Intelligent control method and system for refractory brick production line

Abstract

The application discloses an intelligent control method and system for a firebrick production line, and relates to the technical field of intelligent control methods. A plurality of abnormal features of the firebrick are determined according to abnormal production data and a discharge image of the firebrick. In the optimization process of the firebrick, a plurality of optimization data are generated according to the optimization of the plurality of abnormal features of the firebrick, the optimized production data are determined based on the plurality of optimization data and corresponding abnormal production data, and an intelligent feedback mechanism is generated, thereby ensuring the accuracy of the optimized production data and reducing the defective rate of the next firebrick in the production process. Therefore, in the detection process of the firebrick, intelligent detection of a plurality of detection heads relative to the firebrick is triggered according to the plurality of optimization data and the morphology of the firebrick, and an intelligent detection mechanism is generated. The intelligent control mechanism of the firebrick production line is determined according to the training of the intelligent detection mechanism and the intelligent feedback mechanism, thereby ensuring the accuracy of the intelligent control mechanism of the firebrick production line.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control methods, and particularly relates to an intelligent control method and system for a firebrick production line. BACKGROUND

[0002] With the development of science and technology, firebricks are applied to people's lives and serve as a kind of fire-resistant brick. A firebrick production line is used for continuous production of firebricks and outputs a plurality of firebricks. In the prior art, the firebrick production line has a plurality of processes, each of which is sorted along a preset order. However, each process is relatively independent and does not affect each other. Once the firebricks have abnormal characteristics, the firebricks are optimized in the optimization process and do not affect the production of the production process. However, the production process still produces firebricks with abnormal characteristics, which cannot reduce the defective rate of the next firebrick in the production process. SUMMARY

[0003] The present application aims to overcome the shortcomings of the prior art and provides an intelligent control method and system for a firebrick production line.

[0004] The present application provides an intelligent control method for a firebrick production line, which comprises the following steps: determining a production process, an optimization process and a detection process of firebricks according to a firebrick production line; in the production process of the firebricks, determining abnormal production data according to a production data set of the firebricks, and determining a plurality of abnormal characteristics of the firebricks according to the abnormal production data and an output image of the firebricks; in the optimization process of the firebricks, generating a plurality of optimization data according to optimization of the plurality of abnormal characteristics of the firebricks, determining optimized production data based on the plurality of optimization data and the corresponding abnormal production data, and generating an intelligent feedback mechanism; in the detection process of the firebricks, triggering intelligent detection of a plurality of detection heads relative to the firebricks according to the plurality of optimization data and a shape of the firebricks, and generating an intelligent detection mechanism; and determining an intelligent control mechanism of the firebrick production line according to training of the intelligent detection mechanism and the intelligent feedback mechanism, wherein the intelligent control mechanism of the firebrick production line intelligently controls the production process, the optimization process and the detection process of the firebricks.

[0005] The present application provides an intelligent control system for a firebrick production line, which is applied to the intelligent control method for the firebrick production line as described above. The intelligent control system for the firebrick production line comprises the following components:

[0006] A process module is configured to determine a production process, an optimization process and a detection process of firebricks according to a firebrick production line.

[0007] An abnormal feature module is configured to determine abnormal production data according to a production data set of the refractory bricks in a production process of the refractory bricks, and determine a plurality of abnormal features of the refractory bricks according to the abnormal production data and an outfeed image of the refractory bricks.

[0008] An intelligent feedback module is configured to generate a plurality of optimization data according to optimization of the plurality of abnormal features of the refractory bricks in an optimization process of the refractory bricks, determine optimized production data based on the plurality of optimization data and corresponding abnormal production data, and generate an intelligent feedback mechanism.

[0009] An intelligent detection module is configured to trigger intelligent detection of a plurality of detection heads relative to the refractory bricks according to the plurality of optimization data and a shape of the refractory bricks in a detection process of the refractory bricks, and generate an intelligent detection mechanism.

[0010] An intelligent control module is configured to determine an intelligent control mechanism of a refractory brick production line according to training of the intelligent detection mechanism and the intelligent feedback mechanism, and the intelligent control mechanism of the refractory brick production line intelligently controls the production process, the optimization process and the detection process of the refractory bricks.

[0011] Compared with the prior art, the present application has the following advantages:

[0012] (1) The present application determines the production process, the optimization process and the detection process of the refractory bricks according to the refractory brick production line, determines abnormal production data according to a production data set of the refractory bricks in a production process of the refractory bricks, and determines a plurality of abnormal features of the refractory bricks according to the abnormal production data and an outfeed image of the refractory bricks, generates a plurality of optimization data according to optimization of the plurality of abnormal features of the refractory bricks in an optimization process of the refractory bricks, determines optimized production data based on the plurality of optimization data and corresponding abnormal production data, and generates an intelligent feedback mechanism, which fully utilizes the plurality of optimization data, compares the plurality of optimization data and the abnormal production data, ensures the accuracy of the optimized production data, and realizes intelligent feedback of the optimized production data based on the intelligent feedback mechanism to reduce the defective rate of the next refractory brick in the production process.

[0013] (2) The present application triggers intelligent detection of a plurality of detection heads relative to the refractory bricks according to the plurality of optimization data and a shape of the refractory bricks in a detection process of the refractory bricks, and generates an intelligent detection mechanism, determines an intelligent control mechanism of a refractory brick production line according to training of the intelligent detection mechanism and the intelligent feedback mechanism, and the intelligent control mechanism of the refractory brick production line intelligently controls the production process, the optimization process and the detection process of the refractory bricks, which is compatible with the overall consideration of the intelligent detection mechanism and the intelligent feedback mechanism, ensures the accuracy of the intelligent control mechanism of the refractory brick production line, and intelligently controls the production process, the optimization process and the detection process.

[0014] (3) The present application combines the positions and function types of multiple functional areas, determines the sequence and connection of each production process according to the production process of refractory bricks, identifies the links that need to be optimized and the quality indicators that need to be detected on the basis of the production process, determines the optimization process and the detection process, identifies the key links between the production process and the detection process according to the production process, sets the optimization process after these links, so as to adjust and optimize according to the production results and detection results, and ensures that the optimization process can respond to changes in the production process in a timely manner, improving production efficiency and product quality.

[0015] (4) The present application combines negative feedback paths, positive feedback paths and intelligent control mechanisms to form an overall control path for the refractory brick production line, which will run through the entire production process from raw material preparation to finished product delivery, ensuring that each link can be effectively monitored and adjusted. By integrating various sensors, actuators, control systems and data analysis software, a complete automated control system can be built to monitor the state of the production line in real time and automatically adjust production parameters or take corrective measures according to the decision of the intelligent control mechanism. At the same time, the system also has the ability to learn and optimize itself, ensuring continuous improvement of production efficiency and product quality. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flowchart of the intelligent control method of the refractory brick production line in the embodiment of the present application;

[0017] Figure 2 is a flowchart of step S11 in the intelligent control method of the refractory brick production line in the embodiment of the present application;

[0018] Figure 3 is a flowchart of step S12 in the intelligent control method of the refractory brick production line in the embodiment of the present application;

[0019] Figure 4 is a flowchart of step S13 in the intelligent control method of the refractory brick production line in the embodiment of the present application;

[0020] Figure 5 is a flowchart of step S14 in the intelligent control method of the refractory brick production line in the embodiment of the present application;

[0021] Figure 6 is a flowchart of step S15 in the intelligent control method of the refractory brick production line in the embodiment of the present application;

[0022] Figure 7 is a structural composition diagram of the intelligent control system of the refractory brick production line in the embodiment of the present application. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

[0024] Please refer to Figures 1 to 7 An intelligent control method of a refractory brick production line comprises the following steps:

[0025] Step S11: determining a production process, an optimization process and a detection process of the refractory brick according to the refractory brick production line;

[0026] Step S12: in the production process of the refractory brick, determining abnormal production data according to a production data set of the refractory brick, and determining a plurality of abnormal features of the refractory brick according to the abnormal production data and an outfeed image of the refractory brick;

[0027] Step S13: in the optimization process of the refractory brick, generating a plurality of optimization data according to optimization of the plurality of abnormal features of the refractory brick, determining optimized production data based on the plurality of optimization data and corresponding abnormal production data, and generating an intelligent feedback mechanism;

[0028] Step S14: in the detection process of the refractory brick, triggering intelligent detection of a plurality of detection heads relative to the refractory brick according to the plurality of optimization data and a shape of the refractory brick, and generating an intelligent detection mechanism;

[0029] Step S15: determining an intelligent control mechanism of the refractory brick production line according to training of the intelligent detection mechanism and the intelligent feedback mechanism, and the intelligent control mechanism of the refractory brick production line intelligently controls the production process, the optimization process and the detection process of the refractory brick;

[0030] Reference Figure 2 In step S11, the production process, the optimization process and the detection process of the refractory brick are determined according to the refractory brick production line;

[0031] In the specific implementation process of the present application, the specific steps are as follows:

[0032] S111: collecting a distribution map of the refractory brick production line, and generating a plurality of functional areas according to division of the distribution map of the refractory brick production line;

[0033] S112: in each functional area, determining a functional type of the functional area based on past data of the functional area and past processing events of the functional area, and determining the production process, the optimization process and the detection process of the refractory brick according to positions of the plurality of functional areas and the functional types of the plurality of functional areas;

[0034] S113: in the refractory brick production line, the optimization process is between the production process and the detection process, the optimization process and the production process construct a negative feedback path, the optimization process and the detection process construct a positive feedback path, and the negative feedback path and the positive feedback path are synchronously executed with response of the optimization process.

