A strand surface intelligent guiding grinding system and method

Abstract

The present application belongs to the technical field of metallurgical engineering, and relates to a kind of casting blank surface intelligent guide grinding system and method.The present application is provided with parameter initialization module, parameter compensation module and deviation correction module, which solves the problem of unstable grinding quality caused by grinding force fluctuation, temperature change and material difference in the process of casting blank surface grinding.Specifically, the feed speed is dynamically adjusted by real-time acquisition of grinding force data, the thermal deformation compensation is carried out according to the casting blank temperature and steel characteristics, and the upper limit value is determined based on the quantitative analysis of surface state parameters and historical qualified samples, to realize the adaptive adjustment of the next round of grinding parameters.The present application effectively improves the consistency of grinding depth, prevents the risk of over-grinding during high-temperature casting blank grinding, realizes the closed-loop optimization and continuous improvement of grinding process, improves the stability and reliability of casting blank surface grinding quality, and ensures that the casting blank will not affect the product quality due to surface defects in the subsequent rolling process.

Description

Technical Field

[0001] This invention belongs to the field of metallurgical engineering technology and relates to an intelligent guided grinding system and method for the surface of cast billets. Background Technology

[0002] During the billet production process, defects such as scale, slag inclusions, protruding iron oxide scale, and burrs caused by mechanical scratches often occur on the billet surface due to various process reasons. If these surface protrusion defects are not accurately and uniformly removed before rolling, they will be pressed into the matrix during the subsequent high-pressure rolling process, resulting in product surface quality problems and affecting production safety and product yield.

[0003] In the prior art, such as the efficient identification and finishing method for surface defects of cast billets proposed in Chinese Invention Patent Publication No. CN114260759A, the automatic identification and positioning of defects is achieved by combining magnetic particle display and image recognition, and the grinding wheel group is controlled to perform fixed-point grinding, thereby improving the identification efficiency and automation level.

[0004] However, this method still has the following limitations in practical applications: (1) This scheme relies on preset grinding wheel pressure and time parameters, and lacks the ability to dynamically perceive and adjust the grinding process. In the actual grinding process, the grinding force will continue to fluctuate due to factors such as grinding wheel wear and different defect morphologies. If this fluctuation is not responded to in time, it will easily cause uneven grinding depth and affect the surface grinding quality. However, this scheme does not involve a real-time monitoring and feedback mechanism for grinding force. Therefore, when the actual working conditions do not match the preset conditions, it is difficult to avoid the problem of uneven grinding depth or unstable quality.

[0005] (2) This scheme does not incorporate the actual temperature of the billet and the characteristics of the steel grade into the decision-making logic of the grinding parameters. In actual production lines, the billet temperature is relatively high, and different steel grades have different hardness and grinding characteristics at high temperatures. Billets with higher temperatures or softer materials are more likely to be over-grinded under the same grinding parameters, which may lead to over-grinding. This scheme adopts a uniform parameter control mode and does not consider these changes caused by temperature and material, which affects the control accuracy of the grinding depth.

[0006] (3) The quality assessment process of this scheme is inadequate and cannot form a closed loop for process optimization. Although the scheme can be re-inspected by taking photos twice, its judgment logic is essentially a qualitative judgment on the presence or absence of defects, rather than a quantitative assessment based on key fine-grained parameters such as roughness and residual defect density. This makes it impossible for the system to accurately assess the degree of inadequacy of this round of grinding, and thus it is difficult to provide a precise basis for adjusting the grinding parameters of the next round of the same type of billet, which limits its adaptive and continuous optimization capabilities in mass production. Summary of the Invention

[0007] In view of this, in order to solve the problems mentioned in the background technology, a smart guided grinding system and method for casting surface is proposed.

[0008] The objective of this invention can be achieved through the following technical solution: The first aspect of this invention proposes an intelligent guided grinding system for the surface of a cast billet, comprising the following modules: a parameter initialization module, used to obtain information on the area to be ground and the grinding target value on the surface of the cast billet, divide the grinding process into at least one grinding stage, and configure initial grinding parameters suitable for the first grinding stage.

[0009] The parameter compensation module is used to dynamically correct the initial grinding parameters based on the real-time collected grinding force data during the current grinding stage to obtain the real-time grinding parameters. It also performs thermal deformation compensation on the real-time grinding parameters according to the steel material characteristics of the billet to obtain the compensated grinding parameters. The compensated grinding parameters are then used to grind the area to be ground.

[0010] The deviation correction module is used to detect the surface condition parameters of the ground area after the current grinding cycle is completed, and determine the upper limit of the surface condition parameters based on the historical compliance samples of the same steel grade billet. When the surface condition parameters are greater than the upper limit, the absolute deviation between the surface condition parameters and the grinding compliance value is calculated, and the initial grinding parameters for the next grinding cycle are adaptively adjusted based on the absolute deviation.

