A method and system for intelligent temperature control of a medical glass furnace

Abstract

This invention belongs to the field of furnace temperature control technology, specifically relating to an intelligent temperature control method and system for medical glass furnaces. The method includes: collecting furnace temperature and power data; calculating the standard deviations of temperature and power; obtaining a stability index; and determining the coupling change time based on the stability index. At the coupling change time, the offset is determined using a time-delay cross-correlation function, and a coupling correlation matrix is ​​constructed. Finally, the adjustment vector is solved based on the power control vector output by the PID controller and the correlation matrix to obtain the power setpoint of the heating unit for decoupling control. This invention solves the control mismatch problem caused by the cross-influence of operating condition changes and control loops by dynamically constructing the correlation matrix and performing decoupling control, thus improving the stability and accuracy of temperature control.

Description

Technical Field

[0001] This invention relates to the field of furnace temperature control technology. More specifically, this invention relates to an intelligent temperature control method and system for medical glass furnaces. Background Technology

[0002] Medical glass is a neutral borosilicate glass used to manufacture packaging materials such as medicine bottles and ampoules. Medical glass has stringent quality requirements. The core equipment in the medical glass production process is a large-scale continuous glass furnace, whose temperature control precision and stability directly determine the quality of the final product.

[0003] In related technologies, for example, Chinese patent document with authorization announcement number CN1293436C discloses a comprehensive intelligent furnace temperature control method for glass kilns, including: determining an appropriate oil pressure setpoint using a neural network based on the current temperature of the furnace body and the viscosity of the oil before switching heating; calculating the opening degree of the return oil valve using a neural network in the adjustment mainly based on the oil supply valve to achieve oil pressure compensation, thereby achieving furnace temperature control.

[0004] However, glass furnaces are typically complex thermal systems characterized by high inertia, long lag, and strong coupling of multiple variables. Multiple temperature zones within the furnace, such as the melting zone, refining zone, and working section, interact with each other through the flow of molten glass and thermal radiation. Adjusting the input in one zone can trigger complex, delayed chain reactions in other zones. Faced with these complex characteristics, current technologies have failed to address the problem of the simultaneous cross-influence of adjusting a single heating unit on the temperatures of multiple different zones. Furthermore, their neural network models establish fixed mapping relationships after training, failing to dynamically adjust and update the relationships between different temperature zones based on changes in actual furnace operating conditions. This easily leads to control overshoot and prolonged temperature oscillations, making it difficult to maintain the temperatures required for medical glass production. Consequently, medical glass is prone to defects such as bubbles and streaks, affecting product yield. Summary of the Invention

[0005] To address the technical problems of low temperature control accuracy and poor stability caused by neglecting coupling and hysteresis in existing kiln temperature control, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides an intelligent temperature control method for a medical glass furnace, comprising: obtaining a working condition stability index based on the weighted sum of the standard deviations of temperature within a time window of each temperature measuring point and the weighted sum of the standard deviations of power within a time window of each heating unit; determining the coupling change moment based on the working condition stability index; obtaining the offset between the temperature measuring point and the heating unit based on historical data of any temperature measuring point and historical data of any heating unit using a time-delay cross-correlation function; and obtaining the target time temperature control index based on the standard deviation of the temperature sequence of any temperature measuring point within a target time window, the standard deviation of the power sequence of any heating unit within the offset time window, and the covariance between the two sequences. The coupling between the temperature measuring point and the heating unit is constructed into an association matrix; the offset time is the historical time with the difference between it and the target time as the offset amount; for the deviation between the temperature of each temperature measuring point and its set temperature at the current time, the power control amount required for each temperature measuring point is obtained using a PID algorithm; a system of equations is constructed based on the association matrix corresponding to the current time and the power control vector composed of the power control amount required for each temperature measuring point; the adjustment vector is obtained by solving the system of equations; the power of the heating unit at the previous time is added to its corresponding value in the adjustment vector to obtain the power set value of the current heating unit; the output power of each heating unit is adjusted to the power set value to achieve kiln temperature control.

