Visual online thermal imaging method and system for temperature of metal liquid in crystallizer

Abstract

The invention relates to a visual online thermal imaging method and system for the temperature of metal liquid in a crystallizer. A visual online thermal imaging method for the temperature of metal liquid in a crystallizer comprises the steps that S1, an ultrasonic signal corresponding to the temperature of the metal liquid on the inner wall of the crystallizer in a detection area is obtained in real time through an ultrasonic detection device; s2, converting an ultrasonic signal acquired in real time into corresponding temperature information by using an ultrasonic signal conversion thermal image analysis and recognition algorithm; and S3, visually presenting the temperature change rate of the inner wall of the crystallizer by using a thermal imaging technology according to the obtained temperature distribution information of the metal liquid on the inner wall of the crystallizer in the detection area in combination with the pulling speed, the liquid level, the inlet and outlet water temperature of the crystallizer and steel grade production process parameter process data, and visually presenting the temperature distribution, the abnormal change and the development trend of the crystallizer. And reproduction of continuous casting conditions is realized. The visual online thermal imaging method and system fill the blank of realizing continuous casting condition reproduction based on real-time detection of the temperature of the inner wall of the crystallizer at home and abroad.

Description

Technical Field

[0001] The present invention relates to the field of metal liquid detection in the metallurgical industry, and more particularly to a temperature visualization on-line thermal imaging method and system for metal liquid on the inner wall surface of a mold based on ultrasonic detection. Background Art

[0002] The detection of liquid physical parameters in industries such as metallurgy, chemical engineering, and medicine is an essential part of automatic production process control. For the metallurgical industry, due to the high-temperature and enclosed working conditions in the actual production process, the temperature of the inner wall of the mold is invisible. The single-point temperature detection method of the mold based on thermocouples, due to the limited detection accuracy and response speed of the thermocouples, can only play a role in the application of breakout prediction. For the breakout monitoring and prediction system, the existing single thermocouple is installed in the mold as Figure 1 shown. The copper plate 1 is in contact with the spare plate 2. An installation hole 3 is opened on the spare plate 2, and a corresponding installation hole 5 is machined on the copper plate 1. Then the thermocouple 6 is installed in the installation hole 3. The principle of the existing single-sided copper plate for installing thermocouples is as Figure 2 , a number of installation holes 3 and installation grooves 4 for cable entry and exit are machined on the spare plate 2 in rows and columns. Corresponding installation holes are machined on the copper plate 1, and the corresponding number of thermocouples are installed to form a thermocouple array on the single-sided copper plate. The existing installation method of the inner wall temperature detection of the mold for installing thermocouples requires machining installation holes on the copper tube. Too many installation holes will cause damage to the copper tube and the cost is expensive. Moreover, the installation of the thermocouple into the copper tube and the installation hole is by gluing, which will introduce additional air gaps, thus slowing down the response time of the thermocouple to the temperature change of the copper tube. Furthermore, it takes a certain amount of time for the thermocouple probe and the copper body nearby to thermally equilibrate. This results in a further delay in the corresponding response time when the probe reaches thermal equilibrium with the surrounding copper body. Since each TC probe corresponds to a cable outlet, the installation is complex, and the cable must be disconnected every time the mold is replaced. The thermocouple can only detect the temperature of the copper body at the probe position, that is, the internal temperature of the copper tube wall thickness, and cannot directly contact the molten steel on the inner wall surface of the copper tube. As the distance between the thermocouple and the hot surface increases, the ability of the thermocouple to detect temperature fluctuations will decrease.

[0003] Currently, there is no comprehensive system in the continuous casting field of the steel industry worldwide that can completely reproduce the continuous casting conditions. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the present invention proposes a temperature visualization on-line thermal imaging method for metal liquid in a mold, which can realize the real-time detection of the temperature change of metal liquid during the metal smelting process and completely reproduce the continuous casting conditions.

