Intelligent preparation method of efficient denitration agent

Abstract

The invention provides an intelligent preparation method of a high-efficiency denitration agent, which comprises the following steps: in response to the completion of manufacturing of raw material substrates, numbering the raw material substrates in sequence, and judging whether the raw material substrates meet preset sampling inspection conditions or not; in response to a preset sampling inspection condition, determining the structure, and configuring the light source unit based on the determined structure characteristic; controlling the light source unit to perform light irradiation, and configuring a shooting unit for the raw material substrate; controlling the shooting unit to carry out image shooting to obtain a substrate image; performing region identification on the substrate image, and determining a substrate type corresponding to the raw material substrate based on the obtained region shapes corresponding to the light regions in the substrate image; and calling a preset quality determination strategy corresponding to the type of the substrate to carry out quality detection on the raw material substrate. The detection efficiency is at least improved.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to an intelligent preparation method of a high-efficiency denitrification agent. Background Art

[0002] Denitrification agent, also known as denitrification additive, is a chemical used to remove harmful nitrogen oxides from gases. The following is a detailed introduction to denitrification agent:

[0003] 1. Working principle: Denitrification agent converts nitrogen oxides into harmless gases such as nitrogen and water vapor through chemical reduction reaction with nitrogen oxides, so that the waste gas generated by industrial production meets the emission standards, eliminates the hidden dangers of pollutants from the source, and protects the environment.

[0004] 2. Main types: Denitrification agents are mainly divided into two categories: polymer denitrification agents and low-temperature denitrification agents. In addition, according to the different denitrification principles, they can also be divided into reduction denitrification agents and oxidation denitrification agents. Among them, polymer denitrification agents: contain functional polymer reducing materials and auxiliary components such as emulsifiers, dispersants, slow-release agents, activators and penetrants, which can achieve efficient denitrification effects at lower temperatures, and have the advantages of high denitrification rate, wide temperature adaptation, safe operation and no toxic side effects. Low-temperature denitrification agent: suitable for denitrification processes under low temperature environments, with the characteristics of simple operation and high denitrification efficiency. Reduction denitrification agent: reduces nitrogen oxides (NOx) into nitrogen (N2) and water through the reduction principle. Oxidation denitrification agent: converts nitrogen oxides (NOx) into nitrates through the oxidation principle, thereby achieving the purpose of removing nitrogen oxides. Commonly used oxidation denitrification agents include aluminum powder, iron powder, rare earth aluminum oxide and rare earth iron oxide, which have the advantages of high denitrification efficiency and low cost.

[0005] 3. Application fields: Denitrifiers are widely used in many fields to reduce the emission of nitrogen oxides and protect the environment and human health. Specifically, they can include: Coal-fired power plants: Denitrifiers are added to flue gas to reduce the emission of nitrogen oxides. Industrial production: Many industrial processes also produce nitrogen oxides, and denitrifiers can also reduce their emissions. Automobile exhaust treatment: Denitrifiers can be directly added to automobile exhaust to reduce the emission of nitrogen oxides. Wastewater treatment: Denitrifiers can also be used in wastewater treatment to reduce the content of nitrogen oxides in wastewater.

[0006] The inventors discovered during their research that random inspections should be conducted on denitrifiers during production to ensure that their quality meets production requirements. However, the existing testing process for denitrifiers usually involves multiple steps and complex operations, which increases the time and labor costs of testing, thereby reducing the production efficiency of the corresponding products. Summary of the invention

[0007] Based on the above problems, the present invention is proposed to provide a method and system for intelligently preparing a high-efficiency denitrification agent that overcomes the above problems or at least partially solves the above problems.

[0008] According to one aspect of the present invention, there is provided a method for intelligently preparing a high-efficiency denitrification agent, comprising the following steps:

[0009] In response to the completion of manufacturing the raw material substrate based on the denitrification agent, sequentially numbering the raw material substrate according to the current numbering sequence, and judging whether the raw material substrate meets the preset sampling condition based on the obtained substrate number;

[0010] In response to the raw material substrate meeting the preset sampling condition, the raw material substrate is structurally determined, and a light source unit is configured on the first ventilation surface of the raw material substrate based on the determined structural characteristics;

[0011] Controlling the light source unit to irradiate light toward the first ventilation surface, and configuring a shooting unit on a second ventilation surface of the raw material substrate that is arranged opposite to the first ventilation surface;

[0012] Controlling the shooting unit to shoot an image toward the second ventilation surface to obtain a substrate image;

[0013] Performing region recognition on the substrate image, and determining the type of substrate corresponding to the raw substrate based on the obtained region shapes corresponding to the light regions in the substrate image;

[0014] A preset quality determination strategy corresponding to the type of the substrate is retrieved to perform quality inspection on the raw material substrate.

[0015] Optionally, in the method according to the present invention, judging whether the raw material substrate meets the preset sampling condition based on the obtained substrate number includes:

[0016] Decomposing the substrate number to obtain a primary number and a secondary number;

[0017] Determining a production grade corresponding to the raw material substrate based on the main grade number, and determining a sampling frequency corresponding to the production grade based on a preset grade sampling rule, wherein different production grades correspond to different sampling frequencies;

[0018] Compare the secondary number with each sampling number corresponding to the sampling frequency;

[0019] In response to the secondary number being equal to any sampling number corresponding to the sampling frequency, the raw material substrate is judged to meet the preset sampling condition;

[0020] In response to the secondary number not being equal to any sampling number corresponding to the sampling frequency, determining the sampling number that is in a positive proximity relationship with the secondary number as the target number, and determining the number difference between the target number and the secondary number;

[0021] When the number difference is less than or equal to the preset difference, the sampling result of the substrate number equivalent to the target number is retrieved based on the historical sampling library, and if the sampling result is unqualified, the raw material substrate corresponding to the secondary number is judged to meet the preset sampling conditions.

[0022] Optionally, in the method according to the present invention, the method further comprises:

[0023] In response to the number of random inspections corresponding to the same production level reaching a preset number of random inspections, determining the random inspection results corresponding to the preset number of random inspections as the same random inspection batch;

[0024] In response to the obtained sampling batches reaching a preset batch number, determining the sampling quality corresponding to the sampling batch based on the sampling results in the same sampling batch;

[0025] Sort the sample inspection batches based on the chronological order, and determine whether the quality of each sample inspection shows a gradually decreasing trend or a gradually increasing trend based on the obtained sample inspection order;

[0026] If there is a gradual upward trend, the sampling frequency is reduced through training;

[0027] If there is a gradual downward trend, the sampling frequency is increased through training;

[0028] Among them, the sampling frequency after training is obtained by the following formula:

[0029]

[0030] Among them, q + is the sampling frequency k s Increase the number of training times, ∈ is the sampling frequency k s The training constant value, q - is the sampling frequency k s Reduce the number of training times, k w is the sampling frequency after training.

[0031] Optionally, in the method according to the present invention, in response to the raw material substrate meeting the preset sampling condition, the raw material substrate is structurally determined, and a light source unit is configured on the first ventilation surface of the raw material substrate based on the determined structural characteristics, including:

[0032] Determine any ventilation surface of the raw material substrate as a first ventilation surface, and collect an image of the first ventilation surface to obtain a raw material ventilation image;

[0033] Binarizing the raw material ventilation image to obtain substrate pixel points corresponding to the first pixel value and noise pixel points corresponding to the second pixel value;

[0034] Acquire a first number corresponding to each substrate pixel and a second number corresponding to each noise pixel, and determine a substrate density corresponding to the raw material substrate based on the first number and the second number;

[0035] Performing length detection on the raw material substrate based on a preset extension length strategy to determine a substrate length corresponding to the raw material substrate;

[0036] A fusion calculation is performed based on the substrate density and the substrate length, and a light source unit is configured on the first ventilation surface of the raw substrate based on the obtained substrate coefficient.

