Method and system for dynamically adjusting anti-skid texture of glass container

Abstract

The invention provides a method and a system for dynamically adjusting anti-skid textures of a glass container, which are applied to the technical field of glass manufacturing, and are characterized in that feedforward information and feedback information are acquired in real time, an initial adjustment amount is predicted, a target adjustment amount is obtained through correction according to the feedback information, and then adjustment is executed to cope with dynamic changes in a production process. The method has the advantages of improving texture control precision, stabilizing texture quality and being suitable for high-speed production.

Description

Technical Field

[0001] This application relates to the field of glass manufacturing technology, and in particular to a method and system for dynamically adjusting the anti-slip texture of glass containers. Background Technology

[0002] On high-speed glass blowing production lines, the forming process typically involves dripping precisely measured molten glass into a preheated forming mold. After the mold closes rapidly, the glass is inflated by a main blowing gas, causing it to conform evenly to the inner wall of the mold, thus forming the main shape of the container. Various techniques can be employed to impart specific functional or decorative textures to the outer surface of the glass container, such as microscopic anti-slip textures. One unconventional texture-forming method utilizes an auxiliary gas during the blowing process. Specifically, this system pre-installs multiple micro-channels on the inner wall of the mold. As the glass expands and conforms to the inner wall of the mold, auxiliary gas is blown onto the surface of the glass through these channels. Parameters such as the flow rate, pressure, temperature of the auxiliary gas, as well as the blowing duration and the selection of the blowing channels, locally influence the flow behavior and cooling rate of the glass on the mold surface. This localized intervention can create wrinkles or undulations with specific microstructures on the outer surface of the glass in a semi-solid state, thereby achieving the desired anti-slip or other surface functions.

[0003] However, in actual high-speed blow molding production environments, the quality of microtextures formed through assisted air blowing faces numerous challenges, making it difficult to consistently control texture depth, density, uniformity, and structural consistency. Simultaneously, the shapes of modern glass containers are becoming increasingly diverse, with many containers possessing complex curvature regions, such as the shoulder, neck, or bottom transition areas. In these regions, the design of the air channels within the mold wall and the control of the assisted airflow are more complex. The flow behavior of the glass stock in these complex-shaped regions differs from that in the cylindrical parts of the container, and its sensitivity to blowing parameters also varies. This makes achieving uniform and consistent textures in complex-shaped regions a technical challenge, requiring more refined and regionalized control strategies.

[0004] Therefore, existing technologies have difficulty overcoming the combined effects of multiple dynamic fluctuation sources such as glass material temperature / viscosity, mold temperature, and blowing parameters. It is difficult to ensure that the texture quality remains stable and meets requirements under high-speed production cycles by adjusting auxiliary blowing parameters and local mold temperature in real time and in a coordinated manner, while also adapting to the rapid switching of different product models. Summary of the Invention

[0005] In view of the shortcomings of the prior art, this application provides a method and system for dynamically adjusting the anti-slip texture of glass containers, which has the advantages of responding to dynamic changes in the production process, improving texture control accuracy, stabilizing texture quality, and being suitable for high-speed production.

[0006] In a first aspect, a method for dynamically adjusting the anti-slip texture of a glass container, the method comprising the steps of:

[0007] S1: Real-time acquisition of glass material temperature, mold local temperature, and auxiliary blowing parameters as feedforward information; real-time acquisition of texture deviation of the micro-texture quality of the outer surface of the formed container as feedback information;

[0008] S2: Based on the feedforward information, when predicting the flow and cooling state of the glass material on the mold surface, the initial adjustment amount of the auxiliary blowing parameters and the local temperature of the mold is made.

[0009] S3: Based on the feedback information, correct the initial adjustment amount, and calculate the target adjustment amount for the auxiliary blowing parameters and the local temperature of the mold;

[0010] S4: Execute the target adjustment amount to adjust the parameters of the auxiliary blowing system and the local temperature of the mold, and control the texture formation of the subsequent container.

[0011] Furthermore, step S1 includes:

[0012] S11: The temperature of the glass material at multiple locations within the flow channel is measured non-contactly using an infrared sensor, and the weighted average is used to obtain the glass material temperature.

[0013] S12: By embedding a thermocouple array into the mold surface and fitting the temperature distribution on the mold surface through spatial interpolation, the local temperature of the mold is obtained;

[0014] S13: Using a flow meter, pressure sensor, and temperature sensor, collect flow rate, pressure, and temperature data from the auxiliary blowing pipe, record the opening delay time and blowing duration of the auxiliary blowing pipe, and obtain the auxiliary blowing parameters; use the glass material temperature, the local temperature of the mold, and the auxiliary blowing parameters as the feedforward information.

[0015] S14: Using a visual inspection system, an image of the outer surface of the formed glass container is captured. Texture features are extracted through image processing and compared with preset standard texture features to calculate the texture deviation. The texture deviation is used as the feedback information.

[0016] Furthermore, step S2 includes:

[0017] S21: Construct a predictive model for the flow and cooling state of glass material, and output the predicted values ​​of temperature and flow rate of glass material at multiple key locations on the mold surface;

[0018] S22: Establish a calculation model for the auxiliary blowing parameters and the local temperature adjustment of the mold. Input the predicted values ​​of glass material temperature and flow rate into the calculation model to calculate the preliminary adjustment of the auxiliary blowing parameters and the local temperature of the mold, so that the glass material temperature and flow rate are close to the preset target values.

[0019] Furthermore, step S21 includes:

[0020] S211: The corners of the mold cavity, areas with large curvature changes, and areas directly affected by the auxiliary blowing airflow are designated as key locations on the surface of the glass material on the mold.