[0005] The present invention simultaneously provides a temperature visualization online thermal imaging system for molten metal on the inner wall surface of a mold based on ultrasonic detection. An ultrasonic detection device is arranged in the mold area of the metallurgical casting platform to obtain ultrasonic signals corresponding to the temperature of the molten metal on the inner wall of the mold in the area; through an ultrasonic signal conversion thermal image analysis module, the ultrasonic signals collected in real time are converted into thermal images for real-time preprocessing, decoding, and storage to form a data set of thermal images, realizing the real-time detection of the temperature change of the molten metal during the metal smelting process and realizing the reproduction of continuous casting conditions.

[0006] The present invention fills the gap at home and abroad in realizing the reproduction of continuous casting conditions based on the real-time detection of the inner wall temperature of the mold.

[0007] The technical solution adopted by the present invention: A method for visualizing the temperature of molten metal in a mold online by thermal imaging, and its implementation process includes the following steps: Step S1, using an ultrasonic detection device to obtain ultrasonic signals corresponding to the temperature of the molten metal on the inner wall of the mold in the detection area in real time; Step S2, using an ultrasonic signal conversion thermal image analysis and recognition algorithm to convert the ultrasonic signals collected in real time into corresponding temperature information; Step S3, combining the temperature distribution information of the molten metal on the inner wall of the mold in the obtained detection area with process data such as casting speed, liquid level, inlet and outlet water temperatures of the mold, and steel grade production process parameters, and using thermal imaging technology to visually present the temperature change rate of the inner wall of the mold, intuitively present the temperature distribution, abnormal changes, and development trend of the mold, present the change in the position of the molten steel height, and realize the reproduction of continuous casting conditions.

[0008] In the method for visualizing the temperature of molten metal in a mold online by thermal imaging, in step S2, using an ultrasonic signal conversion thermal image analysis and recognition algorithm, the ultrasonic signals collected in real time are converted into a thermal image of the mold temperature for real-time preprocessing, decoding, and recognition analysis to form a data set of thermal images when the molten metal at the meniscus of the mold in the mold area of the metallurgical casting platform to be detected is formed; the real-time thermal image data set is processed to obtain a characteristic thermal image, and the significant features of the thermal image are identified and analyzed and converted into temperature data information in real time.

[0009] The ultrasonic signal conversion thermal image analysis and recognition algorithm includes: An ultrasonic signal conversion module that processes the ultrasonic signals detected in real time and converts them into the temperature of the molten metal in the mold; An algorithm recognition module that preprocesses the thermal image data set in real time and obtains a characteristic thermal image, identifies and analyzes the significant features of the thermal image, and converts them into temperature data information in real time; When it is detected that the temperature data information exceeds the set temperature threshold, or when it is detected that there is a rapid temperature rise in a specific area on the surface of the molten metal in the mold region, a temperature warning signal is output.

[0010] The algorithm recognition module recognizes, analyzes, preprocesses, and fits the feature thermal image to form an accurate simulation and test standard data set of the feature thermal image; it performs real-time recognition and analysis on the preprocessed and decoded thermal image obtained in real time and the accurate simulation and test standard data set of the feature thermal image, and converts it into temperature data information in real time.

[0011] The ultrasonic signal conversion thermal image analysis recognition algorithm includes: The thermal image preprocessing and fitting module is used to vertically plot the temperature of the mold in the meniscus region for the real-time input. By plotting the temperature values of each mold volume in front of the sensor and the temperature values at their respective vertical positions along the side of the mold, it calculates the local thermal temperature distribution thermal image of the mold region in front of the sensor, preprocesses the thermal image data, and fits the preprocessed thermal image data. The thermal image feature acquisition module is used to acquire the features of the preprocessed and fitted thermal image.

