Control device and method using pyrolysis raw material and product image information

Abstract

To provide a control apparatus and method using a pyrolysis raw material and product image information.SOLUTION: A pyrolysis reaction method comprises: a raw material input step of acquiring first data while inputting a raw material; a pyrolysis reaction step of moving the raw material while pyrolyzing the same; a reaction product discharge step of acquiring second data while discharging a pyrolysis reaction product; an image sample collection step of additionally securing the number of image samples of the first data and the second data; a DB storage step of normalizing the secured image samples into time-series data and storing RGB code values in a DB; a labeling step of assigning labels to the first data and the second data based on the stored DB values; a data processing step of generating third data as estimated moisture content and fourth data as estimated calorific value by utilizing the first data and the second data; and a control value derivation step of deriving control values for pyrolysis temperature and pyrolysis time required to ensure a calorific value-based quality of the pyrolyzed reaction product based on the third data and the fourth data.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a control device and method using pyrolysis raw material and product image information. More specifically, it monitors the input solid raw material and pyrolysis solid product image, and enables efficient pyrolysis reaction through real-time AI analysis and operation variable control.

Background Art

[0002] Biofuels produced from biomass are decomposed into water and carbon dioxide during combustion, and the carbon dioxide thus emitted is reabsorbed into the biomass by photosynthesis as the organisms grow and is once again included in the components of the biomass. Therefore, it is defined as a carbon-neutral fuel. However, due to the characteristics of biomass, the water content is high, resulting in a low calorific value. As a result, the operation efficiency and energy production efficiency during its combustion decrease, and there is a demand for improving the quality of biomass fuel through a semi-carbonization process at a temperature of 250°C to 300°C under anaerobic conditions.

[0003] In addition, the carbon in biochar produced by pyrolyzing biomass at 350°C or higher under limited oxygen conditions is rearranged in a stable structural form and can isolate carbon in the soil for a long time without being decomposed by microorganisms or the like when introduced into the soil. Therefore, in recent years, it has received much attention as a technology for mitigating climate change. Biochar can isolate the carbon source introduced into the soil semi-permanently in the soil due to its safety in the soil, reduce greenhouse gas emissions, and has the effect of increasing crop yields through soil improvement.

[0004] The most common reactor configuration in biomass pyrolysis is the rotary kiln, where heat and material are transferred between biomass particles and a heat source medium by a kiln rotating around a rotating axis. Rotary kiln reactors are a proven technology in various application fields and can directly or indirectly pyrolyze biomass using hot air generated through combustion. The flow of feed material and heat source medium can be configured as parallel or reverse flow, and is generally configured as parallel flow to promote drying. As each biomass particle moves while rotating due to the rotation of the kiln, the residence time of the biomass in the reactor can be adjusted by the rotation speed and inclination of the kiln.

[0005] Image recognition and interpretation technology, made possible by recent advancements in AI technology, enables the recognition and classification of surrounding objects. By learning from diverse types of images, it can identify identifiable features such as hue and form, accurately recognize each feature by cross-referencing with thousands of other images, and assign labels. As a result, meaningful applications of AI object recognition technology are increasing as it is integrated into the quality control processes of recent products.

