Intelligent monitoring method and system for industrial silicon smelting stoking cover fire

Abstract

The invention provides an intelligent monitoring method and system for industrial silicon smelting stoking cover fire, and relates to the technical field of intelligent monitoring, and the method comprises the steps: carrying out the overall scanning of an industrial silicon smelting furnace and hearth materials through a laser scanning device, obtaining a point cloud data set, and constructing a smelting furnace three-dimensional model; performing multi-dimensional data acquisition to obtain a multi-dimensional data acquisition result; rendering to the smelting furnace three-dimensional model to obtain a smelting visual model; reading the monitoring index set, and monitoring to obtain a monitoring parameter set of the monitoring item set; performing smelting anomaly analysis, and extracting a first early warning level of the first monitoring anomaly item; and activating the industrial robot, and performing exception handling on the first monitoring exception item. The technical problems that in the prior art, industrial silicon smelting depends on manual driving of a forklift for fire covering, material pushing, auxiliary material adding, furnace ramming and other operations, and the safety risk of high-temperature radiation and scalding caused by high-temperature furnace material splashing exists are solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to an intelligent monitoring method and system for industrial silicon smelting ramming furnace cover fire. Background Art

[0002] During the industrial silicon smelting process, it is necessary to timely carry out operations such as covering the fire, pushing the materials, adding auxiliary materials, and tamping the furnace according to the conditions of the furnace materials to prevent the occurrence of material piling, empty burning, and agglomeration. At present, the industrial silicon industry relies entirely on manual forklifts to carry out operations such as covering the fire, pushing the materials, adding auxiliary materials, and tamping the furnace. The protection level of the operating area is not high, the space is small, and the operation must withstand nearly 90% of the material at any time.

[0003] Baking at 0℃ high temperature involves high temperature radiation and the safety risk of high temperature charge splashing and scalding. The entire process is completely dependent on the experience of employees, and accidents such as misoperation and damage to the bottom ring and electrode may occur, causing direct and indirect economic losses. Summary of the Invention

[0004] This application provides an intelligent monitoring method and system for industrial silicon smelting furnace cover fire, aiming to solve the technical problem that in the existing technology, industrial silicon smelting relies on manually driven forklifts to perform operations such as covering fire, pushing materials, adding auxiliary materials, and ramming the furnace, which poses safety risks of high-temperature radiation and burns caused by splashing of high-temperature furnace materials.

[0005] The first aspect disclosed in the present application provides an intelligent monitoring method for industrial silicon smelting furnace cover fire, the method comprising: performing an overall scan of the industrial silicon smelting furnace and furnace materials through a laser scanning device to obtain a point cloud data set, and constructing a three-dimensional model of the smelting furnace based on the point cloud data set; performing multi-dimensional data collection on the smelting furnace and furnace materials through multiple types of detection equipment to obtain multi-dimensional data collection results; rendering the multi-dimensional data collection results to the smelting furnace three-dimensional model to obtain a smelting visualization model; reading a monitoring indicator set, monitoring the smelting visualization model based on the monitoring indicator set, and obtaining a monitoring parameter set of a monitoring item set; performing a smelting anomaly analysis on the monitoring parameter set based on a pre-built smelting anomaly analysis model, and extracting a first warning level of a first monitoring anomaly item based on the smelting anomaly analysis result; activating an industrial robot based on the first monitoring anomaly item and the first warning level to perform anomaly processing on the first monitoring anomaly item.

[0006] The second aspect disclosed in the present application provides an intelligent monitoring system for industrial silicon smelting furnace cover fires, the system is used for the above-mentioned intelligent monitoring method for industrial silicon smelting furnace cover fires, the system includes: a three-dimensional model construction module for performing an overall scan of the industrial silicon smelting furnace and the furnace materials through a laser scanning device, obtaining a point cloud data set, and constructing a three-dimensional model of the smelting furnace based on the point cloud data set; a multi-dimensional data acquisition module for performing multi-dimensional data acquisition on the smelting furnace and the furnace materials through multiple types of detection equipment to obtain multi-dimensional data acquisition results; a visualization model construction module for rendering the multi-dimensional data acquisition results to the The three-dimensional model of the smelting furnace is used to obtain a smelting visualization model; a monitoring parameter acquisition module is used to read a monitoring indicator set, monitor the smelting visualization model based on the monitoring indicator set, and obtain a monitoring parameter set of a monitoring item set; a smelting anomaly analysis module is used to perform smelting anomaly analysis on the monitoring parameter set based on a pre-built smelting anomaly analysis model, and extract a first warning level of a first monitoring anomaly item according to the smelting anomaly analysis result; an abnormality handling module is used to activate the industrial robot based on the first monitoring anomaly item and the first warning level, and perform abnormality handling on the first monitoring anomaly item.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The 3D model constructed using laser scanning equipment and point cloud data sets provides a detailed visual representation of the smelting furnace and furnace materials. Multidimensional data from different detection equipment is rendered onto the 3D model, forming a comprehensive smelting visualization model. This allows various types of data to be analyzed and monitored on a unified platform, providing an intuitive operating interface. By reading and monitoring based on a set of monitoring indicators, comprehensive real-time tracking of key parameters in the smelting process is possible. Based on a pre-built smelting anomaly analysis model, in-depth analysis of the monitoring parameter set is performed, anomalies can be accurately identified, and warning levels can be extracted based on the analysis results, improving the accuracy and reliability of anomaly detection. Based on the anomaly detection results, industrial robots are activated for anomaly handling. By introducing industrial robots to replace manual labor in covering the fire, tamping the furnace, pushing materials, and adding materials, employees are prevented from directly contacting high-temperature furnace materials and high-temperature environments, thereby significantly reducing the safety risks of high-temperature radiation and burns, and ensuring the safety of operators.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of an intelligent monitoring method for industrial silicon smelting ramming furnace cover fire provided in an embodiment of the present application;