[0035] In the embodiments of the present application, the distribution map of the refractory brick production line is collected, and a plurality of functional areas are generated according to the division of the distribution map of the refractory brick production line.

[0036] At this time, the detailed layout of the refractory brick production line is collected, which is usually a piece of paper or a digital map showing the positions of all equipment, raw material storage area, product output area, personnel passage and material flow path in the production line; optionally, these information can be obtained by on-site investigation, communication with production line managers or visiting the digital management system of the production line; understanding the physical layout of the production line provides a basis for subsequent functional area division;

[0037] Based on the collected production line distribution map, the production line is divided into a plurality of logically independent but interrelated functional areas; at this time, first identify the key equipment on the production line, such as raw material mixer, forming machine, drying kiln, firing furnace and quality inspection equipment, etc.; according to the positions of these equipment and their roles in the production process, the production line is divided into different functional areas, such as production area, optimization area and detection area, etc.; when dividing the areas, the flow path of the materials on the production line also needs to be considered to ensure that the materials can flow smoothly from one area to the next; dividing the production line into functional areas that are easy to manage and optimize improves production efficiency and product quality.

[0038] Further, in each functional area, the functional type of the functional area is determined based on the past data of the functional area and the past processing events of the functional area, and the production process, optimization process and detection process of the refractory bricks are determined according to the positions of the plurality of functional areas and the functional types of the plurality of functional areas, which is compatible with the overall consideration of the positions of the plurality of functional areas and the functional types of the plurality of functional areas, and ensures the accuracy of the production process, optimization process and detection process of the refractory bricks.

[0039] At this time, for each functional area, collect and analyze its past production data, processing event records, etc. to clarify the main function and role of the area; at this time, collect data from the management system of the production line, equipment log, personnel record, etc.; use statistical analysis, data mining, etc. to identify the main operation, processing flow, production parameter, etc. of each functional area; according to the analysis results, define clear functional types for each functional area, such as raw material mixing, forming and pressing, high-temperature firing, quality detection, etc.; understand the specific function and role of each functional area to provide a basis for subsequent production process planning.

[0040] According to the production process of the refractory brick, the order and connection of each production process are determined. At this time, according to the production process of the refractory brick, the material flow, information transmission and personnel cooperation relationship between each functional area are analyzed. According to the analysis result, the functional areas are connected according to the order of the production process to form a complete production process chain. The smoothness and efficiency of the production process are ensured, and the overall performance of the production line is improved.

[0041] On the basis of the production process, the link needing optimization and the quality index needing detection are identified, and the optimization process and the detection process are determined. At this time, through the analysis of the information such as production data, equipment performance and product quality, the bottleneck, waste and potential improvement point in the production process are identified. According to the product standard and customer demand, the quality index needing detection is determined, such as size, appearance, strength, thermal stability, etc. The optimization process and the detection process are inserted into the appropriate position of the production process chain to form a complete production, optimization and detection process. Through optimization and detection, the product quality and production efficiency are improved, and the production cost is reduced.

[0042] Therefore, in the refractory brick production line, the optimization process is between the production process and the detection process, the optimization process and the production process construct a negative feedback path, and the optimization process and the detection process construct a positive feedback path. The negative feedback path and the positive feedback path are executed synchronously with the response of the optimization process.

[0043] At this time, in the refractory brick production line, the position of the optimization process between the production process and the detection process is clear. According to the production process, the key link between the production process and the detection process is identified, and the optimization process is set after these links in order to adjust and optimize according to the production result and the detection result. It is ensured that the optimization process can respond to the changes in the production process in time, and the production efficiency and product quality are improved.

[0044] A negative feedback path is established between the production process and the optimization process. At this time, the data in the production process is collected in real time, such as raw material mixing ratio, forming pressure, drying temperature, etc. The abnormal value or deviation from the standard value in the production data is identified by using data analysis technology. When the production anomaly is detected, the optimization process is triggered to adjust or optimize the production parameters. After the optimization process is completed, the adjustment result is fed back to the production process to form a closed loop control. Through the negative feedback path, the problems in the production process are found and corrected in time to ensure the stability and controllability of the production process.

[0045] A positive feedback path is established between the optimization process and the detection process; at this time, data in the detection process, such as product size, appearance quality, strength, etc., are collected in real time; according to product standards and customer needs, the detection results are evaluated to determine whether the product quality meets the standards; when the detection results show that the product quality is improved or meets the standards, this information is transmitted as positive feedback to the optimization process, encouraging it to maintain or further improve the current optimization strategy; based on the positive feedback, the optimization process continuously adjusts and optimizes production parameters to improve product quality and production efficiency; through the positive feedback path, the optimization process is encouraged to continuously improve, promoting the continuous optimization and upgrading of the production process.

[0046] Specifically, taking a refractory brick production line as an example, a specific feedback path is constructed: production process: raw materials are mixed, molded, dried, etc., to form a preliminary refractory brick shape; negative feedback path: after the drying process, a sensor is set to monitor the drying temperature and time; if the sensor detects that the drying temperature or time deviates from the standard value, the optimization process is triggered to adjust the drying parameters; after the optimization process is completed, the adjusted drying parameters are fed back to the production process to ensure that the subsequent produced refractory bricks have the appropriate drying degree.

[0047] Optimization process: according to the feedback of the production process and the detection process, the production parameters are adjusted and optimized; for example, adjusting the raw material mixing ratio, molding pressure, drying temperature, etc.; positive feedback path: in the detection process, use quality inspection equipment to detect the size, appearance quality, strength, etc. of the refractory bricks; if the detection results show that the product quality is improved or meets the standards, this information is transmitted as positive feedback to the optimization process; the optimization process continues to maintain or improve the current optimization strategy based on the positive feedback to improve product quality and production efficiency; through such a feedback path, the refractory brick production line can realize real-time monitoring, adjustment and optimization, ensuring the stability and controllability of the production process, while improving product quality and production efficiency.

[0048] Reference Figure 3 In step S12, in the production process of the refractory bricks, abnormal production data is determined according to the production data set of the refractory bricks, and a plurality of abnormal features of the refractory bricks are determined according to the abnormal production data and the discharge image of the refractory bricks;

[0049] In the specific implementation process of the present application, the specific steps are:

[0050] S121: Real-time monitoring of the production process of the refractory bricks, the refractory bricks are produced step by step in the production process, and a plurality of production data of the refractory bricks are output, based on the plurality of production data and the corresponding state image of the refractory bricks, a production data set of the refractory bricks is determined;

[0051] S122: Determine the abnormal surface of each refractory brick from the detection of the production data set of the refractory brick, and determine the corresponding production data from the tracing of each abnormal surface, and determine the production data as abnormal production data;

[0052] S123: When the production process of the refractory brick is completed, the refractory brick is discharged through the discharge port of the refractory brick production line, and the corresponding camera captures the discharge image of the refractory brick, and determines the multiple abnormal features of the refractory brick according to the matching of the discharge image of the refractory brick and the abnormal production data;

[0053] In the embodiments of the present application, the production process of the refractory brick is monitored in real time, the refractory brick is gradually produced in the production process, and multiple production data of the refractory brick are output, and the production data set of the refractory brick is determined based on the multiple production data and the state image of the corresponding refractory brick, which is compatible with the overall consideration of the multiple production data and the state image of the corresponding refractory brick, and ensures the accuracy of the production data set of the refractory brick.

[0054] At this time, the running state of each production link needs to be continuously and uninterruptedly monitored throughout the production process of the refractory brick; At this time, various sensors and monitoring devices installed on the production line, such as temperature sensors, pressure sensors, weight sensors, and high-definition cameras, are used to capture key data in real time during the production process; Ensure a comprehensive understanding of the production process, and timely discover and solve potential problems.

[0055] The production of refractory bricks usually includes multiple steps such as raw material preparation, mixing, molding, drying, and firing, which need to be carried out in a certain order and time interval; At this time, according to the production plan and process requirements, each production link is gradually promoted to ensure that each step is carried out according to the predetermined parameters and conditions; Ensure the production quality and efficiency of the refractory brick.

[0056] During the production process, a large amount of data will be generated, including the proportioning of raw materials, the molding pressure, the drying temperature, the firing time, etc.; At this time, the sensors and monitoring devices will transmit the captured data to the data management system or control center in real time; Provide a basis for subsequent data analysis and quality traceability.

