A method for intelligent control of billet grinding trolley

Abstract

This invention relates to the field of billet grinding trolley control technology, and specifically to an intelligent control method for a billet grinding trolley. The invention identifies surface defect areas on the billet, combines real-time trolley operating status information and grinding process parameters to establish a temporal mapping relationship between pressing depth and grinding trolley speed. It generates speed and grinding pressure adjustment commands based on the current position of the grinding trolley, monitors mechanical feedback signals during the grinding process, calculates the rate of change and directional deflection angle of the three-dimensional cutting force vector, and determines whether the grinding state is in a stable phase. Once the grinding state is stable, it assesses the uniformity of grinding depth distribution based on the actual grinding depth of the billet after the grinding operation, and, combined with historical data on similar billet grinding processes, corrects the speed and grinding pressure adjustment commands. This improves equipment operational stability, reduces repetitive grinding rates, and enhances the overall billet grinding efficiency of the grinding trolley.

Description

Technical Field

[0001] This invention relates to the field of grinding trolley control technology, and specifically to an intelligent control method for a steel billet grinding trolley. Background Technology

[0002] Steel billets are the basic raw material for rolling various types of steel, and their surface quality directly affects the appearance and performance of subsequent rolled products. During continuous casting, heating, and transportation, steel billets are prone to surface defects such as cracks, folds, scale, and scratches, which must be removed through a grinding process to ensure smooth rolling and meet finished product quality standards. The precision of coordinated control of parameters such as the running speed, grinding pressure, and reduction depth of the steel billet grinding trolley, as defect removal equipment, determines the grinding efficiency, grinding uniformity, and steel billet yield.

[0003] In existing technologies, the control of billet grinding trolleys mainly relies on a combination of preset fixed process parameters and manual adjustment based on experience. Existing billet grinding trolley control technology has the following technical problems: 1. Existing technologies use a fixed combination of trolley speed and grinding pressure, without performing three-dimensional identification and quantitative analysis of billet surface defects. This results in an inability to adapt and adjust based on the depth, area, and distribution characteristics of the defect area, leading to incomplete defect removal or over-grinding, increased billet loss, reduced billet yield, and higher subsequent rework costs.

[0004] 2. Existing technologies lack real-time status determination of mechanical feedback signals during the grinding process, making it impossible to identify and judge the stability of grinding. This leads to quality defects such as ripples and burns on the ground surface. Furthermore, it is impossible to dynamically correct control commands based on the actual grinding effect, resulting in poor grinding consistency and a high rate of repeated grinding, which makes it difficult to meet the requirements of continuous production of high-quality steel billets. Summary of the Invention

[0005] This invention aims to overcome the deficiencies in the prior art and provide an intelligent control method for billet grinding trolley, which realizes accurate three-dimensional defect identification, timing coordination of grinding parameters, mechanical stability determination of the grinding process, evaluation of grinding uniformity and closed-loop correction of commands, thereby comprehensively improving the quality and efficiency of billet grinding.

[0006] This invention is achieved through the following technical solution: This invention provides an intelligent control method for a billet grinding trolley, including the following steps: S1, performing a three-dimensional scan on the in-situ billet, identifying the information of the defect area on the billet surface, and constructing a multi-source dataset by combining the real-time operating status information of the billet grinding trolley and the grinding process parameters.

[0007] S2. Based on multi-source datasets, calculate the theoretical dwell time and theoretical grinding amount of the grinding trolley in each defect area, establish the time-series mapping relationship between pressing depth and grinding trolley speed, and generate trolley speed adjustment command and grinding pressure adjustment command in combination with the current position of the grinding trolley.

[0008] S3. Real-time monitoring of mechanical feedback signals during the grinding process, calculation of the rate of change and direction deflection angle of the three-dimensional cutting force vector, and determination of whether the grinding state is in a stable stage.

[0009] S4. Once the grinding state is in a stable phase, assess the uniformity of the grinding depth distribution based on the actual grinding depth distribution of the billet after the grinding operation is completed, combined with information on the defect areas on the billet surface.

[0010] S5. Based on the uniformity of grinding depth distribution and combined with historical data on grinding processes of similar steel billets, the trolley speed adjustment command and grinding pressure adjustment command are revised.

[0011] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention performs three-dimensional scanning on the in-situ steel billet to identify the defect area information on the surface of the steel billet, thereby accurately obtaining the defect location, depth, area and type, providing a data basis for the accurate calculation of subsequent grinding parameters, avoiding the problem of incomplete defect removal or excessive grinding, and improving the steel billet yield and grinding effectiveness.

