A pressure and deformation intelligent feedback control method and system for ceramic flattening equipment

Abstract

The present invention belongs to the technical field of pressure control, and discloses a pressure and deformation intelligent feedback control method and system for ceramic flattening equipment; the method comprises: collecting ceramic physical data of ceramics; dividing the ceramics into m areas; obtaining pressure values ​​required for the m areas during the ceramic flattening process; according to the pressure values ​​required for the m areas, controlling the ceramic flattening equipment to flatten the ceramics, and collecting deformation data of the m areas during the ceramic flattening process and providing real-time feedback; according to the deformation data fed back in real time, judging whether to generate an adjustment instruction, marking the area where the adjustment instruction is generated as the adjustment area, calculating the adjustment pressure of the adjustment area, and controlling the pressure value of the adjustment area in the ceramic flattening equipment to be adjusted to the adjustment pressure; the present invention can realize individualized and refined control of pressure values, effectively reduce the situation where the ceramics are subjected to uneven force, and thus improve the flattening quality and precision of the ceramics.

Description

Technical Field

[0001] The present invention relates to the field of pressure control technology, and more specifically, to a pressure and deformation intelligent feedback control method and system for ceramic flattening equipment. Background Art

[0002] In the production process of ceramic products, in order to improve the flatness and quality of the products, it is necessary to precisely control the shape and size of the products by flattening. Traditional ceramic flattening equipment mainly uses fixed pressure to flatten ceramics, and cannot accurately sense and adjust the deformation of the products during the flattening process in real time, which may cause uneven force on the products, fail to meet the design requirements, and make it difficult to achieve automated production.

[0003] Of course, there are also intelligent pressure regulation methods. For example, a Chinese patent with publication number CN114995541A discloses a method for controlling the internal pressure of a tobacco flavoring roller device; the method comprises: obtaining data on the internal pressure of the flavoring roller device, material flow rate, compressed air pressure for flavor atomization, and dehumidification damper opening, and removing abnormal values ​​from the acquired data; taking the material flow rate, compressed air pressure for flavor atomization, and dehumidification damper opening as inputs, and taking the internal pressure of the device as a response, performing regression analysis to obtain a regression model of the internal pressure of the flavoring roller device; comparing the actual value of the internal pressure of the flavoring roller device with the optimal value; if the actual value does not meet the optimal value requirement, then according to the regression model of the internal pressure of the flavoring roller device, obtaining the dehumidification damper opening corresponding to the optimal value of the internal pressure of the flavoring roller device, and adjusting the dehumidification damper opening through feedback control, so that the internal pressure of the flavoring roller device meets the optimal value requirement; if the dehumidification damper opening exceeds the required value, performing alarm prompts and other steps to stabilize the internal pressure of the flavoring roller device and improve the stability of the tobacco flavoring quality;

[0004] However, the above technology can only compare the actual pressure value with the optimal value and adjust the opening of the moisture exhaust door to ensure that the pressure value in the aroma drum equipment meets the optimal value requirement; if the above technology is applied to the pressure control of the ceramic flattening equipment, it is impossible to monitor and feedback the deformation of the ceramic product during the flattening process in real time; at the same time, only the overall pressure value can be adjusted, and the pressure in different areas cannot be accurately controlled according to the deformation distribution, which leads to uneven pressure distribution and uneven force on the product, thereby affecting the accuracy and stability of the ceramic product;

[0005] In view of this, the present invention proposes a pressure and deformation intelligent feedback control method and system for a ceramic flattening device to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a pressure and deformation intelligent feedback control method of a ceramic flattening device, comprising:

[0007] Collect ceramic physical data of ceramics;

[0008] Divide the ceramic into m regions;

[0009] Collecting the flatness data corresponding to m areas in the ceramic;

[0010] Analyze the ceramic physical data and the flatness data corresponding to the m regions to obtain the pressure values ​​required for the m regions during the ceramic flattening process;

[0011] According to the pressure values ​​required by the m regions, the ceramic flattening equipment is controlled to flatten the ceramics, and the deformation data of the m regions are collected and fed back in real time during the ceramic flattening process;

[0012] Based on the deformation data of the m areas fed back in real time, determine whether to generate an adjustment instruction, mark the area where the adjustment instruction is generated as the adjustment area, analyze the deformation data of the adjustment area, calculate the adjustment pressure, and control the pressure value of the adjustment area in the ceramic flattening equipment to be adjusted to the adjustment pressure.

[0013] Furthermore, the ceramic is a ceramic to be flattened; the physical data of the ceramic includes strength data and elastic modulus;

[0014] The strength data include tensile strength, compressive strength and shear strength;

[0015] The tensile strength is the bearing capacity of the ceramic under tension, the compressive strength is the bearing capacity of the ceramic under load, and the shear strength is the bearing capacity of the ceramic under shear;

[0016] The elastic modulus is the ease with which a ceramic can deform when subjected to force.

[0017] The method for obtaining the intensity data comprises:

[0018] An RFID reader is used to read the RFID tag embedded in the ceramic surface to obtain the corresponding tag ID information, and the tag ID information is sent to the ceramic database to obtain a set of strength data corresponding to the tag ID information, wherein the RFID tag corresponds to the tag ID information one-to-one, and the tag ID information corresponds to the ceramic strength data one-to-one; the ceramic database records the strength data corresponding to the tag ID information of each ceramic.

[0019] Furthermore, the method for obtaining the elastic modulus includes:

[0020] Using an ultrasonic device, the ultrasonic device includes an ultrasonic transmitter and an ultrasonic receiver, the ultrasonic transmitter transmits an ultrasonic pulse to the ceramic, and the ultrasonic receiver receives the ultrasonic pulse reflected from the inside of the ceramic; obtaining an angle value between the ultrasonic device and the ceramic, the angle value including a first angle value and a second angle value, the first angle value being the angle between the ultrasonic transmitter and the ceramic surface, and the second angle value being the angle between the ultrasonic receiver and the ceramic surface;

[0021] Obtaining a first distance value and a second distance value, wherein the first distance value is the distance value between the ultrasonic device and the ceramic surface, and the second distance value is the distance value between the ultrasonic device and the surface of the platform to be processed, wherein the platform to be processed is the platform on which the ceramic is placed in the ceramic lamination device;

[0022] Use a laser scanner to scan the ceramic surface, obtain a scanned image of the ceramic surface, and mark the center point of the scanned image; take the center point as the starting point, measure the distance from the center point to any edge in the scanned image, and mark it as the first edge distance, and mark the edge corresponding to the first edge distance as the first edge; mark any edge corresponding to the first edge distance as the second edge, measure the distance from the center point to the second edge in the scanned image, and mark it as the second edge distance; calculate the ceramic area according to the first edge distance and the second edge distance; the expression of the ceramic area is: M = L'×L", where M is the ceramic area, L' is the first edge distance, and L" is the second edge distance;

[0023] The second distance value is subtracted from the first distance value to obtain the ceramic thickness, and the ceramic area is multiplied by the ceramic thickness to obtain the ceramic volume; the ceramic mass is obtained, and the ceramic mass is divided by the ceramic volume to obtain the ceramic density;

[0024] The first distance value is calculated according to the first angle value and the first distance value. The first distance value is the distance value between the ultrasonic transmitter and the reflection point. The reflection point is the contact point between the laser pulse and the inside of the ceramic. The expression of the first distance value is: Where S1 is the first distance value, L1 is the first distance value, and θ1 is the first angle value. The second distance value is calculated according to the second angle value and the first distance value. The second distance value is the distance value between the ultrasonic receiver and the reflection point. The expression of the second distance value is: Where S2 is the second distance value, θ2 is the second angle value; the first distance value plus the second distance value is taken as the pulse distance, and the pulse distance is the distance traveled by the ultrasonic pulse from the ultrasonic transmitter to the ultrasonic receiver; the time data is collected, and the time data is the time traveled by the ultrasonic pulse from the ultrasonic transmitter to the ultrasonic receiver; the pulse distance is divided by the time data to obtain the elastic wave velocity; the elastic modulus is calculated according to the elastic wave velocity and the ceramic density, and the expression of the elastic modulus is: E = V t 2 ×ρ; where E is the elastic modulus, V tis the elastic wave velocity, and ρ is the ceramic density.

[0025] Furthermore, the m regions are divided according to the position of each pressure pad in the ceramic flattening device, wherein the pressure pad corresponds to the region one by one; the flatness data is the roughness of the ceramic surface; and the method for obtaining the flatness data includes:

[0026] Use a laser scanner to irradiate a laser beam on the ceramic surface for scanning; mark n points in each area of ​​the ceramic surface, where n is an integer greater than 1, and measure the height data of the n points corresponding to each area; calculate the mean value of the height data corresponding to each area, and calculate the standard deviation of the height data of the n points corresponding to each area, and use the standard deviation of the height data of the n points corresponding to each area as the flatness data of each area;

[0027] The expression of flatness data is:

[0028] In the formula, R i is the flatness data of the ith area, H ij is the height data of the jth point in the i-th area, is the mean value of the height data of the ith area, j∈[1,n], n is the number of points, i∈[1,m].

