Sintering batching method and system based on machine vision and LIBS (laser-induced breakdown spectroscopy) technology

Abstract

The invention provides a sintering burdening method and system based on machine vision and an LIBS technology. The method comprises the steps that raw material data for sintering are collected to establish a database; performing batching theoretical calculation by using the raw material data to obtain a theoretical batching table, performing empirical proportion calculation by using the raw material data to obtain an empirical batching table, and inputting the two batching tables into a preset neural network to predict an optimal batching table; feeding according to the optimal proportion data; the shape information of the materials is detected at a time through machine vision for the images of the materials on the belt; main components of the materials are detected at a discharging opening of the feeding machine; carrying out secondary detection from the process to the completion of material mixing, and adjusting material mixing process parameters according to the result of the secondary detection; and detecting for three times after crushing is completed in the process, and adjusting sintering process parameters according to results of the three times of detection, so as to complete the sintering process. Based on the method, the invention further provides a sintering batching system. According to the invention, the batching effect and the sinter quality are improved, and the stable and smooth production is ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of ironmaking, and in particular relates to a sintering batching method and system based on machine vision and LIBS technology. Background Art

[0002] The sintering production process is to provide "fine materials" for blast furnace smelting. Its essence is to mix various powdered iron raw materials with appropriate amounts of fuel and flux, and then mix and granulate them according to a certain proportion, add appropriate amounts of water, mix and granulate them. After a series of physical and chemical changes in the materials on the sintering equipment, the mineral powder particles are sintered and bonded into blocks. The sintering process is an important link in the pre-iron smelting process. However, there are many problems with the raw materials used in sintering production, such as a large variety, large fluctuations in chemical composition and particle size, and unstable sources of supply. These problems result in unstable sintered ore output and quality, large fluctuations in production costs and energy consumption. At the same time, the traditional trial-and-error ore blending model also has problems such as unstable ingredient composition, high cost, and weak ability to cope with changes, which affects the utilization rate of various resources. Therefore, seeking to develop a scientific and reasonable sintering optimization ore blending method is of great significance to improving the competitiveness of steel companies and improving pre-iron technical and economic indicators.

[0003] Sintering batch optimization is to find the batching scheme that meets the requirements of sintering ore output and quality indicators and has the lowest cost by establishing the intrinsic relationship between iron ore and sintering ore output and quality indicators. Intelligent optimization batching is to establish an iron ore sintering optimization batching model by using intelligent algorithms, taking the influence of the physical and chemical properties of iron ore on sintering ore as a constraint factor, selecting a suitable intelligent algorithm to solve the batching model, and finally obtaining the optimal ore batching scheme that meets production requirements and technical indicators. However, the intelligent mathematical models for optimizing batching in the existing technology mostly take sintering production technical indicators as the optimization target. The early intelligent optimization batching of sintering all took the lowest single cost as the optimization target, and then used the raw material supply limit, sintering ore chemical composition indicators and other constraints to establish the model and solve it. With the continuous in-depth research on sintering mechanism and the diversification of optimization targets, a single minimum cost indicator can no longer meet the requirements of steel production enterprises for sintering production technical indicators, especially under the theme of energy saving and consumption reduction, taking into account multi-objective sintering batching optimization issues such as cost, fuel consumption, and pollutant emissions. In view of the current situation where raw materials are unstable and volatile, resulting in large deviations in traditional batching calculations and untimely batching adjustments, it is urgent to establish a new optimization model, propose a new optimization method, and comprehensively consider and systematically solve it. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a sintering batching method and system based on machine vision and LIBS technology, which improves the batching effect and the quality of sintered ore, ensures stable and smooth production, and significantly improves production efficiency.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A sintering batching method based on machine vision and LIBS technology comprises the following steps:

