Coating surface density prediction method, device, system, equipment, medium and product
By acquiring and weighting the coating process data in the coating technology and inputting the surface density prediction model, the problem of poor prediction effect of coating surface density is solved, achieving higher prediction accuracy and lower detection cost.
Patent Information
- Application Number
- CN202411893815.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2025-04-15
AI Technical Summary
In the existing coating technology, the prediction effect of coating surface density is poor, and the detection method has problems such as high equipment cost, high material cost, and long cycle.
By acquiring the coating process data, weighting is performed on several partitions of the coating to be detected, and a surface density prediction model corresponding to the target partition is input to improve the prediction accuracy of the coating surface density.
It improves the accuracy of coating surface density prediction, reduces detection costs and cycles, and enhances the control ability of the coating process.
Smart Images
Figure CN119831093A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of coating technology, and in particular to a method, device, system, equipment, medium and product for predicting coating surface density. Background Art
[0002] Coating technology is mainly used for coating and composite packaging of substrates such as paper, fabric, and film. It has been widely used in various high-tech fields such as adhesive products, packaging materials, semiconductors, flexible printed circuit boards, power and energy storage batteries, etc.
[0003] The coating surface density control is one of the key process control points in the coating process. The current coating surface density detection method is to detect the coating surface density through sensors after the coating is formed after changing the operating parameters. However, this detection method has problems such as high equipment cost, high material cost, and long cycle. Therefore, this field can collect various parameters in the coating process through sensors and predict the coating surface density in real time through the parameters. However, in the coating process, the coating surface density will be affected by various factors, resulting in poor prediction of the coating surface density. Summary of the invention
[0004] In view of this, the present application provides a method, device, system, equipment, medium and product for predicting coating surface density, which improves the accuracy of coating surface density prediction. The technical solution is as follows.
[0005] In view of this, the present invention provides a prediction method, device, system, equipment and medium based on coating surface density to improve the prediction accuracy of coating surface density.
[0006] In a first aspect, the present invention provides a method for predicting coating surface density, the method comprising:
[0007] Acquire coating process data; the coating process data includes first sub-data of a plurality of partitions of the coating to be detected in the lateral direction;
[0008] For a target partition among the plurality of partitions, weighted processing is performed on the first sub-data of the plurality of partitions according to a positional relationship between the plurality of partitions and the target partition to obtain second sub-data of the plurality of partitions;
[0009] The second sub-data of the plurality of partitions are input into a surface density prediction model corresponding to the target partition to obtain a surface density prediction result of the target partition.
[0010] In a possible implementation, the type of the first sub-data includes at least one of: pump speed, slurry flow rate, original die gap, coating pressure, slurry temperature, slurry solid content, and slurry viscosity.
[0011] In a possible implementation, when the first sub-data includes slurry flow rate;
[0012] The step of performing weighted processing on the first sub-data of the plurality of partitions according to the positional relationship between the plurality of partitions and the target partition includes:
[0013] For any candidate partition among the plurality of partitions, when the candidate partition is not a target partition, generating a flow weight corresponding to the candidate partition according to an interval between the candidate partition and the target partition;
[0014] When the candidate partition is a target partition, the flow weight of the candidate partition is determined to be 1;
[0015] The slurry flow rate in the first sub-data is weighted according to the flow rate weights corresponding to the candidate partitions.
[0016] In a possible implementation, the plurality of partitions are numbered according to a specified direction;
[0017] The generating, according to the interval between the candidate partition and the target partition, a flow weight corresponding to the candidate partition, comprises:
[0018] Obtaining the difference between the candidate partition and the target partition;
[0019] The product of the serial number difference and a preset attenuation coefficient is determined as the throughput weight corresponding to the candidate partition.
[0020] In a possible implementation, when the plurality of partitions are numbered from left to right, the type of the first sub-data includes an original die gap, and the original die gap includes a left gap of a die with a smallest number and a right gap of a die with a largest number;
[0021] The weighted processing of the first sub-data of the plurality of partitions includes:
[0022] For any candidate partition among the plurality of partitions, obtaining a first ratio of a number of the candidate partition to a number of partitions of the plurality of partitions;
[0023] Determine the product of the first ratio and the die right gap with the largest number as the die right gap of the candidate partition;
[0024] The product of the difference between 1 and the first ratio and the left gap of the die with the smallest number is determined as the left gap of the die of the candidate partition.
[0025] In a possible implementation, the method further includes:
[0026] Acquire sample process data; the sample process data includes first sample data of a plurality of partitions of the coating to be detected in the transverse direction and label surface density of the plurality of partitions;
[0027] For a target partition of the plurality of partitions, weighted processing is performed on the first sample data of the plurality of partitions according to the positional relationship between the plurality of partitions and the target partition to obtain second sample data of the plurality of partitions;
[0028] Inputting the second sample data of the plurality of partitions into the surface density prediction model corresponding to the target partition to obtain the surface density prediction value corresponding to the target partition;
[0029] According to the predicted value of the surface density corresponding to the target partition and the label surface density corresponding to the target partition, parameters of the surface density prediction model corresponding to the target partition are updated.
