Kiln control method for ceramic firing and automatically regulated kiln
By acquiring real-time and appearance evaluation data in the ceramic firing system and using predictive models for comprehensive evaluation and adaptive adjustment, the problem of multi-zone temperature control was solved, enabling efficient and uniform firing of ceramic products and improving product quality and production efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-03-20
- Publication Date
- 2026-07-10
AI Technical Summary
Existing ceramic firing systems struggle to achieve multi-zone temperature control and cannot adaptively adjust the temperature based on real-time conditions, resulting in inconsistent product quality and high energy consumption, making them unsuitable for flexible production needs involving multiple varieties and small batches.
By acquiring the real-time firing fluctuation coefficient and appearance evaluation coefficient of the temperature control area, and using the firing prediction model to predict future changes, a comprehensive firing evaluation coefficient is generated, enabling adaptive adjustment of temperature and time, and independent control of the temperature field of each area.
It improves the quality consistency and production efficiency of ceramic products, reduces energy consumption, shortens the firing cycle, and ensures the uniformity of the temperature field and the stability of the products.
Smart Images
Figure CN122360152A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of kiln control technology, and more specifically, to a kiln control method and an automatically adjustable kiln for ceramic firing. Background Technology
[0002] Ceramic firing is a crucial step in ceramic product manufacturing, ensuring the stability of the product's physical and chemical properties by controlling firing temperature, time, and heating curves. However, ceramics are extremely sensitive to temperature changes during firing. Furthermore, uneven temperature distribution within the ceramic kiln can lead to inconsistent product quality and increased scrap rates. Traditional ceramic firing systems typically use fixed temperature zones for control, making it difficult to adaptively adjust the temperature distribution based on real-time conditions during firing. Simultaneously, increasing production demands, such as flexible production of multiple varieties in small batches, require more precise temperature control to reduce energy consumption and improve product consistency. Therefore, developing an adaptive firing system with multi-zone temperature control is particularly necessary.
[0003] Existing methods, such as the Chinese patent application CN117109324A, disclose a kiln control method based on visual recognition and a kiln with automatic temperature adjustment. This method inputs ceramic firing raw materials and results into a machine learning model to generate a ceramic processing flow. The kiln is then controlled to use this generated process, and ceramic processing is performed through manual intervention. Historical firing data is retained for further model training. By using this technical solution, the processing technology for ceramic handicrafts is generated through a machine learning model, and the kiln operates in conjunction with this model, thus automating the processing of ceramic handicrafts. While this method automates the processing of ceramic handicrafts, facilitating the batch processing of ceramic handicrafts with different process requirements, research and application of this method and existing technologies have revealed at least the following shortcomings:
[0004] 1. This method is not applicable to multi-zone temperature control. It fails to differentiate between the real-time and future states of different temperature control zones, making it difficult to provide more precise firing control for complex environments.
[0005] 2. Adjusting the temperature solely through curve comparison without a comprehensive assessment of ceramic quality issues makes it difficult to promptly identify and resolve potential problems.
[0006] Therefore, the present invention provides a kiln control method for ceramic firing and an automatically adjustable kiln. Summary of the Invention
[0007] In order to overcome the above-mentioned defects of the prior art, the present invention provides a kiln control method for ceramic firing and an automatically adjustable kiln to solve the problems mentioned in the background art.
[0008] To achieve the above objectives, the present invention provides the following technical solution: Firstly, a kiln control method for ceramic firing, comprising:
[0009] Obtain the real-time firing fluctuation coefficient within the m-th temperature control region, and input the real-time firing fluctuation coefficient into the pre-generated firing prediction model to predict the future firing fluctuation coefficient of the m-th temperature control region; m is an integer greater than zero.
[0010] Obtain the appearance evaluation coefficient of ceramic products in the m-th temperature control zone, and integrate the appearance evaluation coefficient with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient.
[0011] Based on the comprehensive firing evaluation coefficient, it is determined whether there is a quality problem with the ceramic products in the m-th temperature control zone. If a quality problem occurs, a temperature control command is generated. If there is no quality problem and the quality exceeds expectations, a time adjustment command is generated.
[0012] Receive temperature control commands, acquire temperature control data, and adjust the current temperature of the m-th temperature control zone based on the temperature control data;
[0013] Receive time adjustment instructions, acquire time optimization data, and adjust the heating rate of the m-th temperature control zone based on the time optimization data.
[0014] Furthermore, the m-th temperature control zone is obtained by dividing the entire ceramic kiln into M temperature control zones based on a temperature evaluation coefficient, where m = 1, 2, ..., M;
[0015] The methods for obtaining the temperature evaluation coefficient include:
[0016] Step a1: Divide the entire ceramic kiln into K kiln sub-regions according to their area, and obtain the temperature characteristic data of the kth kiln sub-region. The temperature characteristic data includes real-time temperature, temperature gradient and heat conduction rate.
[0017] Step a2: Mark the real-time temperature, temperature gradient, and heat transfer velocity in the temperature characteristic data of the kth kiln sub-region as follows: , and ;
[0018] Step a3: Perform formulaic calculations on the temperature characteristic data to obtain the temperature evaluation coefficient for the kth kiln sub-region. The calculation formula is as follows:
[0019] j;
[0020] In the formula, This represents the temperature assessment coefficient for the k-th kiln sub-region. This indicates the depreciation rate of ceramic kilns. , and These are the corresponding weighting factors; , and All are greater than zero.
[0021] Furthermore, the method for dividing the entire ceramic kiln into M temperature control zones based on a temperature evaluation coefficient for the m-th temperature control zone includes:
[0022] Step b1: Extract the temperature evaluation coefficient of the kth kiln sub-region ;
[0023] Step b2: Set M temperature evaluation coefficient intervals and set M temperature control zones corresponding to the M temperature evaluation coefficient intervals; each temperature evaluation coefficient interval is associated with and bound to one and only one temperature control zone.
[0024] Step b3: Compare the temperature evaluation coefficient of the kth kiln sub-region with each temperature evaluation coefficient interval to obtain the temperature evaluation coefficient interval into which the temperature evaluation coefficient of the kth kiln sub-region falls;
[0025] Step b4: Based on the temperature evaluation coefficient range into which the temperature evaluation coefficient of the kth kiln sub-region falls, classify the kth kiln sub-region into the corresponding temperature control region; and let k = k + 1, then jump back to step b1;
[0026] Step b5: Repeat steps b1 to b4 above until k=M, at which point the loop ends, so that each of the kiln sub-regions is sequentially divided into M temperature control regions.
[0027] Furthermore, the method for obtaining the real-time firing fluctuation coefficient within the m-th temperature control region includes:
[0028] Step c1: Obtain firing characteristic data in the m-th temperature control zone, including heat flow fluctuation rate, temperature evaluation coefficient, furnace humidity and furnace pressure;
[0029] Step c2: Label the heat flux fluctuation rate, temperature evaluation coefficient, furnace humidity, and furnace pressure in the firing characteristic data as follows: , , and ;
[0030] Step c3: Perform dimensionless processing on the firing characteristic data to obtain the real-time firing fluctuation coefficient;
[0031] ;
[0032] In the formula, This represents the real-time firing fluctuation coefficient of the m-th temperature control zone. , , and These are the corresponding weighting factors; , , and All are greater than zero. This represents the logarithmic function with base e.