[0029] Furthermore, the step of obtaining the pressure values ​​required for m regions during the ceramic flattening process includes:

[0030] Step 1: Randomly select an area that is not marked as a selected area from the m areas and mark it as a selected area;

[0031] Step 2: Encode the pressure value, obtain the chromosome, and construct the initial population;

[0032] Step 3: Determine the fitness function;

[0033] Step 4: Natural selection of chromosomes in the population;

[0034] Step 5: Perform crossover recombination on chromosomes in the population;

[0035] Step 6: mutate the chromosomes in the population;

[0036] Step 7: Get a new population, preset the population generation number to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop steps 4 to 6 until the generation number corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is less than or equal to the fitness threshold Q, the loop ends, and the pressure value of the chromosome with the minimum fitness in the new population is obtained, and used as the pressure value corresponding to the selected region in step 1;

[0037] Step 8: Loop steps 1 to 7 until all m regions are marked as selected regions. The loop ends and the pressure values ​​required for the m regions in the ceramic flattening process are obtained.

[0038] Furthermore, in step 2, the pressure value is encoded as X, where X is a chromosome, and the pressure value range [C1, C2] that can be generated by the ceramic flattening device is obtained, where the range of X is [C1, C2]; N chromosomes are randomly generated to form an initial population Z0 = {X1, X2, ..., X N};

[0039] In step 3, the fitness function is expressed as: q =R′ q ;

[0040] In the formula, f q is the fitness corresponding to the qth chromosome, R′ q is the evaluation flatness data corresponding to the qth chromosome, the evaluation flatness data is the roughness of the ceramic surface after flattening, q∈[1,N];

[0041] Methods for obtaining data for evaluating flatness include:

[0042] The ceramic physical data, the flatness data of a region and the pressure value corresponding to chromosome X are used as a set of analysis data; the analysis data are input into the trained flatness prediction model to predict the evaluation flatness data;

[0043] The training process of the flatness prediction model includes:

[0044] Collecting in advance the evaluation flatness data corresponding to a plurality of sets of analysis data, and converting the analysis data and the corresponding evaluation flatness data into a corresponding set of feature vectors;

[0045] Each group of feature vectors is used as the input of the flatness prediction model. The flatness prediction model uses a group of predicted and evaluated flatness data corresponding to each group of analysis data as output, and uses the actual evaluated flatness data corresponding to each group of analysis data as the prediction target, where the actual evaluated flatness data is the pre-set evaluated flatness data corresponding to the analysis data; minimizing the sum of the prediction errors of all the analysis data is used as the training target; the flatness prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped; the flatness prediction model is a deep neural network model;

[0046] In step 4, natural selection is performed by combining the elite method and the rotation method; the elite method generates k1 offspring chromosomes, and for a population with a capacity of N, the fitness corresponding to the N chromosomes is arranged from small to large, and each of the k1 chromosomes at the front generates a offspring chromosome; the rotation method generates k2 offspring chromosomes, that is, N chromosomes generate k2 offspring chromosomes according to the corresponding rotation probability; k1+k2=N;

[0047] The expression of rotation probability is:

[0048] Where P q is the rotation probability corresponding to the qth chromosome;

[0049] In the step 5, U chromosomes are randomly selected from the population for crossover recombination to obtain U new chromosomes; the crossover recombination adopts the PMX method; after the chromosome crossover recombination, the fitness of the U new chromosomes is calculated, the fitness of the U new chromosomes and the fitness of the U chromosomes are sorted from small to large and a sorting table is generated, and the U chromosomes in the sorting table are replaced with the U chromosomes that are crossover recombined in the population in positive order;

[0050] In step 6, the mutation probability is preset to G, and the N chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes.

[0051] Furthermore, the deformation data is the variation of the roughness of the ceramic surface; a collection interval is preset, and the deformation data is collected according to the collection interval;

[0052] The method for acquiring the deformation data comprises:

[0053] An elastic film is added between the ceramic and the pressure pad of the ceramic flattening device, and M flexible strain gauges are embedded in the elastic film, M=m, and the flexible strain gauges correspond to the regions one by one to obtain the strain data of the m regions; a proportional coefficient is preset, and the deformation data is obtained according to the product of the strain data and the proportional coefficient; the expression of the deformation data is: BX i =ψ×YB i ; In the formula, BX i is the deformation data of the ith region, ψ is the proportional coefficient, YB i is the strain data of the ith region.

[0054] Furthermore, the method for determining whether to generate an adjustment instruction includes:

[0055] Calculate the deformation threshold TH of m regions i , the deformation data BX of m regions i Respectively with the corresponding deformation threshold THi Make a comparison;

[0056] If BX i ≥TH i , no adjustment instruction is generated;

[0057] If BX i <TH i , then generate adjustment instructions;

[0058] The calculation method of deformation threshold includes:

[0059] TH i =ω1×R i +ω2×P i ′+ω3×t;

[0060] Where TH i is the deformation threshold of the ith region, P i ′ is the pressure value of the ith area, t is the sampling interval, ω1, ω2, ω3 are the preset weight coefficients;

[0061] Methods for calculating regulated pressure include:

[0062] Subtract the corresponding deformation data from the deformation threshold of the adjustment area to obtain the difference, add the difference to the deformation threshold of the adjustment area to obtain the deformation adjustment data, and calculate the adjustment pressure according to the deformation adjustment data, the acquisition interval and the flatness data; the expression of the adjustment pressure is: Where P h ″ is the regulating pressure of the hth regulating area, TH′ h is the deformation adjustment data of the hth adjustment area, R h is the flatness data of the hth adjustment area, h∈[1,h′], h′ is the number of adjustment areas, 0<h′≤m.

[0063] Furthermore, the flattening time is collected to re-determine whether to generate an adjustment instruction;

[0064] The flattening time is the operating time of the ceramic flattening equipment during one operation;

[0065] Methods for re-determining whether to generate an adjustment instruction include:

[0066] The ceramic flattening process is divided into W sections according to the collection interval, and the duration of each section is the collection interval;

[0067] Set digital labels for W segments in increasing order, with the digital label range being [1, W];

[0068] The digital label corresponding to the segment is marked as the segment label, the segment label and the deformation threshold are used as evaluation data, and the W′ group of evaluation data corresponding to each region is sequentially input into the trained influence analysis model to predict the corresponding influence coefficient, where W′=W;

[0069] The training process of the impact analysis model includes:

[0070] Collecting influence coefficients corresponding to multiple groups of evaluation data in advance, and converting the evaluation data and the corresponding influence coefficients into a corresponding set of feature vectors;

[0071] Each group of feature vectors is used as the input of the impact analysis model. The impact analysis model uses a group of predicted impact coefficients corresponding to each group of evaluation data as output, and uses the actual impact coefficients corresponding to each group of evaluation data as the prediction target. The actual evaluation flatness data is the pre-set impact coefficient corresponding to the evaluation data; minimizing the sum of the prediction deviations of all evaluation data is used as the training target; the impact analysis model is trained until the sum of the prediction deviations reaches convergence and the training is stopped; the impact analysis model is a deep neural network model;

[0072] Multiply the obtained F influence coefficients by the deformation threshold of the corresponding area to obtain new deformation thresholds corresponding to W segments in the m areas, where F = W″×m, W″ is the number of influence coefficients of each area and W″=W′;

[0073] The flattening time corresponding to the deformation data of m regions is divided by the collection interval to obtain the time ratio. The integer part of the time ratio is the corresponding segment label. According to the segment label, the new deformation threshold corresponding to the m regions is obtained. The deformation data of the m regions are compared with the corresponding new deformation threshold. If No adjustment instruction is generated if Generate adjustment instructions, is the new deformation threshold corresponding to the i-th region.

[0074] A pressure and deformation intelligent feedback control system for a ceramic flattening device, and a pressure and deformation intelligent feedback control method for a ceramic flattening device, comprising:

[0075] Ceramic data acquisition module, used to collect ceramic physical data of ceramics;

[0076] A region division module, used for dividing the ceramic into m regions;

[0077] A flatness data acquisition module is used to collect flatness data corresponding to m areas in the ceramic;

[0078] A data analysis module is used to analyze the ceramic physical data and the flatness data corresponding to the m regions, and obtain the pressure values ​​required for the m regions during the ceramic flattening process;

[0079] The ceramic flattening module is used to control the ceramic flattening equipment to flatten the ceramic according to the pressure values ​​required by the m areas, and collect the deformation data of the m areas during the ceramic flattening process and provide real-time feedback;

[0080] The feedback control module is used to determine whether to generate an adjustment instruction based on the deformation data of the m areas fed back in real time, mark the area where the adjustment instruction is generated as the adjustment area, analyze the deformation data of the adjustment area, calculate the adjustment pressure, and control the pressure value of the adjustment area in the ceramic flattening equipment to be adjusted to the adjustment pressure.