[0030] In a possible implementation, updating parameters of the surface density prediction model corresponding to the target partition according to the surface density prediction value corresponding to the target partition and the label surface density corresponding to the target partition includes:
[0031] Generate an update weight according to the difference between the label surface density and the preset surface density;
[0032] According to the difference between the predicted surface density value corresponding to the target partition and the label surface density corresponding to the target partition, and the update weight, the parameters of the surface density prediction model corresponding to the target partition are updated.
[0033] In a second aspect, the present invention provides a device for predicting coating surface density, the device comprising:
[0034] An acquisition module, used for acquiring coating process data; the coating process data includes first sub-data of a plurality of partitions of the coating to be detected in the horizontal direction;
[0035] a weighting module, configured to perform weighted processing on the first sub-data of the plurality of partitions according to a positional relationship between the plurality of partitions and the target partition, so as to obtain the second sub-data of the plurality of partitions;
[0036] The prediction module is used to input the second sub-data of the plurality of partitions into the surface density prediction model corresponding to the target partition to obtain the surface density prediction result of the target partition.
[0037] In a possible implementation, the type of the first sub-data includes at least one of: pump speed, slurry flow rate, original die gap, coating pressure, slurry temperature, slurry solid content, and slurry viscosity.
[0038] In a possible implementation, when the first sub-data includes slurry flow rate; the weighting module is further used to:
[0039] For any candidate partition among the plurality of partitions, when the candidate partition is not a target partition, generating a flow weight corresponding to the candidate partition according to an interval between the candidate partition and the target partition;
[0040] When the candidate partition is a target partition, the flow weight of the candidate partition is determined to be 1;
[0041] The slurry flow rate in the first sub-data is weighted according to the flow rate weights corresponding to the candidate partitions.
[0042] In a possible implementation, the weighting module is further used to: obtain a number difference between the candidate partition and the target partition;
[0043] The product of the serial number difference and a preset attenuation coefficient is determined as the throughput weight corresponding to the candidate partition.
[0044] In a possible implementation, when the plurality of partitions are numbered from left to right, the type of the first sub-data includes an original die gap, and the original die gap includes a left gap of a die with a smallest number and a right gap of a die with a largest number; the weighting module is further used to: for any candidate partition among the plurality of partitions, obtain a first ratio of the number of the candidate partition to the number of partitions of the plurality of partitions;
[0045] Determine the product of the first ratio and the die right gap with the largest number as the die right gap of the candidate partition;
[0046] The product of the difference between 1 and the first ratio and the left gap of the die with the smallest number is determined as the left gap of the die of the candidate partition.
[0047] In a possible implementation manner, the device further includes:
[0048] A sample acquisition module, used to acquire sample process data; the sample process data includes first sample data of a plurality of partitions of the coating to be detected in the transverse direction and label surface density of the plurality of partitions;
[0049] A sample weighting module, for performing weighted processing on first sample data of a plurality of partitions according to a positional relationship between the plurality of partitions and the target partition, to obtain second sample data of the plurality of partitions;
[0050] A sample prediction module, used for inputting the second sample data of the plurality of partitions into a surface density prediction model corresponding to the target partition to obtain a surface density prediction value corresponding to the target partition;
[0051] The model updating module is used to update the parameters of the surface density prediction model corresponding to the target partition according to the surface density prediction value corresponding to the target partition and the label surface density corresponding to the target partition.
[0052] In one possible implementation, the model update module is used to generate update weights based on the difference between the label surface density and the preset surface density; and to update the parameters of the surface density prediction model corresponding to the target partition based on the difference between the surface density prediction value corresponding to the target partition and the label surface density corresponding to the target partition, as well as the update weights.
[0053] In a third aspect, a coating simulation debugging system is provided, the system comprising a micrometer head, a displacement sensor, a controller and a computer device;
[0054] Wherein, the differential head is used to press down the adjustment module to adjust the slurry flow rate of the die head;
[0055] The displacement sensor is used to detect the downward pressing distance of the adjustment module;
[0056] The controller is used to collect sensor data on each die head and transmit the sensor data to a computer device to implement the above-mentioned coating surface density prediction method.
[0057] In a third aspect, a computer device is provided, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the above-mentioned coating surface density prediction method by executing the computer instructions.
[0058] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the above-mentioned method for predicting the coating surface density.
[0059] In a fifth aspect, a computer program product is provided, comprising computer instructions for causing a computer to execute the above-mentioned method for predicting the coating surface density.
