Casting ladle temperature real-time monitoring system and method based on multispectral fusion

Abstract

The invention discloses a casting ladle temperature real-time monitoring system and method based on multispectral fusion, and relates to the technical field of metallurgical casting. The system comprises a binocular temperature measurement thermal infrared imager group, a control box, a processing server and a grading alarm terminal. The measurement method comprises the following steps: S1, arranging a binocular temperature measurement thermal infrared imager group in a temperature measurement area of a casting ladle, and covering the surface of the casting ladle at multiple visual angles; s2, synchronously acquiring an infrared image and a visible light image of the surface of the casting ladle by using the binocular temperature measurement infrared thermal imager set arranged in the step S1; s3, performing multispectral data fusion processing according to the infrared image and the visible light image obtained in the step S2, and comprehensively utilizing infrared temperature distribution information and visible light structure texture information to obtain a space temperature field of the surface of the casting ladle; and S4, carrying out temperature anomaly detection on the surface of the casting ladle according to the space temperature field obtained in the step S3, and carrying out graded alarm according to an abnormal temperature amplitude and distribution characteristics. According to the method, the problem of monitoring blind areas existing in traditional single-point temperature measurement is avoided, the global measurement error in the high-temperature environment is reduced, accurate identification of the abnormal surface temperature of the casting ladle can be achieved, real-time early warning is supported, and the occurrence rate of high-temperature accidents is reduced.

Description

Technical Field

[0001] This invention relates to the field of metallurgical casting technology, and in particular to a real-time monitoring system and method for ladle temperature based on multispectral fusion. Background Technology

[0002] As a core piece of equipment in the metallurgical casting process, the uniformity and stability of the surface temperature field of the ladle directly affect production safety, energy efficiency, and product quality. Localized overheating caused by refractory layer damage may lead to ladle burn-through accidents, while abnormal temperature distribution will exacerbate molten metal oxidation and reduce the yield of castings.

[0003] Thermocouple temperature measurement mainly involves directly measuring the local temperature of the ladle surface and lining by attaching thermocouples, and converting the temperature signal into an electrical signal output using the Seebeck effect. It has the advantage of high single-point measurement accuracy, but the monitoring area of ​​a single thermocouple is small, making it impossible to capture the global temperature distribution on the ladle surface, and it is inconvenient to deploy in mobile ladles. Summary of the Invention

[0004] To address the shortcomings of the aforementioned technologies, the present invention aims to provide a real-time monitoring system and method for ladle temperature based on multispectral fusion, thereby solving the problem that existing mobile ladles are unable to achieve continuous global temperature monitoring of the ladle surface under high-temperature conditions, thus avoiding ladle penetration accidents caused by local overheating.

[0005] To achieve the above objectives, this invention provides a real-time ladle temperature monitoring system and method based on multispectral fusion. The system includes a binocular thermometric infrared thermal imager group, a control box, a processing server, and a graded alarm terminal. The measurement method includes the following steps:

[0006] S1. A binocular thermometric infrared thermal imager group is set up in the ladle temperature measurement area to cover the ladle surface from multiple angles;

[0007] S2. Use the binocular thermometric infrared thermal imager group arranged in step S1 to simultaneously acquire infrared and visible light images of the ladle surface;

[0008] S3. Perform multispectral data fusion processing on the infrared and visible light images obtained in step S2, and comprehensively utilize the infrared temperature distribution information and visible light structure and texture information to obtain the spatial temperature field of the ladle surface;

[0009] S4. Based on the spatial temperature field obtained in step S3, perform temperature anomaly detection on the surface of the ladle and issue graded alarms according to the abnormal temperature amplitude and distribution characteristics.

[0010] Step S1 specifically involves: arranging at least three binocular thermometric infrared thermal imagers at equal intervals along the circumference of the ladle measurement area. Each binocular thermometric infrared thermal imager is arranged in a ring around the center of the ladle, with the included angle between adjacent imagesrs being equal, so that their field of view covers the surface of the ladle and ensures that an overlapping area is formed between adjacent fields of view.

[0011] The thermal imaging acquisition in step S2 specifically involves: allocating an independent thread to each binocular infrared thermal imager to achieve multi-threaded acquisition; the control box sending a synchronous trigger signal to each binocular temperature-measuring infrared thermal imager through a programmable logic controller to enable them to start image acquisition simultaneously; and acquiring the corresponding infrared temperature data, infrared thermal imaging image, and visible light image through the interface program.