[0007] Collecting raw material data for sintering to establish a sintering database; using the raw material data to perform theoretical batching calculations to obtain a theoretical batching table, the theoretical batching table is used to store theoretical proportion data of each material ore; using the raw material data to perform empirical batching calculations to obtain an empirical batching table, the empirical batching table is used to store empirical proportion data of each material ore; inputting the theoretical batching table and the empirical batching table into a preset neural network to predict an optimal batching table; the optimal batching table is used to store optimal proportion data of each material ore;

[0008] Feed materials according to the optimal proportion data of each material in the optimal batching table; obtain images of each material on the belt, and use machine vision to detect the shape information of the material based on the images of each material; and use LIBS at the feeder discharge port to detect the main components of the material at one time;

[0009] After the sintering process is completed to the mixing, machine vision and LIBS are used for secondary inspection, and the mixing process parameters are adjusted according to the results of the secondary inspection; after the sintering process is completed to the crushing, machine vision and LIBS are used for tertiary inspection, and the sintering process parameters are adjusted according to the results of the tertiary inspection to complete the sintering process.

[0010] Furthermore, the raw material data for sintering includes: the type of raw materials for sintering, raw material composition and physicochemical parameters, and material prices.

[0011] Furthermore, the process of using the raw material data to perform theoretical calculation of ingredients to obtain a theoretical ingredient list includes:

[0012] G = 60 × batch size (kg / m) × batching ratio (%) × belt speed (m / min)

[0013] Wherein, G is the feeding amount of any raw material per hour;

[0014] The consumption of any raw material per shift is: M = T × G;

[0015] Among them, M is the consumption of any raw material per shift; T is the loading time during the shift.

[0016] Furthermore, the process of using the raw material data to perform empirical proportioning calculations to obtain an empirical ingredient table includes: assuming a proportioning ratio based on actual production experience, and calculating the chemical composition of the sintered ore based on the original data of various materials to see whether it meets the requirements of specified indicators; the original data includes material moisture, burn loss and chemical composition.

[0017] Furthermore, the preset neural network adopts a symmetrically connected neural network.

[0018] Furthermore, the process of performing LIBS at the feeder discharge port to detect the main components of the material includes: using LIBS technology to emit laser, and directly obtaining the element type and content of the sample based on the spectrum combined with the quantitative analysis algorithm.

[0019] Furthermore, the mixing process includes adding water to moisten, slurry homogenization and balling.

[0020] Furthermore, after adjusting the mixing process parameters according to the results of the secondary test, it also includes spreading, ignition sintering and crushing;

[0021] The process of laying the materials includes laying the base material and the mixed material on the sintering machine trolley;

[0022] During the ignition sintering, the sintering air volume, vacuum degree, material layer thickness, machine speed and sintering end point are precisely controlled;

[0023] The crushing process includes making the sintered block slide onto a single roll crusher to be sheared and crushed.

[0024] Furthermore, the process of adjusting the sintering process parameters according to the results of the three tests includes: comparing the results of the three tests with the sintering database, correcting the ingredient list, adjusting the mixture and adjusting the sintering process parameters.

[0025] The present invention also proposes a sintering batching system based on machine vision and LIBS technology, including: a pre-processing module, a primary detection module and a multiple detection module;

[0026] The pre-processing module is used to collect raw material data for sintering to establish a database; use the raw material data to perform theoretical batching calculations to obtain a theoretical batching table, which is used to store theoretical proportion data of each raw material ore; use the raw material data to perform empirical batching calculations to obtain an empirical batching table, which is used to store empirical proportion data of each raw material ore; input the theoretical batching table and the empirical batching table into a preset neural network to predict an optimal batching table; the optimal batching table is used to store optimal proportion data of each raw material ore;

[0027] The primary detection module is used to feed materials according to the optimal proportion data of each material in the optimal batching table; obtain images of each material on the belt, and use machine vision to detect the shape information of the material based on the images of each material; and perform LIBS at the feeder discharge port to detect the main components of the material;

[0028] The multiple detection module is used to perform secondary detection using machine vision and LIBS after the sintering process is completed to mix the materials, and adjust the mixing process parameters according to the results of the secondary detection; after the sintering process is completed to crush the materials, the machine vision and LIBS are used to perform tertiary detection, and adjust the sintering process parameters according to the results of the tertiary detection to complete the sintering process flow.