[0037] Optionally, in the method according to the present invention, performing length detection on the raw material substrate based on a preset extension length strategy to determine the substrate length corresponding to the raw material substrate includes:

[0038] Taking the image center point of the raw material ventilation image as the coordinate origin, establishing an image coordinate system corresponding to the raw material ventilation image, and acquiring each image coordinate point constituting the raw material ventilation image based on the image coordinate system;

[0039] In each image coordinate point, each coordinate point corresponding to the horizontal coordinate extreme value and the vertical coordinate extreme value is respectively determined as each detection coordinate point, and each detection point position having a mapping relationship with each detection coordinate point is determined in the first ventilation surface corresponding to the raw material ventilation image;

[0040] Determine a ventilation surface of the raw material substrate that is opposite to the first ventilation surface as a second ventilation surface, use the second ventilation surface as a horizontal plane, perform distance detection on each detection point, and obtain each distance detection value corresponding to each detection point;

[0041] Determine the detection value with the smallest corresponding value among the distance detection values ​​as the reference detection value, and determine the detection point corresponding to the reference detection value as the reference point;

[0042] The point-to-surface distance between the second ventilation surface and the reference point along an extension direction perpendicular to the second ventilation surface of the raw material substrate is determined as a substrate length corresponding to the raw material substrate.

[0043] Optionally, in the method according to the present invention, performing a fusion calculation based on the substrate density and the substrate length, and configuring a light source unit on the first ventilation surface of the raw substrate based on the obtained substrate coefficient, comprises:

[0044] Normalizing the substrate density and the substrate length respectively to obtain a corresponding first density coefficient and a first length coefficient;

[0045] The first density coefficient and the first length coefficient are respectively multiplied by the retrieved density weight value and the retrieved length weight value to obtain a second density coefficient and a second length coefficient;

[0046] The second density coefficient and the second length coefficient are superimposed and calculated to obtain a substrate coefficient corresponding to the raw substrate, and a preset light source comparison table is retrieved, wherein the light source comparison table includes different coefficient intervals and light source specifications corresponding to the different coefficient intervals.

[0047] The preset light source comparison table is traversed, and the light source specification corresponding to the substrate coefficient is determined based on the coefficient interval including the substrate coefficient, and the light source unit is configured on the first ventilation surface of the raw substrate based on the light source specification.

[0048] Optionally, in the method according to the present invention, performing region recognition on the substrate image, and determining the type of substrate corresponding to the raw material substrate based on the obtained region shapes corresponding to the light regions in the substrate image, comprises:

[0049] Performing pixel recognition on the obtained substrate image to obtain each light pixel point of the corresponding light source light in the substrate image, and performing pixel connection on each light pixel point based on the presented connectivity to obtain each light area;

[0050] Taking the area center point of each light area as the rotation point, the light area is rotated vertically in sequence to obtain a first rotation area, a second rotation area, and a third rotation area corresponding to the same light area, and the original area shape corresponding to the light area, the first rotation shape corresponding to the first rotation area, the second rotation shape corresponding to the second rotation area, and the third rotation shape corresponding to the third rotation area are determined as a shape comparison group corresponding to the light area;

[0051] Retrieving standard shapes corresponding to honeycomb types, corrugated types, and plate types, and performing similarity comparison between shape comparison groups corresponding to light regions and standard shapes;

[0052] In response to a shape similarity between any shape in any shape comparison group and any standard shape being greater than a preset first similarity, determining a light region corresponding to the shape comparison group as a similar shape corresponding to the standard shape;

[0053] In response to a shape ratio of similar shapes corresponding to any standard shape being greater than a preset ratio, the substrate type corresponding to the raw material substrate is determined as the type corresponding to the standard shape.

[0054] Optionally, in the method according to the present invention, calling a preset quality determination strategy corresponding to the type of the substrate to perform quality inspection on the raw material substrate includes:

[0055] When the substrate type corresponding to the raw substrate is a honeycomb type, the contours of each region corresponding to each light region are obtained, and the contours of the region are decomposed to obtain a horizontal contour line corresponding to the horizontal direction and a longitudinal contour line corresponding to the vertical direction;

[0056] Determine the size of the transverse contour line and the longitudinal contour line respectively, and obtain the transverse size and the longitudinal size corresponding to the same light area;

[0057] Based on the horizontal dimensions corresponding to the light regions, the horizontal dimension with the smallest corresponding value is determined as the horizontal contrast value; based on the vertical dimensions corresponding to the light regions, the vertical dimension with the smallest corresponding value is determined as the vertical contrast value;

[0058] Retrieving a horizontal reference value and a vertical reference value corresponding to the honeycomb type, and comparing the horizontal reference value and the vertical reference value with the horizontal comparison value and the vertical comparison value respectively;

[0059] In response to any comparison value being less than a corresponding reference value, the raw material substrate is determined to be unqualified.

[0060] Optionally, in the method according to the present invention, calling a preset quality determination strategy corresponding to the type of the substrate to perform quality inspection on the raw material substrate includes:

[0061] When the substrate type corresponding to the raw substrate is a plate type or a corrugated type, performing a similarity comparison between each light region, and classifying each light region whose corresponding region similarity is greater than a preset second similarity into the same region shape group, thereby obtaining each region shape group;

[0062] Determine the area of ​​each light region in the same region shape group, and obtain the area of ​​each region corresponding to each light region;

[0063] The area of ​​the region with the smallest corresponding value is determined as the minimum area, and the area of ​​the region with the largest corresponding value is determined as the maximum area, and the average value of the area of ​​all regions except the minimum area and the maximum area is calculated to obtain the area average value;

[0064] In response to an area difference between the extremely small area and / or the extremely large area and the area average being greater than a preset area difference, the raw material substrate is determined to be unqualified.

[0065] According to another aspect of the present invention, there is provided a high-efficiency denitrification agent intelligent preparation system, comprising:

[0066] a sampling inspection determination module, configured to, in response to the completion of the manufacture of the raw material substrate based on the denitrification agent, sequentially number the raw material substrate according to the current numbering sequence, and determine whether the raw material substrate meets the preset sampling inspection condition based on the obtained substrate number;

[0067] a structure determination module configured to determine the structure of the raw material substrate in response to the raw material substrate meeting a preset sampling condition, and configure a light source unit on the first ventilation surface of the raw material substrate based on the determined structural characteristics;

[0068] a shooting configuration module, configured to control the light source unit to irradiate light toward the first ventilation surface, and to configure a shooting unit on a second ventilation surface of the raw material substrate that is arranged opposite to the first ventilation surface;

[0069] An image capturing module is configured to control the capturing unit to capture an image toward the second ventilation surface to obtain an image of the substrate;

[0070] A region recognition module is configured to perform region recognition on the substrate image, and determine the type of substrate corresponding to the raw substrate based on the obtained region shapes corresponding to the light regions in the substrate image;

[0071] The quality detection module is configured to call a preset quality determination strategy corresponding to the type of the substrate to perform quality detection on the raw material substrate.

[0072] According to the solution of the present invention, firstly, the method sequentially numbers the raw material substrates and determines the sampling conditions through an automated process, which greatly improves the production efficiency and management accuracy, ensures that each step of the operation has a basis, and reduces the possibility of human error; secondly, by determining the structure of the raw material substrates that meet the sampling conditions, and accurately configuring the light source unit and the shooting unit, the non-destructive and efficient detection of the internal structural characteristics of the raw material substrate is realized, which not only improves the accuracy of the detection, but also ensures the integrity of the raw material substrate, providing a reliable basis for subsequent processing. Furthermore, by using light irradiation and image shooting technology, combined with advanced area recognition algorithms, the light area and its shape located in the substrate image can be quickly and accurately identified, so as to accurately determine the substrate type of the raw material substrate. The intelligence of this step greatly shortens the time of product classification and enhances the flexibility of the production line. Finally, the corresponding quality inspection strategy is automatically retrieved according to the identified substrate type, making the quality inspection process more scientific and rigorous, effectively ensuring the quality stability and consistency of the denitrification agent products, and improving the market competitiveness of the products and the corresponding detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A flow chart showing a method for intelligently preparing a highly efficient denitrifying agent according to an embodiment of the present invention is shown;

[0074] Figure 2 The figure shows the positional relationship between the light source unit, the raw material substrate and the shooting unit in this embodiment;

[0075] Figure 3 A cross-sectional view of a raw material substrate corresponding to a honeycomb type in this embodiment is shown;

[0076] Figure 4 A cross-sectional view of a raw material substrate of a corresponding plate type in this embodiment is shown;

[0077] Figure 5 A cross-sectional view of a raw material substrate corresponding to the type of corrugation in this embodiment is shown;

[0078] Figure 6 A structural block diagram of a system for intelligently preparing a high-efficiency denitrifying agent according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0079] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0080] Denitrification agent, also known as denitrification additive, is a chemical used to remove harmful nitrogen oxides from gases. The following is a detailed introduction to denitrification agent:

[0081] 1. Working principle: Denitrification agent converts nitrogen oxides into harmless gases such as nitrogen and water vapor through chemical reduction reaction with nitrogen oxides, so that the waste gas generated by industrial production meets the emission standards, eliminates the hidden dangers of pollutants from the source, and protects the environment.