[0006] Conventional rotary kiln technology controls operation, such as hot air flow rate and rotation speed, solely based on reaction temperature measured by thermocouples. This makes it difficult to proactively respond to fluctuations caused by the raw material intake conditions, hindering the assurance of final product quality and yield. Ultimately, the economic viability of product production becomes a problem. Therefore, it is necessary to utilize real-time quality prediction based on image recognition and analysis of raw materials and products as operational parameters for controlling hot air flow rate, kiln rotation speed, and other parameters.Korean Published Patent No. 10-1807077 relates to an indirect rotary kiln reactor, comprising: a dry material inlet for introducing a dry material; a dry material outlet for discharging the introduced dry material; an inner cylinder which is a passage through which the introduced dry material moves; an outer cylinder surrounding the inner cylinder and which is a passage through which hot air supplied from the outside moves; a hot air supply pipe for supplying hot air to the outer cylinder; and a hot air discharge pipe for discharging the supplied hot air, wherein the hot air supply pipe is located in the outer cylinder. Hot air is supplied to at least two or more separate areas, the hot air and the drying material do not come into direct contact, the outer cylinder includes a section plate that divides the outer cylinder into at least two or more areas corresponding to the hot air supplied from the hot air supply pipe, the section plate is formed such that one end is connected to the outer cylinder and the other end is at a certain distance from the inner cylinder, the hot air supplied via the hot air supply pipe can move to adjacent areas through the distance from the inner cylinder, and the section plate in the upper space of the outer cylinder divides the lower space of the outer cylinder The present invention provides an indirect rotary kiln reactor comprising: a hot air branch pipe, which is formed alternately with plates and branches off from the hot air supply pipe to supply the hot air to each section of the outer cylinder divided by the section plates, the hot air branch pipe includes a control valve that can adjust the amount of hot air flowing inside it; a multi-point thermocouple installed in the inner cylinder to sense the internal reaction temperature of the inner cylinder corresponding to the sections divided by the section plates installed in the outer cylinder; and a heat source supply amount control unit that adjusts the control valve based on the internal reaction temperature of the inner cylinder measured by the multi-point thermocouple; a generated gas discharge pipe connected to the inner cylinder for discharging gas products generated during the drying process of the material by the hot air; the generated gas discharge pipe and the hot air discharge pipe are formed in a double structure so that the generated gas discharge pipe passes inside the hot air discharge pipe; and the hot air for drying the material is adjusted by the reaction temperature gradient of the inner cylinder and supplied to the outer cylinder in specific sections.

[0007] Japanese Patent Publication No. 2008-180451 relates to an externally heated rotary kiln and a method of operating the same, and provides an externally heated rotary kiln comprising an inner kiln cylinder that rotates in the axial direction and an outer cylinder through which heating gas flows around the inner kiln cylinder, wherein the workpiece is heated while being transported in the axial direction inside the inner kiln cylinder, the inner kiln cylinder is rotatably supported at a movable end and a fixed end that are movable in the axial direction, and is provided with means for measuring the amount of thermal growth in the axial direction of the inner kiln cylinder and a plurality of non-contact thermometers for measuring the shell temperature at a plurality of positions in the axial direction of the inner kiln cylinder on the peripheral wall of the outer cylinder.

[0008] Japanese Patent Publication No. 6090994 relates to a method for producing carbide and a method for inspecting the quality of carbide, and provides a method for producing carbide from biomass using a carbonizer, characterized in that a first carbonization condition corresponding to carbide having desired pulverability is identified using a profile obtained by measuring the optical properties indicating the hue of the carbide and the pulverability of the carbide for each carbonization condition in which the carbide was produced, the optical properties of the carbide produced in the carbonizer are measured, a second carbonization condition corresponding to the optical properties is identified using the profile, and the carbonizer is controlled so that the second carbonization condition matches the first carbonization condition.

[0009] Korean Registered Patent Publication No. 10-1728665 relates to a method for predicting the calorific value of semi-carbonized biomass using a color difference measurement method, and provides a method for predicting the calorific value of semi-carbonized biomass using a color difference measurement method, characterized by comprising the steps of processing and drying woody biomass; introducing the woody biomass into a semi-carbonization reactor, creating an oxygen-free atmosphere by introducing nitrogen gas into the reactor while heating it, and heat-treating it under high-temperature inert conditions to semi-carbonize it; and crushing the biomass in order to measure the color difference between raw woody biomass and semi-carbonized biomass, compressing the resulting wood powder to make a disc, and measuring this with a colorimeter.

[0010] Conventional technology merely senses the internal reaction temperature of the kiln and controls the reaction temperature through control valves, etc., but does not disclose a method for tracking the heat generation of the product in real time based on its hue using an image viewer such as a camera on the discharge side where the manufactured product is discharged, nor does it disclose a method for controlling the equipment. As a result, it has the problem of having to stop continuous operation of the equipment in order to efficiently manage the energy of the equipment and to check the condition of the product. [Prior art documents] [Patent Documents]

[0011] [Patent Document 1] Korean Registered Patent Publication No. 10-1807077 [Patent Document 2] Japanese Patent Publication No. 2008-180451 [Patent Document 3] Patent No. 6090994 [Patent Document 4] Korean Registered Patent Publication No. 10-1728665 [Overview of the project] [Problems that the invention aims to solve]

[0012] Conventional technology has the problem that it is impossible to control operation through observation in the environment inside the reactor where it is difficult to secure a clear view, and that operation such as the amount of hot air and rotation speed is controlled solely on the reaction temperature measured from thermocouples, making it difficult to ensure the quality of the final product due to the difficulty in responding to fluctuations caused by the intake conditions of the target raw materials.