[0011] Figure 2 Schematic diagram of the structure of the intelligent monitoring system for industrial silicon smelting furnace cover fire provided in an embodiment of the present application.

[0012] Explanation of the accompanying drawings: three-dimensional model construction module 10, multi-dimensional data acquisition module 20, visualization model construction module 30, monitoring parameter acquisition module 40, smelting anomaly analysis module 50, anomaly processing module 60. DETAILED DESCRIPTION

[0013] The embodiments of the present application provide an intelligent monitoring method for industrial silicon smelting furnace cover fire, thereby solving the technical problem in the prior art that industrial silicon smelting relies on manually driven forklifts to perform operations such as covering fire, pushing materials, adding auxiliary materials, and ramming the furnace, which poses safety risks of high-temperature radiation and burns caused by splashing of high-temperature furnace materials.

[0014] After introducing the basic principles of this application, various non-limiting embodiments of this application will be specifically described below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to explain this application and are not used to limit this application.

[0015] like Figure 1 As shown, the embodiment of the present application provides an intelligent monitoring method for industrial silicon smelting furnace cover fire, the method comprising:

[0016] An industrial silicon smelting furnace and furnace materials are scanned as a whole using laser scanning equipment to obtain a point cloud data set, and a three-dimensional model of the smelting furnace is constructed based on the point cloud data set.

[0017] Select laser scanning equipment suitable for the smelting furnace environment, such as a lidar scanner, and use the laser scanning equipment to perform a comprehensive scan of the smelting furnace and the materials inside. Scanning is usually performed from multiple angles to ensure that all details of the furnace body and materials are captured. During the scanning process, the laser scanning equipment illuminates the target object with a laser beam and calculates the distance based on the reflected laser signal to generate a point cloud data set.

[0018] The point cloud data set is subjected to outlier removal, point cloud alignment and splicing. The processed point cloud data is converted into a triangular mesh model using a triangular meshing algorithm, thereby forming a three-dimensional surface model of the smelting furnace and materials. The actual surface texture information, such as material and color, is mapped onto the three-dimensional surface model to make the model more detailed, thus generating a three-dimensional model of the smelting furnace.

[0019] Furthermore, the method of constructing a three-dimensional model of a smelting furnace based on the point cloud data set includes:

[0020] Based on the principle of random sampling consistency, outliers are removed from the point cloud data set to obtain a smooth point cloud data set; point cloud alignment and splicing are performed on the smooth point cloud data set to obtain a constructed point cloud data set; triangulated meshing is performed on the constructed point cloud data set to generate a three-dimensional surface model; surface texture mapping of the silicon smelting furnace and the furnace material is performed based on the three-dimensional surface model to obtain a three-dimensional model of the smelting furnace.

[0021] Random sampling consistency is an iterative method used to estimate mathematical models in a dataset containing a large number of outliers. It randomly selects a subset of data for model fitting and evaluates the conformity of all data points, thereby identifying and removing outliers. Specifically, the type of model to be fitted is determined. In the case of smelting furnace point cloud data, a planar model is usually selected because the surface of the smelting furnace is mostly flat. Parameters such as the number of iterations, model threshold, and minimum number of inliers are set. Several points in the point cloud data are randomly selected for fitting the model. The model parameters are calculated using the selected subset of points. The distance from all points to the fitted model is calculated. Points with a distance less than the preset threshold are classified as inliers, and other points are classified as outliers. The quality of the model fit is evaluated based on the number and distribution of inliers. The above steps are repeated multiple times to find the best model, that is, the model with the most inliers. The model with the most inliers is selected and these points are extracted from the original point cloud. The outliers are removed to obtain a smooth point cloud data set.

