A glass fiber mesh dipped composite control system for wafer cutting grinding wheels

Abstract

The invention relates to the technical field of glass fiber mesh processing, and specifically to a glass fiber mesh dipping composite control system for a wafer cutting grinding wheel, comprising: a parameter setting module, used for obtaining initial parameters of the glass fiber mesh and performing system calibration; the initial parameters include a coating pressure of a laminating rubber roller, a glue liquid level in a glue tank, a target glue layer thickness, a material moving speed and a drying temperature; a coating control module, used for obtaining the glass fiber mesh under corresponding coating pressure and material moving speed, taking the glue liquid level and the target glue layer thickness as constraint conditions, analyzing the current glue dipping amount, and obtaining a glue dipping amount parameter set; a glue feedback module, used for analyzing the glue dipping amount parameter set, determining a glue liquid level change rate corresponding to the glue liquid when the glue dipping amount takes different values, taking the glue liquid level change rate as a feedback condition, and determining a glue liquid feedback control parameter; the production efficiency and the stability of the system are improved.

Description

Technical Field

[0001] The invention relates to the technical field of glass fiber mesh processing, in particular to a glass fiber mesh dipping composite control system for a wafer cutting grinding wheel. Background Art

[0002] As a common composite material, glass fiber mesh is widely used in aerospace, construction, automobile, electronics and other fields. Dipping is an important link in the manufacturing process of glass fiber cloth, which has a decisive influence on the performance and quality of the product. Traditional dipping composite control systems usually rely on fixed parameter settings and manual experience, resulting in the following problems in the production process: Inflexible parameter adjustment: The parameter settings in traditional systems are usually fixed and cannot be dynamically adjusted according to actual production conditions, resulting in unstable product quality.

[0003] Uneven coating: Due to the lack of precise control over the flow rate of the glue and the thickness of the glue layer, uneven coating is prone to occur, affecting the performance of the final product.

[0004] Poor control of drying temperature: The control of drying temperature directly affects the curing effect of the adhesive layer. Traditional methods are difficult to achieve precise control, which may lead to poor curing or over-curing of the adhesive layer.

[0005] Insufficient feedback mechanism: The existing system lacks an effective feedback mechanism and cannot adjust the rate of change of the glue liquid level in time, resulting in instability and waste in the production process.

[0006] In order to solve the above problems, the present invention proposes a glass fiber mesh dipping and compounding control system for wafer cutting grinding wheels, which realizes a more efficient and stable dipping and compounding process by introducing dynamic parameter adjustment, optimization model and feedback control mechanism. Summary of the invention

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is: a glass fiber mesh dipping composite control system for wafer cutting grinding wheels, including: a parameter setting module, used to obtain the initial parameters of the glass fiber mesh and perform system calibration; the initial parameters include the coating pressure of the laminating rubber roller, the glue liquid level in the glue tank, the target glue layer thickness, the material movement speed and the drying temperature.

[0008] The coating control module is used to obtain the glass fiber mesh under the corresponding coating pressure and material moving speed, taking the glue liquid level and the target glue layer thickness as constraints, analyzing the current glue dipping amount of the glue, and obtaining the glue dipping amount parameter set.

[0009] The glue liquid feedback module is used to analyze the glue dipping amount parameter set, determine the glue liquid level change rate corresponding to different values ​​of the glue dipping amount, use the glue liquid level change rate as a feedback condition, and determine the glue liquid feedback control parameters.

[0010] The drying control module is used to optimize and analyze the glue liquid feedback control parameters to obtain the target control parameters, and determine the probability of change of the drying temperature based on the target control parameters. According to the probability of change of the drying temperature, the temperature boundary conditions corresponding to the glue dipping amount are obtained, and the glue dipping amount control strategy is determined based on the temperature boundary conditions and the target control parameters.

[0011] The beneficial effects of the present invention are: 1. The present invention improves the flexibility of parameter setting through system calibration and dynamic parameter adjustment.