[0013] Furthermore, conventional technology requires a certain amount of time to confirm the moisture content of the raw material being drawn in and the calorific value of the product through separate analysis, making it impossible to use it as a parameter for real-time operation control. Therefore, it is necessary to utilize it as an operation parameter for controlling the hot air flow rate and kiln rotation speed through real-time estimation. [Means for solving the problem]

[0014] To achieve this objective, the present invention comprises: a raw material input step in which raw materials are introduced and first data is acquired through a first monitoring means; a pyrolysis reaction step in which the introduced raw materials are moved while being pyrolyzed; a reaction product discharge step in which the pyrolyzed reaction products are discharged and second data is acquired through a second monitoring means; an image sample collection step in which the accuracy of the first and second data is improved and an additional number of image samples are secured; a DB storage step in which the secured image samples are normalized into time-series data and the RGB code values ​​are stored in a DB; a labeling step in which labels are assigned to the first data and second data based on the stored DB values; and the labels of the first data are assigned RG by assigning a moisture content evaluation value. The present invention provides a thermal decomposition reaction method that includes a data processing step in which a correlation between the B value and moisture content is derived, a second data label is assigned a calorific value evaluation value to derive a correlation between the RGB value and calorific value, and after comparing it with the data stored in the DB through simulation, a third data which is an estimated moisture content and a fourth data which is an estimated calorific value are generated using the first and second data; and a control value derivation step in which control values ​​for thermal decomposition temperature and thermal decomposition time necessary to ensure the calorific value standard quality of the reactant reacted by thermal decomposition are derived based on the third and fourth data generated through the data processing step, wherein the thermal decomposition reaction step provides a thermal decomposition reaction method that uses thermal decomposition reaction means in which one or more thermal decompositions from semi-carbonization, biochar reaction, drying, activated carbon reaction and carbonization are carried out.

[0015] Furthermore, in the control value derivation step, the pyrolysis temperature can be controlled by controlling one or more of the following: the rotation speed of the raw material supply motor of the hot blast furnace, the rotation speed of the air supply motor of the hot blast furnace, and the rotation speed of the dilution air supply motor of the gas mixer.

[0016] Furthermore, in the control value derivation step, the pyrolysis time can be controlled by controlling one or more of the rotational speed of the rotary kiln's drive motor and the rotational speed of the raw material supply motor.

[0017] Furthermore, the system includes a raw material input means for acquiring first data through a first monitoring means while raw materials are being input; a pyrolysis reaction step in which the input raw materials are moved while being pyrolyzed; a reaction material discharge means for acquiring second data through a second monitoring means while the pyrolyzed reaction products are discharged; an image sample collection means for improving the accuracy of the first and second data and securing an additional number of image samples; a database storage means for normalizing the secured image samples into time-series data and storing the RGB code values ​​in a database; a labeling step for assigning labels to the first and second data based on the stored database values; and a mechanism for assigning moisture content evaluation values ​​to the labels of the first data to determine the relationship between the RGB values ​​and the moisture content. The invention provides a data processing means that derives a relationship, assigns a calorific value evaluation value to the label of the second data to derive a correlation between the RGB value and the calorific value, compares it with the data stored in the DB through simulation, and then uses the first and second data to generate a third data which is the estimated moisture content and a fourth data which is the estimated calorific value; and a control value derivation means that derives control values ​​for the thermal decomposition temperature and thermal decomposition time necessary to ensure the calorific value standard quality of the reactant reacted by the thermal decomposition reaction based on the third and fourth data generated through the data processing step, and the thermal decomposition reaction step provides a thermal decomposition reaction apparatus that uses a thermal decomposition reaction means in which one or more of the thermal decompositions of semi-carbonization, biochar reaction, drying, activated carbon reaction and carbonization are carried out.