[0081] Technical effects and advantages of the pressure and deformation intelligent feedback control method and system of a ceramic flattening device of the present invention:

[0082] 1. By collecting ceramic physical data and flatness data and dividing the ceramic into multiple areas, the pressure values ​​required for different areas are calculated in real time, and the ceramic flattening equipment is automatically controlled to flatten according to the pressure values ​​required for different areas, thereby realizing individualized control of the pressure value; at the same time, it can monitor the deformation of the ceramic during the flattening process in real time, and collect the corresponding deformation data, and judge whether the pressure value needs to be adjusted according to the deformation data of each area. By independently calculating and adjusting the pressure value of the adjustment area, the refined control of the pressure value can be realized; it can effectively reduce the uneven force on the ceramic, thereby improving the flattening quality and accuracy of the ceramic; it realizes the automated operation of the ceramic flattening process, effectively improves the production efficiency, and lays the foundation for intelligent ceramic manufacturing.

[0083] 2. The ceramic flattening process is divided into multiple time zones, and the influence coefficient corresponding to each section is set; the deformation thresholds of different time stages can be dynamically adjusted to accurately reflect the different deformation standards of ceramics in different flattening stages; the misjudgment problem that may be caused by the use of fixed deformation thresholds is eliminated, and the areas that need to adjust the pressure are more accurately identified, thereby improving the accuracy of the adjusted pressure value; the learning and simulation of the deformation law of the entire flattening process is realized, and the intelligence level of deformation threshold adjustment is improved, thereby improving the pressure control quality and the ceramic flattening effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 This is a schematic diagram of a pressure and deformation intelligent feedback control system for a ceramic flattening device according to Example 1 of the present invention;

[0085] Figure 2 This is a schematic diagram of the elastic modulus acquisition principle of Example 1 of the present invention;

[0086] Figure 3 This is a schematic diagram of a pressure and deformation intelligent feedback control system for a ceramic flattening device according to Embodiment 2 of the present invention;

[0087] Figure 4 This is a flow chart of a method for intelligent feedback control of pressure and deformation of a ceramic flattening device according to Embodiment 3 of the present invention;

[0088] Figure 5 This is a schematic diagram of an electronic device according to Embodiment 4 of the present invention;

[0089] Figure 6 This is a schematic diagram of a storage medium according to Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0090] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] Embodiment 1:

[0092] See also Figure 1 As shown, the pressure value and deformation intelligent feedback control system of a ceramic flattening device described in this embodiment includes a ceramic data acquisition module, a region division module, a flattening data acquisition module, a data analysis module, a ceramic flattening module and a feedback control module; each module is connected by wire and / or wireless means to realize data transmission between modules;

[0093] Ceramic data acquisition module, used to collect ceramic physical data of ceramics;

[0094] The ceramic is a ceramic to be flattened; the physical data of the ceramic includes strength data and elastic modulus;

[0095] Strength data include tensile strength, compressive strength and shear strength;

[0096] The tensile strength is the bearing capacity of ceramics under tension, the compressive strength is the bearing capacity of ceramics under load, and the shear strength is the bearing capacity of ceramics under shear;

[0097] The larger the strength data, the greater the force required to deform the ceramic, that is, when the ceramic is flattened, a larger pressure value needs to be applied to change the shape of the ceramic or adjust the flatness of the ceramic surface; on the contrary, the smaller the strength data, the greater the brittleness of the ceramic, and excessive pressure values ​​will cause the ceramic to break or be damaged, so a smaller pressure value needs to be applied to ensure that the ceramic will not be damaged during the flattening process;

[0098] Methods for obtaining intensity data include:

[0099] An RFID reader is used to read the RFID tag embedded in the ceramic surface to obtain the corresponding tag ID information, and the tag ID information is sent to the ceramic database to obtain a set of strength data corresponding to the tag ID information, wherein the RFID tag corresponds to the tag ID information one-to-one, and the tag ID information corresponds to the ceramic strength data one-to-one; the ceramic database records the strength data corresponding to the tag ID information of each ceramic, and when the ceramic is produced, the corresponding RFID tag is embedded in the surface of each ceramic;

[0100] The elastic modulus is the ease with which ceramics deform when subjected to force. The larger the elastic modulus, the greater the bonding force between molecules. When the ceramic is subjected to the applied pressure, the smaller the deformation degree is. Therefore, a larger pressure value is required to flatten the ceramic, and vice versa.

[0101] Methods for obtaining the elastic modulus include:

[0102] See also Figure 2 As shown, an ultrasonic device installed on the side of the ceramic flattening device is used, the ultrasonic device includes an ultrasonic transmitter and an ultrasonic receiver, the ultrasonic transmitter transmits ultrasonic pulses to the ceramic, and the ultrasonic receiver receives the ultrasonic pulses reflected from the inside of the ceramic; the angle value between the ultrasonic device and the ceramic is obtained, the angle value includes a first angle value and a second angle value, the first angle value is the angle between the ultrasonic transmitter and the ceramic surface, the second angle value is the angle between the ultrasonic receiver and the ceramic surface, and the first angle value and the second angle value are both measured and obtained by a person skilled in the art when installing the ultrasonic device;

[0103] Obtain a first distance value and a second distance value, wherein the first distance value is a distance value between the ultrasonic device and the surface of the ceramic, and the second distance value is a distance value between the ultrasonic device and the surface of the platform to be processed, wherein the platform to be processed is a platform on which the ceramic is placed in the ceramic lamination device; the first distance value and the second distance value are both obtained by a laser distance measuring sensor installed above the platform to be processed, the laser distance measuring sensor is perpendicular to the surface of the platform to be processed, and the laser distance measuring sensor and the ultrasonic device are located at the same horizontal height;

[0104] Use a laser scanner to scan the ceramic surface, obtain a scanned image of the ceramic surface, and mark the center point of the scanned image; take the center point as the starting point, measure the distance from the center point to any edge in the scanned image, and mark it as the first edge distance, and mark the edge corresponding to the first edge distance as the first edge; mark any edge corresponding to the first edge distance as the second edge, measure the distance from the center point to the second edge in the scanned image, and mark it as the second edge distance; calculate the ceramic area according to the first edge distance and the second edge distance; the expression of the ceramic area is: M = L'×L", where M is the ceramic area, L' is the first edge distance, and L" is the second edge distance;

[0105] The second distance value is subtracted from the first distance value to obtain the ceramic thickness, and the ceramic area is multiplied by the ceramic thickness to obtain the ceramic volume; the ceramic mass is obtained by a mass sensor installed in the platform to be processed, and the ceramic mass is divided by the ceramic volume to obtain the ceramic density;

[0106] The first distance value is calculated according to the first angle value and the first distance value. The first distance value is the distance value between the ultrasonic transmitter and the reflection point. The reflection point is the contact point between the laser pulse and the inside of the ceramic. The expression of the first distance value is: Where S1 is the first distance value, L1 is the first distance value, and θ1 is the first angle value. The second distance value is calculated according to the second angle value and the first distance value. The second distance value is the distance value between the ultrasonic receiver and the reflection point. The expression of the second distance value is: Where S2 is the second distance value, θ2 is the second angle value; the first distance value plus the second distance value is taken as the pulse distance, and the pulse distance is the distance traveled by the ultrasonic pulse from the ultrasonic transmitter to the ultrasonic receiver; the time data is collected, and the time data is the time elapsed when the ultrasonic pulse is emitted from the ultrasonic transmitter to the ultrasonic receiver, and the time data is obtained by the built-in timer of the ultrasonic device; the pulse distance is divided by the time data to obtain the elastic wave velocity; the elastic modulus is calculated according to the elastic wave velocity and the ceramic density, and the expression of the elastic modulus is: E = V t 2 ×ρ; where E is the elastic modulus, V t is the elastic wave velocity, ρ is the ceramic density;

[0107] A region division module, used for dividing the ceramic into m regions;

[0108] The m regions are divided according to the position of each pressure pad in the ceramic flattening device, wherein the pressure pads correspond to the regions one by one; the ceramic flattening device includes a plurality of pressure pads, and different forces can be applied to different pressure pads;

[0109] A flatness data acquisition module is used to collect flatness data corresponding to m areas in the ceramic;

[0110] The flatness data is the roughness of the ceramic surface. The larger the flatness data, the greater the roughness of the ceramic surface, the more uneven parts there are on the ceramic surface, and a larger pressure value is needed to press down the convex points to make the ceramic surface flat, and vice versa.