[0033] Furthermore, the method for generating the firing prediction model includes:
[0034] Historical firing sample data is obtained and divided into a firing training set and a firing test set; the historical firing sample data includes a set of firing fluctuation coefficients and their corresponding future firing fluctuation coefficients.
[0035] Construct a first regression network, using the set of firing fluctuation coefficients in the firing training set as the input data of the first regression network, and the future firing fluctuation coefficients in the firing training set as the output data of the first regression network. Train the first regression network to obtain an initial prediction model.
[0036] The initial prediction model is validated using a firing test set, and the initial prediction model whose prediction error is less than or equal to a preset error threshold is used as the firing prediction model; the first regression network is an LSTM neural network or an RNN recurrent neural network.
[0037] Furthermore, the method for obtaining the appearance evaluation coefficient of ceramic products in the m-th temperature control zone includes:
[0038] Step d1: Extract the r-th ceramic image from the firing image set, and extract the image data of the r-th ceramic image; the image data includes crack density, color difference coefficient, and roughness.
[0039] The crack density is obtained by binarizing and calculating the r-th ceramic image. The formula for calculating the crack density is as follows: In the formula, This represents the crack density of the r-th ceramic image; This represents the total number of pixels in the r-th ceramic image. This represents the total number of pixels in the crack area.
[0040] The methods for obtaining the color difference coefficient include:
[0041] Convert the RGB data of the r-th ceramic image to the Lab color space, where L represents luminance and z and x represent color components;
[0042] The color difference coefficient of the r-th ceramic image is calculated using the color difference formula, which is as follows:
[0043] ;
[0044] in, This represents the color difference coefficient of the r-th ceramic image. , and The brightness and color components of the base color; , and Let be the brightness and color components of the r-th ceramic image;
[0045] The roughness is obtained by performing a Fourier transform on the r-th ceramic image and calculating it using the RMS roughness formula, which is as follows: In the formula, This indicates the roughness of the r-th ceramic image. This represents the height value of the i-th pixel. Let I represent the average height of all pixels, and let I be the total number of pixels in the r-th ceramic image.
[0046] Step d2: Input the image data into the pre-built appearance analysis model to obtain the appearance evaluation coefficient of the r-th ceramic image, and let r = r + 1, and return to step d1;
[0047] The expression for the appearance analysis model is as follows:
[0048] ;
[0049] In the formula: Indicates the appearance evaluation coefficient. Indicates crack density. Indicates the color difference coefficient. Indicates the degree of roughness.
[0050] Step d3: Repeat steps d1 to d2 above until r = R, then end the loop and obtain the appearance evaluation coefficient for each ceramic image. Take the largest appearance evaluation coefficient as the appearance evaluation coefficient of the ceramic product in the m-th temperature control area, where R is the total number of ceramic images.
[0051] Furthermore, methods for integrating the appearance evaluation coefficient with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient include:
[0052] The predicted future firing fluctuation coefficient is labeled. The overall firing evaluation coefficient is obtained by integrating the appearance evaluation coefficient with the future firing fluctuation coefficient. The calculation formula is as follows:
[0053] ;
[0054] In the formula, This represents the overall evaluation coefficient for firing. As a balance factor, It is a constant greater than zero.
[0055] Furthermore, methods for determining whether ceramic products in the m-th temperature control zone have quality problems based on the comprehensive firing evaluation coefficient include:
[0056] Preset firing threshold, the firing threshold including and ,in > Compare the overall firing evaluation coefficient with the preset firing threshold.
[0057] like > Then a temperature control command will be generated;
[0058] like If so, no temperature control command or time adjustment command will be generated;
[0059] like Generate time adjustment instructions.
[0060] Furthermore, the method for obtaining the temperature regulation data includes:
[0061] Step e1: Place the test ceramic product in a cooling test environment and place the standard ceramic product in a set standard constant temperature test environment;
[0062] Step e2: Under the cooling test environment, obtain the appearance evaluation coefficient of the tested ceramic product at the f-th degree Celsius, where f is an integer greater than zero;
[0063] Step e3: Under a set standard constant temperature test environment, obtain the appearance evaluation coefficient of the standard ceramic product, and record it as the standard evaluation coefficient;
[0064] Step e4: Take the difference between the appearance evaluation coefficient and the standard evaluation coefficient as the temperature fluctuation difference, and compare the temperature fluctuation difference with the preset temperature fluctuation difference range. If the temperature fluctuation difference does not belong to the preset temperature fluctuation difference range, let f = f + 1 and return to step e2; if the temperature fluctuation difference belongs to the preset temperature fluctuation difference range, take the temperature fluctuation difference as the temperature adjustment data, and bind the f degree Celsius with the temperature adjustment data to obtain the relationship between the temperature adjustment data and the f degree Celsius.
[0065] Step e5: Repeat steps e2 to e4 until the f-th degree Celsius equals the set temperature F, then end the loop to obtain the relationship between the temperature adjustment data and each degree Celsius temperature, where F is a positive integer.
[0066] Furthermore, the method for obtaining the time optimization data includes:
[0067] Step s1: Obtain the first control data of the m-th temperature control zone, take the heating time interval in the first control data as a fixed quantity, the heating rate as a variable, and take the current control value of the heating rate as G;
[0068] Step s2: Let G = G + P, and record the appearance evaluation coefficient of the ceramic product under the control value G;
[0069] Step s3: Repeat step s2. When G equals the preset heating rate threshold, obtain W appearance evaluation coefficients under the first control data, and jump to step s1. W is an integer greater than zero.
[0070] Step s4: Sort the W appearance evaluation coefficients in ascending order of their values;
[0071] Step s5: Use the first control data corresponding to the appearance evaluation coefficient with the smallest value in the sorting as the time optimization data.
[0072] Secondly, the present invention provides an automatically adjustable kiln for implementing the above-described kiln control method for ceramic firing, comprising:
[0073] The prediction module is used to obtain the real-time firing fluctuation coefficient in the m-th temperature control area, and input the real-time firing fluctuation coefficient into the pre-generated firing prediction model to predict the future firing fluctuation coefficient of the m-th temperature control area; m is an integer greater than zero.
[0074] The comprehensive analysis module is used to obtain the appearance evaluation coefficient of ceramic products in the m-th temperature control zone, and to integrate the appearance evaluation coefficient with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient.
[0075] The judgment module determines whether there is a quality problem with the ceramic products in the m-th temperature control zone based on the comprehensive firing evaluation coefficient. If a quality problem occurs, a temperature control command is generated. If there is no quality problem and the quality exceeds expectations, a time adjustment command is generated.
[0076] The temperature adaptive module is used to receive temperature control commands, acquire temperature adjustment data, and adjust the current temperature of the m-th temperature control zone according to the temperature adjustment data.
[0077] The time adaptive module is used to receive time adjustment commands, acquire time optimization data, and adjust the heating rate of the m-th temperature control zone based on the time optimization data.
[0078] Thirdly, the present invention provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;
[0079] The processor executes the aforementioned kiln control method for ceramic firing by calling the computer program stored in the memory.
[0080] Fourthly, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described kiln control method for ceramic firing.
