Intelligent temperature automatic control system for digital glass mold

Abstract

The invention relates to the field of industrial automation and intelligent control, and discloses an intelligent temperature automatic control system for a digital glass mold. The method comprises the following steps: acquiring mold surface temperature and heat flow data through a thermocouple array and a thermal infrared imager, and constructing heat flux and a disturbance coefficient; forming state vectors are established in combination with the forming process parameters and the heat flow distribution, and forming stability levels are generated through support vector regression; building a heat balance deviation model based on the thermophysical parameters and the environment variables, and predicting a temperature trend; a temperature control instruction is generated through the fuzzy neural controller, and temperature dynamic closed-loop control is achieved. The system improves the temperature control precision and thermal field balance of the glass mold, and is suitable for intelligent temperature control management in the glass container forming process.

Description

Technical Field

[0001] The present invention relates to the field of industrial automation and intelligent control, and in particular to a digital glass mold intelligent temperature control system based on heat flow data and intelligent algorithms, belonging to the field of intelligent technology for glass manufacturing equipment. Background Art

[0002] Traditional mold temperature control relies on single-point monitoring: In the existing glass bottle and jar production process, mold temperature control mostly relies on single-point measurement methods such as thermocouples, which can only reflect the local temperature of the mold and lack a comprehensive perception of the heat distribution of the entire mold surface, making it difficult to achieve accurate modeling of the dynamic characteristics of the thermal field.

[0003] Mold heating and cooling operations are mainly based on manual experience: Currently, most glass mold temperature control systems rely on operators to manually adjust the heater or air valve opening degree based on experience. The adjustment frequency is low and the response is delayed, making it impossible to achieve rapid response and dynamic control of temperature fluctuations, which seriously restricts the stability of product quality.

[0004] Lack of a quantitative assessment mechanism for thermal flow disturbances: During glass molding, thermal flow disturbances on the mold surface can lead to uneven cooling of the molten glass and easily produce stress concentration areas. However, existing technologies have not established scientific disturbance quantification indicators, making it impossible to dynamically identify and intervene in mold thermal unevenness.

[0005] Mold thermal balance modeling is still mainly based on static thermal models: The modeling method currently used for glass mold thermal management is mainly static heat conduction model, which is difficult to reflect the dynamic thermal change process of the mold during the molding cycle under the influence of factors such as the impact of glass liquid heat flow and environmental cooling, affecting the accuracy of the temperature control strategy.

[0006] Temperature control strategies lack predictability and intelligent adaptive mechanisms: Most existing temperature control strategies use a fixed threshold adjustment mechanism, and do not integrate temperature change trend prediction and system feedback tuning mechanisms. This results in low temperature control accuracy and poor system robustness, and is not suitable for the dynamic changes required by complex glass forming conditions.

[0007] The temperature control system feedback loop lacks closed-loop correction capabilities: Although some systems have temperature feedback functions, they do not introduce closed-loop control logic, especially the lack of an error self-correction mechanism linked to the heating / cooling execution module. This results in a delayed temperature control response and the inability to continuously maintain control accuracy within the target setting range. Summary of the Invention

[0008] In order to achieve the above-mentioned purpose, the present invention provides the following technical solutions: a digital glass mold intelligent temperature control system, comprising the following modules: Mold heat flow acquisition module, used to obtain temperature data on the surface of glass bottle molds With heat flow data , and perform filtering, interpolation and spatial heat map reconstruction to output stable heat flux parameters and heat flow perturbation coefficient ; Glass forming state recognition module for , heat flux density diagram, glass liquid flow characteristics and cooling rate to construct the molding state vector , and generate molding stability level parameters ; Mold thermal balance modeling module is used to calculate the thermal properties of the mold , environment variables As well as glass heat flux, molding state vector and other data to build mold thermal balance model and calculate dynamic thermal balance deviation value ; A temperature control strategy generation module is used to generate a temperature control strategy based on the thermal balance deviation value. and mold cycle status parameters , generate temperature control instruction vector through fuzzy neural control logic ; Intelligent temperature control execution module, used to control the temperature according to the temperature control instruction vector Control the operation of heating and cooling equipment and collect adjusted temperatures , based on the feedback error The temperature control instructions are automatically corrected, and the temperature feedback information is used to update the mold heat flow acquisition module to achieve closed-loop control.