[0082] 2. Main types: Denitrification agents are mainly divided into two categories: polymer denitrification agents and low-temperature denitrification agents. In addition, according to the different denitrification principles, they can also be divided into reduction denitrification agents and oxidation denitrification agents. Among them, polymer denitrification agents: contain functional polymer reducing materials and auxiliary components such as emulsifiers, dispersants, slow-release agents, activators and penetrants, which can achieve efficient denitrification effects at lower temperatures, and have the advantages of high denitrification rate, wide temperature adaptation, safe operation and no toxic side effects. Low-temperature denitrification agent: suitable for denitrification processes under low temperature environments, with the characteristics of simple operation and high denitrification efficiency. Reduction denitrification agent: reduces nitrogen oxides (NOx) into nitrogen (N2) and water through the reduction principle. Oxidation denitrification agent: converts nitrogen oxides (NOx) into nitrates through the oxidation principle, thereby achieving the purpose of removing nitrogen oxides. Commonly used oxidation denitrification agents include aluminum powder, iron powder, rare earth aluminum oxide and rare earth iron oxide, which have the advantages of high denitrification efficiency and low cost.

[0083] 3. Application fields: Denitrifiers are widely used in many fields to reduce the emission of nitrogen oxides and protect the environment and human health. Specifically, they can include: Coal-fired power plants: Denitrifiers are added to flue gas to reduce the emission of nitrogen oxides. Industrial production: Many industrial processes also produce nitrogen oxides, and denitrifiers can also reduce their emissions. Automobile exhaust treatment: Denitrifiers can be directly added to automobile exhaust to reduce the emission of nitrogen oxides. Wastewater treatment: Denitrifiers can also be used in wastewater treatment to reduce the content of nitrogen oxides in wastewater.

[0084] The inventors discovered during their research that random inspections should be conducted on denitrifiers during production to ensure that their quality meets production requirements. However, the existing testing process for denitrifiers usually involves multiple steps and complex operations, which increases the time and labor costs of testing, thereby reducing the production efficiency of the corresponding products.

[0085] In order to solve the problems existing in the above-mentioned prior art, the inventor proposes the solution of the present invention. One embodiment of the present invention provides a method for intelligently preparing a high-efficiency denitrification agent, which can be executed in a computing device.

[0086] Figure 1A flow chart of a method for intelligently preparing a high-efficiency denitrification agent according to an embodiment of the present invention is shown. The method is suitable for execution in a computing device, wherein the computing device can be understood as a terminal device with data processing capabilities, such as a mobile phone or a computer.

[0087] like Figure 1 As shown, the intelligent preparation method of the efficient denitrification agent proposed in this embodiment starts from step S102. In step S102, the following contents are included:

[0088] In response to the completion of the manufacture of the raw material substrate based on the denitrification agent, the raw material substrate is sequentially numbered according to the current numbering sequence, and it is determined whether the raw material substrate meets the preset sampling condition based on the obtained substrate number.

[0089] For example, in this embodiment, the raw material substrate can be understood as a substrate carrying a denitrification agent, and when the manufacturing of any raw material substrate is completed, the server will sequentially number the raw material substrates according to the current numbering sequence to make a written record of the manufacturing sequence of each raw material substrate; since the detection method for the raw material substrates in this embodiment is random inspection, in order to quickly determine the raw material substrates that need to be sampled during the production process, it is possible to determine whether the raw material substrate meets the preset sampling conditions based on the substrate number, so as to perform corresponding quality inspection on the raw material substrates that meet the preset sampling conditions.

[0090] Furthermore, in this embodiment, the above-mentioned “determining whether the raw substrate meets the preset sampling condition based on the obtained substrate number” may also include the following steps:

[0091] Decomposing the substrate number to obtain a primary number and a secondary number;

[0092] Determining a production grade corresponding to the raw material substrate based on the main grade number, and determining a sampling frequency corresponding to the production grade based on a preset grade sampling rule, wherein different production grades correspond to different sampling frequencies;

[0093] Compare the secondary number with each sampling number corresponding to the sampling frequency;

[0094] In response to the secondary number being equal to any sampling number corresponding to the sampling frequency, the raw material substrate is judged to meet the preset sampling condition;

[0095] In response to the secondary number not being equal to any sampling number corresponding to the sampling frequency, determining the sampling number that is in a positive proximity relationship with the secondary number as the target number, and determining the number difference between the target number and the secondary number;

[0096] When the number difference is less than or equal to the preset difference, the sampling result of the substrate number equivalent to the target number is retrieved based on the historical sampling library, and if the sampling result is unqualified, the raw material substrate corresponding to the secondary number is judged to meet the preset sampling conditions.

[0097] For example, in this embodiment, since the raw material substrate may correspond to different production batches during the production process, the quality requirements that need to be met by each production batch may also be different, resulting in the possibility that the sampling frequency corresponding to each production batch may also be different. Specifically, the higher the quality requirements of any production batch, the higher the corresponding sampling frequency should be, so as to ensure that the quality of the production batch meets the corresponding production conditions. Based on the foregoing, in order to determine the corresponding sampling batch based on the production batch, the relevant substrate number can be generated based on the relevant data of the production batch, specifically including the substrate number can be set to a composite number, including The primary level number and the secondary level number are included, wherein the primary level number specifically corresponds to the production level corresponding to the raw material substrate, and the production level also has a corresponding correspondence with the aforementioned production batch, that is, the higher the production level, the higher the quality requirement of the production batch, and the higher the corresponding sampling frequency; by determining the production level, the corresponding sampling frequency can be determined, for example, the frequency of the production level A is 10 / times, and the sampling frequency of the corresponding production level B can be 20 / times; and the corresponding relationship between the production level and the sampling frequency can be obtained from the preset level sampling rules pre-stored in the server;

[0098] As for the sub-numbers corresponding to the substrate numbers, they mainly correspond to the production sequence, that is, the sub-numbers are produced according to the production sequence. If the production sequence of a batch of raw substrates is 101, the corresponding sub-number can be 101;

[0099] It should be noted that, in this embodiment, illustratively, the corresponding component structure of a certain substrate number may be, for example, level A-101, wherein level A is the primary number and 101 is the secondary number.

[0100] Further, after the determination of the sampling inspection frequency is completed, the secondary number corresponding to the raw material substrate can be compared with the sampling inspection frequency. When the secondary number is equal to any sampling inspection number corresponding to the sampling inspection frequency, the raw material substrate is determined to meet the preset sampling inspection conditions. For example, when the secondary number of a raw material substrate is 10 and the corresponding sampling inspection frequency is 9, the raw material substrate does not meet the preset sampling inspection conditions. When the secondary number of a raw material substrate is 18 and the corresponding sampling inspection frequency is 9, since there is an integer multiple relationship between 18 and 9, the raw material substrate corresponding to the secondary number can be determined to meet the preset sampling inspection conditions.

