Molten steel liquid level monitoring system and method for composite roller remanufacturing

Abstract

The invention relates to the technical field of nondestructive testing, and provides a composite roller remanufacturing molten steel liquid level monitoring system, which adopts a mode of combining an external ultrasonic detection module and a video detection module, and is characterized in that an ultrasonic probe array is attached to the outer wall of a molten steel container through a high-temperature-resistant coupling agent; analyzing the time-frequency characteristics of the ultrasonic signals to detect the liquid level and the steel slag distribution; the high-temperature industrial camera provided with an active cooling structure and a narrow-band infrared optical filter is used for carrying out image acquisition on the surface of the molten steel and identifying a slag layer area. The multi-mode fusion processor overcomes measurement errors caused by molten steel turbulence and slag layer interference. A water cooling channel of the ultrasonic probe is arranged, and long-term stable operation of the sensor in a high-temperature environment is ensured through a circulating cooling medium. The method has the advantages of non-contact measurement, multi-mode redundancy check, self-adaptive compensation and the like, and the requirements for high-precision and high-reliability real-time monitoring of the liquid level of the molten steel in the roller remanufacturing process are met.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of nondestructive testing, and in particular to a molten steel liquid level monitoring system and method for composite roller remanufacturing. BACKGROUND

[0002] In the production of metallurgical industry, accurate monitoring of the molten steel liquid level is a key link in the processes of continuous casting and roller remanufacturing, and directly affects the product quality and production safety. Too high molten steel liquid level can cause overflow accidents, and too low molten steel liquid level can cause slag entrapment defects, so it is necessary to realize real-time monitoring with high precision and high reliability. However, the molten steel liquid level detection in the process of composite roller remanufacturing faces severe technical challenges brought by extreme working conditions.

[0003] Currently, two detection methods are mainly used in industrial sites: contact type and non-contact type. The contact type detection is represented by the eddy current method, which measures the liquid level by installing an electromagnetic induction coil sensor on the side of the molten steel container through drilling, so that the sensor is in direct contact with the molten steel. Although this method has a simple measurement principle, it has obvious defects: the sensor needs to be close to the high-temperature molten steel at 1500-1600℃, and even with a cooling device, the protective layer of the contact surface will still be quickly ablated and damaged, with a very short service life; moreover, the traditional drilling installation method severely limits the flexibility of sensor arrangement, and the maintenance cost is high. More importantly, in special processes such as roller remanufacturing, there is often a composite material overlay on the inner wall of the container, which will significantly affect the stability of the eddy current signal, resulting in increased measurement error.

[0004] The non-contact detection method mainly uses laser ranging and infrared thermal imaging technology. Although this type of method avoids the problem of sensor burning, in actual application, the turbulent fluctuations of the molten steel will cause surface disturbances, and a single detection method is difficult to achieve accurate tracking.

[0005] The current detection means cannot meet the precision and reliability requirements of molten steel liquid level monitoring, and new detection schemes need to be developed to break through multiple technical obstacles such as high temperature and steel slag interference to meet the detection needs of modern metallurgical industry. SUMMARY

[0006] To solve the above problems, the present application provides a molten steel liquid level monitoring method for composite roller remanufacturing, which is implemented as follows:

[0007] A molten steel liquid level monitoring system for composite roller remanufacturing, comprising:

[0008] An external ultrasonic detection module, the external ultrasonic detection module comprises a plurality of ultrasonic probe arrays attached to the outer wall of the molten steel container, and the ultrasonic probe is acoustically coupled to the outer wall of the container through a high-temperature resistant coupling agent;

[0009] A video detection module, which comprises a high-temperature industrial camera aiming at the surface of molten steel, the camera being configured with a narrow-band infrared filter and an active cooling structure;

[0010] A multi-modal fusion processor, which receives time-domain / frequency-domain signals of an ultrasonic array and a video image sequence, and performs the following operations:

[0011] A steel slag feature band in the ultrasonic signal is extracted through wavelet transform;

[0012] A slag layer area in the video is identified based on LBP texture analysis;

[0013] A double-flow neural network is adopted to fuse the ultrasonic transit time and visual edge features, and output a compensated liquid level value;

[0014] A probe cooling module, which is arranged between the outside of the container and the ultrasonic probe, and comprises a water-cooling channel surrounding the ultrasonic probe, and a fluid cooling medium circulating in the channel.