[0007] This invention first calculates the operating condition stability index using the standard deviations of temperature and power, thus dynamically dividing the kiln's operating state. Based on the operating condition stability index, it determines the timing of coupling changes and dynamically schedules model updates. Furthermore, this invention obtains the actual time offset between the heating unit and the temperature measurement point using a time-delay cross-correlation function, solving the heating lag problem. On this basis, it constructs an association matrix using covariance and standard deviation to reflect the true coupling relationship between variables under the current operating condition. Finally, during PID control, this invention uses the association matrix to construct a system of equations to solve the power control vector output by the PID controller, obtaining the decoupled adjustment vector. This allows for the calculation of a power setpoint that simultaneously meets the control requirements of all temperature zones and compensates for the cross-influence of each heating unit. This invention solves the control mismatch and temperature fluctuation problems caused by traditional control methods, which suffer from fixed models, inability to adapt to changing operating conditions, and neglect of cross-influence between control loops. It achieves dynamic decoupled control based on actual operating conditions, improving the stability and accuracy of kiln temperature control.

[0008] Preferably, the operating condition stability index satisfies the following relationship: In the formula, For a moment The kiln operating condition stability index, For a moment Within the time window The standard deviation of temperature data from each temperature measurement point For a moment Within the time window The standard deviation of the input power of each heating unit For the first The weight of each temperature measurement point For the first The weight of each heating unit, For the first The maximum standard deviation of each temperature measurement point For the first Maximum standard deviation of each heating unit The number of temperature measurement points, The number of heating units, It is an exponential function with the natural constant as the base.

[0009] This invention obtains the stability index of the calculated operating conditions by weighted summation of the standard deviation of temperature and the standard deviation of power, highlighting the influence of key areas and realizing the sensitive capture and accurate quantification of changes in the kiln's operating status, providing a reliable basis for subsequent accurate judgment of the timing of coupled changes.

[0010] Preferably, determining the coupling change time based on the operating condition stability index includes: after the kiln has entered real-time operation, the moment when the operating condition stability index first falls below the absolute threshold is taken as the first coupling change time; any moment after the first coupling change time that satisfies the coupling relationship change condition is also taken as the coupling change time.

[0011] This invention determines the timing of coupling changes by setting a stability index, ensuring that the kiln can be immediately identified when it transitions from a stable operating condition to a fluctuating operating condition. This allows for timely updates to the correlation matrix, enabling continuous tracking and dynamic response to changes in operating conditions and ensuring that the control model always matches the actual state of the kiln.

[0012] Preferably, the coupling relationship change condition includes: when the mean of the operating condition stability index within the time window is less than the absolute threshold, and the standard deviation of the operating condition stability index within the time window is greater than the change threshold.

[0013] This invention can determine whether the kiln is in a fluctuating state by using the mean and absolute threshold of the operating condition stability index, and can determine whether the operating condition is undergoing drastic changes by using the standard deviation and change threshold of the operating condition stability index. This distinguishes between stable fluctuation states and dynamic transformation processes, and achieves accurate identification of the moment when the coupling relationship changes substantially, avoiding unnecessary model reconstruction.

[0014] Preferably, the coupling satisfies the following relationship: In the formula, For the target time The temperature measurement point and the first Coupling between heating units For the target time Data sequence within a time window of each temperature measurement point The offset time of the target time Data sequence within the time window of each heating unit For the target time The temperature measurement point and the first The offset between each heating unit Let covariance function be used. It is a function of standard deviation. Parameters to prevent division by zero errors.

[0015] This invention obtains coupling by using the covariance between the temperature sequence and the power sequence at the offset time, as well as the variance of the power sequence. While evaluating the degree of coordinated change in the temperature response to power, it can effectively suppress the interference of large fluctuations in the power of the heating unit itself on the coupling calculation results. This makes the obtained coupling value more accurately reflect the true physical correlation strength between the temperature measurement point and the heating unit, and improves the accuracy of the subsequent correlation matrix.

[0016] Preferably, the offset is the offset that maximizes the time-delay cross-correlation function.

[0017] Preferably, the step of using a PID algorithm to obtain the power control amount required for each temperature measuring point includes: setting a PID controller for each temperature measuring point, inputting the deviation between the temperature of each temperature measuring point and its set temperature into the corresponding PID controller, and obtaining the power control amount required for each temperature measuring point.

[0018] Preferably, the step of constructing a set of equations based on the correlation matrix corresponding to the current time and the power control vector consisting of the power control quantities required by each temperature measurement point includes: using the correlation matrix as the coefficient matrix of the set of equations and using the power control vector as the constant vector of the set of equations.

[0019] Preferably, the correlation matrix corresponding to the current moment is the correlation matrix obtained at the moment of the most recent coupling change.

[0020] Secondly, the present invention provides an intelligent temperature control system for a medical glass furnace, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned intelligent temperature control method for a medical glass furnace is implemented.