[0081] The technical effects and advantages of this invention are as follows:
[0082] 1. This invention obtains the real-time firing fluctuation coefficient of the m-th temperature control zone, uses a firing prediction model to estimate future changes in advance, avoids the development of quality problems of ceramic products to an uncontrollable degree during the firing process, and combines the real-time appearance evaluation coefficient with the predicted temperature fluctuation coefficient to generate a comprehensive firing evaluation coefficient, avoiding errors caused by single index judgment and improving the accuracy of judgment.
[0083] 2. This invention generates different instructions to achieve adaptive switching between temperature control and time adjustment. When quality problems occur, temperature adjustment is prioritized to correct the problems in a timely manner. When there are no quality problems and the quality exceeds expectations, the heating rate is optimized to improve firing efficiency. Differentiated management of different temperature control zones is achieved, saving resources and shortening the firing cycle. Furthermore, independent control of multiple zones ensures uniform temperature field in each zone, reducing the occurrence of firing defects. Attached Figure Description
[0084] Figure 1 This is a schematic diagram of the structure of an automatically adjustable kiln according to Example 1;
[0085] Figure 2 This is a flowchart of the method for obtaining the m-th temperature control zone in Example 1;
[0086] Figure 3 This is a flowchart of the kiln control method for ceramic firing in Example 2. Detailed Implementation
[0087] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0088] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.
[0089] It should be understood that although terms such as "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and a similar second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated items listed.
[0090] Example 1
[0091] Please see Figure 1 As shown, this embodiment discloses an automatically adjustable kiln, including: a prediction module, a comprehensive analysis module, a judgment module, a temperature adaptive module, and a time adaptive module; each module is connected via wired and / or wireless means to realize data transmission between modules.
[0092] The prediction module is used to obtain the real-time firing fluctuation coefficient in the m-th temperature control area, and input the real-time firing fluctuation coefficient into the pre-generated firing prediction model to predict the future firing fluctuation coefficient of the m-th temperature control area; m is an integer greater than zero.
[0093] It should be noted that the m-th temperature control zone is obtained by dividing the entire ceramic kiln into M temperature control zones based on the temperature evaluation coefficient, where m = 1, 2, ..., M.
[0094] It should be understood that ceramic kilns are installed in porcelain manufacturing plants for firing ceramic products. Each ceramic kiln is equipped with thermocouple sensors, humidity sensors, gas concentration sensors, and heating elements. When ceramic products are being fired, the ceramic kiln receives electrical signals sent by the user and uses these signals to collect data during the firing process. These ceramic kilns include shuttle kilns, electric kilns, tunnel kilns, gas-fired kilns, and oil-fired kilns, etc.
[0095] By integrating intelligent sensors and automatic controllers, end-to-end temperature control and energy efficiency optimization can be achieved, which helps to improve the consistency and pass rate of ceramic products.
[0096] The methods for obtaining the temperature evaluation coefficient include:
[0097] Step a1: Divide the entire ceramic kiln into K kiln sub-regions according to their area, and obtain the temperature characteristic data of the kth kiln sub-region. The temperature characteristic data includes real-time temperature, temperature gradient and heat conduction rate.
[0098] It should be noted that the real-time temperature is acquired in real time via a thermocouple sensor; the method for calculating the temperature gradient includes: obtaining the positions of the two temperature sensors. and And the real-time temperature is and The formula for calculating the temperature gradient is: In the formula, Represents the temperature gradient. This indicates the distance between the two temperature sensors.
[0099] The methods for obtaining the heat conduction velocity include:
[0100] Extract the thermal diffusivity and the temperature difference within a preset time period, and calculate the heat conduction velocity using the following formula:
[0101] , ,
[0102] In the formula, Indicates the rate of heat conduction. This indicates the temperature difference within a preset time period. Indicates the preset time. Indicates the thermal diffusivity. This indicates the thermal conductivity of the material used in ceramic kilns. This indicates the material density of the ceramic kiln. This indicates the specific heat capacity of the material used in ceramic kilns.
[0103] Step a2: Mark the real-time temperature, temperature gradient, and heat transfer velocity in the temperature characteristic data of the kth kiln sub-region as follows: , and ;
[0104] Step a3: Perform formulaic calculations on the temperature characteristic data to obtain the temperature evaluation coefficient for the kth kiln sub-region;
[0105] It should be noted that the formula for calculating the temperature assessment coefficient of the kth kiln sub-region is:
[0106] ;
[0107] In the formula, This represents the temperature assessment coefficient for the k-th kiln sub-region. This indicates the depreciation rate of ceramic kilns. , and These are the corresponding weighting factors; , and All are greater than zero.
[0108] It should be understood that the depreciation rate of ceramic kilns is obtained based on existing technology and is pre-stored in the ceramic kiln database, which will not be elaborated here.
[0109] Please see Figure 2 As shown, in practice, the method for dividing the entire ceramic kiln into M temperature control zones based on a temperature evaluation coefficient for the m-th temperature control zone includes:
[0110] Step b1: Extract the temperature evaluation coefficient of the kth kiln sub-region ;
[0111] Step b2: Set M temperature evaluation coefficient intervals and set M temperature control zones corresponding to the M temperature evaluation coefficient intervals; each temperature evaluation coefficient interval is associated with and bound to one and only one temperature control zone.
[0112] Step b3: Compare the temperature evaluation coefficient of the kth kiln sub-region with each temperature evaluation coefficient interval to obtain the temperature evaluation coefficient interval into which the temperature evaluation coefficient of the kth kiln sub-region falls;
[0113] Step b4: Based on the temperature evaluation coefficient range into which the temperature evaluation coefficient of the kth kiln sub-region falls, classify the kth kiln sub-region into the corresponding temperature control region; and let k = k + 1, then jump back to step b1;
[0114] Step b5: Repeat steps b1 to b4 above until k=M, at which point the loop ends, so that each of the kiln sub-regions is sequentially divided into M temperature control regions.
[0115] In practice, methods for obtaining the real-time firing fluctuation coefficient within the m-th temperature control region include:
[0116] Step c1: Obtain firing characteristic data in the m-th temperature control zone, including heat flow fluctuation rate, temperature evaluation coefficient, furnace humidity and furnace pressure;
[0117] It should be noted that: the heat flow fluctuation rate is obtained by measuring and calculating the average heat flow fluctuation within a preset time period using a heat flow sensor; the furnace humidity is obtained by measuring the average humidity within a preset time period using a humidity sensor; and the furnace air pressure is obtained by measuring the average air pressure within a preset time period using a pressure sensor.
[0118] Step c2: Label the heat flux fluctuation rate, temperature evaluation coefficient, furnace humidity, and furnace pressure in the firing characteristic data as follows: , , and ;
[0119] Step c3: Perform dimensionless processing on the firing characteristic data to obtain the real-time firing fluctuation coefficient;
[0120] ;
[0121] In the formula, This represents the real-time firing fluctuation coefficient of the m-th temperature control zone. , , and These are the corresponding weighting factors; , , and All are greater than zero. This represents the logarithmic function with base e.
[0122] It should be noted that higher values for heat flow fluctuation rate, temperature assessment coefficient, furnace humidity, and furnace pressure all contribute to a higher real-time firing fluctuation coefficient. This increases the risk of cracking and shrinkage during firing, significantly impacting the quality of ceramic products. Conversely, lower values for heat flow fluctuation rate, temperature assessment coefficient, furnace humidity, and furnace pressure result in a lower real-time firing fluctuation coefficient, indicating a more stable firing process and reducing the risk of quality problems in ceramic products.