[0011] The second aspect of the present invention provides an intelligent guided grinding method for the surface of a cast billet, comprising the following steps: obtaining information on the area to be ground and the grinding target value on the surface of the cast billet, dividing the grinding process into at least one grinding stage, and configuring initial grinding parameters suitable for the first grinding stage.

[0012] During this round of regrinding, the initial regrinding parameters are dynamically corrected based on the real-time collected grinding force data to obtain the real-time regrinding parameters.

[0013] Based on the steel grade and material characteristics of the billet, the real-time grinding parameters are thermally deformed to obtain the compensated grinding parameters, and the area to be ground is ground using the compensated grinding parameters.

[0014] After this round of grinding is completed, the surface condition parameters of the ground area are tested, and the upper limit of the surface condition parameters is determined based on the historical compliance samples of the same steel grade billet.

[0015] When the surface condition parameter is greater than its upper limit, the absolute deviation between the surface condition parameter and the grinding target value is calculated, and the initial grinding parameters for the next round of grinding are adaptively adjusted based on the absolute deviation.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention performs sliding window analysis on the grinding force time sequence data through the parameter compensation module, dynamically identifies the disturbance state and adjusts the feed speed in real time, effectively solving the problem of grinding force fluctuation caused by grinding wheel wear and defect morphology changes during traditional grinding. Furthermore, by establishing an adaptive correlation between process parameters and dynamic working conditions, the consistency of grinding depth is significantly improved, ensuring uniform surface treatment of the cast billet, avoiding local over-grinding or under-grinding, and greatly improving the stability and reliability of grinding quality.

[0017] (2) This invention combines non-contact infrared thermometry with a material property database to set differentiated thermal deformation compensation depths for different temperature ranges, overcoming the risk of over-grinding caused by material softening during the grinding process of high-temperature cast billets. This thermal deformation compensation mechanism dynamically adjusts the compensation ratio according to the temperature range of the cast billet surface; the higher the temperature, the greater the compensation ratio, thus achieving precise dynamic adjustment of grinding parameters, improving the control accuracy of different grinding depths, and preventing surface quality defects caused by thermal deformation.

[0018] (3) This invention uses a deviation correction module to quantitatively analyze key indicators such as roughness Ra value and residual defect density, and compares them with historical compliant samples to form an adaptive adjustment strategy, breaking through the limitation of traditional methods that only rely on the presence or absence of defects. This mechanism can accurately assess the degree of insufficient grinding, providing a data-driven basis for parameter optimization in the next round of grinding, and realizing closed-loop optimization and continuous improvement of the grinding process. Attached Figure Description

[0019] 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.

[0020] Figure 1 This is a schematic diagram showing the connection of each module in the intelligent guided grinding system for the surface of a cast billet according to the present invention.

[0021] Figure 2 This is a flowchart of the method for obtaining real-time grinding parameters in this invention.

[0022] Figure 3 This is a step diagram of an intelligent guided grinding method for the surface of a cast billet according to the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] Example 1

[0025] Please see Figure 1 As shown, the present invention provides an intelligent guided grinding system for the surface of a cast billet, comprising: a parameter initialization module, a parameter compensation module, and a deviation correction module. The connection relationships between the modules are as follows: the parameter initialization module and the parameter compensation module are connected, the parameter compensation module and the deviation correction module are connected, and the deviation correction module and the parameter initialization module are connected.

[0026] The parameter initialization module is used to obtain information on the area to be refurbished and the refurbishment target value on the surface of the billet, divide the refurbishment process into at least one refurbishment stage, and configure the initial refurbishment parameters applicable to the first refurbishment stage.

[0027] In one embodiment of the present invention, considering that the grinding target and initial execution strategy need to be clearly defined before grinding the billet, given that the surface defects of the billet vary in depth, area and type, if these differences are ignored and fixed parameters are directly used for grinding, it is easy to cause over-grinding or under-grinding. Therefore, by collecting information and configuring parameters in the early stage, the foundation for subsequent accurate grinding can be laid, and blind operation can be avoided.

[0028] Specifically, the step includes: an optical sensor group installed on a fixed bracket upstream of the grinding station, which works in conjunction with an industrial camera and a laser profilometer to collect real-time images of the billet surface and three-dimensional contour data, identify and mark the defect depth, contour area and type label of the area to be ground, and at the same time obtain the current grinding compliance value requirements of the billet from the production management system.