[0060] The technical solution provided by this application may have the following beneficial effects:
[0061] In the coating process, the computer device can obtain the coating process data, that is, the first sub-data of several partitions of the coating to be detected in the horizontal direction, and then the computer device performs surface density detection for each partition separately. For example, when targeting a target partition, the first sub-data of several partitions can be weighted according to the positional relationship between the several partitions and the target partition to obtain the second sub-data of several partitions. At this time, the computer device inputs the second sub-data of several partitions into the surface density prediction model corresponding to the target partition, so that the surface density model corresponding to the target partition predicts the surface density prediction result of the target partition according to the weighted coating process data of each partition; repeating the above process can obtain the surface density prediction result of each partition in the coating process. In the above scheme, the coating is split into multiple partitions horizontally, and each partition is detected separately, taking into account the influence of the coating process data of other partitions on the partition, and through weighting, the influence between the coating process data of different partitions is taken into account in the prediction process, thereby improving the accuracy of the coating surface density prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0063] Figure 1 A schematic diagram of the system structure of a coating control system involved in an embodiment of the present application is shown.
[0064] Figure 2 A schematic diagram of the structure of a coating simulation debugging system provided in an embodiment of the present application is shown.
[0065] Figure 3 A structural detail diagram of the adjustment module involved in the embodiment of the present application is shown.
[0066] Figure 4 The present invention is a flowchart of a method for predicting coating surface density according to an exemplary embodiment.
[0067] Figure 5 A schematic diagram of the partitioning of a coating to be detected involved in the present application is shown.
[0068] Figure 6 The present invention is a flowchart of a method for predicting coating surface density according to an exemplary embodiment.
[0069] Figure 7A schematic diagram showing the relationship between slurry flow rate and surface density partition involved in an embodiment of the present application is shown.
[0070] Figure 8 A schematic diagram showing the relationship between the die gap and the surface density partition involved in the embodiment of the present application is shown.
[0071] Fig. 9 It is a schematic diagram of the structure of a coating surface density prediction device provided in an embodiment of the present application.
[0072] Fig.10 It is a structural schematic diagram of a computer device provided by an optional embodiment of the present invention. DETAILED DESCRIPTION
[0073] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0074] In the description of the embodiments of the present application, the term "corresponding" may indicate a direct or indirect correspondence between two items, or an association relationship between the two items, or a relationship between indication and being indicated, configuration and being configured, and the like.
[0075] Coating technology is mainly used for coating and composite packaging of substrates such as paper, fabrics, and films. It has been widely used in various high-tech fields such as adhesive products, packaging materials, semiconductors, flexible printed circuit boards, power and energy storage batteries, etc. The main difference between coating in different fields lies in the detailed research of coating technology. The coating process is complex and there are many factors that affect the coating effect. The control of coating surface density is one of the key process control points in the coating process. How to quickly obtain the coating surface density is a key technical problem in the coating process control.
[0076] The current coating surface density detection method is to detect the coating surface density through a sensor after the coating is formed after changing the operating parameters. The current detection method has the problems of high equipment cost, high material cost, long cycle, etc. The present invention provides a model-based coating surface density rapid prediction method, device, equipment and storage medium, which can restore the coating process and realize lossless rapid prediction of coating surface density, which is used to solve the technical problems of low detection efficiency and high detection cost in coating process control.
[0077] Please refer to Figure 1 , which shows a system structure diagram of a coating control system involved in an embodiment of the present application.
[0078] like Figure 1 As shown, the coating control system includes a coating device and a computer device. The coating device is provided with a plurality of sensors, and the plurality of sensors on the coating device can collect various data of the coating device during the working process and transmit the data to the computer device.
[0079] Optionally, after the computer device acquires the sensor data corresponding to the coating device, it can process the sensor data, thereby realizing the prediction of the coating surface density of the coating device during operation.
[0080] exist Figure 1 Based on this, the embodiment of the present application also provides a coating simulation debugging system, such as Figure 2 As shown, the coating simulation debugging system includes a micrometer head 201, a displacement sensor 202, a controller 204 and a computer device 206;
[0081] Wherein, the differential head 201 is used to press down the adjustment module to adjust the slurry flow rate of the die head;
[0082] The displacement sensor 202 is used to detect the downward pressing distance of the adjustment module;
[0083] The controller 204 is used to collect sensor data on each die head and transmit the sensor data to a computer device, and the computer device can predict the coating surface density according to the sensor data in the coating model debugging system.
[0084] Optional, such as Figure 2 As shown, the coating simulation debugging system also includes a communication bus connector 203 and a switch 205; wherein the displacement sensor is connected to the communication bus connector (203) via a signal line; the communication bus connector 203 is used to concentrate the displacement sensor signal lines of each adjustment module and transmit the sensor signal to the controller 204; the controller 204 is used to collect the displacement data transmitted from the communication bus and forward it to the switch 205; the switch 205 transmits the sensor displacement data to the computer device 206.
[0085] Please refer to Figure 3 , which shows a detailed structural diagram of the adjustment module involved in the embodiment of the present application. Figure 3 As shown, when the user manually adjusts the differential head 201 to press down the regulating module to adjust the slurry flow rate of the die head, the displacement sensor can detect the pressing distance of the regulating module and transmit it to the computer device 206.