[0057] Combine the real-time output production data with the corresponding refractory brick state image to form a complete and accurate production data set; At this time, image recognition technology is used to analyze the state image and extract feature information related to the quality of the refractory brick, such as color, texture, shape, etc.; Then, associate these feature information with production data to form a production data set containing multiple dimensional information; Provide comprehensive and accurate data support for quality control and anomaly detection of refractory bricks.

[0058] Specifically, assume that a batch of refractory bricks of a specific specification is being produced on a refractory brick production line; in the raw material mixing area, temperature sensors and weight sensors record the temperature of the mixture and the weights of the components in real time; in the forming area, pressure sensors record the pressure output of the forming machine; in the drying area, temperature sensors record the temperature inside the drying kiln; in the firing area, thermocouples record the temperature curve inside the firing furnace.

[0059] After the raw materials are accurately measured, they are mixed uniformly according to the proportion; the mixture is pressed into a specific shape of refractory brick by the forming machine; the formed refractory brick enters the drying kiln for drying treatment; the dried refractory brick enters the firing furnace for high-temperature firing.

[0060] Output production data: temperature in the raw material mixing area: 120℃, weight of each component: high alumina 50kg, quartz sand 30kg, other additives 20kg; pressure in the forming area: 20MPa; temperature in the drying area: 100℃, drying time: 24 hours; temperature curve in the firing area: from room temperature to 1300℃, holding for 4 hours, then dropping to room temperature.

[0061] In each production link, high-definition cameras will capture the state images of the refractory bricks, such as the color of the mixed raw materials, the shape after forming, the surface texture after drying, and the color uniformity after firing, etc.; using image recognition technology, these state images are analyzed to extract feature information such as color, texture, shape, etc.; these feature information is associated with production data (such as temperature, pressure, time, etc.) to form a production data set containing multiple dimensions of information; for example, the production data set of a certain refractory brick is represented as: {raw material mixing temperature: 120℃, raw material proportioning: high alumina 50kg / quartz sand 30kg / other additives 20kg, forming pressure: 20MPa, drying temperature: 100℃, drying time: 24 hours, firing temperature curve: [room temperature→1300℃→room temperature], color uniformity: good, texture clear}; through such real-time monitoring and data collection, the enterprise has a comprehensive understanding of the production process of the refractory bricks, providing strong support for subsequent quality control and anomaly detection.

[0062] Further, according to the detection of the production data set of the refractory bricks, the abnormal surfaces of the state images of each refractory brick are determined, and according to the tracing of each abnormal surface, the corresponding production data is determined, and the production data is determined as abnormal production data, and the abnormal production data is introduced.

[0063] At this time, the state image of the refractory brick is analyzed using image recognition technology and preset quality standards to identify abnormal surfaces such as cracks, color differences, deformations, and pollution. At this time, a model capable of identifying surface defects of the refractory brick is trained through a machine learning algorithm, and then the real-time captured state image is input into the model for detection. Quality problems on the surface of the refractory brick are discovered in a timely manner, providing a basis for subsequent anomaly tracing and quality improvement.

[0064] Once the abnormality on the surface of the refractory brick is detected, the relevant data of the refractory brick in the production process is immediately traced back. At this time, using a data management system or control center, the corresponding production data set is quickly located according to the identification information of the refractory brick (such as batch number, production time, production line position, etc.). The specific reason for the abnormal surface is determined, providing data support for subsequent quality control and production adjustment.

[0065] The traced production data is compared with the preset quality standards or historical data to determine which data is outside the normal range or has abnormalities. At this time, a reasonable threshold or rule is set to compare the production data one by one to identify abnormal values or data deviating from the standard. The specific content of the abnormal production data is determined to provide a clear improvement direction for subsequent quality improvement and production optimization.

[0066] Specifically, assuming that on the refractory brick production line, cracks are detected in a batch of refractory bricks through image recognition technology. At the discharge outlet, a high-definition camera captures the state images of a batch of refractory bricks. The image recognition model analyzes these images and finds that some of the refractory bricks have obvious cracks on their surfaces. According to the batch number and production time of the refractory bricks, the corresponding production data set of the batch of refractory bricks is quickly located. The production data set contains information on raw material ratio, molding pressure, drying temperature, and firing time in multiple dimensions.

[0067] The production data of this batch of refractory bricks is compared with historical data or preset quality standards. It is found that the molding pressure of this batch of refractory bricks is generally low, and the drying temperature is also slightly lower than the standard value. Therefore, it is determined that the molding pressure and drying temperature are the main reasons for the crack problem, and these data are marked as abnormal production data. Through these steps, the enterprise timely discovers the quality problems on the surface of the refractory brick and traces back to the specific production data, providing strong data support for subsequent quality improvement and production optimization. For example, adjust the pressure setting of the molding machine and increase the temperature of the drying kiln to reduce the occurrence of cracks.

[0068] Therefore, when the production process of the refractory brick is completed, the refractory brick is discharged through the discharge port of the refractory brick production line, and the discharge image of the refractory brick is collected by the corresponding camera, and the multiple abnormal features of the refractory brick are determined according to the matching of the discharge image of the refractory brick and the abnormal production data, and the matching of the discharge image of the refractory brick and the abnormal production data is considered as a whole, and the accuracy of the multiple abnormal features of the refractory brick is ensured.

[0069] At this time, a high-definition camera is installed at the end of the refractory brick production line, i.e. at the discharge port, to capture the instant image of the refractory brick being discharged from the production line; at this time, the camera should be set to continuous shooting or trigger shooting mode to ensure that each refractory brick can be clearly captured; the shooting angle and light conditions should be optimized to reduce the influence of shadows and reflections on image quality; the final appearance image of the refractory brick is obtained to provide a basis for subsequent quality detection and abnormal feature recognition.

[0070] The collected discharge image of the refractory brick is matched with the previously determined abnormal production data; the abnormal production data includes improper raw material ratio, insufficient forming pressure, too low drying temperature or insufficient firing time, etc.; at this time, using the data management system or special matching software, the image is associated with the corresponding production data according to the batch number, production time or other unique identifier of the refractory brick; determine which refractory bricks have quality problems related to abnormal production data to provide clues for subsequent feature recognition.

[0071] Combined with the discharge image and abnormal production data, the abnormal features of the refractory brick are identified using image recognition algorithms or manual inspection, including cracks, color differences, deformations, size deviations, etc.; at this time, for image recognition algorithms, abnormal features are detected automatically by trained models; for manual inspection, experienced quality inspectors need to make judgments based on images and data; accurately identify the abnormal features of the refractory brick to provide clear basis for subsequent quality control and product recall.

[0072] Reference Figure 4 In step S13, in the optimization process of the refractory brick, multiple optimization data are generated according to the optimization of the multiple abnormal features of the refractory brick, the optimized production data are determined based on the multiple optimization data and the corresponding abnormal production data, and an intelligent feedback mechanism is generated;

[0073] In the specific implementation process of the present application, the specific steps are as follows:

[0074] S131: When the refractory brick enters the optimization process of the refractory brick, multiple images of the refractory brick when entering the optimization process are collected, multiple pose parameters are determined according to the analysis of the multiple images, and the current pose of the refractory brick is determined according to the multiple pose parameters and the shape of the refractory brick;

[0075] S132: Determine the three-dimensional model of the firebrick according to the synthesis of the multiple images and the morphology of the firebrick, and arrange the three-dimensional model of the firebrick along the current posture of the firebrick; at this time, mark the corresponding multiple abnormal features on the three-dimensional model of the firebrick;

[0076] S133: Determine the corresponding optimization event according to the current posture of the firebrick and the spatial position of the multiple abnormal features, trigger the optimization of the multiple abnormal features, generate multiple optimization data, match the multiple optimization data, the multiple abnormal features and the corresponding abnormal production data, and determine the optimized production data in the matching process;

[0077] S134: Collect the negative feedback path between the optimization process and the production process, and determine the intelligent feedback mechanism based on the multiple training of the negative feedback path, each optimized production data and the production data set of the firebrick.

[0078] In the embodiment of the present application, when the firebrick enters the optimization process of the firebrick, multiple images of the firebrick when entering the optimization process are collected, multiple posture parameters are determined according to the analysis of the multiple images, and the current posture of the firebrick is determined according to the multiple posture parameters and the morphology of the firebrick, which ensures the accuracy of the current posture of the firebrick by considering the multiple posture parameters and the morphology of the firebrick as a whole.

[0079] At this time, multiple high-precision cameras are installed on the entrance or conveyor belt of the optimization process of the firebrick, to ensure that the complete image of the firebrick can be captured from different angles; at this time, the camera should be set to continuous shooting mode, and the shooting frequency should match the moving speed of the firebrick to ensure that each firebrick can be clearly captured; at the same time, the resolution and shooting angle of the camera should be optimized to reduce the influence of image distortion and shadow; multiple angle images of the firebrick when entering the optimization process are obtained, which provides basic data for subsequent posture parameter extraction and current posture determination.