[0029] It should be noted that the information on the area to be ground is used to determine which defects on the surface of the billet need to be treated and their severity. That is, regardless of the severity of the defect, as long as it is marked as an area to be ground, it must be ground.

[0030] The grinding compliance value refers to the pre-set technical threshold that the surface should meet after grinding, based on the steel grade of the cast billet, product specifications, and subsequent process requirements. The grinding compliance value includes the target roughness Ra value and the maximum allowable residual defect density.

[0031] Preferably, in one embodiment of the present invention, the method for dividing the grinding stage includes: retrieving a single-round grinding tolerance threshold corresponding to the billet steel type and grinding compliance value from the process knowledge base. The threshold includes a tolerance defect depth value and a tolerance contour area value. The single-round grinding tolerance threshold is a process safety parameter used to determine whether a single grinding operation can effectively eliminate defects without damaging the equipment or causing over-grinding.

[0032] If the depth of any defect in the area to be refurbished exceeds the tolerance defect depth value, or its outline area exceeds the tolerance outline area value, then the surface of the cast billet is determined to have a severe defect. To avoid excessive stress or poor results from a single-round refurbishment, the refurbishment process is divided into at least two refurbishment stages, adopting a gradual strategy.

[0033] If the depth and contour area of ​​all defects in the area to be refurbished do not exceed their corresponding single-round refurbishment tolerance threshold, then the surface of the billet is determined to have only minor defects. This indicates that the surface defects of the billet are relatively minor and the refurbishment is less difficult. Therefore, the refurbishment process is divided into a single-round refurbishment stage, and the refurbishment is completed in one go using the initial refurbishment parameters.

[0034] This initial division will provide an execution framework for subsequent parameter compensation and deviation correction modules. During the actual grinding process, the system will make closed-loop decisions based on the real-time detection results after each round of grinding, and the final number of grinding stages will be determined by this dynamic evaluation result. For example, even if the initial division is a single-stage process, if the detection fails to meet the standards after grinding, the system will automatically start the next round of grinding.

[0035] Furthermore, the method for configuring the initial grinding parameters includes: matching the corresponding basic grinding depth, basic feed rate, and basic spindle speed from the process knowledge base as basic process parameters based on the defect depth and contour area in the information of the area to be ground, combined with the grinding compliance value; and performing material adaptation correction on the basic process parameters according to the type of steel in the billet.

[0036] Specifically, when the billet steel is a high-hardness steel, the base grinding depth is reduced to a value less than the base grinding depth. When the billet steel is a high-heat-sensitive steel, the base feed rate and base spindle speed are simultaneously reduced to values ​​less than their respective base values.

[0037] Apply equipment operation safety constraints to the corrected parameters, specifically, grinding depth constraints: ensure that the corrected grinding depth does not exceed the maximum cutting depth allowed by the mechanical structure of the grinding equipment;

[0038] Feed rate constraint: Ensure that the corrected feed rate is within the safe operating range of the spindle motor's rated speed; Spindle speed constraint: Ensure that the corrected spindle speed is higher than the minimum critical speed required for the equipment to maintain stable grinding.

[0039] The method for dynamically setting the spacing between adjacent trajectory lines of the grinding path strategy based on the defect contour area includes: dividing the grinding area into different processing levels according to the size of the detected defect contour area, and dynamically adjusting the spacing between adjacent trajectory lines of the grinding tool accordingly.

[0040] Specifically, the baseline spacing value matching the current steel grade and grinding target is retrieved from the process knowledge base. Then, the defect contour area is compared with the preset area threshold range. If the defect area falls into a larger range, the spacing between adjacent trajectory lines is reduced proportionally to ensure complete coverage and effective grinding of large defect areas. Conversely, for smaller defects, the spacing value is adopted or appropriately increased to balance grinding efficiency.

[0041] For example, the spacing between adjacent baseline trajectory lines is set to 2.0 mm, and two area threshold ranges are defined: S < 5 mm² is a small area defect, 5 mm² ≤ S ≤ 15 mm² is a medium area defect, and S > 15 mm² is a large area defect.

[0042] For small defect areas, a baseline spacing of 2.0 mm or increased to 2.2 mm can be used for rapid grinding; for medium-sized defect areas, the spacing should be reduced to 1.8 mm by a proportional factor of 0.9; for large defect areas, the spacing should be reduced to 1.4 mm by a proportional factor of 0.7 to ensure grinding coverage. All adjusted spacing values ​​must meet the hard constraint of not less than the minimum positioning accuracy of the equipment, such as 1.0 mm.

[0043] The grinding depth, feed rate, spindle speed, and regrinding path strategy that meet safety constraints are combined into initial regrinding parameters.