[0021] S212: Establish a finite element model that includes the physical property parameters of the glass material, the structural parameters of the mold, and the auxiliary blowing parameters, and set the initial temperature distribution of the glass material and the temperature boundary conditions of the mold.

[0022] S213: Based on the initial temperature and the temperature boundary conditions, simulate the flow and cooling process of the glass material in the mold cavity, and output the predicted values ​​of the temperature and flow rate of the glass material at multiple key locations on the mold surface.

[0023] Furthermore, step S212 includes:

[0024] S2121: Based on the composition of the glass material, obtain the physical property parameters of the glass material; based on the mold design drawings, obtain the mold structure parameters; refer to the technical parameters of the auxiliary blowing equipment, the auxiliary blowing parameters;

[0025] S2122: Construct a finite element model based on the physical property parameters of the glass material, the structural parameters of the mold, and the auxiliary blowing parameters;

[0026] S2123: Based on the temperature distribution measurement results of the glass material in the flow channel, the initial temperature distribution of the glass material is fitted using an interpolation method as the initial condition of the finite element model.

[0027] S2124: Based on the design parameters of the mold cooling system, set the heat transfer coefficient between the mold and the cooling water as the temperature boundary condition of the finite element model.

[0028] Furthermore, step S213 includes:

[0029] S2131: Import the finite element model into the solver and set the convergence criteria of the solver;

[0030] S2132: Based on the initial temperature distribution of the glass material and the temperature boundary conditions of the mold, iteratively solve the temperature field and velocity field distribution of the glass material in the mold cavity until the energy residual in the iteration process reaches the convergence criterion of the solver.

[0031] S2133: After the convergence criterion is met, record the stable time point at which each key position reaches a stable state;

[0032] S2134: Based on the stable time point, select the corresponding temperature and flow velocity data in the temperature field and velocity field distribution of the glass material in the mold cavity as the predicted values ​​of the temperature and flow velocity at each key location.

[0033] Furthermore, step S22 includes:

[0034] S221: Establish a calculation model for auxiliary blowing parameters and mold local temperature adjustment, including a PID control module and a fuzzy logic control module;

[0035] S222: Input the predicted values ​​of glass material temperature and flow rate into the fuzzy logic control module to calculate the adjustment range of the auxiliary blowing parameters and the PID control parameters of the local temperature of the mold;

[0036] S223: Input the adjustment range of the PID control parameters into the PID control module to calculate the initial adjustment amount of the auxiliary blowing parameters and the local temperature of the mold, so as to make the glass material temperature and flow rate close to the preset target value.

[0037] Furthermore, step S3 includes:

[0038] S31: Construct a mapping model between texture deviation and auxiliary blowing parameters and mold local temperature adjustment amount, input the feedback information into the mapping model, and obtain the correction coefficients of the auxiliary blowing parameters and the mold local temperature;

[0039] S32: Multiply the initial adjustment amount by the correction coefficient to obtain the target adjustment amount of the auxiliary blowing parameters and the local temperature of the mold.

[0040] Furthermore, step S31 includes:

[0041] S311: Set the input layer, the number of hidden layer neurons is set to 10, the Sigmoid activation function is used, the number of output layer neurons is set to 7, and construct a mapping model of texture deviation, auxiliary blowing parameters and mold local temperature adjustment amount containing a BP neural network.

[0042] S312: Normalize the feedback information, map it to the [0,1] interval, input it into the input layer of the trained BP neural network model, calculate the correction coefficient, and output the correction coefficient from the output layer.

[0043] Secondly, a dynamic adjustment system for anti-slip texture of a glass container, applied in the steps of any of the methods described above, the system comprising:

[0044] Acquisition module: acquires glass material temperature, mold local temperature, and auxiliary blowing parameters in real time as feedforward information; acquires texture deviation of the micro-texture quality of the outer surface of the formed container in real time as feedback information;

[0045] Prediction module: Based on the feedforward information, when predicting the flow and cooling state of the glass material on the mold surface, it assists in the initial adjustment of the blowing parameters and the local temperature of the mold.

[0046] Calculation module: Based on the feedback information, corrects the initial adjustment amount, and calculates the target adjustment amount for the auxiliary blowing parameters and the local temperature of the mold;

[0047] Adjustment module: Executes the target adjustment amount to adjust the parameters of the auxiliary blowing system and the local temperature of the mold, thereby controlling the texture formation of the subsequent container.

[0048] Beneficial effects: The method and system for dynamic adjustment of anti-slip texture of glass containers proposed in this application predicts the initial adjustment amount by acquiring feedforward and feedback information in real time, and corrects it according to the feedback information to obtain the target adjustment amount, and then performs the adjustment to cope with dynamic changes in the production process. It has the advantages of improving texture control accuracy, stabilizing texture quality, and being suitable for high-speed production. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating a method for dynamically adjusting the anti-slip texture of a glass container as proposed in this application.

[0050] Figure 2 This is a structural diagram of a dynamic adjustment system for anti-slip texture of a glass container proposed in this application.

[0051] Figure 3 This is a framework diagram of a dynamic adjustment system for anti-slip texture of a glass container proposed in this application.

[0052] Labeling Explanation: 201, Acquisition Module; 202, Prediction Module; 203, Calculation Module; 204, Adjustment Module. Detailed Implementation

[0053] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and marked in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0054] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0055] Please refer to Figure 1 A method for dynamically adjusting the anti-slip texture of a glass container, the method comprising the following steps:

[0056] S1: Real-time acquisition of glass material temperature, mold local temperature, and auxiliary blowing parameters as feedforward information; real-time acquisition of texture deviation of the micro-texture quality of the outer surface of the formed container as feedback information;

[0057] S2: Based on feedforward information, when predicting the flow and cooling state of glass material on the mold surface, the initial adjustment amount of auxiliary blowing parameters and local mold temperature;

[0058] S3: Based on the feedback information, revise the initial adjustment amount and calculate the target adjustment amount for the auxiliary blowing parameters and the local temperature of the mold;

[0059] S4: Execute the target adjustment amount to adjust the auxiliary blowing system parameters and the local temperature of the mold, and control the texture formation of the subsequent container.