[0029] The effects provided in the content of the invention are only the effects of the embodiments, not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:

[0030] The present invention proposes a sintering batching method and system based on machine vision and LIBS technology, the method comprising the following steps: collecting raw material data for sintering to establish a sintering database; using the raw material data to perform theoretical batching calculations to obtain a theoretical batching table, the theoretical batching table is used to store theoretical proportion data of each material ore; using the raw material data to perform empirical batching calculations to obtain an empirical batching table, the empirical batching table is used to store empirical proportion data of each material ore; inputting the theoretical batching table and the empirical batching table into a preset neural network to predict an optimal batching table; the optimal batching table is used to store optimal proportion data of each material ore; according to the optimal batching table, ... The optimal proportion data of each material ore in the optimal batching table is used for feeding; the image of each material on the belt is obtained, and the shape information of the material is detected by machine vision based on the image of each material; and the main components of the material are detected by LIBS at the feeder discharge port; after the sintering process is completed to the mixing, the machine vision and LIBS are used for secondary detection, and the mixing process parameters are adjusted according to the results of the secondary detection; after the sintering process is completed to the crushing, the machine vision and LIBS are used for tertiary detection, and the sintering process parameters are adjusted according to the results of the tertiary detection to complete the sintering process. Based on a sintering batching method based on machine vision and LIBS technology, a sintering batching system based on machine vision and LIBS technology is also proposed. The present invention establishes a database by collecting original data information of various raw materials for sintering, constructs a batching model to perform theoretical computer empirical proportion calculation of the batching, and then performs symmetrical connection neural network calculation, weighs the batching according to the calculation results, and respectively performs three times of detection, analysis, feedback and correction by machine vision and LIBS technology at the feeding port of various raw materials, after mixing, and after sintering and crushing, so as to dynamically adjust the batching and related process parameters in time, effectively avoid the fluctuation of sintered ore quality caused by raw material composition fluctuation, information lag, process parameter change and the like, improve the batching effect and sintered ore quality, ensure the stable and smooth production, significantly improve the production efficiency and reduce the production cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1A flow chart of a sintering batching method based on machine vision and LIBS technology is proposed for Example 1 of the present invention;

[0032] Figure 2 This is a schematic diagram of the symmetrically connected neural network structure proposed in Example 1 of the present invention;

[0033] Figure 3 A schematic diagram of a sintering batching system based on machine vision and LIBS technology is proposed for Example 2 of the present invention. DETAILED DESCRIPTION

[0034] In order to clearly illustrate the technical features of the present solution, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings. The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. In addition, the present invention may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the accompanying drawings are not necessarily drawn to scale. The present invention omits the description of known components and processing techniques and processes to avoid unnecessary limitations on the present invention.

[0035] Example 1

[0036] Embodiment 1 of the present invention proposes a sintering batching method based on machine vision and LIBS technology, which is used to solve the current technical problems that the raw materials are unstable and fluctuate greatly, resulting in large deviations in traditional batching calculations and untimely batching adjustments. Figure 1 A flow chart of a sintering batching method based on machine vision and LIBS technology is proposed for Example 1 of the present invention;

[0037] In step S100, the raw material data for sintering is collected to establish a sintering database, and the original data of the types, components, physicochemical parameters and prices of various raw materials for sintering are collected to establish a database, including various iron-containing raw materials, solvents, fuels and return ore.

[0038] In step S101, the theoretical batching table M is obtained by using the raw material data to perform theoretical batching calculations. 理论 , Theoretical ingredient list M 理论 Used to save the theoretical proportion data of each raw material ore.

