Steel rail bending torsion on-line monitoring control method based on thermal-state 3D full-section contour scanner

Abstract

The invention relates to the technical field of steel rolling production quality detection, in particular to a steel rail bending and torsion online monitoring control method based on a thermal-state 3D full-section contour scanner, which is applied to a finish rolling mill and comprises the following steps: acquiring data of a steel rail, including a bending angle, a torsion angle and temperature data; formulating a corresponding control strategy according to the data of the steel rail; dynamically optimizing a control strategy through a PID (Proportion Integration Differentiation) control algorithm in combination with the monitoring data and the state of the finish rolling mill; and a mapping relation between steel rail deformation and parameters is established, an anti-overregulation mechanism is set, and online monitoring and control of steel rail bending torsion are achieved. Three-dimensional point cloud and temperature data of a plurality of sheet layers of the steel rail are obtained in real time, bending and torsion parameters are calculated through an intelligent algorithm and then transmitted to a finish rolling mill control system in real time through the industrial Ethernet, parameters such as roller pressure and roller gaps are automatically adjusted based on a preset control model, a real-time closed loop of detection-calculation-adjustment is formed, and the real-time control of the steel rail is achieved. And ensuring that the steel rail geometric shape meets the standard in the tapping stage.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of steel rolling production quality detection, and in particular to a rail bending and twisting online monitoring and control method based on a hot 3D full-section profile scanner. BACKGROUND

[0002] In the process of rail hot rolling, the bending (lateral and vertical deviation > 1mm / m) and twisting (> 0.3° / m) defects in the finishing mill steel outlet stage are mainly caused by factors such as uneven roll pressure and roll gap deviation. The traditional monitoring method has the following disadvantages:

[0003] Control lag: offline detection delay > 10min, finishing mill parameter adjustment relies on manual experience, resulting in a large number of defective rails; Lack of linkage: offline detection and finishing mill control run independently, unable to dynamically adjust the steel outlet parameters according to real-time deformation data; Insufficient precision: offline detection cannot cover the full section, and the compensation for thermal deformation is not accurate, with a bending and twisting control error > 1mm / m.

[0004] The prior art lacks deep integration of hot full-section monitoring and rolling mill control, and the finishing mill steel outlet parameters (such as roll pressure and roll gap) are adjusted with a lag, making it difficult to meet the real-time forming requirements of high-speed rail (bending degree ≤ 1mm / m). Therefore, there is an urgent need for an intelligent monitoring system with real-time linkage control capability. SUMMARY

[0005] In view of the above technical problems, a rail bending and twisting online monitoring and control method based on a hot 3D full-section profile scanner is provided. The present application obtains real-time three-dimensional point cloud and temperature data of multiple slices of the rail, calculates the bending and twisting parameters through intelligent algorithms, and transmits them to the finishing mill control system (PLC) in real time through industrial Ethernet. Based on the preset control model, the roll pressure and roll gap parameters are automatically adjusted, forming a real-time closed loop of "detection-computation-adjustment", and ensuring that the geometric shape of the steel outlet stage rail meets the standard.

[0006] The technical means adopted by the present application are as follows:

[0007] A rail bending and twisting online monitoring and control method based on a hot 3D full-section profile scanner, applied to a finishing mill, comprising:

[0008] Obtaining data of the rail, including bending angle, twisting angle and temperature data;

[0009] Formulating a corresponding control strategy according to the data of the rail;

[0010] Combining the monitoring data and the state of the finishing mill, dynamically optimizing the control strategy through a PID control algorithm;

[0011] The mapping relationship between the rail deformation and the parameters is established, and an over-adjustment prevention mechanism is set, so as to realize the online monitoring and control of the rail bending and torsion.

[0012] Further, the data calculation method of the rail is as follows:

[0013] The calculation formula of the bending angle θ is:

[0014]

[0015] θ2 = arctan (Δw / L)

[0016] θ = θ1 / θ2

[0017] Wherein, θ1 represents the vertical bending angle, θ2 represents the horizontal bending angle, Δh: the height difference of the rail in the vertical direction, reflecting the degree of vertical bending deformation; L is the length of the measured bending part of the rail, which is used as the reference length when calculating the bending angle; Δw is the width difference of the rail in the horizontal direction, reflecting the width change caused by the horizontal bending;

[0018] The calculation formula of the torsion angle Δθ is:

[0019] Δθ = ∑arccos(n i ·n i+1 )

[0020] Wherein, n i represents the normal vector of the i-th section, which is perpendicular to the section and is used to describe the spatial direction of the section; n i+1 represents the normal vector of the (i+1)th section, which is adjacent to n i , by calculating the relationship between the normal vectors of two adjacent sections, the torsion angle of the rail is determined.