[0086] The computer device 206 can also make a quick prediction of the coating surface density according to the data of the slurry flow rate of the adjustment module, through the quick prediction of the coating surface density shown in this application, and display the prediction result on the computer device. At this time, the user can know what effect the adjustment of the differential head will have on the coating surface density, thereby realizing the simulation debugging of the coating process. Figure 2 The coating simulation debugging system shown can simulate the coating process by predicting the coating surface density, so that the user can train the debugging method of the micrometer head through the coating simulation debugging system, thereby improving the user's operation accuracy in the actual coating process.
[0087] Figure 4 is a method flow chart of a method for predicting coating surface density according to an exemplary embodiment. The method is executed by a computer device, which may be Figure 1 The computer equipment of the coating control system shown in FIG. Figure 4 As shown, the method for predicting the coating surface density may include the following steps:
[0088] Step 401, obtaining coating process data; the coating process data includes first sub-data of a plurality of partitions of the coating to be detected in the lateral direction.
[0089] Please refer to Figure 5 , which shows a schematic diagram of the partition of a coating to be detected involved in this application. Figure 3 As shown, the coating to be detected has several partitions in the horizontal direction, and there are various types of sensors on the coating equipment, such as thickness sensors, flow sensors, speed sensors, etc. Various types of sensors can respectively collect data from several partitions in the coating to be detected, thereby obtaining the first sub-data in several partitions.
[0090] Step 402 : for a target partition among the plurality of partitions, weighted processing is performed on the first sub-data of the plurality of partitions according to the positional relationship between the plurality of partitions and the target partition, so as to obtain the second sub-data of the plurality of partitions.
[0091] Since the goal of the embodiment of the present application is to predict the coating surface density of the coating to be detected, and since the first sub-data of several partitions have been obtained in step 401, the coating to be detected can be divided into several partitions at this time, and the coating surface density of several partitions can be predicted according to the first sub-data of several partitions.
[0092] However, in the actual coating process, the coating surface density of each partition is not only related to the first sub-data in the partition. For example, when the first sub-data includes the slurry flow rate, the slurry flow rate of the partition will not only affect the coating surface density of the partition, but also affect the coating surface density of the adjacent partitions. The partition farther away from the partition is, the less affected it is. Therefore, in the embodiment of the present application, after obtaining the first sub-data of several partitions, for the target partition, it is necessary to weight the first sub-data of several partitions according to the positional relationship between the several partitions and the target partition, that is, the intensity of the influence of the several partitions on the coating surface density of the target partition, so as to obtain the second sub-data of several partitions.
[0093] Step 403: input the second sub-data of the plurality of partitions into a surface density prediction model corresponding to the target partition to obtain a surface density prediction result of the target partition.
[0094] In the embodiment of the present application, there are several surface density prediction models corresponding to several partitions, and each surface density prediction model is trained to predict the surface density data of its corresponding partition. The process of surface density prediction by each surface density prediction model is similar, so the embodiment of the present application takes the target partition (the target partition can be any one of the several partitions) as an example.
[0095] The surface density model corresponding to the target partition can be a machine learning model with multiple inputs and a single output. The surface density model can simultaneously input the second sub-data of several partitions, and process the second sub-data of several partitions at the same time, and finally obtain the surface density prediction result of the target partition. That is, the surface density model corresponding to the target partition can fuse the features of the second sub-data of each partition, and finally obtain the surface density prediction result of the target partition (that is, the predicted coating surface density within the target partition).
[0096] In summary, in the coating process, the computer device can obtain the coating process data, that is, the first sub-data of several partitions of the coating to be detected in the horizontal direction, and then the computer device performs surface density detection for each partition separately. For example, for the target partition, the first sub-data of several partitions can be weighted according to the positional relationship between the several partitions and the target partition to obtain the second sub-data of several partitions. At this time, the computer device inputs the second sub-data of several partitions into the surface density prediction model corresponding to the target partition, so that the surface density model corresponding to the target partition predicts the surface density prediction result of the target partition according to the weighted coating process data of each partition; repeating the above process can obtain the surface density prediction result of each partition in the coating process. In the above scheme, the coating is split into multiple partitions horizontally, and each partition is detected separately, taking into account the influence of the coating process data of other partitions on the partition, and through weighting, the influence of the coating process data between different partitions is taken into account in the prediction process, thereby improving the accuracy of the coating surface density prediction.
[0097] Figure 6 is a method flow chart of a method for predicting coating surface density according to an exemplary embodiment. The method is executed by a computer device, which may be Figure 1 The computer equipment of the coating control system shown in FIG. Figure 6 As shown, the method for predicting the coating surface density may include the following steps:
[0098] Step 601, obtaining sample process data.
[0099] The sample process data includes first sample data of a plurality of partitions of the coating to be detected in the transverse direction and label surface densities of the plurality of partitions.