[0012] (2) This invention calculates the theoretical dwell time and theoretical grinding amount of the grinding trolley through each defect area, establishes a time-series mapping relationship between the pressing depth and the speed of the grinding trolley, and generates trolley speed adjustment command and grinding pressure adjustment command in combination with the current position of the grinding trolley, so as to realize synchronous and coordinated adjustment of grinding pressure and running speed, improve the smoothness of the grinding process, ensure consistent grinding depth, and reduce the surface quality fluctuation of steel billet and grinding cost.

[0013] (3) By real-time monitoring of the mechanical feedback signal during the grinding operation, the present invention calculates the rate of change and direction deflection angle of the three-dimensional cutting force vector, determines whether the grinding state is in a stable stage, can identify abnormal grinding conditions in a timely manner, effectively avoid quality defects such as surface burns and ripples caused by unstable grinding, improve the safety and stability of equipment operation, and extend the service life of grinding equipment.

[0014] (4) Based on the actual grinding distribution depth of the billet after the grinding operation is completed, combined with the information on the defect area of ​​the billet surface, the present invention evaluates the uniformity of the grinding depth distribution, and combined with historical data of similar billet grinding processes, corrects the trolley speed adjustment command and the grinding pressure adjustment command, effectively improving the consistency and stability of similar billet grinding processes, realizing the effective accumulation and continuous improvement of grinding process parameters, thereby reducing the repeated grinding rate and improving the overall billet grinding efficiency of the grinding trolley. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the method steps of the present invention;

[0017] Figure 2 This is a schematic diagram of the process for determining whether the grinding state is in a stable stage in this invention;

[0018] Figure 3 This is a schematic diagram of the steps for evaluating the uniformity of grinding depth distribution in this invention. Detailed Implementation

[0019] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. Furthermore, it should be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale.

[0020] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification.

[0021] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0022] Please see Figure 1 As shown, the present invention provides an intelligent control method for a billet grinding trolley, including: S1, performing a three-dimensional scan on the in-situ billet to identify the surface defect area information of the billet, and constructing a multi-source dataset by combining the real-time operating status information of the billet grinding trolley and the grinding process parameters.

[0023] S2. Based on multi-source datasets, calculate the theoretical dwell time and theoretical grinding amount of the grinding trolley in each defect area, establish the time-series mapping relationship between pressing depth and grinding trolley speed, and generate trolley speed adjustment command and grinding pressure adjustment command in combination with the current position of the grinding trolley.

[0024] S3. Real-time monitoring of mechanical feedback signals during the grinding process, calculation of the rate of change and direction deflection angle of the three-dimensional cutting force vector, and determination of whether the grinding state is in a stable stage.

[0025] S4. Once the grinding state is in a stable phase, assess the uniformity of the grinding depth distribution based on the actual grinding depth distribution of the billet after the grinding operation is completed, combined with information on the defect areas on the billet surface.

[0026] S5. Based on the uniformity of grinding depth distribution and combined with historical data on grinding processes of similar steel billets, the trolley speed adjustment command and grinding pressure adjustment command are revised.

[0027] Considering that accurate three-dimensional identification of surface defects in steel billets is a prerequisite for the adaptive adjustment of grinding parameters, existing technologies rely solely on manual visual inspection or two-dimensional image detection without performing three-dimensional quantitative analysis of defects. This makes it impossible to obtain accurate data on defect depth, area, and spatial distribution, leading to a mismatch between grinding parameters and defect characteristics. Consequently, this results in incomplete defect removal or over-grinding. Therefore, it is necessary to use three-dimensional point cloud scanning and surface reconstruction technology to achieve high-precision geometric modeling and feature extraction of the defect area, providing a reliable data foundation for subsequent grinding parameter calculations.

[0028] Based on this, the specific implementation of identifying surface defect area information of steel billets in this invention includes: S11, acquiring three-dimensional point cloud data of the in-situ steel billet surface using a three-dimensional scanning device (such as a line laser scanner or structured light scanner), filtering and denoising the three-dimensional point cloud data to remove outlier noise points, and generating a processed point cloud dataset. In this invention, the in-situ steel billet refers to a steel billet that has been fixed in the working area of ​​the grinding trolley and is ready for surface grinding.