[0021] Further, the corresponding control strategy is formulated according to the data of the rail, including bending control and torsion control;

[0022] The bending control means that when the vertical bending θ1 > 3°, the vertical roll pressure is increased by 0.5-1.0 kN / mm; when the horizontal bending angle θ2 > 2°, the horizontal roll gap is adjusted by ±0.3 mm.

[0023] The torsion control means that when the torsion angle Δθ > 1.5° / m, the roll speed compensation is triggered to correct the uneven torque in the rolling process.

[0024] Further, the dynamic optimization of the control strategy by the PID control algorithm specifically includes:

[0025] The thermal deformation compensation model is introduced into the finish rolling temperature influence factor, and the correction formula is:

[0026] (X',Y',Z')=(X,Y,Z)*(1+a*(T-20℃)+β*ΔF / F0)

[0027] Wherein, (X', Y', Z') represents the coordinate value of the corrected rail section profile in three-dimensional space, used to represent the rail shape after considering the influence of thermal deformation and pressure, (X, Y, Z) represents the three-dimensional coordinates of the uncorrected rail section profile; a is the temperature influence coefficient, which reflects the influence degree of temperature change on rail deformation correction, used to quantify the role of temperature factors, T is the finishing temperature, that is, the temperature when the rail finishing process is completed, β is the pressure influence coefficient, used to quantify the parameter of the influence degree of the change of roll pressure on the correction of rail thermal deformation; ΔF is the roll pressure adjustment, which is the value of adjusting the roll pressure in actual production, F0 is the initial roll pressure, which is the pressure reference value set by the roll initially.

[0028] Further, the mapping relationship between the rail deformation and the parameters is that the roll pressure adjustment and the roll gap adjustment are obtained by fitting the historical data;

[0029] The calculation formula of the roll pressure adjustment ΔF is:

[0030] ΔF=k1·θ1

[0031] The calculation formula of the roll gap adjustment ΔS is:

[0032] ΔS=k2·θ2

[0033] Wherein, k1 and k2 are specification coefficients.

[0034] The anti-overadjustment mechanism adjusts the gradient of the parameters, and cooperates with the feedback check to ensure the control stability.

[0035] Compared with the prior art, the present application has the following advantages:

[0036] The rail bending and torsion online monitoring control method based on the hot state 3D full section profile scanner provided by the present application obtains the data of the rail by installing a camera, an integrated infrared thermal imager and the like on the outlet roller of the finishing mill, and formulates a corresponding control strategy according to the data of the rail; the control strategy is dynamically optimized by a PID control algorithm in combination with the monitoring data and the state of the finishing mill; the mapping relationship between the rail deformation and the parameters is established, and an anti-overadjustment mechanism is set, so that the online monitoring and control of the rail bending and torsion are realized.

[0037] The rail bending and torsion online monitoring control method based on the hot state 3D full cross-section contour scanner provided by the application realizes deep linkage of hot state monitoring and final rolling mill tapping control, the detection adjustment delay is less than 500 ms, the lag problem of the traditional method is solved, and the defect rate is reduced by 75%. The application can realize accurate forming control, through full cross-section scanning and intelligent algorithm, the bending control precision is ±0.5 mm / m, the torsion is ±0.5° / m, and the strict requirements of high-speed rail (speed greater than or equal to 350 km / h) are met.

[0038] The rail bending and torsion online monitoring control method based on the hot state 3D full cross-section contour scanner provided by the application has self-adaptive adjustment capability, is based on historical data and a machine learning optimized control model, and automatically adapts to different specifications of rails (43 kg / m-75 kg / m), and the parameter adjustment error is less than 5%. The application can support intelligent production, data tracing and effect checking functions provide quantitative basis for process optimization.

[0039] Based on the above reasons, the application can be widely promoted in the field of steel rolling production quality detection technology. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor under the premise of not paying.

[0041] Figure 1 The rail bending and torsion online monitoring control method based on the hot state 3D full cross-section contour scanner in the application is a flow chart.

[0042] Figure 2 The arrangement structure of the camera on the exit roller of the finishing mill in the application.

[0043] Figure 3 The rail cross-section schematic view in the application. DETAILED DESCRIPTION

[0044] It should be noted that the embodiments in the application and the features in the embodiments can be combined with each other without conflict. The application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0045] 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 only part of the embodiments of the present application, rather than all the embodiments. The description of the at least one example embodiment is actually only illustrative, but not as any limitation on the present application and its application or use. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0046] It should be noted that the terms used herein are only intended to describe specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form, unless the context clearly indicates otherwise, and it should also be understood that, when the terms "comprise" and / or "include" are used in the specification, there is a reference to the presence of a feature, step, operation, device, component and / or combinations thereof.

