Detection system and method for measuring and calculating thickness based on infrared monitoring ladle surface temperature data

Abstract

The invention discloses a detection system and method for measuring and calculating thickness based on infrared monitoring ladle surface temperature data. The system comprises a visible light identification module, an infrared temperature monitoring module, a data processing and analysis module and an early warning and output module. The method comprises the following steps: identifying a ladle number; collecting and processing surface temperature data; based on the steady-state heat conduction model, the real-time thickness of the ladle lining is inversed through the outer surface temperature; predicting the residual life through a thickness loss model; setting a grading early warning threshold value and triggering an alarm; and displaying the monitoring data in real time. The system achieves accurate evaluation and early warning of the thickness and the service life through cooperative temperature measurement of a plurality of infrared devices in combination with initial calibration and black body calibration, and effectively improves the safety monitoring and maintenance decision efficiency of the ladle.

Description

Technical Field

[0001] This invention belongs to the field of ladle wall thickness measurement technology, and particularly relates to a detection system and method for calculating the thickness based on infrared monitoring of ladle surface temperature data. Background Technology

[0002] With the iterative upgrading of modern industrial technology, the surface temperature monitoring of molten iron ladles is undergoing a leapfrog transformation from traditional manual inspection to an intelligent monitoring system. Early methods relying on periodic manual temperature measurements, limited by the time and space constraints of manual operations, not only suffer from monitoring blind spots and time lags, but also struggle to capture real-time dynamic temperature field changes on the ladle surface. This extensive monitoring approach makes it difficult to accurately locate equipment malfunctions, and historical temperature data is fragmented and incomplete, severely hindering the scientific assessment of the ladle's service status and preventative maintenance.

[0003] In existing technologies, thermal imaging network systems based on Internet of Things (IoT) technology can not only achieve 24 / 7 uninterrupted monitoring of the molten iron ladle and its surrounding environment, but also automatically identify abnormal temperature rise areas through AI algorithms, trigger a graded early warning mechanism, and simultaneously generate digital temperature profiles.

[0004] However, this technical solution can only assist technicians in making a relatively scientific assessment of the service status of the molten iron ladle through the early warning function. It cannot accurately and comprehensively analyze the data of the molten iron ladle to form a report, and the assessment of the status of the molten iron ladle is delayed, so it cannot provide early warning of potential dangers. Summary of the Invention

[0005] The purpose of this invention is to provide a detection system and method for calculating the thickness of a molten iron ladle based on infrared monitoring of the surface temperature data, so as to solve the above-mentioned technical problems.

[0006] To solve the above-mentioned technical problems, the specific technical solution of the detection system and method for calculating the thickness based on infrared monitoring of the surface temperature data of molten iron ladle is as follows:

[0007] A detection system for calculating the thickness of a molten iron ladle based on infrared monitoring of surface temperature data includes:

[0008] Visible light recognition module: used to acquire images of molten iron ladles and identify the ladle number using a trained neural network model;

[0009] Infrared temperature monitoring module: includes at least three infrared monitoring devices, respectively arranged on the side and bottom of the molten iron ladle, for collecting surface temperature data;

[0010] Data processing and analysis module: used to receive and process visible light and infrared data, and perform package number identification, temperature analysis, thickness inversion and lifetime prediction;

[0011] Early warning and output module: including audible and visual alarm and display device, used to issue alarm signals when thickness abnormality or lifespan warning is detected, and to display monitoring results in real time.

[0012] Furthermore, all infrared temperature monitoring modules are connected to the local area network via a switch, and the data is wirelessly transmitted to the remote control room via a bridge.

[0013] This invention also discloses a method for calculating the thickness of a molten iron ladle using the aforementioned detection system based on infrared monitoring of the ladle's surface temperature data, comprising the following steps:

[0014] Step 1: Identify the ladle number using a visible light vision algorithm:

[0015] Step 2: Surface temperature data acquisition and processing:

[0016] Step 3: Calculate the real-time thickness of the ladle lining based on infrared temperature data: Obtain the outer surface temperature of the ladle through infrared monitoring equipment, establish a physical model between temperature and lining thickness, and calculate the thickness in real time based on this model.