[0012] Second, the present invention realizes precise control of the flow rate of glue liquid and the thickness of glue layer by optimizing the model and feedback control, thus ensuring the uniformity of coating.

[0013] 3. The present invention realizes precise control of the drying temperature by optimizing the variation probability of the drying temperature and adjusting the temperature boundary conditions.

[0014] Fourth, the present invention introduces glue feedback control parameters to timely adjust the glue level change rate, thereby improving the stability and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0016] Figure 1 The present invention is a system framework diagram of a glass fiber mesh dipped composite control system for wafer cutting grinding wheels.

[0017] Figure 2 The present invention is a system schematic diagram of a glass fiber mesh dipped composite control system for a wafer cutting grinding wheel.

[0018] Figure 3 The present invention is a system schematic diagram for system calibration in a parameter setting module of a glass fiber mesh dipped composite control system for a wafer cutting grinding wheel.

[0019] Figure 4 The present invention is a system schematic diagram of a set of dipping amount parameters in a coating control module of a glass fiber mesh dipping composite control system for wafer cutting grinding wheels.

[0020] Figure 5 The invention discloses a system schematic diagram of glue feedback control parameters in a glue feedback module of a glass fiber mesh glue dipping composite control system for wafer cutting grinding wheels. DETAILED DESCRIPTION

[0021] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. If no specific techniques or conditions are specified in the embodiments, the techniques or conditions described in the literature in the art or the product specifications are used.

[0022] See also Figure 1 , Figure 2 A composite control system for glass fiber mesh dipping for wafer cutting grinding wheels, comprising: a parameter setting module, a coating control module, a glue feedback module and a drying control module; the parameter setting module transmits the selected parameters to the coating control module, and the coating control module determines the selected dipping amount at this time according to the parameters, and transmits the output to the glue feedback module; the glue feedback module determines the liquid level change rate of the glue at this time, and adjusts it, and transmits the adjusted data to the drying control module; the drying control module selects the temperature boundary conditions of the drying temperature according to the previously obtained data, and completes the drying process of the glass fiber mesh.

[0023] The parameter setting module is used to obtain the initial parameters of the glass fiber mesh and perform system calibration; the initial parameters include the coating pressure of the laminating rubber roller, the glue liquid level in the glue tank, the target glue layer thickness, the material moving speed and the drying temperature.

[0024] The coating control module is used to obtain the glass fiber mesh under the corresponding coating pressure and material moving speed, taking the glue liquid level and the target glue layer thickness as constraints, analyzing the current glue dipping amount of the glue, and obtaining the glue dipping amount parameter set.

[0025] The glue liquid feedback module is used to analyze the glue dipping amount parameter set, determine the glue liquid level change rate corresponding to different values ​​of the glue dipping amount, use the glue liquid level change rate as a feedback condition, and determine the glue liquid feedback control parameters.

[0026] The drying control module is used to optimize and analyze the glue liquid feedback control parameters to obtain the target control parameters, and determine the probability of change of the drying temperature based on the target control parameters. According to the probability of change of the drying temperature, the temperature boundary conditions corresponding to the glue dipping amount are obtained, and the glue dipping amount control strategy is determined based on the temperature boundary conditions and the target control parameters.

[0027] The above-mentioned coating pressure indicates the pressure used when coating the glass mesh with glue, so as to determine whether the currently selected glue can be accurately coated on the glass fiber mesh. When coating, a part of the laminating rubber roller is located in the glue groove, and the glue is brought out by rotation and coated on the glass fiber mesh on the side of the laminating rubber roller; the glue liquid level indicates the liquid level height of the stored glue, and this liquid level can also indicate the current corresponding amount of glue, and record the current remaining amount of glue in real time; the target glue layer thickness is used to indicate the coating amount required for the current glass fiber mesh. This target glue layer thickness will select different thicknesses according to the material of the glue used for the glass fiber mesh. At this time, it is necessary to select the target glue layer thickness according to the type of current glue and the design requirements of the glass fiber mesh; the material movement speed indicates the movement speed of the glass fiber mesh during the dipping treatment, and the drying temperature indicates the heating temperature when the glass fiber mesh is heated. This temperature affects whether the current treatment can be completed after the final coating of the glue.