[0012] A temperature visualization online thermal imaging system for the molten metal on the inner wall surface of a mold based on ultrasonic detection, which includes: An ultrasonic detection device that acquires ultrasonic signals in the meniscus region of the mold in real time; The ultrasonic signal conversion thermal image analysis module processes the ultrasonic signals acquired in real time, converts them into a temperature thermal image data set of the molten metal in the meniscus of the mold region during the forming process, and obtains the significant features of the feature thermal image through decoding and recognition analysis, and converts it into temperature data information in real time; A visualization imaging device that realizes the reproduction of continuous casting conditions; An alarm module and device that receives temperature warning information in real time. When it is detected that the temperature data information exceeds the set temperature threshold, or when it is detected that there is a rapid temperature rise in a specific area on the surface of the molten metal in the mold region, a temperature warning signal is output.

[0013] For the temperature visualization online thermal imaging system of the molten metal on the inner wall surface of the mold based on ultrasonic detection, the ultrasonic detection device is arranged in the mold area of the metallurgical casting platform, and faces any outer side or both upper sides of the metallurgical casting platform mold towards the water tank or water jacket in the meniscus region, and is used to acquire the real-time ultrasonic signals of the temperature of the molten metal in the metallurgical casting platform mold.

[0014] The temperature visualization online thermal imaging system for molten metal on the inner wall surface of the mold based on ultrasonic detection, the ultrasonic detection device is simultaneously arranged in the mold area of the metallurgical casting platform, and faces the water tank or water jacket in the upper part of the four sides of the metallurgical casting platform mold towards the meniscus area, for obtaining real-time ultrasonic signals of the temperature of the molten metal in the metallurgical casting platform mold.

[0015] The temperature visualization online thermal imaging system for molten metal on the inner wall surface of the mold based on ultrasonic detection, the ultrasonic detection device is arranged in the mold area of the metallurgical casting platform, and faces the meniscus of the metallurgical casting platform mold, and is arranged perpendicular to the meniscus; what the sensor measures is the "thermal temperature line" of the copper tube in the meniscus area, that is, the thermal temperature distribution of the copper tube along a vertical line in front of the sensor.

[0016] The temperature visualization online thermal imaging system for molten metal on the inner wall surface of the mold based on ultrasonic detection, the ultrasonic signal conversion thermal image analysis module processes the ultrasonic signals, converts the ultrasonic signals into the temperature of the copper tube, differentiates the temperature difference with colors, the copper tube temperature is vertically drawn in the meniscus area, by plotting the temperature values of each copper tube volume in front of the sensor and the temperature values of their respective vertical positions along the side of the copper tube, calculates the local thermal temperature distribution of the copper tube area in front of the sensor, thereby obtaining the thermal temperature profile; adopts a proprietary algorithm based on mathematical transformation and finite element method (FEM) modeling, extracts the meniscus position from the shape of the thermal profile, reconstructs the meniscus thermal curve, and performs 3D drawing, thereby realizing the control of the molten steel liquid level and the online automatic detection of sticking events. Advantages of the invention: 1. The present invention realizes the real-time detection of the temperature change and its height position of the molten metal during the metal smelting process in a non-contact manner, realizes the reproduction of continuous casting conditions, and outputs a temperature warning signal under specific conditions, realizes the real-time warning of sticking quality events in production, and guarantees production safety and production quality. So far, the production process of the mold in the industry belongs to a "black box", and there is no effective means to break through and present the condition of the mold meniscus in real time online, and realize the reproduction of continuous casting conditions. The present invention fills the gap at home and abroad in realizing the reproduction of continuous casting conditions based on the real-time detection of the inner wall temperature of the mold.

[0017] 2. The present invention uses ultrasonic non-contact online measurement of the temperature of the inner wall of the mold, thereby obtaining the molten steel liquid level fluctuation corresponding to the instantaneous temperature change, and combining the temperature distribution information within the area of the inner wall area of the obtained mold, with process data of production process parameters such as casting speed, liquid level, inlet and outlet water temperature of the mold, steel grade, etc., uses thermal imaging technology to visually present the temperature change rate of the inner wall of the mold. This method not only intuitively presents the temperature distribution, abnormal changes and development trends of the mold, presents the change of the molten steel height position, gives a real-time warning of sticking quality events in production, but also realizes the reproduction of continuous casting conditions.