[0017] Step 4: Calculate the remaining service life of the molten iron ladle based on thickness loss: By tracking the change pattern of thickness with the number of uses, a loss model is established to predict the remaining service life and realize risk warning;

[0018] Step 5: Tiered Early Warning: The system sets multi-level early warning thresholds. When the real-time thickness (H_a) approaches or the predicted RUL is lower than the set threshold, the system automatically triggers an audible and visual alarm and highlights the early warning information and the predicted number of remaining uses on the display screen to guide maintenance decisions.

[0019] Step 6: Data Display and Early Warning Output:

[0020] Furthermore, step 1 includes the following steps:

[0021] When the molten iron ladle enters the preset range, the infrared monitoring device 4 identifies the ladle. After detecting that the molten iron ladle has entered the detection area and stays for 3 seconds, it sends out a ladle arrival signal. The visible light camera 3 receives the ladle arrival signal and identifies the ladle number. Multiple detections are used to increase the accuracy of ladle number identification.

[0022] By acquiring real datasets from the field, cropping images using preset bounding boxes to extract target regions, labeling individual digits and organizing the dataset, and finally training a digit detection model based on YOLOv8, a digit detection model is obtained.

[0023] Furthermore, step 1 includes: using a visible light camera 3 to acquire images and crop digital regions, using a neural network to recognize characters, calculating the distance between the midpoints of the characters, merging adjacent characters from left to right to form multi-digit numbers, and finally outputting the most frequently occurring number among all recognition results during the tank's stay as the package number.

[0024] Furthermore, step 2 includes the following steps:

[0025] Infrared monitoring equipment collects thermal radiation data from the surface of molten iron ladles, filters and reduces noise to segment the target area, converts the radiation intensity into temperature values, combines blackbody calibration to correct errors, analyzes temperature distribution characteristics, and extracts the highest temperature, average temperature, and coordinates of high-temperature areas, ultimately generating a visualized temperature map and an over-limit warning signal.

[0026] Furthermore, step 3 includes the following steps:

[0027] Step 3.1: Physical Model Establishment:

[0028] The molten iron ladle is simplified into a multi-layered, flat-walled or cylindrical-walled steady-state heat conduction model. The interior of the ladle lining contains high-temperature molten iron, while the exterior is filled with ambient air. According to Fourier's law of thermal conductivity, the temperature T_s at the measuring point on the outer surface of the ladle has a clear functional relationship with the lining thickness H. For the flat-walled model, the simplified relationship is as follows:

[0029]

[0030] Where T_s is the measured temperature of the outer surface, T_m is the temperature of the molten iron, T_a is the ambient temperature, H is the real-time thickness of the lining to be determined, k is the thermal conductivity of the lining material, and h is the convective heat transfer coefficient between the outer wall of the lining and the air.

[0031] Step 3.2: Model parameter determination and calibration:

[0032] Step 3.2.1: Initial calibration: Before the new ladle is put into use, its initial thickness H_0 and the outer surface temperature T_s0 under stable conditions when filled with molten iron are measured. H_0 and T_s0 are substituted into the above model. Combined with the known T_m and T_a, the effective comprehensive thermal resistance or calibration coefficient under the current working conditions is calculated, and a specific function relationship H = f(T_s, T_m, T_a; k, h) is established.

[0033] Step 3.2.2: Blackbody calibration: Use a blackbody source set up on site to periodically calibrate the absolute temperature measurement value of the infrared monitoring equipment to ensure the accuracy of the T_s data;

[0034] Step 3.3: Real-time thickness calculation:

[0035] When the system is running, the real-time collected T_s, as well as the synchronously acquired T_m and T_a, are substituted into the calibrated functional relationship H = f(T_s, T_m, T_a) to calculate the real-time thickness H corresponding to the measuring point;

[0036] Step 3.4: Multi-regional data fusion:

[0037] After calculating the thickness independently for each infrared monitoring point, the real-time thickness H representing the entire molten iron ladle is obtained by taking the average value, minimum value, or weighted value of a specific area.

[0038] Furthermore, step 4 includes the following steps:

[0039] Step 4.1: Data Recording and Sequence Generation: The system records key data for each use cycle of each molten iron ladle, forming a data sequence: number of uses a, and the real-time thickness of the ladle lining H_a calculated after each use;

[0040] Step 4.2: Calculation of Thickness Loss △H: For the a-th use, the formula for calculating the cumulative thickness loss is:

[0041]

[0042] Where H_0 is the initial thickness, and H_a is the real-time thickness calculated after the a-th use;

[0043] Step 4.3: Loss Model Establishment and Fitting: Using the number of uses a as the independent variable and the cumulative thickness loss △H_a as the dependent variable, the historical data points (a, △H_a) are fitted with a curve;

[0044] Step 4.4: Remaining life prediction and early warning:

[0045] Safety thickness threshold setting: Based on process safety standards, set the critical thickness at which the molten iron ladle needs to be taken offline for maintenance or scrapped;

[0046] Remaining useful life prediction: The fitted functional relationship ΔH = g(a) is solved inversely, i.e.