[0028] Piezoelectric sensors, strain gauge sensors or pressure transmitters are usually used. The sensors are installed at key positions in contact with the material to monitor the actual applied pressure. Photoelectric encoders are installed on or near the shaft of the motor to monitor the material movement speed. Ultrasonic level gauges, radar level gauges, guided wave radar level gauges, capacitive level gauges or float level gauges can be installed in the glue tank to monitor the glue level. Thermocouples, thermal resistors (such as PT100) or infrared temperature sensors are usually installed on heating rollers or other locations where temperature monitoring is required to determine the drying temperature at this time and to obtain temperature changes during dipping and drying. Infrared thickness gauges or laser thickness gauges are usually installed above the coated material to monitor the thickness of the glue layer.

[0029] like Figure 3 As shown, the method of performing system calibration in the parameter setting module is: obtaining the classification index of the current glass fiber mesh, classifying the glass fiber mesh once according to the classification index of the current glass fiber mesh, and obtaining a classification result.

[0030] Obtain the historical database of the current glass fiber mesh, and perform secondary classification on the historical database according to the primary classification result to obtain a secondary classification result; the secondary classification result is realized by calculating the classification feature set of the primary classification result and the historical database, and obtaining the secondary classification result based on the similarity of the classification feature set.

[0031] The initial parameters are adjusted according to the secondary classification results to complete the system calibration of the initial parameters.

[0032] When classifying the above-mentioned primary classification results, it is first necessary to determine the classification index of the current glass fiber mesh. The classification index at this time can be expressed as material type, thickness, density, surface treatment, fiber diameter, and weaving method; classify the current glass fiber mesh according to the current classification index, determine the initial parameters required for the current glass fiber mesh, and the values ​​required to achieve the required material type, thickness, and density, and use the content of this part of the initial parameters after classification according to the classification index as the primary classification result.

[0033] Material type: such as E glass, S glass, etc.

[0034] Thickness: The thickness of the mesh (unit: mm).

[0035] Density: The density of the mesh (unit: g / cm³).

[0036] Surface treatment: whether it has undergone special treatment, such as coating or impregnation.

[0037] Fiber diameter: The diameter of the fiber (unit: μm).

[0038] Weaving method: plain, twill or other weaving methods.

[0039] The purpose of primary and secondary classification at this time is to determine whether the current products meet the requirements after using the initial parameters for production. At this time, it is necessary to adjust the set parameters and control and detect subsequent parameter changes to obtain a complete data control method.

[0040] The method of calculating the classification result of the first classification and the classification feature set of the historical database, and obtaining the secondary classification result based on the similarity retrieval of the classification feature set is as follows: at this time, the value of the classification index in the historical database is compared with the median value of the first classification result. At this time, the first classification result and the corresponding data in the historical data are converted into vectors, and then combined into a classification feature set. At this time, the cosine similarity is calculated according to the data of the first classification result in the classification feature set and the data in the historical database to realize the similarity retrieval of the classification feature set, and the largest group of data calculated by the cosine similarity is used as the secondary classification result output at this time, and the calibrated initial parameters at this time are obtained according to the values ​​of the corresponding initial parameters in the secondary classification result; during the secondary classification processing, the data in the classification feature set corresponds to the classification index of the first classification one by one. At this time, the relevant data are uniformly classified and managed according to these classification indicators during classification.

[0041] In the coating control module, the glue dipping amount parameter set is obtained mainly based on the currently set glue liquid level and target glue layer thickness as parameters to calculate the glue dipping amount used at this time. The glue dipping amount at this time will be expressed as the calculation of the glue liquid flow rate and the glue layer thickness used at this time for analysis as the glue dipping amount parameter set selected at this time.