[0111] The methods for obtaining flatness data include:

[0112] A laser scanner is installed above the platform to be processed, the focus of the laser scanner is adjusted to the platform to be processed, and the laser beam is irradiated on the ceramic surface for scanning; n points are marked in each area of ​​the ceramic surface, where n is an integer greater than 1, and the height data of the n points corresponding to each area are measured; the mean value of the height data corresponding to each area is calculated, and the standard deviation of the height data of the n points corresponding to each area is calculated, and the standard deviation of the height data of the n points corresponding to each area is used as the flatness data of each area;

[0113] The expression of flatness data is:

[0114] In the formula, R i is the flatness data of the ith area, H ij is the height data of the jth point in the i-th area, is the mean value of the height data of the ith area, j∈[1,n], n is n points, i∈[1,m];

[0115] It should be noted that both the laser scanner and the laser range sensor are integrated into the ceramic applanation device;

[0116] A data analysis module is used to analyze the ceramic physical data and the flatness data corresponding to the m regions, and obtain the pressure values ​​required for the m regions during the ceramic flattening process;

[0117] The steps of obtaining the pressure values ​​required for m regions during the ceramic flattening process include:

[0118] Step 1: Randomly select an area that is not marked as a selected area from the m areas and mark it as a selected area;

[0119] Step 2: Encode the pressure value, obtain the chromosome, and construct the initial population;

[0120] Step 3: Determine the fitness function;

[0121] Step 4: Natural selection of chromosomes in the population;

[0122] Step 5: Perform crossover recombination on chromosomes in the population;

[0123] Step 6: mutate the chromosomes in the population;

[0124] Step 7: Obtain a new population, and preset the population generation number to be L, the fitness threshold to be Q, where L is an integer greater than 0, and Q is a real number greater than 0; loop steps 4 to 6 until the generation number corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is less than or equal to the fitness threshold Q, the loop ends, and the pressure value of the chromosome with the minimum fitness in the new population is obtained, and used as the pressure value corresponding to the selected region in step 1; illustratively, if the preset population generation number is 1, the chromosomes in the initial population are subjected to natural selection, crossover recombination and mutation to obtain a new population, and the generation number corresponding to the new population is 1 at this time, so the loop ends;

[0125] Step 8: looping steps 1 to 7 until all m regions are marked as selected regions, the loop ends, and the pressure values ​​required for the m regions in the ceramic flattening process are obtained;

[0126] It should be noted that the fitness threshold Q is preset by those skilled in the art according to the algorithm accuracy, and the population algebra L is obtained by those skilled in the art using a genetic algorithm multiple times under multiple sets of different ceramic physical data and flatness data conditions to obtain the corresponding pressure values. In each genetic algorithm process, when the fitness corresponding to the chromosome in the new population is less than or equal to the fitness threshold Q, the cycle ends and the algebra corresponding to the new population is obtained; the largest algebra among the multiple algebras is taken as the population algebra L;

[0127] In the above step 2, the pressure value is encoded as X, X is the chromosome, and the pressure value range [C1, C2] that the ceramic flattening device can generate is obtained. The range of X is [C1, C2], where the pressure value range is obtained from the technical specifications of the ceramic flattening device; N chromosomes are randomly generated to form an initial population Z0 = {X1, X2, ..., X N};

[0128] In step 3 above, the fitness function is expressed as: f q =R′ q ;

[0129] In the formula, f q is the fitness corresponding to the qth chromosome, R′ q is the evaluation flatness data corresponding to the qth chromosome, the evaluation flatness data is the roughness of the ceramic surface after flattening, q∈[1,N];

[0130] Methods for obtaining data for evaluating flatness include:

[0131] The ceramic physical data, the flatness data of a region and the pressure value corresponding to chromosome X are used as a set of analysis data; the analysis data are input into the trained flatness prediction model to predict the evaluation flatness data;

[0132] The specific training process of the flatness prediction model includes:

[0133] Collecting in advance the evaluation flatness data corresponding to a plurality of sets of analysis data, and converting the analysis data and the corresponding evaluation flatness data into a corresponding set of feature vectors;

[0134] Each set of feature vectors is used as the input of the flatness prediction model. The flatness prediction model uses a set of predicted and evaluated flatness data corresponding to each set of analysis data as output, and uses the actual evaluated flatness data corresponding to each set of analysis data as the prediction target. The actual evaluated flatness data is the pre-set evaluated flatness data corresponding to the analysis data. The training target is to minimize the sum of the prediction errors of all the analysis data. The calculation formula of the prediction error is Z p =(α p -μ p ) 2 , where Z p is the prediction error, p is the group number of the eigenvector corresponding to the analyzed data, α p is the predicted evaluation flatness data corresponding to the pth group of analysis data, μ p is the actual evaluation flatness data corresponding to the pth group of analysis data; the flatness prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped;

[0135] The above-mentioned flatness prediction model is specifically a deep neural network model;

[0136] It should be noted that the evaluation flatness data corresponding to the analysis data is collected by a technician in the field during the historical ceramic flattening process, and multiple sets of ceramic property data are collected. The ceramic property data includes ceramic physical data and flatness data. Under the conditions of each set of ceramic property data, the ceramic is flattened according to the pressure value range, and the evaluation flatness data of the ceramic corresponding to each set of analysis data after flattening is calculated, and used as the evaluation flatness data corresponding to each set of analysis data;

[0137] In the above step 4, natural selection is carried out by combining the elite method and the rotation method; the elite method produces k1 offspring chromosomes, and for a population with a capacity of N, the fitness corresponding to the N chromosomes is arranged from small to large, and the k1 chromosomes at the front each produce a offspring chromosome; the rotation method produces k2 offspring chromosomes, that is, N chromosomes produce k2 offspring chromosomes according to the corresponding rotation probability; k1+k2=N, so as to keep the offspring population capacity N unchanged and the population generation number increases;

[0138] The expression of rotation probability is:

[0139] Where P q is the rotation probability corresponding to the qth chromosome;

[0140] In the above step 5, U chromosomes are randomly selected from the population for crossover recombination to obtain U new chromosomes; the crossover recombination adopts the PMX method, which is a prior art method and will not be described in detail here; after the chromosome crossover recombination, the fitness of the U new chromosomes is calculated, and the fitness of the U new chromosomes and the fitness of the U chromosomes are sorted from small to large to generate a sorting table, and the U chromosomes in the sorting table are replaced with the U chromosomes for crossover recombination in the population in positive order; in this embodiment, U=0.7N is preferred. If the calculated U is not an integer, U is rounded up to ensure that the calculated U is an integer;

[0141] In the above step 6, the mutation probability is preset to G, and the N chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes. The mutation probability is preset by those skilled in the art according to the algorithm efficiency and algorithm accuracy. In this embodiment, G is preferably 0.02;

[0142] The ceramic flattening module is used to control the ceramic flattening equipment to flatten the ceramic according to the pressure values ​​required by the m areas, and collect the deformation data of the m areas during the ceramic flattening process and provide real-time feedback;

[0143] The deformation data is the change in the roughness of the ceramic surface; the collection interval is preset, and the deformation data is collected according to the collection interval; the collection interval is preset by a technician in the field according to the actual situation to ensure that the deformation data can be accurately collected, and the deformation data will not be unable to be collected due to the collection interval being too small;

[0144] Methods for obtaining deformation data include:

[0145] An elastic film is added between the ceramic and the pressure pad of the ceramic flattening device, and M flexible strain gauges are embedded in the elastic film, M=m, and the flexible strain gauges correspond to the regions one by one to obtain the strain data of the m regions; a proportional coefficient is preset, and the deformation data is obtained according to the product of the strain data and the proportional coefficient; the expression of the deformation data is: BX i =ψ×YB i ; In the formula, BX i is the deformation data of the ith region, ψ is the proportional coefficient, YB i is the strain data of the ith region;

[0146] It should be understood that the proportionality coefficient is obtained by a technician in the field during the historical ceramic flattening process, collecting multiple strain data according to the collection interval, and collecting the first flatness data and the second flatness data corresponding to each strain data, subtracting the second flatness data from the first flatness data to obtain the deformation data, wherein the first flatness data is the flatness data of the ceramic when the strain data was collected last time, and the second flatness data is the flatness data of the ceramic when the strain data is collected this time; each deformation data is divided by the corresponding strain data to obtain a ratio, and the average of the multiple ratios is used as the proportionality coefficient;

[0147] It should be noted that the reason for using a flexible strain gauge is that most ordinary strain gauges are rigid structures. After being embedded in an elastic film, the ordinary strain gauge will be damaged due to the pressure during the ceramic flattening process. Therefore, a flexible strain gauge is needed to ensure that the strain gauge will not be damaged while collecting deformation data.