[0047] Unless specifically stated otherwise, the relative arrangements of components and steps, numerical expressions, and numerical values set forth in the examples herein are not intended to limit the scope of the application. At the same time, it should be clear that the sizes of the various parts shown in the drawings are not drawn in proportion. The techniques, methods and devices known to those of ordinary skill in the relevant art can not be discussed in detail, but should be considered as part of the specification under appropriate circumstances. In all examples shown and discussed herein, any specific value should be interpreted as merely exemplary, and not as a limitation. Therefore, other examples of the exemplary embodiments can have different values. It should be noted that similar reference numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0048] As shown in Figure 1 The present application provides a rail bending and torsion online monitoring and control method based on a hot 3D full-profile contour scanner, which is applied to a finishing mill, 8 industrial-grade cameras (Basler acA2040-10gm, 1000fps, IP68) are installed at a 120° angle above and on both sides of the exit roller of the finishing mill, covering the full profile of the rail, as shown in Figure 2 An infrared thermal imager (±2℃) and an encoder (0.01mm resolution) are integrated to synchronously collect the temperature field and rolling speed (5-30m / s). Real-time communication with the finishing mill PLC is realized through the PROFINET protocol, with a communication delay <50ms.

[0049] Data of the steel rail are acquired, including bending angle, torsion angle and temperature data; the cross section of the steel rail is as shown in Figure 3 Figure 3 respectively, single cross section, multiple cross sections, movement, combination, arrangement profile cross section on the roller way and reconstruction of three-dimensional object, including defects. In the implementation, as a preferred embodiment of the present application, the data calculation method of the steel rail is as follows:

[0050] The calculation formula of the bending angle θ is as follows:

[0051]

[0052] θ2=arctan(Δw / L)

[0053] θ=θ1 / θ2

[0054] Wherein, θ1 represents the vertical bending angle, θ2 represents the horizontal bending angle, Δh: the height difference of the steel rail in the vertical direction, reflecting the degree of vertical bending deformation, such as the height difference of the two ends of a section of steel rail in the vertical direction due to bending; L is the length of the measured bending part of the steel rail, which is the reference length for calculating the bending angle, and can be understood as the distance between the two measuring points along the length direction of the steel rail; Δw is the width difference of the steel rail in the horizontal direction, which reflects the width change caused by the horizontal bending, that is, the width difference between the two ends of the steel rail in the horizontal direction when the steel rail is horizontally bent, with the precision of ±0.5°.

[0055] The calculation formula of the torsion angle Δθ is as follows:

[0056] Δθ=∑arccos(n i ·n i+1 )

[0057] Wherein, n i represents the normal vector of the i-th cross section, which is perpendicular to the cross section and is used to describe the spatial direction of the cross section; n i+1 represents the normal vector of the (i+1)-th cross section, which is adjacent to n i . By calculating the relationship between the normal vectors of two adjacent cross sections, the torsion angle of the steel rail is determined, with the resolution of 0.05° / m. According to the specifications of the steel rail (such as 60kg / m, 75kg / m), the threshold value is automatically matched, and manual calibration is supported (error ±5%)

[0058] According to the data of the steel rail, a corresponding control strategy is formulated; in the implementation, as a preferred embodiment of the present application, the corresponding control strategy formulated according to the data of the steel rail includes bending control and torsion control.

[0059] ​The bending control indicates that when the vertical bending θ1>3°, the vertical roll pressure is increased by 0.5-1.0 kN / mm, and is dynamically adjusted according to the deviation; when the transverse bending angle θ2>2°, the horizontal roll gap is adjusted by ±0.3 mm;

[0060] The torsion control indicates that when the torsion angle Δθ>1.5° / m, the rolling speed compensation (±0.5%) of the roll is triggered to correct the uneven torque in the rolling process.

[0061] In combination with the monitoring data and the state of the finishing mill, the control strategy is dynamically optimized through a PID control algorithm.