[0021] By adopting the above technical solution, a computer program for intelligent temperature control of a medical glass furnace is generated and stored in a memory for loading and execution by a processor. Terminal devices are then manufactured based on the memory and processor for convenient use.

[0022] The beneficial effects of this invention are as follows: By collecting multi-dimensional state data of the kiln and calculating the operating condition stability index based on the standard deviations of temperature and power, this invention achieves real-time classification of the kiln's operating conditions. Furthermore, this invention dynamically identifies the moments when the coupling relationship needs updating by observing changes in the operating condition stability index, avoiding unnecessary model reconstruction when the operating condition is stable and ensuring timely response during fluctuations. At the moment of change in the coupling relationship, this invention uses a time-delay cross-correlation function to determine the time offset between the heating unit and the temperature measurement point, and calculates the coupling based on the covariance and power standard deviation, constructing an association matrix that reflects the dynamic characteristics of the current operating condition. The required power control vector is calculated using a PID controller, and the association matrix is ​​used for decoupling to obtain the adjustment power vector for each heating unit, achieving compensation for the cross-influence of multiple temperature zones. This invention can actively compensate for the cross-influence between heating units, rather than merely passively responding to temperature deviations, thereby improving the overall control robustness and temperature uniformity of the medical glass kiln under complex and variable operating conditions. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating an intelligent temperature control method for a medical glass furnace according to the present invention;

[0024] Figure 2 This is a schematic diagram illustrating the comparison of the temperature control effects of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0027] This invention discloses an intelligent temperature control method for a medical glass furnace, referring to... Figure 1 This includes steps S1-S4:

[0028] S1. Collect multi-dimensional status data of the kiln and obtain the operating condition stability index based on the temperature data within the time window of each temperature measurement point and the power data within the time window of each heating unit.

[0029] It should be noted that the operation of a kiln is not static, but includes different operating stages such as stable production, fluctuating feeding, and the heating process. The dynamic characteristics of the kiln system and the correlation between variables differ at each stage. Therefore, before implementing control, the raw data stream needs to be processed to identify the current operating stage of the kiln, laying the foundation for building a precise stage-specific model. Thus, this invention collects multi-dimensional state data of the kiln and obtains the operating condition stability index based on the temperature data within the time window of each temperature measurement point and the power data within the time window of each heating unit.

[0030] Specifically, temperature measurement points are set up in different physical regions of the kiln to collect temperature sequences and input power sequences of each heating unit. Time windows are set up to record the time before each moment. Each data point represents the data within a given time window. The standard deviation of the temperature data within the time window for each temperature measurement point at each time point is obtained, as well as the standard deviation of the input power within the time window for each heating unit at each time point.

[0031] Furthermore, the maximum standard deviation of temperature at each temperature measurement point during the historical normal operation of the kiln is taken as the maximum standard deviation of each temperature measurement point; the maximum standard deviation of input power at each heating unit during the historical normal operation of the kiln is taken as the maximum standard deviation of each heating unit; the ratio between the standard deviation of temperature data within the time window of each temperature measurement point at each moment and the maximum standard deviation of each temperature measurement point is obtained, and the ratios corresponding to each temperature measurement point at each moment are weighted and summed to obtain the weighted sum of temperature indicators; the ratio between the standard deviation of input power within the time window of each heating unit at each moment and the maximum standard deviation of each heating unit is obtained, and the ratios corresponding to each heating unit at each moment are weighted and summed to obtain the weighted sum of power indicators; the operating condition stability index is obtained based on the weighted sum of temperature indicators and the weighted sum of power indicators.

[0032] For example, temperature measuring points are set in the melting zone, clarification zone, and working section of the kiln to collect the input power of the fuel spray gun and heating electrode. The weights of the temperature measuring points are: 0.4 for the melting zone, 0.2 for the clarification zone, and 0.4 for the working section. The weights of the heating units are 0.5 for the fuel spray gun and 0.5 for the heating electrode. The implementers can determine the temperature measuring points and their corresponding weights, as well as the heating units to be monitored and their corresponding weights, according to the actual situation.

[0033] Specifically, the operating condition stability index satisfies the following relationship:

[0034] ;

[0035] In the formula, For a moment The operating condition stability index, For a moment Within the time window The standard deviation of temperature data from each temperature measurement point For a moment Within the time window The standard deviation of the input power of each heating unit For the first The weight of each temperature measurement point For the first The weight of each heating unit, For the first The maximum standard deviation of each temperature measurement point For the first Maximum standard deviation of each heating unit The number of temperature measurement points, The number of heating units, It is an exponential function with the natural constant as the base.