[0029] Preferably, the filtering and noise reduction process employs a radius filtering algorithm, setting a search radius r (e.g., 5 mm) and a neighboring point number threshold N (e.g., 10). If the number of neighboring points of a point cloud within its search radius is less than N, it is determined to be a noise point and removed. The radius filtering algorithm is a well-known existing technique, and will not be elaborated upon here.

[0030] S12. Based on the processed point cloud dataset, surface reconstruction is performed to obtain a three-dimensional model of the billet. The curvature values ​​of each measuring point on the surface of the billet and the height difference between adjacent measuring points are extracted to identify the defect areas on the surface of each billet.

[0031] Preferably, in a specific embodiment of the present invention, the method for identifying the defect areas on the surface of each steel billet is as follows: if the curvature value of a certain measuring point on the surface of the steel billet in the three-dimensional model exceeds the set defect curvature reference range, the set defect curvature reference range is determined according to the steel billet material: for carbon steel, the defect curvature reference range is... For alloy steel, the defect curvature reference range is: If the height difference between adjacent measuring points is greater than the set allowable height difference of the billet, for example, 0.5mm, then the measuring point is marked as a defect measuring point.

[0032] Based on the location of all defect measurement points, the connected domains of all defect measurement points are merged to form the defect regions on the surface of each steel billet.

[0033] S13. Extract the boundary contour of the defect area on the steel billet surface, obtain the depth and area of ​​the defect area, and match the corresponding defect type to the defect area by combining the feature library of steel billet surface defect types.

[0034] S14. Determine the center position of the defect area based on the boundary contour of the defect area, and combine the defect type, depth and area of ​​the defect area to form the defect area information of the steel billet surface.

[0035] It should be noted that the billet surface defect type feature library includes area and depth feature templates for defect types such as cracks, folds, scabs, and scratches. By comparing the depth and area of ​​the defect area with the area and depth feature templates of all defect types, the Euclidean distance between the defect area and all defect types is obtained, and the defect type with the smallest Euclidean distance is selected as the corresponding defect type of the defect area.

[0036] This invention uses three-dimensional scanning of in-situ steel billets to identify defect areas on the billet surface, thereby accurately obtaining the location, depth, area, and type of defects. This provides a data basis for the accurate calculation of subsequent grinding parameters, avoids the problem of incomplete defect removal or over-grinding, and improves the billet yield and grinding effectiveness.

[0037] Considering that the coordinated control of grinding trolley speed and grinding pressure directly affects grinding quality and efficiency, the existing technology uses a fixed combination of trolley speed and grinding pressure without dynamic adaptation according to defect characteristics, resulting in poor smoothness and low depth consistency in the grinding process. It is necessary to establish a time-series mapping relationship between pressing depth and trolley speed to achieve adaptive adjustment and smooth transition of grinding parameters, and ensure that the grinding process is stable and controllable.

[0038] Based on this, the specific implementation of calculating the theoretical dwell time and theoretical grinding amount of the grinding trolley through each defect area in this invention includes: S21, extracting the depth and area of ​​each defect area from the multi-source dataset, comparing them with the allowable grinding depth deviation range, determining the theoretical grinding depth of each defect area, and determining the theoretical grinding amount of the grinding trolley through each defect area in combination with the area of ​​the defect area.

[0039] Preferably, in a specific embodiment of the present invention, the allowable grinding depth deviation range is dynamically set according to the billet yield requirement, generally ±(5%-15%) of the defect depth. When the yield requirement is high, a smaller value is taken. In the present invention, the allowable grinding depth deviation range is set to ±10% of the defect depth. If the defect depth is d, the theoretical grinding depth is taken as 1.1d to ensure complete removal of the defect. The formula for calculating the theoretical grinding amount V of the grinding trolley passing through the defect area is: ,in 1.1d represents the defect area, and 1.1d represents the theoretical grinding depth, thus achieving precise quantification of the grinding volume.

[0040] S22. Based on the rated speed of the grinding head and the specifications of the grinding wheel in the grinding process parameters, determine the amount of material to be ground per unit time. Calculate the ratio between the theoretical grinding amount of each defect area and the material grinding amount per unit time to obtain the theoretical processing time required for the grinding trolley to pass through each defect area.

[0041] It should be noted that the method for determining the amount of material regrinding per unit time is derived from the standard formula for grinding volume removal rate, which is: .