[0044] Through the above operations, initial grinding parameters can be obtained for subsequent grinding stages, providing conditions for the execution of the parameter compensation module.

[0045] The parameter compensation module is used to dynamically correct the initial grinding parameters based on the real-time collected grinding force data during the current grinding stage to obtain the real-time grinding parameters. It also performs thermal deformation compensation on the real-time grinding parameters according to the steel material characteristics of the billet to obtain the compensated grinding parameters. The compensated grinding parameters are then used to grind the area to be ground.

[0046] In one embodiment of the present invention, considering that the uneven surface hardness of the billet and wear of the grinding tool during the grinding process can cause fluctuations in grinding force, if the initial parameters are kept unchanged, local over-grinding or defect residue is likely to occur. Therefore, by real-time grinding force monitoring and dynamic parameter correction, the working conditions can be adaptively responded to and the grinding stability can be guaranteed.

[0047] Specifically, see Figure 2As shown, the step includes: collecting grinding force sensor data at a fixed period of 100ms during the grinding process, and constructing the grinding force sequence within the current sliding time window, wherein the length of the sliding time window is set to 10 cycles.

[0048] Based on the grinding force sequence of the current sliding time window and the historical reference window, the relative rate of change of the fluctuation characteristic between the two windows is calculated. The historical reference window is the grinding force data of the first 20 cycles after the grinding is started, and the fluctuation characteristic is the standard deviation of the grinding force sequence within the current sliding time window.

[0049] When the relative rate of change of at least two consecutive sliding time windows satisfies the monotonically non-decreasing condition, and the relative rate of change of the current sliding time window exceeds 15% of the fluctuation characteristic of the historical benchmark window, the grinding process is determined to have entered a dynamic disturbance state.

[0050] It should be noted that when the relative rate of change of at least two consecutive sliding time windows satisfies the monotonically non-decreasing condition, this indicates that the fluctuation of grinding force shows a trend of continuous enhancement or at least no slowing down, which means that the grinding process is being subjected to a continuous or undiminished external disturbance, rather than a brief random fluctuation.

[0051] Furthermore, when the relative change rate of the current sliding time window exceeds 15% of the fluctuation characteristic of the historical benchmark window, it indicates that the fluctuation amplitude of the current grinding force is significantly greater than the normal fluctuation level at the beginning of the grinding process. The combination of these two factors indicates that the grinding process is not only subject to continuously increasing interference, but the intensity of the interference has also exceeded the normal range. Therefore, the system determines that it has entered a dynamic disturbance state.

[0052] Under dynamic disturbance, the safety adjustment factor is retrieved from the material property database according to the billet steel type. For example, the safety adjustment factor for Q235 steel is 0.03. Based on the product of the relative change rate and the safety adjustment factor, the feed rate in the initial grinding parameters is adjusted in reverse.

[0053] In one specific embodiment, firstly, based on the steel type of the current billet, a safety adjustment factor specific to that steel type is retrieved from a pre-set material property database. This safety adjustment factor is an empirical coefficient determined based on numerous process experiments, and its core function is to ensure that the range of parameter adjustments is limited within the allowable range for equipment safety and processing quality. For example, the empirical value of this factor for Q235 steel is 0.03.

[0054] During adjustment, the product of the relative rate of change and the safety adjustment factor is calculated, and this product value is used as the adjustment amount of the feed rate. This reverse adjustment means that when the grinding force fluctuation increases, i.e., the relative rate of change is positive, the feed rate in the initial regrinding parameters is reduced according to this product value, thereby reducing the instantaneous grinding load and restoring the regrinding process to stability.

[0055] When the relative rate of change of three consecutive sliding time windows does not meet the monotonically non-decreasing condition, exit the dynamic disturbance state and restore the initial grinding parameters; combine the adjusted feed rate with the grinding depth and spindle speed in the initial grinding parameters, and keep the grinding path strategy unchanged to form real-time grinding parameters.

[0056] It should be noted that when the relative rate of change of three consecutive sliding time windows does not meet the monotonically non-decreasing condition, this indicates that the increasing trend of grinding force fluctuation has been interrupted, and the fluctuation level no longer continues to rise, but instead decreases or remains stable. This means that the factors that previously caused the dynamic disturbance have weakened or disappeared, and the regrinding process is returning to a stable operating condition. Based on this, the dynamic disturbance state is determined to have ended, and the initial regrinding parameters are automatically switched back to continue the regrinding operation.

[0057] In one embodiment of the present invention, since the cast billet is at a high temperature during grinding, its surface will be higher than the actual height at room temperature due to thermal expansion. If grinding is performed directly according to the target size, the grinding depth will be insufficient due to material shrinkage after cooling, which may result in residual defects.

