Steel ladle shell full-dimension infrared temperature measurement and refractory weak point prediction system and method

Abstract

The present application relates to a kind of steel ladle shell full dimension infrared temperature measurement and refractory weak point pre-judgment system and method, belong to steelmaking plant steel ladle safety monitoring technical field.The system is through multiple infrared temperature measurement device acquisition steel ladle shell full dimension temperature data, constructs complete temperature field, to this as core fusion steel ladle historical data, argon blowing wire feeding operating condition data, environmental interference data and refractory characteristic data, using machine learning algorithm constructs exclusive pre-judgment model;On the basis of self-learning pre-judgment model, introduce steel ladle operating condition real-time linkage model state optimization mechanism, combine converter tapping, argon blowing wire feeding, steel transfer and other real-time operating condition data dynamic optimization model weight coefficient, simultaneously in refractory erosion thickness quantification based on increase refractory remaining life prediction function.The present application adapts steelmaking severe operating condition, strong anti-interference, practicality is outstanding, fills the precise prediction technology blank of steel ladle refractory.

Description

Technical Field

[0001] This invention belongs to the field of steelmaking plant molten steel ladle safety monitoring technology, and relates to a system and method for full-dimensional infrared temperature measurement of molten steel ladle shell and prediction of refractory material weak points. It is applicable to the scenario of adding and removing the cover at the argon blowing and wire feeding station of the converter tapping line. It can realize full-dimensional temperature monitoring of molten steel ladle shell, accurate prediction of refractory material weak points, real-time optimization of the model state of molten steel ladle operation linkage, and accurate prediction of the remaining life of refractory material, preventing refractory material burn-through accidents and providing a forward-looking decision-making basis for molten steel ladle maintenance and refractory material replacement. Background Technology

[0002] The molten steel ladle is a core piece of equipment in a steel plant, responsible for carrying, transferring, and refining molten steel after it tapps from the converter. The refractory lining inside the ladle shell directly determines its safety. If problems such as weakness, corrosion, or detachment of the refractory lining are not detected in time, the molten steel can easily burn through the ladle shell, causing property damage and potentially leading to a major safety accident. However, in actual production, a margin is often left in the amount of refractory lining used to avoid this burn-through, thus not maximizing its utilization.

[0003] In existing refractory material monitoring technologies for molten steel ladles, some use handheld infrared thermometers for point-based monitoring, which suffers from problems such as missed detections, misjudgments, and low efficiency, and cannot achieve full-area monitoring of the ladle shell. Other technologies use fixed infrared thermometers, but these are mostly deployed in front of the furnace or in the hoisting area, resulting in blind spots in temperature measurement on the side of the molten steel ladle facing away from the thermometer, the area below the trunnion box, and the bottom of the ladle. This makes it impossible to achieve 360° temperature measurement of the molten steel ladle without dead angles, and it is difficult to construct a complete temperature field for the ladle shell.

[0004] Meanwhile, although existing technologies have achieved infrared temperature measurement and early warning functions, they rely solely on single temperature data for refractory condition analysis, without combining historical data from molten steel ladles, on-site environmental interference data, and refractory characteristic data. This results in low accuracy in predicting refractory weakness points, an inability to distinguish the specific causes of temperature anomalies, and a high likelihood of false or missed warnings. Furthermore, existing models lack a real-time operational condition linkage optimization mechanism. When conditions such as converter tapping, argon blowing and wire feeding, and molten steel transfer change, the model cannot dynamically adjust, leading to significant differences in prediction accuracy under different conditions. In addition, existing technologies can only quantify the refractory erosion thickness in a simple way, and cannot combine multi-dimensional data to accurately predict the remaining life of the refractory. They can only rely on subjective assessment of refractory replacement timing based on human experience, which is prone to misjudgment. This can either lead to burn-through risks due to missed judgments or premature removal of refractory materials due to overly conservative judgments, resulting in low utilization rates and wasted production costs.

[0005] Therefore, there is an urgent need for a technology to overcome the shortcomings of existing technologies in monitoring the temperature of refractory materials in molten steel ladles, such as incomplete temperature measurement coverage, low prediction accuracy, poor scenario adaptability, susceptibility to environmental interference, inability to adapt to the needs of ladle movement monitoring, lack of a model optimization mechanism for working condition linkage, and inability to accurately predict the remaining life of refractory materials. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide a system and method for full-dimensional infrared temperature measurement of molten steel ladle shell and prediction of refractory material weak points, so as to realize full-dimensional temperature measurement of molten steel ladle shell, accurately predict refractory material weak points, support real-time optimization of model state in conjunction with working conditions and accurate prediction of the remaining life of refractory material, effectively avoid the risk of burn-through, prevent refractory material from being taken off the production line in advance, and improve the service life of refractory material.