[0042] In the formula, The linear velocity of the grinding wheel. ,in Pi; This refers to the diameter of the grinding wheel, in mm. The rated speed of the grinding head is expressed in r / min; This refers to the effective dressing width of the grinding wheel, expressed in mm. For example, it can be taken as 80% of the grinding wheel width. The theoretical grinding depth is expressed in mm. The material removal efficiency coefficient of the grinding wheel is determined based on the grinding wheel grit size and hardness, and its value range is [value range missing]. .

[0043] Therefore, based on the rated speed of the grinding head in the grinding process parameters and the grinding wheel diameter, effective grinding width, and grinding wheel material removal efficiency coefficient in the grinding wheel specifications, the material grinding amount per unit time is obtained by substituting them into the standard formula for grinding volume removal rate.

[0044] S23. Based on the reference running speed of the grinding trolley in the non-defect area and the center position of each defect area, calculate the theoretical travel time required for the grinding trolley to pass through each defect area. Compare the theoretical processing time with the theoretical travel time, and select the maximum value as the theoretical dwell time of the grinding trolley in each defect area. This ensures that the grinding is sufficient but not excessive.

[0045] Preferably, the specific implementation of the method for establishing the time-series mapping relationship between the pressing depth and the grinding trolley speed in the present invention includes: S24, obtaining the boundary contour area of ​​each defect area on the surface of the billet, and determining the longitudinal start position and longitudinal end position of the boundary contour area.

[0046] S25. Based on the theoretical dwell time of the grinding trolley in each defect area, calculate the target grinding speed of the grinding trolley in each defect area and generate the uniform speed curve of the grinding section.

[0047] It should be noted that, based on the longitudinal start position and longitudinal end position of the boundary contour area, the longitudinal coverage distance of the boundary contour area is obtained. The longitudinal coverage distance is then compared with the theoretical dwell time of the grinding trolley in each defect area to obtain the target grinding speed of the grinding trolley in each defect area. Within the interval from the longitudinal start position to the longitudinal end position, the grinding trolley runs at a constant target grinding speed.

[0048] S26. Based on the benchmark operating speed, determine the speed transition section of the grinding trolley before and after each defect area, and use curve interpolation to generate the speed change curve of the speed transition section.

[0049] Preferably, in a specific embodiment of the present invention, the method of generating the speed change curve of the speed transition segment by curve interpolation is as follows: First, obtain the squared difference between the reference running speed and the target grinding speed, and combine the rated starting acceleration and rated braking deceleration of the grinding trolley drive system to calculate the acceleration distance and deceleration distance using the distance-acceleration formula.

[0050] Secondly, a deceleration transition section is set before the longitudinal starting position in front of the defect area according to the deceleration distance, and an acceleration transition section is set after the longitudinal ending position behind the defect area according to the acceleration distance.

[0051] Finally, using the reference operating speed as the initial speed and the target grinding speed as the final speed, combined with the rated braking deceleration, cubic spline interpolation is used to generate the speed change curve of the deceleration transition section; similarly, using the target grinding speed as the initial speed and the reference operating speed as the final speed, combined with the rated starting acceleration, cubic spline interpolation is used to generate the speed change curve of the acceleration transition section.

[0052] S27. Smoothly splice the speed transition sections before and after each defect area with the grinding section to obtain the speed sequence of the grinding trolley.

[0053] S28. Based on the theoretical grinding amount of the grinding trolley passing through each defect area, determine the pressing depth sequence of the grinding trolley in each defect area, and combine it with the speed sequence of the grinding trolley to establish a time-series mapping relationship between the pressing depth and the speed of the grinding trolley.

[0054] It should be noted that the method for establishing the time-series mapping relationship between the pressing depth and the grinding trolley speed is as follows: S281, calculate the material removal and distribution amount per unit longitudinal coverage length based on the theoretical grinding amount of the defect area and the longitudinal coverage length of the boundary contour area.

[0055] S282. Based on the cross-sectional area of ​​the defect area per unit longitudinal coverage length, the ratio of the material removal and distribution amount per unit longitudinal coverage length to the cross-sectional area of ​​the defect area is calculated to obtain the incremental pressing depth of the grinding head in the longitudinal feed direction.

[0056] S283. Using the longitudinal starting position of the defect area as the starting point of the compression and the longitudinal ending position of the defect area as the ending point of the compression, a compression depth sequence is generated by accumulating point by point according to the compression depth increment.