[0022] Point cloud alignment is to align multiple point cloud data sets into the same coordinate system for easy splicing. The goal of alignment is to make the same objects or areas in multiple point cloud data overlap. Point cloud stitching is to merge the aligned multiple point cloud data sets into a complete point cloud model. Specifically, the overlapping areas in the point cloud data are processed, duplicate points are removed or overlapping areas are merged, and the processed point cloud data are stitched into a complete point cloud data set.

[0023] Triangulation is the process of converting point cloud data into a three-dimensional surface. This is done by connecting the points in the point cloud data into a triangular mesh to form a continuous surface model. A triangulation algorithm, such as one based on the Delaunay criterion, is used to generate a non-overlapping triangular mesh to form a surface mesh model, i.e., a three-dimensional surface model. Each triangle consists of three vertices, which are selected from the point cloud.

[0024] Collect texture images of the smelting furnace and furnace materials. These images come from the shooting of actual scenes. Generate UV coordinates for each point of the 3D model. These coordinates define the position of each pixel in the texture image on the surface of the 3D model. Apply the texture image to the UV coordinates, thereby mapping the 2D texture to the surface of the 3D model to form a complete 3D model of the smelting furnace.

[0025] Multi-dimensional data collection is performed on the smelting furnace and furnace materials through multiple types of detection equipment to obtain multi-dimensional data collection results.

[0026] Various types of detection equipment include temperature sensors, optical imaging equipment, laser rangefinders, etc. The installation location of the detection equipment is determined according to the structure and requirements of the smelting furnace to ensure coverage of key areas. Multi-dimensional data collection is performed through various types of detection equipment, and the data collected by various types of detection equipment are converted into a unified format to integrate and obtain multi-dimensional data collection results.

[0027] The multi-dimensional data collection results are rendered to the three-dimensional model of the smelting furnace to obtain a smelting visualization model.

[0028] The three-dimensional model of the smelting furnace generated based on laser scanning is used as the basic model, and the attributes of the multidimensional data are mapped to the three-dimensional model. For example, the temperature data is mapped to the surface of the model, and a color gradient can be used to represent different temperature areas. By adjusting the color mapping range and details, the changes in the data are clearly displayed. The results of all attribute renderings are combined to generate the final smelting visualization model. This model contains all real-time data and status information, enabling users to more intuitively understand the actual status of the smelting process.

[0029] A monitoring indicator set is read, and the smelting visualization model is monitored based on the monitoring indicator set to obtain a monitoring parameter set of a monitoring item set.

[0030] The monitoring indicator set includes preset fire cover characterization indicators, preset material pushing characterization indicators, preset auxiliary material addition characterization indicators, and preset furnace tamping characterization indicators. The monitoring indicators are associated with the corresponding parts in the smelting visualization model. For example, the furnace temperature indicator is mapped to the temperature distribution area of ​​the furnace in the model. According to the association results, the monitoring parameter set is extracted from the smelting visualization model. The monitoring parameter set includes all actual data values ​​obtained by monitoring according to the monitoring indicators, including temperature data, material height data, material distribution data, etc.

[0031] Furthermore, the method for reading the monitoring indicator set includes:

[0032] Read the monitoring indicator set, wherein the monitoring indicator set includes preset fire cover characterization indicators, preset material pushing characterization indicators, preset auxiliary material addition characterization indicators, and preset furnace tamping characterization indicators; wherein, the preset fire cover characterization indicators include furnace temperature and fire puncture conditions, the preset material pushing characterization indicators include material height and material distribution uniformity, the preset auxiliary material addition characterization indicators include flux usage, reducing agent usage, material melting state, and reaction completion degree, and the preset furnace tamping characterization indicators include material density and material surface state.

[0033] The monitoring indicator set is a set of preset indicators used to evaluate the operating status of the smelting furnace, helping to monitor and evaluate various situations in the production process in real time.

[0034] The preset cover fire characterization indicators are used to monitor the temperature and ignition conditions in the furnace, including monitoring the temperature changes in the furnace, detecting whether it exceeds the normal range, monitoring the ignition conditions in the furnace, and judging whether the reaction is sufficient.

[0035] The preset pushing material characterization indicators are used to monitor the distribution and status of the material in the furnace, including measuring the height of the material in the furnace to determine whether the material is piled too high, evaluating whether the material is evenly distributed in the furnace, and avoiding local overload or collapse.

[0036] The preset auxiliary material addition characterization indicators are used to monitor the usage and effects of auxiliary materials (such as flux and reducing agent), including monitoring the usage of flux and reducing agent to ensure the reaction effect, monitoring the melting state of the material to check whether it reaches the expected state, monitoring the reaction completion, and judging whether it is necessary to continue adding auxiliary materials or adjust the operating parameters.

[0037] Preset tamping characterization indicators are used to monitor the processing status of materials, including measuring the density of the material to evaluate the effect of the tamping operation and checking the status of the material surface to ensure the quality of material processing.

