Sintering ignition intensity intelligent identification method and device based on multi-source data fusion

Abstract

The invention discloses a sintering ignition intensity intelligent identification method and device based on multi-source data fusion, and the method comprises the steps: obtaining waste gas bellows temperature data in a sintering process, carrying out the fitting of the data, and obtaining a grading label of the ignition intensity according to a fitting result; thermal state parameters in the sintering process are obtained, a classification network is built, the relation between the thermal state parameters and the grading labels of the ignition intensity is built, and an overall ignition intensity grading result is obtained; acquiring a surface image of the sintered material layer, detecting two defects of yellow mud and cracks by adopting a method based on machine vision, and counting the existence condition of the defects in a striping manner; and integrating the integral ignition intensity grading result and the striping defect detection result to obtain a refined ignition intensity grading result, thereby realizing sintering ignition intensity intelligent grading based on multi-source data fusion. The method has good effect and practical value on sintering ignition intensity grading through the design of the fusion strategy.

Description

Technical Field

[0001] This application relates to the field of detection technology, and in particular to a method and device for intelligent identification of sintering ignition intensity based on multi-source data fusion. Background Technology

[0002] The iron and steel industry is a pillar industry of the national economy, making outstanding contributions to the growth of my country's GDP. Blast furnace ironmaking, as the main method of modern ironmaking, is a crucial link in iron and steel production. Sintered ore is one of the important raw materials for blast furnace ironmaking in the iron and steel smelting process. Iron ore sintering is a pretreatment process that provides sintered ore to the blast furnace and is an indispensable key step in ironmaking production. Sintering ignition uses the high-temperature flue gas generated by the combustion of blast furnace gas or coke oven gas in the ignition furnace to ignite the fuel in the sintering mixture, forming an initial combustion zone. Simultaneously, under the action of draft gas, the combustion zone burns downwards. Ignition is the beginning of the combustion process in sintering production. Ignition intensity refers to the ignition heat or the amount of gas burned per unit area of ​​the mixture. Suitable ignition intensity is mainly related to factors such as the properties of the mixture, the draft gas volume, and the thermal efficiency of the igniter. Insufficient ignition intensity leads to insufficient strength of the sinter, excessive return ore, and a yellowing surface in the material layer, resulting in a defect known as "yellow mud." Excessive ignition intensity, on the other hand, reduces the permeability of the material surface, decreases productivity, increases energy consumption, and causes the material surface to become too dry, resulting in cracks. Therefore, ignition conditions directly affect sinter quality, yield, and ignition energy consumption. Identifying and appropriately classifying ignition intensity will provide strong guidance for operators' subsequent control strategies.

[0003] Currently, there is a lack of a comprehensive and systematic grading strategy for sintering ignition intensity. Control and adjustment still rely on the experience of on-duty workers, which is inefficient and consumes a lot of manpower. Summary of the Invention

[0004] The purpose of this invention is to provide a method and device for intelligent identification of sintering ignition intensity based on multi-source data fusion, so as to solve the technical problem that the existing control and adjustment relies on the experience of on-duty workers, resulting in low efficiency and high manpower consumption.

[0005] According to a first aspect of the embodiments of this application, a method for intelligent identification of sintering ignition intensity based on multi-source data fusion is provided, comprising: The temperature data of the exhaust gas box during the sintering process is obtained, and the exhaust gas box temperature data is fitted. The ignition intensity classification label is obtained based on the fitting result. The thermal state parameters during the sintering process within the same time period are obtained, a classification network is built, and the relationship between the thermal state parameters and the ignition intensity classification labels is constructed to obtain the overall ignition intensity classification results. Images of the sintered material layer surface were acquired within the same time period. Machine vision-based methods were used to detect two types of defects: yellow mud and cracks. The presence of the two types of defects was statistically analyzed by strip. By integrating the overall ignition intensity grading results with the defect presence data from the strip statistics, a refined ignition intensity grading result is obtained.