[0012] The multispectral fusion method is as follows: based on computer vision methods, the infrared thermal imaging image and the visible light image are spatially registered. The mapping relationship between the two images is established through feature extraction and feature matching. The registered image is segmented to extract the target area of ​​the ladle. Then, the infrared temperature information and the visible light structure information are fused to construct the spatial temperature field of the ladle surface.

[0013] The specific method for detecting temperature anomalies is as follows: statistical analysis of the spatial temperature field is performed to extract the characteristic parameters of the surface temperature. The characteristic parameters include at least the maximum temperature, minimum temperature, average temperature and temperature dispersion. A dynamic threshold model is established based on historical temperature data to identify the abnormal state of the ladle surface temperature. Multi-level alarm strategies are set according to different degrees of anomaly to achieve graded early warning output.

[0014] Based on the above technical solutions, this invention employs infrared thermal imaging technology to achieve non-contact measurement of the ladle surface temperature. It acquires the thermal radiation information of the object under test through online infrared temperature measurement, converts it into temperature data, and realizes real-time acquisition and visualization of temperature information under high-temperature conditions. System calibration is performed using a standard blackbody source, and information fusion is achieved through multispectral data processing algorithms, effectively improving the accuracy of the temperature measurement results.

[0015] Compared with existing technologies, the advantages of this invention are: it upgrades the ladle temperature monitoring method at the system level, transforming it from local discrete measurement to full-field temperature field perception and intelligent analysis, improving the real-time performance and accuracy of temperature anomaly detection, and providing key technical support for ladle health status assessment, risk warning, and intelligent control of the casting process, thus having significant engineering application value. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart of the method of the present invention;

[0018] Figure 2 This is a system diagram of an embodiment of the present invention;

[0019] Figure 3 This is a lens distribution diagram of an industrial binocular infrared thermal imager according to an embodiment of the present invention;

[0020] Figure 4 This is a flowchart of an embodiment of the present invention; Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0023] As shown in the attached diagram, this embodiment provides a real-time monitoring system for ladle temperature based on multispectral fusion. The system includes a binocular thermometric infrared thermal imager group 1, a control box 2, a processing server 4, and a graded alarm terminal 5. The binocular thermometric infrared thermal imager group 1 includes multiple binocular thermometric infrared thermal imagers 6, each including a thermal imaging lens 7 and a visible light lens 8. The multiple binocular thermometric infrared thermal imagers 6 are arranged around the ladle detection area and connected to the processing server 4 via the control box 2. When the transport vehicle carrying the ladle 3 passes through the binocular thermometric infrared thermal imager group 1, each binocular thermometric infrared thermal imager 6 simultaneously images and acquires images of the ladle surface. The processing server 4 receives the infrared and visible light images acquired by the binocular thermometric infrared thermal imager group 1, performs spatial registration of the infrared and visible light images, establishes a mapping relationship through feature point matching, and segments the registered image into target regions to separate the ladle 3 from the background area, thereby obtaining the spatial temperature distribution information of the ladle surface.

[0024] In this embodiment, the binocular temperature-measuring infrared thermal imager group 1 includes three binocular temperature-measuring infrared thermal imagers 6. These three imagers are arranged in a ring around the circumference, with the center of the ladle measurement area as the reference point. The angle between any two adjacent imagers is 120°, and their monitoring areas intersect at the center of the measurement area. When the transport vehicle carrying the ladle 3 passes the center of the measurement area, the ladle surface is within the effective field of view of the three binocular temperature-measuring infrared thermal imagers 6, achieving simultaneous temperature measurement from multiple perspectives.

[0025] The server 4 processes the visible light images captured by the visible light lens 8 to determine whether there are any pouring ladle targets in the monitoring area. In order to achieve standardized management of the pouring ladles, the ladle numbers are identified for easy recording.

[0026] When the ladle 3 enters the monitoring center area of ​​the three binocular thermometric infrared thermal imagers 6, the binocular thermometric infrared thermal imager group 1 starts image acquisition. The processing server 4 reads the infrared thermal imaging video stream and the visible light video stream data respectively, and analyzes them frame by frame. A temperature distribution image of the ladle surface is generated through multispectral image fusion processing. Combined with the temperature data matrix, it is determined whether there are abnormal areas on the ladle surface that exceed the preset safe temperature threshold. For the detected abnormal areas, they are marked in the image and alarm information is generated. Subsequently, the processing server 4 synthesizes the processing results into a video stream and pushes it to the corresponding port for display or storage.