[0047] The predicted total number of uses before reaching the critical thickness is calculated, and the remaining service life is:

[0048] .

[0049] Furthermore, step 5 includes the following steps:

[0050] When the real-time thickness H_a approaches or the predicted RUL falls below the set threshold, the system automatically triggers an audible and visual alarm and displays a warning message and a prediction of the remaining usage times on the screen to guide maintenance decisions.

[0051] Furthermore, step 6 includes the following steps:

[0052] The control room display screen shows real-time temperature heatmaps, thickness curves, remaining lifespan, and warning information; it also supports historical data query, report export, and video playback.

[0053] When the thickness is below the set threshold or the remaining lifespan is insufficient, an audible and visual alarm is triggered and a notification is pushed to the relevant terminal.

[0054] The detection system and method for calculating the thickness of a molten iron ladle based on infrared monitoring of surface temperature data has the following advantages:

[0055] 1. Non-contact, real-time online monitoring of molten iron ladle thickness has been achieved. Utilizing infrared thermometry technology, the ladle lining thickness can be retrieved from surface temperature data without machine shutdown or manual contact, thus achieving continuous and automated monitoring.

[0056] 2. Improved the accuracy and scientific rigor of thickness and lifespan assessment. A physical model based on steady-state thermal conductivity theory was established, and the model parameters were calibrated and corrected through initial calibration and blackbody calibration, overcoming the influence of material differences and environmental interference, thus making thickness inversion and loss calculation more accurate.

[0057] 3. Intelligent prediction and early warning of remaining service life has been achieved. By tracking the change pattern of thickness with the number of uses, a wear model is established to dynamically predict the remaining service life. Combined with a multi-level early warning mechanism, it can proactively alarm when the thickness is abnormal or the service life is nearing its end, guiding preventive maintenance and effectively avoiding safety accidents.

[0058] 5. Improved the informatization and visualization level of molten iron ladle management. The system integrates functions such as automatic ladle number recognition, temperature heat map generation, thickness curve display, and historical data query. The results can be displayed in real time through the display terminal and reports can be exported, realizing centralized management and visual analysis of molten iron ladle data throughout its entire life cycle.

[0059] 6. The system is flexible in deployment and highly adaptable. The monitoring equipment can be networked using wireless transmission, making it easy to install and adjust in complex industrial sites such as cranes and blast furnaces, and it has good engineering applicability and scalability.

[0060] In summary, this invention solves the problems of poor timeliness, discontinuous data, and inability to quantitatively assess thickness and lifespan in traditional manual inspection methods, providing an efficient and reliable technical means for the safe operation and intelligent maintenance of molten iron ladles. Attached Figure Description

[0061] Figure 1 A schematic diagram of a device for identifying the ladle number of molten iron using visible light.

[0062] Figure 2 This is a schematic diagram of an infrared device for monitoring the surface temperature of molten iron ladles.

[0063] Figure 3 This is a diagram of the device topology.

[0064] The markings in the diagram are as follows: 1. Crane boom; 2. Ladle; 3. Visible light camera; 4. Infrared monitoring device one; 5. Infrared monitoring device three; 6. Infrared monitoring device two. Detailed Implementation

[0065] To better understand the purpose, structure, and function of this invention, the following description, in conjunction with the accompanying drawings, provides a more detailed account of a detection system and method for calculating thickness based on infrared monitoring of the surface temperature data of a molten iron ladle.

[0066] The present invention provides a detection system for calculating the thickness of a molten iron ladle based on infrared monitoring of surface temperature data, comprising:

[0067] Visible light recognition module: used to acquire images of molten iron ladles and identify the ladle number using a trained neural network model;

[0068] Infrared temperature monitoring module: includes at least three infrared monitoring devices, respectively arranged on the side and bottom of the molten iron ladle, for collecting surface temperature data;

[0069] Data processing and analysis module: used to receive and process visible light and infrared data, and perform package number identification, temperature analysis, thickness inversion and lifetime prediction;

[0070] Early warning and output module: including audible and visual alarm and display device, used to issue alarm signals when thickness abnormality or lifespan warning is detected, and to display monitoring results in real time.