[0006] According to a second aspect of the embodiments of this application, a smart identification device for sintering ignition intensity based on multi-source data fusion is provided, comprising: The acquisition and processing module is used to acquire the exhaust gas box temperature data during the sintering process, fit the exhaust gas box temperature data, and obtain the ignition intensity classification label based on the fitting result. The grading module is used to acquire the thermal state parameters during the sintering process within the same time period, build a classification network, construct the relationship between the thermal state parameters and the grading labels of the ignition intensity, and obtain the overall ignition intensity grading results. The image processing module is used to acquire images of the sintered material layer surface within the same time period, and to detect two types of defects, namely yellow mud and cracks, using machine vision-based methods, and to statistically analyze the presence of the two types of defects by strip. The fusion module is used to integrate the overall ignition intensity grading results with the defect presence status of the strip statistics to obtain a refined ignition intensity grading result.

[0007] According to a third aspect of the embodiments of this application, an electronic device is provided, comprising: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors perform the method as described in the first aspect.

[0008] According to a fourth aspect of the embodiments of this application, a computer-readable storage medium is provided that stores computer instructions thereon, which, when executed by a processor, implement the steps of the method as described in the first aspect.

[0009] The technical solutions provided by the embodiments of this application may include the following beneficial effects: As can be seen from the above embodiments, this application designs an intelligent identification method for sintering ignition intensity based on multi-source data fusion. It acquires exhaust gas box temperature data during the sintering process, fits the exhaust gas box temperature data, and obtains ignition intensity grading labels based on the fitting results. It also acquires thermal state parameters during the sintering process within the same time period, builds a classification network, and constructs the relationship between the thermal state parameters and the ignition intensity grading labels, enabling prediction and grading of the overall ignition intensity. Furthermore, it statistically analyzes the presence of defects in the strips, selecting two defects reflecting ignition intensity: yellow mud and cracks, which can be used to evaluate the ignition intensity of the strips. Finally, it designs an integration rule for the overall ignition intensity grading results and the strip defect detection results, enabling the acquisition of the overall ignition situation through time-series data and the observation of local ignition situations in the strips through image information, thereby achieving refined ignition intensity grading of the strips and facilitating subsequent refined control.

[0010] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0011] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0012] Figure 1 This is a flowchart illustrating an intelligent identification method for sintering ignition intensity based on multi-source data fusion, according to an exemplary embodiment.

[0013] Figure 2 This is a visualization of the ignition intensity grading of strips according to an exemplary embodiment.

[0014] Figure 3 This is a block diagram illustrating a sintering ignition intensity intelligent identification device based on multi-source data fusion, according to an exemplary embodiment.

[0015] Figure 4 This is a structural diagram of an electronic device according to an exemplary embodiment. Detailed Implementation

[0016] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0017] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0018] Figure 1 This is a flowchart illustrating an intelligent identification method for sintering ignition intensity based on multi-source data fusion, according to an exemplary embodiment. Figure 1 As shown, the method may include the following steps: S1: Obtain the exhaust gas box temperature data during the sintering process, and fit the exhaust gas box temperature data to obtain the ignition intensity classification label based on the fitting result. S2: Obtain the thermal state parameters during the sintering process within the same time period, build a classification network, construct the relationship between the thermal state parameters and the grading labels of the ignition intensity, and obtain the overall ignition intensity grading results; S3: Acquire images of the sintered material layer surface within the same time period, use machine vision-based methods to detect two types of defects: yellow mud and cracks, and statistically analyze the presence of the two types of defects by strip. S4: Integrate the overall ignition intensity grading results with the defect status of the strip statistics to obtain a refined ignition intensity grading result.

[0019] As can be seen from the above embodiments, this application designs an intelligent identification method for sintering ignition intensity based on multi-source data fusion. The relationship between thermal state parameters and ignition intensity grading is constructed. Variables affecting ignition intensity, such as fuel ratio, main flue negative pressure, return ore ratio, material thickness, blending ore ratio, sintering machine speed, light-burned ore ratio, main exhaust air volume, quicklime ratio, and ignition temperature, are selected to predict and grade the overall ignition intensity. The presence of defects in each strip is statistically analyzed, selecting two defects reflecting ignition intensity: yellow mud and cracks, which can be used to evaluate the ignition intensity of each strip. An integration rule is designed to combine the overall ignition intensity grading results with the strip defect detection results. This allows for obtaining the overall ignition situation through time-series data and the local ignition situation of each strip through image information observation, thereby achieving refined ignition intensity grading of each strip, which is beneficial for subsequent refined control.