[0060] Based on the understanding of this method, step S1 involves acquiring process status information, synchronized with the process in time. This includes acquiring the glass material temperature, the temperature at a specific location in the mold, and the auxiliary blowing parameters, providing process input data that affects texture formation. It also involves acquiring the texture deviation value of the formed container, providing actual output data on texture quality. This step provides the data foundation for the adjustment system. For example, sensors can be used to measure temperature and parameter values, and a vision system can be used to capture images and calculate deviation values.

[0061] Further, step S2 utilizes feedforward information for prediction and initial control. Based on input data such as glass charge temperature, temperature at a specific location in the mold, and auxiliary blowing parameters, the flow and cooling states of the glass charge on the mold surface are predicted. Based on the prediction results, initial calculations of the adjustment amounts for the auxiliary blowing parameters and the temperature at the specific location in the mold are calculated. This step provides process input-based regulation, addressing changes before they occur. For example, the state can be predicted using a model, and the initial adjustment amount can be calculated using a control algorithm.

[0062] Therefore, step S3 uses the feedback information to make changes and calculates the final result. Based on the texture deviation value of the formed container, the initial calculation result of the adjustment amount calculated in step S2 is changed. The changed result is the final calculation result of the adjustment amount for the auxiliary blowing parameter value and the temperature value at a specific location in the mold. This step provides changes based on the actual output deviation, making the adjustment result closer to the target. For example, a correction module or algorithm can be used to modify the initial adjustment amount based on the deviation value.

[0063] Finally, step S4 executes the final calculation of the adjustment amount. This final calculation is used to assist the control system of the blowing system and the temperature control system at specific locations in the mold, changing the system's operating parameter values. This step executes control commands, directly affecting the subsequent texture formation process of the glass container. For example, actuators can be used to change system parameters based on the final adjustment amount.

[0064] In some specific implementations, this method is applied to the production of glass bottles with anti-slip textures.

[0065] In step S1, a pyrometer is used to measure the temperature of the glass material in the flow tank, for example, 1100°C. A thermocouple array is used to measure the temperature at a specific location on the mold, for example, 450°C near the textured area. A flow meter and pressure sensor are used to measure auxiliary blowing parameters, for example, pressure 0.2 MPa, flow rate 10 L / min, and blowing duration 0.5 seconds. A vision inspection system is used to capture an image of the outer surface of the formed glass bottle, and image processing is used to calculate the deviation between the texture features and standard features, for example, a deviation of +10 micrometers.

[0066] In step S2, a predictive model is used to predict the temperature and flow velocity of the glass material in the textured area of ​​the mold based on the glass material temperature, the temperature at a specific location in the mold, and the auxiliary blowing parameters. For example, the predicted glass material temperature is 700°C and the flow velocity is 0.1 m / s. A computational model is then used to calculate the initial adjustment amounts for the auxiliary blowing parameters and the temperature at the specific location in the mold based on the predicted values. For example, the calculation shows that the auxiliary blowing pressure should be reduced by 0.01 MPa and the temperature at the specific location in the mold should be reduced by 5°C.

[0067] In step S3, a correction model is used to calculate the change to the initial calculation results based on the texture deviation value +10 micrometers. For example, the calculation shows that the auxiliary blowing pressure should be reduced by an additional 0.005 MPa, and the temperature at a specific location in the mold should be reduced by an additional 2°C. The initial calculation results are added to the change to obtain the final calculation result of the adjustment amount. For example, the adjustment amount of the auxiliary blowing pressure is -0.015 MPa, and the adjustment amount of the temperature at a specific location in the mold is -7°C.

[0068] In step S4, the control system changes the pressure setpoint of the auxiliary blowing system based on the final calculation result of the adjustment amount, for example, by reducing it by 0.015 MPa. Simultaneously, it changes the setpoint of the mold cooling system, lowering the temperature at a specific location on the mold by 7°C. These changes are applied to the subsequent glass bottle production process, affecting the formation of its texture.

[0069] Furthermore, step S1 includes:

[0070] S11: The glass material temperature is obtained by non-contact measurement of multiple locations in the flow channel using an infrared sensor and weighted averaging.

[0071] S12: By embedding a thermocouple array into the mold surface and fitting the temperature distribution on the mold surface through spatial interpolation, the local temperature of the mold can be obtained.

[0072] S13: Using a flow meter, pressure sensor, and temperature sensor, collect flow, pressure, and temperature data from the auxiliary blowing pipe, record the opening delay time and blowing duration of the auxiliary blowing pipe, and obtain the auxiliary blowing parameters; use the glass material temperature, mold local temperature, and auxiliary blowing parameters as feedforward information.

[0073] S14: Using a vision inspection system, images of the outer surface of the formed glass container are captured. Texture features are extracted through image processing and compared with preset standard texture features to calculate texture deviation. The texture deviation is used as feedback information.

[0074] Specifically, in order to solve the problem of how to accurately, in real time and reliably acquire the feedforward and feedback information required for the dynamic adjustment process, this method provides a specific technical solution.

[0075] Non-contact measurement of high-temperature glass material using infrared sensors avoids interference with the glass material's condition. Multi-point measurement and weighted averaging improve the representativeness and accuracy of temperature data, providing reliable glass material condition input for feedforward control.