[0080] Use image processing algorithms (such as edge detection, feature extraction, etc.) to analyze the collected multiple images, and extract feature information such as edge contour and key point position of the firebrick; at this time, first, pre-process the image, such as denoising, contrast enhancement, etc., to improve the accuracy of feature extraction; then, use feature matching algorithms (such as SIFT, SURF, etc.) to find the same feature points in multiple images, and calculate multiple posture parameters (such as rotation angle, inclination angle, displacement, etc.) of the firebrick according to the positional relationship of these feature points; through image analysis and feature extraction, multiple posture parameters of the firebrick in three-dimensional space are determined, which provides the basis for subsequent determination of the current posture.

[0081] In combination with the known shape of the refractory brick (such as cuboid, cylinder, etc.) and the extracted attitude parameters, the precise attitude of the refractory brick in the current space is determined by using a three-dimensional reconstruction algorithm or an attitude estimation algorithm; at this time, according to the shape characteristics of the refractory brick (such as aspect ratio, symmetry, etc.), the extracted attitude parameters are checked and adjusted to ensure the accuracy and consistency of the attitude; then, the three-dimensional model of the refractory brick is reconstructed according to the adjusted attitude parameters by using a three-dimensional modeling software or an attitude estimation model, so as to determine its current attitude; by combining the shape of the refractory brick and the extracted attitude parameters, the precise attitude of the refractory brick in the current space is determined, which provides accurate positioning information for subsequent optimization processing.

[0082] Further, the three-dimensional model of the refractory brick is determined according to the synthesis of the multiple images and the shape of the refractory brick, and the three-dimensional model of the refractory brick is arranged along the current attitude of the refractory brick; at this time, the corresponding multiple abnormal features are marked on the three-dimensional model of the refractory brick, which takes into account the overall consideration of the synthesis of the multiple images and the shape of the refractory brick, and ensures the accuracy of the three-dimensional model of the refractory brick.

[0083] At this time, the three-dimensional model of the refractory brick is synthesized by using the refractory brick images collected from multiple angles and the known shape of the refractory brick (such as cuboid, cylinder, etc.) through three-dimensional reconstruction technology; at this time, the collected images are preprocessed, including denoising, contrast enhancement, distortion correction, etc., to improve the image quality; then, the three-dimensional coordinates of the surface of the refractory brick are calculated by using stereo vision technology (such as structured light, binocular vision, etc.) or three-dimensional scanning technology in combination with feature point matching in the images; finally, according to these three-dimensional coordinates and the shape characteristics of the refractory brick, the three-dimensional model of the refractory brick is constructed by using three-dimensional modeling software (such as Blender, SolidWorks, etc.); by combining the images and the shape, a three-dimensional model accurately reflecting the actual shape and size of the refractory brick is generated, which provides a basis for subsequent analysis and processing.

[0084] The synthesized three-dimensional model of the refractory brick is arranged according to the previously determined current attitude (such as rotation angle, inclination angle, displacement, etc.), to ensure that the model is consistent with the position and orientation of the actual refractory brick in space; at this time, the three-dimensional model is transformed by using a three-dimensional transformation matrix (including a rotation matrix, a translation matrix, etc.) to match the current attitude of the refractory brick, which usually involves rotating, translating, etc. the model to ensure that the coordinate system of the model is consistent with the actual coordinate system of the refractory brick; the three-dimensional model can accurately reflect the position and orientation of the refractory brick in the actual production process, which provides accurate positioning information for subsequent analysis and processing.

[0085] According to the previously detected abnormal features of the refractory brick (such as cracks, color differences, deformations, etc.), mark the corresponding positions on the three-dimensional model; at this time, use the marking tool in the image processing or three-dimensional modeling software to map the feature points to the three-dimensional model according to the position information of the abnormal features in the image, and mark on the model; the marking adopts color, texture, highlight display, etc. to facilitate quick identification during subsequent analysis and processing; by marking the abnormal features on the three-dimensional model, intuitive visual guidance is provided for subsequent optimization processing, helping operators quickly locate and handle problems.

[0086] Further, according to the current posture of the refractory brick and the spatial positions of the multiple abnormal features, determine the corresponding optimization event and trigger the optimization of the multiple abnormal features to generate multiple optimization data, match the multiple optimization data, the multiple abnormal features and the corresponding abnormal production data, and determine the optimized production data in the matching process, which takes into account the current posture of the refractory brick and the spatial positions of the multiple abnormal features, ensuring the accuracy of the corresponding optimization event.

[0087] At this time, analyze the current posture of the refractory brick (such as rotation angle, inclination angle, displacement, etc.) and the specific positions of the multiple abnormal features (such as cracks, color differences, deformations, etc.) in three-dimensional space to determine the optimization operation or event that needs to be performed; at this time, use three-dimensional modeling software or a special optimization analysis tool to visually analyze the three-dimensional model of the refractory brick and the abnormal features; according to the type, position and severity of the abnormal features, combined with the material, production process and other factors of the refractory brick, formulate the corresponding optimization strategy; for example, for cracks, polishing or repair is needed; for color difference, the raw material ratio or firing temperature needs to be adjusted; ensure that the optimization operation can effectively improve the quality and production efficiency of the refractory brick for specific problems.

[0088] According to the determined optimization event, trigger the corresponding optimization equipment or tool to optimize the abnormal features on the refractory brick; at this time, use automatic control system or manual operation to position the optimization equipment or tool to the position of the abnormal features and operate according to the predetermined optimization strategy; for example, use a polisher to polish cracks and use a spraying device to repair color difference areas; through optimization processing, eliminate or reduce the abnormal features on the refractory brick to improve its overall quality and appearance.

[0089] During the optimization process, record relevant optimization data such as optimization time, optimization intensity, materials or tools used, etc.; at this time, use sensors, data recorders and other devices to collect various data in the optimization process in real time, which are stored in the database for subsequent analysis and query; provide data support for the subsequent matching process, and help evaluate the optimization effect and improve the optimization strategy.

[0090] The generated optimization data, multiple abnormal features and corresponding abnormal production data are matched, and the relationship and mutual influence therebetween are analyzed; according to the matching result, the production data (such as raw material ratio, molding pressure, firing temperature, etc.) is adjusted and optimized to improve the quality and production efficiency of the refractory brick; at this time, the data analysis software or algorithm is used to perform statistical analysis, trend prediction and other operations on the matched data; according to the analysis result, an improvement scheme is formulated, and the production data is updated; through data matching and analysis, the continuous improvement and optimization of the production process are realized, and the overall quality and market competitiveness of the refractory brick are improved.

[0091] Therefore, the negative feedback path between the optimization process and the production process is collected, the intelligent feedback mechanism is determined based on the multiple training of the negative feedback path, the optimized production data and the production data set of the refractory brick, the overall consideration of the multiple training of the negative feedback path, the optimized production data and the production data set of the refractory brick is compatible, the accuracy of the intelligent feedback mechanism is guaranteed, at the same time, the multiple optimization data is fully utilized, and the multiple optimization data and the abnormal production data are compared, the accuracy of the optimized production data is guaranteed, and the intelligent feedback of the optimized production data is realized based on the intelligent feedback mechanism, so as to reduce the defective rate of the next refractory brick in the production process.

[0092] At this time, the feedback information from the optimization process to the production process is collected, including the performance of the optimized refractory brick in the production process (such as quality stability, whether the abnormal feature is reduced, etc.) and the change of the production data (such as the adjustment of the raw material ratio, the molding pressure, the firing temperature, etc.); at this time, monitoring points are set on the production line, and sensors and data acquisition systems are used to record the key parameters and indicators in the production process in real time; at the same time, a feedback mechanism is established to associate and compare the results of the optimization process and the feedback information of the production process; the effect of the optimization measures in actual production is understood, and data support is provided for the subsequent intelligent feedback mechanism.

[0093] The collected negative feedback path data, optimized production data and production data set of the refractory brick are trained by using machine learning or deep learning algorithm, so as to construct an intelligent feedback mechanism which can automatically identify abnormal features, predict optimization demand and adjust production data; at this time, the data is preprocessed, including data cleaning, feature extraction and label annotation, etc.; then, a suitable machine learning model (such as decision tree, support vector machine, neural network, etc.) or deep learning framework (such as TensorFlow, PyTorch, etc.) is selected, and the processed data is input into the model for training; in the training process, the parameters and structure of the model are constantly adjusted to improve its prediction and optimization ability; through training, the intelligent feedback mechanism can accurately identify the abnormal features of the refractory brick, predict the optimization demand, and automatically adjust the production data, so as to realize the intelligent control of the production process.

[0094] After the training is completed, the intelligent feedback mechanism is verified and tested to ensure that it can accurately identify abnormal features, predict optimization needs, and automatically adjust production data in actual application; at the same time, according to the results of verification and testing, the intelligent feedback mechanism is further optimized and adjusted; at this time, the intelligent feedback mechanism is tested using the verification data set to evaluate its prediction accuracy, optimization effect, and stability; according to the test results, the parameters, structure or algorithm of the model are adjusted to improve its performance; ensure that the intelligent feedback mechanism can work stably and accurately in actual application, and improve the production quality and efficiency of refractory bricks.