[0015] Further, the ultrasonic probe array is at least two layers of annular probe arrays uniformly distributed along the circumference of the outer wall of the container, each probe emits a coded pulse signal, and receives reflected waves or transmitted waves.

[0016] Further, the ultrasonic probe is arranged in three layers of annular arrays along the height direction of the container, and the spacing between each layer is one-third of the maximum fluctuation height of the molten steel.

[0017] Further, a plurality of ultrasonic probes are arranged in each layer, adjacent probes are at an included angle of 45°, and upper and lower layer probes are staggered by 22.5°, forming a three-dimensional detection network.

[0018] Further, the water-cooling channel is a double-layer spiral structure, the inner channel directly contacts the piezoelectric wafer, the outer channel covers the probe shell, and the flow directions of the two media are opposite.

[0019] Further, when the multi-modal fusion processor performs steel slag compensation, it specifically includes:

[0020] The slag layer thickness is calculated by the amplitude attenuation rate of the transmitted wave and the time delay difference of the reflected wave;

[0021] When the video recognition slag coverage is greater than 30%, the infrared thermal image fitting liquid surface is enabled.

[0022] Further, the exposure time of the high-temperature industrial camera and the ultrasonic pulse emission are synchronized by FPGA hardware.

[0023] The application also discloses a composite roller remanufacturing molten steel liquid level monitoring method, which adopts the detection system described above, and the specific steps include:

[0024] S1. Emit coded pulses to the container wall by external ultrasonic array, and receive reflected / transmitted signals;

[0025] S2. Synchronously collect multi-spectral video images of the liquid surface, and extract dynamic optical flow field;

[0026] S3. Identify slag interference area based on the spectrum slope of ultrasonic signals, and calculate acoustic compensation amount;

[0027] S4. Input ultrasonic transit time, video edge features and slag layer compensation amount into a pre-trained double-flow neural network, and output liquid level value;

[0028] S5. Stabilize the probe temperature in the set temperature range by closed-loop control of the water cooling system flow rate. Further, the reflected signal is a first echo received by the liquid surface / slag layer, and the transit time is measured;

[0029] The transmitted signal is a penetrating signal received by the opposite probe, and the attenuation coefficient is used to distinguish between liquid steel and slag.

[0030] Compared with the prior art, the present application has the following beneficial effects:

[0031] The present application effectively solves the problem of high-temperature molten steel liquid level detection by organically combining the dual-mode detection method, and the specific effects include:

[0032] First, the present application innovatively adopts an external ultrasonic detection scheme, and the ultrasonic probe array is attached to the outer wall of the liquid steel container, so that detection can be realized without drilling holes on the container. Through the water cooling area heat conduction control system, the working environment temperature of the sensor is significantly reduced, the service life of the probe is greatly prolonged, and the reliability and stability of the detection are ensured.

[0033] Second, the present application designs a multi-layer ultrasonic probe array structure, and constructs a three-dimensional detection network on the outer wall of the liquid steel container, which improves the detection coverage. An optimized spatial arrangement mode and dynamic recombination technology are adopted to ensure that multiple probes work cooperatively at any time, realize omnidirectional monitoring of the molten steel liquid surface, and effectively eliminate the blind area problem existing in the traditional detection method.

[0034] Third, the present application combines the external ultrasonic array with the video detection system, and the video detection adopts a specially configured high-temperature industrial camera. Through optical filtering and image enhancement technology, the background thermal radiation interference is significantly suppressed, and the imaging clarity is greatly improved. The precise synchronization and intelligent fusion processing of the ultrasonic and visual systems realize high-precision liquid level measurement and excellent slag layer recognition performance. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the examples of the present application or the prior art or the descriptions in the prior art required to be used in the drawings, it is obvious that, for those skilled in the art, without creative labor, other drawings can also be obtained according to these drawings.

[0036] Figure 1 The flow chart of the molten steel liquid level monitoring method for the composite roller remanufacturing steel.

[0037] Figure 2 The detection working condition schematic diagram. DETAILED DESCRIPTION

[0038] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application.

