An automatic feeding control method and system for glass furnaces

Abstract

This application relates to the field of automatic feeding control technology, and in particular to an automatic feeding control method and system for glass furnaces; the method includes the following steps: collecting data from the glass furnace within the current control cycle, obtaining a deviation term and a mean absolute temperature difference term, and using the difference between the deviation term and the mean absolute temperature difference term as the furnace heat index; calculating a dynamic time constant; using the dynamic time constant to perform first-order inertial filtering on the feeding speed to obtain the main control quantity; based on the main control quantity and the feedback correction quantity, obtaining the feeding speed command for each control cycle; the feedback correction quantity characterizes the liquid level deviation of the molten glass; outputting the feeding speed command to the feeding system for execution, thereby controlling the automatic feeding of the glass furnace; thus improving the effect of automatic feeding control of the glass furnace.

Description

Technical Field

[0001] This application relates to the field of automatic feeding control technology, and in particular to an automatic feeding control method and system for glass furnaces. Background Technology

[0002] In the continuous production of high-end medical glass such as ampoules, the automatic feeding system of the glass furnace needs to work closely with the downstream tube drawing process. When the output of the tube drawing machine changes due to production plan adjustments, the upstream feeding system must respond quickly to maintain a constant glass melt level within the glass furnace. To achieve this, the industry commonly employs a feedforward-feedback hybrid control strategy. Specifically, to avoid thermal shock and liquid surface disturbance caused by sudden and drastic changes in the feeding rate, the target feeding rate change calculated by the feedforward is usually smoothed. Specifically, a fixed ramp function is used to smoothly transition to the new value within a preset time.

[0003] However, the slope of the static smoothing strategy is a fixed value. If the slope is set very gently to ensure the stability of the glass furnace when the discharge rate is significantly adjusted, then when the discharge rate is only slightly adjusted, this overly gentle transition process will be sluggish, causing the feed rate to lag behind the discharge rate for a long time, resulting in unnecessary liquid level drops. Conversely, if the slope is set steeply to cope with minor adjustments in order to pursue a rapid response, then when faced with significant adjustments, this aggressive change in the feed rate will disrupt the thermal balance of the melting zone, causing violent fluctuations in the liquid level. Therefore, existing automatic feed control methods for glass furnaces generally suffer from the problem of difficulty in balancing response and stability. Since the feed rate adjustment relies on fixed parameters, it fails to comprehensively consider various factors such as the furnace thermal state, discharge changes, and liquid level feedback, leading to phenomena such as slow response or thermal shock under different disturbance intensities. At the same time, the heat and material flow inside the furnace have significant nonlinear coupling characteristics, and traditional linear control strategies are unable to reflect the dynamic characteristics of the system. Summary of the Invention

[0004] To address the problem of poor automatic feeding performance in glass furnaces, this application provides an automatic feeding control method and system for glass furnaces.

[0005] In the first aspect, this application provides an automatic feeding control method for glass furnaces, which adopts the following technical solution:

[0006] Collect data from the glass furnace during the current control cycle, including: real-time average temperature, real-time liquid level, feeding rate, and discharge rate;

[0007] The deviation between the real-time average temperature and the preset process standard temperature is taken as the deviation item.

[0008] Calculate the average absolute temperature difference term of the glass furnace over all previous control cycles, including the current control cycle, and use the difference between the deviation term and the average absolute temperature difference term as the furnace heat index.

[0009] The difference in output between the current control cycle and the previous adjacent control cycle, along with the furnace heat index, are used to obtain the dynamic time constant through a nonlinear mapping function.

[0010] The feeding rate is processed by first-order inertial filtering using the dynamic time constant to obtain the main control quantity; based on the main control quantity and the feedback correction quantity, the feeding rate command for each control cycle is obtained; the feedback correction quantity characterizes the liquid level deviation of the molten glass.

[0011] Input the feeding speed command into the feeding system to control the automatic feeding of the glass furnace.

[0012] The beneficial effects are as follows: by collecting furnace temperature, liquid level, feeding speed and discharge rate in real time, and constructing heat index and feedback correction, multi-parameter coupled control is realized, so that the feeding adjustment can respond to discharge disturbances and maintain thermal stability, which significantly improves the overall performance of automatic feeding.