[0036] in, This is a weighted sum of temperature indicators; the value represents the overall temperature variation of the kiln. A larger value indicates that the temperature variation occurred at time [time missing]. The more drastic the temperature changes within the time window corresponding to each temperature measurement point, the more likely the kiln is undergoing a transition in operating conditions. The smaller the operating condition stability index, the better; a smaller value indicates that at time... The more gradual the temperature data changes within the time window corresponding to each temperature measurement point, the more likely the kiln is in a stable production process. The higher the working condition stability index, the better.

[0037] The power index is a weighted sum, representing the overall power variation of the kiln. A larger value indicates that the kiln control system is likely to be performing drastic and frequent adjustments, and the kiln is likely to be in a dynamically changing heating state, with a smaller operating condition stability index. Conversely, a smaller value indicates that the kiln control system is likely not performing adjustments, and the kiln is likely to be in a stable heating state, with a larger operating condition stability index.

[0038] S2. Obtain the timing of coupling changes based on the operating condition stability index.

[0039] It should be noted that as the kiln system operates, the coupling relationship between the various temperature measuring points and the heating units in the kiln will also change. However, the small data fluctuations generated during normal kiln operation do not represent a substantial change in the thermodynamic coupling characteristics between the temperature measuring points and the heating units inside the kiln. It is not necessary to re-evaluate the coupling relationship between the temperature measuring points and the heating units at every moment to adjust the kiln temperature. Therefore, this invention obtains the coupling change time based on the operating condition stability index.

[0040] Specifically, after the kiln is in real-time operation, the moment when the operating condition stability index first falls below the absolute threshold is taken as the first coupling change moment; any moment after the first coupling change moment that satisfies the coupling relationship change condition is also taken as the coupling change moment. The coupling relationship change condition includes: when the mean of the operating condition stability index within the time window is less than the absolute threshold, and the standard deviation of the operating condition stability index within the time window is greater than the change threshold.

[0041] For example, the absolute threshold is 0.6 and the variable threshold is 0.1. Implementers can determine the absolute threshold and variable threshold according to the actual situation.

[0042] S3. At the moment of coupling change, construct an association matrix based on the temperature data of the temperature measurement point and the power data of the heating unit.

[0043] It should be noted that the coupling relationships between various variables within a glass furnace are not static but dynamically change with the operating conditions. For example, during the feeding fluctuation phase, the newly added cold material temporarily enhances the thermodynamic coupling between the melting and refining zones. When the coupling relationship between the temperature measuring point and the heating unit changes, using the control model from the stable production phase will lead to control mismatch. Therefore, this invention utilizes the actual operating data of the furnace to construct a correlation matrix that reflects its current dynamic characteristics.

[0044] Specifically, taking any coupling change moment as the target moment, the temperature sequence composed of all data before the target moment at any temperature measuring point and the power sequence composed of all data before the target moment at any heating unit are obtained. The time-delay cross-correlation function between the two sequences is obtained, and the offset that makes the time-delay cross-correlation function reach its maximum value is obtained. This offset is used as the offset between the heating unit and the temperature measuring point. Based on the covariance between the temperature sequence within the time window of any temperature measuring point at the target moment and the power sequence of any heating unit within the time window of the offset moment at the target moment, as well as the standard deviation of the power sequence within the time window of the offset moment at the target moment, the coupling between the temperature measuring point and the heating unit is obtained. The offset moment of the target moment is the moment before the target moment, and the difference between the offset moment and the target moment is the offset between the heating unit and the temperature measuring point.

[0045] Furthermore, the coupling between each temperature measuring point and each heating unit is combined into an association matrix. For example, the coupling between the first temperature measuring point and the first heating unit constitutes the value in the first row and first column of the association matrix, the coupling between the first temperature measuring point and the second heating unit constitutes the value in the first row and second column of the association matrix, and the coupling between the second temperature measuring point and the first heating unit constitutes the value in the second row and first column of the association matrix.

[0046] Specifically, the coupling between the temperature measuring point and the heating unit satisfies the following relationship:

[0047] ;

[0048] In the formula, For the target time The temperature measurement point and the first Coupling between heating units For the target time Data sequence within a time window of each temperature measurement point The offset time of the target time Data sequence within the time window of each heating unit For the target time The temperature measurement point and the first The offset between each heating unit Let covariance function be used. It is a function of standard deviation. To prevent division by zero errors in parameters, this embodiment... The value is 0.001, and the implementers can adjust it according to the actual situation. The value of .