[0100] In an embodiment of the present application, in order to predict the surface density of the coating during the coating process, it is first necessary to train surface density prediction models corresponding to several partitions. Therefore, in an embodiment of the present application, in order to train the surface density prediction model, sample process data can be obtained, that is, data collected by the coating equipment during the historical coating process. The sample process data includes both the first sample data of several partitions and the label surface density of several partitions actually detected. At this time, the first sample data can be used as input data for model training, and the label surface density can be used as a label for model training, so as to train a machine learning model that can accurately predict the coating surface density.
[0101] In the embodiment of the present application, the type of the first sample data includes: at least one of pump speed, slurry flow rate, original die gap, coating pressure, slurry temperature, slurry solid content, and slurry viscosity.
[0102] Among them, the surface density data is the lateral surface density data, which can refer to the dry film surface density or the wet film surface density. For example, in the embodiment of the present application, one data can be collected every 1 mm, and one surface density data contains n lateral data. Among them, the slurry flow data type can include but is not limited to the block position, T block position, slurry partition pipeline flow control valve position (corresponding to different slurry discharge methods, that is, different types of die heads); a total of 1 partition adjustment area slurry flow adjustment data.
[0103] After obtaining the n horizontal data, it is necessary to convert the n horizontal data into the first sample data of l partitions through data alignment, data deduplication and partition conversion operations.
[0104] Specifically, in an embodiment of the present application, each data collection item can be aligned according to the collection time to obtain modeling data after data alignment; then the aligned data can be deduplicated according to the surface density data, and the n surface density data at each moment are compared. If there is data with completely consistent surface density, the first data is retained, and the remaining duplicate data is eliminated to obtain the deduplicated data.
[0105] At this time, the surface density data at each moment contains a total of n millimeter-level surface density data. According to the preset partition width k, the n data are partitioned into l surface density adjustment areas. The surface density in each adjustment area is averaged as the surface density value of the current adjustment area. The original n millimeter-level surface density data are converted to l regional surface density values, which are used as label values for subsequent model training.
[0106] Step 602 : for a target partition of a plurality of partitions, weighted processing is performed on the first sample data of the plurality of partitions according to the positional relationship between the plurality of partitions and the target partition, so as to obtain second sample data of the plurality of partitions.
[0107] However, in the actual coating process, the coating surface density of each partition is not only related to the first sample data in the partition. Therefore, for the target partition, it is necessary to weight the first sample data of several partitions according to the positional relationship between the several partitions and the target partition, that is, the intensity of the influence of the several partitions on the coating surface density of the target partition. Please refer to Figure 7 , which shows a schematic diagram of the relationship between the slurry flow rate and the surface density partition involved in the embodiment of the present application. Figure 7 As shown,.
[0108] Optionally, when the first sample data includes slurry circulation, the weighted processing flow includes: for any candidate partition among several partitions, when the candidate partition is not a target partition, generating a circulation weight corresponding to the candidate partition according to the interval between the candidate partition and the target partition; when the candidate partition is the target partition, determining the circulation weight of the candidate partition to be 1; and weighting the slurry circulation in the first sample data according to the circulation weights corresponding to each candidate partition.
[0109] When the coating surface density of the target partition needs to be predicted, the flow weight of the candidate partition can be determined based on the interval between the target partition and each candidate partition in several partitions, that is, the influence intensity of the flow of the candidate partition on the target partition. Since the candidate partition is defined as any one of several partitions, the candidate partition may also be the target partition. In this case, the flow weight can be directly defined as 1; and for any candidate partition that is not the target partition, the interval between the candidate partition and the target partition is directly obtained, and the corresponding flow weight is generated according to the interval.
[0110] Furthermore, several partitions may be numbered, and the interval between each partition and the target partition may be determined based on the number difference. Specifically, several partitions are numbered in a specified direction; the number difference between the candidate partition and the target partition is obtained; and the product of the number difference and a preset attenuation coefficient is determined as the flow weight corresponding to the candidate partition.
[0111] For example, the circulation weight is: beta = beta 0 *dr*(|i-num t |), where beta 0 is the initial coefficient, dr is the attenuation coefficient, the product of the two is the above preset attenuation coefficient, num t is the number of the adjustment module, at this time T final = beta*T; where T final =[t f1 ,t f2 ,…,t fl ] refers to the slurry flow rate used for modeling after flow rate weight adjustment, T = [t 1 ,t 2 ,…,t l ] refers to the collected raw data of slurry flow rate.
[0112] Optionally, when the plurality of partitions are numbered from left to right, the type of the first sample data includes an original die gap, and the original die gap includes a left gap of a die with a smallest number and a right gap of a die with a largest number; at this time, the weighted processing flow further includes:
[0113] For any candidate partition among several partitions, obtain a first ratio of the number of the candidate partition and the number of partitions of the several partitions; determine the product of the first ratio and the right gap of the die with the largest number as the right gap of the die of the candidate partition; determine the product of the difference between 1 and the first ratio and the left gap of the die with the smallest number as the left gap of the die of the candidate partition.