[0101] Here, when the secondary number corresponding to any raw material substrate is not equal to any sampling number of the corresponding sampling frequency, the sampling number that is in a positive proximity relationship with it can also be determined as the target number. Here, the positive proximity relationship should be understood as the numbering sequence of the secondary number is greater than the numbering sequence of the corresponding sampling number. For example, when the sampling frequency is 10 / times, the corresponding secondary number is 11, then the sampling number of 10 can be determined as being in a positive proximity relationship with the secondary number; or when the sampling frequency is 10 / times, the corresponding sampling number is 29, then the sampling number of 20 can be determined as being in a positive proximity relationship with the secondary number;

[0102] After the target number is determined, the target number and the secondary number can be calculated to obtain the corresponding number difference; at the same time, the obtained number difference is compared with the preset difference. When the number difference is less than the preset difference, it indicates that the production sequence of the raw material substrate corresponding to the secondary number is close to the production sequence of the raw material substrate corresponding to the target number. At this time, the server will retrieve the sampling result of the substrate number equivalent to the target number based on the historical sampling library, and determine the corresponding sampling result;

[0103] However, if the sampling result is unqualified, the raw material substrate corresponding to the secondary number can be judged as meeting the preset sampling conditions, and the secondary number can be used for another sampling inspection to further improve the quality sampling inspection effect; and when the sampling inspection result is qualified, there is no need to use the secondary number for another sampling inspection.

[0104] Furthermore, in this embodiment, based on the above content, it can be known that each production level has a corresponding sampling batch, and the setting of the corresponding sampling batch can be based on human experience. In actual operation, the quality inspection of the raw material substrate is mainly to continuously improve the production process of the raw material substrate, so as to ensure that the relevant quality requirements are met. In order to dynamically adjust the sampling batch based on the actual situation to ensure that the production process is supervised by sampling, it can be implemented based on the following scheme:

[0105] In response to the number of random inspections corresponding to the same production level reaching a preset number of random inspections, determining the random inspection results corresponding to the preset number of random inspections as the same random inspection batch;

[0106] In response to the obtained sampling batches reaching a preset batch number, determining the sampling quality corresponding to the sampling batch based on the sampling results in the same sampling batch;

[0107] Sort the sample inspection batches based on the chronological order, and determine whether the quality of each sample inspection shows a gradually decreasing trend or a gradually increasing trend based on the obtained sample inspection order;

[0108] If there is a gradual upward trend, the sampling frequency is reduced through training;

[0109] If there is a gradual downward trend, the sampling frequency is increased through training;

[0110] Among them, the sampling frequency after training is obtained by the following formula:

[0111]

[0112] Among them, q + is the sampling frequency k s Increase the number of training times, ∈ is the sampling frequency k s The training constant value, q- is the sampling frequency k s Reduce the number of training times, k w is the sampling frequency after training.

[0113] For example, when the number of random inspections corresponding to the same production level reaches the preset number of random inspections, the random inspection results corresponding to the preset number of random inspections can be determined as the same random inspection batch; further, when the obtained random inspection batches reach the preset number of batches, it is necessary to determine the random inspection quality corresponding to the random inspection batch based on the random inspection results in the same random inspection batch; for example, the preset number of random inspections can be set to 20, and when the number of random inspections corresponding to level A is 20, it can be determined as the same random inspection batch, and if there are 100 corresponding random inspection results in total, 5 random inspection batches arranged in chronological order can be obtained, and then the random inspection quality corresponding to each random inspection batch can be obtained in turn, that is, the product qualification rate can be determined;

[0114] After completing the determination of the sampling quality corresponding to each sampling batch, each sampling batch can be sorted based on the chronological order to obtain the corresponding sampling order, and further judge the changing trend of each sampling quality according to the sampling order. If there is a gradually decreasing trend, it means that the corresponding product qualification rate is gradually decreasing. At this time, the corresponding sampling frequency should be increased to improve the supervision of the production process; if there is a gradually increasing trend, it means that the corresponding product qualification rate is gradually increasing. At this time, the corresponding sampling frequency can be reduced to place the focus of supervision on other production levels.

[0115] In step S104, the following contents are included:

[0116] In response to the raw material substrate meeting the preset sampling conditions, the raw material substrate is structurally determined, and a light source unit is configured on the first ventilation surface of the raw material substrate based on the determined structural characteristics.

[0117] For example, when the raw material substrate meets the preset sampling conditions, the raw material substrate can be subjected to corresponding quality inspection. In the present embodiment, the sampling inspection of the raw material substrate can be carried out based on image acquisition. The reasons are as follows: the raw material substrate produced using a denitrification agent generally has a ventilation structure with a fixed structure (including ventilation surfaces on both sides and a ventilation space between the ventilation surfaces on both sides). If the shape of the ventilation structure is different from the preset shape, it indicates that the raw material substrate will have a certain quality deviation. Therefore, the quality inspection of the raw material substrate can be completed based on the corresponding image acquisition of the ventilation surface. In addition, since the raw material substrate has a certain size and shape, and the corresponding ventilation structure generally has a plurality of ventilation holes arranged in an array, based on the scattering, transmission and reflection characteristics of light, if When capturing images of the raw material substrate under natural light, the natural light will be scattered, transmitted and reflected accordingly in the ventilation structure of the raw material substrate, thereby gradually reducing the luminous flux emitted from the ventilation surface, and the image content in the ventilation structure cannot be accurately acquired, thus affecting the detection effect of the quality inspection; based on this, the light source unit can be configured on the first ventilation surface of the raw material substrate that constitutes the ventilation structure, so as to increase the brightness of the ambient light so as to be able to capture the integrity of the ventilation structure; in addition, since the scattering, transmission and reflection of light in the ventilation structure will be affected by the structural characteristics of the ventilation structure (such as the length of the substrate and the density of the substrate corresponding to the ventilation holes), therefore, when configuring the light source unit, it is necessary to determine the structural characteristics accordingly.

[0118] Further, in this embodiment, the above-mentioned “in response to the raw material substrate meeting the preset sampling condition, determining the structure of the raw material substrate, and configuring the light source unit on the first ventilation surface of the raw material substrate based on the determined structural characteristics” may also include the following steps:

[0119] Determine any ventilation surface of the raw material substrate as a first ventilation surface, and collect an image of the first ventilation surface to obtain a raw material ventilation image;

[0120] Binarizing the raw material ventilation image to obtain substrate pixel points corresponding to the first pixel value and noise pixel points corresponding to the second pixel value;

[0121] Acquire a first number corresponding to each substrate pixel and a second number corresponding to each noise pixel, and determine a substrate density corresponding to the raw material substrate based on the first number and the second number;

[0122] Performing length detection on the raw material substrate based on a preset extension length strategy to determine a substrate length corresponding to the raw material substrate;

[0123] A fusion calculation is performed based on the substrate density and the substrate length, and a light source unit is configured on the first ventilation surface of the raw substrate based on the obtained substrate coefficient.

[0124] For example, in this embodiment, since the ventilation structure of the raw material substrate generally has two opposite ventilation surfaces, when setting the light source unit for the ventilation surface, any ventilation surface of the raw material substrate can be determined as the first ventilation surface, and the image of the first ventilation surface can be captured to obtain the corresponding raw material ventilation image. Here, the image capture method can be based on placing a corresponding camera directly in front of the first ventilation surface to perform the corresponding configuration operation; after obtaining the raw material ventilation image, in order to determine the structure of the raw material substrate, the raw material ventilation image can be binarized to obtain each substrate pixel point corresponding to the first pixel value and each noise pixel point corresponding to the second pixel value, wherein each noise pixel point can be understood as a pixel point other than the raw material. Pixel points corresponding to other areas outside the substrate; by means of binarization processing, the pixel points of the corresponding substrate area in the raw material ventilation image can be quickly acquired, and the substrate density of the corresponding raw material substrate is further determined by dividing the first number of corresponding substrate pixel points and the second number of corresponding noise pixel points; after completing the determination of the substrate density, it is also necessary to detect the substrate length of the corresponding raw material substrate, and here, the corresponding preset extension length strategy is selected to perform the length detection of the raw material substrate, so as to obtain the substrate length corresponding to the raw material substrate; finally, by fusing the acquired substrate density and substrate length, the light source unit can be configured for the first ventilation surface of the raw material substrate based on the possible substrate coefficient.