[0007] This invention achieves 360° temperature measurement of the molten steel ladle shell and complete temperature field construction by adding detection points on the opposite side of the ladle and optimizing the deployment of eight sets of temperature measuring devices. Combined with multi-dimensional data, it accurately predicts weak points in the refractory material. It introduces a model-state optimization mechanism for real-time linkage with the molten steel ladle's operating conditions, automatically optimizing model weights when operating conditions change, such as argon blowing and wire feeding, and molten steel transfer, thus improving prediction accuracy under different conditions. Based on the quantification of corrosion thickness, it constructs a sub-model for predicting the remaining service life of the refractory material, accurately predicting the remaining number of uses / duration of the refractory material. This invention effectively avoids molten steel ladle burn-through safety accidents and provides a forward-looking decision-making basis for molten steel ladle maintenance and refractory material replacement, preventing unnecessary premature removal of refractory materials, maximizing refractory material service life, increasing refractory material utilization, reducing production costs, and adapting to harsh steelmaking conditions and mobile molten steel ladle scenarios.

[0008] To achieve the above objectives, the present invention provides the following technical solution: Solution 1: A system for full-dimensional infrared temperature measurement and refractory material weakness prediction of molten steel ladle shell, comprising an infrared temperature measurement module, an environmental sensing module, a data acquisition module, a data fusion module, a prediction and early warning module, and a database module. These modules work collaboratively to achieve full-dimensional temperature measurement of the molten steel ladle shell, accurate prediction of refractory material weakness, real-time model optimization, and accurate prediction of the remaining life of the refractory material. The specific structure is as follows: Infrared temperature measurement modules are deployed on the columns of the cover-opening platform at the argon-feeding station of the converter tapping line, beside the cover-opening platform, and on a newly added infrared temperature measurement platform directly opposite the cover-opening platform of the converter argon-feeding station (or on the wall of the continuous casting ladle rotary table). These modules collect comprehensive temperature data of the molten steel ladle shell and transmit it to the data acquisition module. If there is no ladle rotary table directly opposite the converter tapping line and no existing installation platform, a new temperature measurement platform is specially constructed as the installation foundation for the temperature measurement device. This new platform is flush with the cover-opening platform, fixed to a concrete foundation, and uses a steel structure equipped with heat insulation and vibration damping features. It is suitable for the harsh working conditions of high temperature, vibration, and dust in the steelmaking site and does not obstruct the steel ladle transport passage. If the converter tapping line is directly opposite the continuous casting ladle rotary table, the rotary table partition wall can be used as the installation platform for the infrared thermometer.

[0009] The environmental sensing module, installed on the side of the lifting platform, integrates the acquisition of three types of parameters: dust concentration, flue gas concentration, and ambient temperature. It is used to detect environmental interference data (dust concentration, flue gas concentration, and ambient temperature) at the installation site of the infrared temperature measurement module, assisting in correcting the detection data of the infrared temperature measurement module and improving the detection accuracy of the infrared temperature measurement system.

[0010] The data acquisition module is connected to the infrared temperature measurement module, the environmental sensing module, the refractory parameter input terminal, and the secondary system terminal in the steelmaking workshop. It is used to collect five types of data, including the full-dimensional temperature data of the molten steel ladle shell transmitted by the infrared temperature measurement module, the on-site environmental interference data, the refractory material characteristic data, the historical data of the molten steel ladle, and the real-time operating condition data of the molten steel ladle. The module also preprocesses the collected raw data (i.e., noise reduction, filtering, removal of invalid data, and addition of dynamic feature tags to the real-time operating condition data) and transmits it to the data fusion module.

[0011] The data fusion module receives various types of data transmitted from the data acquisition module. It uses data calibration and feature extraction algorithms to correct for environmental interference in the temperature field data (combining dust, flue gas concentration, and ambient temperature). Then, it deeply integrates the data with refractory material property data, historical data of molten steel ladles, and real-time operating condition data of molten steel ladles with dynamic feature tags to establish a multi-dimensional data association model. The fused dataset is then output to the prediction and early warning module. During the fusion process, the dynamic characteristics of the real-time operating condition data are preserved, enabling precise linkage with the modal optimization mechanism of the prediction model.

[0012] The prediction and early warning module has a built-in model for predicting weak points in refractory materials, which is trained based on machine learning algorithms (random forest or neural network). This model has a model state optimization mechanism that links the working conditions of the molten steel ladle in real time, and integrates a sub-model for predicting the remaining life of refractory materials. It is used to locate weak points in refractory materials, quantify the corrosion thickness, and provide early warning of burn-through risk.

[0013] The database module is used to store historical data, refractory material property data, temperature measurement data, environmental data, real-time operating condition data, and prediction and life prediction results of the molten steel ladle. It establishes a full life cycle data archive for the molten steel ladle, supports data query and traceability, and provides data support for the continuous self-learning of the prediction and early warning model and the optimization of the remaining life prediction sub-model, realizing dedicated data management for a single molten steel ladle.