[0095] Specifically, assuming that the crack and color difference abnormal features have been optimized on the refractory brick production line, and the negative feedback path data between the optimization process and the production process has been collected; sensors and data collection systems are set up on the production line to record key parameters such as raw material ratio, molding pressure, and firing temperature in real time; at the same time, the quality of the optimized refractory bricks is detected to record feedback information such as quality stability and whether the abnormal features are reduced; for example, 100 batches of refractory brick production data are recorded, including raw material ratio (A material: B material: C material = X: Y: Z), molding pressure P, and firing temperature T; at the same time, the quality detection results of each batch of refractory bricks are recorded, including the number of cracks and the degree of color difference.

[0096] The collected negative feedback path data, optimized production data, and refractory brick production data are preprocessed, key features are extracted, and labels are labeled; select a suitable machine learning model (such as a neural network) and input the processed data into the model for training; during the training process, the parameters and structure of the model are constantly adjusted to improve its prediction and optimization capabilities; for example, a three-layer neural network model is selected, the input layer contains features such as raw material ratio, molding pressure, and firing temperature, and the output layer is the quality detection results of the refractory bricks (such as the number of cracks and the degree of color difference); through training, the model can learn the mapping relationship between these features and the quality detection results; after the training is completed, the intelligent feedback mechanism is tested using the verification data set to evaluate its prediction accuracy, optimization effect, and stability; according to the test results, the parameters, structure or algorithm of the model are adjusted to improve its performance; for example, it is found that the prediction accuracy of the model fluctuates greatly within a certain batch range, so the parameters of the model are fine-tuned to improve its stability and prediction accuracy.

[0097] Finally, an intelligent feedback mechanism that can automatically identify abnormal features, predict optimization needs, and automatically adjust production data is determined; in actual application, this mechanism can automatically adjust parameters such as raw material ratio, molding pressure, and firing temperature according to real-time monitoring of production data and feedback information to improve the production quality and efficiency of refractory bricks.

[0098] Reference Figure 5 In step S14, in the detection process of the refractory brick, intelligent detection of the plurality of detection heads relative to the refractory brick is triggered according to the plurality of optimization data and the morphology of the refractory brick, and an intelligent detection mechanism is generated;

[0099] In the specific implementation of the present application, the specific steps are:

[0100] S141: When the refractory brick enters the detection process of the refractory brick, the optimized surface of the refractory brick is determined according to the analysis of the plurality of optimization data, and the optimization distribution map of the refractory brick is determined according to the position of the optimized surface and the morphology of the refractory brick;

[0101] S142: Mark the corresponding optimization event in the optimization distribution map of the refractory brick, and determine the detection order of the optimized surface according to each optimization event, the corresponding optimized surface and the morphology of the refractory brick, and trigger intelligent detection of the plurality of detection heads relative to the refractory brick according to the detection order of the optimized surface;

[0102] S143: Collect the positive feedback path between the optimization process and the detection process, and determine the intelligent detection mechanism based on the positive feedback path, the multiple training of each optimized surface and the plurality of detection heads.

[0103] In the embodiment of the present application, when the refractory brick enters the detection process of the refractory brick, the optimized surface of the refractory brick is determined according to the analysis of the plurality of optimization data, and the optimization distribution map of the refractory brick is determined according to the position of the optimized surface and the morphology of the refractory brick, which is compatible with the overall consideration of the position of the optimized surface and the morphology of the refractory brick, and ensures the accuracy of the optimization distribution map of the refractory brick.

[0104] At this time, in the process of the refractory brick production line, when the refractory bricks complete all optimization processes, they will be transported to the detection process, which marks the transition of the refractory bricks from the optimization stage to the quality verification stage; at the same time, before the detection process starts, all data related to the optimization of the refractory bricks need to be collected and analyzed, including but not limited to: records of the optimization process (such as polishing time, types of repair materials, heat treatment temperature curve, etc.), quality detection results (such as preliminary size measurement, surface hardness test, etc.), and any sensor data directly related to the optimization of the refractory bricks (such as temperature sensors, pressure sensors, etc.); At this time, these data are usually stored in databases or data files and need to be extracted and sorted through specific data analysis software or algorithms; during the analysis process, the accuracy, integrity and consistency of the data need to be ensured.

[0105] Based on the analysis of the optimization data, identify the surfaces on the refractory bricks that have been optimized, which usually involves in-depth analysis and comparison of the data to determine which surfaces have been subjected to specific optimization measures; At this time, image processing techniques (such as edge detection, texture analysis) or machine learning algorithms (such as classifiers) are used to assist in identifying optimized surfaces; In addition, according to the records of the optimization process, the optimized surfaces are directly marked.

[0106] Based on the determination of the optimized surface, combined with the morphology of the refractory brick (such as size, shape, weight, etc.), draw the optimization distribution map of the refractory brick, which should clearly show which surfaces are optimized and the type and degree of optimization; At this time, use CAD (Computer Aided Design) software or similar drawing tools to create the optimization distribution map; In the map, the optimized surfaces are represented by different colors, lines or symbols to distinguish different optimization types and degrees; At the same time, add annotations or labels to provide more detailed information.

[0107] Further, mark the corresponding optimization events in the optimization distribution map of the refractory brick, and determine the detection order of the optimized surface according to each optimization event, the corresponding optimized surface and the morphology of the refractory brick, and trigger the intelligent detection of the multiple detection heads relative to the refractory brick according to the detection order of the optimized surface, which is compatible with the overall consideration of each optimization event, the corresponding optimized surface and the morphology of the refractory brick, ensuring the accuracy of the detection order of the optimized surface.

[0108] At this time, according to the optimization distribution map of the refractory brick, mark the specific optimization events that occur on each optimized surface, which include but are not limited to: polishing, repairing, coating, heat treatment, etc.; At this time, use different symbols, colors or text annotations to mark different optimization events on the optimization distribution map; For example, use a circle to represent the polishing event and use an arrow to point to the corresponding optimized surface; Use a square to represent the coating event and label the type and color of the coating next to it.

[0109] Based on the optimization events, the location of the optimized surface and the morphology of the refractory brick, a reasonable detection order is developed, which should take into account the efficiency, accuracy of detection and accessibility of the detection head; At this time, use priority sorting algorithm or expert system to determine the detection order; For example, first detect those optimization surfaces that have the greatest impact on the performance of the refractory brick (such as surfaces that bear high temperature and wear), and then detect other surfaces; At the same time, the layout and movement path of the detection head should also be considered to ensure efficient coverage of all optimized surfaces.

[0110] According to the determined detection sequence, control multiple detection heads (such as visual sensors, laser range finders, infrared thermographs, etc.) to perform intelligent detection on the refractory bricks. These detection heads should be able to automatically adjust their positions, angles, and detection parameters to adapt to different shapes and optimized surfaces of the refractory bricks. At this time, use an automated control system or robotic technology to achieve intelligent control of the detection heads. For example, use machine vision algorithms to guide the detection heads to position at precise locations of the optimized surfaces. Use laser ranging technology to measure the flatness of the surfaces. Use infrared thermographs to detect the temperature distribution of the surfaces. These detection data are collected and analyzed in real time to evaluate the quality of the refractory bricks and the optimization effect.

[0111] Specifically, assume that a refractory brick for a glass furnace is being produced, which has an irregular polyhedral shape with multiple optimized surfaces. During production, the following optimization events have been performed on these surfaces: Top surface: Sprayed with a wear-resistant coating; Side A: High-temperature heat treatment to improve strength; Side B: Repair to fix defects during production; Bottom surface: Polished with anti-slip texture; Now, intelligent detection needs to be triggered according to these optimization events.

[0112] Use CAD software to draw an optimization distribution map of the refractory brick, and mark the optimization events on each optimized surface in the map. For example, mark the wear-resistant coating event on the top surface with a circle and point to the surface with an arrow; mark the high-temperature heat treatment event on side A with a square and label "HT" next to it to indicate high temperature; mark the repair event on side B with a triangle and indicate the repair area with a dashed line; mark the polishing event of the anti-slip texture on the bottom surface with a wavy line.

[0113] According to the importance of the optimization events and the accessibility of the detection heads, the following detection sequence is developed: First, detect the wear-resistant coating on the top surface (because the coating is crucial to the wear resistance of the refractory brick), then detect the high-temperature heat treatment effect on side A (because side A is exposed to high temperature for a long time), then detect the repair quality on side B (because there are defects in the repair area), and finally detect the anti-slip texture on the bottom surface (because the bottom surface mainly affects the stability of the refractory brick when placed).