[0018] 3. The temperature visualization online thermal imaging system for molten metal on the inner wall surface of the mold based on ultrasonic detection of the present invention reduces the installation and maintenance workload, is non-contact, does not require invasive and destructive processing of the copper tube, and reduces the processing of the water jacket. It is applicable to any billet type. There is no need to use different detection means according to different billet types. The external shape design of the sensor is modular, and multiple modules can be installed on the mold, on one or more surfaces, depending on the usage requirements and the size of the mold.

[0019] 3. The present invention uses a non-contact ultrasonic detector for temperature measurement. By not contacting the copper plate and performing a small amount of processing on the water jacket, the temperature detection of molten steel at different positions of the copper plate can be realized, reducing the processing workload and the installation and maintenance workload. It greatly reduces the enterprise cost and the equipment usage and maintenance amount. It breaks through the limitations of the traditional liquid level detection technology, further improves the detection response speed, and can cover all application environments.

[0020] 4. The temperature visualization online thermal imaging method and system for molten metal on the inner wall surface of the mold based on ultrasonic detection of the present invention has the "actual probe" as the ultrasonic velocity itself, realizing zero "probe preheating delay". At the same time, the ultrasonic probe detects the average temperature of the volume near the entire copper body, including the contribution of the hot surface of the copper tube in contact with the molten steel. Therefore, the temperature detection response speed is fast and the time is short. It provides a real-time "thermal temperature profile" of the copper tube at the meniscus, which can be used to detect the lubrication effect under different steel grades and different continuous casting drawing speeds, improving the surface quality of the billet. There is no need to perform full-layer calibration before each continuous casting operation, saving time and personnel activities. Description of the Drawings

[0021] Figure 1 The figure shows a schematic diagram of installing a single thermocouple into the mold in the prior art; Figure 2 The figure shows a schematic diagram of the principle of installing a thermocouple on a single-sided copper plate in the prior art; Figure 3 The figure shows a schematic diagram of the installation method of the ultrasonic detection device adopted by the present invention; Figure 4 The figure shows a schematic cross-sectional installation method of installing the ultrasonic detection device adopted by the present invention into the water jacket; Figure 5 The figure shows a schematic diagram of the cable wiring method adopted by the present invention; Figure 6 The figure shows the composition of the temperature visualization online thermal imaging system of the present invention. Detailed Embodiments

[0022] To make the technical concept and advantages for the realization of the invention of the present invention clearer and more understandable, the technical solution of the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the following embodiments are only for explaining and illustrating the preferred embodiments of the present invention, and should not be regarded as and do not constitute a limitation on the scope of patent protection required by the present invention.

[0023] Embodiment 1: The method for visualizing the online thermal imaging of the temperature of the molten metal in the mold of the present invention includes the following steps: Step S1, using an ultrasonic detection device to obtain in real time the ultrasonic signal corresponding to the temperature of the molten metal on the inner wall of the mold in the detection area; Step S2, using an ultrasonic signal conversion thermal image analysis and recognition algorithm to convert the ultrasonic signal collected in real time into corresponding temperature information; Step S3, combining the obtained temperature distribution information of the molten metal on the inner wall of the mold in the detection area with process data such as casting speed, liquid level, inlet and outlet water temperatures of the mold, and steel grade production process parameters, and using thermal imaging technology to visually present the temperature change rate of the inner wall of the mold, intuitively present the temperature distribution, abnormal changes and development trends of the mold, present the change of the molten steel height position, and realize the reproduction of continuous casting conditions.