[0014] Preferably, the infrared temperature measurement module has a total of 8 sets of infrared temperature measurement devices. Among them, 2 sets are installed on the columns of the lifting platform, at a height directly above the trunnion of the molten steel ladle; 3 sets are installed on the side of the lifting platform, with the left and right sets used to scan one side of the ladle, and the middle set installed at an upward tilt of 30°~45° to scan the bottom of the molten steel ladle; the remaining 3 sets are installed on the newly added infrared temperature measurement platform opposite the lifting platform or on the wall of the continuous casting ladle turntable, with functions corresponding one-to-one with the 3 sets on the side of the lifting platform, used to scan the other side of the ladle. The 8 sets of infrared temperature measurement devices work together to achieve full-dimensional and 360° temperature measurement of the molten steel ladle shell (above and below the trunnion, bottom of the ladle, and both sides of the ladle), and collect temperature data (including the coordinates of the temperature measurement point, temperature value, and rate of temperature change) of each temperature measurement point on the molten steel ladle shell in real time, and transmit it to the data acquisition module.

[0015] Preferably, all eight infrared temperature measurement devices adopt a high-temperature resistant sealed structure, equipped with a high-temperature resistant quartz glass lens, a lens dust cover, and an automatic blowing device. Compressed air is used to periodically blow the lens surface (blowing frequency 30 seconds / time) to prevent dust and smoke adhesion. The sealed design also resists the penetration of dust and smoke from the site, ensuring temperature measurement accuracy. The temperature measurement device installed diagonally upwards from the middle of the platform side has an additional optimized lens dust cover sealing level to address dust accumulation issues associated with upward-facing installations. The infrared temperature measurement device has a temperature measurement range of 200℃~1800℃, a horizontal scanning angle of 60°, a vertical scanning angle of 45°, and a scanning speed ≥30 frames / second. This is suitable for the speed at which molten steel tanks are carried by steel tank trucks or lifted by cranes, allowing for full-dimensional scanning of the molten steel tank shell without requiring the vehicle or crane to stop.

[0016] Preferably, the environmental sensing module adopts an integrated sealed design with an automatic purging interface; it has a high temperature resistance range of 25℃~150℃, fully covering the steelmaking site environment; it has a dustproof rating of IP67+ and a vibration resistance rating of IK10, adapting to the vibration-proof and high-temperature-proof structure of the newly added platform; the measurement accuracy meets the usage requirements (dust concentration error ≤±2%, flue gas concentration error ≤±3%, ambient temperature error ≤±0.5℃).

[0017] Preferably, the data acquisition module connects to the infrared temperature measurement module, the environmental sensing module, the refractory parameter input terminal, and the secondary system terminal in the steelmaking workshop, respectively, to collect five types of core data. These include: ① Full-dimensional temperature data of the molten steel ladle shell transmitted by the infrared temperature measurement module; ② On-site environmental interference data (on-site dust concentration, flue gas concentration, ambient temperature); ③ Refractory property data (physicochemical material parameters and masonry process parameters of the refractory materials in the molten steel ladle); ④ Historical data of the molten steel ladle (number of times the molten steel ladle has been used, molten steel temperature, usage time, refractory masonry thickness, maintenance records, and refractory replacement cycle); ⑤ Real-time operating data of the molten steel ladle (converter tapping data, argon blowing and wire feeding, argon blowing station production data, molten steel transfer speed, and dynamic operating data such as sudden changes in molten steel temperature). The data acquisition module performs noise reduction and filtering on the collected raw data, removes invalid data, adds dynamic feature tags to the real-time operating data, and transmits it to the data fusion module.

[0018] Preferably, the prediction and early warning module incorporates a refractory material weakness prediction model trained using machine learning algorithms (random forest or neural network). This model possesses self-learning capabilities and a model-state optimization mechanism that links real-time steel ladle operating conditions. It also integrates a refractory material remaining life prediction sub-model. Receiving fused data from the data fusion module, the module performs model calculations to achieve four core functions: ① Accurately locating refractory material weaknesses (distinguishing between temperature anomalies caused by refractory material weakness, or by operating condition fluctuations and environmental interference); ② Quantifying refractory material erosion thickness (inverting the remaining refractory material thickness with an error ≤3%); ③ Predict the risk level of burn-through (divided into three levels: minor, moderate, and severe); when a moderate or higher risk is predicted, an audible and visual alarm will be automatically issued, and the warning information will be transmitted to the central control room and maintenance terminal; ④ Refractories remaining life prediction: through the built-in remaining life prediction sub-model, based on the quantified refractory corrosion thickness, and integrating the frequency of use of the molten steel ladle, molten steel operating conditions, refractory characteristics, historical loss data and real-time operating condition data, a refractory loss rate correlation model is constructed to accurately predict the remaining number of uses / duration of refractory materials, providing a forward-looking decision-making basis for molten steel ladle maintenance and refractory material replacement.