[0020] In the specific implementation of S1, the temperature data of the exhaust gas box during the sintering process is acquired, and the exhaust gas box temperature data is fitted. Based on the fitting result, the ignition intensity grading label is obtained, including: S11: Obtain the temperature data of the exhaust gas box during the sintering process; Specifically, the exhaust gas blower temperature data of a sintering plant of a Chinese steel company was used. The blowers were arranged under the sintering machine, with a total of 24 blowers. The number of blowers was determined according to the requirements. S12: Select the temperatures of several of the aforementioned exhaust gas boxes and fit them to obtain the fitting curve results; Specifically, the exhaust gas temperature distribution curve of the wind box can be obtained by polynomial fitting. In this embodiment, the fitting curve is obtained by quadratic fitting of the last five wind boxes. S13: Take the maximum value point of the fitted curve to obtain the sintering endpoint position, and substitute the sintering endpoint position into the grading rule of ignition intensity to obtain the grading label of ignition intensity, wherein the grading rule divides the ignition intensity into several levels according to the sintering endpoint position information.

[0021] Specifically, since the sintering endpoint represents the point where the sintered material layer is fully burned, the fuel at the sintering endpoint must be completely burned. Therefore, the temperature of the material layer at this point should be the highest point of the solid phase temperature of the entire sintered material. Thus, the maximum value of the fitted curve is taken as the sintering endpoint. When the sintering endpoint is located at the second-to-last wind box, i.e., wind box number 23, the area of ​​the sintering machine can be utilized to the maximum extent.

[0022] The hierarchical rules in this embodiment are as follows: .

[0023] Specifically, ignition intensity affects the sintering process conditions. Sintering process conditions refer to the combustion state of the sintering process over a period of time under certain operating settings and power support. Based on the degree of completion of the mixture's combustion, these conditions can be divided into three categories: overburning, normal, and underburning. The sintering endpoint is an important indicator of the operating conditions. If the sintering endpoint is reached prematurely, the effective area of ​​the sintering machine is not fully utilized, resulting in low utilization; if the sintering endpoint is delayed, the mixture is not completely burned during unloading, leading to increased return ore and low output. Therefore, the sintering endpoint position can represent the ignition intensity. An early sintering endpoint (defined as before 22.5 mm in this embodiment) indicates that the sintered material is prematurely burned, in an overburning condition, and the ignition intensity is too strong; a delayed sintering endpoint (defined as after 23.5 mm in this embodiment) indicates that the sintered material is not completely burned, in an underburning condition, and the ignition intensity is too weak; a sintering endpoint within the normal range (defined as between 22.5 and 23.5 mm in this embodiment) indicates that the sintered material is normally burned, in a normal operating condition, and the ignition intensity is normal.

[0024] In the specific implementation of S2, thermal state parameters during the sintering process within the same time period are obtained, a classification network is built, and the relationship between the thermal state parameters and the ignition intensity classification labels is constructed to obtain the overall ignition intensity classification results, including: S21: Obtain the thermal state parameters of the sintering process within the same time period, including: fuel ratio, main flue negative pressure, return ore ratio, material thickness, mixed ore ratio, sintering machine speed, light calcined ore ratio, main exhaust volume, quicklime ratio and ignition temperature; Specifically, parameters such as fuel ratio, main flue negative pressure, return ore ratio, material thickness, blending ore ratio, sintering machine speed, light calcined ore ratio, main exhaust air volume, quicklime ratio, and ignition temperature can reflect the thermal state of the sintering process. These parameters will affect the ignition intensity and can therefore be used to predict the ignition intensity. S22: Construct a classification network, with the thermal state parameters as input and the ignition intensity classification label as output; Specifically, a multilayer perceptron network is constructed, with the thermal state parameters described in S21 as the input and the ignition intensity classification label described in S13 as the output, to establish the relationship between the thermal state parameters and the ignition intensity. S23: Train the classification network, input the thermal state parameters into the trained network, and obtain the predicted overall ignition intensity classification result.

[0025] Specifically, the classification network, after training, has the ability to model thermal state parameters and ignition intensity. By inputting the thermal state parameters into the trained network, the overall ignition intensity classification result can be predicted.