[0123] In implementation, the method for generating the firing prediction model includes:
[0124] Historical firing sample data is obtained and divided into a firing training set and a firing test set; the historical firing sample data includes a set of firing fluctuation coefficients and their corresponding future firing fluctuation coefficients.
[0125] Construct a first regression network, using the set of firing fluctuation coefficients in the firing training set as the input data of the first regression network, and the future firing fluctuation coefficients in the firing training set as the output data of the first regression network. Train the first regression network to obtain an initial prediction model.
[0126] The initial prediction model is validated using a firing test set, and the initial prediction model whose prediction error is less than or equal to a preset error threshold is used as the firing prediction model; the first regression network is an LSTM neural network or an RNN recurrent neural network.
[0127] The comprehensive analysis module is used to obtain the appearance evaluation coefficient of ceramic products in the m-th temperature control zone, and to integrate the appearance evaluation coefficient with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient.
[0128] It should be noted that the methods for obtaining the appearance evaluation coefficient of ceramic products in the m-th temperature control zone include:
[0129] Step d1: Extract the r-th ceramic image from the firing image set, and extract the image data of the r-th ceramic image; the image data includes crack density, color difference coefficient, and roughness.
[0130] It should be noted that the crack density is obtained by binarizing and calculating the r-th ceramic image, and the formula for calculating the crack density is as follows: In the formula, This represents the crack density of the r-th ceramic image; This represents the total number of pixels in the r-th ceramic image. This represents the total number of pixels in the cracked area.
[0131] The methods for obtaining the color difference coefficient include:
[0132] Convert the RGB data of the r-th ceramic image to the Lab color space, where L represents luminance and z and x represent color components;
[0133] The color difference coefficient of the r-th ceramic image is calculated using the color difference formula, which is as follows:
[0134] ;
[0135] in, This represents the color difference coefficient of the r-th ceramic image. , and The brightness and color components of the base color; , and Let be the brightness and color components of the r-th ceramic image;
[0136] The roughness is obtained by performing a Fourier transform on the r-th ceramic image and calculating it using the RMS roughness formula, which is as follows: In the formula, This indicates the roughness of the r-th ceramic image. This represents the height value of the i-th pixel. Let I represent the average height of all pixels, and let I be the total number of pixels in the r-th ceramic image.
[0137] Step d2: Input the image data into the pre-built appearance analysis model to obtain the appearance evaluation coefficient of the r-th ceramic image, and let r = r + 1, and return to step d1;
[0138] The expression for the appearance analysis model is as follows:
[0139] ;
[0140] In the formula: Indicates the appearance evaluation coefficient. Indicates crack density. Indicates the color difference coefficient. Indicates the degree of roughness.
[0141] Step d3: Repeat steps d1 to d2 above until r = R, then end the loop and obtain the appearance evaluation coefficient for each ceramic image. Take the largest appearance evaluation coefficient as the appearance evaluation coefficient of the ceramic product in the m-th temperature control area, where R is the total number of ceramic images.
[0142] In practice, methods for integrating the appearance evaluation coefficient with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient include:
[0143] The predicted future firing fluctuation coefficient is labeled. The overall firing evaluation coefficient is obtained by integrating the appearance evaluation coefficient with the future firing fluctuation coefficient. The calculation formula is as follows:
[0144] ;
[0145] In the formula, This represents the overall evaluation coefficient for firing. As a balance factor, It is a constant greater than zero.
[0146] It should be noted that: the larger the value of the appearance evaluation coefficient, the larger the value of the future firing fluctuation coefficient, and the larger the value of the comprehensive firing evaluation coefficient, the greater the probability of quality problems in ceramic products in the m-th temperature control zone. Conversely, the smaller the value of the appearance evaluation coefficient, the smaller the value of the future firing fluctuation coefficient, and the smaller the value of the comprehensive firing evaluation coefficient, the lower the probability of quality problems in ceramic products in the m-th temperature control zone.
[0147] The judgment module is used to determine whether there is a quality problem with the ceramic products in the m-th temperature control zone based on the comprehensive firing evaluation coefficient. If a quality problem occurs, a temperature control command is generated; if there is no quality problem and the quality exceeds expectations, a time adjustment command is generated.
[0148] In practice, methods for determining whether ceramic products in the m-th temperature control zone have quality problems based on the comprehensive firing evaluation coefficient include:
[0149] Preset firing threshold, the firing threshold including and ,in > Compare the overall firing evaluation coefficient with the preset firing threshold.
[0150] like > If the temperature control command is generated, it indicates that the temperature, time or other parameters of the ceramic product in the m-th temperature control zone have deviated from the reasonable process range during the firing process, causing quality problems in the ceramic product. At this time, the temperature of the ceramic product in the m-th temperature control zone is adjusted by generating the temperature control command.
[0151] like If no temperature control command or time adjustment command is generated, it indicates that there are no quality problems with the ceramic products in the m-th temperature control zone.
[0152] like The generation of a time adjustment command indicates that the quality control of the firing process is very stable, even exceeding the expected standards. In this case, continuing to operate with the current parameters may result in unnecessary time waste. Therefore, the system generates a time adjustment command to shorten the firing time, improve production efficiency without affecting product quality. To this end, the firing time of ceramic products in the m-th temperature control zone is adjusted using the time adjustment command.
[0153] It should be noted that temperature is a core parameter affecting the quality of ceramic firing; excessively high or low temperatures can lead to defects such as cracks or discoloration. Therefore, when the overall firing evaluation coefficient exceeds the maximum preset firing threshold, temperature adjustment measures should be prioritized to reduce risk. When the overall firing evaluation coefficient is below the minimum preset firing threshold, it means that production is at its optimal state, and it is not necessary to maintain the original firing time. Instead, resources can be saved and production efficiency improved by shortening the firing time.
[0154] The temperature adaptive module is used to receive temperature control commands, acquire temperature adjustment data, and adjust the current temperature of the m-th temperature control zone according to the temperature adjustment data.
[0155] In practice, the method for obtaining the temperature regulation data includes:
[0156] Step e1: Place the test ceramic product in a cooling test environment and place the standard ceramic product in a set standard constant temperature test environment;
[0157] It should be noted that: the cooling test environment is a test environment in which the temperature is gradually reduced, used to observe the appearance quality of ceramic products; the set standard constant temperature test environment refers to a test environment with a constant temperature, in which the temperature is strictly controlled within a preset standard range in order to eliminate the influence of temperature fluctuations on the test results.
[0158] Step e2: Under the cooling test environment, obtain the appearance evaluation coefficient of the tested ceramic product at the f-th degree Celsius, where f is an integer greater than zero;
[0159] Step e3: Under a set standard constant temperature test environment, obtain the appearance evaluation coefficient of the standard ceramic product, and record it as the standard evaluation coefficient;
[0160] Step e4: Take the difference between the appearance evaluation coefficient and the standard evaluation coefficient as the temperature fluctuation difference, and compare the temperature fluctuation difference with the preset temperature fluctuation difference range. If the temperature fluctuation difference does not belong to the preset temperature fluctuation difference range, let f = f + 1 and return to step e2; if the temperature fluctuation difference belongs to the preset temperature fluctuation difference range, take the temperature fluctuation difference as the temperature adjustment data, and bind the f degree Celsius with the temperature adjustment data to obtain the relationship between the temperature adjustment data and the f degree Celsius.