[0038] Furthermore, the method of monitoring the smelting visualization model based on the monitoring indicator set to obtain a monitoring parameter set of a monitoring item set includes:

[0039] Through a set of intelligent monitoring equipment, the furnace temperature and the ignition condition are monitored to obtain the cover fire characterization state monitoring parameters; through a set of intelligent monitoring equipment, the material height and the material distribution uniformity are monitored to obtain the push material characterization state monitoring parameters; through a set of intelligent monitoring equipment, the flux usage, the reducing agent usage, the material melting state, and the reaction completion are monitored to obtain the auxiliary material addition characterization state monitoring parameters; through a set of intelligent monitoring equipment, the material density and the material surface state are monitored to obtain the ramming furnace characterization state monitoring parameters; the cover fire characterization state monitoring parameters, the push material characterization state monitoring parameters, the auxiliary material addition characterization state monitoring parameters, and the ramming furnace characterization state monitoring parameters are integrated to obtain the monitoring parameter set.

[0040] Use high-temperature sensors, such as thermocouples and infrared temperature sensors, to measure the temperature inside the furnace and the ignition status in real time, and generate fire cover status monitoring parameters to determine whether to adjust the fire cover operation.

[0041] Use a laser rangefinder, such as a laser scanner, to measure the height of the material. Use a camera or image processing system to capture images of the interior of the furnace. Use image analysis technology to calculate the distribution uniformity of the material, obtain material height and distribution data, and generate pusher status monitoring parameters to determine whether the pusher operation needs to be adjusted.

[0042] Use flow meters to monitor the flow of flux and reducing agent, use melting point sensors to monitor the melting state of materials, use chemical reaction monitoring equipment to evaluate the completion of the reaction, obtain the flux and reducing agent usage and material melting state data, and generate auxiliary material addition characterization status monitoring parameters to determine whether the auxiliary material addition amount needs to be adjusted.

[0043] Use density sensors to measure the density of the material, and use surface condition sensors, such as laser scanners and high-definition cameras, to check the surface condition of the material, obtain material density and surface condition data, and generate ramming characterization status monitoring parameters to determine whether ramming operations are required.

[0044] The obtained monitoring parameters representing the state of fire cover, the state of material pushing, the state of auxiliary material adding and the state of furnace pounding are integrated into one data set to obtain the monitoring parameter set.

[0045] Based on a pre-built smelting anomaly analysis model, a smelting anomaly analysis is performed on the monitoring parameter set, and a first warning level of a first monitoring anomaly item is extracted according to the smelting anomaly analysis result.

[0046] Collect monitoring data from historical smelting processes, obtain historical abnormal conditions and corresponding monitoring parameters, and use them as training data for the analysis model. Use the training data to train the anomaly analysis model to ensure that the model can identify normal and abnormal states, as well as the warning level of abnormal states. Verify the model and use cross-validation and other methods to evaluate the performance of the model to ensure its accuracy and reliability, and finally obtain a smelting anomaly analysis model.

[0047] The monitoring parameter set is used as a model input and input into the smelting anomaly analysis model for analysis. The anomaly analysis results are output, which include the anomalies present in the monitoring parameter set and the warning level of the anomaly, such as minor, warning, or severe. A first monitoring anomaly item is extracted from the anomaly analysis results. The first monitoring anomaly item is any one of multiple anomalies. By extracting the first monitoring anomaly item, an anomaly handling process for that item is prepared, thereby achieving sequential handling of multiple anomalies.

[0048] Based on the first monitored abnormal item and the first warning level, the industrial robot is activated to perform abnormal processing on the first monitored abnormal item.

[0049] According to the first monitoring abnormality item, the specific abnormality that needs to be handled is clarified, for example, it can be abnormal temperature, material height is too high or collapsed, severe fire, etc. According to the abnormal items and warning levels, a specific treatment plan is formulated. For example, when the furnace temperature rises to a certain level or the oxidation reaction is too intense, it is necessary to use the industrial robot's fire covering tool to perform a fire covering operation. Specifically, the fire covering tool is used to evenly spread covering materials, such as graphite powder or other covering agents, on the surface of the charge. The covering material can form a protective film to reduce the contact between oxygen and the charge, thereby controlling the oxidation reaction.

[0050] Furthermore, the industrial robot integrates a smelting tool set, and the smelting tool set includes a fire covering tool, a furnace tamping tool, a material pushing tool and a material adding tool.

[0051] Fire-covering tools are used to address fire issues in smelting furnaces; tamping tools are used to stir, compact, or otherwise manipulate the materials in the furnace to improve fluidity and reaction quality; pusher tools are used to move materials within or outside the furnace, ensuring proper distribution and smooth entry and exit processes; and charging tools are used to add auxiliary materials, such as flux and reducing agents, to the smelting furnace to promote chemical reactions or adjust material properties. Industrial robots integrate multiple tools. Their modular design allows for rapid tool replacement and adjustment when needed. Through intelligent tool coordination, robots can respond to various production issues in real time, improving the efficiency, safety, and quality of the smelting process.