[0125] Furthermore, in this embodiment, the above-mentioned “performing length detection on the raw material substrate based on a preset extension length strategy to determine the substrate length corresponding to the raw material substrate” may also include the following steps:

[0126] Taking the image center point of the raw material ventilation image as the coordinate origin, establishing an image coordinate system corresponding to the raw material ventilation image, and acquiring each image coordinate point constituting the raw material ventilation image based on the image coordinate system;

[0127] In each image coordinate point, each coordinate point corresponding to the horizontal coordinate extreme value and the vertical coordinate extreme value is respectively determined as each detection coordinate point, and each detection point position having a mapping relationship with each detection coordinate point is determined in the first ventilation surface corresponding to the raw material ventilation image;

[0128] Determine a ventilation surface of the raw material substrate that is opposite to the first ventilation surface as a second ventilation surface, use the second ventilation surface as a horizontal plane, perform distance detection on each detection point, and obtain each distance detection value corresponding to each detection point;

[0129] Determine the detection value with the smallest corresponding value among the distance detection values ​​as the reference detection value, and determine the detection point corresponding to the reference detection value as the reference point;

[0130] The point-to-surface distance between the second ventilation surface and the reference point along an extension direction perpendicular to the second ventilation surface of the raw material substrate is determined as a substrate length corresponding to the raw material substrate.

[0131] For example, in this embodiment, when the raw material substrate is produced, due to different manufacturing processes, the two ventilation surfaces included in the ventilation structure of the raw material substrate may not be parallel to each other, so that the substrate length of the raw material substrate may not be a constant value. In this case, it is necessary to perform the length detection of the raw material substrate based on the preset extension length strategy, wherein the specific process may be as follows:

[0132] Firstly, the image center point of the raw material ventilation image can be used as the coordinate origin to establish an image coordinate system corresponding to the raw material ventilation image, and further based on the image coordinate system, the image coordinate points constituting the raw material ventilation image can be obtained;

[0133] Next, each coordinate point corresponding to the transverse coordinate extreme value (including the corresponding maximum value and the minimum value) and the longitudinal coordinate extreme value (including the corresponding maximum value and the minimum value) in each image coordinate point is determined as each detection coordinate point, wherein, since each detection coordinate point corresponds to a coordinate extreme value, each detection coordinate point can be understood to be located at the image contour of the raw material ventilation image; further, since the raw material ventilation image is obtained based on the first ventilation surface, there is a corresponding mapping relationship between the raw material ventilation image and the first ventilation surface, based on which each detection point position having a mapping relationship with each detection coordinate point can be obtained on the first ventilation surface;

[0134] Next, a ventilation surface of the raw material substrate that is opposite to the first ventilation surface is determined as a second ventilation surface, and further, the second ventilation surface is used as a horizontal plane to perform distance detection on each detection point to obtain each distance detection value corresponding to each detection point; specifically, the corresponding distance detection process may include: pre-setting an infrared sensor at a position above the first ventilation surface, and performing distance detection on each detection point based on the infrared sensor to obtain corresponding distance detection values;

[0135] Then, the detection value with the smallest corresponding value among the distance detection values ​​can be determined as the reference detection value, and the detection point corresponding to the reference detection value can be determined as the reference point at the same time; it can be explained that the smaller the corresponding distance detection value is, the longer the substrate length corresponding to the detection point should be;

[0136] Finally, the point-to-surface distance between the second ventilation surface and the reference point can be obtained along the extension direction perpendicular to the second ventilation surface of the raw material substrate, and the point-to-surface distance can be further determined as the substrate length corresponding to the raw material substrate; it should be noted that since the specific value of the substrate length will affect the configuration of the light source unit, the longer the substrate length, the higher the light source configuration of the required light source unit should be. Therefore, it is necessary to determine the distance detection value with the smallest corresponding value in each distance detection value as the corresponding reference detection value, and further determine the corresponding substrate length based on the reference detection value to ensure that when the light source unit is configured subsequently, the collection requirements for the integrity of the ventilation structure can be met.

[0137] In this embodiment, it can be known from the above content that after the substrate density and the substrate length are determined, a fusion calculation can be performed based on the substrate density and the substrate length, and then the light source unit can be configured on the first ventilation surface of the raw substrate based on the obtained substrate coefficient. The specific implementation method can be implemented based on the following steps:

[0138] Normalizing the substrate density and the substrate length respectively to obtain a corresponding first density coefficient and a first length coefficient;

[0139] The first density coefficient and the first length coefficient are respectively multiplied by the retrieved density weight value and the retrieved length weight value to obtain a second density coefficient and a second length coefficient;

[0140] The second density coefficient and the second length coefficient are superimposed and calculated to obtain a substrate coefficient corresponding to the raw substrate, and a preset light source comparison table is retrieved, wherein the light source comparison table includes different coefficient intervals and light source specifications corresponding to the different coefficient intervals.

[0141] The preset light source comparison table is traversed, and the light source specification corresponding to the substrate coefficient is determined based on the coefficient interval including the substrate coefficient, and the light source unit is configured on the first ventilation surface of the raw substrate based on the light source specification.

[0142] For example, in this embodiment, when the substrate coefficient is obtained, since the substrate density and the substrate length correspond to different measurement units, in order to perform subsequent fusion calculation, the basic density and the substrate length need to be normalized respectively. The specific steps may include calling the density normalization value and the length normalization value, and further multiplying the substrate density and the substrate length with the density normalization value and the length normalization value respectively, so as to obtain the corresponding first density coefficient and the first length coefficient;

[0143] Further, after obtaining the first density coefficient and the first length coefficient, the corresponding density weight value and the length weight value may be pre-set based on the influence of the density and the length on the configuration of the light source unit. For example, the density weight value may be 0.6, and the length weight value may be 0.4. The first density coefficient and the first length coefficient are respectively multiplied by the density weight value and the length weight value to obtain the corresponding second density coefficient and the second length coefficient.

[0144] Next, the second density coefficient and the second length coefficient are superimposed and calculated, and then the substrate coefficient corresponding to the raw material substrate can be obtained. In order to configure the light source unit based on the substrate coefficient, a corresponding preset light source comparison table can also be stored in the server, wherein the preset light source comparison table includes different coefficient intervals and light source specifications corresponding to each different coefficient interval;

[0145] Finally, by traversing the preset light source comparison table, the coefficient interval including the substrate coefficient is determined, and the light source specification corresponding to the substrate coefficient is further determined based on the coefficient interval, so that the light source unit of the first ventilation surface of the raw material substrate can be configured based on the light source specification; it should be noted that the coefficient interval included in the preset light source comparison table may, for example, include interval A[1, 2], interval B[3, 4] and interval C[5, 6]. When the corresponding substrate coefficient is 3, it can be determined that interval B includes the substrate coefficient. At this time, the corresponding light source unit can be selected based on the light source specification corresponding to interval B. The light source specification can be understood as the relevant configuration of the light intensity of the corresponding light source unit.

[0146] In step S106, the following contents are included:

[0147] The light source unit is controlled to irradiate light toward the first ventilation surface, and a shooting unit is configured on a second ventilation surface of the raw material substrate that is opposite to the first ventilation surface.

[0148] For example, after completing the configuration of the light source unit, the light source unit can be controlled to irradiate light toward the first ventilation surface, so that the light from the light source can be transmitted in the ventilation structure of the raw material substrate. At the same time, in order to be able to obtain the light irradiation conditions, it is also necessary to set a corresponding shooting unit at the second ventilation surface opposite to the first ventilation surface to realize image capture of the light irradiation conditions.

[0149] In step S108, the following contents are included:

[0150] The shooting unit is controlled to shoot an image toward the second ventilation surface to obtain a substrate image.

[0151] For example, after the setting of the shooting unit is completed, the shooting unit can be controlled to shoot images toward the second ventilation surface under the premise that the light source unit irradiates light toward the first ventilation surface, so as to obtain the corresponding substrate image.