[0161] Step e5: Repeat steps e2 to e4 until the f-th degree Celsius equals the set temperature F, then end the loop to obtain the relationship between the temperature adjustment data and each degree Celsius temperature, where F is a positive integer.
[0162] This step ensures the reliability and accuracy of temperature regulation data through continuous comparison and adjustment, and uses standard constant temperature conditions as a benchmark to avoid measurement errors caused by environmental fluctuations.
[0163] The time adaptive module is used to receive time adjustment commands, acquire time optimization data, and adjust the heating rate of the m-th temperature control zone according to the time optimization data;
[0164] In implementation, the method for obtaining the time optimization data includes:
[0165] Step s1: Obtain the first control data of the m-th temperature control zone, take the heating time interval in the first control data as a fixed quantity, the heating rate as a variable, and take the current control value of the heating rate as G;
[0166] It should be noted that the first control data includes a heating time interval and a heating rate. The heating time interval is defined as follows: in the m-th temperature control zone, the initial heating time is recorded, the target temperature is set, and the time when the temperature sensor detects that the temperature has reached the target temperature is recorded. The heating time interval is obtained by calculating the difference between the time when the target temperature is reached and the initial time. The heating rate refers to the rate of temperature change within this heating time interval.
[0167] Step s2: Let G = G + P, and record the appearance evaluation coefficient of the ceramic product under the control value G;
[0168] It should be noted that the value of P is set by those skilled in the art based on practical experience, or the value of P is determined using binary search or genetic algorithm. The binary search and genetic algorithm are existing technologies, and this embodiment will not elaborate on them further.
[0169] Step s3: Repeat step s2. When G equals the preset heating rate threshold, obtain W appearance evaluation coefficients under the first control data, and jump to step s1. W is an integer greater than zero.
[0170] Step s4: Sort the W appearance evaluation coefficients in ascending order of their values;
[0171] Step s5: Use the first control data corresponding to the appearance evaluation coefficient with the smallest value in the sorting as the time optimization data.
[0172] This embodiment obtains the real-time firing fluctuation coefficient of the m-th temperature control zone and uses a firing prediction model to estimate future changes in advance, thus preventing the quality problems of ceramic products from developing to an uncontrollable degree during the firing process. The real-time appearance evaluation coefficient is combined with the predicted temperature fluctuation coefficient to generate a comprehensive firing evaluation coefficient, avoiding errors caused by a single indicator and improving the accuracy of the judgment.
[0173] This embodiment generates different instructions to achieve adaptive switching between temperature control and time adjustment. When quality problems occur, temperature adjustment is prioritized to correct the problems in a timely manner. When there are no quality problems and the quality exceeds expectations, the heating rate is optimized to improve firing efficiency. Differentiated management of different temperature control zones is achieved, saving resources and shortening the firing cycle. Furthermore, independent control of multiple zones ensures uniform temperature field in each zone, reducing the occurrence of firing defects.
[0174] Example 2
[0175] Please see Figure 3 As shown, this embodiment provides a kiln control method for ceramic firing, the method comprising:
[0176] Obtain the real-time firing fluctuation coefficient within the m-th temperature control region, and input the real-time firing fluctuation coefficient into the pre-generated firing prediction model to predict the future firing fluctuation coefficient of the m-th temperature control region; m is an integer greater than zero.
[0177] It should be noted that the m-th temperature control zone is obtained by dividing the entire ceramic kiln into M temperature control zones based on the temperature evaluation coefficient, where m = 1, 2, ..., M.
[0178] It should be understood that ceramic kilns are installed in porcelain manufacturing plants for firing ceramic products. Each ceramic kiln is equipped with thermocouple sensors, humidity sensors, gas concentration sensors, and heating elements. When ceramic products are being fired, the ceramic kiln receives electrical signals sent by the user and uses these signals to collect data during the firing process. These ceramic kilns include shuttle kilns, electric kilns, tunnel kilns, gas-fired kilns, and oil-fired kilns, etc.
[0179] By integrating intelligent sensors and automatic controllers, end-to-end temperature control and energy efficiency optimization can be achieved, which helps to improve the consistency and pass rate of ceramic products.
[0180] The methods for obtaining the temperature evaluation coefficient include:
[0181] Step a1: Divide the entire ceramic kiln into K kiln sub-regions according to their area, and obtain the temperature characteristic data of the kth kiln sub-region. The temperature characteristic data includes real-time temperature, temperature gradient and heat conduction rate.
[0182] It should be noted that the real-time temperature is acquired in real time via a thermocouple sensor; the method for calculating the temperature gradient includes: obtaining the positions of the two temperature sensors. and And the real-time temperature is and The formula for calculating the temperature gradient is: In the formula, Represents the temperature gradient. This indicates the distance between the two temperature sensors.
[0183] The methods for obtaining the heat conduction velocity include:
[0184] Extract the thermal diffusivity and the temperature difference within a preset time period, and calculate the heat conduction velocity using the following formula:
[0185] , ,
[0186] In the formula, Indicates the rate of heat conduction. This indicates the temperature difference within a preset time period. Indicates the preset time. Indicates the thermal diffusivity. This indicates the thermal conductivity of the material used in ceramic kilns. This indicates the material density of the ceramic kiln. This indicates the specific heat capacity of the material used in ceramic kilns.
[0187] Step a2: Mark the real-time temperature, temperature gradient, and heat transfer velocity in the temperature characteristic data of the kth kiln sub-region as follows: , and ;
[0188] Step a3: Perform formulaic calculations on the temperature characteristic data to obtain the temperature evaluation coefficient for the kth kiln sub-region;
[0189] It should be noted that the formula for calculating the temperature assessment coefficient of the kth kiln sub-region is:
[0190] ;
[0191] In the formula, This represents the temperature assessment coefficient for the k-th kiln sub-region. This indicates the depreciation rate of ceramic kilns. , and These are the corresponding weighting factors; , and All are greater than zero.
[0192] It should be understood that the depreciation rate of ceramic kilns is obtained based on existing technology and is pre-stored in the ceramic kiln database, which will not be elaborated here.
[0193] In practice, the method for dividing the entire ceramic kiln into M temperature control zones based on a temperature evaluation coefficient for the m-th temperature control zone includes:
[0194] Step b1: Extract the temperature evaluation coefficient of the kth kiln sub-region ;
[0195] Step b2: Set M temperature evaluation coefficient intervals and set M temperature control zones corresponding to the M temperature evaluation coefficient intervals; each temperature evaluation coefficient interval is associated with and bound to one and only one temperature control zone.
[0196] Step b3: Compare the temperature evaluation coefficient of the kth kiln sub-region with each temperature evaluation coefficient interval to obtain the temperature evaluation coefficient interval into which the temperature evaluation coefficient of the kth kiln sub-region falls;
[0197] Step b4: Based on the temperature evaluation coefficient range into which the temperature evaluation coefficient of the kth kiln sub-region falls, classify the kth kiln sub-region into the corresponding temperature control region; and let k = k + 1, then jump back to step b1;
[0198] Step b5: Repeat steps b1 to b4 above until k=M, at which point the loop ends, so that each of the kiln sub-regions is sequentially divided into M temperature control regions.