[0013] Furthermore, the method for obtaining the deviation term is as follows:

[0014] For the current control cycle, calculate the difference between the real-time average temperature and the preset process standard temperature, and use the ratio of the difference to the preset process standard temperature as the deviation term.

[0015] The beneficial effects are as follows: defining the ratio of real-time average temperature to standard temperature as the deviation term enables the normalized assessment of furnace heat reserves, making temperature deviations under different operating conditions uniformly comparable, and facilitating the quantitative calculation of dynamic control parameters.

[0016] Furthermore, the method for obtaining the mean absolute temperature difference term is as follows:

[0017] Calculate the average real-time average temperature of all previous control cycles, including the current control cycle, and calculate the absolute value of the difference between the real-time average temperature of each control cycle and the average value. The average of the ratios of all absolute differences to the preset process standard temperature is taken as the average absolute temperature difference term.

[0018] The beneficial effects are: by reflecting the intensity of thermal fluctuations in the furnace through the average absolute temperature difference term of the historical control cycle temperature, the system can identify the stability of the thermal state, thereby achieving adaptive heat regulation and preventing control instability caused by drastic temperature fluctuations.

[0019] Furthermore, the method for obtaining the dynamic time constant is as follows:

[0020] Calculate the difference between the output of each control cycle and the output of the previous control cycle, calculate the absolute value of the ratio of the difference to the preset standard output, obtain the nonlinear mapping function value of the difference between the absolute value and the furnace heat index of the control cycle, and use the product of the sum of the value and the nonlinear mapping function value and the preset basic transition time as the dynamic time constant of each control cycle.

[0021] The beneficial effects are as follows: by obtaining the dynamic time constant through the nonlinear mapping between the difference in material output and the heat index of the furnace, the transition time of the feeding rate can be automatically changed with the intensity of disturbance and the level of heat reserve, thus solving the problem of response lag or overshoot of the traditional fixed time constant.

[0022] Furthermore, the nonlinear mapping function is a hyperbolic tangent function.

[0023] Furthermore, the method for obtaining the main control quantity is as follows:

[0024] For each control cycle, the ratio of the control cycle duration to the dynamic time constant is calculated, and the difference between the feeding rate and the feeding rate of the previous adjacent control cycle is calculated. The sum of the product of the ratio and the difference and the feeding rate of the previous adjacent control cycle is used as the main adjustment variable.

[0025] The beneficial effects are as follows: by performing first-order inertial filtering on the feeding rate through dynamic time constant, the main control quantity is obtained, making the feeding rate change smoother, reducing thermal shock and liquid surface disturbance, and improving the stable control performance of the system.

[0026] Furthermore, the method for obtaining the feeding speed command is as follows:

[0027] For each control cycle, the liquid level difference between the real-time liquid level and the preset target liquid level is calculated, the product of the preset liquid level feedback coefficient and the liquid level difference is calculated, and the difference between the main adjustment amount and the product is used as the feeding speed command for each control cycle.

[0028] Furthermore, inputting the feeding speed command into the feeding system includes: performing a safety constraint judgment on the feeding speed command, and sending the feeding speed command after the safety constraint judgment to the feeding system for execution.

[0029] Furthermore, the safety constraint determination includes: comparing with the preset maximum speed and preset minimum speed of the feeding system; if the feeding speed command exceeds the range formed by the preset maximum speed and preset minimum speed, then limiting its value to that range.

[0030] Secondly, this application provides an automatic feeding control system for glass furnaces, which adopts the following technical solution:

[0031] An automatic feeding control system for a glass furnace includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the automatic feeding control method for a glass furnace as described above.

[0032] The above-mentioned automatic feeding control method for glass furnaces is generated into a computer program and stored in a memory so that it can be loaded and executed by a processor. Thus, a system can be made based on the memory and processor for convenient use.

