A 1000MPa grade bainitic steel cold rolling production method

Abstract

The present application relates to the technical field of alloy steel, and specifically to a cold rolling production method for 1000MPa grade bainitic steel. The method comprises: smelting and continuous casting according to the set composition to form a cast billet; hot rolling the obtained cast billet to obtain a hot-rolled plate; obtaining the pickling time and three-dimensional point cloud of the last hot-rolled plate after pickling, obtaining the convex and concave parts based on the three-dimensional point cloud data, obtaining the vector of the boundary data point based on its characteristics, obtaining the supervoxel feature, obtaining the surface sharpness and the consistent surface feature coefficient, and dividing the supervoxel into corrosion and oxidation based on this to obtain the pickling coefficient, and then obtaining the pickling time of the next hot-rolled plate for pickling; cold rolling the hot-rolled plate after pickling to obtain a cold-rolled plate; then annealing; pickling the annealed cold-rolled plate to obtain a rigid body; electroplating a layer of nickel on the rigid body; and completing the cold rolling production of bainitic steel. The present application improves the strength and toughness of the bainitic steel products obtained in subsequent production.

Description

Technical Field

[0001] The present application relates to the technical field of alloy steel, and in particular to a cold rolling production method for 1000MPa grade bainitic steel. Background Art

[0002] At present, 1000MPa grade cold-rolled advanced bainitic steel is widely used in car bodies, such as TR980, etc. These steels are currently widely used in reinforcements such as body B-pillars and door frame reinforcement plates, as well as more complex body structural parts such as front and rear longitudinal beams. As automobile factories continue to increase their requirements for the strength of automobile structural parts, the production processes of hot rolling and cold rolling will become more complicated. High-strength performance requirements mean higher alloy composition design, but high alloy composition design will seriously affect the product surface quality grade during industrial production. Therefore, developing and designing a production method for cold-rolled advanced bainitic steel that meets both the high-strength requirements of automobile structural design and the high surface requirements of automobile personalization has become a crucial research direction, providing solid technical support for the lightweight engineering of the new generation of automobiles.

[0003] For 1000MPa high surface grade cold-rolled advanced bainitic steel, the control method using a single process is not very effective, and surface control must be performed based on a full-process process coupling design. Therefore, it is of great significance to explore a production method for high surface grade cold-rolled advanced bainitic steel on the basis of meeting performance requirements.

[0004] In the cold rolling production of bainitic steel, the pickling process is a key link that affects the surface quality of bainitic steel products. The choice of pickling time in the pickling process affects the pickling results, and further affects the strength and toughness of the produced bainitic steel products. However, the existing pickling process of bainitic steel is usually carried out according to several preset pickling times, which is prone to under-pickling and over-pickling in the pickling process. Summary of the invention

[0005] In order to solve the technical problems of under-pickling and over-pickling in the pickling process, the present application provides a 1000MPa grade bainitic steel cold rolling production method, and the technical scheme adopted is as follows:

[0006] The present application proposes a 1000MPa grade bainite steel cold rolling production method, the method comprising the following steps:

[0007] Smelting and continuous casting into ingots according to the set composition;

[0008] The obtained ingot is hot-rolled to obtain a hot-rolled plate;

[0009] The pickling time of the next hot-rolled plate is obtained based on the characteristics of the hot-rolled plate after the previous pickling, and the hot-rolled plate is pickled based on the obtained pickling time. The method for obtaining the pickling time is as follows:

[0010] S1, record the last pickled hot-rolled plate as the plate to be processed, and obtain the pickling time and three-dimensional point cloud data of the plate to be processed;

[0011] S2, clustering the three-dimensional point cloud data of the plate to be processed to obtain a number of supervoxels, obtaining the protruding and concave volumes of the supervoxels based on the height of the three-dimensional point cloud of the hot-rolled plate after normal pickling, and obtaining the surface concave-convex degree of the supervoxels based on the sum of the volumes;

[0012] S3, screening supervoxels based on the surface concavity and convexity, obtaining the fitted surface and boundary data points of the screened supervoxels, and obtaining the consistent surface feature coefficient of the supervoxels according to the similarity of the vectors obtained on the fitted surface of all boundary data points in the neighborhood of the boundary data points;

[0013] S4, obtaining the normal vector of each boundary data point, and obtaining the surface sharpness of the supervoxel according to the similarity of the normal vectors of the boundary data point and the adjacent data points;

[0014] S5, obtaining a regional characteristic value based on the consistent surface characteristic coefficient and surface sharpness of the supervoxel; selecting the supervoxel according to the regional characteristic value, and dividing the supervoxel into a corroded supervoxel and an oxidized supervoxel in combination with the protruding volume and the concave volume of the supervoxel; obtaining the pickling coefficient of the to-be-processed board according to the difference in the degree of surface concavity and convexity between the corroded supervoxel and the oxidized supervoxel;

[0015] S6, obtaining the pickling time of the next hot-rolled plate based on the pickling coefficient and pickling time of the plate to be processed;

[0016] The pickled hot-rolled sheet is cold-rolled to obtain a cold-rolled sheet;

[0017] Annealing the cold rolled sheet;

[0018] The annealed cold-rolled sheet is pickled again to obtain a rigid body;

[0019] The steel body is pickled in a short process and then electroplated with a nickel layer to complete the cold rolling production of bainitic steel.