[0024] In step S2, using an ultrasonic signal conversion thermal image analysis and recognition algorithm, the ultrasonic signal collected in real time is converted into a thermal image of the mold temperature, and preprocessing, decoding and recognition analysis are carried out in real time to form a thermal image data set of the temperature of the molten metal at the meniscus of the mold in the mold area of the detected metallurgical casting platform; the real-time thermal image data set is processed to obtain a characteristic thermal image, and the significant features of the thermal image are identified and analyzed and converted into temperature data information in real time.

[0025] Embodiment 2: The difference between this embodiment and Embodiment 1 is that: further, in step S2, the ultrasonic signal conversion thermal image analysis and recognition algorithm includes: An ultrasonic signal conversion module that processes the ultrasonic signal detected in real time and converts it into the temperature of the molten metal in the mold; An algorithm recognition module that preprocesses the thermal image data set in real time and obtains a characteristic thermal image, identifies and analyzes the significant features of the thermal image, and converts them into temperature data information in real time; When it is detected that the temperature data information exceeds the set temperature threshold, or when it is detected that a rapid temperature rise occurs in a specific area on the surface of the molten metal in the mold area, a temperature warning signal is output.

[0026] The algorithm recognition module recognizes and analyzes the preprocessed and fitted characteristic thermal image to form an accurate simulation and test standard data set of the characteristic thermal image; and performs real-time recognition and analysis on the preprocessed and decoded thermal image obtained in real time and the accurate simulation and test standard data set of the characteristic thermal image, and converts it into temperature data information in real time.

[0027] Embodiment 3: The method for visualizing the temperature of molten metal in a mold in real time in this embodiment is different from Embodiments 1 and 2 in that the ultrasonic signal conversion thermal image analysis recognition algorithm includes: The thermal image preprocessing and fitting module is used to vertically plot the temperature of the mold in the meniscus region for the real-time input. By plotting the temperature values of each mold volume in front of the sensor and the temperature values at their respective vertical positions along the side of the mold, the local thermal temperature distribution thermal image of the mold region in front of the sensor is calculated, the thermal image data is preprocessed, and the preprocessed thermal image data is fitted. The thermal image feature acquisition module is used to acquire the features of the preprocessed and fitted thermal image.

[0028] Embodiment 4: Refer to Figure 6 This embodiment is a temperature visualization on-line thermal imaging system for molten metal on the inner wall surface of a mold based on ultrasonic detection. The system includes: The ultrasonic detection device acquires the ultrasonic signal in the meniscus region of the mold in real time; The ultrasonic signal conversion thermal image analysis module performs signal processing on the ultrasonic signal acquired in real time, converts it into a temperature thermal image data set of the molten metal forming the meniscus in the mold region, and obtains the significant features of the characteristic thermal image through decoding and recognition analysis, and converts it into temperature data information in real time; The visualization imaging device realizes the reproduction of continuous casting conditions; The alarm module and device receive temperature warning information in real time. When it is detected that the temperature data information exceeds the set temperature threshold, or when it is detected that there is a rapid temperature rise in a specific area on the surface of the molten metal in the mold region, a temperature warning signal is output.

[0029] The ultrasonic detection device is arranged in the mold area of the metallurgical casting platform, and faces the meniscus of the metallurgical casting platform mold, and is arranged perpendicular to the meniscus; the sensor measures the "thermal temperature line" of the copper tube in the meniscus region, that is, the thermal temperature distribution of the copper tube in front of the sensor along a vertical line.