[0018] Furthermore, the present invention can also be provided in forms that combine various means for solving the aforementioned problems. [Effects of the Invention]

[0019] The present invention enables real-time analysis of the moisture content of raw materials fed into the pyrolysis reactor and the heat generated by the discharged reactants, allowing for efficient energy management and operation. [Brief explanation of the drawing]

[0020] [Figure 1]This is a diagram showing a pyrolysis reactor in one embodiment of the present invention. [Figure 2] In one embodiment of the present invention, it is a photograph of the form of semicarbide, which is a reaction product depending on the reaction temperature of the raw material input during semi-carbonization. [Figure 3] In one embodiment of the present invention, it is the experimental results of the calorific value, yield and elemental composition of EFB (Empty Fruit Bunch) semicarbide depending on the reaction temperature. [Figure 4] In one embodiment of the present invention, it is the experimental results of the calorific value, yield and elemental composition of EFB (Empty Fruit Bunch) semicarbide depending on the reaction time. [Figure 5] In one embodiment of the present invention, it is the experimental results of the calorific value of EFB (Empty Fruit Bunch) semicarbide depending on the reaction temperature and reaction time. [Figure 6] In one embodiment of the present invention, it is a diagram of the step of dividing the images of the first data and the second data, converting the image hue for each cell into RGB code values, and then deriving the average value. [Figure 7] In one embodiment of the present invention, it is a diagram of the step of overlapping the images of the first data and the second data to generate an image. [Figure 8] It is a diagram of the labeling step of grouping RGB data code values and assigning labels. [Figure 9] It is a diagram of the data processing step of deriving the correlation between RGB values, moisture content and calorific value, and estimating the moisture content and calorific value corresponding to any RFG value through simulation.

Mode for Carrying Out the Invention

[0021] Hereinafter, with reference to the attached drawings, embodiments that allow a person with ordinary skill in the art to carry out the present invention will be described in detail. However, in describing the operating principle of a preferred embodiment of the present invention in detail, if it is determined that a specific description of a related known function or configuration may obscure the gist of the present invention, such a detailed description will be omitted.

[0022] Furthermore, the same reference numerals shall be used for parts that have similar functions and operations throughout the entire drawing. Throughout the specification, when one part is said to be connected to another part, this includes not only direct connections but also indirect connections through other elements in between. Also, the inclusion of one component does not exclude other components unless otherwise stated, but rather means that other components may be included.

[0023] Furthermore, limitations or additions to one embodiment described herein may apply not only to that particular embodiment but also to other embodiments.

[0024] Furthermore, throughout the description of this invention and the claims, the singular notation includes plural nouns unless otherwise specified.

[0025] Figure 1 shows a pyrolysis reaction apparatus in one embodiment of the present invention; Figure 2 is a photograph of the form of the semi-carbonized product, which is the reaction product, depending on the reaction temperature of the raw materials introduced during semi-carbonization in one embodiment of the present invention; Figure 3 shows experimental results for the heat generation, yield, and elemental composition of EFB (Empty Fruit Bunch) semi-carbonized product depending on the reaction temperature in one embodiment of the present invention; Figure 4 shows experimental results for the heat generation, yield, and elemental composition of EFB (Empty Fruit Bunch) semi-carbonized product depending on the reaction time in one embodiment of the present invention; Figure 5 shows experimental results for EFB (Empty Fruit Bunch) semi-carbonized product depending on the reaction temperature and reaction time in one embodiment of the present invention. In the experimental results for the calorific value of the semi-carbide, Figure 6 illustrates the steps in one embodiment of the present invention in which the images of the first and second data are divided, the image hue for each cell is converted to an RGB code value, and then the average value is derived. Figure 7 illustrates the steps in one embodiment of the present invention in which the images of the first and second data are overlapped to generate an image. Figure 8 illustrates the labeling steps in which the RGB data code values ​​are grouped and labels are assigned. Figure 9 illustrates the data processing steps in which the correlation between RGB values, moisture content, and calorific value is derived, and the moisture content and calorific value corresponding to an arbitrary RFG value are estimated through simulation.