[0114] The automation control system is used to control multiple detection heads for intelligent detection of refractory bricks. For example, a machine vision algorithm is used to guide a visual sensor to position to the wear-resistant coating position on the top surface, and image processing technology is used to evaluate the uniformity and thickness of the coating. A laser range finder is used to measure the dimensional change and flatness of side A after high-temperature heat treatment. An infrared thermal imager is used to detect the temperature distribution of the repaired area on side B to evaluate the thermal stability of the repair material. A contact-type measuring instrument is used to detect the depth and consistency of the anti-slip texture on the bottom surface. These detection data are collected in real time and input into quality analysis software to evaluate the quality of the refractory bricks and optimize the effect. Through this process, each optimized surface is properly detected to verify the effectiveness of the optimization event and provide important information for subsequent quality control and improvement.

[0115] Therefore, the positive feedback path between the optimization process and the detection process is collected, and the intelligent detection mechanism is determined based on the positive feedback path, the multiple training of each optimized surface and the multiple detection heads, which is compatible with the overall consideration of the positive feedback path, the multiple training of each optimized surface and the multiple detection heads, and ensures the accuracy of the intelligent detection mechanism.

[0116] At this time, the positive feedback path refers to a closed-loop feedback system from the optimization process to the detection process and back to the optimization process. In this system, the results of the detection process are used to evaluate the effectiveness of the optimization process and provide improvement direction for subsequent optimization. Therefore, all data and information generated during this process need to be collected, including the parameters of the optimization process, the results of the detection process and their associated information. At this time, these data are collected and stored through databases, data warehouses or data stream processing platforms. At the same time, the accuracy and completeness of the data need to be ensured for subsequent analysis and training.

[0117] Using the collected positive feedback path data, the intelligent detection mechanism is trained, which involves multiple aspects, including the selection of detection heads, the setting of detection parameters, the interpretation of detection results and the adjustment of optimization measures, etc. At this time, machine learning, deep learning and other algorithms are used to train the intelligent detection mechanism. First, the training objectives and indicators need to be defined, such as detection accuracy, missed detection rate, false detection rate, etc. Then, according to the positive feedback path data, a training data set is constructed, and a suitable training algorithm and model are selected. During the training process, the parameters and structure of the model need to be constantly adjusted to improve its generalization ability and accuracy.

[0118] After multiple training, a stable, accurate and efficient intelligent detection mechanism is determined, which should be able to automatically select appropriate detection heads and detection parameters according to the morphology of the refractory brick, the optimized surface and the detection requirements, and output accurate detection results; at this time, the accuracy and stability of the intelligent detection mechanism are verified by cross-validation, test set evaluation and other methods; at the same time, according to the actual needs, the intelligent detection mechanism is further optimized and adjusted to improve its adaptability and flexibility.

[0119] Specifically, suppose a refractory brick for steel smelting is being produced, which is a cuboid with multiple optimized surfaces; In the production process, various optimization measures are adopted, such as wear-resistant coating spraying, high-temperature heat treatment, repair, etc.; In order to verify the effect of these optimization measures, relevant data need to be collected in the detection process, and the intelligent detection mechanism is determined based on these data.

[0120] First, a database is established to store all data generated by the optimization process and the detection process; The data of the optimization process includes the spraying parameters of the wear-resistant coating (such as coating thickness, spraying speed, etc.), the temperature curve and time parameters of the high-temperature heat treatment, the types and amounts of repair materials, etc.; The data of the detection process includes image data of visual sensors, size measurement data of laser range finders, temperature distribution data of infrared thermographs, etc.; At the same time, the batch number, production time, detection time and other related information of each refractory brick are also recorded for subsequent data analysis and training.

[0121] Support vector machine (SVM) in machine learning algorithm is selected to train the intelligent detection mechanism; First, the training data set is constructed according to the positive feedback path data, including the parameters of the optimization process and the results of the detection process; Then, the SVM model is trained, and its parameters (such as penalty parameter C, kernel function parameter γ, etc.) are constantly adjusted to improve the accuracy and generalization ability of the model; In the training process, cross-validation method is also used to evaluate the performance of the model, and the optimal model parameters are selected; After multiple training, a stable, accurate and efficient intelligent detection mechanism is determined, which can automatically select appropriate detection heads and detection parameters according to the morphology of the refractory brick, the optimized surface and the detection requirements; For example, when the uniformity and thickness of the wear-resistant coating need to be detected, the mechanism will automatically select the visual sensor and set the corresponding detection parameters; When the size change needs to be measured, the mechanism will select the laser range finder and adjust its measurement range; At the same time, the intelligent detection mechanism can also output accurate detection results, and give suggestions or adjustment directions for optimization measures according to the detection results; Through this process, the intelligent detection mechanism is successfully determined, which provides strong support for the quality control and optimization of refractory bricks.

[0122] Reference Figure 6In step S15, the intelligent control mechanism of the firebrick production line is determined according to the training of the intelligent detection mechanism and the intelligent feedback mechanism, and the intelligent control mechanism of the firebrick production line intelligently controls the production process, the optimization process and the detection process of the firebrick.

[0123] In the specific implementation of the present application, the specific steps are:

[0124] S151: Collect the intelligent detection mechanism and the intelligent feedback mechanism, connect the intelligent detection mechanism and the intelligent feedback mechanism based on the optimization process of the firebrick, and determine the intelligent control mechanism of the firebrick production line according to the training of the optimization process of the firebrick, the intelligent detection mechanism and the intelligent feedback mechanism;

[0125] S152: Configure a negative feedback path in the intelligent feedback mechanism and a positive feedback path in the intelligent detection mechanism, and form the overall control path of the firebrick production line based on the compatibility of the intelligent control mechanism of the firebrick production line with the negative feedback path and the positive feedback path.

[0126] S153: Determine a plurality of intelligent control nodes according to the division of the overall control path, and the plurality of intelligent control nodes are respectively distributed in the production process, the optimization process and the detection process of the firebrick.

[0127] In the embodiment of the present application, the intelligent detection mechanism and the intelligent feedback mechanism are collected, the intelligent detection mechanism and the intelligent feedback mechanism are connected based on the optimization process of the firebrick, and the intelligent control mechanism of the firebrick production line is determined according to the training of the optimization process of the firebrick, the intelligent detection mechanism and the intelligent feedback mechanism. The overall consideration of the training of the optimization process of the firebrick, the intelligent detection mechanism and the intelligent feedback mechanism is compatible, and the accuracy of the intelligent control mechanism of the firebrick production line is ensured.

[0128] At this time, the intelligent detection mechanism refers to a series of automatic equipment and algorithms for real-time monitoring of key parameters and quality indicators in the production process of the firebrick, which includes but is not limited to size, shape, surface quality, material composition, etc. At this time, through the deployment of sensors, visual detection systems, laser range finders and other equipment on the production line, the data of the firebrick at each production stage is collected in real time. At the same time, data analysis software is used to preprocess and clean the data to ensure the accuracy and integrity of the data.

[0129] The intelligent feedback mechanism refers to a mechanism for automatically adjusting production parameters or taking corrective measures based on the data provided by the intelligent detection mechanism, which includes the state of the production equipment, the effect evaluation of the optimization process, etc. At this time, by connecting the control system of the production equipment and the intelligent detection mechanism, feedback information in the production process is obtained in real time. At the same time, machine learning algorithms are used to analyze and predict the feedback information to determine the production parameters that need to be adjusted or the corrective measures that need to be taken.

[0130] Intelligent detection mechanisms and intelligent feedback mechanisms are combined with the optimization process to form a closed-loop control system that can monitor the quality of refractory bricks in real time and automatically adjust the parameters of the optimization process based on feedback results. At this point, first, the optimization process of refractory bricks and key quality indicators are determined. Then, intelligent detection mechanisms are deployed at key locations in the optimization process to monitor quality indicators in real time. Next, intelligent feedback mechanisms are connected to detection mechanisms to automatically adjust the parameters of the optimization process based on detection results. Finally, the effectiveness and stability of the closed-loop control system are verified through simulation and testing. The data collected by intelligent detection mechanisms and intelligent feedback mechanisms are used to train and optimize the intelligent control mechanism, which aims to improve the accuracy and robustness of the control system. Machine learning algorithms such as supervised learning and reinforcement learning are used to train the intelligent control mechanism. During the training process, historical data is used as a training set to optimize the performance of the control system by continuously adjusting model parameters. Cross-validation and test set evaluation methods are used to verify the accuracy and generalization ability of the model. After training, the optimized intelligent control mechanism is deployed on the production line for practical application.

[0131] Further, a negative feedback path is configured in the intelligent feedback mechanism, and a positive feedback path is configured in the intelligent detection mechanism. The intelligent control mechanism of the refractory brick production line is compatible with the negative feedback path and the positive feedback path, and forms the overall control path of the refractory brick production line.

[0132] At this point, the negative feedback path refers to a series of automatic adjustment measures triggered by the intelligent feedback mechanism when the intelligent detection mechanism finds that the quality or production parameters of refractory bricks deviate from the preset standard, in order to correct the deviation and restore the stable state of the production line. At this point, first, determine which production parameters or quality indicators need to be monitored and set the corresponding threshold. When the intelligent detection mechanism detects that these parameters or indicators exceed the threshold, the intelligent feedback mechanism will start the negative feedback path. Specific measures include adjusting the parameters of production equipment, suspending production for manual inspection or adjusting the optimization process, etc.