[0030] The sensor generates multiple ultrasonic beams. After these ultrasonic beams are propagated to the copper tube wall and reflected, they are finally received by the sensor. Since the ultrasonic velocity in the copper tube depends on the thermal state of the medium, special signal processing is performed on the ultrasonic signal. By converting the ultrasonic signal into the copper tube temperature (differential temperature is distinguished by color), the copper tube temperature is vertically plotted in the meniscus region. By plotting the temperature values of each copper tube volume in front of the sensor and the temperature values at their respective vertical positions along the side of the copper tube, the local thermal temperature distribution of the copper tube area in front of the sensor is calculated, thereby obtaining the thermal temperature profile. Through a proprietary algorithm based on mathematical transformation and finite element method (FEM) modeling, the meniscus position is extracted from the shape of the thermal profile, the meniscus thermal curve is reconstructed, and 3D plotting is performed, so as to realize the control of the molten steel liquid level, and online automatic detection of sticking events, real-time alarm, and sending the sticking detection alarm to the upper-level PLC, enabling the operator to take appropriate corrective measures, such as reducing the continuous casting drawing speed, providing sufficient time for the solidified solid shell to form again, and closing the fracture, thereby avoiding the occurrence of steel leakage accidents.

[0031] The online thermal imaging system for visualizing the temperature of the molten metal on the inner wall surface of the mold based on ultrasonic detection in the present invention reduces the amount of installation and maintenance work. It is non-contact, does not require invasive and destructive processing of the copper tube, and reduces the processing of the water jacket. It is applicable to any billet type. There is no need to use different detection means for different billet types. The outer shape design of the sensor is modular, and multiple modules can be installed on the mold, on one or more surfaces, depending on the usage requirements and the size of the mold.

[0032] The ultrasonic detection device is arranged in the mold area of the metallurgical casting platform mold, and faces the water tank or the water jacket in the upper part of any outer side of the metallurgical casting platform mold towards the meniscus area, for acquiring the real-time ultrasonic signal of the liquid temperature when the molten metal of the metallurgical casting platform mold is formed.

[0033] The ultrasonic detection device is simultaneously arranged in the mold area of the metallurgical casting platform mold, and faces the water tank or the water jacket in the upper parts of both sides of the metallurgical casting platform mold towards the meniscus area, for acquiring the real-time ultrasonic signal when the molten metal of the metallurgical casting platform mold is formed.

[0034] The installation method is as Figure 3 (the original equipment needs to be modified): The copper tube a is installed into the water jacket b. Both the copper tube a and the water jacket b are installed on the outer shell c. A slot e is opened along the water jacket b, and threaded mounting holes f are machined around the slot e. A flange d is installed on the outer shell c. The installation of the sensor g into the water jacket b is shown in Figure 4 , and the front end face of the sensor g is flush with the inner wall of the water jacket b. The sensor cable h is connected to the sockets inside the sensor g and inside the flange d (not marked in the figure) as shown in Figure 5 .

[0035] The method and system of the present invention realize the real-time detection of the temperature change of molten metal during the metal smelting process in a non-contact manner, reproduce the continuous casting conditions, and output a temperature warning signal under specific conditions to realize the real-time warning of the bonding quality event in production, ensuring production safety and production quality.

[0036] During the working process of the whole device: after the sensor is fixedly installed, the gap between the water jacket and the copper tube will not affect the water flow. The water flow from bottom to top is used to cool the copper tube, ensuring uniform heat dissipation of the copper tube.

[0037] The ultrasonic signal conversion thermal image analysis device includes an algorithm recognition module, which is used to preprocess the thermal image data set in real time and obtain the characteristic thermal image, identify and analyze the significant features of the obtained thermal image and convert them into temperature data information in real time, and output a temperature warning signal. The alarm signal includes an audible and visual alarm signal, and different audio intensity signals can be used to represent different alarm level signals, and red, yellow, and blue light signals represent different alarm level signals.