[0009] Preferably, the mold heat flow collection module includes: The original temperature data of the mold surface collected by the thermocouple array and infrared thermal imager assembly , noise is suppressed by Gaussian filter function to obtain the temperature data of the mold surface The spatial heat flow data of the mold surface is obtained by bilinear interpolation reconstruction Among them, the original temperature data of the mold surface , mold surface temperature data , spatial heat flow data of the mold surface middle, and represents the mold surface coordinates, Indicates the collection time; Based on the Fourier heat conduction inversion algorithm, heat flux parameters are constructed : ; in, is the thermal conductivity of the mold material, is the temperature gradient; Calculate the heat flow perturbation coefficient : ; in, Spatial heat flow data for the mold surface The standard deviation of heat flow, Spatial heat flow data for the mold surface The average value of the heat flow disturbance coefficient Used to reflect the degree of uneven heating of the local mold.

[0010] Preferably, the glass forming state recognition module includes: Through the multi-frame heat map sequence of temperature data , , ... calculate the mold cooling rate curve , and combined with the heat flow disturbance coefficient And bottle molding process parameters Constructing the forming state vector : ; in, is the temperature data, is the average cooling rate, is the heat flux per unit area of the mold, is the heat flow disturbance coefficient, is a set of standardized process parameters; Generate molding stability level parameters based on support vector regression algorithm : ; in, It is the support vector regression algorithm model, the molding stability level parameter The value of reflects whether there are problems of overcooling, overheating, and inhomogeneity in glass forming, and serves as an input parameter for subsequent thermal balance modeling.

[0011] Preferably, the mold thermal balance modeling module includes: The mold heat flux parameter , heat flow disturbance coefficient , molding stability grade parameters , mold thermal properties parameters , and environment variables , the mold thermal properties parameters ,in is the specific heat capacity of the mold, is the mold density, is the mold thermal conductivity, the environmental variable =[ ],in is the ambient temperature, is the cooling wind speed, and the dynamic thermal balance deviation value is constructed based on the above data : ; in, The product of the specific heat capacity of the mold, the mold density and the mold temperature data versus time The partial differential of reflects the instantaneous heat change rate of the mold during heat conduction, so that the temperature prediction value at the future time point can be obtained , is the surface convection heat transfer coefficient, For the mold area; The deviation value It is used to quantify the difference between the current thermal state of the mold and the ideal steady state, and provide input for the generation of temperature control strategies.

[0012] Preferably, the temperature control strategy generation module includes: The temperature peak sequence collected in multiple cycles and valley sequence Perform fitting analysis and calculate the average amplitude , cycle time , and finally the cycle state parameters of the mold are obtained: ; Based on the dynamic thermal balance deviation value , molding stability grade parameters , mold molding cycle state parameters , construct the fuzzy neural controller input vector ; Obtaining a temperature prediction value at a future time point based on claim 4 , including the air valve opening through the adaptive control algorithm: ; in, is the expected temperature to be set; Generate temperature control command vector using adaptive fuzzy neural inference network ,in is the heating power value, is the inherent power of the heating system, is the opening of the cooling air valve; Preferably, the intelligent temperature control execution module includes: Real-time collection of temperature feedback data after execution , and the expected temperature Compare the errors and get the feedback error : ; The error is corrected by incremental PID temperature control instruction: ; is the proportional gain, reflecting the immediate response of the current error to the temperature control adjustment; is the integral gain, which is used to eliminate the steady-state error of the system; It is a differential gain that responds to the error change trend to improve the forward-looking nature of the regulation; Indicates the accumulated error, which helps in long-term error correction; is the error change rate, reflecting the error growth or convergence trend; Updated , and feeds back to the temperature control strategy generation module to achieve closed-loop adaptive adjustment; Feedback temperature data at the same time Send back to the mold heat flow acquisition module to update the heat flux parameters , realizing data closed loop and dynamic stability of temperature control within the system.

[0013] Compared with the prior art, the present invention has the following beneficial effects: Realize comprehensive dynamic perception of the mold's thermal field and improve thermal distribution accuracy: This invention combines multi-point infrared thermal imaging with mold modeling technology to construct a three-dimensional thermal distribution map, breaking the limitations of traditional single-point thermocouple monitoring, realizing dynamic perception and data reconstruction of the mold's full-surface thermal field, and significantly improving the spatial accuracy and real-time response of thermal field monitoring.