[0152] For example, Figure 2 It is shown in the figure that when the raw material substrate is irradiated with light based on the light source unit, the shooting unit is controlled to shoot an image of the second ventilation surface of the raw material substrate, wherein the light source unit is arranged on the left side of the raw material substrate and the corresponding number is set to 6, and the shooting unit is arranged on the right side of the raw material substrate.

[0153] In step S110, the following contents are included:

[0154] The substrate image is subjected to region recognition, and the type of substrate corresponding to the raw substrate is determined based on the obtained region shapes corresponding to the light regions in the substrate image.

[0155] For example, in this embodiment, after the corresponding substrate image is obtained, the corresponding light irradiation conditions can be obtained based on the substrate image. Since the raw material substrate may include multiple substrate types, such as honeycomb type, corrugated type and plate type, in order to perform quality inspection on the raw material substrate based on the substrate image, it is necessary to determine the corresponding substrate type based on the substrate image. Here, since the substrate structures corresponding to different substrate types are different, the substrate image can be regionally identified, and the substrate type corresponding to the raw material substrate can be determined based on the shapes of the regions corresponding to the light regions in the substrate image.

[0156] Furthermore, in this embodiment, the above-mentioned “performing region recognition on the substrate image, and determining the type of substrate corresponding to the raw substrate based on the obtained region shapes corresponding to the light regions in the substrate image” may also include the following steps:

[0157] Performing pixel recognition on the obtained substrate image to obtain each light pixel point of the corresponding light source light in the substrate image, and performing pixel connection on each light pixel point based on the presented connectivity to obtain each light area;

[0158] Taking the area center point of each light area as the rotation point, the light area is rotated vertically in sequence to obtain a first rotation area, a second rotation area, and a third rotation area corresponding to the same light area, and the original area shape corresponding to the light area, the first rotation shape corresponding to the first rotation area, the second rotation shape corresponding to the second rotation area, and the third rotation shape corresponding to the third rotation area are determined as a shape comparison group corresponding to the light area;

[0159] Retrieving standard shapes corresponding to honeycomb types, corrugated types, and plate types, and performing similarity comparison between shape comparison groups corresponding to light regions and standard shapes;

[0160] In response to a shape similarity between any shape in any shape comparison group and any standard shape being greater than a preset first similarity, determining a light region corresponding to the shape comparison group as a similar shape corresponding to the standard shape;

[0161] In response to a shape ratio of similar shapes corresponding to any standard shape being greater than a preset ratio, the substrate type corresponding to the raw material substrate is determined as the type corresponding to the standard shape.

[0162] For example, in this embodiment, when performing region recognition on a substrate image, it can be implemented based on the following process:

[0163] First, it is necessary to perform pixel recognition on the obtained substrate image to obtain the light pixel points of the corresponding light source relationship pattern located in the substrate image, and further perform pixel connection on adjacent pixel points to obtain the corresponding light areas;

[0164] Next, since the substrate structure included in any substrate type is generally symmetrical, in order to group the light regions located in the substrate image based on the same region shape, the region center point of each light region can be used as the rotation point, and the light regions can be vertically rotated in sequence to obtain the first rotation region, the second rotation region and the third rotation region corresponding to the same light region, and the original region shape corresponding to the light region, the first rotation shape corresponding to the first rotation region, the second rotation shape corresponding to the second rotation region and the third rotation shape corresponding to the third rotation region are determined as the shape comparison group corresponding to the light region;

[0165] Next, in order to determine the type of substrate corresponding to the raw material substrate based on the regional shape, it is necessary to retrieve the standard shapes corresponding to the honeycomb type, corrugated type and plate type, and at the same time, perform a similarity comparison between the shape comparison group corresponding to each light region and each standard shape;

[0166] Then, when the shape similarity between any shape in any shape comparison group and any standard shape is greater than a preset first similarity, the light region corresponding to the shape comparison group may be determined as a similar shape corresponding to the standard shape;

[0167] Finally, it is also necessary to determine the shape ratio of similar shapes corresponding to each standard shape. When the shape ratio corresponding to any standard shape is greater than the preset ratio, it indicates that most of the light areas of the corresponding raw material substrate are similar to the standard shape. At this time, the substrate type corresponding to the raw material substrate can be determined as the type corresponding to the standard shape.

[0168] For example, Figure 3 The raw material substrate corresponding to the honeycomb type is shown. Figure 4 The raw material substrates corresponding to the plate type are shown. Figure 5 The raw material substrate corresponding to the corrugation type is shown.

[0169] In step S112, the following contents are included:

[0170] A preset quality determination strategy corresponding to the type of the substrate is retrieved to perform quality inspection on the raw material substrate.

[0171] For example, in this embodiment, after the substrate type of the raw material substrate is determined, a preset quality determination strategy corresponding to the substrate type can be retrieved from the server to perform quality inspection on the raw material substrate, thereby determining whether the raw material substrate is qualified or unqualified.

[0172] Furthermore, in this embodiment, the above-mentioned “calling a preset quality determination strategy corresponding to the type of the substrate to perform quality inspection on the raw substrate” may also include the following steps:

[0173] When the substrate type corresponding to the raw substrate is a honeycomb type, the contours of each region corresponding to each light region are obtained, and the contours of the region are decomposed to obtain a horizontal contour line corresponding to the horizontal direction and a longitudinal contour line corresponding to the vertical direction;

[0174] Determine the size of the transverse contour line and the longitudinal contour line respectively, and obtain the transverse size and the longitudinal size corresponding to the same light area;

[0175] Based on the horizontal dimensions corresponding to the light regions, the horizontal dimension with the smallest corresponding value is determined as the horizontal contrast value; based on the vertical dimensions corresponding to the light regions, the vertical dimension with the smallest corresponding value is determined as the vertical contrast value;

[0176] Retrieving a horizontal reference value and a vertical reference value corresponding to the honeycomb type, and comparing the horizontal reference value and the vertical reference value with the horizontal comparison value and the vertical comparison value respectively;

[0177] In response to any comparison value being less than a corresponding reference value, the raw material substrate is determined to be unqualified.

[0178] For example, in this embodiment, when the substrate type of the raw material substrate is determined to be a honeycomb type, based on the structural characteristics of the honeycomb type, it can be known that the ventilation surface of the honeycomb type is generally composed of ventilation holes of various shapes arranged in an array. Therefore, the size of each rectangular area can be determined to perform a quality test on the raw material substrate; specifically, the following implementation process can be included:

[0179] First, the contours of each region corresponding to each light region can be obtained, and each region contour can be decomposed to obtain a horizontal contour line corresponding to the same region contour and a vertical contour line corresponding to the same region contour, wherein the same region contour should include two horizontal contour lines and two vertical contour lines;

[0180] Next, by determining the size of the transverse contour line and the longitudinal contour line, the transverse size and the longitudinal size corresponding to the same light area are obtained, and based on the transverse sizes corresponding to the light areas located on the raw material substrate, the transverse size with the smallest corresponding value is determined and determined as the transverse comparison value. Similarly, based on the longitudinal sizes corresponding to the light areas, the longitudinal size with the smallest corresponding value is determined as the longitudinal comparison value.

[0181] Next, the horizontal reference value and the vertical reference value of the corresponding cell type stored in the server are retrieved, and the horizontal reference value and the vertical reference value are compared with the horizontal comparison value and the vertical comparison value respectively;

[0182] Finally, when any contrast value (lateral contrast value and / or longitudinal contrast value) is less than the corresponding reference value, it indicates that the size of each light area included in the raw substrate does not meet the standard, so the structure of the quality test of the raw substrate can be determined as unqualified.