[0033] This application has the following technical effects:

[0034] The automatic feeding control method and system for glass furnaces proposed in this application achieves adaptive control of the feeding rate. By setting the difference between the deviation term and the average absolute temperature difference term to form the furnace heat index, the system can comprehensively reflect the current heat reserve level and thermal stability of the furnace, avoiding misjudgments caused by adjusting based solely on instantaneous temperature. The introduction of a nonlinear dynamic time constant calculation method based on the hyperbolic tangent function automatically extends the feeding transition time when the output rate changes drastically and automatically shortens the adjustment response when the disturbance is small, achieving a dynamic balance between thermal stability and response speed. A first-order inertial filtering mechanism further ensures the smoothness of feeding rate changes, avoiding thermal shock to the furnace caused by sudden changes. The introduction of liquid level feedback correction allows the system to sense the glass liquid level deviation in real time, promptly correct the feeding rate, and prevent the accumulation of errors due to liquid level fluctuations. The safety constraint module ensures that the final executed command is within the physical safety range, preventing equipment overload operation. Through a layered progressive mechanism, intelligent control of multi-dimensional parameters such as temperature, output rate, and liquid level is achieved. Compared with the traditional fixed slope smoothing method, this application significantly improves the response sensitivity and thermal stability of automatic feeding in glass furnaces, reduces the risk of liquid level oscillation and thermal stress, extends equipment life, and improves production efficiency and glass product quality. It has the beneficial effects of clear structure, achievable algorithm, and precise control. Attached Figure Description

[0035] Figure 1 This is a flowchart of an automatic feeding control method for a glass furnace according to this application. Detailed Implementation

[0036] This application discloses an automatic feeding control method for glass furnaces. The method involves collecting data from the glass furnace within the current control cycle, obtaining a deviation term and a mean absolute temperature difference term, and using the difference between the deviation term and the mean absolute temperature difference term as the furnace heat index. A dynamic time constant is calculated. The feeding speed is then subjected to a first-order inertial filter using the dynamic time constant to obtain the main control variable. Based on the main control variable and the feedback correction variable, a feeding speed command for each control cycle is obtained. The feedback correction variable characterizes the liquid level deviation of the molten glass. The feeding speed command is output to the feeding system for execution, thereby controlling the automatic feeding of the glass furnace. This improves the effectiveness of automatic feeding control for glass furnaces.

[0037] Reference Figure 1 An automatic feeding control method for a glass furnace includes steps S1-S4.

[0038] Step S1: Collect data of the glass furnace during the current control cycle, including: real-time average temperature, real-time liquid level, feeding rate and discharge rate.

[0039] Because the glass furnace charging process has strong inertia and large lag characteristics, it requires periodic control based on a stable time base to avoid computational redundancy caused by continuous control. It should be noted that the control method in this application is not a one-time calculation process, but rather operates continuously in a periodic manner. A complete charging process is divided into a series of continuous and equal-length time units, each time unit being called a control cycle. It should be noted that the following calculations are not performed during the first control cycle of a complete feeding process. In one embodiment of this application, the duration of the control cycle is 10 minutes; implementers may select other values ​​based on actual circumstances.

[0040] Furthermore, in each control cycle At the adjacent sampling time before the start, the following data is collected as data for each control cycle: real-time average temperature of the key melting zone of the glass furnace. The real-time average temperature is measured by an array of thermocouples arranged in the melting zone and averaged by the control system; the real-time liquid level of the molten glass in the glass furnace is also measured. The real-time liquid level is measured by a laser level gauge or a capacitive level sensor; the feeding rate is controlled periodically. The feeding speed is obtained from feedback from the motor encoder of the feeding mechanism or from output commands from the control system; each control cycle The amount of material discharged at the adjacent sampling time before the start The output is determined by the traction speed of the downstream pipe drawing machine and the measured pipe diameter, based on... The calculation yielded, where The density of the molten glass, The cross-sectional area of ​​the pipe is... This refers to the traction speed.

[0041] Step S2: The deviation between the real-time average temperature and the preset process standard temperature is taken as the deviation term; the average absolute temperature difference term of the glass furnace over all previous control cycles, including the current control cycle, is calculated, and the difference between the deviation term and the average absolute temperature difference term is taken as the furnace heat index.

[0042] It should be noted that since data from each control cycle alone cannot fully assess the thermal state of the glass furnace, both heat reserves and operational stability must be considered. Specifically, the deviation between the current temperature and the set value reflects the current heat reserve level of the glass furnace, while the average absolute temperature difference over all past control cycles reflects the relative severity of temperature fluctuations in the glass furnace.