[0076] By embedding a thermocouple array on the mold surface and performing spatial interpolation, the local temperature distribution in key areas of the mold can be obtained, which is crucial for understanding and controlling the cooling behavior of the glass material on the mold surface.

[0077] By using multiple sensors and time recording, key parameters of the assisted blowing process were comprehensively collected, including the state of the gas and the blowing sequence. These parameters directly affect the impact of assisted blowing on the local flow and cooling of the glass material, providing detailed blowing condition inputs for feedforward control.

[0078] By objectively measuring the texture of the formed container through a vision inspection system, image processing and contrast calculation provide real-time, quantitative texture quality feedback information, which is the basis for realizing closed-loop control and correcting feedforward prediction.

[0079] By acquiring information from multiple sources in real time, we can ensure that subsequent predictions, calculations and adjustments are based on accurate data, thereby effectively responding to dynamic fluctuations in the production process and achieving stable and precise control of the anti-slip texture of glass containers.

[0080] Furthermore, step S2 includes:

[0081] S21: Construct a predictive model for the flow and cooling state of glass material, and output the predicted values ​​of temperature and flow rate of glass material at multiple key locations on the mold surface;

[0082] S22: Establish a calculation model for the auxiliary blowing parameters and the local temperature adjustment of the mold. Input the predicted values ​​of glass material temperature and flow rate into the calculation model to calculate the preliminary adjustment of the auxiliary blowing parameters and the local temperature of the mold, so that the glass material temperature and flow rate are close to the preset target values.

[0083] Specifically, this method addresses the problem of how to accurately and effectively predict the initial adjustment amount.

[0084] First, the glass material temperature, local mold temperature, and auxiliary blowing parameters are acquired as feedforward information. Then, using this feedforward information, a predictive model for the glass material flow and cooling state is constructed. This model simulates the flow and cooling process of the glass material within the mold cavity, outputting predicted values ​​for the temperature and flow velocity of the glass material at key locations on the mold surface. These predicted values ​​characterize the actual physical state of the glass material during the forming process.

[0085] Next, these predicted temperature and flow rate values ​​are input into a calculation model for auxiliary blowing parameters and mold local temperature adjustments. Based on the predicted state of the glass material, this model calculates the initial adjustments needed to the auxiliary blowing parameters (such as flow rate, pressure, and duration) and the mold local temperature. The goal of these initial adjustments is to bring the temperature and flow rate of the glass material at key locations to a preset ideal state, thereby promoting the formation of the desired anti-slip texture.

[0086] By predicting the state of the glass material first and then calculating the adjustment amount, the determination of the initial adjustment amount is more based on the process mechanism, which improves the accuracy and effectiveness of the adjustment.

[0087] Furthermore, step S21 includes:

[0088] S211: The corners of the mold cavity, areas with large curvature changes, and areas directly affected by the auxiliary blowing airflow are designated as key locations on the surface of the glass material on the mold.

[0089] S212: Establish a finite element model that includes the physical property parameters of the glass material, the structural parameters of the mold, and the auxiliary blowing parameters, and set the initial temperature distribution of the glass material and the temperature boundary conditions of the mold.

[0090] S213: Based on the initial temperature and temperature boundary conditions, simulate the flow and cooling process of glass material in the mold cavity, and output the predicted values ​​of temperature and flow rate of glass material at multiple key locations on the mold surface.

[0091] The key locations are defined by analyzing the geometric model of the mold. Corners and areas with large curvature changes can be identified by calculating the curvature distribution of the inner wall surface of the mold. The areas directly affected by the auxiliary airflow are determined based on the outlet location and blowing direction of the auxiliary air duct in the mold design. These areas significantly affect the flow and cooling behavior of the glass material and are key areas for texture formation.

[0092] Building a finite element model involves discretizing the mold cavity and glass charge into a finite number of elements. The physical properties of the glass charge, such as viscosity, density, thermal conductivity, and specific heat capacity, are input into the model; these parameters are typically functions of temperature.

[0093] Mold structure parameters include the thermophysical properties of the mold material and the geometric dimensions of the mold. Auxiliary blowing parameters, such as gas flow rate, temperature, pressure, blowing duration, and effective area, are used as inputs or boundary conditions for the model.

[0094] The initial temperature distribution of the glass charge can be obtained by measuring the temperature at multiple locations within the flow channel and performing spatial interpolation.

[0095] The temperature boundary conditions of a mold are usually set as the convective heat transfer conditions between the outer surface of the mold and the cooling medium (such as cooling water or air). The heat transfer coefficient is determined according to the design and operating parameters of the cooling system.

[0096] The simulation process is achieved by solving a set of coupled partial differential equations describing glass flow (e.g., the Navier-Stokes equations) and heat transfer (e.g., the energy conservation equations). The solver uses an iterative method to calculate the temperature and velocity fields of the glass during the filling and cooling of the mold cavity. After the simulation, the temperature and flow velocity values ​​at predetermined key locations and specific time points (e.g., when the glass completely adheres to the mold wall or when blowing ends) are extracted as prediction results.

[0097] Thus, through physical modeling and numerical simulation, the state information of the glass material in the key areas was obtained.

[0098] Specifically, to address the combined impact of multiple dynamic fluctuations, such as glass material temperature / viscosity, mold temperature, and blowing parameters, the auxiliary blowing parameters and local mold temperature are adjusted in real time and in a coordinated manner to ensure that the texture quality remains stable and meets requirements under high-speed production cycles, while also adapting to the rapid switching between different product models.

[0099] This solution addresses the issue of insufficient accuracy in prediction models by constructing a predictive model of glass flow and cooling state. This model enables a more accurate understanding of the dynamic behavior of the glass during the forming process, thereby supporting more precise and effective texture control and adjustment, and improving the stability and consistency of texture quality.