[0183] Similarly, in this embodiment, when the substrate type corresponding to the raw material substrate is a plate type or a corrugated type, the above-mentioned "calling a preset quality determination strategy corresponding to the substrate type to perform quality inspection on the raw material substrate" may also include the following steps:

[0184] When the substrate type corresponding to the raw substrate is a plate type or a corrugated type, performing a similarity comparison between each light region, and classifying each light region whose corresponding region similarity is greater than a preset second similarity into the same region shape group, thereby obtaining each region shape group;

[0185] Determine the area of ​​each light region in the same region shape group, and obtain the area of ​​each region corresponding to each light region;

[0186] The area of ​​the region with the smallest corresponding value is determined as the minimum area, and the area of ​​the region with the largest corresponding value is determined as the maximum area, and the average value of the area of ​​all regions except the minimum area and the maximum area is calculated to obtain the area average value;

[0187] In response to an area difference between the extremely small area and / or the extremely large area and the area average being greater than a preset area difference, the raw material substrate is determined to be unqualified.

[0188] For example, it can be known from the above content that the substrate structure included in any type of substrate is generally symmetrical, so the same raw material substrate includes multiple light areas with the same or similar shapes. At this time, the light areas can be compared with each other respectively, and when the similarity of any area obtained is greater than the preset second similarity stored in the server, it indicates that the two light areas corresponding to the area similarity are the same or similar shapes. At this time, the two can be divided into the same area shape group, and then the area shape groups corresponding to different shapes are obtained; further, the area of ​​each light area in the same area shape group can be determined respectively to obtain the area of ​​each area corresponding to each light area, and the area of ​​the area with the smallest corresponding value is determined as the minimum area, and the area of ​​the area with the largest corresponding value is determined as the maximum area, and the average of all the remaining areas is further calculated to obtain the corresponding area average. Finally, the difference between the minimum area and the maximum area and the area average is calculated to obtain the corresponding area difference, and when any area difference is greater than the preset area difference, it indicates that the raw material substrate has light areas with large size differences, and the raw material substrate needs to be determined as unqualified.

[0189] According to the solution of the present invention, firstly, the method sequentially numbers the raw material substrates and determines the sampling conditions through an automated process, which greatly improves the production efficiency and management accuracy, ensures that each step of the operation has a basis, and reduces the possibility of human error; secondly, by determining the structure of the raw material substrates that meet the sampling conditions, and accurately configuring the light source unit and the shooting unit, the non-destructive and efficient detection of the internal structural characteristics of the raw material substrate is realized, which not only improves the accuracy of the detection, but also ensures the integrity of the raw material substrate, and provides a reliable basis for subsequent processing. Furthermore, by using light irradiation and image shooting technology, combined with advanced area recognition algorithms, the light area and its shape located in the substrate image can be quickly and accurately identified, so as to accurately determine the substrate type of the raw material substrate. The intelligence of this step greatly shortens the time of product classification and enhances the flexibility of the production line. Finally, the corresponding quality inspection strategy is automatically retrieved according to the identified substrate type, making the quality inspection process more scientific and rigorous, effectively ensuring the quality stability and consistency of the denitrification agent products, and improving the market competitiveness of the products and the corresponding detection efficiency.

[0190] Another embodiment of the present invention provides a highly efficient denitrification agent intelligent preparation system, Figure 6 The corresponding system block diagram is as follows: Figure 6 As shown, the system includes:

[0191] a sampling inspection determination module, configured to, in response to the completion of the manufacture of the raw material substrate based on the denitrification agent, sequentially number the raw material substrate according to the current numbering sequence, and determine whether the raw material substrate meets the preset sampling inspection condition based on the obtained substrate number;

[0192] a structure determination module configured to determine the structure of the raw material substrate in response to the raw material substrate meeting a preset sampling condition, and configure a light source unit on the first ventilation surface of the raw material substrate based on the determined structural characteristics;

[0193] a shooting configuration module, configured to control the light source unit to irradiate light toward the first ventilation surface, and to configure a shooting unit on a second ventilation surface of the raw material substrate that is arranged opposite to the first ventilation surface;

[0194] An image capturing module is configured to control the capturing unit to capture an image toward the second ventilation surface to obtain an image of the substrate;

[0195] A region recognition module is configured to perform region recognition on the substrate image, and determine the type of substrate corresponding to the raw substrate based on the obtained region shapes corresponding to the light regions in the substrate image;

[0196] The quality detection module is configured to call a preset quality determination strategy corresponding to the type of the substrate to perform quality detection on the raw material substrate.

[0197] In the description provided herein, algorithms and displays are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the examples of the present invention. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to implement the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the preferred embodiment of the present invention.

[0198] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0199] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof.

[0200] Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in the devices described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or may be divided into multiple submodules.

[0201] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of submodules or subunits or subcomponents.

[0202] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features from different embodiments is meant to be within the scope of the present invention and to form different embodiments.

[0203] In addition, some of the embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices that perform the functions. Therefore, a processor with necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein of the device embodiments are examples of devices for implementing the functions performed by the elements for the purpose of implementing the invention.

[0204] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved, and is not intended to imply that the objects so described must have a given order in time, space, order, or in any other manner.

[0205] Although the present invention has been described according to a limited number of embodiments, it will be apparent to those skilled in the art, with the benefit of the above description, that other embodiments may be envisioned within the scope of the invention thus described. In addition, it should be noted that the language used in this specification is primarily selected for readability and instructional purposes, rather than for explaining or limiting the subject matter of the present invention.

Claims

1. A method for preparing an efficient denitrifying agent intelligently, characterized in that: The following steps are involved: In response to the completion of manufacturing the raw material substrate based on the denitrification agent, sequentially numbering the raw material substrate according to the current numbering sequence, and judging whether the raw material substrate meets the preset sampling condition based on the obtained substrate number; In response to the raw material substrate meeting the preset sampling condition, the raw material substrate is structurally determined, and a light source unit is configured on the first ventilation surface of the raw material substrate based on the determined structural characteristics; Controlling the light source unit to irradiate light toward the first ventilation surface, and configuring a shooting unit on a second ventilation surface of the raw material substrate that is arranged opposite to the first ventilation surface; Controlling the shooting unit to shoot an image toward the second ventilation surface to obtain a substrate image; Performing region recognition on the substrate image, and determining the type of substrate corresponding to the raw substrate based on the obtained region shapes corresponding to the light regions in the substrate image; A preset quality determination strategy corresponding to the type of the substrate is retrieved to perform quality inspection on the raw material substrate.

2. The intelligent preparation method of a high-efficiency denitrification agent according to claim 1, characterized in that: Judging whether the raw material substrate meets the preset sampling conditions based on the obtained substrate number includes: Decomposing the substrate number to obtain a primary number and a secondary number; Determining a production grade corresponding to the raw material substrate based on the main grade number, and determining a sampling frequency corresponding to the production grade based on a preset grade sampling rule, wherein different production grades correspond to different sampling frequencies; Compare the secondary number with each sampling number corresponding to the sampling frequency; In response to the secondary number being equal to any sampling number corresponding to the sampling frequency, the raw material substrate is judged to meet the preset sampling condition; In response to the secondary number not being equal to any sampling number corresponding to the sampling frequency, determining the sampling number that is in a positive proximity relationship with the secondary number as the target number, and determining the number difference between the target number and the secondary number; When the number difference is less than or equal to the preset difference, the sampling result of the substrate number equivalent to the target number is retrieved based on the historical sampling library, and if the sampling result is unqualified, the raw material substrate corresponding to the secondary number is judged to meet the preset sampling conditions.

3. The intelligent preparation method of a high-efficiency denitrification agent according to claim 2, characterized in that: The method further comprises: In response to the number of random inspections corresponding to the same production level reaching a preset number of random inspections, determining the random inspection results corresponding to the preset number of random inspections as the same random inspection batch; In response to the obtained sampling batches reaching a preset batch number, determining the sampling quality corresponding to the sampling batch based on the sampling results in the same sampling batch; Sort the sample inspection batches based on the chronological order, and determine whether the quality of each sample inspection shows a gradually decreasing trend or a gradually increasing trend based on the obtained sample inspection order; If there is a gradual upward trend, the sampling frequency is reduced through training; If there is a gradual downward trend, the sampling frequency is increased through training; Among them, the sampling frequency after training is obtained by the following formula: Among them, q + is the sampling frequency k s Increase the number of training times, ∈ is the sampling frequency k s The training constant value, q- is the sampling frequency k s Reduce the number of training times, k w is the sampling frequency after training.