[0058] Therefore, based on the real-time surface temperature of the billet and the specific thermal expansion coefficients of the steel grade, the grinding parameters are dynamically compensated. By adding a compensation depth, the amount of cooling shrinkage is offset, thereby ensuring that the surface dimensions of the billet are consistent with the grinding target after it cools to room temperature.

[0059] Specifically, the steps include: collecting real-time surface temperature data of the billet using a non-contact infrared thermometer installed upstream of the grinding station; retrieving the corresponding temperature-thermal deformation relationship from the material property database according to the billet steel type; superimposing the grinding depth and thermal deformation compensation depth in the real-time grinding parameters, while keeping the feed rate, spindle speed, and grinding path strategy in the initial grinding parameters unchanged, to generate the compensated grinding parameters.

[0060] Preferably, in one embodiment of the present invention, the method for setting the thermal deformation compensation depth includes: when the surface temperature of the billet is in a first temperature range of 300-500℃, setting the thermal deformation compensation depth to 5% of the grinding depth in the real-time grinding parameters; when the surface temperature of the billet is in a second temperature range of 500-700℃, setting the thermal deformation compensation depth to 8% of the grinding depth in the real-time grinding parameters; and when the surface temperature of the billet is in a third temperature range of 700-900℃, setting the thermal deformation compensation depth to 12% of the grinding depth in the real-time grinding parameters.

[0061] The division of the temperature range and the setting of the compensation ratio are determined based on experimental data of the thermal expansion characteristics of different steel grades in the corresponding temperature ranges. The higher the temperature, the greater the thermal deformation of the billet surface, and the more the required compensation depth is increased, thereby offsetting the insufficient grinding amount caused by thermal deformation.

[0062] For example, assuming the grinding depth in the current real-time grinding parameters is 0.2 mm, if the surface temperature of the billet is detected to be 550°C, which is in the second temperature range, then the thermal deformation compensation depth is 0.2 mm × 8% = 0.016 mm, and the actual grinding depth after compensation is 0.216 mm.

[0063] In another embodiment, the compensation ratio can be adjusted adaptively for different steel grades. For example, for stainless steel grades with a high coefficient of thermal expansion, the compensation ratio can be multiplied by a steel grade correction factor of 1.1 to 1.3.

[0064] Through the above operations, accurate grinding parameters that take into account the effects of thermal deformation can be obtained, providing high-quality grinding treatment for the surface condition detection of the deviation correction module.

[0065] The deviation correction module is used to detect the surface condition parameters of the ground area after the current grinding cycle is completed, and determine the upper limit of the surface condition parameters based on the historical compliance samples of the same steel grade billet. When the surface condition parameters are greater than the upper limit, the absolute deviation between the surface condition parameters and the grinding compliance value is calculated, and the initial grinding parameters for the next grinding cycle are adaptively adjusted based on the absolute deviation.

[0066] It should be explained that the end of this round of grinding refers to the time after the grinding equipment has completed the coverage of all grinding areas in the current round according to the planned grinding path strategy.

[0067] In one embodiment of the present invention, considering that the grinding standard needs to be verified in conjunction with historical data, this avoids the limitations of a single standard. If there are sufficient historical samples, a reasonable upper limit can be dynamically set through statistical methods; if there are insufficient samples, an industry standard preset threshold is adopted to ensure the objectivity and accuracy of the judgment of insufficient grinding.

[0068] Specifically, the steps include: after the current grinding is completed, the surface condition of the ground area is detected by the detection device installed downstream of the grinding station to obtain surface condition parameters including surface roughness Ra value and residual defect density; at the same time, the historical qualified sample data of the same steel grade billet are queried from the quality database, the distribution characteristics of the surface condition parameters are statistically analyzed, and the upper limit threshold of each parameter is determined.

[0069] Preferably, in one embodiment of the present invention, the method by which the deviation correction module determines whether the surface state parameter is greater than its upper limit value includes: when the number of historical qualified samples of the same steel grade billet meets the statistical requirements, such as the number of samples ≥ 30, the requirements of the central limit theorem are met, so that the parameter distribution is close to the normal distribution.

[0070] The upper limit is dynamically calculated based on the grinding compliance value and the standard deviation of historical data. The calculation formula is: Upper limit = Grinding compliance value + k × Standard deviation, where k is the confidence coefficient, usually taken as 2.5. This coefficient is set based on statistical principles, indicating that under normal circumstances, 99% of the compliant cast billets will have parameters lower than this upper limit, thus ensuring high reliability of the judgment.