[0042] like Figure 4 As shown, the implementation method of the dipping amount parameter set is shown below.

[0043] Obtain a glue flow model and a glue layer thickness model corresponding to the glue level and the target glue layer thickness, divide the glue flow model and the glue layer thickness model according to the constraints, and obtain the glue flow output by the glue flow model and the glue layer thickness output by the glue layer thickness model respectively.

[0044] By evaluating the glue flow rate and glue layer thickness according to the parameter memory library, the efficiency gain value and speed gain value of the glue coating amount are calculated respectively; according to the efficiency gain value and the speed gain value, multiple optimization parameters are obtained as the output glue dipping amount parameter set.

[0045] The data used in the glue flow model and the glue layer thickness model will be standardized before calculation to prevent the problem of inconsistent dimensions during calculation.

[0046] For the glue flow model, the coating pressure, material moving speed, glue level, diameter of the laminating rubber roller, depth of the laminating rubber roller immersed in the glue tank, and viscosity of the glue are obtained to obtain the glue flow. The glue flow is expressed as shown below.

[0047] ;in, Indicates the glue flow rate, It represents the empirical constant, which is obtained based on the fitting of the change of the glue flow rate. In this case, the empirical constant represents the average ratio of the residual change rate of the glue flow rate in the historical data; Indicates coating pressure, Indicates the material moving speed, Indicates the glue liquid level. Indicates the diameter of the laminating roller. Indicates the depth of the laminating rubber roller immersed in the rubber tank. Indicates the viscosity of the glue. It represents the influence function of the diameter of the laminating rubber roller, the depth of the laminating rubber roller immersed in the glue tank, and the viscosity of the glue on the glue flow rate.

[0048] The depth of the laminating rubber roller immersed in the rubber tank can be achieved in the following ways.

[0049] At this time, the depth of the glue tank is determined by the diameter of the laminating rubber roller and the glue level. At the same time, the selected glue level represents the height of a liquid level, not the flow rate of the current liquid level.

[0050] The influence function of the diameter of the laminating rubber roller, the depth of the laminating rubber roller immersed in the rubber tank, and the viscosity of the glue on the glue flow rate can be expressed in the following form.

[0051] ;in, , , It represents the adjustment coefficient. In the present invention, the values ​​set by the adjustment coefficient are respectively expressed as 0.8, 0.2, and -0.1.

[0052] The above-mentioned average ratio of the change rate of the glue liquid flow residual is to first calculate the average value of the glue liquid flow residual, then calculate the standard deviation of the glue liquid flow residual, subtract the standard deviation from the value of each glue liquid flow residual to obtain the average value, and then divide this value by the average value of the glue liquid flow residual to obtain the average ratio of the change rate of the glue liquid flow residual. At this time, the main focus is on calculating the change rate of the residual in the historical prediction. The residual is the difference between the actual value and the predicted value. At this time, the current data is extrapolated through historical data.

[0053] For the glue layer thickness model, the glue flow rate, material moving speed and material width are obtained to determine the glue layer thickness at this time.

[0054] ;in, Indicates the thickness of the glue layer. Indicates the glue flow rate, Indicates the material moving speed, Indicates material width, which indicates the width of the glass fiber mesh coated at this time.

[0055] At the same time the corresponding It can be expressed as shown below.

[0056] ; This form is used to determine the value that can affect the thickness of the glue layer.

[0057] For the glue dipping amount parameter set, when using the glue liquid flow model and the glue layer thickness model to determine the corresponding glue dipping amount parameters, the currently output glue dipping amount parameter set can also be adjusted by judging the efficiency gain value of the coating and the speed gain value of the speed to the glue dipping amount; the purpose of the adjustment here is to find the relevant data from the parameter memory library through the glue liquid flow and the glue layer thickness and evaluate it. At this time, the evaluation method is to calculate the efficiency gain value and the speed gain for evaluation to determine whether there is a normal value situation at this time.