4. The intelligent preparation method of a high-efficiency denitrification agent according to claim 1, characterized in that: In response to the raw material substrate meeting the preset sampling condition, the raw material substrate is structurally determined, and a light source unit is configured on the first ventilation surface of the raw material substrate based on the determined structural characteristics, including: Determine any ventilation surface of the raw material substrate as a first ventilation surface, and collect an image of the first ventilation surface to obtain a raw material ventilation image; Binarizing the raw material ventilation image to obtain substrate pixel points corresponding to the first pixel value and noise pixel points corresponding to the second pixel value; Acquire a first number corresponding to each substrate pixel and a second number corresponding to each noise pixel, and determine a substrate density corresponding to the raw material substrate based on the first number and the second number; Performing length detection on the raw material substrate based on a preset extension length strategy to determine a substrate length corresponding to the raw material substrate; A fusion calculation is performed based on the substrate density and the substrate length, and a light source unit is configured on the first ventilation surface of the raw substrate based on the obtained substrate coefficient.

5. The intelligent preparation method of a high-efficiency denitrification agent according to claim 4, characterized in that: Performing length detection on the raw material substrate based on a preset extension length strategy to determine a substrate length corresponding to the raw material substrate includes: Taking the image center point of the raw material ventilation image as the coordinate origin, establishing an image coordinate system corresponding to the raw material ventilation image, and acquiring each image coordinate point constituting the raw material ventilation image based on the image coordinate system; In each image coordinate point, each coordinate point corresponding to the horizontal coordinate extreme value and the vertical coordinate extreme value is respectively determined as each detection coordinate point, and each detection point position having a mapping relationship with each detection coordinate point is determined in the first ventilation surface corresponding to the raw material ventilation image; Determine a ventilation surface of the raw material substrate that is opposite to the first ventilation surface as a second ventilation surface, use the second ventilation surface as a horizontal plane, perform distance detection on each detection point, and obtain each distance detection value corresponding to each detection point; Determine the detection value with the smallest corresponding value among the distance detection values ​​as the reference detection value, and determine the detection point corresponding to the reference detection value as the reference point; The point-to-surface distance between the second ventilation surface and the reference point along an extension direction perpendicular to the second ventilation surface of the raw material substrate is determined as a substrate length corresponding to the raw material substrate.

6. The intelligent preparation method of a high-efficiency denitrification agent according to claim 4, characterized in that: Performing a fusion calculation based on the substrate density and the substrate length, and configuring a light source unit on the first ventilation surface of the raw substrate based on the obtained substrate coefficient, including: Normalizing the substrate density and the substrate length respectively to obtain a corresponding first density coefficient and a first length coefficient; The first density coefficient and the first length coefficient are respectively multiplied by the retrieved density weight value and the retrieved length weight value to obtain a second density coefficient and a second length coefficient; The second density coefficient and the second length coefficient are superimposed and calculated to obtain a substrate coefficient corresponding to the raw substrate, and a preset light source comparison table is retrieved, wherein the light source comparison table includes different coefficient intervals and light source specifications corresponding to different coefficient intervals. The preset light source comparison table is traversed, and the light source specification corresponding to the substrate coefficient is determined based on the coefficient interval including the substrate coefficient, and the light source unit is configured on the first ventilation surface of the raw substrate based on the light source specification.

7. The intelligent preparation method of a high-efficiency denitrification agent according to claim 1, characterized in that: Performing region recognition on the substrate image, and determining the type of substrate corresponding to the raw substrate based on the obtained region shapes corresponding to the light regions in the substrate image, including: Performing pixel recognition on the obtained substrate image to obtain each light pixel point of the corresponding light source light in the substrate image, and performing pixel connection on each light pixel point based on the presented connectivity to obtain each light area; Taking the area center point of each light area as the rotation point, the light area is rotated vertically in sequence to obtain a first rotation area, a second rotation area, and a third rotation area corresponding to the same light area, and the original area shape corresponding to the light area, the first rotation shape corresponding to the first rotation area, the second rotation shape corresponding to the second rotation area, and the third rotation shape corresponding to the third rotation area are determined as a shape comparison group corresponding to the light area; Retrieving standard shapes corresponding to honeycomb types, corrugated types and plate types, and performing similarity comparison between shape comparison groups corresponding to light regions and standard shapes; In response to a shape similarity between any shape in any shape comparison group and any standard shape being greater than a preset first similarity, determining a light region corresponding to the shape comparison group as a similar shape corresponding to the standard shape; In response to a shape ratio of similar shapes corresponding to any standard shape being greater than a preset ratio, the substrate type corresponding to the raw material substrate is determined as the type corresponding to the standard shape.

8. The intelligent preparation method of a high-efficiency denitrification agent according to claim 7, characterized in that: Retrieving a preset quality determination strategy corresponding to the type of substrate to perform quality inspection on the raw substrate includes: When the substrate type corresponding to the raw substrate is a honeycomb type, the contours of each region corresponding to each light region are obtained, and the contours of the region are decomposed to obtain a horizontal contour line corresponding to the horizontal direction and a longitudinal contour line corresponding to the vertical direction; Determine the size of the transverse contour line and the longitudinal contour line respectively, and obtain the transverse size and the longitudinal size corresponding to the same light area; Based on the horizontal dimensions corresponding to the light regions, the horizontal dimension with the smallest corresponding value is determined as the horizontal contrast value; based on the vertical dimensions corresponding to the light regions, the vertical dimension with the smallest corresponding value is determined as the vertical contrast value; Retrieving a horizontal reference value and a vertical reference value corresponding to the honeycomb type, and comparing the horizontal reference value and the vertical reference value with the horizontal comparison value and the vertical comparison value respectively; In response to any comparison value being less than a corresponding reference value, the raw material substrate is determined to be unqualified.

9. The intelligent preparation method of a high-efficiency denitrification agent according to claim 7, characterized in that: Retrieving a preset quality determination strategy corresponding to the type of substrate to perform quality inspection on the raw substrate includes: When the substrate type corresponding to the raw substrate is a plate type or a corrugated type, performing a similarity comparison between each light region, and classifying each light region whose corresponding region similarity is greater than a preset second similarity into the same region shape group, to obtain each region shape group; Determine the area of ​​each light region in the same region shape group, and obtain the area of ​​each region corresponding to each light region; The area of ​​the region with the smallest corresponding value is determined as the minimum area, and the area of ​​the region with the largest corresponding value is determined as the maximum area, and the average value of the area of ​​all regions except the minimum area and the maximum area is calculated to obtain the area average value; In response to an area difference between the extremely small area and / or the extremely large area and the area average being greater than a preset area difference, the raw material substrate is determined to be unqualified.

10. An efficient denitrification agent intelligent preparation system, characterized in that: include: a sampling inspection determination module, configured to, in response to completion of manufacturing of a raw material substrate based on a denitrifying agent, sequentially number the raw material substrate according to a current numbering sequence, and determine whether the raw material substrate meets a preset sampling inspection condition based on the obtained substrate number; a structure determination module configured to determine the structure of the raw material substrate in response to the raw material substrate meeting a preset sampling condition, and configure a light source unit on the first ventilation surface of the raw material substrate based on the determined structural characteristics; a shooting configuration module, configured to control the light source unit to irradiate light toward the first ventilation surface, and to configure a shooting unit on a second ventilation surface of the raw material substrate that is arranged opposite to the first ventilation surface; An image capturing module is configured to control the capturing unit to capture an image toward the second ventilation surface to obtain an image of the substrate; A region recognition module is configured to perform region recognition on the substrate image, and determine the type of substrate corresponding to the raw substrate based on the obtained region shapes corresponding to the light regions in the substrate image; The quality detection module is configured to call a preset quality determination strategy corresponding to the type of the substrate to perform quality detection on the raw material substrate.

Images