Arc-shaped continuous casting machine vibration monitoring method based on visual enhanced image technology

Abstract

The invention discloses an arc-shaped continuous casting machine vibration monitoring method based on a visual enhanced image technology. The method comprises the following steps that a measurement target object is arranged on an arc-shaped continuous casting vibration table; the vibration state of the measurement target object is monitored through the vision enhancement image system, and vibration data of the measurement target object is collected; preprocessing the collected vibration data of the measurement target object; according to the technological requirements of the arc-shaped continuous casting machine, alarm threshold values in the up-down direction, the left-right direction and the front-back direction of the outer arc and the inner arc are set. The arc continuous casting machine vibration monitoring method based on the visual enhanced image technology can stably and reliably work in a severe environment of a continuous casting production line, periodic continuous monitoring of vibration of the arc continuous casting machine in different directions is achieved, data collection is conducted under the complex working condition of actual operation of the continuous casting machine, and the accuracy of monitoring of the vibration of the arc continuous casting machine is improved. The data validity is improved, production personnel can make more accurate decisions, the production quality and efficiency are improved, the alarm threshold value is set, and smooth production is guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of continuous casting machine monitoring, and particularly relates to a vibration monitoring method for an arc-shaped continuous casting machine based on visual enhanced image technology. BACKGROUND

[0002] In the production process of a steel mill, an arc-shaped continuous casting machine plays a vital role. The working process is as follows: first, molten steel in a ladle is poured into a tundish, and then the molten steel is poured into a crystallizer through adjustment of a tundish slide gate. A vibration table is a core device for driving the crystallizer to make reciprocating motion, and through accurate control of the vibration parameters of the crystallizer, the crystallizer is prevented from being bonded with the molten steel, so as to ensure smooth stripping of the casting blank and improve the surface quality.

[0003] When the vibration table drives the crystallizer to make reciprocating motion, mainly relies on the up-down direction motion, but at the same time, inevitable front-back and left-right direction vibrations are generated. Therefore, effective detection of the up-down, left-right and front-back vibrations of the outer arc and inner arc of the vibration table, and control of the vibration size within a reasonable error range, play a key role in smooth stripping of the casting blank and improvement of the surface quality.

[0004] However, the current continuous casting production line site environment brings great challenges to vibration detection. The high temperature of the site environment makes it difficult to directly measure the vibration displacement of the three directions by installing and deploying displacement sensors on the vibration table. At the same time, the measurement method by deploying infrared sensors and the like is affected by complex environmental factors such as high temperature, dust and heat waves, and cannot achieve effective measurement.

[0005] The current detection method mainly detects the up-down, left-right and front-back vibration data of the outer arc and inner arc of the vibration table by vibration displacement sensors when the machine is stopped and runs empty. However, this method has many problems. The data collected during shutdown is quite different from the complex working environment and conditions of the continuous casting machine during operation, and the data effectiveness is poor. Moreover, the collection method is non-periodic collection, which cannot provide effective support for production personnel to adjust parameters in real time, and seriously affects the improvement of production efficiency and product quality. SUMMARY

[0006] The purpose of the present application is to provide a vibration monitoring method for an arc-shaped continuous casting machine based on visual enhanced image technology, which can work stably and reliably in the harsh environment of the continuous casting production line, realize periodic and continuous monitoring of the three direction vibrations of the outer arc and inner arc of the arc-shaped continuous casting machine, collect data under the complex working conditions of the actual operation of the continuous casting machine, improve data effectiveness, help production personnel make more accurate decisions, improve production quality and efficiency, set an alarm threshold, and ensure smooth production.

[0007] To achieve the above purpose, the present application provides a vibration monitoring method for an arc-shaped continuous casting machine based on visual enhanced image technology, comprising the following steps:

[0008] Step S1, setting a measurement target on the arc-shaped continuous casting vibration table;

[0009] Step S2, monitoring the vibration state of the measurement target by a visual enhancement image system, and collecting vibration data of the measurement target;

[0010] Step S3, preprocessing the collected vibration data of the measurement target;

[0011] Step S4, setting alarm thresholds in the up-down, left-right and front-back directions of the outer arc and the inner arc according to the process requirements of the arc-shaped continuous casting machine.

[0012] Preferably, in step S1, the measurement target is set on the arc-shaped continuous casting vibration table, specifically including:

[0013] First, a hole is made on the protective cover on the surface of the arc-shaped continuous casting vibration table, and the measurement target is set;

[0014] Then, the measurement target is fixed on the arc-shaped continuous casting vibration table, and the motion trajectory directly reflects the three-dimensional vibration state of the arc-shaped continuous casting machine;

[0015] Finally, the vibration parameters are inverted by tracking the displacement change of the measurement target.