[0038] This system can detect the tiny temperature change of the measured volume because ultrasonic waves pass through the entire volume of the copper tube from the inner wall to the outer wall at a high frequency and ultrasonic speed. The sensor measures the temperature value hundreds of times per second, providing real-time information about what is happening on the hot surface (including the molten steel level fluctuation corresponding to the instantaneous temperature change), providing real-time information about the thermal profile of the mold and the temperature trend of the copper tube, providing quality tracking and benchmark of the billet, indicating the part where the temperature deviation or sudden temperature peak of the product occurs, which may be caused by bonding and sticking of steel, leakage of steel, abnormal liquid level fluctuation, incorrect position of the submerged entry nozzle (SEN), incorrect lubrication, etc. These key information can be used as indicators for correct solidification, stable process, and optimization of continuous casting parameter settings. By analyzing this temperature data, continuous casting practice and process variables can be optimized, so as to realize reproducible continuous casting conditions and improve the quality of the billet.

[0039] This detection system can realize the real-time detection of the temperature change of molten metal during the metal smelting process, reproduce the continuous casting conditions, and ensure production safety and production quality.

[0040] The above is only the preferred embodiment of the present invention and does not constitute a limitation to the present invention. Under the guidance of the prior art, those skilled in the art can make other modifications to the implementation of the present invention without creative labor. Any modification made within the spirit and principle of the present invention or any simple replacement or equivalent replacement using the conventional technical means in the art shall be included in the protection scope of the present invention.

Claims

1. A method for visualizing the temperature of liquid metal in a crystallizer by online thermal imaging, characterized in that: The steps include: Step S1, using an ultrasonic detection device to obtain in real time an ultrasonic signal corresponding to the temperature of the metal liquid on the inner wall of the crystallizer in the detection area; Step S2, using an ultrasonic signal conversion thermal image analysis and recognition algorithm to convert the real-time collected ultrasonic signal into corresponding temperature information; Step S3, combining the obtained temperature distribution information of the metal liquid on the inner wall of the crystallizer in the detection area with the casting speed, liquid level, water inlet and outlet temperature of the crystallizer, and process data of steel production process parameters, uses thermal imaging technology to visualize the temperature change rate of the inner wall of the crystallizer, intuitively present the crystallizer temperature distribution, abnormal changes and development trends, and present the changes in the height position of the molten steel, so as to reproduce the continuous casting conditions.

2. The method for visualizing the temperature of liquid metal in a crystallizer according to claim 1, characterized in that: In step S2, the ultrasonic signal is converted into a thermal image analysis and recognition algorithm to convert the real-time collected ultrasonic signal into a crystallizer temperature thermal image for real-time preprocessing, decoding and recognition analysis to form a thermal image data set of the metal liquid temperature of the meniscus of the crystallizer in the crystallizer area of ​​the metallurgical casting platform; the real-time thermal image data set is processed to obtain a characteristic thermal image, and the significant features of the thermal image are recognized and analyzed and converted into temperature data information in real time.

3. The method for visualizing the temperature of liquid metal in a crystallizer online thermal imaging according to claim 2, characterized in that: In step S2, the ultrasonic signal is converted into a thermal image analysis and recognition algorithm, including: The ultrasonic signal conversion module processes the ultrasonic signal detected in real time and converts it into the temperature of the metal liquid in the crystallizer; An algorithm recognition module pre-processes the thermal image data set and obtains characteristic thermal images in real time, recognizes and analyzes the significant features of the thermal images, and converts them into temperature data information in real time; When it is detected that the temperature data information exceeds a set temperature threshold, or when it is detected that a rapid temperature rise occurs in a specific area on the surface of the metal liquid in the crystallizer area, a temperature warning signal is output.

4. The method for visualizing the temperature of liquid metal in a crystallizer online thermal imaging according to claim 3, characterized in that: The algorithm recognition module recognizes and analyzes the characteristic thermal image after preprocessing and fitting, and forms an accurate simulation and test standard data set of the characteristic thermal image; The real-time acquisition, pre-processing and decoding of the thermal image and the accurate simulation and test standard data set of the characteristic thermal image are used for real-time identification and analysis, and converted into temperature data information in real time.