[0071] System hardware configuration such as Figure 1 Figure 2 As shown, the molten iron ladle 2 is hung on the crane boom 1. The molten iron ladle 2 has a ladle number. The visible light camera 3 is set on the side facing the ladle number to identify the ladle number. Infrared monitoring device 1 4 and infrared monitoring device 2 6 are respectively set on both sides of the molten iron ladle. Infrared monitoring device 3 5 is set on the bottom of the molten iron ladle to identify the surface temperature of the sides and bottom of the molten iron ladle.

[0072] The system's hardware connection topology is as follows Figure 3 As shown, all monitoring devices are connected to the local area network via a switch, and the data is wirelessly transmitted to the remote control room via a bridge.

[0073] The present invention provides a method for calculating the thickness of a molten iron ladle based on infrared monitoring of surface temperature data, the method comprising the following essential steps:

[0074] Step 1: Identify the ladle number using a visible light vision algorithm:

[0075] When the molten iron ladle enters the preset range, the infrared monitoring device 4 identifies the ladle. After detecting that the molten iron ladle has entered the detection area and stays for 3 seconds, it sends out a ladle arrival signal. The visible light camera 3 receives the ladle arrival signal and identifies the ladle number. Multiple detections are used to increase the accuracy of ladle number identification.

[0076] By acquiring a large amount of real-world data from the site, using pre-defined cropping boxes to extract target regions, labeling individual digits, and organizing the dataset, a digit detection model was finally trained based on YOLOv8. In practical applications, a visible light camera was used to capture images and crop digit regions. After recognizing characters using a neural network, the distance between the midpoints of the characters was calculated, and neighboring characters were merged from left to right to form multi-digit numbers. Finally, the most frequently occurring digit among all recognition results during the tank's stay was used as the package number for output.

[0077] Step 2: Surface temperature data acquisition and processing:

[0078] Infrared monitoring equipment collects thermal radiation data from the surface of molten iron ladles, filters and reduces noise to segment the target area, converts the radiation intensity into temperature values, combines blackbody calibration to correct errors, analyzes temperature distribution characteristics, and extracts the highest temperature, average temperature, and coordinates of high-temperature areas, ultimately generating a visualized temperature map and an over-limit warning signal.

[0079] Step 3: Calculate the real-time thickness of the ladle lining based on infrared temperature data: Obtain the outer surface temperature of the ladle through infrared monitoring equipment, establish a physical model between temperature and lining thickness, and calculate the thickness in real time based on this model.

[0080] Step 3.1: Physical Model Establishment:

[0081] The molten iron ladle is simplified into a multi-layered, flat-walled or cylindrical-walled steady-state heat conduction model. The interior of the ladle contains high-temperature molten iron (a constant heat source, temperature T_m), while the exterior is ambient air. According to Fourier's law of thermal conductivity, there is a clear functional relationship between the temperature (T_s) at the measuring point on the outer surface of the ladle and the thickness (H) of the ladle lining. For the flat-walled model, the simplified relationship is:

[0082]

[0083] Where T_s is the measured temperature of the outer surface, T_m is the temperature of the molten iron (which can be obtained through process parameters or measured briefly), T_a is the ambient temperature, H is the real-time thickness of the lining to be determined, k is the thermal conductivity of the lining material, and h is the convective heat transfer coefficient between the outer wall of the lining and the air.

[0084] Step 3.2: Model parameter determination and calibration:

[0085] Step 3.2.1: Initial Calibration: Before putting a new ladle into use, measure its initial thickness (H_0) and the outer surface temperature (T_s0) under stable conditions containing molten iron. Substitute H_0 and T_s0 into the above model, and combine them with the known T_m and T_a, to back-calculate the effective comprehensive thermal resistance or calibration coefficient (such as the actual equivalent values ​​of k and h) under the current operating conditions, and establish a specific function relationship H = f(T_s,T_m,T_a; k,h).

[0086] Step 3.2.2: Blackbody calibration: Use a blackbody source set up on site to periodically calibrate the absolute temperature measurement value of the infrared monitoring equipment to ensure the accuracy of the T_s data.