[0039] Theoretical ingredient table M is obtained by using raw material data to perform theoretical calculation of ingredients. 理论 The process includes:

[0040] G = 60 × batch size (kg / m) × batching ratio (%) × belt speed (m / min)

[0041] Wherein, G is the feeding amount of any raw material per hour;

[0042] The consumption of any raw material per shift is: M = T × G;

[0043] Among them, M is the consumption of any raw material per shift; T is the loading time during the shift.

[0044] The theoretical ratio of each raw material can be calculated according to the sintering process regulations. For example:

[0045] 1. Burning residue = dry material ratio × (1-burning loss)

[0046] 2. TFe in the batch = TFe content of the raw material × the proportion of the raw material

[0047] 3. SiO2 in the batch = SiO2 content of the raw material × the proportion of the raw material

[0048] 4. Ca0 in the batch = Ca0 content of the raw material × the proportion of the raw material

[0049] 5. Mg0 in the batch = Mg0 content of the raw material × the proportion of the raw material

[0050] 6. Mn in the batch = Mn content of the raw material × the proportion of the raw material

[0051] 7. Chemical composition of sintered ore

[0052] Sintered ore TFe = sum of TFe brought in by various raw materials / total burn residue

[0053] Sintered ore SiO2 = the sum of SiO2 brought in by various raw materials / total burnt residue

[0054] Sintered ore CaO = the sum of CaO introduced by various raw materials / total burnt residue

[0055] Sintered ore Mg0 = the sum of MgO brought in by various raw materials / total burn residue

[0056] Sintered ore Mn = the sum of Mn brought in by various raw materials / total burn residue

[0057] Those skilled in the art can select the required calculation formula according to actual conditions.

[0058] In step S102, the empirical proportioning calculation is performed using the raw material data to obtain the empirical proportioning table M. 经验 , Experience Ingredients Table M 经验 Used to save the empirical proportion data of each material ore;

[0059] The verification algorithm is used, and the specific process is: assuming the batching ratio based on actual production experience, and calculating the chemical composition of the sintered ore based on the original data of various materials to see whether it meets the requirements of the specified indicators; the original data includes material moisture, burn loss and chemical composition.

[0060] Before calculating the ingredients, the following data must be known:

[0061] (1) Relevant physical properties and chemical composition of various raw materials; (2) Technical conditions of sintered ore; (3) Circulation volume of sintered ore; (4) Storage and supply of raw materials; (5) Capacity of batching equipment.

[0062] The steps are:

[0063] (1) Determine the proportion of raw materials: Determine the proportion of various raw materials according to the requirements of sintering technology, raw material supply plan and chemical composition.

[0064] (2) Calculate the amount of dry material:

[0065] Dry material quantity (%) = wet material ratio × (100-water content); Total dry material quantity Σdry material = sum of various dry material quantities;

[0066] (3) Calculate the remaining amount (burned amount):

[0067] Residual amount (%) = dry material amount × (100-burnt loss);

[0068] Residual coke powder (%) = coke powder to dry material ratio × (100-burning loss) = coke powder to dry material ratio × ash content;

[0069] Total residual quantityΣ residual quantity=the sum of residual quantities of various raw materials;

[0070] (4) Calculate the various components that go into the batching:

[0071] Iron content of various raw materials in the batch (%):

[0072] W(TFe) = iron content of a certain raw material × dry material ratio of a certain raw material;

[0073] Total iron content ΣW (TFe) = the sum of the iron content of each raw material;

[0074] SiO2 content of various raw materials in the batch (%):

[0075] W(SiO2) = mass fraction of SiO2 in a certain raw material × dry material ratio of a certain raw material;

[0076] Total SiO2 content ΣW(SiO2) = the sum of SiO2 content of each raw material;

[0077] CaO content of various raw materials in the batching (%):

[0078] W(CaO) = mass fraction of CaO in a certain raw material x dry material ratio of a certain raw material;

[0079] Total CaO content ΣW(CaO) = the sum of CaO content of each raw material;

[0080] (5) Calculate sintered ore R (basicity):

[0081] ∑W(CaO)R=∑W(SiO2);

[0082] (6) Calculate the chemical composition of sintered ore:

[0083] W(TFe) = the sum of TFe brought in by various raw materials / total residual amount;

[0084] W (SiO2 ore) = the sum of SiO2 brought in by various materials ÷ total residual amount;

[0085] W (CaO ore) = the sum of CaO brought in by various materials / total residual amount;

[0086] If the calculated sintered ore basicity and sintered ore composition meet the requirements, the predetermined batching ratio is the actual batching ratio.