[0026] Figure 6 illustrates the process of dividing the images of the first and second data sets evenly into cells of 9 points or more, converting the image hue of each cell into RGB (Red, Green, Blue) code values, and then deriving the average value for each cell.

[0027] Figure 7 illustrates the process of acquiring images of the first and second data sets four times at 2.5-second intervals, overlapping them in a 2:8 ratio to create a single image, and then generating the image to be applied to Figure 6.

[0028] Figure 8 illustrates the process of classifying each code value of the RGB data into 16 groups, and then assigning a total of 4,069 labels through combinations of the classified R, G, and B groups.

[0029] Furthermore, after the image and labeling generated through Figure 7, one or more of the calorific value and water content can be predicted through a sequence including the following steps.

[0030] 1) The stage of preparing labeled data so that collected image data can be used for training (this may include data augmentation).

[0031] 2) The stage of training a model to predict heat generation / moisture content.

[0032] 3) Adjusting the model's weights and learning the relationships between heat generation, water content, image, and the label.

[0033] 4) The stage of adjusting the model's hyperparameters or performing tuning work to improve the model's performance through cumulative data.

[0034] The present invention comprises: a raw material input step in which raw materials are introduced and first data is acquired through a first monitoring means; a pyrolysis reaction step in which the introduced raw materials are moved while being pyrolyzed; a reaction product discharge step in which the pyrolyzed reaction products are discharged and second data is acquired through a second monitoring means; an image sample collection step in which the accuracy of the first and second data is improved and an additional number of image samples are secured; a DB storage step in which the secured image samples are normalized into time-series data and the RGB code values ​​are stored in a DB; a labeling step in which labels are assigned to the first data and second data based on the stored DB values; and the labels of the first data are assigned a moisture content evaluation value, thereby combining the RGB value and moisture content. The method provides a thermal decomposition reaction method that derives a correlation between the RGB values ​​and the amount of heat generated by deriving a correlation between the RGB values ​​and the amount of heat generated by assigning a heat generation evaluation value to the label of the second data, comparing it with the data stored in the DB through simulation, and using the first and second data to generate a third data which is the estimated moisture content and a fourth data which is the estimated amount of heat generated; and a control value derivation step that derives control values ​​for the thermal decomposition temperature and thermal decomposition time necessary to ensure the heat generation standard quality of the reactant reacted by the thermal decomposition reaction based on the third and fourth data generated through the data processing step, wherein the thermal decomposition reaction step provides a thermal decomposition reaction method that uses a thermal decomposition reaction means in which one or more thermal decompositions from semi-carbonization, biochar reaction, drying, activated carbon reaction and carbonization are carried out.

[0035] Furthermore, in the control value derivation step, the pyrolysis temperature can be controlled by controlling one or more of the following: the rotation speed of the raw material supply motor of the hot blast furnace, the rotation speed of the air supply motor of the hot blast furnace, and the rotation speed of the dilution air supply motor of the gas mixer.

[0036] Furthermore, in the control value derivation step, the pyrolysis time can be controlled by controlling one or more of the rotational speed of the rotary kiln's drive motor and the rotational speed of the raw material supply motor.

[0037] Furthermore, the first and second data may be visual information, and the visual information may be an image.

[0038] Furthermore, the first monitoring means can divide an image into a predetermined number of cells in order to acquire the first data, which is an image.

[0039] Furthermore, the second monitoring means can divide the image into a predetermined number of cells in order to acquire the second data, which is an image.

[0040] Furthermore, the images of the first and second data can be evenly divided into cells of 9 points or more.

[0041] Furthermore, after converting the image hue of each of the equally divided cells into RGB code values, the average value of each can be derived.

[0042] Furthermore, in the image sample acquisition stage, an 80% overlap based on 2.5 seconds can be applied to improve accuracy and secure an additional number of image samples.

[0043] The 2.5 seconds set in the image sample acquisition stage can be adjusted to allow for 10 seconds of dwell time analysis of the input fuel and discharged reactants through the relationship between the travel distance and motor rotation speed when using the conveyor belt. Therefore, it can be appropriately applied to situations where 2.5 seconds is not necessarily required.