[0071] Otherwise, when the number of historical samples meeting the standard is insufficient, a preset fixed threshold is used as the upper limit. For example, when a new steel grade is put into production or the number of samples is less than 30, the system will automatically switch to the fixed threshold mode. Taking the target roughness Ra value of 0.8μm as an example, its fixed upper limit is 1.04μm (0.8×1.3), which is 1.3 times the target value, ensuring a conservative judgment in the case of insufficient data.

[0072] When any surface condition parameter exceeds its upper limit, it is determined that the grinding is insufficient. This indicates that the current grinding operation has failed to meet the preset grinding quality standard, and the surface of the cast billet still has excessive roughness or too many residual defects. The next round of grinding must be started, and the process parameters must be adjusted accordingly. If multiple surface condition parameters exceed the limit simultaneously, the parameter with the largest absolute deviation will be dealt with first.

[0073] Otherwise, if all surface condition parameters are not greater than their respective upper limits, it indicates that the current grinding operation is effective and the surface quality of the billet has reached or exceeded the grinding standard value, and the grinding process of the billet can be terminated.

[0074] For example, for Q235 steel billets, when the number of historical samples meeting the standard is 45, the target roughness Ra value is 0.8μm, and the historical standard deviation is 0.12μm, the upper limit of roughness is calculated as 0.8 + 2.5 × 0.12 = 1.1μm; if the surface roughness after grinding is detected to be 1.25μm, which is greater than the upper limit of 1.1μm, it is determined that there is insufficient grinding.

[0075] In one embodiment of the present invention, considering that the causes of insufficient grinding are different, the parameters need to be adjusted in a targeted manner to avoid blindly increasing the grinding stage. At the same time, by limiting the equipment safety threshold, the parameter adjustment can be prevented from exceeding the equipment's operating capacity.

[0076] Specifically, the step includes: when insufficient grinding is detected, calculating the absolute deviation between the surface condition parameters and the grinding target value, and adjusting the initial parameters for the next round of grinding according to the type and magnitude of the deviation.

[0077] Preferably, in one embodiment of the present invention, the adaptive adjustment includes: calculating the absolute deviation between the surface state parameters and the grinding target value, wherein the surface state parameters include the roughness Ra value and the residual defect density, and the grinding target value includes the target roughness Ra value and the maximum allowable residual defect density.

[0078] When the absolute deviation of the roughness Ra value is greater than zero, the grinding depth of the next round is increased proportionally by a coefficient of 0.3. If the absolute deviation of the residual defect density is greater than zero, the spacing between adjacent trajectory lines of the grinding path is reduced in the high-density defect area by 20% of the original spacing. The above-mentioned parameter variation ratio is limited by the equipment's safe operation threshold.

[0079] When the absolute deviation of the roughness Ra value is greater than zero, it indicates that the surface roughness after the current grinding has not reached the preset target roughness value, and the surface is still too rough. It is necessary to improve the surface finish by increasing the amount of material removed in the next grinding cycle. If it is not greater than zero, it indicates that the surface roughness has met or exceeded the requirements.

[0080] The principle of increasing the grinding depth proportionally in the next round, with a proportionality coefficient set at 0.3, is based on the following: this coefficient is an empirical value determined through numerous process experiments, achieving an effective balance between improving surface roughness and preventing over-grinding. Simultaneously, the increase in grinding depth must be ensured to remain within the equipment's safe operating threshold. Implementers can make adaptive adjustments within the range of 0.2 to 0.5 based on the characteristics of different steel grades.

[0081] If the absolute deviation of the residual defect density is greater than zero, it indicates that too many defects remain on the surface of the cast billet after grinding, and the grinding coverage is insufficient. If it is not greater than zero, it indicates that the defect removal effect has met the standard.

[0082] The principle behind reducing the spacing between adjacent trajectory lines in a high-density defect area is to ensure that unremoved defects can be effectively ground by increasing the coverage density of the grinding trajectory in that area. The reduction ratio is 20% of the original spacing. This ratio is determined based on engineering practices that balance grinding thoroughness and operational efficiency, and the adjusted spacing should not be less than the minimum trajectory spacing that the equipment can execute.

[0083] For example, if the original spacing between adjacent trajectory lines is 2.0 mm, the adjusted spacing is 2.0 × (1 - 20%) = 1.6 mm.

[0084] Furthermore, the deviation correction module also includes: when any surface state parameter is greater than its corresponding upper limit value, and the grinding stage reaches the preset upper limit threshold, which is usually 3 rounds, the current grinding process is terminated, the current billet is marked as abnormal and removed from the grinding station, a manual intervention command is triggered, and the process is transferred to the next billet or the machine is stopped.