[0019] The real-time linkage model optimization mechanism is as follows: the refractory material weak point prediction model identifies the dynamic feature labels of the working conditions in the fusion data in real time. When the working conditions such as converter tapping parameters (such as changes in tapping temperature and tapping quantity), argon blowing and wire feeding parameters (such as adjustments to bottom blowing pressure and flow rate of molten steel ladle), molten steel transfer (such as speed changes), or molten steel temperature (sudden rise / drop) change, the refractory material weak point prediction model automatically identifies the type and magnitude of the working condition change, dynamically adjusts the weight coefficients of various influencing factors such as temperature field, environment, working conditions, or historical data, realizes real-time optimization of the model state, and improves the accuracy of refractory material prediction under different working conditions.

[0020] Among them, the refractory remaining life prediction sub-model is based on the quantified refractory erosion thickness, and integrates the frequency of use of molten steel ladle, molten steel operating conditions, refractory characteristics, historical loss data and real-time operating condition data. Through the algorithm, a loss rate correlation model is constructed to achieve accurate prediction of the remaining number of uses or duration of refractory.

[0021] Option 2: A method for full-dimensional infrared thermography of molten steel ladle shells and prediction of weak points in refractory materials, the specific steps of which are as follows: Step 1: Deployment and debugging of infrared temperature measurement: First, set up the new infrared temperature measurement platform directly opposite the lifting platform of the converter argon blowing and wire feeding station and at a suitable position at a distance from the running track of the molten steel ladle (ensuring no obstruction, optimal temperature measurement angle, no obstruction of the transportation channel and no impact on the hoisting of the molten steel ladle). Fix it firmly and debug the horizontal and verticality of the platform to ensure no vibration and no tilting. If the steel outlet line is directly opposite the ladle turret, the wall of the ladle turret can also be used as the installation carrier. Next, install 8 sets of infrared temperature measuring devices on the columns of the lifting platform, the sides of the platform, and the walls of the newly added platform (or the continuous casting ladle turret). Adjust the installation angles according to their functions: 2 sets on the columns should be aligned with the trunnion and above; 2 sets on the left and right sides of the platform should be aligned with the sides of the ladle; 1 set in the middle of the platform should be tilted upwards at 30°~45° and aligned with the bottom of the ladle; the 3 sets installed on the newly added platform should correspond one-to-one with the 3 sets on the lifting platform, aligned with the other side of the ladle. Adjust the device parameters (temperature range, scanning speed, scanning angle), and start the automatic purging device (set the purging frequency to 30 seconds / cycle) to ensure the temperature measuring devices work properly and are adapted to the moving speed of the molten steel ladle on the molten steel ladle car or when lifted by a crane. Install the integrated environmental sensor and start the automatic purging device.

[0022] Step 2: Multi-dimensional data acquisition: Real-time acquisition of full-dimensional temperature data of the molten steel ladle shell using an infrared temperature measurement module to construct a preliminary temperature field; Simultaneous acquisition of on-site environmental interference data, refractory material property data, historical data of the molten steel ladle, and real-time operating data of the molten steel ladle (dynamic data such as converter tapping, argon blowing and wire feeding, and molten steel transfer) using a data acquisition module; Noise reduction and filtering of all raw data to remove invalid data; and addition of dynamic feature tags to the real-time operating data.

[0023] Step 3: Data Fusion and Calibration: Through the data fusion module, environmental interference is corrected for the temperature field data. Then, the modified temperature field data is combined with other types of data for feature extraction and correlation fusion to form a multi-dimensional fused dataset. Dynamic feature labels are added to the real-time operating condition data to improve the linkage between the data and the model-state optimization mechanism and improve data accuracy.

[0024] Step 4: Prediction of Refractory Weak Points, Model Optimization, and Remaining Life Prediction: The fused dataset is input into the prediction model of the prediction and early warning module. The model first identifies the real-time dynamic feature labels of the fused data. If changes in the working conditions such as converter tapping parameters, argon blowing pressure and flow rate adjustment, sudden rise in molten steel temperature, and change in transfer speed are detected, the weight coefficients of each influencing factor are automatically adjusted to complete the real-time optimization of the model state. The optimized model completes the accurate location of refractory weak points, quantification of refractory erosion thickness (error ≤3%), and prediction of burn-through risk level through calculation. At the same time, the model automatically inputs the quantified refractory erosion thickness data into the built-in refractory remaining life prediction sub-model. The sub-model integrates the frequency of use of molten steel ladle, molten steel working conditions, refractory characteristics, historical loss data, and real-time working condition data to construct a refractory loss rate correlation model, accurately predicting the remaining number of uses / duration of refractory.

[0025] Step 5: Early Warning, Life Feedback, and Data Storage: When a moderate or higher risk of burn-through is predicted, an audible and visual alarm is automatically issued, and the early warning information is simultaneously transmitted to the central control room and maintenance terminal. At the same time, the remaining life prediction results of the refractory materials (remaining usage times / duration) are pushed to the steel ladle operation and maintenance management terminal to provide a forward-looking decision-making basis for maintenance and refractory material replacement. All collected data, model optimization parameters, prediction results, and life prediction results are stored in the database module, and the full life cycle data archive of the steel ladle is updated to provide data support for the continuous optimization of the subsequent prediction model and the remaining life prediction sub-model.