[0014] Possessing the ability to identify and locate thermal disturbances based on deviation analysis: Different from the existing technology that only performs simple alarms or threshold judgments on temperature changes, the present invention constructs a thermal disturbance identification model to quantitatively identify local overheating or overcooling areas in the mold thermal field, achieve accurate positioning of thermal anomaly distribution, and provide a scientific basis for temperature control.

[0015] Introducing a dynamic thermal balance adjustment strategy to improve temperature control adaptability: This invention integrates historical temperature change data with the current thermal state, and combines it with an adjustment strategy module based on a rule engine to dynamically allocate mold heating and cooling resources to form a temperature control closed-loop system that can adaptively adjust as the glass molding cycle changes, effectively avoiding over-adjustment or lagging adjustment.

[0016] Construct a feedback closed-loop and error correction mechanism to improve temperature control accuracy and stability: After the temperature control instruction is generated, the present invention performs closed-loop feedback control by real-time recovery of temperature response data, and introduces a self-correction mechanism based on error residual correction, which significantly improves the stability and convergence accuracy of the temperature control system under complex thermal load changes, and is superior to traditional open-loop or semi-open-loop control methods.

[0017] Possessing the ability to predict thermal fields and realize feedforward temperature control and regulation: The temperature control prediction mechanism proposed in the present invention integrates multi-dimensional data such as the mold structure model, historical working conditions and the glass liquid entering the mold temperature to construct a temperature change trend prediction model, which can realize prediction and adjustment preparation before temperature anomalies, which is fundamentally different from the "passive response" control method of the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of the system module flow provided for this application; Figure 2 Schematic diagram of the system modules provided for this application. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0020] refer to Figure 1-Figure 2 The embodiment of the present invention provides a digital glass mold intelligent temperature control system, including the following modules: Mold heat flow acquisition module, used to obtain temperature data on the surface of glass bottle molds With heat flow data , and perform filtering, interpolation and spatial heat map reconstruction to output stable heat flux parameters and heat flow perturbation coefficient .

[0021] In the mold heat flow acquisition module, during the production and molding process of glass bottles and jars, a thermocouple array is evenly distributed on multiple key thermal control areas on the outer surface of the glass mold. At the same time, a high-resolution infrared thermal imager component is set outside the mold to periodically and synchronously collect the original temperature data on the two-dimensional space of the mold surface, which is recorded as ,in and represents the mold surface coordinates, Indicates the acquisition time. Thermocouple data is mainly used to accurately capture the temperature change trend of a point, while the infrared thermal imager can provide an image of the overall thermal field distribution of the mold. The two can work together to compensate for the accuracy or coverage defects of a single sensor. Considering that the infrared image may be affected by reflections, surface stains, etc. during the acquisition process, resulting in high-frequency noise in the temperature data, Apply two-dimensional Gaussian filter function for convolution processing to achieve smoothing and noise suppression of temperature signal, and obtain the mold surface temperature distribution after filtering ; After obtaining the smoothed mold temperature distribution Finally, the bilinear interpolation method is used to fill in the sparse areas in the temperature field, reconstruct the continuous spatial thermal distribution data, and then further deduce the heat flux per unit area of the mold based on the principle of heat conduction. The heat flux calculation uses the following Fourier heat conduction inversion model: ; in, is the thermal conductivity of the mold material, is the temperature gradient, which can be obtained by querying in a two-dimensional grid using the central difference method; The final result Represents the heat flux density distribution on the mold surface at different spatial positions and time points for subsequent thermal disturbance analysis.

[0022] In order to quantitatively evaluate the degree of balance of thermal field distribution on the mold surface, the thermal flow disturbance coefficient is constructed. , and its calculation formula is: ; in, Spatial heat flow data for the mold surface The standard deviation of heat flow, Spatial heat flow data for the mold surface The average value of .

[0023] Glass forming state recognition module for , heat flux density diagram, glass liquid flow characteristics and cooling rate to construct the molding state vector , and generate molding stability level parameters .

[0024] In the glass forming state recognition module, the system first calls the mold heat flow acquisition module to obtain the mold temperature heat map sequence in a continuous time period, setting the sampling period to 100ms, and collecting a total of Mold surface temperature data after frame smoothing , and for a sequence of continuous heatmap frames ~ , calculate the mold cooling rate curve , and combined with the heat flow disturbance coefficient And bottle molding process parameters Constructing the forming state vector : ; in, is the temperature data, is the average cooling rate, is the heat flux per unit area of the mold, is the heat flow disturbance coefficient, To standardize the process parameter set, the vector structure can be uniformly and normally represented according to the actual process standard.