[0020] In the above scheme, the production control scheme of the present application proposes a control idea for the whole process high surface design from the perspective of hot rolling, cold rolling and continuous annealing processes, by analyzing the different distribution characteristics of the iron oxide film area, corrosion area, indentation and pit area on the surface of the hot-rolled plate in the point cloud data, combining the surface feature consistency and surface sharpness constructed by the abnormal supervoxel set obtained by the surface concave and convex degree to construct the regional feature value, and based on the convex volume and concave volume of the supervoxel in the supervoxel set obtained by the regional feature value, the oxidized supervoxel and the corroded supervoxel are constructed respectively. The pickling time adjustment value is obtained by using the pickling adjustment coefficient. Its beneficial effect is that it reduces the influence of the indentations and pit areas on the surface of the hot-rolled plate caused by the hot rolling process on the subsequent evaluation accuracy of the residual degree of the iron oxide film on the surface of the hot-rolled plate and the corrosion degree of the corrosion area. Therefore, the pickling time of the subsequent hot-rolled plates of the same batch can be controlled more accurately according to the residual degree of the iron oxide film on the surface of the hot-rolled plate and the corrosion degree of the corrosion area, so as to avoid the subsequent hot-rolled plates from being under-pickled or over-pickled, thereby improving the strength and toughness of the bainitic steel products produced subsequently.

[0021] In one embodiment, the chemical composition of the ingot comprises, by mass percentage, C: 0.19% to 0.23%, Si: 1.50% to 2.00%, Mn: 2.10% to 2.50%, Als: 0.02% to 0.06%, P: 0.02%, S≤0.005%, N≤0.005%, and the remainder is Fe and other inevitable impurities.

[0022] In one embodiment, during the hot rolling treatment, the heating temperature of the ingot is 1230-1280°C, and the insulation time is more than 200 minutes; the hot rolling start temperature is greater than 1080°C, and the final rolling temperature is 930-970°C; the coiling temperature is 540-580°C, and the hot rolling starts two stages of descaling, and the descaling water pressure is ≥18MPa.

[0023] In one embodiment, the method of clustering the three-dimensional point cloud data of the plate to be processed to obtain a plurality of supervoxels, obtaining the protruding and recessed volumes of the supervoxels based on the height of the three-dimensional point cloud of the hot-rolled plate after normal pickling, and obtaining the surface concave-convex degree of the supervoxels based on the sum of the volumes is as follows:

[0024] The three-dimensional point cloud data of the plate to be processed is divided into a plurality of supervoxels by using a supervoxel clustering point cloud segmentation algorithm;

[0025] The median of the coordinate values ​​of all data points in the three-dimensional point cloud data of the hot-rolled plate of normal pickling is taken as the standard height; the data points in the supervoxel are divided into two parts based on the standard height, which are recorded as convex point cloud data and concave point cloud data;

[0026] The volumes of the convex area and the concave area enclosed by the convex point cloud data and the concave point cloud data and the standard height are obtained through the surface reconstruction algorithm, and the sum of the volumes of the convex area and the concave area is used as the surface concaveness of each supervoxel.

[0027] In one embodiment, the method of screening supervoxels based on surface concavity, obtaining the fitted surface and boundary data points of the screened supervoxels, and obtaining the consistent surface feature coefficient of the supervoxels according to the similarity of the vectors obtained on the fitted surface of all boundary data points in the neighborhood of the boundary data points is:

[0028] All surface concavity and convexity levels are used as the input of the maximum inter-class variance to obtain the partition threshold, and the set of supervoxels greater than the partition threshold is recorded as an abnormal supervoxel set;

[0029] For the supervoxels of the abnormal supervoxel set, a surface fitting algorithm is used to obtain a fitting surface;

[0030] The boundary data points are obtained by using a contour extraction algorithm, and the Gaussian curvature and the average curvature of the boundary data points in the fitting surface are obtained, and the vector composed of the obtained Gaussian curvature and the average curvature is recorded as the local surface feature vector of the boundary data points;

[0031] For any boundary pixel point, a neighborhood is constructed with it as the center, and a preset number of boundary data points closest to it are obtained as neighborhood data points. The cosine similarity of the local surface feature vectors between all neighborhood data points is obtained, and the mean of all cosine similarities is taken as the surface feature consistency of the boundary data points. The consistent degrees of all surface features in the supervoxel are averaged to obtain the consistent surface feature coefficient of the supervoxel.