[0058] At this time, the efficiency gain value and the speed gain value are implemented by obtaining multiple optimization parameters as follows: according to the values ​​of the efficiency gain value and the speed gain value, the efficiency gain value and the speed gain value are adjusted in sequence in the form of continuous time points, and the adjusted efficiency gain value and the speed gain value are output as optimization parameters; wherein the corresponding numbers of the efficiency gain values ​​and the speed gain values ​​are consistent.

[0059] The efficiency gain value is expressed as follows.

[0060] ;in, represents the adjusted efficiency gain value, Indicates the maximum value of the efficiency gain, represents the initial value of the efficiency gain value, represents the average value of the efficiency gain, Represents the residual of the efficiency gain at the i-th time point. The residual is the difference between the actual value currently obtained and the predicted value. represents the residual of the efficiency gain at the i-1th time point, Represents the number of time points, where i ranges from 2 to n.

[0061] The above efficiency gain value is obtained according to the glue flow rate and the glue layer thickness.

[0062] ;in, represents the efficiency gain value, Indicates the glue flow rate, Indicates the maximum value of the glue flow rate, Indicates the thickness of the glue layer. Indicates the maximum value of the glue layer thickness, represents the weight of the glue flow rate, Indicates the weight of the glue layer thickness. The weights set at this time are selected as 0.7 and 0.3 in the order of glue flow rate and glue layer thickness.

[0063] The speed gain value is expressed as: the speed gain value is the ratio of the change in the glue flow rate to the change in the material moving speed.

[0064] ;in, Expressed as the adjusted speed gain value, Indicates the average value of the speed gain value, Indicates the minimum value of the speed gain. represents the residual of the velocity gain value at the i-th time point, Represents the residual of the velocity gain value at the i-1th time point, Represents the number of time points, where i ranges from 2 to n.

[0065] When the adjusted efficiency gain value and speed gain value are greater than the corresponding average values ​​in the preset historical data, the data of glue flow, coating pressure, material movement speed, and glue liquid level corresponding to the efficiency gain value and speed gain value are used as output optimization parameters, and the optimization parameters are combined into a glue dipping amount parameter set.

[0066] In the glue liquid feedback module, the reason for selecting different values ​​of the liquid level change rate and the glue dipping amount is to determine whether consistent glue dipping control can be maintained when different material batches change, so as to obtain the glue liquid feedback control parameters at this time.

[0067] like Figure 5 As shown, the implementation method of obtaining the glue feedback control parameters at this time is shown below.

[0068] The liquid level change rate is obtained, and the correlation coefficient between the liquid level change rate and the amount of dipping glue is determined.

[0069] Get the execution priority of the dipping amount parameter set.

[0070] According to the correlation coefficient between the liquid level change and the glue dipping amount, when the execution priority is consistent with the sorting order of the correlation coefficient from large to small, the data of the glue dipping amount parameter set corresponding to the correlation coefficient is used as a feature vector, where each feature vector corresponds to a liquid level change rate, and the feature vector and the liquid level change rate are used to form glue feedback control parameters according to the result of glue liquid control.

[0071] When the execution priority is inconsistent with the descending sorting order of the correlation coefficient, the data corresponding to the maximum value of the correlation coefficient is obtained as the output glue feedback control parameter.

[0072] In the glue feedback module, the glue dipping amount can be obtained by multiplying the glue flow rate and time.

[0073] The correlation coefficient between the liquid level change rate and the amount of glue dipping is calculated by taking the liquid level change rate and the amount of glue dipping as inputs, and using the calculation method of the Pearson correlation coefficient to determine the value between the liquid level change rate and the amount of glue dipping when different values ​​are selected for the liquid level change rate and the amount of glue dipping, and when the liquid level change rate changes continuously within a unit time.

[0074] The above execution priority represents the weight of the corresponding data in the dipping amount parameter set to quantify the priority between different parameters at this time.