[0025] The shaping state vector Input is a pre-trained support vector regression (SVR) model that has been trained using a large amount of historical molding data for supervised learning. The learning goal is to predict the thermal stability level of the molding process.

[0026] The output parameter of the SVR model is the molding stability level value , this value is a continuous floating point number, which is used to reflect the degree of thermal abnormality during the glass bottle molding process: ; in, It is the support vector regression algorithm model, the molding stability level parameter The value of reflects whether there are problems of overcooling, overheating, and inhomogeneity in glass forming, and serves as an input parameter for subsequent thermal balance modeling.

[0027] Mold thermal balance modeling module is used to calculate the thermal properties of the mold , environment variables As well as glass heat flux, molding state vector and other data to build mold thermal balance model and calculate dynamic thermal balance deviation value .

[0028] In the mold heat balance modeling module, the mold heat flux parameter , heat flow disturbance coefficient , molding stability grade parameters , mold thermal properties parameters , and environment variables , the mold thermal properties parameters ,in is the specific heat capacity of the mold, is the mold density, is the mold thermal conductivity, the environmental variable =[ ],in is the ambient temperature, is the cooling wind speed.

[0029] The system is based on the specific heat capacity of the mold ,density, And mold surface temperature field , for this function with respect to time Calculate the partial derivative to get the spatial position of the mold per unit time The rate of heat change: ; This item represents the instantaneous heat accumulation or dissipation of the mold due to internal energy changes and is used to evaluate the temperature dynamic behavior of the mold in a local area.

[0030] The system further introduces surface convection heat exchange items based on environmental variable modeling: ; Among them, the heat transfer coefficient The empirical model can be used to determine the mold material, surface roughness, wind speed This item reflects the heat exchange process between the mold and the ambient air due to the temperature difference.

[0031] Finally, the mold heat balance modeling module combines the above heat input and output data and calculates the entire mold area through the following integral model Thermal equilibrium deviation on : ; The model actually reflects the thermal energy balance of the mold per unit area in a short period of time, that is, the difference between the input heat flux and the heat accumulation and heat loss inside the mold.

[0032] Calculated thermal balance deviation Characterizes the degree of deviation between the current mold thermal state and the ideal thermal equilibrium state. If: , it means that the mold is in a trend of overheating, and the system can control the cooling module to increase the cooling intensity; , indicating that the mold is cooling too quickly. You can reduce the cooling airflow or increase local heating; , indicating that the mold thermal state is stable and the current control strategy can be maintained.

[0033] The system will As a key input parameter, it is passed to the temperature control strategy generation module to achieve closed-loop control of the mold thermal state.

[0034] A temperature control strategy generation module is used to generate a temperature control strategy based on the thermal balance deviation value. and mold cycle status parameters , generate temperature control instruction vector through fuzzy neural control logic .

[0035] In the temperature control strategy generation module, the mold temperature sensor collects mold temperature data at a fixed frequency and extracts the temperature peak value sequence and valley value sequence within multiple consecutive mold temperature cycles through a sliding time window: The peak sequence represents the time sequence of the maximum temperature of the mold in each molding cycle, which is recorded as: ; The valley sequence represents the time sequence of the lowest mold temperature in the corresponding cycle, which is recorded as: ; The system calculates the average temperature amplitude through the above data fitting analysis and molding cycle time : ; Further calculate the mold molding cycle state parameters: ; This parameter reflects the temperature fluctuation intensity of the mold per unit time, thereby indirectly evaluating the stability of the mold thermal cycle.

[0036] Then, the system obtains from the previous module: dynamic thermal balance deviation value , molding stability grade parameters , mold molding cycle state parameters , construct the fuzzy neural controller input vector ; The system is based on the temperature partial differential model built in the thermal balance modeling module, combined with the current mold heat flux and thermophysical parameters, to predict the mold temperature at future times. , according to the expected temperature setting value , the system uses an adaptive control strategy to dynamically adjust the opening of the cooling air valve and obtain the cooling control instruction: ; This formula means that when the predicted temperature is high, the system automatically increases the air valve opening (i.e., the cooling intensity is improved), and when the predicted temperature is low, the air valve opening is reduced (i.e., the cooling capacity is reduced).