[0062] In specific implementation, as a preferred embodiment of the present application, the dynamic optimization of the control strategy through the PID control algorithm specifically includes:

[0063] The thermal deformation compensation model is introduced into the finishing temperature influencing factor, and the formula is corrected as:

[0064] (X',Y',Z')=(X,Y,Z)*(1+a*(T-20℃)+β*ΔF / F0)

[0065] Wherein, (X',Y',Z') represents the coordinate values of the corrected rail section profile in the three-dimensional space (which can be understood as a coordinate system), used to represent the rail shape after considering the influence of thermal deformation and pressure, (X,Y,Z) represents the uncorrected (original) rail section profile three-dimensional coordinates; a is the temperature influence coefficient, which reflects the influence degree of temperature change on the rail deformation correction, and is used to quantify the role of temperature factors, T is the finishing temperature, i.e. the temperature when the rail finishing process is completed, β is the pressure influence coefficient, which is used to quantify the parameter of the degree of rail hot deformation correction caused by the change of roll pressure; ΔF is the roll pressure adjustment, which is the value of adjusting the roll pressure in actual production, and F0 is the initial roll pressure, which is the pressure reference value set by the roll initially.

[0066] The mapping relationship between the rail deformation and the parameters is established, and an anti-overadjustment mechanism is set to realize the online monitoring and control of the rail bending and torsion.

[0067] In specific implementation, as a preferred embodiment of the present application, the establishment of the mapping relationship between the rail deformation and the parameters indicates that the roll pressure adjustment and the roll gap adjustment are obtained through historical data fitting.

[0068] The calculation formula of the roll pressure adjustment ΔF is:

[0069] ΔF=k1·θ1

[0070] The calculation formula of the roll gap adjustment ΔS is:

[0071] ΔS=k2·θ2

[0072] wherein k1, k2 are specification coefficients, optimized by machine learning.

[0073] The anti-overadjustment mechanism adjusts the gradient by setting parameters, such as an adjustment amount of ≤10% of the maximum range each time, and cooperates with feedback verification, and the deformation is retested within 1s after adjustment to ensure control stability.

[0074] The system architecture in the application is also optimized, and the optimized system comprises a perception layer, an algorithm layer, an application layer and a data tracing layer.

[0075] A finishing mill state sensor (roll pressure sensor, roll gap displacement sensor) is added to the perception layer, and the precision is ±0.5% FS and ±0.01 mm respectively, and the equipment operation parameters are collected in real time.

[0076] In the algorithm layer, the monitoring data and the finishing mill state are fused, the adjustment strategy is dynamically optimized by the PID control algorithm, the response time is <100 ms, and the thermal deformation compensation model is introduced into the finishing temperature (800-1200℃) influencing factor.

[0077] In the application layer, a linkage control interface is developed, the finishing mill parameters (pressure / roll gap / speed) and the rail deformation curve are displayed in real time, and manual intervention and automatic control mode switching are supported;

[0078] In the data tracing layer, linkage adjustment records (including time stamp, adjustment parameter, retest result) are stored, and the tracing accuracy is 0.1s.

[0079] The key steps of the system architecture and the finishing mill linkage include:

[0080] (1) Real-time monitoring: the scanner collects the full-section point cloud at a frequency of 150Hz, and completes the deformation calculation within 100ms;

[0081] (2) Deviation determination: compare θ 1 , θ 2 , Δθ with the preset threshold value, and generate an adjustment instruction (such as “increase the vertical roll pressure by 0.8kN / mm”);

[0082] (3) Parameter adjustment: send the instruction to the finishing mill PLC through PROFINET, and the PLC completes the roll pressure / roll gap adjustment within 500ms;

[0083] (4) Effect verification: continuously monitor for 3 scanning periods (300ms) after adjustment, and if the deviation does not converge, trigger secondary adjustment (10% increase).

[0084] Embodiment

[0085] In this embodiment, taking 60 kg / m rail hot rolling as an example, the rail bending and torsion online monitoring and control method based on the hot 3D full-face profile scanner in the application is used to control the finishing mill.

[0086] (1) Device and system linkage configuration

[0087] Finishing mill parameters: vertical roller initial pressure F = 500 kN / mm, horizontal roller gap S = 150 mm, rolling speed v = 10 m / s. v h

[0088] Scanner deployment: 8 cameras cover the full face of the rail, and an infrared thermal imager monitors the rail head temperature (1050±20℃) in real time.

[0089] (2) Linkage control example

[0090] Bending over-limit processing: detected vertical bending θ1 = 3.5° (threshold 3°), system calculates ΔF = 0.5×(3.5-3.0) = 0.25 kN / mm, sends instructions to PLC; PLC adjusts F to 500.25 kN / mm within 150 ms, 3s later, re-measured θ1 = 2.8°, meets the standard. v v

[0091] Torsion over-limit processing: if Δθ = 2.0° / m (threshold 1.5° / m), the system triggers the roller speed compensation +0.5%, while fine-tuning the horizontal roller gap ±0.3mm, 2s later, the torsion angle drops to 1.2° / m.