[0052] Furthermore, constructing the smelting anomaly analysis model includes:

[0053] Based on the historical smelting monitoring records, the historical cover fire characterization status monitoring parameter set, the historical material pushing characterization status monitoring parameter set, the historical auxiliary material addition characterization status monitoring parameter set, and the historical furnace ramming characterization status monitoring parameter set are extracted; based on the historical cover fire characterization status monitoring parameter set, the historical material pushing characterization status monitoring parameter set, the historical auxiliary material addition characterization status monitoring parameter set, and the historical furnace ramming characterization status monitoring parameter set, a cover fire abnormality classification early warning device, a material pushing abnormality classification early warning device, an auxiliary material addition abnormality classification early warning device, and a furnace ramming abnormality classification early warning device are constructed, and the smelting abnormality analysis model is obtained through integration.

[0054] Collect historical monitoring records from monitoring systems, data recorders, databases and other channels during the smelting process, classify historical data according to monitoring indicators, and extract historical fire cover characterization status monitoring parameter sets, historical material pushing characterization status monitoring parameter sets, historical auxiliary material addition characterization status monitoring parameter sets, and historical furnace pounding characterization status monitoring parameter sets.

[0055] Analyze the abnormal patterns and trends of historical fire cover characterization status monitoring parameters (such as furnace temperature and sparking conditions), identify fire cover abnormal patterns in historical data, including temperature abnormalities, frequent sparking, etc. Based on the identified abnormal patterns, build a fire cover abnormality classification early warning device for anomaly detection of real-time data. Use the same method to analyze the historical push characterization status monitoring parameter set, the historical auxiliary material addition characterization status monitoring parameter set, and the historical ramming characterization status monitoring parameter set in turn, and build a push abnormality classification early warning device, an auxiliary material addition abnormality classification early warning device, and a ramming abnormality classification early warning device. Integrate the above-mentioned various abnormality classification early warning devices into a comprehensive smelting abnormality analysis model for analyzing different abnormal situations.

[0056] Furthermore, the method for constructing the cover fire abnormality classification early warning device includes:

[0057] Perform a status analysis on the historical cover fire characterization status monitoring parameter set to obtain a historical cover fire characterization status analysis result, wherein the historical cover fire characterization status analysis result includes multiple historical abnormal cover fire characterization states and multiple historical state deviation amplitudes of the multiple historical abnormal cover fire characterization states; mark the multiple historical state deviation amplitudes with warning levels to obtain multiple historical warning levels; use the historical cover fire characterization status monitoring parameter set and the multiple historical abnormal cover fire characterization states to construct a cover fire characterization status abnormality analysis branch; use the historical cover fire characterization status monitoring parameter set and the multiple historical warning levels to construct a cover fire warning level analysis branch; integrate the cover fire characterization status abnormality analysis branch and the cover fire warning level analysis branch to obtain the cover fire abnormality classification alarm.

[0058] The set of historical fire-capping characterization status monitoring parameters includes historical monitoring data such as furnace temperature and sparking conditions. The normal range is defined according to actual conditions and specific needs. For example, the furnace temperature should be maintained within a certain range under normal operation. Identify abnormal conditions in historical data, including furnace temperature that is significantly higher or lower than the normal range, etc. Identify and extract abnormal fire-capping conditions that have occurred in historical data. These abnormal conditions include abnormal parameter values, such as abnormal temperature and frequent sparking. Calculate the deviation of the abnormal state from the normal state, such as calculating the absolute difference between the abnormal state parameter value and the normal state parameter value, to generate the historical state deviation amplitude.

[0059] Alert levels are defined based on abnormal conditions and deviations in historical data. For example, if there are no abnormal conditions or the deviation is within an acceptable range, it is marked as normal; if the deviation is outside the normal range but not serious and requires attention, it is marked as warning; if the deviation is significant and may cause safety issues or production interruptions and requires immediate attention, it is marked as critical. Statistical analysis is performed on the deviations of multiple historical conditions, and each abnormal condition is assigned a warning level, resulting in multiple historical warning levels.

[0060] The system uses a set of historical monitoring parameters representing fire capping states as input data, and multiple historical abnormal fire capping states as output data, which includes labeled abnormal state samples indicating abnormal conditions. The data is divided into training, validation, and test sets. The neural network model is trained on the training set and validated on the validation set to prevent overfitting. The neural network's hyperparameters are adjusted to optimize model performance, and the model's accuracy (i.e., the proportion of correct classifications) is evaluated on the test set. Based on the evaluation results, the number of network layers and the number of neurons in each layer are adjusted to improve model performance. Ultimately, a model that meets the preset accuracy is obtained. This branch serves as the fire capping state abnormality analysis branch, which can learn abnormal patterns from historical data and perform anomaly detection on real-time data, supporting the safety and efficiency of the smelting process.