[0037] The system uses fuzzy neural reasoning network to input vector As the input layer signal of the forward neural network, after fuzzy rules and weight adaptive adjustment, the temperature control instruction vector is generated: ; in is the heating power value, is the inherent power of the heating system, is the opening of the cooling air valve; During the continuous operation of the system, the mold temperature data is updated in real time, and the controller re-acquires the new Input and keep adjusting Output enables self-learning and optimization of control strategies. This approach effectively ensures dynamic stability of mold temperature within the target range, helping to improve thermal stability and product consistency of glass molding.

[0038] Intelligent temperature control execution module, used to control the temperature according to the temperature control instruction vector Control the operation of heating and cooling equipment and collect adjusted temperatures , based on the feedback error The temperature control instructions are automatically corrected, and the temperature feedback information is used to update the mold heat flow acquisition module to achieve closed-loop control.

[0039] In the intelligent temperature control execution module, the module realizes dynamic adjustment of temperature control instructions and implements data closed loop within the system through real-time feedback error calculation and incremental PID control.

[0040] First, during the operation of the glass mold, the system periodically collects the two-dimensional spatiotemporal distribution data of the mold surface temperature through the embedded temperature sensor network, which is recorded as: ; The temperature feedback value represents the current time point Next, the mold is at the spatial point The actual temperature at .

[0041] At the same time, the system generates the temperature control target given by the module based on the temperature control strategy to obtain the expected temperature distribution of each area of the mold: ; By comparing the current feedback value with the expected value point by point, the system calculates the time point Temperature control error: ; The feedback error Input to the incremental PID control algorithm to calculate the temperature control correction required at the current time point: ; in: is the proportional gain, reflecting the immediate response of the current error to the temperature control adjustment; is the integral gain, which is used to eliminate the steady-state error of the system; It is a differential gain that responds to the error change trend to improve the forward-looking nature of the regulation; Indicates the accumulated error, which helps in long-term error correction; is the error change rate, reflecting the error growth or convergence trend.

[0042] Calculated correction value Superimposed with the temperature control instruction of the previous cycle, the temperature control output at the next time point is updated: ; Updated temperature control instructions This information is sent to the lower-level control layer, which makes fine adjustments to the heating and cooling systems, such as adjusting heater power or controlling air valve opening. After these adjustments, new temperature feedback is collected and the next control cycle begins, achieving adaptive closed-loop control.

[0043] At the same time, the current temperature feedback data It is also synchronously transmitted back to the mold heat flux acquisition module for dynamic updating of mold heat flux parameters. The change in heat flux reflects the direct impact of external cooling effects and heating behavior on the thermal state of the mold, thus providing accurate input for the next round of thermal balance modeling.

[0044] Through the above implementation, the intelligent temperature control execution module not only completes the error correction and precise execution of the temperature control instructions, but also realizes the following two data closed-loop mechanisms: Temperature control strategy closed loop: The PID-corrected instructions are fed back to the temperature control strategy generation module, so that the formation of future strategies takes into account historical control effects and improves the accuracy of adaptive adjustment; Thermal modeling data closed loop: Temperature feedback is used to correct mold heat flow estimation, making the mold thermal modeling module continuously close to the actual state, thereby improving the fit between the predicted temperature and the actual temperature.

[0045] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions therein may be combined with each other.

[0046] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present invention specification and drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. A digital glass mold intelligent temperature control system, characterized in that: Includes the following modules: Mold heat flow acquisition module, used to obtain temperature data on the surface of glass bottle molds With heat flow data , and perform filtering, interpolation and spatial heat map reconstruction to output stable heat flux parameters and heat flow disturbance coefficient ; A glass forming state recognition module is used to identify the glass forming state based on the temperature data. , heat flux density diagram, glass liquid flow characteristics and cooling rate to construct the molding state vector , and generate molding stability level parameters ; Mold thermal balance modeling module is used to calculate the thermal properties of the mold , environment variables As well as glass heat flux, molding state vector and other data to build mold thermal balance model and calculate dynamic thermal balance deviation value ; A temperature control strategy generation module is used to generate a temperature control strategy based on the thermal balance deviation value. and mold cycle status parameters , generate temperature control instruction vector through fuzzy neural control logic ; Intelligent temperature control execution module, used to control the temperature according to the temperature control instruction vector Control the operation of heating and cooling equipment and collect adjusted temperatures , based on the feedback error The temperature control instructions are automatically corrected, and the temperature feedback information is used to update the mold heat flow acquisition module to achieve closed-loop control.