[0016] Preferably, a mark pattern is designed on the surface of the measurement target, including but not limited to a checkerboard, concentric circles and a fluorescent code; the mark pattern is recognized and enhanced by a machine vision algorithm.

[0017] Preferably, in step S2, the vibration state of the measurement target is monitored by a visual enhancement image system, and vibration data of the measurement target is collected, and the specific process is as follows:

[0018] First, the visual enhancement image system, as a non-contact measuring instrument, shoots a video image of the measurement target;

[0019] Then, the video is processed by a visual algorithm enhancement technology to realize video enhancement, and the micro-vibration is amplified to a visible vibration form; each pixel point in the interface of the visual enhancement image system is equivalent to a displacement sensor, and there are millions of displacement sensors simultaneously collecting vibration data;

[0020] Finally, the vibration data of the point, including the waveform, frequency spectrum and axis trajectory, are obtained by selecting the position of interest on the video interface of the visual enhancement image system, i.e. the position corresponding to the measurement target; the visual enhancement image system continuously monitors the measurement target to realize periodic and continuous monitoring of the vibration in the up-down, left-right and front-back directions of the outer arc and the inner arc of the arc-shaped continuous casting machine vibration table.

[0021] Preferably, in step S3, the collected vibration data of the measurement target is pre-processed, and the specific process is as follows:

[0022] In step S31, the steam, dust and thermal radiation noise is eliminated by the frequency domain-physical joint denoising method.

[0023] In step S32, the SIFT algorithm is used to perform sub-pixel level positioning and measure the characteristic points of the target for quadratic polynomial fitting.

[0024] In step S33, the image coordinates are dynamically converted into world coordinates according to the curvature radius of the arc continuous casting machine, the rigid transformation matrix is calculated through the non-collinear feature points on the target, the translation / rotation components are separated, and the nonlinear displacement error caused by the arc structure is eliminated.

[0025] In step S34, the missing frames are reconstructed by motion compensation based on Temporal PS, the data is phase-aligned with the vibration period as the window, and the vibration period normalization is realized.

[0026] Preferably, in step S31, the steam, dust and thermal radiation noise is eliminated by the frequency domain-physical joint denoising method, which specifically includes:

[0027] Firstly, a band-pass filter is constructed based on Fourier transform to obtain the vibration fundamental frequency signal; and then, based on the atmospheric scattering model, the clear image of the measurement target is restored by utilizing the characteristics of near-infrared band penetrating steam.

[0028] Preferably, in step S4, according to the process requirements of the arc continuous casting machine, the alarm threshold values in the up-down, left-right and front-back directions of the outer arc and the inner arc are set.

[0029] When the vibration exceeds the set alarm threshold value, the system will timely remind the production operator, and the production operator will timely adjust the process parameters according to the alarm information to ensure the smooth demolding of the casting blank.

[0030] Therefore, the arc continuous casting machine vibration monitoring method based on the visual enhanced image technology has the following beneficial effects:

[0031] (1) Adapt to harsh environment: the visual enhanced image technology is adopted, which avoids the difficulties in installing and using traditional sensors in complex environments such as high temperature, dust and heat wave, and can work stably and reliably in the harsh environment of the continuous casting production line.

[0032] (2) Real-time continuous monitoring: the periodic continuous monitoring of the vibration of the outer arc and the inner arc of the arc continuous casting machine in three directions is realized, compared with the traditional stoppage non-periodic acquisition method, the vibration data can be provided to the production personnel in real time, so that the production personnel can timely understand the running state of the continuous casting machine, and provide timely and effective support for process parameter adjustment.

[0033] (3) Improve data validity: the vibration data obtained by the present application is collected under the complex working conditions of the actual operation of the continuous casting machine, which can better reflect the real working state of the continuous casting machine compared with the data collected when the machine is stopped and unloaded, greatly improving the data validity, helping production personnel to make more accurate decisions, and improving production quality and efficiency;

[0034] (4) Accurate alarm function: set alarm threshold according to the process requirements of the continuous casting machine, and timely remind the production operator when the vibration is abnormal, which can effectively prevent the quality problems and production accidents of the cast slab caused by excessive vibration, and ensure the smooth production.

[0035] The technical solutions of the present application will be further described in detail below through the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 A flowchart of an arc continuous casting machine vibration monitoring method based on visual enhanced image technology. DETAILED DESCRIPTION

[0037] The technical solutions of the present application will be further described in detail below through the drawings and examples.

[0038] As shown in Figure 1 An arc continuous casting machine vibration monitoring method based on visual enhanced image technology, comprising the following steps:

[0039] Step S1, setting a measurement target on the arc continuous casting vibration table;

[0040] Step S2, monitoring the vibration state of the measurement target through a visual enhanced image system, and collecting vibration data of the measurement target;

[0041] Step S3, preprocessing the collected vibration data of the measurement target;

[0042] Step S4, setting alarm thresholds for the outer arc, inner arc and three directions according to the process requirements of the arc continuous casting machine.