[0100] Furthermore, step S212 includes:

[0101] S2121: Obtain the physical property parameters of the glass material based on its composition; obtain the mold structure parameters based on the mold design drawings; and consult the technical parameters of the auxiliary blowing equipment to obtain the auxiliary blowing parameters.

[0102] S2122: Construct a finite element model based on the physical properties of the glass material, the structural parameters of the mold, and the auxiliary blowing parameters;

[0103] S2123: Based on the temperature distribution measurement results of the glass material in the flow channel, the initial temperature distribution of the glass material is fitted using an interpolation method as the initial condition of the finite element model.

[0104] S2124: Based on the design parameters of the mold cooling system, set the heat transfer coefficient between the mold and the cooling water as the temperature boundary condition of the finite element model.

[0105] The physical properties of the glass frit are obtained by analyzing its chemical composition and consulting relevant material databases or literature, such as the viscosity of the glass as a function of temperature, specific heat capacity, and thermal conductivity. The mold structure parameters are obtained by reading the computer-aided design drawings of the mold, extracting information such as the geometry, dimensions, air passage location, and cooling water channel layout of the mold cavity. The auxiliary blowing parameters are obtained by consulting the technical manual of the auxiliary blowing system to understand its performance indicators such as maximum flow rate, pressure range, temperature range, and control response time.

[0106] Constructing a finite element model involves selecting appropriate element types and meshing strategies to accurately discretize the computational domains of the glass bloc, mold, and gas. Fitting the initial temperature distribution of the glass bloc involves arranging multiple temperature sensors in the flow channel to measure the glass bloc temperature at different locations, and then using mathematical interpolation algorithms, such as linear interpolation or spline interpolation, to construct a continuous temperature field as the starting state for the simulation. The heat transfer coefficient between the mold and cooling water is determined based on the geometry of the mold's cooling water channels, the flow rate and temperature of the cooling water, and the thermophysical properties of the mold material and the cooling water. The convective heat transfer coefficient can be determined using empirical formulas or numerical calculation methods.

[0107] Specifically, this solution addresses the problem that existing technologies lack detailed methods for obtaining the parameters required for the finite element model and setting initial / boundary conditions when constructing predictive models for glass flow and cooling states, resulting in insufficient model accuracy.

[0108] First, by explicitly specifying the basic data required for constructing the finite element model from the glass composition, mold design drawings, and auxiliary blowing equipment technical parameters, the accuracy of the input parameters was ensured. Next, these accurate parameters were used to construct the finite element model, laying the foundation for subsequent numerical simulations. Furthermore, by actually measuring the temperature distribution of the glass in the flow channel and interpolating the data, initial temperature conditions close to actual production conditions were set.

[0109] Therefore, by setting the heat transfer boundary conditions between the mold and cooling water according to the design parameters of the mold cooling system, the influence of mold cooling on the glass forming process is accurately reflected. Through the above steps, the constructed finite element model can more accurately simulate the flow and cooling process of glass material in the mold cavity, outputting reliable temperature and flow velocity prediction values. This improves the accuracy of subsequent calculations of auxiliary blowing parameters and preliminary adjustments to the local temperature of the mold, and helps to achieve precise control of the anti-slip texture of glass containers.

[0110] Furthermore, step S213 includes:

[0111] S2131: Import the finite element model into the solver and set the convergence criteria for the solver;

[0112] S2132: Based on the initial temperature distribution of the glass material and the temperature boundary conditions of the mold, iteratively solve the temperature field and velocity field distribution of the glass material in the mold cavity until the energy residual in the iteration process reaches the convergence criterion of the solver.

[0113] S2133: After the convergence criteria are met, record the stable time point at which each key position reaches a stable state;

[0114] S2134: Based on the stable time point, select the corresponding temperature and flow velocity data in the temperature and velocity field distribution of the glass material in the mold cavity as the predicted values ​​of temperature and flow velocity at each key location.

[0115] Specifically, in order to solve the problem of accurately extracting the predicted values ​​of key locations from the simulation results of glass flow and cooling, the finite element model containing the physical properties of the glass, the mold structure parameters, and the auxiliary blowing parameters is first imported into professional numerical solution software, and the convergence criteria for the solution calculation are set to ensure the accuracy of the simulation results.

[0116] Next, based on the initial temperature distribution of the glass material upon entering the mold cavity and the temperature boundary conditions set according to the heat exchange conditions between the mold and the cooling system, the software begins iterative calculations of the temperature and flow velocity distribution of the glass material within the mold cavity over time. This iterative process continues until the calculation results meet preset convergence criteria, such as an energy conservation error less than a set value. After the simulation is complete, a series of glass material temperature and velocity field data at different time points are obtained.

[0117] To determine the state of the glass charge at critical forming moments, this scheme introduces a step to identify the "steady-state time point". After the simulation reaches convergence, the system analyzes the temperature and velocity changes over time at each critical location (e.g., mold corners, air blowing area) to determine the time point when these physical quantities tend to stabilize.

[0118] Finally, based on the stable time points of each identified key location, the temperature and flow rate values ​​corresponding to these specific time points are precisely extracted from the complete time series data obtained from the simulation calculations. These extracted values ​​are then determined as the predicted temperature and flow rate values ​​of the glass frit at that key location, reflecting the state of the glass frit when the texture is formed or when it reaches its final shape. This improves the accuracy of the prediction and provides a reliable basis for subsequent parameter adjustments.