[0039] Reference is made to the accompanying drawings Figure 2 In the composite roller remanufacturing working condition, the roller 10 is generally placed in the molten steel container, and the molten steel 30 is injected into the container 20 for casting. In this working condition, the present application discloses a molten steel liquid level monitoring system for composite roller remanufacturing, which comprises:

[0040] The external ultrasonic detection module 40 comprises a plurality of ultrasonic probe arrays attached to the outer wall of the molten steel container. The ultrasonic probes are acoustically coupled to the outer wall of the container through a high-temperature resistant coupling agent. Specifically, it further comprises the arrangement of the ultrasonic probe array and the setting of the specific signal acquisition mode.

[0041] The video detection module 50 comprises a high-temperature industrial camera aligned with the surface of the molten steel. The camera is equipped with a narrow-band infrared filter and a cooling structure. The narrow-band infrared filter suppresses background radiation, and the active cooling structure can use a semiconductor cooler to maintain the camera temperature. Further, a nitrogen curtain can be set to prevent mirror contamination.

[0042] The multi-modal fusion processor receives the time domain / frequency domain signals of the ultrasonic array and the video image sequence, and performs the following operations:

[0043] The steel slag characteristic frequency band in the ultrasonic signal is extracted by wavelet transform;

[0044] The slag layer area in the video is identified based on LBP texture analysis;

[0045] The dual-flow neural network fuses the ultrasonic transit time and visual edge features to output a compensated liquid level value.

[0046] A probe cooling module 60 is arranged between the outside of the container and the ultrasonic probe, and comprises a water-cooled channel surrounding the ultrasonic probe, in which a fluid cooling medium is circulated.

[0047] The detection system must adapt to the curved surface geometry, and the traditional linear probe array arrangement will produce a detection blind area. At the same time, the difference in acoustic characteristics between the roll base material and the overlay layer will further interfere with the accuracy of the measurement signal. Based on this, in the embodiment, the ultrasonic probe array is evenly distributed along the outer wall of the container in at least two layers of annular probe arrays, each probe emits a coded pulse signal, and receives reflected or transmitted waves.

[0048] In the axial arrangement, the ultrasonic probe is arranged in three layers of annular arrays along the height direction of the container, and the interval between each layer is one third of the maximum fluctuation height of the liquid steel. For example, if the liquid level fluctuation range is ±150mm, the layer spacing is set to 100mm. In the circumferential distribution, a plurality of ultrasonic probes are arranged in each layer, the adjacent probes are at an angle of 45°, and the upper and lower probes are staggered by 22.5°, forming a three-dimensional detection network. This design ensures that at least two ultrasonic probes can detect the liquid surface at the same time at any time, eliminating the detection blind area.

[0049] The ultrasonic probe is installed on a flexible substrate. Specifically, the ultrasonic probe can be fixed on a copper alloy flexible substrate, and the probe is arranged on the outer wall of the liquid steel container by applying pressure through a hydraulic cylinder to form a self-adaptive curved surface fitting.

[0050] Further, the water-cooled channel is a double-layer spiral structure, the inner channel directly contacts the piezoelectric wafer, and the outer channel covers the probe shell, and the flow directions of the two media are opposite. Forming a reverse temperature lifting, reducing the thermal flow superposition effect. The inner channel quickly cools the piezoelectric wafer, and the outer channel maintains the temperature balance of the structure. Through secondary heat exchange, the temperature of the outer layer is balanced, and the system energy efficiency is improved.

[0051] Further, when the multi-modal fusion processor performs slag compensation, it specifically includes:

[0052] The slag layer thickness is calculated by the transmission wave amplitude attenuation rate and the reflection wave time delay difference;

[0053] When the video recognition slag coverage is greater than 30%, the infrared thermal image fitting liquid surface is enabled.

[0054] Further, the exposure time of the high-temperature industrial camera and the ultrasonic pulse emission are synchronized by FPGA hardware.

[0055] The application further discloses a molten steel liquid level monitoring method for composite roller remanufacturing.

[0056] S1. Emitting coded pulses to the container wall through an external ultrasonic array, and receiving reflected / transmitted signals;

[0057] S2. Synchronously collecting multi-spectral video images of the steel liquid surface, and extracting a dynamic light flow field;

[0058] S3. Identifying a steel slag interference area based on a spectrum slope of the ultrasonic signals, and calculating an acoustic compensation amount;

[0059] S4. Inputting the ultrasonic transit time, video edge features and slag layer compensation amount into a pre-trained double-flow neural network, and outputting a liquid level value;

[0060] S5. Stabilizing the probe temperature in a set temperature range through closed-loop control of the water cooling system flow rate. Further, the reflected signal is a first echo received by the liquid surface / slag layer, and the transit time is measured;

[0061] The transmitted signal is a penetrating signal received by the opposite probe, and the steel liquid and the steel slag are distinguished through an attenuation coefficient.