[0061] The same training method as that for the abnormal analysis branch of the fire characterization state is used to train the fire warning level analysis branch. For the sake of brevity of the specification, it will not be repeated here.

[0062] The cover fire characterization state abnormality analysis branch and the cover fire warning level analysis branch are integrated to form a comprehensive cover fire abnormality classification early warning device, which can comprehensively consider various abnormal information and generate accurate early warning results, thereby improving the safety and operational efficiency of the smelting process.

[0063] In summary, the intelligent monitoring method for industrial silicon smelting furnace cover fire provided by the embodiment of the present application has the following technical effects:

[0064] The 3D model constructed using laser scanning equipment and point cloud data sets provides a detailed visual representation of the smelting furnace and furnace materials. Multidimensional data from different detection equipment is rendered onto the 3D model, forming a comprehensive smelting visualization model. This allows various types of data to be analyzed and monitored on a unified platform, providing an intuitive operating interface. By reading and monitoring based on a set of monitoring indicators, comprehensive real-time tracking of key parameters in the smelting process is possible. Based on a pre-built smelting anomaly analysis model, in-depth analysis of the monitoring parameter set is performed, anomalies can be accurately identified, and warning levels can be extracted based on the analysis results, improving the accuracy and reliability of anomaly detection. Based on the anomaly detection results, industrial robots are activated for anomaly handling. By introducing industrial robots to replace manual labor in covering the fire, tamping the furnace, pushing materials, and adding materials, employees are prevented from directly contacting high-temperature furnace materials and high-temperature environments, thereby significantly reducing the safety risks of high-temperature radiation and burns, and ensuring the safety of operators.

[0065] Based on the same inventive concept as the intelligent monitoring method for industrial silicon smelting ramming furnace cover fire in the aforementioned embodiment, Figure 2 As shown, the embodiment of the present application provides an intelligent monitoring system for industrial silicon smelting furnace cover fire, the system comprising:

[0066] A three-dimensional model construction module 10 is used to perform an overall scan of the industrial silicon smelting furnace and furnace materials through laser scanning equipment, obtain a point cloud data set, and construct a three-dimensional model of the smelting furnace based on the point cloud data set; a multi-dimensional data acquisition module 20 is used to perform multi-dimensional data acquisition on the smelting furnace and furnace materials through multiple types of detection equipment to obtain multi-dimensional data acquisition results; a visualization model construction module 30 is used to render the multi-dimensional data acquisition results to the smelting furnace three-dimensional model to obtain a smelting visualization model; a monitoring parameter acquisition module 40 is used to read a monitoring indicator set, monitor the smelting visualization model based on the monitoring indicator set, and obtain a monitoring parameter set of a monitoring item set; a smelting anomaly analysis module 50 is used to perform smelting anomaly analysis on the monitoring parameter set based on a pre-constructed smelting anomaly analysis model, and extract a first warning level of a first monitoring anomaly item based on the smelting anomaly analysis result; an anomaly handling module 60 is used to activate the industrial robot based on the first monitoring anomaly item and the first warning level, and perform anomaly handling of the first monitoring anomaly item.

[0067] Furthermore, the industrial robot integrates a smelting tool set, and the smelting tool set includes a fire covering tool, a furnace tamping tool, a material pushing tool and a material adding tool.

[0068] Furthermore, the system further includes a model building module to perform the following steps:

[0069] Based on the principle of random sampling consistency, outliers are removed from the point cloud data set to obtain a smooth point cloud data set; point cloud alignment and splicing are performed on the smooth point cloud data set to obtain a constructed point cloud data set; triangulated meshing is performed on the constructed point cloud data set to generate a three-dimensional surface model; surface texture mapping of the silicon smelting furnace and the furnace material is performed based on the three-dimensional surface model to obtain a three-dimensional model of the smelting furnace.

[0070] Furthermore, the system further includes a monitoring indicator reading module to perform the following operation steps:

[0071] Read the monitoring indicator set, wherein the monitoring indicator set includes preset fire cover characterization indicators, preset material pushing characterization indicators, preset auxiliary material addition characterization indicators, and preset furnace tamping characterization indicators; wherein, the preset fire cover characterization indicators include furnace temperature and fire puncture conditions, the preset material pushing characterization indicators include material height and material distribution uniformity, the preset auxiliary material addition characterization indicators include flux usage, reducing agent usage, material melting state, and reaction completion degree, and the preset furnace tamping characterization indicators include material density and material surface state.