[0114] Please refer to Figure 8 , which shows a schematic diagram of the relationship between the die head gap and the surface density partition involved in the embodiment of the present application. Figure 8 As shown, the first sample data generally includes the left gap of the leftmost die head and the right gap of the rightmost die head, that is, GAP left and GAP right In the embodiment of the present application, the die left gap and die right gap of each candidate partition can be obtained by weighting. Specifically, the die left gap weight is: weight left(i) =1-i / l; Die right gap weight adjustment: weight right(i) =i / l; the final surface density modeling die head gap modeling data is:
[0115] GAP left,final =weight left(i) *GAP left ;
[0116] GAP right,final =weight right(i) *GAP right ;
[0117] Among them, GAP *,final Refers to the die gap data that ultimately participates in the modeling of the i-th surface density model (that is, the die gap data of the i-th candidate partition, including GAP left,final and GAP right,final ), GAP * Refers to the original die gap data, i.e. GAP left and GAP right .
[0118] Step 603: input the second sample data of the plurality of partitions into the surface density prediction model corresponding to the target partition to obtain the surface density prediction value corresponding to the target partition.
[0119] The execution principle of step 603 is detailed in step 402 and will not be repeated here.
[0120] Step 604: update the parameters of the surface density prediction model corresponding to the target partition according to the surface density prediction value corresponding to the target partition and the label surface density corresponding to the target partition.
[0121] Optionally, the computer device generates an update weight based on the difference between the label surface density and the preset surface density; and updates the parameters of the surface density prediction model corresponding to the target partition based on the difference between the surface density prediction value corresponding to the target partition and the label surface density corresponding to the target partition, as well as the update weight.
[0122] When training the surface density prediction model corresponding to the target partition, in order to improve the training effect, the difference between the label surface density and the preset surface density can be calculated to characterize the degree of deviation of each sample surface density data from the set value, and the update weight of the model corresponding to the samples with high deviation can be increased to increase the diversity of modeling data.
[0123] Specifically, the following formula can be used to express the updated weight:
[0124] (1)y bias =|y i -y s |;
[0125] (2)y exp =e^y bias ;
[0126] (3)y weight =y exp / ∑y exp ;
[0127] Among them, y bias is the deviation between the ith label surface density and the preset surface density, y weight is the updated weight of the i-th sample.
[0128] Step 605, obtaining coating process data; the coating process data includes first sub-data of a plurality of partitions of the coating to be detected in the horizontal direction.
[0129] Optionally, the type of the first sub-data includes at least one of: pump speed, slurry flow rate, original die gap, coating pressure, slurry temperature, slurry solid content, and slurry viscosity.
[0130] Step 606: for a target partition among the plurality of partitions, weighted processing is performed on the first sub-data of the plurality of partitions according to the positional relationship between the plurality of partitions and the target partition, so as to obtain the second sub-data of the plurality of partitions.
[0131] Optionally, when the first sub-data includes slurry circulation, the weighted processing flow includes: for any candidate partition among several partitions, when the candidate partition is not the target partition, generating a circulation weight corresponding to the candidate partition according to the interval between the candidate partition and the target partition; when the candidate partition is the target partition, determining the circulation weight of the candidate partition to be 1; and weighting the slurry circulation in the first sub-data according to the circulation weights corresponding to each candidate partition.
[0132] Furthermore, several partitions may be numbered, and the interval between each partition and the target partition may be determined based on the number difference. Specifically, several partitions are numbered in a specified direction; the number difference between the candidate partition and the target partition is obtained; and the product of the number difference and a preset attenuation coefficient is determined as the flow weight corresponding to the candidate partition.
[0133] Optionally, when the plurality of partitions are numbered from left to right, the type of the first sub-data includes an original die gap, and the original die gap includes a left gap of a die with a smallest number and a right gap of a die with a largest number; at this time, the weighted processing flow further includes:
[0134] For any candidate partition among several partitions, obtain a first ratio of the number of the candidate partition and the number of partitions of the several partitions; determine the product of the first ratio and the right gap of the die with the largest number as the right gap of the die of the candidate partition; determine the product of the difference between 1 and the first ratio and the left gap of the die with the smallest number as the left gap of the die of the candidate partition.
[0135] Step 607: input the second sub-data of the plurality of partitions into the surface density prediction model corresponding to the target partition to obtain the surface density prediction result of the target partition.
[0136] Among them, the above steps 605 to 607 are the model application process corresponding to the model training process of steps 601 to 603. Therefore, the specific execution method of steps 605 to 607 is similar to that of steps 601 to 603, and will not be repeated here.
[0137] In summary, in the coating process, the computer device can obtain the coating process data, that is, the first sub-data of several partitions of the coating to be detected in the horizontal direction, and then the computer device performs surface density detection for each partition separately. For example, for the target partition, the first sub-data of several partitions can be weighted according to the positional relationship between the several partitions and the target partition to obtain the second sub-data of several partitions. At this time, the computer device inputs the second sub-data of several partitions into the surface density prediction model corresponding to the target partition, so that the surface density model corresponding to the target partition predicts the surface density prediction result of the target partition according to the weighted coating process data of each partition; repeating the above process can obtain the surface density prediction result of each partition in the coating process. In the above scheme, the coating is split into multiple partitions horizontally, and each partition is detected separately, taking into account the influence of the coating process data of other partitions on the partition, and through weighting, the influence of the coating process data between different partitions is taken into account in the prediction process, thereby improving the accuracy of the coating surface density prediction.