[0085] Example 2

[0086] See Figure 3 As shown, the present invention proposes an intelligent guided grinding method for the surface of a cast billet, comprising the following steps: obtaining information on the area to be ground and the grinding target value on the surface of the cast billet, dividing the grinding process into at least one grinding stage, and configuring initial grinding parameters suitable for the first grinding stage.

[0087] During this round of regrinding, the initial regrinding parameters are dynamically corrected based on the real-time collected grinding force data to obtain the real-time regrinding parameters.

[0088] Based on the steel grade and material characteristics of the billet, the real-time grinding parameters are thermally deformed to obtain the compensated grinding parameters, and the area to be ground is ground using the compensated grinding parameters.

[0089] After this round of grinding is completed, the surface condition parameters of the ground area are tested, and the upper limit of the surface condition parameters is determined based on the historical compliance samples of the same steel grade billet.

[0090] When the surface condition parameter is greater than its upper limit, the absolute deviation between the surface condition parameter and the grinding target value is calculated, and the initial grinding parameters for the next round of grinding are adaptively adjusted based on the absolute deviation.

[0091] In summary, this invention detects surface condition parameters after each round of grinding to determine whether the parameters meet the standards or reach the upper limit of the grinding cycle. Then, based on the deviation, it adjusts the parameters for the next round or terminates the process, ultimately ensuring that the cast billet either meets the grinding quality standards or is promptly identified and removed from the production line. This method establishes a closed-loop control process of detection-judgment-adjustment-re-execution, achieving intelligent guidance of the cast billet surface grinding process, improving grinding accuracy and efficiency, reducing the need for manual intervention, and adapting to complex working conditions with different steel material characteristics and surface defect types, effectively ensuring the stability and consistency of the cast billet surface quality.

[0092] 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.

[0093] Those skilled in the art will recognize that the algorithmic 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 implementations should not be considered beyond the scope of this application.

[0094] 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.

[0095] 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.

[0096] 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 strand surface intelligent guided dressing system, characterized by, The method comprises the following modules: A parameter initialization module is configured to obtain information of a region to be ground on the surface of a casting blank and a grinding target value, divide a grinding process into at least one grinding stage, and configure initial grinding parameters suitable for the first grinding stage; A parameter compensation module is configured to, in the current grinding stage, dynamically correct the initial grinding parameters based on real-time collected grinding force data to obtain real-time grinding parameters, compensate the real-time grinding parameters according to the material properties of the casting blank to obtain compensated grinding parameters, and use the compensated grinding parameters to grind the region to be ground; A deviation correction module is configured to, after the current grinding stage, detect a surface state parameter of the ground region, determine an upper limit value of the surface state parameter according to historical samples of the same type of casting blank, calculate an absolute deviation between the surface state parameter and the grinding target value when the surface state parameter is greater than the upper limit value, and adaptively adjust the initial grinding parameters for the next grinding stage based on the absolute deviation. The parameter initialization module is configured to obtain the information of the region to be ground on the surface of the casting blank by an optical sensor group installed on a fixed support upstream of a grinding station, wherein the information of the region to be ground includes defect depth, profile area and type label, and the optical sensor group includes an industrial camera and a laser profilometer. The method for dividing the grinding stage is to retrieve a tolerable defect depth value and a tolerable profile area value corresponding to the type of the casting blank and the grinding target value from a process knowledge base, divide the grinding process into at least two grinding stages if the depth of any defect in the region to be ground is greater than the tolerable defect depth value or the profile area of any defect is greater than the tolerable profile area value, or divide the grinding process into a single grinding stage. The method for configuring the initial grinding parameters is to match corresponding basic grinding depth, basic feed speed and basic spindle speed as basic process parameters from the process knowledge base according to the defect depth and profile area in the information of the region to be ground in combination with the grinding target value, adaptively correct the basic process parameters according to the type of the casting blank, apply equipment operation safety constraints to the corrected parameters, dynamically set the distance between adjacent track lines of the grinding path strategy based on the defect profile area, and combine the grinding depth, feed speed, spindle speed and grinding path strategy that meet the safety constraints into the initial grinding parameters. The method for obtaining the real-time grinding parameters comprises the following steps:

2. A strand surface intelligent guided trimming system as claimed in claim 1, characterized in that, In the current grinding stage, collect grinding force time series data at a fixed period, and construct a grinding force sequence in a current sliding time window; Calculate the relative change rate of fluctuation characteristic quantities between the current sliding time window and a historical reference window based on the grinding force sequences of the two windows, wherein the historical reference window is the grinding force data of the first N periods after the start of grinding, and the fluctuation characteristic quantity is the standard deviation of the grinding force sequence in the current sliding time window; When the relative change rates of at least two consecutive sliding time windows meet the monotone non-decreasing condition, and the relative change rate of the current sliding time window exceeds a preset proportion of the fluctuation characteristic quantity of the historical reference window, it is determined that the grinding process enters a dynamic disturbance state. ​ In the dynamic disturbance state, the safety adjustment factor is retrieved from the material property database according to the steel grade of the casting blank, and the feed speed in the initial grinding parameter is inversely adjusted based on the product of the relative change rate and the safety adjustment factor; When the relative change rates of the three consecutive sliding time windows do not satisfy the monotonic non-decreasing condition, the dynamic disturbance state is exited and the initial grinding parameter is restored; The adjusted feed speed is combined with the grinding depth and the spindle speed in the initial grinding parameter, and the grinding path strategy is kept unchanged to form the real-time grinding parameter.

3. A strand surface intelligent guided trimming system as claimed in claim 1, wherein, The method for obtaining the compensated grinding parameter is: Real-time acquisition of casting blank surface temperature data by a non-contact infrared temperature measuring instrument installed upstream of the grinding station; According to the steel grade of the casting blank, the corresponding temperature-thermal deformation relationship is retrieved from the material property database; When the casting blank surface temperature is in the first temperature interval, the thermal deformation compensation depth is set to the first proportion of the grinding depth in the real-time grinding parameter; When the casting blank surface temperature is in the second temperature interval, the thermal deformation compensation depth is set to the second proportion of the grinding depth in the real-time grinding parameter; When the casting blank surface temperature is in the third temperature interval, the thermal deformation compensation depth is set to the third proportion of the grinding depth in the real-time grinding parameter; Wherein, the upper limit values of the first temperature interval, the second temperature interval and the third temperature interval are sequentially increased, and the values of the first proportion, the second proportion and the third proportion are sequentially increased; The grinding depth in the real-time grinding parameter is superimposed with the thermal deformation compensation depth, while the feed speed, the spindle speed and the grinding path strategy in the initial grinding parameter are maintained unchanged to generate the compensated grinding parameter.

4. A strand surface intelligent guided trimming system as claimed in claim 1, characterized in that, The method for the deviation correction module to judge whether the surface state parameter is greater than its upper limit value includes: When the number of historical qualified samples of the same steel grade casting blank meets the statistical requirements, the upper limit value is dynamically calculated based on the grinding qualified value and the historical data standard deviation; otherwise, a preset fixed threshold is used; When any surface state parameter is greater than its upper limit value, it is determined that there is insufficient grinding.

5. A strand surface intelligent guided trimming system as claimed in claim 1, wherein, The adaptive adjustment includes: Calculate the absolute deviation of the surface state parameter and the grinding qualified value, wherein the surface state parameter includes the roughness Ra value and the residual defect density, and the grinding qualified value includes the target roughness Ra value and the maximum allowed residual defect density; When the absolute deviation value of the roughness Ra value is greater than zero, the grinding depth of the next round is increased in direct proportion; If the absolute deviation of the residual defect density is greater than zero, the spacing between adjacent track lines of the grinding path is reduced in the defect high-density area; The adjusted parameter variation proportion is limited by the equipment safety operation threshold.

6. A strand surface intelligent guided trimming system as claimed in claim 1, characterized in that, The deviation correction module further includes the following content: When any surface state parameter is greater than its corresponding upper limit value and the grinding stage reaches a preset upper limit threshold, the current grinding process is terminated, the current casting blank is marked as an abnormal state and removed from the grinding station, an artificial intervention instruction is triggered, and the processing flow of the next casting blank or the machine is entered.

7. A method of intelligent guided conditioning of a strand surface, by means of a system for intelligent guided conditioning of a strand surface according to any one of claims 1 to 6, characterized in that the following steps are performed, The method includes the following steps: Obtain the information of the casting blank surface to be ground and the grinding qualified value, divide the grinding process into at least one round of grinding stage, and configure the initial grinding parameter suitable for the first round of grinding stage; In the current grinding stage, the initial grinding parameters are dynamically corrected based on the real-time collected grinding force data to obtain real-time grinding parameters; According to the steel grade material characteristics of the casting blank, the real-time grinding parameters are compensated for thermal deformation to obtain compensated grinding parameters, and the compensated grinding parameters are used to grind the region to be ground; After the current grinding is completed, the surface state parameters of the ground region are detected, and the upper limit value of the surface state parameters is determined according to the historical qualified sample situation of the same steel grade casting blank; When the surface state parameter is greater than the upper limit value, the absolute deviation of the surface state parameter from the grinding qualified value is calculated, and the initial grinding parameters of the next round of grinding are adaptively adjusted based on the absolute deviation.

Images