[0072] Furthermore, the system further includes a monitoring parameter set acquisition module to perform the following operation steps:

[0073] Through a set of intelligent monitoring equipment, the furnace temperature and the ignition condition are monitored to obtain the cover fire characterization state monitoring parameters; through a set of intelligent monitoring equipment, the material height and the material distribution uniformity are monitored to obtain the push material characterization state monitoring parameters; through a set of intelligent monitoring equipment, the flux usage, the reducing agent usage, the material melting state, and the reaction completion are monitored to obtain the auxiliary material addition characterization state monitoring parameters; through a set of intelligent monitoring equipment, the material density and the material surface state are monitored to obtain the ramming furnace characterization state monitoring parameters; the cover fire characterization state monitoring parameters, the push material characterization state monitoring parameters, the auxiliary material addition characterization state monitoring parameters, and the ramming furnace characterization state monitoring parameters are integrated to obtain the monitoring parameter set.

[0074] Furthermore, the system also includes a smelting anomaly analysis model building module to perform the following steps:

[0075] Based on the historical smelting monitoring records, the historical cover fire characterization status monitoring parameter set, the historical material pushing characterization status monitoring parameter set, the historical auxiliary material addition characterization status monitoring parameter set, and the historical furnace ramming characterization status monitoring parameter set are extracted; based on the historical cover fire characterization status monitoring parameter set, the historical material pushing characterization status monitoring parameter set, the historical auxiliary material addition characterization status monitoring parameter set, and the historical furnace ramming characterization status monitoring parameter set, a cover fire abnormality classification early warning device, a material pushing abnormality classification early warning device, an auxiliary material addition abnormality classification early warning device, and a furnace ramming abnormality classification early warning device are constructed, and the smelting abnormality analysis model is obtained through integration.

[0076] Furthermore, the system further includes a fire anomaly classification early warning device construction module to perform the following operation steps:

[0077] Perform a status analysis on the historical cover fire characterization status monitoring parameter set to obtain a historical cover fire characterization status analysis result, wherein the historical cover fire characterization status analysis result includes multiple historical abnormal cover fire characterization states and multiple historical state deviation amplitudes of the multiple historical abnormal cover fire characterization states; mark the multiple historical state deviation amplitudes with warning levels to obtain multiple historical warning levels; use the historical cover fire characterization status monitoring parameter set and the multiple historical abnormal cover fire characterization states to construct a cover fire characterization status abnormality analysis branch; use the historical cover fire characterization status monitoring parameter set and the multiple historical warning levels to construct a cover fire warning level analysis branch; integrate the cover fire characterization status abnormality analysis branch and the cover fire warning level analysis branch to obtain the cover fire abnormality classification alarm.

[0078] Through the detailed description of the intelligent monitoring method for industrial silicon smelting ramming furnace cover fire in the above specification, those skilled in the art can clearly understand the intelligent monitoring system for industrial silicon smelting ramming furnace cover fire in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0079] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent monitoring method for industrial silicon smelting furnace cover fire, characterized in that: The method comprises: Scan the industrial silicon smelting furnace and the furnace materials as a whole by using a laser scanning device to obtain a point cloud data set, and construct a three-dimensional model of the smelting furnace according to the point cloud data set; Multi-dimensional data collection of smelting furnace and furnace materials is carried out through various types of detection equipment to obtain multi-dimensional data collection results; Rendering the multi-dimensional data collection results to the three-dimensional model of the smelting furnace to obtain a smelting visualization model; Reading a monitoring indicator set, monitoring the smelting visualization model based on the monitoring indicator set, and obtaining a monitoring parameter set of a monitoring item set; Based on a pre-built smelting anomaly analysis model, performing smelting anomaly analysis on the monitoring parameter set, and extracting a first warning level of a first monitoring anomaly item according to the smelting anomaly analysis result; Based on the first monitored abnormal item and the first warning level, the industrial robot is activated to perform abnormal processing on the first monitored abnormal item.

2. The intelligent monitoring method for industrial silicon smelting furnace cover fire according to claim 1, characterized in that: The industrial robot integrates a smelting tool set, and the smelting tool set includes a fire covering tool, a furnace tamping tool, a material pushing tool and a material feeding tool.

3. The intelligent monitoring method for industrial silicon smelting furnace cover fire according to claim 1, characterized in that: The method of constructing a three-dimensional model of a smelting furnace according to the point cloud data set includes: Based on the principle of random sampling consistency, outliers are removed from the point cloud data set to obtain a smooth point cloud data set; Performing point cloud alignment and splicing on the smoothed point cloud data set to obtain a constructed point cloud data set; Performing triangulation mesh processing on the constructed point cloud data set to generate a three-dimensional surface model; The surface texture mapping of the silicon smelting furnace and the furnace material is performed based on the three-dimensional surface model to obtain the three-dimensional model of the smelting furnace.