[0138] In the embodiments of the present application, a device for predicting the surface density of a coating is also provided, and the device is used to implement the above-mentioned embodiments and preferred embodiments, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.
[0139] The present application embodiment provides a device for predicting coating surface density. Fig. 9 : is a schematic diagram of a coating surface density prediction device provided in an embodiment of the present application, the device comprising:
[0140] The acquisition module 901 is used to acquire coating process data; the coating process data includes first sub-data of a plurality of partitions of the coating to be detected in the horizontal direction;
[0141] A weighting module 902 is used for performing weighted processing on the first sub-data of the plurality of partitions according to the position relationship between the plurality of partitions and the target partition, so as to obtain the second sub-data of the plurality of partitions;
[0142] The prediction module 903 is used to input the second sub-data of the plurality of partitions into the surface density prediction model corresponding to the target partition to obtain the surface density prediction result of the target partition.
[0143] In a possible implementation, the type of the first sub-data includes at least one of: pump speed, slurry flow rate, original die gap, coating pressure, slurry temperature, slurry solid content, and slurry viscosity.
[0144] In a possible implementation, when the first sub-data includes slurry flow rate, the weighting module 902 is further configured to:
[0145] For any candidate partition among the plurality of partitions, when the candidate partition is not a target partition, generating a flow weight corresponding to the candidate partition according to an interval between the candidate partition and the target partition;
[0146] When the candidate partition is a target partition, the flow weight of the candidate partition is determined to be 1;
[0147] The slurry flow rate in the first sub-data is weighted according to the flow rate weights corresponding to the candidate partitions.
[0148] In a possible implementation, the weighting module 902 is further used to: obtain a number difference between the candidate partition and the target partition;
[0149] The product of the serial number difference and a preset attenuation coefficient is determined as the throughput weight corresponding to the candidate partition.
[0150] In a possible implementation, when the plurality of partitions are numbered from left to right, the type of the first sub-data includes an original die gap, and the original die gap includes a left gap of a die with a smallest number and a right gap of a die with a largest number; the weighting module 902 is further used to: for any candidate partition among the plurality of partitions, obtain a first ratio of the number of the candidate partition to the number of partitions of the plurality of partitions;
[0151] Determine the product of the first ratio and the die right gap with the largest number as the die right gap of the candidate partition;
[0152] The product of the difference between 1 and the first ratio and the left gap of the die with the smallest number is determined as the left gap of the die of the candidate partition.
[0153] In a possible implementation manner, the device further includes:
[0154] A sample acquisition module, used to acquire sample process data; the sample process data includes first sample data of a plurality of partitions of the coating to be detected in the transverse direction and label surface density of the plurality of partitions;
[0155] A sample weighting module, for performing weighted processing on first sample data of a plurality of partitions according to a positional relationship between the plurality of partitions and the target partition, to obtain second sample data of the plurality of partitions;
[0156] A sample prediction module, used for inputting the second sample data of the plurality of partitions into a surface density prediction model corresponding to the target partition to obtain a surface density prediction value corresponding to the target partition;
[0157] The model updating module is used to update the parameters of the surface density prediction model corresponding to the target partition according to the surface density prediction value corresponding to the target partition and the label surface density corresponding to the target partition.
[0158] In one possible implementation, the model update module is used to generate update weights based on the difference between the label surface density and the preset surface density; and to update the parameters of the surface density prediction model corresponding to the target partition based on the difference between the surface density prediction value corresponding to the target partition and the label surface density corresponding to the target partition, as well as the update weights.
[0159] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.
[0160] In summary, in the coating process, the computer device can obtain the coating process data, that is, the first sub-data of several partitions of the coating to be detected in the horizontal direction, and then the computer device performs surface density detection for each partition separately. For example, for the target partition, the first sub-data of several partitions can be weighted according to the positional relationship between the several partitions and the target partition to obtain the second sub-data of several partitions. At this time, the computer device inputs the second sub-data of several partitions into the surface density prediction model corresponding to the target partition, so that the surface density model corresponding to the target partition predicts the surface density prediction result of the target partition according to the weighted coating process data of each partition; repeating the above process can obtain the surface density prediction result of each partition in the coating process. In the above scheme, the coating is split into multiple partitions horizontally, and each partition is detected separately, taking into account the influence of the coating process data of other partitions on the partition, and through weighting, the influence of the coating process data between different partitions is taken into account in the prediction process, thereby improving the accuracy of the coating surface density prediction.
[0161] The coating surface density prediction device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.
[0162] An embodiment of the present invention further provides a computer device, which can be implemented as a computer device in the above-mentioned coating control system.