[0133] The positive feedback path refers to the automatic confirmation and reinforcement of the current production state by the intelligent feedback mechanism when the intelligent detection mechanism confirms that the quality or production parameters of refractory bricks meet the preset standard, in order to maintain efficient and stable production. At this point, similarly, it is necessary to determine which production parameters or quality indicators meet the standard and set the corresponding confirmation conditions. When the intelligent detection mechanism detects that these parameters or indicators meet the confirmation conditions, the intelligent feedback mechanism will start the positive feedback path. Specific measures include confirming the effectiveness of the current production parameters, continuing production, or recording successful experiences for future reference.

[0134] The intelligent control mechanism needs to be able to handle the information of both negative feedback and positive feedback paths simultaneously and make reasonable decisions and adjustments based on these information; at this time, in the design of the intelligent control mechanism, it needs to ensure that it can receive real-time data from the intelligent detection mechanism and the intelligent feedback mechanism, and judge whether the current production state needs to be adjusted according to these data; at the same time, the intelligent control mechanism also needs to have the ability to handle multiple feedback paths to ensure that correct decisions can be made in different situations.

[0135] The combination of negative feedback path, positive feedback path and intelligent control mechanism forms the overall control path of the refractory brick production line, which will run through the entire production process from raw material preparation to finished product delivery, ensuring that each link can be effectively monitored and adjusted; at this time, by integrating various sensors, actuators, control systems and data analysis software, a complete automatic control system is constructed, which can monitor the state of the production line in real time and automatically adjust the production parameters or take corrective measures according to the decision of the intelligent control mechanism; at the same time, the system also needs to have the ability of self-learning and optimization to continuously improve the production efficiency and product quality.

[0136] Specifically, suppose a high-performance refractory brick is being produced, and its production process includes raw material proportioning, molding, sintering and quality inspection, etc. In order to form the overall control path of the refractory brick production line, in the raw material proportioning link, an automatic weighing system is set up to monitor the input amount of various raw materials; when the system detects that the input amount of a certain raw material exceeds the preset range, it will automatically adjust the parameters of the feeding machine to restore the correct proportioning.

[0137] In the molding link, a visual detection system is deployed to monitor the shape and size of the refractory brick; when the system detects that the shape or size does not meet the standard, it will automatically stop the molding machine and issue an alarm for the operator to make manual adjustments; in the sintering link, temperature sensors and infrared thermometers are used to monitor the temperature and temperature distribution in the sintering furnace; when the system detects that the temperature exceeds the preset range or the temperature distribution is uneven, it will automatically adjust the parameters of the sintering furnace to optimize the sintering effect.

[0138] In the quality inspection link, laser range finders and X-ray detection equipment are used to detect the dimensional accuracy and internal defects of the refractory brick; when the system detects that these parameters meet the standard, it will automatically confirm the quality of the current production batch and allow the production of the next batch to continue; at the same time, a production data statistics system is set up to record the successful experience and failure lessons of each production batch; when the system detects that the quality of a certain production batch is stable and the efficiency is high, it will automatically record these successful experiences and use them as a reference in subsequent production.

[0139] An intelligent control mechanism based on machine learning is designed, which can receive real-time data from various production links and determine whether the current production state needs to be adjusted according to these data; when the intelligent control mechanism detects negative feedback in a certain production link, it will automatically trigger the corresponding adjustment measures and monitor the effect after adjustment; if the adjustment is effective, continue production; if not, further adjustment or alarm; when the intelligent control mechanism detects positive feedback in a certain production link, it will confirm the effectiveness of the current production parameters and record successful experience for subsequent reference; at the same time, it will also continuously optimize the production process according to these successful experiences to improve production efficiency and product quality; at the same time, by integrating sensors, actuators, control systems and data analysis software of each production link, a complete automation control system is constructed, which can monitor the state of the production line in real time and automatically adjust production parameters or take corrective measures according to the decision of the intelligent control mechanism; at the same time, the system also has the ability of self-learning and optimization, which can continuously adapt to changes and challenges in the production process; through this process, the negative feedback path and the positive feedback path are successfully configured, and the overall control path of the refractory brick production line is formed, which not only improves the production efficiency and product quality, but also reduces the production cost and the frequency of manual intervention, providing strong support for the sustainable development of the enterprise.

[0140] Therefore, according to the division of the overall control path, a plurality of intelligent control nodes are determined, which are distributed in the production process, optimization process and detection process of the refractory brick, compatible with the overall consideration of the intelligent detection mechanism and the intelligent feedback mechanism, ensuring the accuracy of the intelligent control mechanism of the refractory brick production line, and intelligently controlling the production process, optimization process and detection process.

[0141] At this time, before this step, the overall control path of the refractory brick production line has been established, which includes the negative feedback path and the positive feedback path, as well as their integration with the intelligent control mechanism; the division of the overall control path is based on the logical sequence of the production process and the quality control points.

[0142] According to the division of the overall control path, it is necessary to determine how many intelligent control nodes are needed in the production process, optimization process and detection process, and the number and location of these nodes should be able to fully cover the key quality control points in the production process and ensure the stability and controllability of the production process; the number of intelligent control nodes depends on the complexity of the production process and the quality control requirements; too many nodes will make the system too complex and difficult to maintain, while too few nodes will not provide enough information to control the production process.

[0143] Depending on the location of the intelligent control nodes and the required functionality, appropriate intelligent control technologies and devices need to be selected, which include sensors, actuators, controllers, data analysis software, etc.; for example, in the production process, temperature sensors, pressure sensors, and flow sensors are needed to monitor the processing of raw materials; in the optimization process, machine vision systems are needed to monitor and adjust the uniformity and thickness of the coating; in the detection process, X-ray detection equipment is needed to detect internal defects of refractory bricks.

[0144] Once the number and location of intelligent control nodes are determined and appropriate intelligent control technologies and devices are selected, the deployment and integration of these nodes begin, which usually involves steps such as installing sensors and actuators, configuring control systems and data analysis software, etc.; during the deployment process, it is necessary to ensure that each node can effectively communicate and exchange data with other nodes and intelligent control mechanisms in the overall control path; after the deployment is completed, each intelligent control node needs to be tested to ensure that they can correctly monitor and control the production process, which includes steps such as simulating production conditions, verifying data accuracy, adjusting control parameters, etc.; based on the test results, the intelligent control nodes need to be optimized to improve their accuracy and reliability, which involves adjusting sensor positions, improving data analysis algorithms, optimizing control strategies, etc.

[0145] Specifically, suppose a high-performance refractory brick is being produced, and its production process includes raw material preparation, molding, sintering, and quality inspection, etc. multiple processes; in order to determine and deploy intelligent control nodes, the following steps are taken: the overall control path of the refractory brick production line has been established, which includes monitoring and controlling the raw material preparation, molding, sintering, and quality inspection processes, etc. The logical relationship and quality control points between these processes have been clearly defined.

[0146] In the raw material preparation process, two intelligent control nodes are determined: one for monitoring the proportioning and mixing process of raw materials, and the other for monitoring the temperature and humidity of raw materials; in the molding process, one intelligent control node is determined to monitor and adjust the pressure and speed of the molding machine to ensure that the shape and size of the refractory brick meet the standards; in the sintering process, two intelligent control nodes are determined: one for monitoring the temperature and temperature distribution of the sintering furnace, and the other for monitoring the gas composition and flow during the sintering process; in the quality inspection process, one intelligent control node is determined to detect internal defects of refractory bricks using X-ray detection equipment.

[0147] In the raw material preparation process, temperature sensors, humidity sensors and automatic weighing systems are selected to monitor the proportioning, mixing process and temperature and humidity of raw materials; in the forming process, pressure sensors and speed sensors are selected to monitor the pressure and speed of the forming machine, and an automatic adjustment system is equipped to adjust these parameters; in the sintering process, temperature sensors, infrared thermal imagers and gas analyzers are selected to monitor the temperature, temperature distribution and gas composition and flow of the sintering furnace; in the quality inspection process, X-ray detection equipment is selected to detect the internal defects of the refractory bricks, and image analysis software is equipped to automatically identify and classify defects.

[0148] According to the determined positions and quantities, sensors and actuators are installed on the production line, and corresponding control systems and data analysis software are configured; it is ensured that each intelligent control node can effectively communicate and exchange data with other nodes and intelligent control mechanisms in the overall control path;

[0149] Each intelligent control node is tested to verify its accuracy and reliability; according to the test results, the positions and parameters of some nodes are adjusted to improve their monitoring and control effect; the intelligent control system of the entire production line is also optimized to improve its overall performance and stability; through this process, multiple intelligent control nodes are successfully determined and deployed in the production process, optimization process and detection process of the refractory bricks, which can effectively monitor and control the production process, improve production efficiency and product quality, and reduce production cost and the frequency of manual intervention.