[0087] In step S103, the theoretical batching table and the empirical batching table are input into a preset neural network to predict the optimal batching table; the optimal batching table is used to store the optimal proportion data of each material ore;

[0088] The neural network preset in this application adopts a symmetrically connected neural network. Figure 2 This is a schematic diagram of the symmetrically connected neural network structure proposed in Example 1 of the present invention; the symmetrically connected network is a bit like a recurrent network, but the connections between units are symmetrical (they have the same weights in both directions). Compared with recurrent networks, symmetrically connected networks are easier to analyze. Among them, X1 to Xn are input terminals; Y1 to Yn are output terminals.

[0089] In step S104, feed the materials according to the optimal proportion data of each material in the optimal batching table; obtain images of each material on the belt, and use machine vision to detect the shape information of the materials based on the images of each material; and use LIBS to detect the main components of the materials at the feeder discharge port;

[0090] The symmetrical neural network calculates the proportion of each ore and the weighing system starts weighing to the belt.

[0091] Measurement and judgment are carried out at the discharge port of the feeder for each raw material. The raw materials on the belt are captured by the lens and image processed. Various operations are performed to extract the characteristics of the target based on information such as pixel distribution, brightness and color. The judgment results are then output based on the preset judgment criteria and compared with the data information based on the initial calculation of the raw material. Feedback is promptly provided to the model for real-time dynamic adjustment and correction.

[0092] LIBS detection is carried out at the feeder discharge port of each raw material to timely detect and analyze the main components of the raw materials on the belt, and also compared with the data information of the raw materials as the basis for initial calculation, and fed back to the model for real-time dynamic adjustment and correction. Machine vision uses LIBS technology to emit lasers, and directly obtains the element type and content of the sample based on the spectrum combined with quantitative analysis algorithms.

[0093] In step S105, water is added for wetting, mud homogenization and balling according to the process requirements. A secondary mixing process is adopted, in which wetting and mixing are performed in the first mixing, and the material can also be preheated when the ore is heated and returned; then secondary mixing is performed: mixing and balling are continued to improve the air permeability of the sintering material layer.

[0094] After the secondary mixing, machine vision and LIBS technology are used to perform secondary detection, analysis, and judgment on the feeding belt, and compare and feedback with the target data in the model database, so as to timely correct the ingredient list and adjust the process parameters of the first and second mixing, such as mixing time, moisture content, etc.

[0095] After adjusting the mixing process parameters according to the results of the secondary test, it also includes material distribution, ignition sintering and crushing;

[0096] In step S106, a material laying process is performed, and the material laying process includes laying the base material and the mixed material on the sintering machine trolley.

[0097] According to the process requirements, the base material and the mixed material are spread on the sintering machine trolley. Before spreading the mixed material, a layer of small pieces of sintered ore is spread as the base material, the purpose of which is to protect the grate, reduce the dust removal load, extend the life of the fan rotor, and reduce or eliminate the grate sticking. After the base material is spread, the material is spread. When spreading the material, the particle size and chemical composition of the mixed material are required to be evenly distributed along the longitudinal and transverse directions of the trolley, and have a certain degree of looseness and a smooth surface.

[0098] In step S107, ignition and sintering are performed, and the sintering air volume, vacuum degree, material layer thickness, machine speed and sintering end point are precisely controlled during the ignition and sintering.