[0057] S284. Extract the timestamps corresponding to each coordinate point in the speed sequence of the grinding trolley, align each pressing depth value in the pressing depth sequence with the timestamp of the corresponding coordinate point, and establish a temporal mapping relationship between pressing depth and grinding trolley speed.

[0058] Preferably, the method for generating the trolley speed adjustment command and the grinding pressure adjustment command in the present invention is as follows: First, based on the current position of the grinding trolley, the target grinding trolley speed and target pressing depth corresponding to the current position of the grinding trolley are selected from the time-series mapping relationship between pressing depth and grinding trolley speed.

[0059] Secondly, the real-time running speed of the grinding trolley is obtained, and the difference between the real-time running speed and the target grinding trolley speed is compared. Based on the speed difference, a trolley speed adjustment command is generated.

[0060] Then, if the grinding trolley is currently in the grinding section of a defect area, the current pressing depth of the grinding trolley is obtained, and the depth difference between the current pressing depth and the target pressing depth is obtained by comparison.

[0061] Finally, based on the characteristic curves of hydraulic pressure and pressing depth of the grinding head corresponding to the grinding trolley, the regulating hydraulic pressure corresponding to the depth difference is determined, and the grinding pressure regulating command is generated.

[0062] It should be noted that the hydraulic pressure and pressing depth characteristic curves of the grinding head corresponding to the grinding trolley are calibrated by the factory test of the grinding trolley, which characterize the steady-state hydraulic pressure requirements corresponding to different pressing depths.

[0063] This invention calculates the theoretical dwell time and theoretical grinding amount of the grinding trolley in each defect area, establishes a time-series mapping relationship between the pressing depth and the speed of the grinding trolley, and generates trolley speed adjustment commands and grinding pressure adjustment commands in combination with the current position of the grinding trolley. This enables synchronous and coordinated adjustment of grinding pressure and running speed, improves the smoothness of the grinding process, ensures consistent grinding depth, and reduces surface quality fluctuations and grinding costs of steel billets.

[0064] Considering the fluctuations in cutting force caused by factors such as grinding wheel wear and uneven hardness of steel billets during the grinding process, failure to monitor grinding stability in real time will prevent the identification of abnormal grinding conditions, leading to quality defects such as burns and ripples on the ground surface, and even equipment damage. Furthermore, monitoring only the force signal in a single direction will not fully reflect the spatial variation characteristics of the three-dimensional cutting state. Therefore, dynamic analysis of the three-dimensional cutting force vector is necessary to accurately determine grinding stability.

[0065] Based on this, such as Figure 2 As shown, the method for determining whether the grinding state is in a stable stage in this invention is as follows: S31, by using the force sensor array built into the grinding head, the mechanical feedback signal during the grinding operation is monitored in real time, wherein the mechanical feedback signal includes the cutting force components in the vertical direction, the feed direction and the transverse swing direction, forming a three-dimensional cutting force vector.

[0066] S32. Compare the three-dimensional cutting force vector at the current acquisition time with that at its adjacent acquisition time, and calculate the rate of change and direction deflection angle of the three-dimensional cutting force vector.

[0067] It should be noted that the rate of change of the three-dimensional cutting force vector is calculated as follows: the difference in magnitude between the three-dimensional cutting force vector at the current acquisition time and the adjacent acquisition time is calculated, and the difference in magnitude is compared with the acquisition time interval. The ratio is taken as the rate of change of the three-dimensional cutting force vector.

[0068] The angle between the three-dimensional cutting force vectors at the current acquisition time and those at adjacent acquisition times is calculated by vector dot product, and this angle is used as the directional deflection angle.

[0069] S33. If the rate of change and the direction deflection angle of the three-dimensional cutting force vector are both within the preset stable fluctuation range, the grinding state is determined to be in a stable stage; otherwise, the grinding state is determined to be in an unstable stage, triggering an abnormal warning and suspending the grinding operation, waiting for manual intervention to adjust the grinding parameters.

[0070] Preferably, in a specific embodiment of the present invention, the preset stable fluctuation range is determined by statistical analysis of the force signals of the historical stable grinding stage, taking the 95th percentile of the force change rate distribution in the stable stage as the stable fluctuation range of the change rate, and taking the 95th percentile of the direction deflection angle distribution as the stable fluctuation range of the direction deflection angle.