[0044] Furthermore, by applying an overlap method that superimposes multiple images, a representative image can be generated, thereby improving system accuracy.

[0045] Furthermore, after classifying the RGB code values ​​into 16 groups, 4,096 labels can be assigned through combinations of the classified R, G, and B groups.

[0046] (Example 1) Control of pyrolysis temperature

[0047] Under operating conditions where the calorific value of the pyrolysis product derived from the fourth data point is 4,150 kcal / kg, the control step of the pyrolysis temperature-related operating factors to achieve the target calorific value of 4,500 kcal / kg.

[0048] (1) Increase the rotation speed of the raw material supply motor for the hot air furnace by 5%

[0049] (2) Increase the rotation speed of the hot air furnace's air supply motor by 7%

[0050] (3) Reduce the rotation speed of the dilution air supply motor of the gas mixer by 3%.

[0051] (Example 2) Control of pyrolysis time

[0052] Under operating conditions where the calorific value of the pyrolysis product derived from the fourth data point is 4,950 kcal / kg, the control step of the pyrolysis time-related operating factors to achieve the target calorific value of 4,500 kcal / kg.

[0053] (1) Increase the rotational speed of the rotary kiln's drive motor by 8%.

[0054] (2) Increase the rotation speed of the raw material supply motor by 5%

[0055] Furthermore, the system includes a raw material input means for acquiring first data through a first monitoring means while raw materials are being input; a pyrolysis reaction step in which the input raw materials are moved while being pyrolyzed; a reaction material discharge means for acquiring second data through a second monitoring means while the pyrolyzed reaction products are discharged; an image sample collection means for improving the accuracy of the first and second data and securing an additional number of image samples; a database storage means for normalizing the secured image samples into time-series data and storing the RGB code values ​​in a database; a labeling step for assigning labels to the first and second data based on the stored database values; and a mechanism for assigning moisture content evaluation values ​​to the labels of the first data to determine the relationship between the RGB values ​​and the moisture content. The invention provides a data processing means that derives a relationship, assigns a calorific value evaluation value to the label of the second data to derive a correlation between the RGB value and the calorific value, compares it with the data stored in the DB through simulation, and then uses the first and second data to generate a third data which is the estimated moisture content and a fourth data which is the estimated calorific value; and a control value derivation means that derives control values ​​for the thermal decomposition temperature and thermal decomposition time necessary to ensure the calorific value standard quality of the reactant reacted by the thermal decomposition reaction based on the third and fourth data generated through the data processing step, and the thermal decomposition reaction step provides a thermal decomposition reaction apparatus that uses a thermal decomposition reaction means in which one or more of the thermal decompositions of semi-carbonization, biochar reaction, drying, activated carbon reaction and carbonization are carried out.

[0056] Furthermore, the present invention provides a kiln-type indirect pyrolysis reaction means formed by an outer cylinder 200 that receives hot air from the outside and an inner cylinder 300 through which raw materials move; a pyrolysis reaction control means 900 for driving the pyrolysis means; a raw material inlet 110 located on one side of the pyrolysis reaction means into which raw materials are introduced; a reaction material outlet 120 located on the opposite side of the raw material inlet for discharging pyrolysis reaction products, which are the raw materials that have been pyrolyzed; a hot air supply pipe 400 for supplying hot air to the outer cylinder; a multi-point thermocouple 430 located on one side of the inner cylinder for measuring the internal temperature of the reactor; a heat source supply amount control means 440 connected to the multi-point thermocouple and hot air supply pipe of the inner cylinder for controlling the hot air supplied to the outer cylinder; a generated gas discharge pipe 600 located on one side of the upper part of the pyrolysis reaction means for discharging gases generated inside; and a means installed near the raw material inlet for introducing... The rotary pyrolysis reactor 100 includes a first monitoring means 700 for monitoring the raw materials being discharged and transmitting first data obtained through monitoring; a second monitoring means 710 installed near the reaction product discharge port for monitoring the discharged reaction products and transmitting second data obtained through monitoring; and an AI analysis means 800 that receives data transmitted from the first and second monitoring means, generates third and fourth data based on the received first and second data, provides the third data to the heat source supply amount control means and provides the fourth data to the pyrolysis reaction control means, wherein the outer cylinder includes area plates 210 for dividing into predetermined areas, and one side of the hot air supply piping and one side of the outer cylinder are connected by a hot air branch pipe 410 including a control valve 420.