[0043] EMBODIMENT

[0044] Step S1, setting a measurement target on the arc continuous casting vibration table.

[0045] The surface of the arc continuous casting vibration table is usually provided with a protective cover to prevent molten steel from spilling. Therefore, when setting the measurement target, the protective cover needs to be holed according to the size of the measurement target.

[0046] The measurement target is fixed on the curved continuous casting vibration table, and its motion trajectory directly reflects the three-dimensional vibration state of the curved continuous casting machine. By tracking the displacement change of the measurement target, the vibration parameters are inverted. The measurement target replaces the traditional contact sensor, avoiding signal distortion caused by high temperature and dust.

[0047] The surface of the measurement target is designed with a special pattern (such as a chessboard, concentric circles, and fluorescent coding), and the recognition robustness is enhanced through machine vision algorithms. For example, in the 850nm infrared band, the reflectivity of the ceramic-based fluorescent coating is over 90%, significantly improving the imaging quality in low-visibility environments.

[0048] The known geometric features (such as symmetry and spacing) of the measurement target are used to calibrate the camera lens distortion and eliminate the image nonlinear error caused by the curved structure. Therefore, the measurement target has high contrast and is easy to identify, making it convenient for the vision-enhanced image system to accurately capture its vibration information.

[0049] Step S2, monitor the vibration state of the measurement target through the vision-enhanced image system, and collect the vibration data of the measurement target.

[0050] Firstly, the vision-enhanced image system, as a non-contact measuring instrument, takes a video of the measurement target.

[0051] Then, the video is processed through vision algorithm enhancement technology to realize video enhancement and amplify the micro-vibration to a visible vibration form. Each pixel point in the interface of the vision-enhanced image system is equivalent to a displacement sensor, and there are millions of displacement sensors simultaneously collecting vibration data.

[0052] Finally, by selecting the position of interest on the video interface of the vision-enhanced image system, which corresponds to the position of the measurement target, the vibration data of the point, including waveform, frequency spectrum, and axis trajectory, are obtained. The vision-enhanced image system continuously monitors the measurement target, realizing periodic and continuous monitoring of the up-down, left-right, and forward-backward vibration of the outer arc and inner arc of the curved continuous casting machine vibration table.

[0053] Step S3, pre-process the collected vibration data of the measurement target.

[0054] Step S31, eliminate steam, dust, and thermal radiation noise through frequency domain-physical joint denoising method.

[0055] Firstly, a band-pass filter is constructed based on Fourier transform to obtain the vibration fundamental frequency signal. Then, based on the atmospheric scattering model, the clear image of the measurement target is restored by utilizing the near-infrared band's characteristic of penetrating steam.

[0056] Step S32, using SIFT algorithm, sub-pixel level positioning measurement target feature point quadratic polynomial fitting.

[0057] Step S33, according to the curvature radius of the arc continuous casting machine, the image coordinates are dynamically converted into world coordinates, the rigid transformation matrix is calculated through at least three non-collinear feature points on the target, the translation / rotation components are separated, and the nonlinear displacement error caused by the arc structure is eliminated.

[0058] Step S34, based on the Temporal PS, the missing frames are motion compensated and reconstructed, the vibration period of the target is taken as a window, the data is phase aligned, and the vibration period is normalized.

[0059] Step S4, according to the process requirements of the arc continuous casting machine, the alarm thresholds in the up-down, left-right and front-back directions of the outer arc and the inner arc are reasonably set. When the vibration exceeds the set alarm threshold, the system will timely remind the production operator, and the production operator can adjust the process parameters in time according to the alarm information to ensure the smooth demolding of the casting blank and improve the production quality.

[0060] Therefore, the arc continuous casting machine vibration monitoring method based on the visual enhanced image technology has the following beneficial effects:

[0061] 1. Adapt to harsh environment: using visual enhanced image technology, the problems of installing and using traditional sensors in high temperature, dust, heat wave and other complex environments are avoided, and the system can work stably and reliably in the harsh environment of the continuous casting production line.

[0062] 2. Real-time continuous monitoring: the present application realizes the periodic continuous monitoring of the vibration of the outer arc and the inner arc of the arc continuous casting machine in three directions, compared with the traditional stop non-periodic acquisition method, the vibration data can be provided in real time for the production personnel, so that the production personnel can understand the running state of the continuous casting machine in time, and provide timely and effective support for the adjustment of process parameters.

[0063] 3. Improve data effectiveness: the vibration data obtained by the present application is collected under the complex working conditions of the continuous casting machine, which can better reflect the real working state of the continuous casting machine compared with the data collected when the machine is stopped and unloaded, and the data effectiveness is greatly improved, which helps the production personnel to make more accurate decisions and improves the production quality and efficiency.