[0049] in, This represents the heating lag of the kiln heating system, reflecting the first When the energy applied by the heating unit changes, it is reflected in the first... The time required for each temperature measurement point, through Power data that matches the temperature data at the target time can be found, thus enabling better acquisition of the first... The heating unit and the first The coupling relationship between the temperature measurement points.

[0050] It represents and The degree of coordinated change between them; the larger the value, the more significant the change. The temperature at the first temperature measuring point affects the... The greater the power regulation response of the heating unit, the better the response of the heating unit. The temperature measurement point and the first The greater the coupling between heating units, the smaller the value indicates that the first heating unit... The temperature at the first temperature measuring point affects the... The smaller the power regulation response of the heating unit, the better. The temperature measurement point and the first The smaller the coupling between heating units;

[0051] In the calculation of covariance, the greater the difference in fluctuation between two data series, the more likely it is that even if the correlation between the two data series is not high, there is still a large covariance between them. In kiln control, power is usually a parameter that fluctuates significantly over a short period of time. Therefore, through… Divide by This method reduces the impact of power fluctuations on the coupling between the temperature measuring point and the heating unit.

[0052] S4. Obtain the power setting value of each heating unit based on the correlation matrix and the temperature deviation of each temperature measuring point, and control the kiln temperature based on the power setting value of each heating unit.

[0053] It should be noted that the reason why existing control methods perform poorly in multi-temperature zone coordinated control is that they fail to effectively handle the cross-influence between control loops. For example, performing an independent temperature adjustment on the melting zone will inevitably cause unexpected temperature fluctuations in downstream temperature zones such as the clarification zone due to heat transfer. Therefore, this invention obtains the power setpoint of each heating unit based on the correlation matrix and the temperature deviation of each temperature measuring point, and improves the situation of unexpected temperature fluctuations through the correlation matrix.

[0054] Specifically, a PID controller is set up for each temperature measuring point to obtain the difference between the temperature of each temperature measuring point and the temperature setpoint at the current moment. The difference between the temperature of each temperature measuring point and the temperature setpoint is input into the PID controller of the corresponding temperature measuring point to obtain the power control quantity required for each temperature measuring point. The power control quantity of each temperature measuring point is constructed into a power control vector.

[0055] For example, if there are two temperature measurement points, the power control quantity of the first temperature measurement point is 10kW and the power control quantity of the second temperature measurement point is 7kW, then the power control vector is (10,7).

[0056] Furthermore, the correlation matrix obtained at the closest coupling change moment to the current moment is used as the correlation matrix at the current moment. The correlation matrix at the current moment is used as the coefficient matrix of the equation system. The power control vector at the current moment is used as the constant vector. The adjustment vector is used as the unknown vector. The equation system is solved to obtain the adjustment vector. Each value in the adjustment vector is the power adjustment amount of each heating unit. The power setpoint of each heating unit at the previous moment is added to its corresponding value in the adjustment vector to obtain the power setpoint of the heating unit at the current moment.

[0057] Furthermore, the output power of each heating unit is adjusted to the power set value to achieve temperature control of the kiln.

[0058] Specifically, the adjustment vector satisfies the following relation:

[0059] ;

[0060] In the formula, Let this be the adjustment vector at the current moment. This is the correlation matrix at the current time. This is the power control vector at the current moment.

[0061] in, This represents the power required for the temperature at each measuring point to change to the set value. However, each measuring point is affected by all heating units. Therefore, it is necessary to solve for the power adjustment of each heating unit based on the influence (correlation matrix) of each heating unit on each measuring point. and Construct the adjustment vector The system of equations provides the power adjustment amount of each heating unit to meet the temperature change requirements of each temperature measuring point.

[0062] For example, if the number of temperature measuring points and the number of heating units are both 2, the correlation matrix... Power control vector ,pass and The constructed system of equations is The solution obtained is the adjustment vector. , and This refers to the power adjustment amount of the first heating unit and the power adjustment amount of the second heating unit.

[0063] For example, Figure 2 This is a comparison diagram of the temperature control effect of the present invention. In the diagram, the temperature control of the present invention makes the temperature closer to the set temperature. The gray area in the diagram corresponds to the time of disturbance of the kiln. It can be seen from the diagram that the temperature control of the kiln of the present invention is more resistant to disturbance interference and has a better control effect.

[0064] This invention also discloses an intelligent temperature control system for a medical glass furnace, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement an intelligent temperature control method for a medical glass furnace according to the present invention.

[0065] 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.