[0119] In some specific implementations, ANSYS Fluent is used as the solver, and the energy residual convergence criterion is set to 1e-6. After the simulation is completed, the temperature change curve over time is analyzed for a key location on the mold shoulder. If the temperature change rate at this location is less than 0.1℃ / ms for 10 consecutive time steps at a simulation time of 0.5 seconds, then 0.5 seconds is recorded as the time point when the location reaches a steady state. Then, the temperature value (e.g., 750℃) and flow velocity value (e.g., 0.05m / s) at this location are extracted from the simulation results at 0.5 seconds as the predicted temperature and flow velocity values ​​for this key location. Further, step S22 includes:

[0120] S221: Establish a calculation model for auxiliary blowing parameters and mold local temperature adjustment, including a PID control module and a fuzzy logic control module;

[0121] S222: Input the predicted values ​​of glass material temperature and flow rate into the fuzzy logic control module to calculate the adjustment range of PID control parameters for auxiliary blowing parameters and local temperature of the mold;

[0122] S223: Input the PID control parameter adjustment range into the PID control module to calculate the initial adjustment amount of the auxiliary blowing parameters and the local temperature of the mold, so as to make the glass material temperature and flow rate close to the preset target value.

[0123] Specifically, this method determines the initial adjustment amount by constructing a computational model that combines fuzzy logic and PID control. First, a glass flow and cooling state prediction model outputs predicted values ​​for the temperature and flow velocity of the glass at key locations on the mold surface. These predicted values ​​are input into the fuzzy logic control module within the computational model. Based on preset fuzzy rules, the fuzzy logic module transforms these predicted values ​​into adjustment ranges for the PID control module parameters (such as proportional gain P, integral gain I, and derivative gain D). For example, if the predicted temperature is much higher than the target and the flow velocity is much lower than the target, the fuzzy logic might calculate a significant increase in the P gain range for blowing pressure control and a decrease in the I gain range for mold temperature control.

[0124] Then, the adjustment ranges of these PID parameters calculated by fuzzy logic are passed to the PID control module. The PID control module uses the predicted glass charge temperature and flow rate as process variables, the preset target temperature and flow rate as setpoints, and the parameter adjustment ranges provided by fuzzy logic to calculate preliminary adjustments to auxiliary blowing parameters (such as blowing pressure and duration) and the local temperature of the mold. These preliminary adjustments aim to reduce the deviation between the predicted and target values, making the state of the glass charge on the mold surface closer to ideal conditions, thus providing a favorable foundation for subsequent texture formation.

[0125] In some specific implementations, the calculation model for auxiliary blowing parameters and local mold temperature adjustment can be implemented as follows: The inputs to the fuzzy logic control module can be the predicted temperature deviation (predicted temperature minus target temperature) and the predicted speed deviation (predicted speed minus target speed). Fuzzy sets are defined for these inputs, such as "negative large", "negative small", "zero", "positive small", and "positive large". Output fuzzy sets are defined for the adjustment ranges of the PID control parameters (e.g., the P, I, and D gains of the blowing pressure controller and the P, I, and D gains of the mold temperature controller), such as "significantly reduced", "reduced", "unchanged", "increased", and "significantly increased". A fuzzy rule base is established, for example, "If the predicted temperature deviation is 'positive large' and the predicted speed deviation is 'negative large', then the blowing pressure P gain adjustment range is 'significantly increased', and the mold temperature I gain adjustment range is 'significantly reduced'". Through fuzzy inference and defuzzification processes, specific PID parameter adjustment range values ​​are obtained. The PID control module receives these adjustment ranges and calculates the initial adjustment amount based on the predicted deviation. For example, the initial adjustment amount of the blowing pressure = Kp * predicted temperature deviation + Ki * ∫(predicted temperature deviation) dt + Kd * d(predicted temperature deviation) / dt, where Kp, Ki, and Kd are the PID gains determined based on the adjustment range of the fuzzy logic output. The initial adjustment amount of the local temperature of the mold is calculated similarly. Therefore, the calculated initial adjustment amount is used to guide subsequent control execution.

[0126] Furthermore, step S3 includes:

[0127] S31: Construct a mapping model between texture deviation and auxiliary blowing parameters and mold local temperature adjustment amount, input feedback information into the mapping model, and obtain the correction coefficients of auxiliary blowing parameters and mold local temperature;

[0128] S32: Multiply the initial adjustment amount by the correction coefficient to obtain the target adjustment amount for the auxiliary blowing parameters and the local temperature of the mold.

[0129] Specifically, to address the combined impact of multiple dynamic fluctuations, such as glass material temperature / viscosity, mold temperature, and blowing parameters, this method ensures that the texture quality remains stable and meets requirements under high-speed production cycles by adjusting auxiliary blowing parameters and local mold temperature in real time and collaboratively. This also allows for rapid switching between different product models. The method constructs a mapping model between texture deviation and the adjustment amounts of auxiliary blowing parameters and local mold temperature, and inputs the texture deviation of the micro-texture quality of the outer surface of the formed container as feedback information into the mapping model.

[0130] The mapping model calculates correction coefficients for the auxiliary blowing parameters and the local mold temperature based on the input texture deviation. Then, the initial adjustments to the auxiliary blowing parameters and local mold temperature, predicted based on feedforward information, are multiplied by these correction coefficients. The result of this multiplication is the target adjustment for the auxiliary blowing parameters and local mold temperature. Executing this target adjustment adjusts the auxiliary blowing system parameters and the local mold temperature, controlling the texture formation of the subsequent container. This process quantifies the actual texture quality deviation as a correction to the control parameter adjustments, improving the accuracy of feedback control and thus stabilizing texture quality.

[0131] Furthermore, step S31 includes:

[0132] S311: Set the input layer, the number of hidden layer neurons is set to 10, the Sigmoid activation function is used, the number of output layer neurons is set to 7, and construct a mapping model of texture deviation, auxiliary blowing parameters and mold local temperature adjustment amount containing a BP neural network.