[0163] See also Fig.10 , Fig.10 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Fig.10 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphic information in a graphical user interface on an external input / output device (such as a display device coupled to an interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig.10 A processor 10 is taken as an example.
[0164] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.
[0165] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.
[0166] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0167] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.
[0168] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.
[0169] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.
[0170] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.
[0171] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.
Claims
1. A method for predicting coating surface density, characterized in that: The method comprises: Acquire coating process data; the coating process data includes first sub-data of a plurality of partitions of the coating to be detected in the lateral direction; For a target partition among the plurality of partitions, weighted processing is performed on the first sub-data of the plurality of partitions according to a positional relationship between the plurality of partitions and the target partition to obtain second sub-data of the plurality of partitions; The second sub-data of the plurality of partitions are input into a surface density prediction model corresponding to the target partition to obtain a surface density prediction result of the target partition.
2. The method according to claim 1, characterized in that The type of the first sub-data includes: at least one of pump speed, slurry flow rate, original die gap, coating pressure, slurry temperature, slurry solid content, and slurry viscosity.
3. The method according to claim 2, characterized in that When the first sub-data includes slurry flow rate; The step of performing weighted processing on the first sub-data of the plurality of partitions according to the positional relationship between the plurality of partitions and the target partition includes: For any candidate partition among the plurality of partitions, when the candidate partition is not a target partition, generating a flow weight corresponding to the candidate partition according to an interval between the candidate partition and the target partition; When the candidate partition is a target partition, the flow weight of the candidate partition is determined to be 1; The slurry flow rate in the first sub-data is weighted according to the flow rate weight corresponding to each candidate partition.
4. The method according to claim 3, characterized in that The plurality of partitions are numbered according to a specified direction; The generating, according to the interval between the candidate partition and the target partition, a flow weight corresponding to the candidate partition, comprises: Obtaining the difference between the candidate partition and the target partition; The product of the serial number difference and a preset attenuation coefficient is determined as the throughput weight corresponding to the candidate partition.
5. The method according to claim 2, characterized in that: When the plurality of partitions are numbered from left to right, the type of the first sub-data includes an original die gap, and the original die gap includes a left gap of a die with the smallest number and a right gap of a die with the largest number; The weighted processing of the first sub-data of the plurality of partitions includes: For any candidate partition among the plurality of partitions, obtaining a first ratio of a number of the candidate partition to a number of partitions of the plurality of partitions; Determine the product of the first ratio and the die right gap with the largest number as the die right gap of the candidate partition; The product of the difference between 1 and the first ratio and the left gap of the die with the smallest number is determined as the left gap of the die of the candidate partition.
6. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: Acquire sample process data; the sample process data includes first sample data of a plurality of partitions of the coating to be detected in the transverse direction and label surface density of the plurality of partitions; For a target partition of the plurality of partitions, weighted processing is performed on the first sample data of the plurality of partitions according to the positional relationship between the plurality of partitions and the target partition to obtain second sample data of the plurality of partitions; Inputting the second sample data of the plurality of partitions into the surface density prediction model corresponding to the target partition to obtain the surface density prediction value corresponding to the target partition; According to the predicted value of the surface density corresponding to the target partition and the label surface density corresponding to the target partition, parameters of the surface density prediction model corresponding to the target partition are updated.
7. The method according to claim 6, characterized in that According to the surface density prediction value corresponding to the target partition and the label surface density corresponding to the target partition, updating parameters of the surface density prediction model corresponding to the target partition includes: Generate an update weight according to the difference between the label surface density and the preset surface density; According to the difference between the predicted surface density value corresponding to the target partition and the label surface density corresponding to the target partition, and the update weight, the parameters of the surface density prediction model corresponding to the target partition are updated.
8. A device for predicting coating surface density, characterized in that: The device comprises: An acquisition module, used for acquiring coating process data; the coating process data includes first sub-data of a plurality of partitions of the coating to be detected in the horizontal direction; a weighting module, configured to perform weighted processing on the first sub-data of the plurality of partitions according to a positional relationship between the plurality of partitions and the target partition, so as to obtain the second sub-data of the plurality of partitions; The prediction module is used to input the second sub-data of the plurality of partitions into the surface density prediction model corresponding to the target partition to obtain the surface density prediction result of the target partition.
9. A coating simulation debugging system, characterized in that: The system includes a micrometer head, a displacement sensor, a controller and a computer device; Wherein, the differential head is used to press down the adjustment module to adjust the slurry flow rate of the die head; The displacement sensor is used to detect the downward pressing distance of the adjustment module; The controller is used to collect sensor data on each die head and transmit the sensor data to a computer device to implement the coating surface density prediction method as described in any one of claims 1 to 7.
10. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the coating surface density prediction method described in any one of claims 1 to 7 by executing the computer instructions.
11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for predicting the coating surface density according to any one of claims 1 to 7.
12. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the method for predicting the coating surface density according to any one of claims 1 to 7.