[0062] The above merely describes the preferred embodiments of the application and is not used to limit the application, and the application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall be included in the protection scope of the application.

Claims

1. A composite roll remanufacturing molten steel level monitoring system, characterized by, The application relates to a molten steel level detection system, which comprises the following parts: An external ultrasonic detection module, which comprises a plurality of ultrasonic probe arrays attached to the outer wall of a molten steel container, and the ultrasonic probes are acoustically coupled to the outer wall of the container through high-temperature coupling agents; A video detection module, which comprises a high-temperature industrial camera aimed at the surface of the molten steel, and the camera is provided with a narrow-band infrared filter and a cooling structure; A multi-modal fusion processor, which receives the time-domain / frequency-domain signals of the ultrasonic array and the video image sequence, and performs the following operations: Extracting the slag feature frequency band in the ultrasonic signal through wavelet transform; Identifying the slag layer area in the video based on LBP texture analysis; Fusing the ultrasonic transit time and the visual edge features through a double-flow neural network to output the compensated liquid level value; A probe cooling module, which comprises a water-cooled channel surrounding the ultrasonic probe, and a fluid cooling medium is circulated in the channel.

2. The composite roll remanufacturing molten steel level monitoring system according to claim 1, characterized in that, The ultrasonic probe array comprises at least two layers of annular probe arrays uniformly distributed along the circumference of the container outer wall, each ultrasonic probe emits a coded pulse signal and receives reflected waves or transmitted waves.

3. The composite roll remanufacturing molten steel level monitoring system according to claim 2, characterized in that, The ultrasonic probes are arranged in three layers along the height direction of the container, and the interval between each layer is one-third of the maximum fluctuation height of the molten steel.

4. The composite roll remanufacturing molten steel level monitoring system according to claim 3, characterized in that, Each layer is provided with a plurality of ultrasonic probes, adjacent probes are at an angle of 45 degrees, and upper and lower probes are staggered by 22.5 degrees, forming a three-dimensional detection network.

5. The composite roll remanufacturing molten steel level monitoring system according to claim 1 or 2, characterized in that, The water-cooled channel has a double-layer spiral structure, the inner channel directly contacts the piezoelectric wafer, the outer channel covers the probe shell, and the flow directions of the two media are opposite.

6. The composite roll remanufacturing molten steel level monitoring system according to claim 1, characterized in that, When the multi-modal fusion processor performs slag compensation, it specifically comprises: Calculating the slag layer thickness through the attenuation rate of the transmitted wave amplitude and the time delay difference of the reflected wave; When the video-identified slag coverage is greater than 30%, the infrared thermal image is used to fit the liquid level.

7. The composite roll remanufacturing molten steel level monitoring system according to claim 1, characterized in that, The exposure time of the high-temperature industrial camera and the ultrasonic pulse emission are synchronized by FPGA hardware.

8. A method of monitoring the liquid level of molten steel for a composite roll remanufacturing steel using the detection system according to any one of claims 1 to 5, characterized by, The specific steps comprise: S1. Emitting coded pulses to the container wall through the external ultrasonic array, and receiving reflected / transmitted signals; S2. Synchronously collecting multispectral video images of the molten steel surface, and extracting dynamic optical flow fields; S3. Identifying the steel slag interference area based on the spectral slope of the ultrasonic signal, and calculating the acoustic compensation amount; S4. Inputting the ultrasonic transit time, video edge features and slag layer compensation amount into a pre-trained double-flow neural network to output the liquid level value; S5. Controlling the flow rate of the water cooling system through a closed loop to stabilize the probe temperature within a set temperature range.

9. The method of claim 8, wherein the method is characterized by: The reflected signal is a first echo received by the liquid surface / slag layer, and the transit time is measured; The transmitted signal is a penetrating signal received by the opposite probe, and the attenuation coefficient is used to distinguish the molten steel and the steel slag.

Images