[0057] Furthermore, the raw materials may be solid raw materials that require thermal decomposition such as drying, semi-carbonization, biochar, or carbonization, and may include, for example, biomass (bagasse, EFB, wood chips, etc.), organic waste (livestock manure, sewage sludge, food waste, etc.) and combustible waste (waste plastics, waste paper, waste rubber, etc.).

[0058] Furthermore, the predetermined areas separated by the area plates may consist of two or more areas.

[0059] Furthermore, multi-point thermocouples may be arranged to sense the temperature (T1 to T6) inside the inner cylinder in accordance with predetermined areas separated by the area plates.

[0060] Furthermore, the internal temperature, i.e., the reaction temperature, can be between 100°C and 500°C.

[0061] Furthermore, one or more of the multi-point thermocouples may be inserted near the raw material inlet or the reactant outlet, or they may be installed so as to penetrate near both the raw material inlet and the reactant outlet.

[0062] Furthermore, the control valve can adjust the amount of hot air supplied.

[0063] Furthermore, the outer cylinder may be installed in a manner that surrounds the inner cylinder with a certain distance between them, so as to form a passage for hot air to move between it and the inner cylinder.

[0064] Furthermore, one side of the outer cylinder may further include a hot air exhaust pipe 500 so that the supplied hot air can be discharged.

[0065] Furthermore, the hot air exhaust piping is formed in a double-walled structure that surrounds the generated gas exhaust piping, allowing the temperature of the generated gas exhaust piping to be maintained above a certain temperature, and enabling the generated gas to be used as fuel.

[0066] Furthermore, the hot air supply piping may be formed in a double-walled structure so that the generated gas discharge piping can be installed passing through the inside of the hot air supply piping.

[0067] Furthermore, when the hot air supply piping is located at the lower end of the pyrolysis reaction means, hot air can be supplied to the lower part of the pyrolysis reaction means so that indirect contact with the raw materials occurs most frequently. In this case, the utilization efficiency of the heat source can be slightly increased compared to when it is located at the upper end.

[0068] Furthermore, the thermal decomposition reaction means can be used in one or more of the following: partial carbonization, drying, biochar reaction, activated carbon reaction, and carbonization.

[0069] Furthermore, one or more of the first data and second data transmitted from the AI ​​analysis means can be based on visual information.

[0070] Furthermore, the first monitoring means can obtain moisture content information from visual information of the raw materials being input.

[0071] Furthermore, the second monitoring means can obtain heat generation information from visual information of the discharged reactants.

[0072] Furthermore, the heat source supply amount control means, upon receiving the third data, can adjust the control valve installed in the hot air branch pipe.

[0073] Furthermore, the pyrolysis reaction control means, upon receiving the fourth data, can adjust the rotation speed of the pyrolysis reaction means.

[0074] Furthermore, the hot air branch pipes can be connected to correspond to predetermined areas of the outer cylinder that are divided by the area plates.

[0075] A method for a pyrolysis reaction using the pyrolysis reactor may be provided, comprising: a raw material input step in which raw materials are introduced and first data is acquired through a first monitoring means; a pyrolysis reaction step in which the introduced raw materials are moved while being pyrolyzed; a reactant discharge step in which the pyrolyzed reactants are discharged and second data is acquired through a second monitoring means; a data processing step in which third and fourth data are generated based on the first and second data; and a pyrolysis reaction and apparatus control step in which pyrolysis and the apparatus are controlled based on the third and fourth data generated through the data processing step.

[0076] The aforementioned RGB refers to the three primary colors of light, and is a compound word of red, green, and blue. The RGB color model is a method of representing colors using the three primary colors of light.