[0043] Therefore, based on the above analysis, the deviation between the real-time average temperature of the key melting zone collected before the start of each control cycle and the process standard temperature is normalized, and combined with the average absolute temperature difference term of temperature fluctuations in all past control cycles, a comprehensive furnace heat index for each control cycle is constructed. The calculation formula is:

[0044]

[0045] In the formula, It is the furnace heat index for each control cycle; It is the real-time average temperature of each control cycle; It is the preset process standard temperature of the glass furnace; It is the position sequence of each control cycle, that is, each control cycle is the Nth control cycle; It is the first before each control cycle Real-time average temperature collected within each control cycle; It is the real-time average temperature of all previous control cycles, including each control cycle. The arithmetic mean; in one embodiment of this application, the preset process standard temperature is 1500 degrees Celsius, and the implementer may select other values ​​based on the actual situation.

[0046] It should be noted that, It is the normalized temperature deviation term, when When this item is positive, it indicates that the glass furnace currently has sufficient heat and strong resistance to disturbances; the furnace heat index... The higher; when When this item is negative, it indicates that the glass furnace is experiencing a heat deficit and insufficient heat reserves, resulting in a negative furnace heat index. The lower; This is the mean absolute temperature difference term, reflecting the intensity of temperature fluctuations in the critical melting zone of the glass furnace over all past control cycles. A larger value indicates a more unstable thermal state in the glass furnace, and a higher furnace heat index. The lower the value, the more stable the thermal state of the glass furnace; conversely, the smaller the value, the better the furnace heat index. The higher.

[0047] For example, for the furnace heat index, set... Assume the total number of control cycles that have been completed is... The temperatures collected in the past three cycles were respectively The real-time average temperature collected at the start of the current cycle. First, calculate the historical average temperature: The furnace heat index is then:

[0048] .

[0049] Step S3: Obtain the dynamic time constant by using a nonlinear mapping function to analyze the difference in material output between the current control cycle and the previous adjacent control cycle, as well as the furnace heat index.

[0050] Based on the above steps, the furnace heat index for each control cycle was obtained. However, while the heat state index can be used to assess the state of the glass furnace itself, it cannot detect the intensity of disturbances caused by changes in the external discharge rate. Specifically, when the discharge rate changes more significantly and the furnace heat index is lower, it indicates a stronger disturbance and weaker resistance to disturbances in the system. In this case, the transition time of the feeding rate must be further extended. Conversely, when the discharge rate changes less significantly and the heat state index is higher, it indicates a minor disturbance and stable furnace condition, with good thermal buffering capacity. In this case, the transition time of the feeding rate can be appropriately shortened.

[0051] Based on the above analysis, the difference between the discharge rate of each control cycle and the discharge rate at the moment before the start of the previous control cycle is normalized and used as a positive driving term, and the furnace heat index is... As a negative adjustment term, the dynamic time constant is calculated through the nonlinear mapping of the hyperbolic tangent function. The calculation expression is as follows:

[0052]

[0053] In the formula, It is the dynamic time constant of the feeding rate transition in each control cycle; This is the preset basic transition time, representing the default transition duration under normal circumstances; This refers to the output volume in each control cycle; It is the output of the control cycle preceding each adjacent control cycle; It is the preset standard output of the glass furnace, used to normalize the output difference; It is the furnace heat index for each control cycle; It is a hyperbolic tangent function. In one embodiment of this application, the preset basic transition time and the preset standard output amount are respectively 300s, Implementers can choose other values ​​based on the actual situation.

[0054] It should be noted that, This is the normalized magnitude of the difference in output flow rate, serving as a positive driving term. A larger value indicates a more severe external disturbance in the output flow, and a higher risk of system impact; therefore, the dynamic time constant... The dynamic time constant should be increased to prolong the transition time of the feeding rate and suppress thermal disturbance; conversely, the smaller this value, the smoother the change in the output rate and the less the system disturbance. It can reduce and shorten the transition time of feeding speed, improve response speed, and avoid continuous deviation of liquid level caused by feeding lag.