[0099] Sufficient ignition temperature, appropriate high temperature holding time, uniform ignition along the width of the trolley. Accurately control the sintering air volume, vacuum degree, material layer thickness, machine speed and sintering end point according to process requirements.

[0100] In step S108, a crushing process is performed, and the crushing process includes making the sintered block slide onto a single-roll crusher to be sheared and crushed.

[0101] When the trolley moves to the end of the sintering machine, the sintering process is completed and the red-hot sintered blocks slide onto the single-roll crusher to be sheared and crushed.

[0102] In step S109, after the sintering process is completed to the mixing, the machine vision and LIBS are used to perform a secondary inspection respectively, and the mixing process parameters are adjusted according to the results of the secondary inspection; after the sintering process is completed to the crushing, the machine vision and LIBS are used to perform a tertiary inspection respectively, and the sintering process parameters are adjusted according to the results of the tertiary inspection to complete the sintering process flow.

[0103] Machine vision and LIBS technology are used to conduct a third inspection, analysis and judgment on the crushed finished sintered ore. The inspection results are fed back to the model database in a timely manner for comparison, so that the ingredient list can be corrected in time and the relevant process parameters of mixing and sintering can be adjusted.

[0104] Embodiment 1 of the present invention proposes a sintering batching method based on machine vision and LIBS technology, which establishes a database by collecting original data information of various raw materials for sintering, constructs a batching model to perform theoretical computer empirical proportion calculation of the batching, and then performs symmetrical connection neural network calculation, weighs the batching according to the calculation results, and performs three times of machine vision and LIBS technology detection, analysis, feedback, and correction at the feeding port of various raw materials, after mixing, and after sintering and crushing, respectively, to dynamically adjust the batching and related process parameters in time, effectively avoiding fluctuations in the quality of sintered ore caused by fluctuations in raw material composition, information lag, and changes in process parameters, improving the batching effect and the quality of sintered ore, ensuring stable and smooth production, and significantly improving production efficiency and reducing production costs.

[0105] Example 2

[0106] Based on the sintering batching method based on machine vision and LIBS technology proposed in Example 1 of the present invention, Example 2 of the present invention further proposes a sintering batching system based on machine vision and LIBS technology. Figure 3 A schematic diagram of a sintering batching system based on machine vision and LIBS technology is proposed for Example 2 of the present invention, the system comprising: a pre-processing module, a primary detection module and a multiple detection module;

[0107] The preprocessing module is used to collect raw material data for sintering to establish a database; use the raw material data to perform theoretical batching calculations to obtain a theoretical batching table, which is used to store theoretical proportion data of each material ore; use the raw material data to perform empirical batching calculations to obtain an empirical batching table, which is used to store empirical proportion data of each material ore; input the theoretical batching table and the empirical batching table into a preset neural network to predict an optimal batching table; the optimal batching table is used to store optimal proportion data of each material ore;

[0108] The primary detection module is used to feed materials according to the optimal proportion data of each material in the optimal batching table; obtain the image of each material on the belt, and use machine vision to detect the shape information of the material based on the image of each material; and use LIBS to detect the main components of the material at the feeder discharge port;

[0109] The multiple detection module is used to perform secondary detection using machine vision and LIBS after the sintering process is completed to mix the materials, and to adjust the mixing process parameters according to the results of the secondary detection; after the sintering process is completed to crush the materials, the machine vision and LIBS are used to perform tertiary detection, and to adjust the sintering process parameters according to the results of the tertiary detection to complete the sintering process.

[0110] During the execution of the pre-processing module of the present invention, the raw material data for sintering includes: the type of raw material for sintering, the raw material composition and physicochemical parameters, and the material price.

[0111] The process of using raw material data to perform theoretical calculations to obtain a theoretical ingredient list includes:

[0112] G = 60 × batch size (kg / m) × batching ratio (%) × belt speed (m / min)

[0113] Wherein, G is the feeding amount of any raw material per hour;

[0114] The consumption of any raw material per shift is: M = T × G;

[0115] Among them, M is the consumption of any raw material per shift; T is the loading time during the shift.