[0071] This invention monitors the mechanical feedback signals during the grinding process in real time, calculates the rate of change and directional deflection angle of the three-dimensional cutting force vector, and determines whether the grinding state is in a stable stage. It can promptly identify abnormal grinding conditions, effectively avoid quality defects such as surface burns and ripples caused by grinding instability, improve equipment operation safety and stability, and extend the service life of grinding equipment.

[0072] Considering that without a quantitative evaluation of the actual grinding effect after grinding, it will be impossible to identify quality problems such as uneven grinding depth and localized missed grinding, leading to residual defects or reduced yield; and if only single-point measurement or visual inspection is used, it will be difficult to obtain grinding depth distribution information for the entire surface. Therefore, it is necessary to achieve accurate measurement and uniformity evaluation of grinding depth by registering and aligning the 3D point clouds before and after grinding.

[0073] Based on this, such as Figure 3 As shown, the evaluation method for the uniformity of grinding depth distribution in this invention is as follows: S41, the three-dimensional point cloud data of the steel billet surface after the grinding operation is completed is obtained by scanning again with a three-dimensional scanning device, and the data is registered and aligned with the point cloud data of the corresponding defect area before grinding in the defect area information of the steel billet surface. The height difference of each point cloud position is calculated to obtain the actual grinding distribution depth of the steel billet.

[0074] S42. Extract the mean, range, and standard deviation of grinding depth from the actual grinding distribution depth of the billet. Based on the mean and standard deviation of grinding depth, use the 3σ criterion to determine the allowable range of grinding depth deviation.

[0075] S43. If the difference in grinding depth is within the allowable range of grinding depth deviation, the uniformity of grinding depth distribution is deemed acceptable; otherwise, the uniformity of grinding depth distribution is deemed unacceptable.

[0076] This invention achieves full-surface quantitative evaluation of grinding depth distribution by registering and aligning three-dimensional point cloud data before and after grinding, obtaining the uniformity of grinding depth distribution, realizing quantitative evaluation of grinding quality, providing data basis for subsequent process parameter optimization, and effectively improving the scientific nature of grinding quality evaluation.

[0077] Considering that there may be parameter deviations in a single grinding process, if the process parameters are not continuously optimized without comparison and correction with historical high-quality process data, it will be impossible to achieve good consistency in grinding of similar steel billets. Therefore, it is necessary to combine historical grinding process data of similar steel billets to achieve closed-loop correction of the current process parameters, thereby continuously optimizing the grinding process.

[0078] Based on this, the method for correcting the trolley speed adjustment command and the grinding pressure adjustment command in this invention is as follows: S51, when the uniformity of grinding depth distribution is not up to standard, select historical grinding process records with similar steel billet surface defect area information and uniformity of grinding depth distribution from the steel billet historical grinding database.

[0079] S52. Extract the speed and pressing depth sequences of the grinding trolley from all historical grinding process records, and perform a difference analysis between them and the current speed and pressing depth sequences of the grinding trolley during the grinding operation.

[0080] S53. Based on the difference analysis results, the trolley speed adjustment command and the trolley pressure adjustment command during the grinding operation are corrected, and the corrected trolley speed adjustment command and the trolley pressure adjustment command are generated.

[0081] Preferably, in a specific embodiment of the present invention, the method for correcting the trolley speed adjustment command and the grinding pressure adjustment command is as follows: S531, based on the grinding trolley speed recorded in all historical grinding processes, determine the grinding trolley speed reference range corresponding to the historical grinding process.

[0082] S532. If the current grinding trolley speed is outside the grinding trolley speed reference range, then obtain the average value of the grinding trolley speed reference range and use it as the corrected trolley speed adjustment command; otherwise, there is no need to correct the trolley speed adjustment command.

[0083] S533. Use cosine similarity calculation to obtain the similarity between the pressing depth sequence of all historical grinding process records and the current pressing depth sequence, filter the historical pressing depth sequence with the highest similarity, and obtain the depth difference between the historical pressing depth sequence and the current pressing depth sequence.

[0084] S534. Based on the depth difference, correct the pressing depth at the corresponding position in the current pressing depth sequence, and generate a corrected grinding pressure adjustment command.

[0085] This invention assesses the uniformity of grinding depth distribution based on the actual grinding depth distribution of the billet after the grinding operation is completed, combined with information on the defect area on the billet surface. It also modifies the trolley speed adjustment command and the grinding pressure adjustment command by combining historical data of similar billet grinding processes. This effectively improves the consistency and stability of similar billet grinding processes, enables the effective accumulation and continuous improvement of grinding process parameters, thereby reducing the rate of repeated grinding and improving the overall billet grinding efficiency of the grinding trolley.