[0055] It should also be noted that, This refers to the furnace heat index. A higher furnace heat index indicates sufficient heat and thermal stability in the glass furnace, strong resistance to disturbances, and the ability to withstand rapid changes in feed volume. Therefore, the dynamic time constant... Decrease; conversely, when the value of the furnace heat index is smaller or negative, it indicates that the heat of the glass furnace is insufficient or fluctuates drastically, requiring a reduction in the rate of change of the feeding speed and the dynamic time constant. Increase.

[0056] For example, setting a basic transition time Standard output Assuming the current output volume Previous cycle output Using the calculation from the previous step Then the dynamic time constant is: .

[0057] Step S4: The feeding rate is subjected to first-order inertial filtering using the dynamic time constant to obtain the main control quantity; based on the main control quantity and the feedback correction quantity, the feeding rate command for each control cycle is obtained; the feedback correction quantity represents the liquid level deviation of the glass melt; the feeding rate command is input into the feeding system to control the automatic feeding of the glass furnace.

[0058] Since the dynamic time constant alone cannot correct for the actual fluctuations in the molten glass level during the transition process in real time, it may cause cumulative errors. Specifically, the feeding rate needs to smoothly track the target value to avoid thermal shock; at the same time, the deviation between the real-time molten glass level and the target molten glass level can reflect material imbalance.

[0059] Based on the above analysis, a first-order inertial filter is applied to the target feeding speed using a dynamic time constant to form a smooth transition main control term. Simultaneously, the real-time glass melt level deviation is introduced as a negative feedback correction term to suppress glass melt level fluctuations. The feeding speed command for each control cycle is obtained by superimposing both parameters. It should be noted that the unit of the feeding speed command is kg / min, which represents the feeding rate rather than the velocity in a kinematic sense. The calculation formula is as follows:

[0060]

[0061] In the formula, These are the feeding speed commands for each control cycle; It is the feeding rate of the control cycle preceding the previous control cycle; It refers to the duration of the control cycle; It is the dynamic time constant of the feeding rate transition in each control cycle; It refers to the feeding speed in each control cycle; It is the preset liquid level feedback coefficient, which indicates how many kg / min of feeding speed needs to be adjusted in reverse to correct the deviation of the standard liquid level by 1 meter. It is the real-time liquid level for each control cycle; This refers to the preset target liquid level of the molten glass. In one embodiment of this application, the values ​​of the preset liquid level feedback coefficient and the preset target liquid level are respectively... 1m, implementers can choose other values ​​based on the actual situation.

[0062] It should be noted that, It is the feeding rate executed in the previous cycle, which serves as the initial state for the inertial filtering in this cycle, ensuring that the feeding rate changes continuously without abrupt changes; This is a first-order inertial filter term, representing the adjustment amount for the current feeding speed. A larger value indicates a larger adjustment step, and thus a larger feeding speed command for each control cycle. It will accelerate the feeding speed towards the target. Closer; the smaller this value, the smaller the adjustment step, and the faster the feeding speed command in each control cycle. The slower the change, the better. Through dynamic adjustment... The system can automatically determine the transition speed of the feeding rate from the current value to the target value, achieving coordination between response speed and smooth operation.

[0063] It should also be noted that, It is a feedback correction term consisting of the deviation between the real-time liquid glass level and the target liquid glass level. When This indicates that the liquid level is too high, and the feeding rate command in each control cycle needs to be reduced. To cause the liquid level to drop; when This indicates that the liquid level is too low, and the feeding rate command in each control cycle needs to be increased. To raise the liquid level.

[0064] For example, setting a control cycle Target liquid level Liquid level feedback coefficient Assuming the feeding rate of the previous cycle... The target feeding speed that matches the current output rate Current real-time liquid level Using the calculation from the previous step The feeding speed command for each control cycle is as follows: The system obtains the feeding speed commands for each control cycle. Furthermore, since the feeding speed commands are purely mathematical calculations and may exceed the physical execution capabilities of the feeding system, direct output could damage equipment or cause system instability. Therefore, this application compares and limits the calculated feeding speed commands with the system's preset maximum and minimum speeds to achieve safety constraints. The final command, after safety assessment, is sent to the actuator and recorded as the current feeding speed for the next cycle, ensuring the continuity of the control process. Specifically:

[0065] Safety constraint judgment: The feeding speed command for each control cycle... With the preset maximum speed allowed by the feeding system Preset minimum speed Compare the values. If the value is outside the acceptable range, limit it to a safe boundary.