[0148] A feedback control module is used to determine whether to generate an adjustment instruction based on the deformation data of the m areas fed back in real time, mark the area where the adjustment instruction is generated as the adjustment area, analyze the deformation data of the adjustment area, calculate the adjustment pressure, and control the pressure value of the adjustment area in the ceramic flattening device to be adjusted to the adjustment pressure;

[0149] The method for determining whether to generate an adjustment instruction includes:

[0150] Calculate the deformation threshold TH of m regions i , the deformation data BX of m regions i Respectively with the corresponding deformation threshold TH i Make a comparison;

[0151] If BX i ≥TH i , no adjustment instruction is generated, indicating that the current pressure of the corresponding area meets the pressure requirement of the area, and there is no need to adjust the pressure value;

[0152] If BX i <TH i , an adjustment instruction is generated, indicating that the current pressure of the corresponding area cannot meet the pressure demand of the area, and the pressure value needs to be increased to improve the flattening effect of the area;

[0153] The calculation method of deformation threshold includes:

[0154] TH i =ω1×R i +ω2×P i ′+ω3×t;

[0155] Where TH i is the deformation threshold of the ith region, Pi ′ is the pressure value of the ith area, t is the sampling interval, ω1, ω2, ω3 are the preset weight coefficients;

[0156] The specific value of the weight coefficient in the formula can be set according to the actual situation. The weight coefficient reflects the influence of the flatness data, the pressure value and the acquisition interval on the deformation threshold. Those skilled in the art can preset the corresponding weight coefficient according to the actual influence of the flatness data, the pressure value and the acquisition interval on the deformation threshold, so as to accurately evaluate the deformation threshold of different areas;

[0157] It should be noted that the flatness data, pressure value and acquisition interval are all influencing factors of the deformation threshold; the larger the flatness data, the greater the roughness of the ceramic surface, and the larger the deformation data during the flattening operation, that is, the larger the deformation threshold, and vice versa; the larger the pressure value, the greater the pressure acting on the ceramic surface during the flattening operation, resulting in larger deformation data of the ceramic, that is, the larger the deformation threshold, and vice versa; the larger the acquisition interval, the longer the time for flattening the ceramic during the flattening operation, still resulting in larger deformation data of the ceramic, that is, the larger the deformation threshold, and vice versa; since the deformation threshold is only a value for judging whether to generate an adjustment instruction, the calculation of the deformation threshold is a dimensionless calculation;

[0158] Methods for calculating regulated pressure include:

[0159] Subtract the corresponding deformation data from the deformation threshold of the adjustment area to obtain the difference, add the difference to the deformation threshold of the adjustment area to obtain the deformation adjustment data, and calculate the adjustment pressure according to the deformation adjustment data, the acquisition interval and the flatness data; the expression of the adjustment pressure is: In the formula, P″ h is the regulating pressure of the hth regulating area, TH′ h is the deformation adjustment data of the hth adjustment area, R h is the flatness data of the hth adjustment area, h∈[1,h′], h′ is the number of adjustment areas, 0<h′≤m;

[0160] It should be noted that after long-term use, the actual pressure value generated by the ceramic flattening equipment may be less than the set pressure value. After the ceramic flattening is completed, the flatness data of the ceramic may not meet the requirements. Therefore, it is necessary to detect the deformation in real time during the ceramic flattening process and adjust the pressure value accordingly to ensure that after the ceramic flattening is completed, the flatness data of the ceramic can meet the set requirements.

[0161] This embodiment collects ceramic physical data and flatness data, divides the ceramic into multiple areas, calculates the pressure values ​​required for different areas in real time, and automatically controls the ceramic flattening equipment to flatten according to the pressure values ​​required for different areas, thereby realizing individualized control of the pressure value; at the same time, it can monitor the deformation of the ceramic during the flattening process in real time, and collect the corresponding deformation data, and judge whether the pressure value needs to be adjusted according to the deformation data of each area, and realize fine control of the pressure value by independently calculating and adjusting the pressure value of the adjustment area; it can effectively reduce the uneven force on the ceramic, thereby improving the flattening quality and precision of the ceramic; it realizes the automated operation in the ceramic flattening process, effectively improves the production efficiency, and lays the foundation for intelligent ceramic manufacturing.

[0162] Embodiment 2:

[0163] See also Figure 3 As shown, this embodiment is further improved on the basis of the embodiment 1. Since the flatness data of the ceramic is relatively large at the beginning of the ceramic flattening, the deformation data in the collection interval is relatively large, and when the ceramic flattening is about to end, the flatness data of the ceramic is relatively small, so the deformation data in the collection interval is also relatively small. If a fixed deformation threshold is used in the ceramic flattening process, the adjustment area cannot be effectively determined, resulting in an erroneous pressure value adjustment of the adjustment area; therefore, this embodiment provides a pressure and deformation intelligent feedback control system for a ceramic flattening device, which also includes:

[0164] The threshold adjustment module is used to collect the flattening time and re-determine whether to generate an adjustment instruction;

[0165] The flattening time is the operating time of the ceramic flattening equipment during one operation;

[0166] The flattening time is obtained by the built-in counter of the ceramic flattening device. When the ceramic flattening device starts to run, the timer starts to count, and when the ceramic flattening device stops running, the timer stops counting and is reset to zero.

[0167] Methods for re-determining whether to generate an adjustment instruction include:

[0168] The ceramic flattening process is divided into W sections according to the collection interval, and the duration of each section is the collection interval;

[0169] The W segments are sequentially and incrementally assigned digital labels, and the digital label range is [1, W], where the digital label corresponding to the segment at the beginning of the ceramic flattening process is 1, and the digital label corresponding to the segment near the end of the ceramic flattening process is W;

[0170] The digital label corresponding to the segment is marked as the segment label, the segment label and the deformation threshold are used as evaluation data, and the W′ group of evaluation data corresponding to each region is sequentially input into the trained influence analysis model to predict the corresponding influence coefficient, where W′=W;

[0171] The specific training process of the impact analysis model includes:

[0172] Collecting influence coefficients corresponding to multiple groups of evaluation data in advance, and converting the evaluation data and the corresponding influence coefficients into a corresponding set of feature vectors;

[0173] Each set of feature vectors is used as the input of the impact analysis model. The impact analysis model uses a set of predicted impact coefficients corresponding to each set of evaluation data as output, and the actual impact coefficients corresponding to each set of evaluation data as the prediction target. The actual evaluation flatness data is the pre-set impact coefficient corresponding to the evaluation data; the training target is to minimize the sum of the prediction deviations of all evaluation data; wherein the calculation formula for the prediction deviation is Z y =(α y -μ y ) 2 , where Z y is the prediction deviation, y is the group number of the feature vector corresponding to the evaluation data, α y is the predicted impact coefficient corresponding to the y-th group of evaluation data, μ y is the actual impact coefficient corresponding to the yth group of evaluation data; the impact analysis model is trained until the sum of the prediction deviations reaches convergence and the training is stopped;

[0174] The above-mentioned impact analysis model is specifically a deep neural network model;

[0175] It should be noted that the influence coefficient corresponding to the evaluation data is calculated by a technician in the field during the historical ceramic flattening process, and multiple groups of deformation thresholds are calculated. Under the condition of multiple groups of deformation thresholds, the deformation data of different sections of the ceramic corresponding to each group of deformation thresholds during the flattening process are collected, and the deformation data of different sections are divided by the corresponding deformation threshold to obtain the influence coefficient, which is used as the influence coefficient corresponding to the evaluation data;

[0176] Multiply the obtained F influence coefficients by the deformation threshold of the corresponding area to obtain new deformation thresholds corresponding to W segments in the m areas, where F = W″×m, W″ is the number of influence coefficients of each area and W″=W′;

[0177] The flattening time corresponding to the deformation data of m regions is divided by the collection interval to obtain the time ratio. The integer part of the time ratio is the corresponding segment label. According to the segment label, the new deformation threshold corresponding to the m regions is obtained. The deformation data of the m regions are compared with the corresponding new deformation threshold. If No adjustment instruction is generated if Generate adjustment instructions, is the new deformation threshold corresponding to the i-th region;

[0178] This embodiment divides the ceramic flattening process into multiple time zones and sets the influence coefficient corresponding to each section; it can dynamically adjust the deformation thresholds of different time stages to accurately reflect the different deformation standards of ceramics in different flattening stages; eliminate the misjudgment problem that may be caused by using a fixed deformation threshold, more accurately identify the area where pressure adjustment is required, and improve the accuracy of the adjusted pressure value; realize the learning and simulation of the deformation law of the entire flattening process, enhance the intelligence level of deformation threshold adjustment, thereby improving the pressure control quality and the ceramic flattening effect.

[0179] Embodiment 3:

[0180] See also Figure 4 As shown, the part not described in detail in this embodiment can be found in the description of embodiments 1 and 2, and a method for intelligent feedback control of pressure and deformation of a ceramic flattening device is provided, the method comprising:

[0181] Collect ceramic physical data of ceramics;

[0182] Divide the ceramic into m regions;

[0183] Collecting the flatness data corresponding to m areas in the ceramic;

[0184] Analyze the ceramic physical data and the flatness data corresponding to the m regions to obtain the pressure values ​​required for the m regions during the ceramic flattening process;

[0185] According to the pressure values ​​required by the m regions, the ceramic flattening equipment is controlled to flatten the ceramics, and the deformation data of the m regions are collected and fed back in real time during the ceramic flattening process;

[0186] Based on the deformation data of the m areas fed back in real time, determine whether to generate an adjustment instruction, mark the area where the adjustment instruction is generated as the adjustment area, analyze the deformation data of the adjustment area, calculate the adjustment pressure, and control the pressure value of the adjustment area in the ceramic flattening equipment to be adjusted to the adjustment pressure.

[0187] Furthermore, the ceramic is a ceramic to be flattened; the physical data of the ceramic includes strength data and elastic modulus;

[0188] The strength data include tensile strength, compressive strength and shear strength;

[0189] The tensile strength is the bearing capacity of the ceramic under tension, the compressive strength is the bearing capacity of the ceramic under load, and the shear strength is the bearing capacity of the ceramic under shear;

[0190] The elastic modulus is the ease with which a ceramic can deform when subjected to force.