[0086] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0087] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0088] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0089] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0090] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent control of a billet grinding trolley, characterized in that, include: Three-dimensional scanning of in-situ steel billets was performed to identify surface defect areas. Combined with real-time operating status information of the steel billet grinding trolley and grinding process parameters, a multi-source dataset was constructed. Based on multi-source datasets, the theoretical dwell time and theoretical grinding amount of the grinding trolley in each defect area are calculated. A time-series mapping relationship between pressing depth and grinding trolley speed is established. Combined with the current position of the grinding trolley, trolley speed adjustment command and grinding pressure adjustment command are generated. Real-time monitoring of mechanical feedback signals during the grinding process, calculation of the rate of change and direction deflection angle of the three-dimensional cutting force vector, and determination of whether the grinding state is in a stable stage; Once the grinding state has reached a stable stage, the uniformity of the grinding depth distribution is evaluated based on the actual grinding depth distribution of the billet after the grinding operation is completed, combined with information on the defect area on the billet surface. Based on the uniformity of grinding depth distribution and combined with historical data on similar steel billet grinding processes, the trolley speed adjustment command and the grinding pressure adjustment command were revised. The method for identifying surface defect areas of steel billets is as follows: The three-dimensional point cloud data of the surface of the steel billet is obtained by a three-dimensional scanning device, and the three-dimensional point cloud data is filtered and noise reduced to generate a processed point cloud dataset. Based on the processed point cloud dataset, a three-dimensional model of the billet is obtained by surface reconstruction. The curvature values ​​of each measuring point on the surface of the billet and the height difference between adjacent measuring points are extracted to identify the defect areas on the surface of each billet. The boundary contour of the defect area on the steel billet surface is extracted to obtain the depth and area of ​​the defect area. Combined with the feature library of steel billet surface defect types, the defect area is matched with the corresponding defect type. The center position of the defect area is determined based on the boundary contour of the defect area. Combined with the defect type, the depth and area of ​​the defect area, the information of the defect area on the surface of the steel billet is constructed. The method for calculating the theoretical dwell time and theoretical grinding amount of the grinding trolley in each defect area is as follows: The depth and area of ​​each defect region are extracted from the multi-source dataset, compared with the allowable grinding depth deviation range, to determine the theoretical grinding depth of each defect region. Combined with the area of ​​the defect region, the theoretical grinding amount of the grinding trolley passing through each defect region is determined. Based on the rated speed of the grinding head and the specifications of the grinding wheel in the grinding process parameters, the amount of material grinding per unit time is determined. The theoretical grinding amount of each defect area is calculated by ratio to the amount of material grinding per unit time, and the theoretical processing time required for the grinding trolley to pass through each defect area is obtained. Based on the reference running speed of the grinding trolley in the non-defect area and the center position of each defect area, the theoretical travel time required for the grinding trolley to pass through each defect area is calculated. The theoretical processing time is compared with the theoretical travel time, and the maximum value is selected as the theoretical dwell time of the grinding trolley in each defect area.

2. The intelligent control method for a billet grinding trolley according to claim 1, characterized in that, The method for identifying defect areas on the surface of each steel billet is as follows: If the curvature value of a certain measuring point on the surface of the billet exceeds the set defect curvature reference range, or the height difference between its adjacent measuring points is greater than the set allowable height difference of the billet, then the measuring point is marked as a defect measuring point. Based on the location of all defect measurement points, the connected domains of all defect measurement points are merged to form the defect regions on the surface of each steel billet.

3. The intelligent control method for a billet grinding trolley according to claim 1, characterized in that, The time-series mapping relationship between the pressing depth and the grinding trolley speed is established as follows: Obtain the boundary contour area of ​​each defect region on the surface of the steel billet, and determine the longitudinal start position and longitudinal end position of the boundary contour area; Based on the theoretical dwell time of the grinding trolley in each defect area, the target grinding speed of the grinding trolley in each defect area is calculated, and a uniform speed curve of the grinding section is generated. Based on the benchmark operating speed, the speed transition section of the grinding trolley before and after each defect area is determined, and the speed change curve of the speed transition section is generated by curve interpolation. The speed transition sections before and after each defect area are smoothly spliced ​​with the grinding section to obtain the speed sequence of the grinding trolley. Based on the theoretical grinding amount of the grinding trolley passing through each defect area, the pressing depth sequence of the grinding trolley in each defect area is determined. Combined with the speed sequence of the grinding trolley, a time-series mapping relationship between the pressing depth and the speed of the grinding trolley is established.

