer is installed in the gob forming section to measure the temperature of the glass supplied to the mold (hereinafter referred to as “gob temperature”), and to measure the temperature of the same product during the previous production. The difference between the Gob temperature and the current Gob temperature is used as the antecedent of the fuzzy inference rule, and fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance.
The mold temperature is estimated, and the temporal change amount of the difference between the current estimated value and the previous estimated value of the estimated mold temperature is set as the antecedent of the fuzzy inference rule, and the membership function of the fuzzy set is set in advance. Fuzzy inference based on the inference rules
Controls the mold cooling air volume.

In addition to the above, in addition to detecting the flow rate of glass supplied per unit time to the glass forming machine, a section operating number detector is installed in the glass forming machine to detect the section operating number. The amount of temporal change in the difference between the current measurement value and the previous measurement value of each of the glass flow rate and section operation number is used as the antecedent of the fuzzy inference rule, and the membership function and inference of the fuzzy set are set in advance. Based on the rules, fuzzy inference is performed, and the above glass flow rate and the number of sections operated at the previous production of the same product, the current glass flow rate, and the deviation from the section operation table are taken as the antecedent of the fuzzy inference rules. Fuzzy inference is performed based on the set fuzzy set membership function and inference rules to estimate the mold temperature,
The temporal change of the difference between the current estimated value and the previous estimated value of the estimated mold temperature is defined as the antecedent part of the fuzzy inference rule, and the fuzzy set membership function and the inference rule are used for the fuzzy inference rule. Inference is performed to control the mold cooling air volume.

In the above description, the method of controlling the amount of the mold temperature control wind by the damper opening has been described. Alternatively, the rotation speed of the blower motor may be changed, or the molding machine timing device may be controlled. Can also be done.

[0100]

FIG. 24 shows the mold temperature and the mold temperature estimation result when only the mold temperature inference of the present invention is used during the conventional control during the production of the product A (900 ml, sake bottle: 390 g). From the measurement results, it can be seen that the behaviors of the mold temperature and the mold estimated value are the same.

FIG. 25 shows the relationship between pressure and mold temperature in the system control data.

FIG. 26 shows product A (900 ml sake bottle: 3
90g) shows the air temperature, pressure, absolute humidity, and mold temperature when the mold cooling air volume is controlled constant (pressure value) by conventional control during production.

FIG. 27 shows the measurement results of the wind temperature, pressure and mold temperature when the method of the present invention (control using the time-dependent change of the wind temperature as a variable) is employed during production of the same product. In the control method of the present invention, the fluctuation of the mold temperature is controlled by ± 4.8 ° C. by the mold stabilization control even when the same disturbance factor is received. Thus, it can be seen that the controllability of the steady control during normal production by the fuzzy control of the present invention is excellent.

FIG. 28 shows a case where the method of the present invention (control using other factors as variables in addition to wind temperature for fine control) during production of the same product is employed (die temperature stabilization control).
2 shows the measurement results of the same measurement point. Looking at the measurement results,
Disturbance factor in conventional control (change in disturbance of wind temperature and absolute humidity)
However, in the control method of the present invention, the mold temperature fluctuation is controlled by ± 2.5 ° C. by the mold temperature stabilization control even when a similar disturbance factor is received. doing.

FIG. 29 shows the results of measurement of the mold temperature and pressure under the conventional control when the product B (one shovel bottle: 950 g) was changed to the product A (900 ml sake bottle: 390 g).

FIG. 30 shows the results of measurement of the mold temperature and pressure when the method of the present invention is employed when the same mold change is performed. Looking at the measurement results, in the conventional control, it took 55 minutes for the mold temperature to stabilize, but when the method of the present invention was adopted, 30 minutes was achieved for the mold temperature to stabilize.
Thus, it can be seen that the controllability of the start-up setting control in the type change by the fuzzy control of the present invention is excellent.

FIG. 31 shows product A (900 ml sake bottle: 3
90g) shows the result of detecting the number of operating sections (FIG. 31) and the control result of not detecting the number of operating sections (top of FIG. 31) when the method of the present invention is employed during production. Looking at the measurement results, in the control that does not detect the number of operating sections, the temperature of the mold is 2.5 ° C due to the stop of the section.
It takes 10 minutes to become unstable and stable, but if a control method for detecting the number of operating sections is adopted, the mold temperature remains stable. As described above, it is understood that detecting and controlling the number of operating sections has a remarkable effect on control.