[0191] The method for obtaining the intensity data comprises:

[0192] An RFID reader is used to read the RFID tag embedded in the ceramic surface to obtain the corresponding tag ID information, and the tag ID information is sent to the ceramic database to obtain a set of strength data corresponding to the tag ID information, wherein the RFID tag corresponds to the tag ID information one-to-one, and the tag ID information corresponds to the ceramic strength data one-to-one; the ceramic database records the strength data corresponding to the tag ID information of each ceramic.

[0193] Furthermore, the method for obtaining the elastic modulus includes:

[0194] Using an ultrasonic device, the ultrasonic device includes an ultrasonic transmitter and an ultrasonic receiver, the ultrasonic transmitter transmits an ultrasonic pulse to the ceramic, and the ultrasonic receiver receives the ultrasonic pulse reflected from the inside of the ceramic; obtaining an angle value between the ultrasonic device and the ceramic, the angle value including a first angle value and a second angle value, the first angle value being the angle between the ultrasonic transmitter and the ceramic surface, and the second angle value being the angle between the ultrasonic receiver and the ceramic surface;

[0195] Obtaining a first distance value and a second distance value, wherein the first distance value is the distance value between the ultrasonic device and the ceramic surface, and the second distance value is the distance value between the ultrasonic device and the surface of the platform to be processed, wherein the platform to be processed is the platform on which the ceramic is placed in the ceramic lamination device;

[0196] Use a laser scanner to scan the ceramic surface, obtain a scanned image of the ceramic surface, and mark the center point of the scanned image; take the center point as the starting point, measure the distance from the center point to any edge in the scanned image, and mark it as the first edge distance, and mark the edge corresponding to the first edge distance as the first edge; mark any edge corresponding to the first edge distance as the second edge, measure the distance from the center point to the second edge in the scanned image, and mark it as the second edge distance; calculate the ceramic area according to the first edge distance and the second edge distance; the expression of the ceramic area is: M = L'×L", where M is the ceramic area, L' is the first edge distance, and L" is the second edge distance;

[0197] The second distance value is subtracted from the first distance value to obtain the ceramic thickness, and the ceramic area is multiplied by the ceramic thickness to obtain the ceramic volume; the ceramic mass is obtained, and the ceramic mass is divided by the ceramic volume to obtain the ceramic density;

[0198] The first distance value is calculated according to the first angle value and the first distance value. The first distance value is the distance value between the ultrasonic transmitter and the reflection point. The reflection point is the contact point between the laser pulse and the inside of the ceramic. The expression of the first distance value is: Where S1 is the first distance value, L1 is the first distance value, and θ1 is the first angle value. The second distance value is calculated according to the second angle value and the first distance value. The second distance value is the distance value between the ultrasonic receiver and the reflection point. The expression of the second distance value is: Where S2 is the second distance value, θ2 is the second angle value; the first distance value plus the second distance value is taken as the pulse distance, and the pulse distance is the distance traveled by the ultrasonic pulse from the ultrasonic transmitter to the ultrasonic receiver; the time data is collected, and the time data is the time traveled by the ultrasonic pulse from the ultrasonic transmitter to the ultrasonic receiver; the pulse distance is divided by the time data to obtain the elastic wave velocity; the elastic modulus is calculated according to the elastic wave velocity and the ceramic density, and the expression of the elastic modulus is: E = V t 2 ×ρ; where E is the elastic modulus, V t is the elastic wave velocity, and ρ is the ceramic density.

[0199] Furthermore, the m regions are divided according to the position of each pressure pad in the ceramic flattening device, wherein the pressure pad corresponds to the region one by one; the flatness data is the roughness of the ceramic surface; and the method for obtaining the flatness data includes:

[0200] Use a laser scanner to irradiate a laser beam on the ceramic surface for scanning; mark n points in each area of ​​the ceramic surface, where n is an integer greater than 1, and measure the height data of the n points corresponding to each area; calculate the mean value of the height data corresponding to each area, and calculate the standard deviation of the height data of the n points corresponding to each area, and use the standard deviation of the height data of the n points corresponding to each area as the flatness data of each area;

[0201] The expression of flatness data is:

[0202] In the formula, R i is the flatness data of the ith area, H ij is the height data of the jth point in the i-th area, is the mean value of the height data of the ith area, j∈[1,n], n is the number of points, i∈[1,m].

[0203] Furthermore, the step of obtaining the pressure values ​​required for m regions during the ceramic flattening process includes:

[0204] Step 1: Randomly select an area that is not marked as a selected area from the m areas and mark it as a selected area;

[0205] Step 2: Encode the pressure value, obtain the chromosome, and construct the initial population;

[0206] Step 3: Determine the fitness function;

[0207] Step 4: Natural selection of chromosomes in the population;

[0208] Step 5: Perform crossover recombination on chromosomes in the population;

[0209] Step 6: mutate the chromosomes in the population;

[0210] Step 7: Get a new population, preset the population generation number to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop steps 4 to 6 until the generation number corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is less than or equal to the fitness threshold Q, the loop ends, and the pressure value of the chromosome with the minimum fitness in the new population is obtained, and used as the pressure value corresponding to the selected region in step 1;

[0211] Step 8: Loop steps 1 to 7 until all m regions are marked as selected regions. The loop ends and the pressure values ​​required for the m regions in the ceramic flattening process are obtained.

[0212] Furthermore, in step 2, the pressure value is encoded as X, where X is a chromosome, and the pressure value range [C1, C2] that can be generated by the ceramic flattening device is obtained, where the range of X is [C1, C2]; N chromosomes are randomly generated to form an initial population Z0 = {X1, X2, ..., X N};

[0213] In step 3, the fitness function is expressed as: q =R′ q ;

[0214] In the formula, f q is the fitness corresponding to the qth chromosome, R′ q is the evaluation flatness data corresponding to the qth chromosome, the evaluation flatness data is the roughness of the ceramic surface after flattening, q∈[1,N];

[0215] Methods for obtaining data for evaluating flatness include:

[0216] The ceramic physical data, the flatness data of a region and the pressure value corresponding to chromosome X are used as a set of analysis data; the analysis data are input into the trained flatness prediction model to predict the evaluation flatness data;

[0217] The training process of the flatness prediction model includes:

[0218] Collecting in advance the evaluation flatness data corresponding to a plurality of sets of analysis data, and converting the analysis data and the corresponding evaluation flatness data into a corresponding set of feature vectors;

[0219] Each group of feature vectors is used as the input of the flatness prediction model. The flatness prediction model uses a group of predicted and evaluated flatness data corresponding to each group of analysis data as output, and uses the actual evaluated flatness data corresponding to each group of analysis data as the prediction target, where the actual evaluated flatness data is the pre-set evaluated flatness data corresponding to the analysis data; minimizing the sum of the prediction errors of all the analysis data is used as the training target; the flatness prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped; the flatness prediction model is a deep neural network model;

[0220] In step 4, natural selection is performed by combining the elite method and the rotation method; the elite method generates k1 offspring chromosomes, and for a population with a capacity of N, the fitness corresponding to the N chromosomes is arranged from small to large, and each of the k1 chromosomes at the front generates a offspring chromosome; the rotation method generates k2 offspring chromosomes, that is, N chromosomes generate k2 offspring chromosomes according to the corresponding rotation probability; k1+k2=N;

[0221] The expression of rotation probability is:

[0222] Where P q is the rotation probability corresponding to the qth chromosome;

[0223] In the step 5, U chromosomes are randomly selected from the population for crossover recombination to obtain U new chromosomes; the crossover recombination adopts the PMX method; after the chromosome crossover recombination, the fitness of the U new chromosomes is calculated, the fitness of the U new chromosomes and the fitness of the U chromosomes are sorted from small to large and a sorting table is generated, and the U chromosomes in the sorting table are replaced with the U chromosomes that are crossover recombined in the population in positive order;

[0224] In step 6, the mutation probability is preset to G, and the N chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes.

[0225] Furthermore, the deformation data is the variation of the roughness of the ceramic surface; a collection interval is preset, and the deformation data is collected according to the collection interval;

[0226] The method for acquiring the deformation data comprises:

[0227] An elastic film is added between the ceramic and the pressure pad of the ceramic flattening device, and M flexible strain gauges are embedded in the elastic film, M=m, and the flexible strain gauges correspond to the regions one by one to obtain the strain data of the m regions; a proportional coefficient is preset, and the deformation data is obtained according to the product of the strain data and the proportional coefficient; the expression of the deformation data is: BX i =ψ×YBi ; In the formula, BX i is the deformation data of the ith region, ψ is the proportional coefficient, YB i is the strain data of the ith region.