[0087] Step 3.3: Real-time thickness calculation:

[0088] When the system is running, the real-time collected T_s, as well as the synchronously acquired T_m and T_a, are substituted into the calibrated functional relationship H = f(T_s, T_m, T_a) to calculate the real-time thickness (H) corresponding to the measuring point.

[0089] Step 3.4: Multi-regional data fusion:

[0090] Multiple infrared monitoring points are typically arranged on the outer surface of the molten iron ladle. After calculating the thickness of each point independently, the real-time thickness (H) representing the entire ladle can be obtained by taking the average value, minimum value, or weighted value for a specific area.

[0091] Note: The relationship between thickness (H) and T (surface temperature of molten iron ladle) needs to be re-verified for molten iron ladles made of different materials.

[0092] Step 4: Calculate the remaining service life of the molten iron ladle based on thickness loss: By tracking the change pattern of thickness with the number of uses, a loss model is established to predict the remaining service life and realize risk warning.

[0093] Step 4.1: Data Recording and Sequence Generation: The system records key data for each ladle in each usage cycle (a complete cycle of filling, transporting, and dumping) to form a data sequence: number of uses (a), and the real-time thickness of the ladle lining calculated after each use (H_a).

[0094] Step 4.2: Thickness Loss (ΔH) Calculation: For the a-th use, the cumulative thickness loss is calculated using the following formula:

[0095]

[0096] Where H_0 is the initial thickness and H_a is the real-time thickness calculated after the a-th use.

[0097] Step 4.3: Loss Model Establishment and Fitting: Using the number of uses (a) as the independent variable and the cumulative thickness loss (ΔH_a) as the dependent variable, perform curve fitting on the historical data points (a, ΔH_a). Typically, the loss process may exhibit linear or nonlinear characteristics.

[0098] For example:

[0099] Linear model:

[0100]

[0101] Where β is the average loss thickness (slope) per use, and C is the initial deviation (intercept, ideally 0).

[0102] Nonlinear models (such as exponential decay):

[0103]

[0104] Where γ is the loss coefficient.

[0105] The optimal model and parameters (β, γ, etc.) are determined by regression analysis (such as least squares method).

[0106] Step 4.4: Remaining life prediction and early warning:

[0107] Safety thickness threshold (H_critical) setting: Based on process safety standards, set the critical thickness at which the molten iron ladle needs to be taken offline for maintenance or scrapped.

[0108] Remaining useful life (RUL) prediction: The fitted functional relationship ΔH = g(a) is solved inversely, i.e.

[0109] Calculate the predicted total number of uses before reaching the critical thickness. The remaining service life is:

[0110] .

[0111] Step 5: Tiered Early Warning: The system sets multi-level early warning thresholds (such as warning and alarm). When the real-time thickness (H_a) approaches or the predicted RUL is lower than the set threshold, the system automatically triggers an audible and visual alarm and highlights the early warning information and the predicted remaining number of uses on the display screen to guide maintenance decisions.

[0112] Note: The functional relationship between △H (loss thickness) and a (number of times the ladle is used) is related to the material of the ladle.

[0113] Step 6: Data Display and Early Warning Output:

[0114] The temperature heat map, thickness curve, remaining lifespan, and early warning information are displayed in real time on the control room screen.

[0115] Supports historical data query, report export, and video playback;

[0116] When the thickness is below the set threshold or the remaining lifespan is insufficient, an audible and visual alarm is triggered and a notification is pushed to the relevant terminal.

[0117] Example:

[0118] Step 1: When the molten iron ladle enters the preset range, the infrared monitoring device 4 detects the ladle. After detecting that the molten iron ladle has entered the detection area and stays for 3 seconds, it sends out a ladle arrival signal. The visible light camera 3 receives the ladle arrival signal and identifies the ladle number. Multiple detections are used to increase the accuracy of ladle number identification.

[0119] Step 2: After the package number is identified, the infrared monitoring device begins to measure the temperature data and simultaneously filters out areas outside the target range.

[0120] Step 3: Deploy a pair of network bridges and a video decoder in the operator's cab. Video data is transmitted to the decoder via the network bridges and displayed in real time on the display screen in the operator's cab. A specially developed PC software client is installed on a computer in the remote control room. The software client can realize configuration functions such as real-time preview, multi-channel video monitoring, detection of start and stop buttons, video image overlay acquisition time and acquisition results, early warning function, recording function, equipment network anomaly alarm, historical query, and export.