[0077] Anyone with ordinary skill in the field to which this invention belongs will be able to make various applications and modifications within the scope of this invention based on the above. [Explanation of symbols]

[0078] 100 Rotary pyrolysis reactor 110 Raw material input port 120 Reactant discharge port 200 Outer cylinder 210 Area Board 300 inner cylinder 400 Hot air supply piping 410 Hot air branch pipe 420 Control valve 430 Multi-point thermocouple 440 Heat source supply amount control means 500 Hot air exhaust piping 600 Gas generation exhaust piping 700 First monitoring means 710 Second monitoring means 800 AI analysis tools 900 Thermal decomposition reaction control means

Claims

1. Raw material input stage: While raw materials are being introduced, first data, which is an image sample, is acquired through the first monitoring means; A thermal decomposition reaction step in which the aforementioned raw materials are moved while being thermally decomposed; A reaction product discharge step in which, while discharging the reaction product that has undergone the thermal decomposition reaction, a second data, which is an image sample, is acquired through a second monitoring means; An image sample acquisition step in which, in order to improve the accuracy of the first data and the second data, the images of the first data and the second data are overlapped to secure an additional number of image samples; The DB storage step involves dividing the secured image sample into cells, converting the image hue for each divided cell into an RGB code value, and storing the converted RGB code value in a DB; A labeling step in which labels are assigned to the first data and the second data based on the stored RGB code values; A data processing step in which a correlation between RGB values ​​and moisture content is learned using the first data with pre-prepared labels and the moisture content of the raw materials, a correlation between RGB values ​​and calorific value is learned using the second data with pre-prepared labels and the calorific value of the reactants, and a data processing step in which a third data, which is the estimated moisture content, and a fourth data, which is the estimated calorific value, are generated from the first and second data with labels using a trained model for calorific value / moisture content prediction that has been tuned by the RGB code values ​​stored in the DB; and A control value derivation step is included for deriving control values ​​for the thermal decomposition temperature and thermal decomposition time necessary to ensure the calorific value standard quality of the reactant reacted by thermal decomposition based on the third and fourth data generated through the data processing step; The aforementioned thermal decomposition reaction step is a thermal decomposition reaction method using thermal decomposition reaction means that carries out one or more thermal decompositions of semi-carbonization and carbonization.

2. The pyrolysis reaction method according to claim 1, wherein in the control value derivation step, the pyrolysis temperature is controlled by controlling one or more of the following: the rotational speed of the raw material supply motor of the hot blast furnace, the rotational speed of the air supply motor of the hot blast furnace, and the rotational speed of the dilution air supply motor of the gas mixer.

3. The pyrolysis reaction method according to claim 1, wherein in the control value derivation step, the pyrolysis time is controlled by controlling one or more of the rotation speed of the rotary kiln's drive motor and the rotation speed of the raw material supply motor.

4. A raw material input means that acquires first data, which is an image sample, through a first monitoring means while raw materials are being input; A thermal decomposition reaction means in which the introduced raw materials are moved while being thermally decomposed; A reactant discharge means that discharges the reaction product that has undergone the thermal decomposition reaction while acquiring second data, which is an image sample, through a second monitoring means; Image sample acquisition means for increasing the number of image samples by overlapping the images of the first and second data in order to improve the accuracy of the first and second data; A database storage means that divides the secured image sample into cells, converts the image hue for each divided cell into an RGB code value, and stores the converted RGB code values ​​in a database; Labeling means for assigning labels to the first data and the second data based on the stored RGB code values; A data processing means that learns the correlation between RGB values ​​and moisture content using the first data with pre-prepared labels and the moisture content of the raw materials, learns the correlation between RGB values ​​and calorific value using the second data with pre-prepared labels and the calorific value of the reactants, and generates a third data which is the estimated moisture content and a fourth data which is the estimated calorific value from the first data and the second data using a trained model for calorific value / moisture content prediction that has been tuned by the RGB code values ​​stored in the DB; and Includes a control value derivation means for deriving control values ​​for the thermal decomposition temperature and thermal decomposition time necessary to ensure the calorific value standard quality of the reactant reacted by thermal decomposition based on the third data and fourth data generated through the data processing means; The aforementioned pyrolysis reaction means is a pyrolysis reaction apparatus that uses a pyrolysis reaction means in which one or more of semi-carbonization and carbonization are carried out.

Images