[0228] Furthermore, the method for determining whether to generate an adjustment instruction includes:

[0229] Calculate the deformation threshold TH of m regions i , the deformation data BX of m regions i Respectively with the corresponding deformation threshold TH i Make a comparison;

[0230] If BX i ≥TH i , no adjustment instruction is generated;

[0231] If BX i <TH i , then generate adjustment instructions;

[0232] The calculation method of deformation threshold includes:

[0233] TH i =ω1×R i +ω2×P i ′+ω3×t;

[0234] Where TH i is the deformation threshold of the ith region, P i ′ is the pressure value of the ith area, t is the sampling interval, ω1, ω2, ω3 are the preset weight coefficients;

[0235] Methods for calculating regulated pressure include:

[0236] Subtract the corresponding deformation data from the deformation threshold of the adjustment area to obtain the difference, add the difference to the deformation threshold of the adjustment area to obtain the deformation adjustment data, and calculate the adjustment pressure according to the deformation adjustment data, the acquisition interval and the flatness data; the expression of the adjustment pressure is: Where P h ″ is the regulating pressure of the hth regulating area, TH′ h is the deformation adjustment data of the hth adjustment area, R h is the flatness data of the hth adjustment area, h∈[1,h′], h′ is the number of adjustment areas, 0<h′≤m.

[0237] Furthermore, the flattening time is collected to re-determine whether to generate an adjustment instruction;

[0238] The flattening time is the operating time of the ceramic flattening equipment during one operation;

[0239] Methods for re-determining whether to generate an adjustment instruction include:

[0240] The ceramic flattening process is divided into W sections according to the collection interval, and the duration of each section is the collection interval;

[0241] Set digital labels for W segments in increasing order, with the digital label range being [1, W];

[0242] The digital label corresponding to the segment is marked as the segment label, the segment label and the deformation threshold are used as evaluation data, and the W′ group of evaluation data corresponding to each region is sequentially input into the trained influence analysis model to predict the corresponding influence coefficient, where W′=W;

[0243] The training process of the impact analysis model includes:

[0244] Collecting influence coefficients corresponding to multiple groups of evaluation data in advance, and converting the evaluation data and the corresponding influence coefficients into a corresponding set of feature vectors;

[0245] Each group of feature vectors is used as the input of the impact analysis model. The impact analysis model uses a group of predicted impact coefficients corresponding to each group of evaluation data as output, and uses the actual impact coefficients corresponding to each group of evaluation data as the prediction target. The actual evaluation flatness data is the pre-set impact coefficient corresponding to the evaluation data; minimizing the sum of the prediction deviations of all evaluation data is used as the training target; the impact analysis model is trained until the sum of the prediction deviations reaches convergence and the training is stopped; the impact analysis model is a deep neural network model;

[0246] Multiply the obtained F influence coefficients by the deformation threshold of the corresponding area to obtain new deformation thresholds corresponding to W segments in the m areas, where F = W″×m, W″ is the number of influence coefficients of each area and W″=W′;

[0247] The flattening time corresponding to the deformation data of m regions is divided by the collection interval to obtain the time ratio. The integer part of the time ratio is the corresponding segment label. According to the segment label, the new deformation threshold corresponding to the m regions is obtained. The deformation data of the m regions are compared with the corresponding new deformation threshold. If No adjustment instruction is generated if Generate adjustment instructions, is the new deformation threshold corresponding to the i-th region.

[0248] Embodiment 4:

[0249] See also Figure 5As shown, according to another aspect of the present application, an electronic device 500 is also provided. The electronic device 500 may include one or more processors and one or more memories. The memories store computer readable codes, and when the computer readable codes are run by one or more processors, the pressure and deformation intelligent feedback control method of a ceramic flattening device as described above can be executed.

[0250] The method or system according to the embodiment of the present application can also be used by Figure 5 The electronic device architecture shown in FIG. Figure 5 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, ROM 503, RAM 504, a communication port 505 connected to a network, an input / output 506, a hard disk 507, etc. The storage device in the electronic device 500, such as ROM 503 or hard disk 507, may store a pressure and deformation intelligent feedback control method of a ceramic flattening device provided in the present application. Furthermore, the electronic device 500 may also include a user interface 508. Of course, Figure 5 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 5 One or more components of an electronic device are shown.

[0251] Embodiment 5:

[0252] See also Figure 6 As shown, a computer-readable storage medium 600 according to an embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are executed by the processor, a pressure and deformation intelligent feedback control method of a ceramic flattening device according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0253] In addition, according to the embodiments of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: a pressure and deformation intelligent feedback control method for a ceramic flattening device. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0254] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0255] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for intelligent feedback control of pressure and deformation of a ceramic flattening device, characterized in that: include: Collect ceramic physical data of ceramics; Divide the ceramic into m regions; Collecting the flatness data corresponding to m areas in the ceramic; Analyze the ceramic physical data and the flatness data corresponding to the m regions to obtain the pressure values ​​required for the m regions during the ceramic flattening process; According to the pressure values ​​required by the m regions, the ceramic flattening equipment is controlled to flatten the ceramics, and the deformation data of the m regions are collected and fed back in real time during the ceramic flattening process; According to the deformation data of the m regions fed back in real time, it is determined whether to generate an adjustment instruction, the region where the adjustment instruction is generated is marked as an adjustment region, the deformation data of the adjustment region is analyzed, the adjustment pressure is calculated, and the pressure value of the adjustment region in the ceramic flattening device is controlled to be adjusted to the adjustment pressure; The m regions are divided according to the position of each pressure pad in the ceramic flattening device, wherein the pressure pad corresponds to the region one by one; the flatness data is the roughness of the ceramic surface; and the method for obtaining the flatness data includes: Use a laser scanner to irradiate a laser beam on the ceramic surface for scanning; mark n points in each area of ​​the ceramic surface, where n is an integer greater than 1, and measure the height data of the n points corresponding to each area; calculate the mean value of the height data corresponding to each area, and calculate the standard deviation of the height data of the n points corresponding to each area, and use the standard deviation of the height data of the n points corresponding to each area as the flatness data of each area; The expression of flatness data is: In the formula, R i is the flatness data of the ith area, H ij is the height data of the jth point in the i-th area, is the mean value of the height data of the ith area, j∈[1,n], n is n points, i∈[1,m]; The step of obtaining the pressure values ​​required for m regions during the ceramic flattening process comprises: Step 1: Randomly select an area that is not marked as a selected area from the m areas and mark it as a selected area; Step 2: Encode the pressure value, obtain the chromosome, and construct the initial population; Step 3: Determine the fitness function; Step 4: Natural selection of chromosomes in the population; Step 5: Perform crossover recombination on chromosomes in the population; Step 6: mutate the chromosomes in the population; Step 7: Get a new population, preset the population generation number to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop steps 4 to 6 until the generation number corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is less than or equal to the fitness threshold Q, the loop ends, and the pressure value of the chromosome with the minimum fitness in the new population is obtained, and used as the pressure value corresponding to the selected region in step 1; Step 8: looping steps 1 to 7 until all m regions are marked as selected regions, the loop ends, and the pressure values ​​required for the m regions in the ceramic flattening process are obtained; In step 2, the pressure value is encoded as X, where X is the chromosome, and the pressure value range [C1, C2] that can be generated by the ceramic flattening device is obtained, where the range of X is [C1, C2]; N chromosomes are randomly generated to form an initial population Z0 = {X1, X2, ..., X N }; In step 3, the fitness function is expressed as: q =R′ q ; In the formula, f q is the fitness corresponding to the qth chromosome, R′ q is the evaluation flatness data corresponding to the qth chromosome, the evaluation flatness data is the roughness of the ceramic surface after flattening, q∈[1,N]; Methods for obtaining data for evaluating flatness include: The ceramic physical data, the flatness data of a region and the pressure value corresponding to chromosome X are used as a set of analysis data; the analysis data are input into the trained flatness prediction model to predict the evaluation flatness data; The training process of the flatness prediction model includes: Collecting in advance the evaluation flatness data corresponding to a plurality of sets of analysis data, and converting the analysis data and the corresponding evaluation flatness data into a corresponding set of feature vectors; Each group of feature vectors is used as the input of the flatness prediction model. The flatness prediction model uses a group of predicted and evaluated flatness data corresponding to each group of analysis data as output, and uses the actual evaluated flatness data corresponding to each group of analysis data as the prediction target, where the actual evaluated flatness data is the pre-set evaluated flatness data corresponding to the analysis data; minimizing the sum of the prediction errors of all the analysis data is used as the training target; the flatness prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped; the flatness prediction model is a deep neural network model; In step 4, natural selection is performed by combining the elite method and the rotation method; the elite method generates k1 offspring chromosomes, and for a population with a capacity of N, the fitness corresponding to the N chromosomes is arranged from small to large, and each of the k1 chromosomes at the front generates a offspring chromosome; the rotation method generates k2 offspring chromosomes, that is, N chromosomes generate k2 offspring chromosomes according to the corresponding rotation probability; k1+k2=N; The expression of rotation probability is: Where P q is the rotation probability corresponding to the qth chromosome; In the step 5, U chromosomes are randomly selected from the population for crossover recombination to obtain U new chromosomes; the crossover recombination adopts the PMX method; after the chromosome crossover recombination, the fitness of the U new chromosomes is calculated, the fitness of the U new chromosomes and the fitness of the U chromosomes are sorted from small to large and a sorting table is generated, and the U chromosomes in the sorting table are replaced with the U chromosomes that are crossover recombined in the population in positive order; In step 6, the mutation probability is preset to G, and the N chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes; The deformation data is the variation of the roughness of the ceramic surface; a collection interval is preset, and the deformation data is collected according to the collection interval; The method for acquiring the deformation data comprises: An elastic film is added between the ceramic and the pressure pad of the ceramic flattening device, and M flexible strain gauges are embedded in the elastic film, M=m, and the flexible strain gauges correspond to the regions one by one to obtain the strain data of the m regions; a proportional coefficient is preset, and the deformation data is obtained according to the product of the strain data and the proportional coefficient; the expression of the deformation data is: BX i =ψ×YB i ; In the formula, BX i is the deformation data of the ith region, ψ is the proportional coefficient, YB i is the strain data of the ith region; The method for determining whether to generate an adjustment instruction includes: Calculate the deformation threshold TH of m regions i , the deformation data BX of m regions i Respectively with the corresponding deformation threshold TH i Make a comparison; If BX i ≥TH i , no adjustment instruction is generated; If BX i <TH i , then generate adjustment instructions; The calculation method of deformation threshold includes: TH i =ω1×R i +ω2×P i ′+ω3×t; Where TH i is the deformation threshold of the ith region, P i ′ is the pressure value of the ith area, t is the sampling interval, ω1, ω2, ω3 are the preset weight coefficients; Methods for calculating regulated pressure include: Subtract the corresponding deformation data from the deformation threshold of the adjustment area to obtain the difference, add the difference to the deformation threshold of the adjustment area to obtain the deformation adjustment data, and calculate the adjustment pressure according to the deformation adjustment data, the acquisition interval and the flatness data; the expression of the adjustment pressure is: Where P h ″ is the regulating pressure of the hth regulating area, TH′ h is the deformation adjustment data of the hth adjustment area, R h is the flatness data of the hth adjustment area, h∈[1,h′], h′ is the number of adjustment areas, 0<h′≤m.