[0075] The method for forming glue liquid feedback control parameters from the characteristic vector and the liquid level change rate according to the result of glue liquid control is to cluster the characteristic vector according to the liquid level change rate to generate multiple clusters, wherein each cluster is conditioned by the consistency of the liquid level change rate, and after clustering the characteristic vector according to multiple clusters, the clusters are output as glue liquid feedback control parameters.

[0076] At this time, the consistency of the liquid level change rate is expressed as data that changes according to the corresponding numerical value when the liquid level change rate value changes. For example, the corresponding feature vector of the liquid level change rate of 1% is taken as one category, and the corresponding feature vector of the liquid level change rate of 5% is taken as another category to obtain data related to the liquid level change rate.

[0077] When the execution priority is inconsistent with the order of sorting the correlation coefficient from large to small, in order to ensure that the control system can respond to the liquid level change quickly and accurately, the parameter set with the largest correlation coefficient is selected as the output glue feedback control parameter. The parameters selected at this time are to select data that is more obvious with the liquid level change rate, and use this part of the data to control the liquid level change rate, so as to adjust the current glue dipping amount. When the liquid level change rate is large, the glue flow rate will be affected by the liquid level height, resulting in a problem of reduced glue flow rate. At this time, it is necessary to determine the corresponding glue dipping amount to determine whether it meets the normal processing steps for the glass fiber mesh.

[0078] The target control parameter in the drying control module is represented by the weighted average value of the glue feedback control parameter after normalization.

[0079] At this time, the selection method of the dipping amount control strategy is shown below.

[0080] The target control parameter is used as the retrieval condition to determine the probability of change of the drying temperature when the target control parameter takes the corresponding value; it is judged whether the probability of change of the drying temperature at this time satisfies the normal distribution. When it satisfies the normal distribution, the value of the temperature boundary condition is determined based on the mean and standard deviation of the drying temperature, and the temperature boundary condition is adjusted. The target control parameter and the adjusted temperature boundary condition are used as the control strategy for the amount of glue impregnation.

[0081] The implementation of determining the value of the temperature boundary condition based on the mean and standard deviation of the drying temperature is shown below.

[0082] ;in, represents the upper limit of the temperature boundary condition, Indicates the average value of drying temperature. Represents the standard deviation of the drying temperature.

[0083] in, Indicates the upper limit value of the temperature boundary condition.

[0084] The method for adjusting the temperature boundary conditions is as shown below: the weights corresponding to the temperature boundary conditions are obtained, and the temperature boundary conditions are adjusted in chronological order; at this time, the data at the time points with the same efficiency gain values ​​are processed, and the values ​​of these data are determined to obtain the final adjusted temperature.

[0085] ;in, represents the value of the adjusted temperature boundary condition, represents the minimum value of the temperature boundary condition, represents the value of the temperature boundary condition at the i-th time point, represents the average value of the temperature boundary condition, represents the weight of the value of the temperature boundary condition at the i-th time point, represents the maximum value of the weight of the value of the temperature boundary condition, Indicates the number of time points. At this time, the value range of i is 1 to n; the temperature boundary condition selected at this time is divided into an upper limit value and a lower limit value, and the upper limit value and the lower limit value are adjusted by the calculation method described above. At this time, the average value and the minimum value of the corresponding temperature boundary condition are determined by adjusting the upper limit value and the lower limit value of the temperature boundary condition respectively, and the corresponding average value and the minimum value can be calculated from the data processed by the temperature boundary condition in the history to obtain the average value and the minimum value, so as to complete the adjustment of the temperature boundary condition.

[0086] After completing the selection of the corresponding temperature boundary conditions and target control parameters, the current control system situation can be judged based on these two values. The target control parameters can reflect the corresponding situation of the glass fiber mesh production at this time, and the temperature boundary conditions can display the preheating and heating temperatures when drying the glass fiber mesh at this time to judge whether the current implementation process meets the requirements. At the same time, according to these two values, the scheme with the same current parameter values ​​can be selected from the normally controlled schemes in the database to adjust the overall production process to improve production efficiency and safety.