[0026] The beneficial effects of this invention are as follows: 1) Highly adaptable to various scenarios, with no blind spots in temperature measurement coverage: Suitable for the process layout of most existing steelmaking workshops, utilizing the steel ladle with a lifting platform at the argon-blowing wire feeding station and adding a new temperature measurement platform (or using the partition wall of the continuous casting ladle turntable) as the installation carrier for temperature measurement devices and integrated environmental sensors, deploying 8 sets of infrared temperature measurement devices. The 8 sets of devices have clear functional divisions and are symmetrically deployed, accurately covering all areas above and below the trunnion of the steel ladle, the bottom of the ladle, and both sides of the ladle body, completely solving the problem of the original temperature measurement blind spots; adaptable to mobile steel ladles, it can complete full-dimensional scanning without the steel ladle stopping, and all devices are deployed on the platforms on both sides, without obstructing the steel ladle truck transportation channel, which conforms to the layout specifications of steelmaking workshops, and is different from the existing general temperature measurement layout, with adaptability far exceeding existing technologies.

[0027] 2) High prediction accuracy, strong anti-interference ability, and excellent adaptability to working conditions: Based on the full-dimensional temperature field data of the molten steel ladle shell, and integrating five types of key data such as environmental interference data, a prediction model is constructed by combining machine learning algorithms. It can accurately distinguish the specific causes of temperature anomalies, accurately locate the weak points of refractory materials and quantify the corrosion thickness, with small prediction errors. At the same time, through the sealing, automatic purging, and lens dust cover design of the temperature measuring device, it effectively resists the interference of dust and smoke on site. The high temperature and vibration protection design of the newly added platform further ensures the stability of the temperature measuring device. The innovative introduction of a model state optimization mechanism with real-time linkage to working conditions means that when the working conditions such as argon blowing and wire feeding and molten steel transfer change, the model automatically optimizes the weight coefficients, completely solving the problem of large differences in prediction accuracy under different working conditions of the existing model, and achieving high-precision prediction under all working conditions.

[0028] 3) New remaining life prediction function, with strong forward-looking operation and maintenance decision-making: Based on the accurate quantification of refractory corrosion thickness, a sub-model for predicting the remaining life of refractory is constructed. It integrates the usage frequency of the molten steel ladle, the working condition of the molten steel, the characteristics of the refractory material and real-time working condition data to accurately predict the remaining number of times / duration of refractory material to be used. This breaks away from the extensive mode of traditional manual experience assessment and provides a scientific and forward-looking decision-making basis for the formulation of molten steel ladle maintenance plans and the selection of the timing of refractory material replacement. It avoids cost waste caused by blind replacement and also eliminates safety accidents caused by delayed replacement.

[0029] 4) Highly practical and with excellent safety assurance: The new platform is easy to install and securely fixed, requiring no large-scale modification of on-site equipment, and the cost is controllable; the device is adapted to the harsh working conditions of steel plants, enabling early prediction and graded warning of weak points in refractory materials, avoiding steel ladle burn-through accidents, providing accurate basis for steel ladle maintenance, extending the service life of steel ladle, reducing production costs, and improving the safety of steelmaking production.

[0030] 5) Significantly improved economic benefits and refractory material utilization: Accurate assessment of refractory material condition and prediction of remaining lifespan prevents premature removal from production, greatly extending refractory material lifespan. This changes the traditional, extensive management model that relies solely on manual experience to determine whether refractory materials need repair or removal from production. By using a model to precisely quantify refractory material corrosion thickness, locate weak points, and predict remaining lifespan, it clearly understands the true state of refractory material wear. This avoids both the risk of burn-through due to missed assessments and the unnecessary premature removal of refractory materials due to overly conservative judgments. While ensuring safe production, it maximizes the effective lifespan of refractory materials, improves refractory material utilization, reduces refractory material consumption and maintenance costs, and enhances the turnover efficiency of molten steel ladles.

[0031] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0032] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the installation of the steel ladle shell full-dimensional infrared temperature measurement and refractory weak point prediction system of the present invention; Figure 2 A schematic cross-sectional view of the infrared thermometer installation. Figure 3 This is a schematic longitudinal section diagram of the infrared thermometer installation.

[0033] Reference numerals: 1-Integrated environmental sensor; 2-Infrared thermometer; 3-Molten steel ladle; 4-Molten steel ladle truck; 5-PLC control system; 6-Molten steel ladle hoisting crane. Detailed Implementation

[0034] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0035] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0036] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0037] Please see Figures 1-3 This invention provides a system for full-dimensional infrared temperature measurement and refractory material weakness prediction of molten steel ladle shell, including an infrared temperature measurement module, an environmental sensing module, a data acquisition module, a data fusion module, a prediction and early warning module, and a database module. The modules work together to achieve full-dimensional temperature measurement of molten steel ladle shell and accurate prediction of refractory material weakness, real-time model optimization, and accurate prediction of the remaining life of refractory material.