5. The method for visualizing the temperature of liquid metal in a crystallizer online thermal imaging according to claim 2, 3 or 4, characterized in that: The ultrasonic signal conversion thermal image analysis and recognition algorithm includes: A thermal image preprocessing and fitting module is used to vertically draw the real-time input crystallizer temperature in the meniscus area, calculate the local thermal temperature distribution thermal image of the crystallizer area in front of the sensor by drawing the temperature value of each crystallizer volume in front of the sensor and the temperature value of each vertical position along the side of the crystallizer, preprocess the thermal image data, and fit the preprocessed thermal image data; The thermal image feature acquisition module is used to acquire the thermal image features after preprocessing and fitting.

6. An online thermal imaging system for visualizing the temperature of liquid metal on the inner wall of a crystallizer based on ultrasonic detection according to claim 1, characterized in that: The system comprises: Ultrasonic detection device, to obtain ultrasonic signals from the meniscus area of ​​the crystallizer in real time; The ultrasonic signal conversion thermal image analysis module processes the ultrasonic signal acquired in real time and converts it into a temperature thermal image data set of the metal liquid in the meniscus of the crystallizer area during molding. Through decoding and recognition analysis, the significant features of the characteristic thermal image are obtained and converted into temperature data information in real time; Visual imaging device to reproduce continuous casting conditions; The alarm module and device receive temperature warning information in real time, and output a temperature warning signal when it is detected that the temperature data information exceeds the set temperature threshold, or when a rapid temperature rise is detected in a specific area on the surface of the metal liquid in the crystallizer area.

7. The temperature visualization online thermal imaging system of the metal liquid on the inner wall of the crystallizer based on ultrasonic detection according to claim 6 is characterized in that: The ultrasonic detection device is arranged in the mold area of ​​the metallurgical casting platform crystallizer, and is facing any outer side or the upper part of both sides of the metallurgical casting platform crystallizer towards the water tank or water jacket in the meniscus area, and is used to obtain real-time ultrasonic signals when the metal liquid is formed in the metallurgical casting platform crystallizer.

8. The temperature visualization online thermal imaging system of the metal liquid on the inner wall of the crystallizer based on ultrasonic detection according to claim 6 is characterized in that: The ultrasonic detection device is also arranged in the mold area of ​​the metallurgical casting platform crystallizer, and is facing the water tank or water jacket in the meniscus area on the upper four sides of the metallurgical casting platform crystallizer, so as to obtain the real-time ultrasonic signal of the liquid temperature when the metal liquid of the metallurgical casting platform crystallizer is formed.

9. The temperature visualization online thermal imaging system for metal liquid on the inner wall of a crystallizer based on ultrasonic detection according to claim 6, 7 or 8, characterized in that: The ultrasonic detection device is arranged in the crystallizer mold area of ​​the metallurgical casting platform, facing the meniscus of the crystallizer of the metallurgical casting platform, and is arranged perpendicular to the meniscus; the sensor measures the "thermal temperature line" of the copper tube in the meniscus area, that is, the thermal temperature distribution of the copper tube along a vertical line in front of the sensor.

10. The temperature visualization online thermal imaging system of the metal liquid on the inner wall of the crystallizer based on ultrasonic detection according to claim 9 is characterized in that: The ultrasonic signal conversion thermal image analysis module processes the ultrasonic signal and converts it into the temperature of the copper tube. The temperature difference is distinguished by color. The temperature of the copper tube is plotted vertically in the meniscus area. By plotting the temperature value of each copper tube volume in front of the sensor and the temperature value of each vertical position along the side of the copper tube, the local thermal temperature distribution of the copper tube area in front of the sensor is calculated, thereby obtaining the thermal temperature profile. A proprietary algorithm based on mathematical conversion and finite element modeling is used to extract the meniscus position from the shape of the thermal profile, reconstruct the meniscus thermal curve, and perform 3D drawing, thereby achieving the control of the molten steel level and online automatic detection of viscosity events.

Images