[0087] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention and they are still covered by the protection scope of the present invention.

Claims

1. A glass fiber mesh dipped composite control system for wafer cutting grinding wheels, characterized in that: include: The parameter setting module is used to obtain the initial parameters of the glass fiber mesh and perform system calibration; the initial parameters include the coating pressure of the laminating rubber roller, the glue liquid level in the glue tank, the target glue layer thickness, the material moving speed and the drying temperature; The coating control module is used to obtain the glass fiber mesh under the corresponding coating pressure and material moving speed, taking the glue liquid level and the target glue layer thickness as constraint conditions, analyzing the glue dipping amount of the current glue liquid, and obtaining the glue dipping amount parameter set; The glue liquid feedback module is used to analyze the glue dipping amount parameter set, determine the glue liquid level change rate corresponding to different values ​​of the glue dipping amount, and use the glue liquid level change rate as a feedback condition to determine the glue liquid feedback control parameter; The drying control module is used to optimize and analyze the glue feedback control parameters to obtain the target control parameters, and determine the probability of change of the drying temperature according to the target control parameters. According to the probability of change of the drying temperature, the temperature boundary conditions corresponding to the glue dipping amount are obtained, and the glue dipping amount control strategy is determined based on the temperature boundary conditions and the target control parameters; The way to calibrate the system in the parameter setting module is: Obtaining the classification index of the current glass fiber mesh, classifying the glass fiber mesh once according to the classification index of the current glass fiber mesh, and obtaining a classification result; Obtain a historical database of the current glass fiber mesh, and perform secondary classification on the historical database according to the primary classification result to obtain a secondary classification result; the secondary classification result is realized by calculating a classification feature set between the primary classification result and the historical database, and retrieving the secondary classification result based on the similarity of the classification feature set; The initial parameters are adjusted according to the secondary classification results to complete the system calibration of the initial parameters; The implementation of the dipping amount parameter set is as follows: Obtaining a glue flow model and a glue layer thickness model corresponding to the glue liquid level and the target glue layer thickness, dividing the glue flow model and the glue layer thickness model according to the constraint conditions, and respectively obtaining the glue flow output by the glue flow model and the glue layer thickness output by the glue layer thickness model; By evaluating the glue flow rate and glue layer thickness according to the parameter memory library, the efficiency gain value and speed gain value of the glue coating amount are calculated respectively.

2. The glass fiber mesh dipped composite control system for wafer cutting grinding wheel according to claim 1, characterized in that: According to the efficiency gain value and the speed gain value, a plurality of optimization parameters are obtained as an output dipping amount parameter set.

3. The glass fiber mesh dipped composite control system for wafer cutting grinding wheel according to claim 1, characterized in that: For the glue flow model, the coating pressure, material moving speed, glue level, the diameter of the laminating rubber roller, the depth of the laminating rubber roller immersed in the glue tank, and the viscosity of the glue are obtained to obtain the glue flow. The glue flow is expressed as follows: ； in, Indicates the glue flow rate, It represents the empirical constant, which is obtained based on the fitting of the change of the glue flow rate. In this case, the empirical constant represents the average ratio of the residual change rate of the glue flow rate in the historical data; Indicates coating pressure, Indicates the material moving speed, Indicates the glue liquid level. Indicates the diameter of the laminating rubber roller. Indicates the depth of the laminating rubber roller immersed in the rubber tank. Indicates the viscosity of the glue. It represents the influence function of the diameter of the laminating rubber roller, the depth of the laminating rubber roller immersed in the glue tank, and the viscosity of the glue on the glue flow rate.