FIG. 32 shows product A (900 ml sake bottle: 3
90g) is a graph showing how much improvement in productivity has occurred due to the difference in controllability between the conventional control and the present invention. Conventionally, the production efficiency was poor due to the unstable mold temperature. However, the present invention has stabilized the mold temperature and improved the production efficiency by 4%. Also, the mold cooling adjustment work (changing the cooling time of each section, adjusting the cooling air volume of each section) in order to prevent defects (cold rolling) caused by a change in the mold temperature by the control of the present invention as compared with the conventional method. It decreased by 0.26 (times / day).

As shown in FIG. 33, regarding the period comparison,
The control results of the conventional control and the control according to the present invention over three months, and the results of investigations on productivity have also obtained good results. . The needs for glass products are lighter, higher quality,
Amidst diversification, the tendency of glass product production to be high-mix low-volume production (increase in the number of mold replacements) is increasing, leading to increased costs.

As described above, according to the present invention, the remodeling time is shortened, and the mold temperature fluctuation value is reduced. Stabilizing the mold temperature, reducing the mold temperature cooling adjustment work, and stabilizing the production process, such as reducing the occurrence of defective products due to the mold temperature stabilization.
Significantly contribute to the improvement of the yield.

[Brief description of the drawings]

FIG. 1 is a schematic structural explanatory view of a mold cooling device for carrying out a method of the present invention.

FIG. 2 is a schematic structural explanatory view of another mold cooling device for carrying out the method of the present invention.

FIG. 3 is a perspective view of a mold of a mold cooling device for carrying out the method of the present invention.

FIG. 4 is a control block explanatory diagram of fuzzy control for implementing the method of the present invention.

FIG. 5 is an explanatory diagram of a control device configuration.

FIG. 6 is an explanatory diagram of a control mode switching method.

FIG. 7 is a membership function.

FIG. 8 is a membership function.

FIG. 9 is a membership function.

FIG. 10 is a membership function.

FIG. 11 is an explanatory diagram of inference regulation (control rule).

FIG. 12 is an explanatory diagram of stabilization control of mold temperature.

FIG. 13 is a membership function.

FIG. 14 is a membership function.

FIG. 15 is a membership function.

FIG. 16 is a membership function.

FIG. 17 is a membership function.

FIG. 18 shows a membership function.

FIG. 19 is a membership function.

FIG. 20 is a membership function.

FIG. 21 is a membership function.

FIG. 22 is a membership function.

FIG. 23 is a membership function.

FIG. 24 is an explanatory diagram of mold temperature estimation data.

FIG. 25 is an explanatory diagram of system control data (mold temperature).

FIG. 26 is an explanatory diagram of steady control data (conventional control) during normal production.

FIG. 27 is an explanatory diagram of steady control data (system control) (wind temperature) during normal production.

FIG. 28 is an explanatory diagram of steady control data (system control) during normal production.

FIG. 29 is an explanatory diagram of start-up settling control data (conventional control) in type change.

FIG. 30 is an explanatory diagram of start-up settling control data (system control) in a type change.

FIG. 31 is an explanatory diagram of control data in which the number of operating sections is detected / not detected.

FIG. 32: Comparison control result before / after system operation (product A)
It is.

FIG. 33 is a comparison control result (period comparison) before / after system operation.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 For hearth 2 Feeder 3 Gob 4 Molding machine 5 Section 8 Duct 9 Damper 10 Pressure gauge 11 Wind temperature gauge 12 Hygrometer 13 Feeder thermometer 14 Thermometer 15 Molding machine timing device

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Takayuki Kasamatsu 7-55 Minamienkajima, Taisho-ku, Osaka City, Osaka Prefecture In-house Engineering Co., Ltd. Address (72) Inventor Masahiro Konishi 685-1 Okami Young, Ichinomiya City, Aichi Prefecture (56) References JP-A-62-119127 (JP, A) JP-A-7-17721 (JP, A) JP-A-3 -228833 (JP, A) JP-A-63-255703 (JP, A) JP-A-63-190727 (JP, A) JP-B-52-28808 (JP, B1) JP-B-2-1776 (JP, B2) )

Claims (12)