[0038] The specific implementation method of this system is as follows: 1) Deployment of the new platform and infrared temperature measurement module: First, set up a new temperature measurement platform directly opposite the cover-opening platform of the converter argon blowing and wire feeding station, at a suitable distance from the steel ladle's running track (ensuring no obstruction, optimal temperature measurement angle, no obstruction of the transport channel, and no impact on the steel ladle hoisting). The platform is made of steel, with its height completely flush with the cover-opening platform, and its bottom is fixed to a concrete foundation with expansion bolts, equipped with a vibration-damping buffer structure. Install one set of infrared temperature measurement devices on each of the two columns on both sides of the cover-opening platform (2 sets in total), at a height directly above the trunnion of the steel ladle; install three sets of infrared temperature measurement devices on the side of the cover-opening platform, with two sets on the left and right sides aligned with one side of the ladle, and the middle set installed at a 35° upward angle, aligned with the bottom of the steel ladle; install three sets of infrared temperature measurement devices on the new platform, corresponding one-to-one with the three sets on the side of the cover-opening platform, aligned with the other side of the ladle, for a total of eight sets. An industrial high-temperature infrared thermometer with a scanning speed of 30 frames per second is selected, with a temperature measurement range of 200℃~1800℃, a horizontal scanning angle of 60° and a vertical scanning angle of 45°. It is equipped with a high-temperature resistant quartz glass lens, a lens dust cover and an automatic blowing device. The blowing frequency is set to once every 30 seconds to ensure that there is no dust or smoke adhering to the lens. The installation angle is adjusted so that a single temperature measuring device can cover the tank area at a height of 1~2.5m. The eight sets of devices work together to achieve 360° coverage of the molten steel tank shell without dead angles, and are adapted to the movement speed of the molten steel tank on the molten steel tank truck or when it is lifted by a crane.

[0039] 2) Deployment of Integrated Environmental Sensing Module: An integrated environmental sensing module is installed on the side of the lifting platform. It integrates the acquisition of three parameters: dust concentration, flue gas concentration, and ambient temperature. It features an integrated sealed design (high-temperature resistant quartz sealed shell, consistent with the lens protection of the infrared temperature measuring device), and a built-in automatic purging interface (which can be linked with the automatic purging system of the infrared temperature measuring device, purging synchronously once every 30 seconds to prevent dust from adhering to the sensor probe). It has a high-temperature resistance range of 25℃~150℃, fully covering the steelmaking site environment; a dustproof rating of IP67, and a vibration resistance rating of IK10, adapting to the vibration-resistant and high-temperature-resistant structure of the newly added platform; the measurement accuracy meets the usage requirements (dust concentration error ≤ ±2%, flue gas concentration error ≤ ±3%, ambient temperature error ≤ ±0.5℃).

[0040] 3) Data acquisition module debugging: Connect the data acquisition module to the terminal for inputting 8 sets of infrared temperature measuring devices, integrated environmental sensing module, refractory parameters and real-time operating conditions of molten steel ladle. Set the data acquisition frequency to 1 time / second. Use wavelet denoising algorithm to denoise the acquired raw data, remove invalid data such as outliers and missing values, and add dynamic feature labels to the real-time operating conditions data according to preset rules.

[0041] 4) Data Fusion and Model Training: The data fusion module uses principal component analysis to extract core features from various data types. First, it corrects the temperature field data by incorporating dust concentration, flue gas concentration, and ambient temperature. Then, it fuses the data with historical data from molten steel ladles, argon blowing and wire feeding operation data, refractory material characteristic data, and real-time operation data with dynamic feature labels. A random forest algorithm is used to train the refractory material weak point prediction model, incorporating a model-state optimization mechanism that links real-time operation conditions and setting rules for adjusting weight coefficients when operating conditions change. Simultaneously, a sub-model for predicting the remaining life of refractory materials is trained. 1000 sets of molten steel ladle operation, maintenance, and loss data are imported for training, ensuring that the model's refractory material erosion thickness inversion error is ≤3%, risk prediction accuracy is ≥98%, and the refractory material remaining life prediction error is ≤5%. The model utilizes historical data for self-learning to improve computational accuracy.