[0133] S312: Normalize the feedback information, map it to the [0,1] interval, input it into the input layer of the trained BP neural network model, calculate the correction coefficient, and output the correction coefficient from the output layer.

[0134] This technical solution introduces a BP neural network to construct a mapping model between texture deviation and auxiliary blowing parameters and mold local temperature adjustment, thus solving the problem that traditional mapping methods cannot accurately handle complex nonlinear relationships.

[0135] The input layer receives texture deviation feedback information. The hidden layer has 10 neurons and uses the sigmoid activation function, enabling the neural network to handle nonlinear relationships and learn and approximate the complex nonlinear mapping relationship between texture deviation and correction coefficients. The output layer has 7 neurons, corresponding to the number of correction coefficients for the auxiliary blowing parameters and local mold temperature that need to be calculated, directly outputting the correction results. A mapping model incorporating a BP neural network is constructed, utilizing its learning ability to learn the correspondence between texture deviation and the required correction coefficients through training data. The feedback information is normalized and mapped to the [0,1] interval to ensure that the input data is within the applicable range of the neural network, improving the network's training efficiency and computational accuracy. The normalized feedback information is input into the trained BP neural network model, and the network calculates the corresponding correction coefficients based on the learned mapping relationship. The calculated correction coefficients are output from the output layer; these coefficients are used to correct the initial adjustment amount.

[0136] Please refer to Figure 2 , Figure 3 A dynamic adjustment system for anti-slip texture of glass containers, applied in the steps of any of the above methods, the system comprising:

[0137] Acquisition module 201: acquires glass material temperature, mold local temperature, and auxiliary blowing parameters in real time as feedforward information; acquires texture deviation of the micro-texture quality of the outer surface of the formed container in real time as feedback information;

[0138] Prediction module 202: Based on feedforward information, when predicting the flow and cooling state of glass material on the mold surface, it assists in the initial adjustment of blowing parameters and local mold temperature.

[0139] Calculation module 203: Based on the feedback information, corrects the initial adjustment amount and calculates the target adjustment amount for the auxiliary blowing parameters and the local temperature of the mold;

[0140] Adjustment module 204: Executes the target adjustment amount to adjust the parameters of the auxiliary blowing system and the local temperature of the mold, thereby controlling the texture formation of the subsequent container.

[0141] The acquisition module 201 is responsible for collecting key data in the production process in real time, including glass material temperature, local mold temperature, and auxiliary blowing parameters as feedforward information, and texture deviation of the formed container as feedback information. This real-time data forms the basis for subsequent prediction, calculation, and adjustment, thereby ensuring that the system can perceive the current production status and product quality. The technical contribution of the acquisition module 201 lies in providing comprehensive and real-time process status and result quality data input.

[0142] The prediction module 202 receives feedforward information from the acquisition module and predicts the flow and cooling state of the glass material on the mold surface based on this information. It then calculates the initial adjustments to the auxiliary blowing parameters and the local mold temperature. This module uses a prediction model and feedforward information to anticipate potential texture problems and provide adjustment suggestions in advance, thus demonstrating the system's predictive control capability. The technical contribution of the prediction module 202 lies in using feedforward information for prediction and preliminary decision-making, thereby improving the system's response speed.

[0143] The calculation module 203 receives the preliminary adjustment amount output by the prediction module and the feedback information (texture deviation) provided by the acquisition module. It corrects the preliminary adjustment amount based on the texture deviation, calculating the final target adjustment amounts for the auxiliary blowing parameters and the local temperature of the mold. This module combines feedforward prediction and feedback correction, thereby ensuring that the adjustment amount considers both the prediction of the process state and the correction of the impact of actual texture deviation, improving the accuracy of the adjustment.

[0144] The technical contribution of the calculation module 203 lies in revising the initial adjustment by incorporating feedback information to arrive at a more precise target adjustment amount. The adjustment module 204 receives the target adjustment amount output by the calculation module 203 and executes these adjustment instructions to actually change the parameters of the auxiliary blowing system and the local temperature of the mold.

[0145] By executing these precisely calculated adjustments, the system can intervene in the glass container forming process in real time, controlling the subsequent texture formation of the container to ensure it meets quality requirements. The technical contribution of the adjustment module 204 lies in translating the calculation results into actual control actions, directly impacting the production process and product quality.

[0146] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0147] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for dynamically adjusting the anti-slip texture of a glass container, characterized in that, The method includes the following steps: S1: Real-time acquisition of glass material temperature, mold local temperature, and auxiliary blowing parameters as feedforward information; real-time acquisition of texture deviation of the micro-texture quality of the outer surface of the formed container as feedback information; S2: Based on the feedforward information, when predicting the flow and cooling state of the glass material on the mold surface, the initial adjustment amount of the auxiliary blowing parameters and the local temperature of the mold is made. S3: Based on the feedback information, correct the initial adjustment amount, and calculate the target adjustment amount for the auxiliary blowing parameters and the local temperature of the mold; S4: Execute the target adjustment amount to adjust the parameters of the auxiliary blowing system and the local temperature of the mold, and control the texture formation of the subsequent container.

2. The method for dynamically adjusting the anti-slip texture of a glass container according to claim 1, characterized in that, Step S1 includes: S11: The temperature of the glass material at multiple locations within the flow channel is measured non-contactly using an infrared sensor, and the weighted average is used to obtain the glass material temperature. S12: By embedding a thermocouple array into the mold surface and fitting the temperature distribution on the mold surface through spatial interpolation, the local temperature of the mold is obtained; S13: Using a flow meter, pressure sensor, and temperature sensor, collect flow rate, pressure, and temperature data from the auxiliary blowing pipe, record the opening delay time and blowing duration of the auxiliary blowing pipe, and obtain the auxiliary blowing parameters; use the glass material temperature, the local temperature of the mold, and the auxiliary blowing parameters as the feedforward information. S14: Using a visual inspection system, an image of the outer surface of the formed glass container is captured. Texture features are extracted through image processing and compared with preset standard texture features to calculate the texture deviation. The texture deviation is used as the feedback information.