[0199] In practice, methods for obtaining the real-time firing fluctuation coefficient within the m-th temperature control region include:
[0200] Step c1: Obtain firing characteristic data in the m-th temperature control zone, including heat flow fluctuation rate, temperature evaluation coefficient, furnace humidity and furnace pressure;
[0201] It should be noted that: the heat flow fluctuation rate is obtained by measuring and calculating the average heat flow fluctuation within a preset time period using a heat flow sensor; the furnace humidity is obtained by measuring the average humidity within a preset time period using a humidity sensor; and the furnace air pressure is obtained by measuring the average air pressure within a preset time period using a pressure sensor.
[0202] Step c2: Label the heat flux fluctuation rate, temperature evaluation coefficient, furnace humidity, and furnace pressure in the firing characteristic data as follows: , , and ;
[0203] Step c3: Perform dimensionless processing on the firing characteristic data to obtain the real-time firing fluctuation coefficient;
[0204] ;
[0205] In the formula, This represents the real-time firing fluctuation coefficient of the m-th temperature control zone. , , and These are the corresponding weighting factors; , , and All are greater than zero. This represents the logarithmic function with base e.
[0206] It should be noted that higher values for heat flow fluctuation rate, temperature assessment coefficient, furnace humidity, and furnace pressure all contribute to a larger real-time firing fluctuation coefficient. This increases the risk of cracking and shrinkage during firing, significantly impacting the quality of ceramic products. Conversely, lower values for heat flow fluctuation rate, temperature assessment coefficient, furnace humidity, and furnace pressure result in a smaller real-time firing fluctuation coefficient, indicating a more stable firing process and less impact on the quality of ceramic products.
[0207] In implementation, the method for generating the firing prediction model includes:
[0208] Historical firing sample data is obtained and divided into a firing training set and a firing test set; the historical firing sample data includes a set of firing fluctuation coefficients and their corresponding future firing fluctuation coefficients.
[0209] Construct a first regression network, using the set of firing fluctuation coefficients in the firing training set as the input data of the first regression network, and the future firing fluctuation coefficients in the firing training set as the output data of the first regression network. Train the first regression network to obtain an initial prediction model.
[0210] The initial prediction model is validated using a firing test set, and the initial prediction model whose prediction error is less than or equal to a preset error threshold is used as the firing prediction model; the first regression network is an LSTM neural network or an RNN recurrent neural network.
[0211] Furthermore, the appearance evaluation coefficient of the ceramic product in the m-th temperature control area is obtained, and the appearance evaluation coefficient is integrated with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient.
[0212] It should be noted that the methods for obtaining the appearance evaluation coefficient of ceramic products in the m-th temperature control zone include:
[0213] Step d1: Extract the r-th ceramic image from the firing image set, and extract the image data of the r-th ceramic image; the image data includes crack density, color difference coefficient, and roughness.
[0214] It should be noted that the crack density is obtained by binarizing and calculating the r-th ceramic image, and the formula for calculating the crack density is as follows: In the formula, This represents the crack density of the r-th ceramic image; This represents the total number of pixels in the r-th ceramic image. This represents the total number of pixels in the cracked area.
[0215] The methods for obtaining the color difference coefficient include:
[0216] Convert the RGB data of the r-th ceramic image to the Lab color space, where L represents luminance and z and x represent color components;
[0217] The color difference coefficient of the r-th ceramic image is calculated using the color difference formula, which is as follows:
[0218] ;
[0219] in, This represents the color difference coefficient of the r-th ceramic image. , and The brightness and color components of the base color; , and Let be the brightness and color components of the r-th ceramic image;
[0220] The roughness is obtained by performing a Fourier transform on the r-th ceramic image and calculating it using the RMS roughness formula, which is as follows: In the formula, This indicates the roughness of the r-th ceramic image. This represents the height value of the i-th pixel. Let I represent the average height of all pixels, and let I be the total number of pixels in the r-th ceramic image.
[0221] Step d2: Input the image data into the pre-built appearance analysis model to obtain the appearance evaluation coefficient of the r-th ceramic image, and let r = r + 1, and return to step d1;
[0222] The expression for the appearance analysis model is as follows:
[0223] ;
[0224] In the formula: Indicates the appearance evaluation coefficient. Indicates crack density. Indicates the color difference coefficient. Indicates the degree of roughness.
[0225] Step d3: Repeat steps d1 to d2 above until r = R, then end the loop and obtain the appearance evaluation coefficient for each ceramic image. Take the largest appearance evaluation coefficient as the appearance evaluation coefficient of the ceramic product in the m-th temperature control area, where R is the total number of ceramic images.
[0226] In practice, methods for integrating the appearance evaluation coefficient with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient include:
[0227] The predicted future firing fluctuation coefficient is labeled. The overall firing evaluation coefficient is obtained by integrating the appearance evaluation coefficient with the future firing fluctuation coefficient. The calculation formula is as follows:
[0228] ;
[0229] In the formula, This represents the overall evaluation coefficient for firing. As a balance factor, It is a constant greater than zero.
[0230] It should be noted that: the larger the value of the appearance evaluation coefficient, the larger the value of the future firing fluctuation coefficient, and the larger the value of the comprehensive firing evaluation coefficient, the greater the probability of quality problems in ceramic products in the m-th temperature control zone. Conversely, the smaller the value of the appearance evaluation coefficient, the smaller the value of the future firing fluctuation coefficient, and the smaller the value of the comprehensive firing evaluation coefficient, the lower the probability of quality problems in ceramic products in the m-th temperature control zone.
[0231] Furthermore, based on the comprehensive firing evaluation coefficient, it is determined whether there is a quality problem with the ceramic products in the m-th temperature control zone. If a quality problem occurs, a temperature control command is generated; if there is no quality problem and the quality exceeds expectations, a time adjustment command is generated.
[0232] In practice, methods for determining whether ceramic products in the m-th temperature control zone have quality problems based on the comprehensive firing evaluation coefficient include:
[0233] Preset firing threshold, the firing threshold including and ,in > Compare the overall firing evaluation coefficient with the preset firing threshold.
[0234] like > If the temperature control command is generated, it indicates that the temperature, time or other parameters of the ceramic product in the m-th temperature control zone have deviated from the reasonable process range during the firing process, causing quality problems in the ceramic product. At this time, the temperature of the ceramic product in the m-th temperature control zone is adjusted by generating the temperature control command.
[0235] like If no temperature control command or time adjustment command is generated, it indicates that there are no quality problems with the ceramic products in the m-th temperature control zone.
[0236] like The generation of a time adjustment command indicates that the quality control of the firing process is very stable, even exceeding the expected standards. In this case, continuing to operate with the current parameters may result in unnecessary time waste. Therefore, the system generates a time adjustment command to shorten the firing time, improve production efficiency without affecting product quality. To this end, the firing time of ceramic products in the m-th temperature control zone is adjusted using the time adjustment command.
[0237] It should be noted that temperature is a core parameter affecting the quality of ceramic firing; excessively high or low temperatures can lead to defects such as cracks or discoloration. Therefore, when the overall firing evaluation coefficient exceeds the maximum preset firing threshold, temperature adjustment measures should be prioritized to reduce risk. When the overall firing evaluation coefficient is below the minimum preset firing threshold, it means that production is at its optimal state, and it is not necessary to maintain the original firing time. Instead, resources can be saved and production efficiency improved by shortening the firing time.