2. The method for intelligent feedback control of pressure and deformation of ceramic flattening equipment according to claim 1, characterized in that: The ceramic is a ceramic to be flattened; the physical data of the ceramic includes strength data and elastic modulus; The strength data include tensile strength, compressive strength and shear strength; The tensile strength is the bearing capacity of the ceramic under tension, the compressive strength is the bearing capacity of the ceramic under load, and the shear strength is the bearing capacity of the ceramic under shear; The elastic modulus is the ease with which a ceramic deforms when subjected to force; The method for obtaining the intensity data comprises: An RFID reader is used to read the RFID tag embedded in the ceramic surface to obtain the corresponding tag ID information, and the tag ID information is sent to the ceramic database to obtain a set of strength data corresponding to the tag ID information, wherein the RFID tag corresponds to the tag ID information one-to-one, and the tag ID information corresponds to the ceramic strength data one-to-one; the ceramic database records the strength data corresponding to the tag ID information of each ceramic.

3. The method for intelligent feedback control of pressure and deformation of ceramic flattening equipment according to claim 2, characterized in that: The method for obtaining the elastic modulus comprises: Using an ultrasonic device, the ultrasonic device includes an ultrasonic transmitter and an ultrasonic receiver, the ultrasonic transmitter transmits an ultrasonic pulse to the ceramic, and the ultrasonic receiver receives the ultrasonic pulse reflected from the inside of the ceramic; obtaining an angle value between the ultrasonic device and the ceramic, the angle value including a first angle value and a second angle value, the first angle value being the angle between the ultrasonic transmitter and the ceramic surface, and the second angle value being the angle between the ultrasonic receiver and the ceramic surface; Obtaining a first distance value and a second distance value, wherein the first distance value is the distance value between the ultrasonic device and the ceramic surface, and the second distance value is the distance value between the ultrasonic device and the surface of the platform to be processed, wherein the platform to be processed is the platform on which the ceramic is placed in the ceramic lamination device; Use a laser scanner to scan the ceramic surface, obtain a scanned image of the ceramic surface, and mark the center point of the scanned image; take the center point as the starting point, measure the distance from the center point to any edge in the scanned image, and mark it as the first edge distance, and mark the edge corresponding to the first edge distance as the first edge; mark any edge corresponding to the first edge distance as the second edge, measure the distance from the center point to the second edge in the scanned image, and mark it as the second edge distance; calculate the ceramic area according to the first edge distance and the second edge distance; the expression of the ceramic area is: M = L'×L", where M is the ceramic area, L' is the first edge distance, and L" is the second edge distance; The second distance value is subtracted from the first distance value to obtain the ceramic thickness, and the ceramic area is multiplied by the ceramic thickness to obtain the ceramic volume; the ceramic mass is obtained, and the ceramic mass is divided by the ceramic volume to obtain the ceramic density; The first distance value is calculated according to the first angle value and the first distance value. The first distance value is the distance value between the ultrasonic transmitter and the reflection point. The reflection point is the contact point between the laser pulse and the inside of the ceramic. The expression of the first distance value is: Where S1 is the first distance value, L1 is the first distance value, and θ1 is the first angle value. The second distance value is calculated according to the second angle value and the first distance value. The second distance value is the distance value between the ultrasonic receiver and the reflection point. The expression of the second distance value is: Where S2 is the second distance value, θ2 is the second angle value; the first distance value plus the second distance value is taken as the pulse distance, and the pulse distance is the distance traveled by the ultrasonic pulse from the ultrasonic transmitter to the ultrasonic receiver; the time data is collected, and the time data is the time traveled by the ultrasonic pulse from the ultrasonic transmitter to the ultrasonic receiver; the pulse distance is divided by the time data to obtain the elastic wave velocity; the elastic modulus is calculated according to the elastic wave velocity and the ceramic density, and the expression of the elastic modulus is: E = V t 2 ×ρ; where E is the elastic modulus, V t is the elastic wave velocity, and ρ is the ceramic density.

4. The method for intelligent feedback control of pressure and deformation of ceramic flattening equipment according to claim 3, characterized in that: Collect the flattening time and re-determine whether to generate an adjustment instruction; The flattening time is the operating time of the ceramic flattening equipment during one operation; Methods for re-determining whether to generate an adjustment instruction include: The ceramic flattening process is divided into W sections according to the collection interval, and the duration of each section is the collection interval; Set digital labels for W segments in increasing order, with the digital label range being [1, W]; The digital label corresponding to the segment is marked as the segment label, the segment label and the deformation threshold are used as evaluation data, and the W′ group of evaluation data corresponding to each region is sequentially input into the trained influence analysis model to predict the corresponding influence coefficient, where W′=W; The training process of the impact analysis model includes: Collecting influence coefficients corresponding to multiple groups of evaluation data in advance, and converting the evaluation data and the corresponding influence coefficients into a corresponding set of feature vectors; Each group of feature vectors is used as the input of the impact analysis model. The impact analysis model uses a group of predicted impact coefficients corresponding to each group of evaluation data as output, and uses the actual impact coefficients corresponding to each group of evaluation data as the prediction target. The actual evaluation flatness data is the pre-set impact coefficient corresponding to the evaluation data; minimizing the sum of the prediction deviations of all evaluation data is used as the training target; the impact analysis model is trained until the sum of the prediction deviations reaches convergence and the training is stopped; the impact analysis model is a deep neural network model; Multiply the obtained F influence coefficients by the deformation threshold of the corresponding area to obtain new deformation thresholds corresponding to W segments in the m areas, where F = W″×m, W″ is the number of influence coefficients of each area and W″=W′; The flattening time corresponding to the deformation data of m regions is divided by the collection interval to obtain the time ratio. The integer part of the time ratio is the corresponding segment label. According to the segment label, the new deformation threshold corresponding to the m regions is obtained. The deformation data of the m regions are compared with the corresponding new deformation threshold. If No adjustment instruction is generated if Generate adjustment instructions, is the new deformation threshold corresponding to the i-th region.

5. A pressure and deformation intelligent feedback control system for a ceramic flattening device, implementing a pressure and deformation intelligent feedback control method for a ceramic flattening device as claimed in any one of claims 1 to 4, characterized in that: include: Ceramic data acquisition module, used to collect ceramic physical data of ceramics; A region division module, used for dividing the ceramic into m regions; A flatness data acquisition module is used to collect flatness data corresponding to m areas in the ceramic; A data analysis module is used to analyze the ceramic physical data and the flatness data corresponding to the m regions, and obtain the pressure values ​​required for the m regions during the ceramic flattening process; The ceramic flattening module is used to control the ceramic flattening equipment to flatten the ceramic according to the pressure values ​​required by the m areas, and collect the deformation data of the m areas during the ceramic flattening process and provide real-time feedback; The feedback control module is used to determine whether to generate an adjustment instruction based on the deformation data of the m areas fed back in real time, mark the area where the adjustment instruction is generated as the adjustment area, analyze the deformation data of the adjustment area, calculate the adjustment pressure, and control the pressure value of the adjustment area in the ceramic flattening equipment to be adjusted to the adjustment pressure.

Images