[0027] The core of the temperature anomaly detection technology lies in acquiring a temperature data matrix for statistical analysis. Abnormal temperature states are identified through extreme value analysis, mean calculation, and temperature dispersion assessment. The anomaly judgment threshold is adaptively updated based on historical operating data. A multi-level alarm mechanism is then constructed: a level one warning is triggered when the ladle surface temperature deviates from the normal fluctuation range; a level two alarm is triggered when the abnormal state persists; and a level three emergency alarm is triggered when the temperature exceeds the critical safety threshold.

[0028] In this embodiment, referring to the accompanying drawings, the processing server 4 pre-sets a safe temperature threshold for the ladle surface. When the ladle 3 enters the measurement area with the transport vehicle, the system triggers multiple binocular thermometric infrared thermal imagers 6 to start synchronous acquisition and obtain corresponding data. The processing server 4 fuses the data from each binocular thermometric infrared thermal imager to generate a spatial temperature field on the ladle surface. The system compares and analyzes the spatial temperature field with the preset safe temperature threshold. When it determines that there is a temperature exceeding the limit area, it triggers an alarm of the corresponding level according to the degree of abnormality, and records and stores the spatial location of the abnormal area and the alarm information.

Claims

1. A real-time monitoring system and method for ladle temperature based on multispectral fusion, characterized in that, The system includes a binocular thermometric infrared thermal imager assembly, a control box, a processing server, and a tiered alarm terminal. The measurement method includes the following steps: S1. A binocular thermometric infrared thermal imager group is set up in the ladle temperature measurement area to cover the ladle surface from multiple angles; S2. Use the binocular thermometric infrared thermal imager group arranged in step S1 to simultaneously acquire infrared and visible light images of the ladle surface; S3. Perform multispectral data fusion processing on the infrared and visible light images obtained in step S2, and comprehensively utilize the infrared temperature distribution information and visible light structure and texture information to obtain the spatial temperature field of the ladle surface; S4. Based on the spatial temperature field obtained in step S3, perform temperature anomaly detection on the surface of the ladle and issue graded alarms according to the abnormal temperature amplitude and distribution characteristics.

2. The real-time ladle temperature monitoring system and method based on multispectral fusion as described in claim 1, characterized in that, Step S1 specifically involves: arranging at least three binocular thermometric infrared thermal imagers at equal intervals along the circumference of the ladle measurement area. Each binocular thermometric infrared thermal imager is arranged in a ring around the center of the ladle, with the included angle between adjacent imagesrs being equal, so that their field of view covers the surface of the ladle and ensures that an overlapping area is formed between adjacent fields of view.

3. The real-time ladle temperature monitoring system and method based on multispectral fusion according to claim 2, characterized in that, The image acquisition method in step S2 is as follows: each binocular infrared thermal imager is assigned an independent thread to achieve multi-threaded acquisition. The control box sends a synchronous trigger signal to each binocular temperature-measuring infrared thermal imager through a programmable logic controller, so that they can start image acquisition at the same time. The corresponding infrared temperature data, infrared thermal imaging images and visible light images are obtained through the interface program.

4. The real-time monitoring system and method for ladle temperature based on multispectral fusion according to claim 3, characterized in that, The multispectral data fusion processing in step S3 is as follows: spatial registration of infrared and visible light images is performed based on computer vision methods. The mapping relationship between the two images is established through feature extraction and feature matching. The registered images are segmented to extract the target area of ​​the ladle. Then, infrared temperature information and visible light structure information are fused to construct the spatial temperature field of the ladle surface.

5. The real-time ladle temperature monitoring system and method based on multispectral fusion according to claim 4, characterized in that, The temperature anomaly detection method is as follows: statistical analysis of the spatial temperature field is performed to extract the characteristic parameters of the surface temperature. The characteristic parameters include at least the maximum temperature, minimum temperature, average temperature and temperature dispersion. A dynamic threshold model is established based on historical temperature data to identify the abnormal state of the ladle surface temperature. Multi-level alarm strategies are set according to different degrees of anomaly to achieve graded early warning output.

Images