[0064] 4. Accurate alarm function: according to the process requirements of the continuous casting machine, the alarm threshold is set, and when the vibration is abnormal, the production operator is reminded in time, which can effectively prevent the casting blank quality problems and production accidents caused by excessive vibration, and ensure the smooth production.

[0065] It should be pointed out finally that the above examples are only used to illustrate the technical solutions of the present application but not to limit it, and although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can still be modified or replaced equivalently, and these modifications or equivalent replacements should not make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present application.

Claims

1. A method for monitoring the vibrations of an arc continuous caster based on visual enhancement imaging technology, characterized in that, The method comprises the following steps: Step S1, setting a measurement target on the arc continuous casting vibration table; Step S2, monitoring the vibration state of the measurement target through a visual enhanced image system, and collecting vibration data of the measurement target; Step S3, preprocessing the collected vibration data of the measurement target; Step S4, setting alarm thresholds in the up-down, left-right and front-back directions of the outer arc and the inner arc according to the process requirements of the arc continuous casting machine.

2. A method of monitoring the vibration of an arc continuous caster based on visual augmented image technology according to claim 1, characterized in that, In step S1, the measurement target is set on the arc continuous casting vibration table, which specifically comprises: First, a hole is opened on the protective cover on the surface of the arc continuous casting vibration table to set the measurement target; Then, the measurement target is fixed on the arc continuous casting vibration table, and the motion track directly reflects the three-dimensional vibration state of the arc continuous casting machine; Finally, the vibration parameters are inverted by tracking the displacement change of the measurement target.

3. A method of monitoring the vibrations of an arc continuous caster based on vision-enhanced video technology according to claim 2, characterized in that, The surface of the measurement target is designed with a mark pattern, including but not limited to a checkerboard, concentric circles and a fluorescent code; The mark pattern is recognized and enhanced through a machine vision algorithm.

4. The method of claim 1, wherein the method is a visual enhanced image based method of monitoring the arc of a continuous caster. In step S2, the vibration state of the measurement target is monitored through the visual enhanced image system, and the vibration data of the measurement target is collected, and the specific process is as follows: First, the visual enhanced image system is a non-contact measuring instrument, which shoots a video image of the measurement target; Then, the video is processed through visual algorithm enhancement technology to realize video enhancement, and the micro-vibration is amplified to a visible vibration form; each pixel point in the interface of the visual enhanced image system is equivalent to a displacement sensor, which is equivalent to millions of displacement sensors simultaneously collecting vibration data; Finally, the vibration data of the point, including the waveform, frequency spectrum and axis track, are obtained by selecting the position of interest on the video interface of the visual enhanced image system, i.e. the position corresponding to the measurement target; the visual enhanced image system continuously monitors the measurement target to realize periodic and continuous monitoring of the vibration in the up-down, left-right and front-back directions of the outer arc and the inner arc of the arc continuous casting machine vibration table.

5. The method of claim 1, wherein the method is based on visual enhanced image technology of an arc continuous caster vibration monitoring method. In step S3, the collected vibration data of the measurement target is preprocessed, and the specific process is as follows: Step S31, eliminate steam, dust and thermal radiation noise through a frequency domain-physical joint denoising method; Step S32, use SIFT algorithm to perform sub-pixel level positioning and measure the feature points of the measurement target twice polynomial fitting; Step S33, convert the image coordinates to world coordinates dynamically according to the curvature radius of the arc continuous casting machine, calculate the rigid transformation matrix through the non-collinear feature points on the target, separate the translation / rotation components, and eliminate the nonlinear displacement error caused by the arc structure; Step S34, based on TemporalPS, reconstruct the missing frames by motion compensation, take the vibration period of the measurement target as the window, and align the data in phase to realize vibration period normalization.

6. A method of monitoring the vibration of an arc continuous caster based on vision-enhanced video technology according to claim 5, characterized in that, In step S31, the steam, dust and thermal radiation noise are eliminated through the frequency domain-physical joint denoising method, which specifically comprises: First, a band-pass filter is constructed based on Fourier transform to obtain the vibration fundamental frequency signal; then, based on the atmospheric scattering model, the clear image of the measurement target is restored by using the characteristic of near-infrared band penetrating steam.

7. The method of claim 1, wherein the method is a visual enhanced image based method of monitoring the arc of a continuous caster. In step S4, according to the arc-shaped continuous casting machine process requirements, the upper and lower, left and right, front and back three direction alarm thresholds of the outer arc and the inner arc are set; When the vibration exceeds the set alarm threshold, the system will timely remind the production operator, and the production operator will timely adjust the process parameters according to the alarm information to ensure the smooth stripping of the casting blank.

Images