[0121] Step 4: Calculate the real-time thickness (H) ∝ T (ladle surface temperature) function relationship using the initial molten iron temperature, ladle surface temperature, and ambient temperature, and correct for errors through multiple calculations. Calculate the ladle wall thickness in different areas using infrared monitoring of real-time temperature data and display the results in real time. Set up a ladle refractory thickness early warning system, define alarm thresholds, and issue warnings for refractory thickness thresholds.

[0122] Step 5: Calculate the functional relationship between ΔH (loss thickness) and a (number of times the ladle has been used) using historical temperature data of the molten iron ladle. Correct the error through multiple calculations, generate a curve showing the change in ladle wall thickness, assess the condition of the molten iron ladle, evaluate its service life, and provide early warnings for potential hazards.

[0123] Step 6: For different series of molten iron ladles, refractory thickness prediction templates can be set and directly called.

[0124] Step 7: Monitoring and display equipment can be flexibly deployed and adjusted to adapt to the site environment using wireless bridges, wireless access points, etc. This case study uses... Figure 3 As shown in the diagram, the display screen is placed in locations where wiring is difficult, such as the aircraft control room.

[0125] It is understood that the present invention has been described through some embodiments, and those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.

Claims

1. A detection system for calculating thickness based on infrared monitoring of molten iron ladle surface temperature data, characterized in that, include: Visible light recognition module: used to acquire images of molten iron ladles and identify the ladle number using a trained neural network model; Infrared temperature monitoring module: includes at least three infrared monitoring devices, respectively arranged on the side and bottom of the molten iron ladle, for collecting surface temperature data; Data processing and analysis module: used to receive and process visible light and infrared data, and perform package number identification, temperature analysis, thickness inversion and lifetime prediction; Early warning and output module: including audible and visual alarm and display device, used to issue alarm signals when thickness abnormality or lifespan warning is detected, and to display monitoring results in real time.

2. The detection system for calculating thickness based on infrared monitoring of molten iron ladle surface temperature data according to claim 1, characterized in that, All infrared temperature monitoring modules are connected to the local area network via a switch, and the data is wirelessly transmitted to the remote control room via a bridge.

3. A method for calculating the thickness of a molten iron ladle using the detection system for calculating thickness based on infrared monitoring of the ladle surface temperature data as described in claim 1 or 2, characterized in that, Includes the following steps: Step 1: Identify the ladle number using a visible light vision algorithm: Step 2: Surface temperature data acquisition and processing: Step 3: Calculate the real-time thickness of the ladle lining based on infrared temperature data: Obtain the outer surface temperature of the ladle through infrared monitoring equipment, establish a physical model between temperature and lining thickness, and calculate the thickness in real time based on this model. Step 4: Calculate the remaining service life of the molten iron ladle based on thickness loss: By tracking the change pattern of thickness with the number of uses, a loss model is established to predict the remaining service life and realize risk warning; Step 5: Tiered Early Warning: The system sets multi-level early warning thresholds. When the real-time thickness (H_a) approaches or the predicted RUL is lower than the set threshold, the system automatically triggers an audible and visual alarm and highlights the early warning information and the predicted number of remaining uses on the display screen to guide maintenance decisions. Step 6: Data display and early warning output.

4. The method for calculating the thickness of a molten iron ladle surface temperature using the detection system according to claim 3, characterized in that, Step 1 includes the following steps: When the molten iron ladle enters the preset range, the infrared monitoring device 4 identifies the ladle. After detecting that the molten iron ladle has entered the detection area and stays for 3 seconds, it sends out a ladle arrival signal. The visible light camera 3 receives the ladle arrival signal and identifies the ladle number. Multiple detections are used to increase the accuracy of ladle number identification. By acquiring real datasets from the field, cropping images using preset bounding boxes to extract target regions, labeling individual digits and organizing the dataset, and finally training a digit detection model based on YOLOv8, a digit detection model is obtained.

5. The method for calculating the thickness of a molten iron ladle surface temperature using the detection system according to claim 3, characterized in that, Step 1 includes: using a visible light camera 3 to acquire images and crop digital regions, using a neural network to recognize characters, calculating the distance between the midpoints of the characters, merging adjacent characters from left to right to form multi-digit numbers, and finally outputting the most frequently occurring number among all recognition results during the tank's stay as the package number.