[0238] Furthermore, it receives temperature control commands, acquires temperature control data, and adjusts the current temperature of the m-th temperature control zone based on the temperature control data;
[0239] In practice, the method for obtaining the temperature regulation data includes:
[0240] Step e1: Place the test ceramic product in a cooling test environment and place the standard ceramic product in a set standard constant temperature test environment;
[0241] Step e2: Under the cooling test environment, obtain the appearance evaluation coefficient of the tested ceramic product at the f-th degree Celsius, where f is an integer greater than zero;
[0242] Step e3: Under a set standard constant temperature test environment, obtain the appearance evaluation coefficient of the standard ceramic product, and record it as the standard evaluation coefficient;
[0243] Step e4: Take the difference between the appearance evaluation coefficient and the standard evaluation coefficient as the temperature fluctuation difference, and compare the temperature fluctuation difference with the preset temperature fluctuation difference range. If the temperature fluctuation difference does not belong to the preset temperature fluctuation difference range, let f = f + 1 and return to step e2; if the temperature fluctuation difference belongs to the preset temperature fluctuation difference range, take the temperature fluctuation difference as the temperature adjustment data, and bind the f degree Celsius with the temperature adjustment data to obtain the relationship between the temperature adjustment data and the f degree Celsius.
[0244] Step e5: Repeat steps e2 to e4 until the f-th degree Celsius equals the set temperature F, then end the loop to obtain the relationship between the temperature adjustment data and each degree Celsius temperature, where F is a positive integer.
[0245] This step ensures the reliability and accuracy of temperature regulation data through continuous comparison and adjustment, and uses standard constant temperature conditions as a benchmark to avoid measurement errors caused by environmental fluctuations.
[0246] Furthermore, it receives time adjustment instructions, acquires time optimization data, and adjusts the heating rate of the m-th temperature control zone based on the time optimization data;
[0247] In implementation, the method for obtaining the time optimization data includes:
[0248] Step s1: Obtain the first control data of the m-th temperature control zone, take the heating time interval in the first control data as a fixed quantity, the heating rate as a variable, and take the current control value of the heating rate as G;
[0249] Step s2: Let G = G + P, and record the appearance evaluation coefficient of the ceramic product under the control value G;
[0250] It should be noted that the value of P is set by those skilled in the art based on practical experience, or the value of P is determined using binary search or genetic algorithm. The binary search and genetic algorithm are existing technologies, and this embodiment will not elaborate on them further.
[0251] Step s3: Repeat step s2. When G equals the preset heating rate threshold, obtain W appearance evaluation coefficients under the first control data, and jump to step s1. W is an integer greater than zero.
[0252] Step s4: Sort the W appearance evaluation coefficients in ascending order of their values;
[0253] Step s5: Use the first control data corresponding to the appearance evaluation coefficient with the smallest value in the sorting as the time optimization data.
[0254] Example 3
[0255] This embodiment provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable code, which, when executed by the one or more processors, can perform the kiln control method for ceramic firing as described above.
[0256] The system according to the embodiments of this application can also be implemented using the architecture of the electronic device described below. The electronic device may include a bus, one or more CPUs, read-only memory (ROM), random access memory (RAM), a communication port connected to a network, input / output, a hard disk, etc. The storage device in the electronic device, such as ROM or hard disk, may store the kiln control method for ceramic firing provided in this application. Furthermore, the electronic device may also include a user interface. Of course, the above architecture is merely exemplary; when implementing different devices, one or more components of the above-described electronic device may be omitted according to actual needs.
[0257] Example 4
[0258] This embodiment provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the kiln control method for ceramic firing according to Embodiment 1.
[0259] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0260] In conclusion, 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 within the protection scope of the present invention.
Claims
1. A kiln control method for ceramic firing, characterized in that, include: Obtain the real-time firing fluctuation coefficient in the m-th temperature control area, and input the real-time firing fluctuation coefficient into the pre-generated firing prediction model to predict the future firing fluctuation coefficient of the m-th temperature control area. m is an integer greater than zero; Obtain the appearance evaluation coefficient of ceramic products in the m-th temperature control zone, and integrate the appearance evaluation coefficient with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient. The firing comprehensive evaluation coefficient is used to determine whether there is a quality problem with the ceramic products in the m-th temperature control zone. If there is a quality problem, a temperature control command is generated. If there is no quality problem and the quality exceeds expectations, a time adjustment command is generated. Receive temperature control commands, acquire temperature control data, and adjust the current temperature of the m-th temperature control zone based on the temperature control data; Receive time adjustment instructions, acquire time optimization data, and adjust the heating rate of the m-th temperature control zone based on the time optimization data.
2. The kiln control method for ceramic firing according to claim 1, characterized in that, The m-th temperature control zone is obtained by dividing the entire ceramic kiln into M temperature control zones based on a temperature evaluation coefficient, where m = 1, 2, ..., M; The methods for obtaining the temperature evaluation coefficient include: Step a1: Divide the entire ceramic kiln into K kiln sub-regions according to their area, and obtain the temperature characteristic data of the kth kiln sub-region. The temperature characteristic data includes real-time temperature, temperature gradient and heat conduction rate. Step a2: Mark the real-time temperature, temperature gradient, and heat transfer velocity in the temperature characteristic data of the kth kiln sub-region as follows: , and ; Step a3: Perform formulaic calculations on the temperature characteristic data to obtain the temperature evaluation coefficient for the kth kiln sub-region. The calculation formula is as follows: j； In the formula, This represents the temperature assessment coefficient for the k-th kiln sub-region. This indicates the depreciation rate of ceramic kilns. , and These are the corresponding weighting factors; , and All are greater than zero.
3. The kiln control method for ceramic firing according to claim 2, characterized in that, The method for dividing the entire ceramic kiln into M temperature control zones based on a temperature evaluation coefficient includes: Step b1: Extract the temperature evaluation coefficient of the kth kiln sub-region ; Step b2: Set M temperature evaluation coefficient intervals and set M temperature control zones corresponding to the M temperature evaluation coefficient intervals; each temperature evaluation coefficient interval is associated with and bound to one and only one temperature control zone. Step b3: Compare the temperature evaluation coefficient of the kth kiln sub-region with each temperature evaluation coefficient interval to obtain the temperature evaluation coefficient interval into which the temperature evaluation coefficient of the kth kiln sub-region falls; Step b4: Based on the temperature evaluation coefficient range into which the temperature evaluation coefficient of the kth kiln sub-region falls, classify the kth kiln sub-region into the corresponding temperature control region; and let k = k + 1, then jump back to step b1; Step b5: Repeat steps b1 to b4 above until k=M, at which point the loop ends, so that each of the kiln sub-regions is sequentially divided into M temperature control regions.
4. The kiln control method for ceramic firing according to claim 3, characterized in that, Methods for obtaining the real-time firing fluctuation coefficient within the m-th temperature control region include: Step c1: Obtain firing characteristic data in the m-th temperature control zone, including heat flow fluctuation rate, temperature evaluation coefficient, furnace humidity and furnace pressure; Step c2: Label the heat flux fluctuation rate, temperature evaluation coefficient, furnace humidity, and furnace pressure in the firing characteristic data as follows: , , and ; Step c3: Perform dimensionless processing on the firing characteristic data to obtain the real-time firing fluctuation coefficient; ； In the formula, This represents the real-time firing fluctuation coefficient of the m-th temperature control zone. , , and These are the corresponding weighting factors; , , and All are greater than zero. This represents the logarithmic function with base e.