[0066] In one embodiment, the specific process of confining it within a safety boundary if it exceeds the range is as follows:

[0067] like Greater than At that time, a mandatory order ;like Less than At that time, a mandatory order This ensures that the feeding speed commands for each control cycle are within safe limits.

[0068] Command output: Feeding speed command after safety constraint judgment It is sent to the actuator of the feeding system.

[0069] Status update: Update the output instructions to reflect the next control cycle. The entire process will be in the next control cycle. Then repeat the process.

[0070] This application also discloses an automatic feeding control system for a glass furnace, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, an automatic feeding control method for a glass furnace according to this application is implemented.

[0071] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0072] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for automatic batch charging control of a glass furnace, characterized by, The steps include: collecting data from the glass furnace during the current control cycle, including: real-time average temperature, real-time liquid level, feeding rate, and discharge rate; The deviation between the real-time average temperature and the preset process standard temperature is taken as the deviation item. Calculate the average absolute temperature difference term of the glass furnace over all previous control cycles, including the current control cycle, and use the difference between the deviation term and the average absolute temperature difference term as the furnace heat index. The difference in material output between the current control cycle and the previous adjacent control cycle, along with the furnace heat index, are used to obtain the dynamic time constant through a nonlinear mapping function; the method for obtaining the dynamic time constant is as follows: Calculate the difference between the output of each control cycle and the output of the previous control cycle, calculate the absolute value of the ratio of the difference to the preset standard output, obtain the nonlinear mapping function value of the difference between the absolute value and the furnace heat index of the control cycle, and use the product of the sum of the value 1 and the nonlinear mapping function value and the preset basic transition time as the dynamic time constant of each control cycle. The feeding rate is processed by first-order inertial filtering using the dynamic time constant to obtain the main control quantity; based on the main control quantity and the feedback correction quantity, the feeding rate command for each control cycle is obtained; the feedback correction quantity characterizes the liquid level deviation of the molten glass. Input the feeding speed command into the feeding system to control the automatic feeding of the glass furnace.

2. The method of claim 1, wherein the method further comprises: The method for obtaining the deviation term is as follows: For the current control cycle, calculate the difference between the real-time average temperature and the preset process standard temperature, and use the ratio of the difference to the preset process standard temperature as the deviation term.

3. The method of claim 1, wherein the method further comprises: The method for obtaining the mean absolute temperature difference term is as follows: Calculate the average real-time average temperature of all previous control cycles, including the current control cycle, and calculate the absolute value of the difference between the real-time average temperature of each control cycle and the average value. The average of the ratios of all absolute differences to the preset process standard temperature is taken as the average absolute temperature difference term.

4. The method of claim 1, wherein the method further comprises: The nonlinear mapping function is a hyperbolic tangent function.

5. The method of claim 1, wherein the method further comprises: The method for obtaining the main control variable is as follows: For each control cycle, the ratio of the control cycle duration to the dynamic time constant is calculated, and the difference between the feeding rate and the feeding rate of the previous adjacent control cycle is calculated. The sum of the product of the ratio and the difference and the feeding rate of the previous adjacent control cycle is used as the main adjustment variable.

6. The method of claim 1, wherein, The method for obtaining the feeding speed command is as follows: For each control cycle, the liquid level difference between the real-time liquid level and the preset target liquid level is calculated, the product of the preset liquid level feedback coefficient and the liquid level difference is calculated, and the difference between the main adjustment amount and the product is used as the feeding speed command for each control cycle.

7. The method of claim 1, wherein the method further comprises: The step of inputting the feeding speed command into the feeding system includes: performing a safety constraint judgment on the feeding speed command, and sending the feeding speed command after the safety constraint judgment to the feeding system for execution.

8. The automatic feeding control method for a glass furnace according to claim 7, characterized in that, The safety constraint determination includes: comparing the value with the preset maximum speed and preset minimum speed of the feeding system; if the feeding speed command exceeds the range formed by the preset maximum speed and preset minimum speed, then limiting its value to that range.

9. A glass furnace automatic charging control system characterized by, include: A processor and a memory, the memory storing computer program instructions which, when executed by the processor, implement a method for automatic glass furnace charging control according to any one of claims 1-8.

Images