(57) [Claims] 1. A glass melted in a glass melting furnace is fed from a fore hearth to a gob forming section via a feeder to form a gob, and the gob is fed to a mold of a glass forming machine. In a mold temperature control method for a glass forming machine in which the amount of cooling air of a mold is controlled to control a mold temperature, (a) a suction damper and a blower are installed in an air duct to supply cooling air; (B) A thermometer is installed in a ventilation duct between the blower and the glass forming machine to measure the temperature (hereinafter referred to as “wind temperature”). Fuzzy inference is performed based on the fuzzy inference rule based on the membership function of the fuzzy set and the inference rule, and the mold temperature is estimated. It is characterized by controlling the mold cooling air volume. Temperature control method for a glass forming machine. 2. A pressure gauge is installed in a blower duct between the blower and the glass forming machine to measure a pressure (hereinafter, referred to as “source pressure”), and a hygrometer is installed in a blower duct between the blower and the glass forming machine. Install and measure the humidity (hereinafter referred to as “humidity”), and use the time change of the difference between the current measured value of the original pressure and the humidity and the previous measured value as the antecedent of the fuzzy inference rule. 2. The method according to claim 1, wherein a fuzzy inference is performed based on a membership function of a fuzzy set and an inference rule set in advance, a mold temperature is estimated, and a mold cooling air volume is controlled. Mold temperature control method for glass forming machine. 3. A thermometer is installed in the gob forming section to measure the temperature of the glass supplied to the mold (hereinafter referred to as “gob temperature”), and to install a thermometer near the glass forming machine. , Temperature (hereinafter, referred to as “temperature”), and the temporal change of the difference between the current measurement value and the previous measurement value of the Gob temperature and the temperature is defined as the antecedent of the fuzzy inference rule, 3. The glass according to claim 2, wherein a fuzzy inference is performed based on a membership function of a fuzzy set and an inference rule set in advance, a mold temperature is estimated, and a mold cooling air volume is controlled. Mold temperature control method for molding machine. 4. In addition to detecting the flow rate of glass supplied to the glass forming machine per unit time, installing a detector for the number of operating sections in the glass forming machine to detect the number of operating sections, and detecting the flow rate of the glass and operating the section. The temporal change of the difference between the current measurement value and the previous measurement value of each measurement value of the number is set as the antecedent of the fuzzy inference rule, and based on the membership function of the fuzzy set and the inference rule set in advance. Make inferences, estimate mold temperatures,
4. The method of controlling a temperature of a mold of a glass forming machine according to claim 2, wherein the amount of cooling air of the mold is controlled. 5. The glass melted in a glass melting furnace is fed from a foreheart via a feeder to a gob forming section to form a gob, and the gob is fed to a mold of a glass forming machine. In a mold temperature control method for a glass forming machine in which a mold cooling air volume is controlled to control a mold temperature, (a) a suction damper and a blower are installed in an air duct to supply cooling air; (B) A thermometer is installed in a ventilation duct between the blower and the glass forming machine to measure the temperature (hereinafter, referred to as “wind temperature”). (C) The measurement value of the measurement value of the above-mentioned wind temperature and the previous measurement value The amount of temporal change in the difference from the value is the antecedent of the fuzzy inference rule,
Fuzzy inference is performed based on a preset fuzzy set membership function and inference rules to estimate the mold temperature, and (d) the difference between the current estimated value and the previous estimated value of the estimated mold temperature. Is used as the antecedent of the fuzzy inference rule, and fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance. (E) Estimated mold temperature and set temperature Is used as the antecedent of the fuzzy inference rule, fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance, and (f) deviation value between the changed set temperature and the set temperature Is the antecedent of the fuzzy inference rules, and based on the membership function of the fuzzy set and the inference rules set in advance, perform fuzzy inference, estimate the mold temperature, and control the mold cooling air volume. Mold temperature control method for a glass forming machine, wherein. 6. A pressure gauge is installed in a blower duct between the blower and the glass forming machine to measure a pressure (hereinafter, referred to as “source pressure”), and a hygrometer is installed in the blower duct between the blower and the glass forming machine. Install and measure the humidity (hereinafter referred to as “humidity”), and use the time change of the difference between the current measured value of the original pressure and the humidity and the previous measured value as the antecedent of the fuzzy inference rule. The fuzzy inference is performed based on a membership function of a fuzzy set and an inference rule set in advance, a mold temperature is estimated, and a mold cooling air flow is controlled. Mold temperature control method for glass forming machine. 