[0116] The process of using raw material data to perform empirical proportion calculations to obtain an empirical ingredient list includes: assuming the proportion based on actual production experience, and calculating the chemical composition of the sintered ore based on the original data of various materials to see whether it meets the requirements of the specified indicators; the original data includes material moisture, burn loss and chemical composition.

[0117] During the execution of the primary detection module, the process of performing LIBS at the feeder discharge port to detect the main components of the material includes: using LIBS technology to emit laser, and directly obtaining the element type and content of the sample based on the spectrum combined with the quantitative analysis algorithm.

[0118] During the execution of multiple detection modules, the mixing process includes adding water to moisten, mud homogenization and balling.

[0119] After adjusting the mixing process parameters according to the results of the secondary test, it also includes distribution, ignition sintering and crushing; the distribution process includes laying the base material and the mixed material on the sintering machine trolley; during the ignition and sintering, the sintering air volume, vacuum degree, material layer thickness, machine speed and sintering end point are accurately controlled; the crushing process includes making the sintered block slide onto the single-roll crusher for shearing and crushing.

[0120] The process of adjusting the sintering process parameters according to the results of the three tests includes: comparing the results of the three tests with the sintering database, correcting the ingredient list, adjusting the mixture and adjusting the sintering process parameters.

[0121] Embodiment 2 of the present invention proposes a sintering batching system based on machine vision and LIBS technology, which establishes a database by collecting original data information of various raw materials for sintering, constructs a batching model to perform theoretical computer empirical proportion calculation of the batching, and then performs symmetrical connection neural network calculation, weighs the batching according to the calculation results, and performs three times of machine vision and LIBS technology detection, analysis, feedback, and correction at the feeding port of various raw materials, after mixing, and after sintering and crushing, respectively, to dynamically adjust the batching and related process parameters in time, effectively avoiding fluctuations in the quality of sintered ore caused by fluctuations in raw material composition, information lag, and changes in process parameters, improving the batching effect and the quality of sintered ore, ensuring stable and smooth production, and significantly improving production efficiency and reducing production costs.

[0122] The description of the relevant parts of the sintering batching system based on machine vision and LIBS technology provided in Example 2 of the present application can be found in the detailed description of the corresponding parts of the sintering batching method based on machine vision and LIBS technology provided in Example 1 of the present application, and will not be repeated here.

[0123] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment that includes a series of elements are inherent to the elements. In the absence of more restrictions, the elements limited by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or equipment that includes the elements. In addition, the above-mentioned technical solution provided in the embodiment of the present application is consistent with the corresponding technical solution in the prior art in principle, and the part is not described in detail, so as not to repeat too much.

[0124] Although the above describes the specific implementation of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. For those skilled in the art, other different forms of modifications or deformations can be made on the basis of the above description. It is not necessary and impossible to list all the implementation methods here. On the basis of the technical solution of the present invention, various modifications or deformations that can be made by those skilled in the art without creative work are still within the scope of protection of the present invention.

Claims

1. A sintering batching method based on machine vision and LIBS technology, characterized in that: The following steps are involved: Collecting raw material data for sintering to establish a sintering database; using the raw material data to perform theoretical batching calculations to obtain a theoretical batching table, the theoretical batching table is used to store theoretical proportion data of each material ore; using the raw material data to perform empirical batching calculations to obtain an empirical batching table, the empirical batching table is used to store empirical proportion data of each material ore; inputting the theoretical batching table and the empirical batching table into a preset neural network to predict an optimal batching table; the optimal batching table is used to store optimal proportion data of each material ore; Feed materials according to the optimal proportion data of each material in the optimal batching table; obtain images of each material on the belt, and use machine vision to detect the shape information of the material based on the images of each material; and use LIBS at the feeder discharge port to detect the main components of the material at one time; After the sintering process is completed and the mixing is completed, machine vision and LIBS are used for secondary inspection, and the mixing process parameters are adjusted according to the results of the secondary inspection; After the sintering process is completed and the crushing is completed, machine vision and LIBS are used to perform three inspections respectively, and the sintering process parameters are adjusted according to the results of the three inspections to complete the sintering process.