3. The method for dynamically adjusting the anti-slip texture of a glass container according to claim 1, characterized in that, Step S2 includes: S21: Construct a predictive model for the flow and cooling state of glass material, and output the predicted values ​​of temperature and flow rate of glass material at multiple key locations on the mold surface; S22: Establish a calculation model for the auxiliary blowing parameters and the local temperature adjustment of the mold. Input the predicted values ​​of glass material temperature and flow rate into the calculation model to calculate the preliminary adjustment of the auxiliary blowing parameters and the local temperature of the mold, so that the glass material temperature and flow rate are close to the preset target values.

4. The method for dynamically adjusting the anti-slip texture of a glass container according to claim 3, characterized in that, Step S21 includes: S211: The corners of the mold cavity, areas with large curvature changes, and areas directly affected by the auxiliary blowing airflow are designated as key locations on the surface of the glass material on the mold. S212: Establish a finite element model that includes the physical property parameters of the glass material, the structural parameters of the mold, and the auxiliary blowing parameters, and set the initial temperature distribution of the glass material and the temperature boundary conditions of the mold. S213: Based on the initial temperature and the temperature boundary conditions, simulate the flow and cooling process of the glass material in the mold cavity, and output the predicted values ​​of the temperature and flow rate of the glass material at multiple key locations on the mold surface.

5. The method for dynamically adjusting the anti-slip texture of a glass container according to claim 4, characterized in that, Step S212 includes: S2121: Obtain the physical property parameters of the glass material based on its composition; obtain the mold structure parameters based on the mold design drawings; consult the technical parameters of the auxiliary blowing equipment, specifically the auxiliary blowing parameters. S2122: Construct a finite element model based on the physical property parameters of the glass material, the structural parameters of the mold, and the auxiliary blowing parameters; S2123: Based on the temperature distribution measurement results of the glass material in the flow channel, the initial temperature distribution of the glass material is fitted using an interpolation method as the initial condition of the finite element model. S2124: Based on the design parameters of the mold cooling system, set the heat transfer coefficient between the mold and the cooling water as the temperature boundary condition of the finite element model.

6. The method for dynamically adjusting the anti-slip texture of a glass container according to claim 4, characterized in that, Step S213 includes: S2131: Import the finite element model into the solver and set the convergence criteria of the solver; S2132: Based on the initial temperature distribution of the glass material and the temperature boundary conditions of the mold, iteratively solve the temperature field and velocity field distribution of the glass material in the mold cavity until the energy residual in the iteration process reaches the convergence criterion of the solver. S2133: After the convergence criterion is met, record the stable time point when each key position reaches a stable state; S2134: Based on the stable time point, select the corresponding temperature and flow velocity data in the temperature field and velocity field distribution of the glass material in the mold cavity as the predicted values ​​of the temperature and flow velocity at each key location.

7. The method for dynamically adjusting the anti-slip texture of a glass container according to claim 3, characterized in that, Step S22 includes: S221: Establish a calculation model for auxiliary blowing parameters and local temperature adjustment of the mold, including a PID control module and a fuzzy logic control module; S222: Input the predicted values ​​of glass material temperature and flow rate into the fuzzy logic control module to calculate the adjustment range of the auxiliary blowing parameters and the PID control parameters of the local temperature of the mold; S223: Input the adjustment range of the PID control parameters into the PID control module to calculate the initial adjustment amount of the auxiliary blowing parameters and the local temperature of the mold, so as to make the glass material temperature and flow rate close to the preset target value.

8. The method for dynamically adjusting the anti-slip texture of a glass container according to claim 1, characterized in that, Step S3 includes: S31: Construct a mapping model between texture deviation and auxiliary blowing parameters and mold local temperature adjustment amount, input the feedback information into the mapping model, and obtain the correction coefficients of the auxiliary blowing parameters and the mold local temperature; S32: Multiply the initial adjustment amount by the correction coefficient to obtain the target adjustment amount of the auxiliary blowing parameters and the local temperature of the mold.

9. The method for dynamically adjusting the anti-slip texture of a glass container according to claim 8, characterized in that, Step S31 includes: S311: Set the input layer, the number of hidden layer neurons is set to 10, the Sigmoid activation function is used, the number of output layer neurons is set to 7, and construct a mapping model of texture deviation, auxiliary blowing parameters and mold local temperature adjustment amount containing a BP neural network. S312: Normalize the feedback information, map it to the [0,1] interval, input it into the input layer of the trained BP neural network model, calculate the correction coefficient, and output the correction coefficient from the output layer.

10. A dynamic adjustment system for anti-slip texture of glass containers, characterized in that, The system, applied in the steps of the method according to any one of claims 1-9, comprises: Acquisition module: acquires glass material temperature, mold local temperature, and auxiliary blowing parameters in real time as feedforward information; acquires texture deviation of the micro-texture quality of the outer surface of the formed container in real time as feedback information; Prediction module: Based on the feedforward information, when predicting the flow and cooling state of the glass material on the mold surface, it assists in the initial adjustment of the blowing parameters and the local temperature of the mold. Calculation module: Based on the feedback information, corrects the initial adjustment amount, and calculates the target adjustment amount for the auxiliary blowing parameters and the local temperature of the mold; Adjustment module: Executes the target adjustment amount to adjust the parameters of the auxiliary blowing system and the local temperature of the mold, thereby controlling the texture formation of the subsequent container.

Images