[0092] (3) Control effect verification

[0093] After 100 batches of testing, the rail end bending over-limit rate within 5m is reduced from 18% to 4.5%, and the torsion qualification rate is increased from 82% to 97% after the finishing mill linkage;

[0094] Control delay is stabilized at 180ms, and deformation convergence time is <5s after parameter adjustment, which is 10 times more efficient than traditional manual adjustment.

[0095] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the application, but not to limit them; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent substitutions for part or all of the technical features; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the application.​​​​

Claims

1. A rail bending and torsion online monitoring control method based on a hot state 3D full cross-section contour scanner, applied to a finishing mill, characterized in that, The application relates to a method for controlling the bending and twisting of a rail, and belongs to the technical field of rail rolling. The method comprises the following steps: acquiring data of the rail, including a bending angle, a twisting angle and temperature data; formulating a corresponding control strategy according to the data of the rail; dynamically optimizing the control strategy through a PID control algorithm in combination with monitoring data and the state of a finishing mill; 2. The rail bending and twisting online monitoring control method based on the hot state 3D full-face profile scanner according to claim 1, characterized in that, establishing a mapping relationship between rail deformation and parameters, and setting an anti-overadjustment mechanism to realize online monitoring and control of the bending and twisting of the rail. The data calculation method of the rail is as follows: The calculation formula of the bending angle theta is as follows: theta2 = arctan (Delta w / L) theta = theta1 / theta2 Wherein, theta1 represents a vertical bending angle, theta2 represents a horizontal bending angle, Delta h is a height difference value of the rail along a vertical direction, reflecting a vertical bending deformation degree; L is a length of a measured bending part of the rail, used as a reference length when the bending angle is calculated; Delta w is a width difference value of the rail along a horizontal direction, reflecting a width change caused by horizontal bending; Δθ = ∑arccos(n i ·n i+1 ) wherein n i represents the normal vector of the i-th section, the normal vector is perpendicular to the section, and is used to describe the spatial direction of the section; n i+1 represents the normal vector of the (i+1)-th section, and is adjacent to n i By calculating the relationship between the normal vectors of two adjacent sections, the torsion angle of the rail is determined.

3. The rail bend and twist online monitoring control method based on hot state 3D full profile scanner according to claim 1, characterized in that, The calculation formula of the twisting angle Delta theta is as follows: The control strategy according to the data of the rail comprises bending control and twisting control; The bending control means that when the vertical bending theta1 is greater than 3 DEG, the vertical roll pressure is increased by 0.5-1.0 kN / mm; and when the horizontal bending angle theta2 is greater than 2 DEG, the horizontal roll gap is adjusted by plus or minus 0.3 mm; 4. The rail bend and twist online monitoring control method based on hot state 3D full section profile scanner according to claim 1, characterized in that, The twisting control means that when the twisting angle Delta theta is greater than 1.5 DEG / m, the roll rotating speed compensation is triggered to correct the torque unevenness in the rolling process. The dynamic optimization of the control strategy through the PID control algorithm specifically comprises the following steps: The hot deformation compensation model is introduced into a finishing temperature influencing factor, and the correction formula is as follows: (X', Y', Z') = (X, Y, Z) * (1 + a * (T-20 DEG C) + beta * Delta F / F0) 5. The rail bend and twist online monitoring control method based on hot state 3D full profile scanner according to claim 1, characterized in that, Wherein, (X', Y', Z') represents the three-dimensional coordinate values of the corrected rail section profile in a three-dimensional space, used for representing the rail shape after considering the influences of hot deformation and pressure, (X, Y, Z) represents the three-dimensional coordinates of the uncorrected rail section profile; a is a temperature influencing coefficient, reflecting the influence degree of temperature change on the deformation correction of the rail, used for quantifying the influence of the temperature factor, T is a finishing temperature, that is, the temperature when the rail finishing process is completed, beta is a pressure influencing coefficient, used for quantifying the parameter of the deformation correction degree of the rail hot state caused by the change of the roll pressure; Delta F is a roll pressure adjustment value, which is the numerical value of the roll pressure adjustment in actual production, and F0 is an initial roll pressure, which is the pressure reference value set by the roll initially. The mapping relationship between the rail deformation and the parameters is obtained by fitting the historical data to obtain the roll pressure adjustment value and the roll gap adjustment value. The calculation formula of the roll pressure adjustment value Delta F is as follows: Delta F = k1 * theta1 The calculation formula of the roll gap adjustment value Delta S is as follows: Delta S = k2 * theta2 Wherein, k1 and k2 are specification coefficients. The anti-overadjustment mechanism sets a parameter adjustment gradient and cooperates with feedback checking to ensure the control stability.

Images