4. The glass fiber mesh dipped composite control system for wafer cutting grinding wheel according to claim 1, characterized in that: For the glue layer thickness model, the glue flow rate, material moving speed and material width are obtained to determine the glue layer thickness at this time; ； in, Indicates the thickness of the glue layer, Indicates the glue flow rate, Indicates the material moving speed, Indicates material width, which indicates the width of the glass fiber mesh coated at this time.

5. The glass fiber mesh dipped composite control system for wafer cutting grinding wheel according to claim 2, characterized in that: The efficiency gain value and the speed gain value are implemented by obtaining multiple optimization parameters as follows: according to the values ​​of the efficiency gain value and the speed gain value, the efficiency gain value and the speed gain value are adjusted in sequence in the form of continuous time points, and the adjusted efficiency gain value and the speed gain value are output as optimization parameters; wherein the corresponding numbers of the efficiency gain value and the speed gain value are consistent; The efficiency gain value is expressed as follows: ； in, represents the adjusted efficiency gain value, Indicates the maximum value of the efficiency gain, represents the initial value of the efficiency gain value, represents the average value of the efficiency gain, Represents the residual of the efficiency gain at the i-th time point. The residual is the difference between the actual value currently obtained and the predicted value. represents the residual of the efficiency gain at the i-1th time point, Represents the number of time points, where i ranges from 2 to n.

6. The glass fiber mesh dipped composite control system for wafer cutting grinding wheel according to claim 2, characterized in that: The speed gain value is expressed as: The speed gain value is the ratio of the change in the glue flow rate to the change in the material moving speed: ;in, Expressed as the adjusted speed gain value, Indicates the average value of the speed gain value, Indicates the minimum value of the speed gain. represents the residual of the velocity gain value at the i-th time point, Represents the residual of the velocity gain value at the i-1th time point, Represents the number of time points, where i ranges from 2 to n.

7. The glass fiber mesh dipped composite control system for wafer cutting grinding wheel according to claim 1, characterized in that: The implementation method of the glue feedback control parameters is as follows: Obtain the liquid level change rate, and determine the correlation coefficient between the liquid level change rate and the amount of dipping; Get the execution priority of the dipping amount parameter set; According to the correlation coefficient between the liquid level change and the glue dipping amount, when the execution priority is consistent with the sorting order of the correlation coefficient from large to small, the data of the glue dipping amount parameter set corresponding to the correlation coefficient is used as a feature vector, where each feature vector corresponds to a liquid level change rate, and the feature vector and the liquid level change rate are used to form a glue feedback control parameter according to the result of glue liquid control; When the execution priority order is inconsistent with the order of sorting the correlation coefficient from large to small, the data corresponding to the maximum value of the correlation coefficient is obtained as the output glue feedback control parameter.

8. The glass fiber mesh dipped composite control system for wafer cutting grinding wheel according to claim 1, characterized in that: The selection method of the dipping amount control strategy is as follows: The target control parameter is used as the retrieval condition to determine the probability of change of the drying temperature when the target control parameter takes the corresponding value; it is judged whether the probability of change of the drying temperature at this time satisfies the normal distribution. When it satisfies the normal distribution, the value of the temperature boundary condition is determined based on the mean and standard deviation of the drying temperature, and the temperature boundary condition is adjusted. The target control parameter and the adjusted temperature boundary condition are used as the control strategy for the amount of glue impregnation.

9. The glass fiber mesh dipped composite control system for wafer cutting grinding wheel according to claim 8, characterized in that: The method of adjusting the temperature boundary conditions is as follows: obtaining the weight corresponding to the temperature boundary conditions, and adjusting the temperature boundary conditions in chronological order; ； in, represents the value of the adjusted temperature boundary condition, represents the minimum value of the temperature boundary condition, represents the value of the temperature boundary condition at the i-th time point, represents the average value of the temperature boundary condition, represents the weight of the value of the temperature boundary condition at the i-th time point, represents the maximum value of the weight of the value of the temperature boundary condition, Indicates the number of time points; the value of i ranges from 1 to n.

Images