6. The method for calculating the thickness of a molten iron ladle surface temperature using the detection system according to claim 3, characterized in that, Step 2 includes the following steps: Infrared monitoring equipment collects thermal radiation data from the surface of molten iron ladles, filters and reduces noise to segment the target area, converts the radiation intensity into temperature values, combines blackbody calibration to correct errors, analyzes temperature distribution characteristics, and extracts the highest temperature, average temperature, and coordinates of high-temperature areas, ultimately generating a visualized temperature map and an over-limit warning signal.

7. The method for calculating the thickness of a molten iron ladle surface temperature using the detection system according to claim 3, characterized in that, Step 3 includes the following steps: Step 3.1: Physical Model Establishment: The molten iron ladle is simplified into a multi-layered, flat-walled or cylindrical-walled steady-state heat conduction model. The interior of the ladle lining contains high-temperature molten iron, while the exterior is filled with ambient air. According to Fourier's law of thermal conductivity, the temperature T_s at the measuring point on the outer surface of the ladle has a clear functional relationship with the lining thickness H. For the flat-walled model, the simplified relationship is as follows: ， Where T_s is the measured temperature of the outer surface, T_m is the temperature of the molten iron, T_a is the ambient temperature, H is the real-time thickness of the lining to be determined, k is the thermal conductivity of the lining material, and h is the convective heat transfer coefficient between the outer wall of the lining and the air. Step 3.2: Model parameter determination and calibration: Step 3.2.1: Initial calibration: Before the new ladle is put into use, its initial thickness H_0 and the outer surface temperature T_s0 under stable conditions when filled with molten iron are measured. H_0 and T_s0 are substituted into the above model. Combined with the known T_m and T_a, the effective comprehensive thermal resistance or calibration coefficient under the current working conditions is calculated, and a specific function relationship H = f(T_s, T_m, T_a; k, h) is established. Step 3.2.2: Blackbody calibration: Use a blackbody source set up on site to periodically calibrate the absolute temperature measurement value of the infrared monitoring equipment to ensure the accuracy of the T_s data; Step 3.3: Real-time thickness calculation: When the system is running, the real-time collected T_s, as well as the synchronously acquired T_m and T_a, are substituted into the calibrated functional relationship H = f(T_s, T_m, T_a) to calculate the real-time thickness H corresponding to the measuring point; Step 3.4: Multi-regional data fusion: After calculating the thickness independently for each infrared monitoring point, the real-time thickness H representing the entire molten iron ladle is obtained by taking the average value, minimum value, or weighted value of a specific area.

8. The method for calculating the thickness of a molten iron ladle surface temperature using the detection system according to claim 3, characterized in that, Step 4 includes the following steps: Step 4.1: Data Recording and Sequence Generation: The system records key data for each use cycle of each molten iron ladle, forming a data sequence: number of uses a, and the real-time thickness of the ladle lining H_a calculated after each use; Step 4.2: Calculation of Thickness Loss △H: For the a-th use, the formula for calculating the cumulative thickness loss is: ， Where H_0 is the initial thickness, and H_a is the real-time thickness calculated after the a-th use; Step 4.3: Loss Model Establishment and Fitting: Using the number of uses a as the independent variable and the cumulative thickness loss △H_a as the dependent variable, the historical data points (a, △H_a) are fitted with a curve; Step 4.4: Remaining life prediction and early warning: Safety thickness threshold setting: Based on process safety standards, set the critical thickness at which the molten iron ladle needs to be taken offline for maintenance or scrapped; Remaining useful life prediction: The fitted functional relationship ΔH = g(a) is solved inversely, i.e. , The predicted total number of uses before reaching the critical thickness is calculated, and the remaining service life is: 。 9. The method for calculating the thickness of a molten iron ladle surface temperature using the detection system according to claim 3, characterized in that, Step 5 includes the following steps: When the real-time thickness H_a approaches or the predicted RUL falls below the set threshold, the system automatically triggers an audible and visual alarm and displays a warning message and a prediction of the remaining usage times on the screen to guide maintenance decisions.

10. The method for calculating the thickness of a molten iron ladle surface temperature using the detection system according to claim 3, characterized in that, Step 6 includes the following steps: The control room display screen shows real-time temperature heatmaps, thickness curves, remaining lifespan, and warning information; it also supports historical data query, report export, and video playback. When the thickness is below the set threshold or the remaining lifespan is insufficient, an audible and visual alarm is triggered and a notification is pushed to the relevant terminal.

Images