[0026] In the specific implementation of S3, images of the sintered material layer surface are acquired within the same time period. Machine vision-based methods are used to detect two types of defects: yellow mud and cracks. The presence of these two defects is statistically analyzed by strip, including: S31: Acquire images of the sintered material layer surface within the same time period, and use a color-based detection method to detect the location of yellow mud and a semantic segmentation-based detection method to detect the distribution of cracks. Specifically, yellow mud indicates a situation of insufficient ignition intensity, while cracks indicate a situation of excessive ignition intensity. Yellow mud has a distinct color characteristic, allowing for the location of yellow mud in an image to be detected using color-based methods; cracks are characterized by their elongated and narrow shape, allowing for the detection of crack distribution using semantic segmentation-based methods. S32: Divide the surface of the sintered material layer into several strips from left to right. Combine the results of yellow mud and crack detection to count the presence of yellow mud and the area of ​​cracks in each strip, and obtain the strip defect detection results.

[0027] Specifically, there are six ignition nozzles distributed from left to right on the sintering machine, which divide the surface of the sintering material layer into six strips from left to right. Combined with the detection results of yellow mud and cracks, the presence or absence of yellow mud on each strip and the number of pixels of cracks, i.e. the area of ​​cracks, can be obtained.

[0028] In the specific implementation of S4, the overall ignition intensity grading result is integrated with the defect presence status of the strip statistics to obtain a refined ignition intensity grading result, including: The overall ignition intensity grading results are integrated with the strip defect detection results. The integration rule is as follows: the overall ignition intensity grading results are combined with the defect presence information of the strip statistics one by one. The overall ignition intensity grading results are used as the overall ignition intensity information, and the defect presence information of the strip statistics are used as the local ignition intensity information, so as to obtain a refined ignition intensity grading result.

[0029] The integration rules in this embodiment are as follows: A crack is considered significant if the number of pixels in each crack strip is greater than 550; otherwise, it is considered insignificant. The area of ​​the crack is used as an indicator of its severity, and further as an indicator of excessive local ignition intensity.

[0030] By incorporating the overall ignition intensity grading results and the strip defect detection results into the integration rules, a refined ignition intensity grading result can be obtained, thereby realizing intelligent grading of sintering ignition intensity based on multi-source data fusion. Figure 2 This is a visualization of the ignition intensity grading by strip, according to an exemplary embodiment. The overall ignition intensity predicted by the thermal state parameters at this moment is "excessive". The results of the strip defect detection are as follows: the first strip from left to right is "no yellow mud & inconspicuous cracks", and the remaining strips are "no yellow mud & obvious cracks". Therefore, the refined grading result of the first strip is "overall excessive, locally normal", represented by light red; the refined grading result of the remaining strips is "overall excessive, locally excessive", represented by dark red.

[0031] Corresponding to the aforementioned embodiments of the intelligent identification method for sintering ignition intensity based on multi-source data fusion, the present invention also provides embodiments of an intelligent identification device for sintering ignition intensity based on multi-source data fusion.

[0032] Figure 3 This is a block diagram illustrating an intelligent identification device for sintering ignition intensity based on multi-source data fusion, according to an exemplary embodiment. (Refer to...) Figure 3 The device includes: The acquisition and processing module 1 is used to acquire the exhaust gas box temperature data during the sintering process, fit the exhaust gas box temperature data, and obtain the ignition intensity classification label based on the fitting result. The grading module 2 is used to acquire the thermal state parameters during the sintering process within the same time period, build a classification network, construct the relationship between the thermal state parameters and the grading labels of the ignition intensity, and obtain the overall ignition intensity grading result. Image processing module 3 is used to acquire images of the sintered material layer surface within the same time period, and to detect two types of defects, namely yellow mud and cracks, using machine vision-based methods, and to statistically analyze the presence of the two types of defects by strip. The fusion module 4 is used to integrate the overall ignition intensity grading results with the defect existence status of the strip statistics to obtain a refined ignition intensity grading result.

[0033] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0034] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0035] Accordingly, this application also provides an electronic device, including: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the above-described intelligent identification method for sintering ignition intensity based on multi-source data fusion. Figure 4 The diagram shown is a hardware structure diagram of any device with data processing capabilities, including a sintering ignition intensity intelligent identification device based on multi-source data fusion provided in an embodiment of the present invention. (Except for...) Figure 4 In addition to the processor and memory shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0036] Accordingly, this application also provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the aforementioned intelligent identification method for sintering ignition intensity based on multi-source data fusion. The computer-readable storage medium can be an internal storage unit of any data-processing device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device for a wind turbine, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., mounted on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data-processing device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data-processing device, and can also be used to temporarily store data that has been output or will be output.