[0042] 5) System Operation and Verification: After the converter taps steel, when the molten steel ladle car passes the lifting and unloading platform at a speed of 20-30 m / s, the infrared temperature measuring device on the column of the lifting and unloading platform scans the area above the trunnion of the molten steel ladle. When the molten steel ladle car transports the molten steel ladle to the crane lifting position, and the crane lifts the molten steel ladle from the car, the other 6 sets of infrared temperature measuring devices scan simultaneously (2 sets on both sides of the newly added platform and 2 sets on both sides of the lifting and unloading platform symmetrically scan both sides of the ladle body, and 1 set in the middle of the newly added platform and the lifting and unloading platform scans the bottom of the ladle). This completes the full-dimensional temperature acquisition of the molten steel ladle shell without requiring the vehicle or crane to stop. The data acquisition module simultaneously collects environmental data such as the ambient dust concentration of 50 mg / m³ and the ambient temperature of 45℃. The data acquisition module reads real-time operating conditions from the converter secondary system, including the converter tapping temperature of 1630℃, the tapping amount of 200t, the argon blowing pressure at the bottom of the molten steel ladle of 0.6 MPa, the wire feeding length of 12m, the wire feeding station processing time of 6min, and the molten steel ladle transfer speed of 25m / min. Data: When the argon blowing pressure is adjusted to 0.4 MPa (operating condition changes), the model automatically identifies the characteristics of this operating condition and adjusts the weight coefficients to complete model state optimization. After the data fusion module completes the data fusion, it transmits the data to the prediction and early warning module. The optimized model accurately identifies one weak point in the refractory material, inverts the remaining thickness of the working layer to 120 mm, predicts a slight risk, and does not issue an audible or visual alarm. At the same time, the remaining life prediction sub-model integrates the operating condition data and the frequency of use of the molten steel ladle, predicting that the remaining number of uses of the refractory material in the working layer is 50 heats, which is consistent with the actual maintenance and wear trend. When the simulated weak point in the refractory material is further eroded and the remaining working layer thickness drops to 100 mm, the model predicts a moderate risk, automatically issues an audible and visual alarm, synchronously transmits the early warning information to the central control room, and updates the remaining working layer life prediction result to 15 heats, pushing it to the operation and maintenance management terminal. This verifies the effectiveness and accuracy of the system in avoiding burn-through risk, accurately judging the refractory material condition, accurately predicting the remaining life, and avoiding premature shutdown.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A system for full-dimensional infrared temperature measurement and refractory material weak point prediction of molten steel ladle shell, characterized in that, It includes an infrared temperature measurement module, an environmental sensing module, a data acquisition module, a data fusion module, and a prediction and early warning module; The infrared temperature measurement module is deployed on the column of the cover-opening platform of the converter tapping line argon feeding station, on the side of the cover-opening platform, on the newly added infrared temperature measurement platform directly opposite the cover-opening platform of the converter argon feeding station, or on the wall of the continuous casting ladle rotary table. It is used to collect full-dimensional temperature data of the molten steel ladle shell and transmit it to the data acquisition module. The environmental sensing module is used to detect environmental interference data at the installation site of the infrared temperature measurement module, assist in correcting the detection data of the infrared temperature measurement module, and improve the detection accuracy of the infrared temperature measurement system. The data acquisition module is connected to the infrared temperature measurement module, the environmental sensing module, the refractory parameter input terminal, and the secondary system terminal of the steelmaking workshop, respectively. It is used to collect five types of data, including the full-dimensional temperature data of the molten steel ladle shell transmitted by the infrared temperature measurement module, the on-site environmental interference data, the refractory property data, the historical data of the molten steel ladle, and the real-time operating condition data of the molten steel ladle. The module also preprocesses the collected raw data and transmits it to the data fusion module. The data fusion module uses data calibration and feature extraction algorithms to correct environmental interference in the temperature field data. Then, it deeply integrates the temperature field data with refractory material property data, historical data of molten steel ladle, and real-time operating condition data of molten steel ladle to establish a multi-dimensional data association model and output the fused dataset to the prediction and early warning module. The prediction and early warning module has a built-in refractory weak point prediction model trained based on machine learning algorithms. This model has a model state optimization mechanism that links the working conditions of the molten steel ladle in real time, and integrates a refractory remaining life prediction sub-model to realize the location of refractory weak points, quantification of corrosion thickness, and early warning of burn-through risk.

2. The omnidirectional infrared temperature measurement and refractory weak point prediction system for molten steel ladle shells according to claim 1, characterized in that, The infrared temperature measurement module consists of eight sets of infrared temperature measurement devices. Two sets are installed on the columns of the lifting platform, at a height directly above the trunnion of the molten steel ladle. Three sets are installed on the side of the lifting platform: the left and right sets are used to scan one side of the ladle, and the middle set is installed at an upward angle of 30° to 45° to scan the bottom of the molten steel ladle. The remaining three sets are installed on the newly added infrared temperature measurement platform opposite the lifting platform or on the wall of the continuous casting ladle turntable. Their functions correspond one-to-one with the three sets on the side of the lifting platform, and they are used to scan the other side of the ladle. The eight sets of infrared temperature measurement devices work together to achieve full-dimensional and 360° temperature measurement of the molten steel ladle shell without blind spots.

3. The omnidirectional infrared temperature measurement and refractory weak point prediction system for molten steel ladle shells according to claim 2, characterized in that, All eight infrared temperature measurement devices adopt a high-temperature resistant sealed structure.

4. The omnidirectional infrared temperature measurement and refractory weak point prediction system for molten steel ladle shells according to claim 1, characterized in that, The environmental sensing module adopts an integrated sealed design and comes with an automatic purging interface.

5. The omnidirectional infrared temperature measurement and refractory weak point prediction system for molten steel ladle shells according to claim 1, characterized in that, The data acquisition module preprocesses the acquired raw data, specifically including noise reduction and filtering, removing invalid data, and adding dynamic feature labels to the real-time operating data.