4. The intelligent control method for a billet grinding trolley according to claim 3, characterized in that: The method for generating the trolley speed adjustment command and the grinding pressure adjustment command is as follows: Based on the current position of the grinding trolley, the target grinding trolley speed and target pressing depth corresponding to the current position of the grinding trolley are selected from the time-series mapping relationship between pressing depth and grinding trolley speed. The real-time running speed of the grinding trolley is obtained, and the difference between the real-time running speed and the target grinding trolley speed is compared. Based on the speed difference, a trolley speed adjustment command is generated. If the current position of the grinding trolley is in the grinding section of a defect area, obtain the current pressing depth of the grinding trolley and compare it with the depth difference between the current pressing depth and the target pressing depth. Based on the characteristic curves of hydraulic pressure and pressing depth of the grinding head corresponding to the grinding trolley, the regulating hydraulic pressure corresponding to the depth difference is determined, and the grinding pressure regulating command is generated.

5. The intelligent control method for a billet grinding trolley according to claim 1, characterized in that: The method for determining whether the grinding state is in a stable stage is as follows: The mechanical feedback signal during the grinding operation is monitored in real time by the built-in force sensor array in the grinding head. The mechanical feedback signal includes the cutting force components in the vertical direction, feed direction and lateral swing direction, which constitute a three-dimensional cutting force vector. The current acquisition time is compared with the three-dimensional cutting force vector of its adjacent acquisition time, and the rate of change and direction deflection angle of the three-dimensional cutting force vector are calculated. If the rate of change and the direction deflection angle of the three-dimensional cutting force vector are both within the preset stable fluctuation range, then the grinding state is determined to be in a stable stage; otherwise, the grinding state is determined to be in an unstable stage.

6. The intelligent control method for a billet grinding trolley according to claim 1, characterized in that: The method for evaluating the uniformity of the grinding depth distribution is as follows: The three-dimensional point cloud data of the steel billet surface after the grinding operation is obtained by scanning again with a three-dimensional scanning device. The data is then registered and aligned with the point cloud data of the corresponding defect area before grinding in the defect area information of the steel billet surface. The height difference of each point cloud position is calculated to obtain the actual grinding distribution depth of the steel billet. Extract the mean, range, and standard deviation of grinding depth from the actual grinding distribution depth of the steel billet, and determine the allowable grinding depth deviation range based on the mean and standard deviation of grinding depth. If the difference in grinding depth is within the allowable range of grinding depth deviation, the uniformity of grinding depth distribution is deemed acceptable; otherwise, the uniformity of grinding depth distribution is deemed unacceptable.

7. The intelligent control method for a billet grinding trolley according to claim 6, characterized in that: The method for correcting the trolley speed adjustment command and the grinding pressure adjustment command is as follows: When the uniformity of grinding depth distribution is not up to standard, the historical grinding process records with similar surface defect areas and uniform grinding depth distribution are selected from the billet historical grinding database. Extract the sequence of grinding trolley speed and pressing depth from all historical grinding process records, and perform a difference analysis between them and the current sequence of grinding trolley speed and pressing depth during the grinding operation. Based on the difference analysis results, the trolley speed adjustment command and the grinding pressure adjustment command during the grinding operation are corrected to generate the corrected trolley speed adjustment command and the grinding pressure adjustment command.

8. The intelligent control method for a billet grinding trolley according to claim 7, characterized in that: The method for correcting the trolley speed adjustment command and the grinding pressure adjustment command is as follows: Based on the grinding trolley speeds recorded for all historical grinding processes, a reference range for the grinding trolley speeds corresponding to the historical grinding processes is determined. If the current grinding trolley speed is outside the grinding trolley speed reference range, then the average value of the grinding trolley speed reference range is obtained and used as the corrected trolley speed adjustment command; otherwise, there is no need to correct the trolley speed adjustment command. Obtain the similarity between the pressing depth sequence of all historical grinding process records and the current pressing depth sequence, filter the historical pressing depth sequence with the highest similarity, and obtain the depth difference between the historical pressing depth sequence and the current pressing depth sequence; The pressing depth at the corresponding position in the current pressing depth sequence is corrected based on the depth difference, and a corrected grinding pressure adjustment command is generated.

Images