2. A digital glass mold intelligent temperature control system according to claim 1, characterized in that: The mold heat flow acquisition module includes: The original temperature data of the mold surface collected by the thermocouple array and infrared thermal imager assembly , noise is suppressed by Gaussian filter function to obtain the temperature data of the mold surface The spatial heat flow data of the mold surface is obtained by bilinear interpolation reconstruction , wherein the original temperature data on the mold surface , mold surface temperature data , spatial heat flow data of mold surface middle, and represents the mold surface coordinates, Indicates the collection time; Based on the Fourier heat conduction inversion algorithm, heat flux parameters are constructed : ； in, is the thermal conductivity of the mold material, is the temperature gradient, which can be obtained by querying in a two-dimensional grid using the central difference method; Calculate the heat flow perturbation coefficient : ； in, Spatial heat flow data for the mold surface The standard deviation of heat flow, Spatial heat flow data for the mold surface The average value of the heat flow disturbance coefficient Used to reflect the degree of uneven heating of the local mold.

3. The digital glass mold intelligent temperature control system according to claim 1, characterized in that: The glass forming state recognition module includes: Through the multi-frame heat map sequence of temperature data , , ... calculate the mold cooling rate curve , and combined with the heat flow disturbance coefficient And bottle molding process parameters Constructing the forming state vector : ； in, is the temperature data, is the average cooling rate, is the heat flux per unit area of the mold, is the heat flow disturbance coefficient, is a set of standardized process parameters; Generate molding stability level parameters based on support vector regression algorithm : ； in, It is the support vector regression algorithm model, the molding stability level parameter The value of reflects whether there are problems of overcooling, overheating, and inhomogeneity in glass forming, and serves as an input parameter for subsequent thermal balance modeling.

4. The digital glass mold intelligent temperature control system according to claim 1, characterized in that: The mold thermal balance modeling module includes: The mold heat flux parameter , heat flow disturbance coefficient , molding stability grade parameters , mold thermal properties parameters , and environment variables , the mold thermal properties parameters ,in is the specific heat capacity of the mold, is the mold density, is the mold thermal conductivity, the environmental variable =[ ],in is the ambient temperature, is the cooling wind speed, and the dynamic thermal balance deviation value is constructed based on the above data : ； in, The product of the specific heat capacity of the mold, the mold density and the mold temperature data versus time The partial differential of reflects the instantaneous heat change rate of the mold during heat conduction, so that the temperature prediction value at the future time point can be obtained , is the surface convection heat transfer coefficient, For the mold area; The deviation value It is used to quantify the difference between the current thermal state of the mold and the ideal steady state, and provide input for the generation of temperature control strategies.

5. The digital glass mold intelligent temperature control system according to claim 1, characterized in that: The temperature control strategy generation module includes: The temperature peak sequence collected in multiple cycles and valley sequence Perform fitting analysis and calculate the average amplitude , cycle time , and finally the cycle state parameters of the mold are obtained: ； Based on the dynamic thermal balance deviation value , molding stability grade parameters , mold molding cycle state parameters , construct the fuzzy neural controller input vector ; Temperature prediction based on future time points , including the air valve opening through the adaptive control algorithm: ； in, is the expected temperature to be set; Generate temperature control command vector using adaptive fuzzy neural inference network ,in is the heating power value, is the inherent power of the heating system, is the cooling air valve opening.

6. The digital glass mold intelligent temperature control system according to claim 1 is characterized in that: The intelligent temperature control execution module includes: Real-time collection of temperature feedback data after execution , and the expected temperature Compare the errors and get the feedback error : ； The error is corrected by incremental PID temperature control instruction: ； is the proportional gain, reflecting the immediate response of the current error to the temperature control adjustment; is the integral gain, which is used to eliminate the steady-state error of the system; It is a differential gain that responds to the error change trend to improve the forward-looking nature of the regulation; Indicates the accumulated error, which helps in long-term error correction; is the error change rate, reflecting the error growth or convergence trend; Updated , and feeds back to the temperature control strategy generation module to achieve closed-loop adaptive adjustment; Feedback temperature data at the same time Send back to the mold heat flow acquisition module to update the heat flux parameters , achieving data closed loop and dynamic stability of temperature control within the system.

Images