6. The omnidirectional infrared temperature measurement and refractory weak point prediction system for molten steel ladle shells according to claim 1, characterized in that, The environmental interference data includes dust concentration, flue gas concentration, and ambient temperature; The refractory material property data includes the physical and chemical material parameters and masonry process parameters of the refractory material for molten steel ladles; The historical data of the molten steel ladle includes the number of times the molten steel ladle was used, the temperature of the molten steel, the duration of use, the thickness of the refractory lining, maintenance records, and the replacement cycle of the refractory materials; The real-time operating data of the molten steel ladle includes converter tapping data, argon blowing station production data, molten steel transfer speed, and molten steel temperature change data.

7. The omnidirectional infrared temperature measurement and refractory weak point prediction system for molten steel ladle shells according to claim 5, characterized in that, In the data fusion module, the dynamic characteristics of real-time operating data are retained during the fusion process, enabling precise linkage with the modal optimization mechanism of the prediction model.

8. The omnidirectional infrared temperature measurement and refractory weak point prediction system for molten steel ladle shells according to claim 1, characterized in that, In the prediction and early warning module, the model state optimization mechanism for real-time linkage of working conditions is as follows: the refractory weak point prediction model identifies the dynamic feature tags of the working conditions in the fused data in real time. When the working conditions such as converter tapping parameters, argon blowing and wire feeding parameters, molten steel transfer or molten steel temperature change, the refractory weak point prediction model automatically identifies the type and magnitude of the working condition change, dynamically adjusts the weight coefficients of each influencing factor, realizes real-time optimization of the model state, and improves the accuracy of refractory prediction under different working conditions. The refractory remaining life prediction sub-model is based on the quantified refractory erosion thickness, and integrates the frequency of use of the molten steel ladle, molten steel operating conditions, refractory characteristics, historical loss data and real-time operating condition data. It constructs a loss rate correlation model through an algorithm to achieve accurate prediction of the remaining number of uses or duration of refractory.

9. The system for full-dimensional infrared temperature measurement and refractory material weak point prediction of steel ladle shell according to any one of claims 1 to 8, characterized in that, The system also includes a database module for storing historical data of the molten steel ladle, refractory material property data, temperature measurement data, environmental data, real-time operating condition data, and prediction and life prediction results. It establishes a full life cycle data archive for the molten steel ladle, supports data query and traceability, and provides data support for the continuous self-learning of the prediction and early warning model and the optimization of the remaining life prediction sub-model, thus realizing dedicated data management for a single molten steel ladle.

10. The omnidirectional infrared temperature measurement and refractory weak point prediction system for molten steel ladle shells according to claim 9, characterized in that, The implementation steps of this system are as follows: Step 1: Infrared Temperature Measurement Deployment and Debugging: First, set up and fix the new infrared temperature measurement device installation platform directly opposite the cover-opening platform of the converter argon blowing and wire feeding station. The clearance between the new platform and the cover-opening platform should meet the hoisting requirements of the molten steel ladle. If the steel outlet line is directly opposite the ladle turret, the wall of the ladle turret can be used as the installation carrier. Then, install 8 sets of infrared temperature measurement devices on the columns of the cover-opening platform, the side of the platform, and the wall of the new platform or the continuous casting ladle turret. Debug the installation angle according to the functional division and start the automatic purging device. Install the integrated environmental sensor and start the automatic purging device. Step 2: Multi-dimensional data acquisition: Real-time acquisition of full-dimensional temperature data of the molten steel ladle shell through the infrared temperature measurement module to construct a preliminary temperature field; Simultaneous acquisition of on-site environmental interference data, refractory material property data, historical data of the molten steel ladle, and real-time operating condition data of the molten steel ladle through the data acquisition module; Noise reduction and filtering of all raw data, removal of invalid data, and addition of dynamic feature labels to the real-time operating condition data. Step 3: Data Fusion and Calibration: Through the data fusion module, environmental interference is corrected for the temperature field data. Then, the modified temperature field data is combined with other types of data for feature extraction and correlation fusion to form a multi-dimensional fused dataset. Dynamic feature labels are added to the real-time operating condition data. Step 4: Prediction of refractory material weak points. The fused dataset is input into the prediction model of the prediction and early warning module. The model first identifies real-time operating conditions. If the operating conditions change, the weight coefficients are automatically optimized to adjust the model state. Then, the weak point is located, the corrosion thickness is quantified, and the risk is predicted. At the same time, the quantified corrosion thickness data is input into the refractory material remaining life prediction sub-model. The sub-model integrates the steel ladle usage frequency, steel operating conditions, refractory material characteristics, historical loss data, and real-time operating condition data to construct a refractory material loss rate correlation model to predict the remaining number of uses or duration of refractory material. Step 5: Early warning and data storage. Issue tiered early warnings and simultaneously output the remaining life prediction results of refractory materials, and update the full life cycle data archive of molten steel ladles.

Images