5. The kiln control method for ceramic firing according to claim 4, characterized in that, The method for generating the firing prediction model includes: Historical firing sample data is obtained and divided into a firing training set and a firing test set; the historical firing sample data includes a set of firing fluctuation coefficients and their corresponding future firing fluctuation coefficients. Construct a first regression network, using the set of firing fluctuation coefficients in the firing training set as the input data of the first regression network, and the future firing fluctuation coefficients in the firing training set as the output data of the first regression network. Train the first regression network to obtain an initial prediction model. The initial prediction model is validated using a firing test set, and the initial prediction model whose prediction error is less than or equal to a preset error threshold is used as the firing prediction model; the first regression network is an LSTM neural network or an RNN recurrent neural network.
6. The kiln control method for ceramic firing according to claim 5, characterized in that, The methods for obtaining the appearance evaluation coefficient of ceramic products in the m-th temperature control zone include: Step d1: Extract the r-th ceramic image from the firing image set, and extract the image data of the r-th ceramic image; the image data includes crack density, color difference coefficient, and roughness. The crack density is obtained by binarizing and calculating the r-th ceramic image. The formula for calculating the crack density is as follows: In the formula, This represents the crack density of the r-th ceramic image; This represents the total number of pixels in the r-th ceramic image. This represents the total number of pixels in the crack area. The methods for obtaining the color difference coefficient include: Convert the RGB data of the r-th ceramic image to the Lab color space, where L represents luminance and z and x represent color components; The color difference coefficient of the r-th ceramic image is calculated using the color difference formula, which is as follows: ； in, This represents the color difference coefficient of the r-th ceramic image. , and The brightness and color components of the base color; , and Let be the brightness and color components of the r-th ceramic image; The roughness is obtained by performing a Fourier transform on the r-th ceramic image and calculating it using the RMS roughness formula, which is as follows: In the formula, This indicates the roughness of the r-th ceramic image. This represents the height value of the i-th pixel. Let I represent the average height of all pixels, and let I be the total number of pixels in the r-th ceramic image. Step d2: Input the image data into the pre-built appearance analysis model to obtain the appearance evaluation coefficient of the r-th ceramic image, and let r = r + 1, and return to step d1; The expression for the appearance analysis model is as follows: ； In the formula: Indicates the appearance evaluation coefficient. Indicates crack density. Indicates the color difference coefficient. Indicates the degree of roughness; Step d3: Repeat steps d1 to d2 above until r = R, then end the loop and obtain the appearance evaluation coefficient for each ceramic image. Take the largest appearance evaluation coefficient as the appearance evaluation coefficient of the ceramic product in the m-th temperature control area, where R is the total number of ceramic images.
7. The kiln control method for ceramic firing according to claim 6, characterized in that, Methods for integrating appearance evaluation coefficients with future firing fluctuation coefficients to obtain a comprehensive firing evaluation coefficient include: The predicted future firing fluctuation coefficient is labeled. The overall firing evaluation coefficient is obtained by integrating the appearance evaluation coefficient with the future firing fluctuation coefficient. The calculation formula is as follows: ； In the formula, This represents the overall evaluation coefficient for firing. As a balance factor, It is a constant greater than zero.
8. The kiln control method for ceramic firing according to claim 7, characterized in that, Methods for determining whether ceramic products in the m-th temperature control zone have quality problems based on the comprehensive firing evaluation coefficient include: Preset firing threshold, the firing threshold including and ,in > Compare the overall firing evaluation coefficient with the preset firing threshold. like > Then a temperature control command will be generated; like If so, no temperature control command or time adjustment command will be generated; like Generate time adjustment instructions.
9. The kiln control method for ceramic firing according to claim 8, characterized in that, The method for obtaining the temperature regulation data includes: Step e1: Place the test ceramic product in a cooling test environment and place the standard ceramic product in a set standard constant temperature test environment; Step e2: Under the cooling test environment, obtain the appearance evaluation coefficient of the tested ceramic product at the f-th degree Celsius, where f is an integer greater than zero; Step e3: Under a set standard constant temperature test environment, obtain the appearance evaluation coefficient of the standard ceramic product, and record it as the standard evaluation coefficient; Step e4: Take the difference between the appearance evaluation coefficient and the standard evaluation coefficient as the temperature fluctuation difference, and compare the temperature fluctuation difference with the preset temperature fluctuation difference range. If the temperature fluctuation difference does not belong to the preset temperature fluctuation difference range, let f = f + 1 and return to step e2; if the temperature fluctuation difference belongs to the preset temperature fluctuation difference range, take the temperature fluctuation difference as the temperature adjustment data, and bind the f degree Celsius with the temperature adjustment data to obtain the relationship between the temperature adjustment data and the f degree Celsius. Step e5: Repeat steps e2 to e4 until the f-th degree Celsius equals the set temperature F, then end the loop to obtain the relationship between the temperature adjustment data and each degree Celsius temperature, where F is a positive integer.
10. The kiln control method for ceramic firing according to claim 9, characterized in that, The method for obtaining the time optimization data includes: Step s1: Obtain the first control data of the m-th temperature control zone, take the heating time interval in the first control data as a fixed quantity, the heating rate as a variable, and take the current control value of the heating rate as G; Step s2: Let G = G + P, and record the appearance evaluation coefficient of the ceramic product under the control value G; Step s3: Repeat step s2. When G equals the preset heating rate threshold, obtain W appearance evaluation coefficients under the first control data, and jump to step s1. W is an integer greater than zero. Step s4: Sort the W appearance evaluation coefficients from smallest to largest value; Step s5: Use the first control data corresponding to the appearance evaluation coefficient with the smallest value in the sorting as the time optimization data.
11. An automatically adjustable kiln for implementing the kiln control method for ceramic firing according to any one of claims 1-10, characterized in that, include: The prediction module is used to obtain the real-time firing fluctuation coefficient in the m-th temperature control area and input the real-time firing fluctuation coefficient into the pre-generated firing prediction model to predict the future firing fluctuation coefficient of the m-th temperature control area. m is an integer greater than zero; The comprehensive analysis module is used to obtain the appearance evaluation coefficient of ceramic products in the m-th temperature control zone, and to integrate the appearance evaluation coefficient with the future firing fluctuation coefficient to obtain the comprehensive firing evaluation coefficient. The judgment module determines whether there is a quality problem with the ceramic products in the m-th temperature control zone based on the comprehensive firing evaluation coefficient. If a quality problem occurs, a temperature control command is generated. If there is no quality problem and the quality exceeds expectations, a time adjustment command is generated. The temperature adaptive module is used to receive temperature control commands, acquire temperature adjustment data, and adjust the current temperature of the m-th temperature control zone according to the temperature adjustment data. The time adaptive module is used to receive time adjustment commands, acquire time optimization data, and adjust the heating rate of the m-th temperature control zone based on the time optimization data.
12. An electronic device, characterized in that, include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the kiln control method for ceramic firing as described in any one of claims 1-10 by calling the computer program stored in the memory.
13. A computer-readable storage medium, characterized in that, The system stores instructions that, when executed on a computer, cause the computer to perform the kiln control method for ceramic firing as described in any one of claims 1-10.