[0037] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0038] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for intelligent identification of sintering ignition intensity based on multi-source data fusion, characterized in that, include: The temperature data of the exhaust gas box during the sintering process is obtained, and the exhaust gas box temperature data is fitted. The ignition intensity classification label is obtained based on the fitting result. The thermal state parameters during the sintering process within the same time period are obtained, a classification network is built, and the relationship between the thermal state parameters and the ignition intensity classification labels is constructed to obtain the overall ignition intensity classification results. Images of the sintered material layer surface were acquired within the same time period. Machine vision-based methods were used to detect two types of defects: yellow mud and cracks. The presence of the two types of defects was statistically analyzed by strip. By integrating the overall ignition intensity grading results with the defect presence data from the strip statistics, a refined ignition intensity grading result is obtained.

2. The method according to claim 1, characterized in that, Acquire exhaust gas box temperature data during the sintering process, fit the exhaust gas box temperature data, and obtain ignition intensity grading labels based on the fitting results, including: Acquire the temperature data of the exhaust gas fan box during the sintering process; The temperatures of several of the aforementioned exhaust gas boxes were selected for fitting, and the fitting curve results were obtained. The maximum value of the fitted curve is taken to obtain the sintering endpoint position. The sintering endpoint position is then substituted into the ignition intensity grading rule to obtain the ignition intensity grading label. The grading rule is to divide the ignition intensity into several levels based on the sintering endpoint position information.

3. The method according to claim 1, characterized in that, The thermal state parameters during the sintering process within the same time period are obtained, a classification network is built, and the relationship between the thermal state parameters and the ignition intensity classification labels is constructed to obtain the overall ignition intensity classification results, including: The thermal state parameters of the sintering process within the same time period are obtained, including: fuel ratio, main flue negative pressure, return ore ratio, material thickness, mixed ore ratio, sintering machine speed, light calcined ore ratio, main exhaust air volume, quicklime ratio and ignition temperature. Construct a classification network, with the thermal state parameters as input and the ignition intensity classification label as output; The classification network is trained, and the thermal state parameters are input into the trained network to obtain the predicted overall ignition intensity classification result.

4. The method according to claim 1, characterized in that, Images of the sintered material layer surface within the same time period were acquired. Machine vision-based methods were used to detect two types of defects: yellow mud and cracks. The presence of these two defects was statistically analyzed by strip, including: Images of the sintered material layer surface were acquired within the same time period. Color-based detection method was used to detect the location of yellow mud, and semantic segmentation-based detection method was used to detect the distribution of cracks. The surface of the sintered material layer is divided into several strips from left to right. Based on the detection results of yellow mud and cracks, the presence of yellow mud and the area of ​​cracks in each strip are statistically analyzed to obtain the strip defect detection results.

5. The method according to claim 1, characterized in that, By integrating the overall ignition intensity grading results with the defect presence data from the strip statistics, a refined ignition intensity grading result is obtained, including: The overall ignition intensity grading results are integrated with the strip defect detection results. The integration rule is as follows: the overall ignition intensity grading results are combined with the defect presence information of the strip statistics one by one. The overall ignition intensity grading results are used as overall ignition intensity information, and the defect presence information of the strip statistics are used as local ignition intensity information to form a refined ignition intensity grading result.

6. A smart identification device for sintering ignition intensity based on multi-source data fusion, characterized in that, include: The acquisition and processing module is used to acquire the exhaust gas box temperature data during the sintering process, fit the exhaust gas box temperature data, and obtain the ignition intensity classification label based on the fitting result. The grading module is used to acquire the thermal state parameters during the sintering process within the same time period, build a classification network, construct the relationship between the thermal state parameters and the grading labels of the ignition intensity, and obtain the overall ignition intensity grading results. The image processing module is used to acquire images of the sintered material layer surface within the same time period, and to detect two types of defects, namely yellow mud and cracks, using machine vision-based methods, and to statistically analyze the presence of the two types of defects by strip. The fusion module is used to integrate the overall ignition intensity grading results with the defect presence status of the strip statistics to obtain a refined ignition intensity grading result.

7. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-5.

8. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by the processor, this instruction implements the steps of the method as described in any one of claims 1-5.

Images