[0150] Please refer to Figure 7 , Figure 7 is a structural composition diagram of the intelligent control system of the refractory brick production line in the embodiment of the application; the intelligent control system of the refractory brick production line comprises:

[0151] a process module 21 for determining the production process, optimization process and detection process of the refractory bricks according to the refractory brick production line;

[0152] an abnormal feature module 22 for determining abnormal production data according to a production data set of the refractory bricks in the production process of the refractory bricks, and determining multiple abnormal features of the refractory bricks according to the abnormal production data and an outfeed image of the refractory bricks;

[0153] an intelligent feedback module 23 for generating multiple optimization data according to optimization of the multiple abnormal features of the refractory bricks in the optimization process of the refractory bricks, determining optimized production data based on the multiple optimization data and corresponding abnormal production data, and generating an intelligent feedback mechanism;

[0154] The intelligent detection module 24 is configured to trigger intelligent detection of the detection heads relative to the refractory bricks according to the optimized data and the shape of the refractory bricks in the detection process of the refractory bricks, and generate an intelligent detection mechanism.

[0155] The intelligent control module 25 is configured to determine an intelligent control mechanism of the refractory brick production line according to the training of the intelligent detection mechanism and the intelligent feedback mechanism, and the intelligent control mechanism of the refractory brick production line intelligently controls the production process, the optimization process and the detection process of the refractory bricks.

[0156] Any combination of the technical features of the above embodiments is possible. In order to make the description concise, not all combinations of the technical features in the above embodiments are described, however, as long as the combinations of the technical features do not exist, they should be considered as the scope of the present disclosure.

Claims

1. An intelligent control method for a refractory brick production line, characterized in that, include: Determine the production process, optimization process, and testing process for refractory bricks based on the refractory brick production line; In the production process of refractory bricks, abnormal production data is determined based on the production data set of refractory bricks, and multiple abnormal characteristics of refractory bricks are determined based on the abnormal production data and the discharge images of refractory bricks. In the optimization process of refractory bricks, multiple optimization data are generated based on the optimization of multiple abnormal characteristics of refractory bricks. Based on the multiple optimization data and the corresponding abnormal production data, the optimized production data is determined and an intelligent feedback mechanism is generated. In the refractory brick inspection process, multiple inspection heads are triggered to perform intelligent inspection relative to the refractory brick based on multiple optimization data and the morphology of the refractory brick, and an intelligent inspection mechanism is generated. This includes: when the refractory brick enters the inspection process, determining the optimized surface of the refractory brick based on the analysis of multiple optimization data, and determining the optimized distribution map of the refractory brick based on the position of the optimized surface and the morphology of the refractory brick; marking the corresponding optimization events in the optimized distribution map of the refractory brick, and determining the inspection order of the optimized surfaces based on each optimization event, the corresponding optimized surface, and the morphology of the refractory brick, and triggering intelligent inspection of the refractory brick relative to the refractory brick based on the inspection order of the optimized surfaces; collecting the positive feedback path between the optimization process and the inspection process, and determining the intelligent inspection mechanism based on the positive feedback path, each optimized surface, and multiple inspection heads through multiple training. The intelligent control mechanism of the refractory brick production line is determined based on the training of the intelligent detection mechanism and the intelligent feedback mechanism. This intelligent control mechanism of the refractory brick production line intelligently controls the production process, optimization process and detection process of refractory bricks.

2. The intelligent control method for a refractory brick production line according to claim 1, characterized in that, The process of determining the production process, optimization process, and testing process of refractory bricks based on the refractory brick production line includes: Collect the distribution map of the refractory brick production line, and generate multiple functional areas based on the division of the distribution map of the refractory brick production line; In each functional area, the functional type of the functional area is determined based on the previous data and previous processing events of the functional area, and the production process, optimization process and testing process of refractory bricks are determined according to the location and functional type of multiple functional areas. In a refractory brick production line, the optimization process is located between the production process and the testing process. The optimization process forms a negative feedback path with the production process and a positive feedback path with the testing process. The negative and positive feedback paths are executed synchronously with the response of the optimization process.

3. The intelligent control method for a refractory brick production line according to claim 1, characterized in that, In the refractory brick production process, abnormal production data is determined based on the refractory brick production data set, and multiple abnormal characteristics of the refractory bricks are determined based on the abnormal production data and the refractory brick discharge images, including: The production process of refractory bricks is monitored in real time. Refractory bricks are produced step by step in the production process, and multiple production data of refractory bricks are output. The production data set of refractory bricks is determined based on multiple production data and the corresponding state images of refractory bricks. Based on the detection of the production data set of refractory bricks, the abnormal surfaces of the state images of each refractory brick are identified, and the corresponding production data is determined by tracing each abnormal surface, and the production data is identified as abnormal production data. During the production process of refractory bricks, the refractory bricks are discharged through the discharge port of the refractory brick production line, and the corresponding camera captures the discharge image of the refractory bricks. Based on the matching of the discharge image of the refractory bricks and abnormal production data, multiple abnormal characteristics of the refractory bricks are determined.

4. The intelligent control method for a refractory brick production line according to claim 1, characterized in that, In the refractory brick optimization process, multiple optimization data are generated based on the optimization of multiple abnormal characteristics of the refractory brick. Based on these multiple optimization data and corresponding abnormal production data, optimized production data is determined, and an intelligent feedback mechanism is generated, including: When the refractory bricks enter the optimization process, multiple images of the refractory bricks are collected. Multiple posture parameters are determined based on the analysis of the multiple images, and the current posture of the refractory bricks is determined based on the multiple posture parameters and the shape of the refractory bricks. A three-dimensional model of the refractory brick is determined by synthesizing multiple images and the morphology of the refractory brick, and the three-dimensional model of the refractory brick is arranged along the current posture of the refractory brick; at this time, multiple abnormal features corresponding to the three-dimensional model of the refractory brick are marked.

5. The intelligent control method for a refractory brick production line according to claim 4, characterized in that, In the refractory brick optimization process, multiple optimization data are generated based on the optimization of multiple abnormal characteristics of the refractory brick. Based on these multiple optimization data and corresponding abnormal production data, optimized production data is determined, and an intelligent feedback mechanism is generated. The process also includes: Based on the current posture of the refractory brick and the spatial location of multiple abnormal features, the corresponding optimization event is determined, and the optimization of multiple abnormal features is triggered to generate multiple optimization data. The multiple optimization data, multiple abnormal features and corresponding abnormal production data are matched, and the optimized production data is determined during the matching process. The negative feedback path between the optimization process and the production process is collected, and the intelligent feedback mechanism is determined based on the multiple training of the negative feedback path, the various optimized production data and the production data set of refractory bricks.

6. The intelligent control method for a refractory brick production line according to claim 1, characterized in that, The intelligent control mechanism for the refractory brick production line, determined through training based on intelligent detection and feedback mechanisms, intelligently controls the production, optimization, and testing processes of the refractory bricks, including: The system collects intelligent detection and feedback mechanisms, connects these mechanisms with the optimized processes of refractory bricks, and determines the intelligent control mechanism for the refractory brick production line based on the training of the optimized processes, intelligent detection, and intelligent feedback mechanisms.

7. The intelligent control method for a refractory brick production line according to claim 6, characterized in that, The intelligent control mechanism for the refractory brick production line, determined by training based on the intelligent detection and feedback mechanisms, intelligently controls the production, optimization, and testing processes of the refractory bricks, and further includes: The intelligent feedback mechanism is configured with a negative feedback path, and the intelligent detection mechanism is configured with a positive feedback path. The intelligent control mechanism based on the refractory brick production line is compatible with both negative and positive feedback paths, and forms the overall control path of the refractory brick production line. Multiple intelligent control nodes are determined based on the division of the overall control path, and these nodes are distributed in the production process, optimization process, and testing process of refractory bricks.

8. An intelligent control system for a refractory brick production line, characterized in that, The intelligent control system of the refractory brick production line is applied to the intelligent control method of the refractory brick production line as described in any one of claims 1-7, and the intelligent control system of the refractory brick production line includes: The process module is used to determine the production process, optimization process, and testing process for refractory bricks based on the refractory brick production line. The abnormal feature module is used to determine abnormal production data based on the production data set of refractory bricks in the refractory brick production process, and to determine multiple abnormal features of refractory bricks based on the abnormal production data and the output image of refractory bricks. The intelligent feedback module is used to generate multiple optimization data based on the optimization of multiple abnormal characteristics of refractory bricks in the optimization process of refractory bricks, determine the optimized production data based on the multiple optimization data and the corresponding abnormal production data, and generate an intelligent feedback mechanism. The intelligent detection module is used in the refractory brick inspection process to trigger multiple detection heads to perform intelligent detection relative to the refractory brick based on multiple optimized data and the morphology of the refractory brick, and to generate an intelligent detection mechanism. The intelligent control module is used to determine the intelligent control mechanism of the refractory brick production line based on the training of the intelligent detection mechanism and the intelligent feedback mechanism. The intelligent control mechanism of the refractory brick production line performs intelligent control on the production process, optimization process and detection process of refractory bricks.

Images