4. The intelligent monitoring method for industrial silicon smelting furnace cover fire according to claim 1, characterized in that: The method for reading the monitoring indicator set includes: Reading a monitoring index set, wherein the monitoring index set includes a preset fire cover characterization index, a preset material pushing characterization index, a preset auxiliary material adding characterization index, and a preset furnace pounding characterization index; Among them, the preset fire cover characterization indicators include furnace temperature and fire puncture conditions, the preset material pushing characterization indicators include material height and material distribution uniformity, the preset auxiliary material addition characterization indicators include flux usage, reducing agent usage, material melting state, reaction completion degree, and the preset furnace pounding characterization indicators include material density and material surface state.

5. The intelligent monitoring method for industrial silicon smelting furnace cover fire according to claim 4, characterized in that: The method of monitoring the smelting visualization model based on the monitoring indicator set to obtain a monitoring parameter set of a monitoring item set includes: Through a collection of intelligent monitoring equipment, the furnace temperature and the ignition condition are monitored to obtain the monitoring parameters of the fire characterization state; Through a collection of intelligent monitoring equipment, the material height and material distribution uniformity are monitored to obtain monitoring parameters of the material pushing state; Through a collection of intelligent monitoring equipment, the amount of flux used, the amount of reducing agent used, the melting state of the material, and the degree of completion of the reaction are monitored to obtain monitoring parameters characterizing the state of auxiliary material addition; Through a collection of intelligent monitoring equipment, the density and surface state of the material are monitored to obtain monitoring parameters representing the state of the ramming furnace; The monitoring parameter set is obtained by integrating the fire cover status monitoring parameters, material pushing status monitoring parameters, auxiliary material adding status monitoring parameters, and furnace pounding status monitoring parameters.

6. The intelligent monitoring method for industrial silicon smelting furnace cover fire according to claim 1, characterized in that: Constructing the smelting anomaly analysis model includes: Based on the historical smelting monitoring records, extract the historical fire cover characterization state monitoring parameter set, the historical material pushing characterization state monitoring parameter set, the historical auxiliary material addition characterization state monitoring parameter set, and the historical furnace pounding characterization state monitoring parameter set; Based on the historical set of monitoring parameters for characterizing the state of covering fire, the historical set of monitoring parameters for characterizing the state of pushing material, the historical set of monitoring parameters for characterizing the state of auxiliary material addition, and the historical set of monitoring parameters for characterizing the state of ramming furnace, a cover fire abnormality classification early warning device, a push material abnormality classification early warning device, an auxiliary material addition abnormality classification early warning device, and a ramming furnace abnormality classification early warning device are constructed, and the smelting abnormality analysis model is obtained by integration.

7. The intelligent monitoring method for industrial silicon smelting furnace cover fire according to claim 6, characterized in that: The method for constructing the cover fire abnormality classification early warning device includes: Performing state analysis on the historical fire cover characterization state monitoring parameter set to obtain a historical fire cover characterization state analysis result, wherein the historical fire cover characterization state analysis result includes a plurality of historical abnormal fire cover characterization states and a plurality of historical state deviation amplitudes of the plurality of historical abnormal fire cover characterization states; Marking the multiple historical state deviation amplitudes with warning levels to obtain multiple historical warning levels; Using the historical cover fire characterization state monitoring parameter set and the multiple historical abnormal cover fire characterization states, construct a cover fire characterization state abnormality analysis branch; Using the historical fire characterization state monitoring parameter set and the multiple historical warning levels, constructing a fire warning level analysis branch; The cover fire characterization state abnormality analysis branch and the cover fire warning level analysis branch are integrated to obtain the cover fire abnormality classification warning device.

8. Intelligent monitoring system for industrial silicon smelting furnace cover fire, characterized in that: The intelligent monitoring method for industrial silicon smelting ramming furnace cover fire according to any one of claims 1 to 7 is used, and the system comprises: A three-dimensional model building module is used to perform an overall scan of an industrial silicon smelting furnace and furnace materials using a laser scanning device to obtain a point cloud data set, and to build a three-dimensional model of the smelting furnace based on the point cloud data set; The multi-dimensional data acquisition module is used to collect multi-dimensional data of the smelting furnace and furnace materials through multiple types of detection equipment to obtain multi-dimensional data acquisition results; A visualization model building module, used for rendering the multi-dimensional data collection results to the three-dimensional model of the smelting furnace to obtain a smelting visualization model; A monitoring parameter acquisition module, used for reading a monitoring indicator set, monitoring the smelting visualization model based on the monitoring indicator set, and obtaining a monitoring parameter set of a monitoring project set; A smelting anomaly analysis module, used to perform smelting anomaly analysis on the monitoring parameter set based on a pre-built smelting anomaly analysis model, and extract a first warning level of a first monitoring anomaly item according to the smelting anomaly analysis result; An exception handling module is used to activate the industrial robot and perform exception handling of the first monitored abnormal item based on the first monitored abnormal item and the first warning level.

Images