2. The sintering batching method based on machine vision and LIBS technology according to claim 1, characterized in that: The raw material data for sintering include: the type of raw materials for sintering, raw material composition and physicochemical parameters, and material prices.

3. The sintering batching method based on machine vision and LIBS technology according to claim 1, characterized in that: The process of using the raw material data to perform theoretical calculation of ingredients to obtain a theoretical ingredient list includes: G = 60 × batch size (kg / m) × batching ratio (%) × belt speed (m / min) Wherein, G is the feeding amount of any raw material per hour; The consumption of any raw material per shift is: M = T × G; Among them, M is the consumption of any raw material per shift; T is the loading time during the shift.

4. The sintering batching method based on machine vision and LIBS technology according to claim 1, characterized in that: The process of using the raw material data to perform empirical proportioning calculations to obtain an empirical proportioning table includes: assuming proportioning ratios based on actual production experience, and calculating the chemical composition of the sintered ore based on the original data of various materials to see whether it meets the requirements of the specified indicators; the original data includes material moisture, burn loss and chemical composition.

5. The sintering batching method based on machine vision and LIBS technology according to claim 1, characterized in that: The preset neural network adopts a symmetrically connected neural network.

6. The sintering batching method based on machine vision and LIBS technology according to claim 1, characterized in that: The process of using LIBS to detect the main components of the material at the feeder discharge port includes: using LIBS technology to emit laser, and directly obtaining the element type and content of the sample based on the spectrum combined with the quantitative analysis algorithm.

7. The sintering batching method based on machine vision and LIBS technology according to claim 1, characterized in that: The mixing process includes adding water to moisten, mudding and balling.

8. The sintering batching method based on machine vision and LIBS technology according to claim 1, characterized in that: After adjusting the mixing process parameters according to the results of the secondary test, it also includes laying, ignition sintering and crushing; The process of laying the materials includes laying the base material and the mixed material on the sintering machine trolley; During the ignition sintering, the sintering air volume, vacuum degree, material layer thickness, machine speed and sintering end point are precisely controlled; The crushing process includes making the sintered block slide onto a single roll crusher to be sheared and crushed.

9. The sintering batching method based on machine vision and LIBS technology according to claim 1, characterized in that: The process of adjusting the sintering process parameters according to the results of the three tests includes: comparing the results of the three tests with the sintering database, correcting the ingredient list, adjusting the mixture and adjusting the sintering process parameters.

10. A sintering batching system based on machine vision and LIBS technology, characterized in that: include: A preprocessing module, a one-time detection module, and a multiple-time detection module; The pre-processing module is used to collect raw material data for sintering to establish a database; use the raw material data to perform theoretical batching calculations to obtain a theoretical batching table, which is used to store theoretical proportion data of each raw material ore; use the raw material data to perform empirical batching calculations to obtain an empirical batching table, which is used to store empirical proportion data of each raw material ore; input the theoretical batching table and the empirical batching table into a preset neural network to predict an optimal batching table; the optimal batching table is used to store optimal proportion data of each raw material ore; The primary detection module is used to feed materials according to the optimal proportion data of each material in the optimal batching table; obtain images of each material on the belt, and use machine vision to detect the shape information of the material based on the images of each material; and perform LIBS at the feeder discharge port to detect the main components of the material; The multiple detection module is used to perform secondary detection using machine vision and LIBS respectively after the sintering process is completed to complete the mixing, and adjust the mixing process parameters according to the results of the secondary detection; After the sintering process is completed and the crushing is completed, machine vision and LIBS are used to perform three inspections respectively, and the sintering process parameters are adjusted according to the results of the three inspections to complete the sintering process.

Images