7. A thermometer is installed in the gob forming section to measure the temperature of the glass supplied to the mold (hereinafter referred to as “gob temperature”), and to install a thermometer near the glass forming machine. , Temperature (hereinafter, referred to as “temperature”), and the temporal change of the difference between the current measurement value and the previous measurement value of the Gob temperature and the temperature is defined as the antecedent of the fuzzy inference rule, 7. The glass according to claim 6, wherein fuzzy inference is performed based on a preset fuzzy set membership function and inference rules, a mold temperature is estimated, and a mold cooling air volume is controlled. Mold temperature control method for molding machine. 8. In addition to detecting the flow rate of glass supplied to the glass forming machine per unit time, installing a detector for the number of operating sections in the glass forming machine to detect the number of operating sections. The temporal change of the difference between the current measurement value and the previous measurement value of each measurement value of the number is set as the antecedent part of the fuzzy inference rule, and based on the membership function of the fuzzy set and the inference rule set in advance, 7. The method according to claim 6, wherein a fuzzy inference is performed to estimate a mold temperature and control a mold cooling air volume.
Or a mold temperature control method for a glass forming machine according to 7. 9. The glass melted in the glass melting furnace is fed from a fore hearth to a gob forming unit via a feeder to form a gob, and the gob is fed to a mold of a glass forming machine. In a mold temperature control method for a glass forming machine in which a mold cooling air volume is controlled to control a mold temperature, (a) a suction damper and a blower are installed in an air duct to supply cooling air; (B) Install a pressure gauge in the duct between the blower and the glass forming machine to measure the pressure (hereinafter referred to as “source pressure”). (C) Install a hygrometer in the duct between the blower and the glass forming machine. To measure humidity (hereinafter referred to as “humidity”), (d) installing a thermometer near the glass forming machine to measure the temperature (hereinafter referred to as “temperature”), and (e) measuring the same product. The above original pressure, humidity and temperature at the previous production and the current original pressure and humidity , The amount of deviation from the temperature as the antecedent of the fuzzy inference rule, fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance, and the mold temperature is estimated and estimated. Fuzzy inference is performed based on the membership of the fuzzy set and the inference rules based on the membership of the fuzzy set and the fuzzy inference rule. And a method of controlling a mold cooling air flow rate. 10. A thermometer is installed in a duct between a blower and a glass forming machine to measure a temperature (hereinafter referred to as “wind temperature”), and to measure the above-mentioned wind temperature at the time of previous production of the same product and the current wind temperature. The fuzzy inference is performed based on the fuzzy inference rule, and the fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule, and the estimated mold temperature is estimated. The temporal change in the difference between the current estimated value and the previous estimated value of the temperature is used as the antecedent of the fuzzy inference rule, and fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance. The method of controlling a temperature of a mold of a glass forming machine according to claim 9, wherein a flow rate of a mold cooling air is controlled. 11. A thermometer is installed in a gob forming section to measure a temperature of a glass supplied to a mold (hereinafter, referred to as “gob temperature”), and to compare the gob temperature of the same product in the previous production with the current gob temperature. The deviation from the Gob temperature is used as the antecedent of the fuzzy inference rule, fuzzy inference is performed based on the membership function of the fuzzy set and the inference rule set in advance, and the mold temperature is estimated and estimated. Fuzzy inference is carried out based on the fuzzy inference rule based on the membership of the fuzzy set and the fuzzy inference rules. The method of controlling the temperature of a mold of a glass forming machine according to claim 9, wherein the amount of cooling air of the mold is controlled. 12. In addition to detecting the flow rate of glass supplied to the glass forming machine per unit time, installing a detector for the number of operating sections in the glass forming machine to detect the number of operating sections. The temporal change of the difference between the current measurement value and the previous measurement value of each measurement value of the number is set as the antecedent part of the fuzzy inference rule, and based on the membership function of the fuzzy set and the inference rule set in advance, Fuzzy inference is performed, and the amount of deviation between the above-mentioned glass flow rate and section operation number at the previous production of the same product and the current glass flow rate and section operation number is used as the antecedent of the fuzzy inference rule, and a preset fuzzy set. Fuzzy inference is performed based on the membership function and the inference rule of, and the mold temperature is estimated. Fuzzy inference based on the fuzzy inference rule and the fuzzy set membership function and inference rule, and control the die cooling air volume. The method according to claim 9, 10 or 11, wherein the method is performed.