Claims

1. A method for intelligent temperature control of a medical glass furnace, characterized in that, include: The ratio between the standard deviation and the maximum standard deviation of temperature data within the time window of each temperature measurement point is obtained. The ratios corresponding to each temperature measurement point are weighted and summed to obtain the weighted sum of temperature indicators. The ratio between the standard deviation of input power within the time window of each heating unit and the maximum standard deviation of each heating unit is obtained. The ratios corresponding to each heating unit are weighted and summed to obtain the weighted sum of power indicators. The operating condition stability index is obtained based on the weighted sum of temperature indicators and the weighted sum of power indicators. The coupling change time is determined based on the operating condition stability index, and the coupling change time is taken as the target time. The offset between the temperature measuring point and the heating unit is obtained by using the time-delay cross-correlation function based on the historical data of any temperature measuring point and the historical data of any heating unit before the target time. Based on the standard deviation of the temperature sequence of any temperature measurement point within the target time window, the standard deviation of the power sequence of any heating unit within the offset time window, and the covariance between the two sequences, the coupling between the temperature measurement point and the heating unit at the target time is obtained, and the coupling between each temperature measurement point and each heating unit is combined into an correlation matrix; the offset time is the historical time with a difference of offset from the target time. For the deviation between the temperature of each temperature measuring point at the current moment and its set temperature, the PID algorithm is used to obtain the power control amount required for each temperature measuring point; a set of equations is constructed based on the correlation matrix corresponding to the current moment and the power control vector composed of the power control amount required for each temperature measuring point; the adjustment vector of each heating unit is obtained by solving the set of equations; the power of each heating unit at the previous moment is added to its corresponding value in the adjustment vector to obtain the power set value of the current heating unit; the correlation matrix corresponding to the current moment is the correlation matrix obtained at the moment of the closest coupling change to the current moment. The output power of each heating unit is adjusted to the power set value to achieve kiln temperature control.

2. The intelligent temperature control method for a medical glass furnace according to claim 1, characterized in that, The operating condition stability index satisfies the following relationship: ； In the formula, For a moment The kiln operating condition stability index, For a moment Within the time window The standard deviation of temperature data from each temperature measurement point For a moment Within the time window The standard deviation of the input power of each heating unit For the first The weight of each temperature measurement point For the first The weight of each heating unit, For the first The maximum standard deviation of each temperature measurement point For the first Maximum standard deviation of each heating unit The number of temperature measurement points, The number of heating units, It is an exponential function with the natural constant as the base.

3. The intelligent temperature control method for a medical glass furnace according to claim 1, characterized in that, The determination of the coupling change time based on the operating condition stability index includes: after the kiln is in real-time operation, the moment when the operating condition stability index first falls below the absolute threshold is taken as the first coupling change time; any moment after the first coupling change time that satisfies the coupling relationship change condition is also taken as the coupling change time; the coupling relationship change condition includes: when the mean of the operating condition stability index within the time window is less than the absolute threshold, and the standard deviation of the operating condition stability index within the time window is greater than the change threshold.

4. The intelligent temperature control method for a medical glass furnace according to claim 1, characterized in that, The coupling satisfies the following relationship: ； In the formula, For the target time The temperature measurement point and the first Coupling between heating units For the target time Data sequence within a time window of each temperature measurement point The offset time of the target time Data sequence within the time window of each heating unit For the target time The temperature measurement point and the first The offset between each heating unit Let covariance function be used. It is a function of standard deviation. Parameters to prevent division by zero errors.

5. The intelligent temperature control method for a medical glass furnace according to claim 1, characterized in that, The offset is the offset that makes the time-delay cross-correlation function reach its maximum value.

6. The intelligent temperature control method for a medical glass furnace according to claim 1, characterized in that, The step of using a PID algorithm to obtain the power control amount required for each temperature measurement point includes: setting a PID controller for each temperature measurement point, inputting the deviation between the temperature of each temperature measurement point and its set temperature into the corresponding PID controller, and obtaining the power control amount required for each temperature measurement point.

7. The intelligent temperature control method for a medical glass furnace according to claim 1, characterized in that, The process of constructing a set of equations based on the correlation matrix corresponding to the current moment and the power control vector consisting of the power control quantities required by each temperature measurement point includes: using the correlation matrix as the coefficient matrix of the set of equations and the power control vector as the constant vector of the set of equations.

8. A smart temperature control system for a medical glass furnace, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a method for intelligent temperature control of a medical glass furnace according to any one of claims 1-7.

Images