[0032] In one embodiment, the method of obtaining the normal vector of each boundary data point and obtaining the surface sharpness of the supervoxel according to the similarity between the normal vectors of the boundary data point and the adjacent data points is:

[0033] A circle of preset size is constructed with the boundary data point as the center, the data points in the circle are recorded as adjacent points, and the normal vectors of the fitting surface corresponding to the supervoxel at the boundary data point and each adjacent point are obtained;

[0034] Calculate the cosine similarity between the normal vector of the boundary data point and the normal vector of each of its adjacent points, and obtain the local sharpness of each boundary data point based on all cosine similarities;

[0035] Local sharpness is negatively correlated with all cosine similarities;

[0036] The local sharpness of all boundary data points in the supervoxel is normalized by taking the mean value to obtain the surface sharpness of the supervoxel.

[0037] In one embodiment, the method of obtaining regional characteristic values ​​based on the consistent surface characteristic coefficient and surface sharpness of supervoxels; screening supervoxels according to the regional characteristic values, and dividing supervoxels into corroded supervoxels and oxidized supervoxels in combination with the protruding volume and the concave volume of the supervoxels; and obtaining the pickling coefficient of the to-be-processed board according to the difference in the surface concave-convexity degree of the corroded supervoxels and the oxidized supervoxels is as follows:

[0038] The regional eigenvalue is negatively correlated with the consistent surface characteristic coefficient and the surface sharpness;

[0039] A characteristic value threshold is preset, and a set consisting of all supervoxels whose regional characteristic values ​​in the abnormal supervoxel set are greater than the characteristic value threshold is recorded as a first supervoxel set of the plate to be processed;

[0040] For each supervoxel in the first supervoxel set, the ratio of the convex volume to the concave volume in the supervoxel is recorded as the concave-convex feature value of the supervoxel;

[0041] The concave-convex eigenvalues ​​of all supervoxels in the first supervoxel set are used as the input of the maximum inter-class variance algorithm, and the concave-convex eigenvalue threshold is output, and the supervoxels whose concave-convex eigenvalues ​​are less than the concave-convex eigenvalue threshold are recorded as oxidized supervoxels, and the remaining supervoxels in all supervoxels are recorded as eroded supervoxels;

[0042] The surface concave-convex degrees of all oxidized supervoxels in the first supervoxel set are summed to obtain a first oxidized concave-convex value of the first supervoxel, and the surface concave-convex degrees of all corroded supervoxels in the first supervoxel set are summed to obtain a first corroded concave-convex value of the first supervoxel; and the pickling coefficient of the to-be-processed board is obtained based on the first oxidized concave-convex value and the first corroded concave-convex value;

[0043] The pickling coefficient of the plate to be processed is positively correlated with the first oxidation concave-convex value, and negatively correlated with the first corrosion concave-convex value.

[0044] In one embodiment, the method for obtaining the pickling time of the next hot-rolled plate based on the pickling coefficient and pickling time of the plate to be processed is:

[0045] , Indicates the pickling time of the board to be treated, Indicates the preset parameters for adjusting the pickling time. is the rounding function, Indicates the pickling coefficient of the plate to be treated, represents the normalization function, Indicates the pickling time of the next hot-rolled plate to be processed.

[0046] In one embodiment, during the annealing treatment, the heating temperature is 850-880°C, the holding time is 120-150s, the temperature is slowly cooled to 730-760°C, and the temperature is quickly cooled to 220-240°C. Tempering, the tempering temperature is 350-370°C, and the holding time is 380s-510s.

[0047] In one embodiment, in the step of pickling again to obtain the rigid body, the pickling concentration is 12-16% dilute hydrochloric acid, the pickling temperature is 60-70° C., and the pickling time is ≥10s.

[0048] The beneficial effects of this application are:

[0049] The production control scheme of this application proposes a control idea for the whole process high surface design from the perspective of hot rolling, cold rolling and continuous annealing processes. By analyzing the different distribution characteristics of the iron oxide film area, corrosion area, indentation and pit area on the surface of the hot-rolled plate in the point cloud data, the surface feature consistency and surface sharpness of the abnormal supervoxel set obtained by the surface concave and convex degree are combined to construct the regional feature value, and the convex and concave feature values ​​constructed by the convex volume and concave volume of the supervoxel in the supervoxel set obtained based on the regional feature value are used to obtain the pickling adjustment of the oxidation supervoxel and the corrosion supervoxel. The pickling time adjustment value is obtained by using a coefficient, which has the beneficial effect of reducing the influence of the indentations and pit areas on the surface of the hot-rolled plate caused by the hot rolling process on the subsequent evaluation accuracy of the residual degree of the iron oxide film on the surface of the hot-rolled plate and the corrosion degree of the corrosion area, so that the pickling time of the subsequent hot-rolled plates of the same batch can be controlled more accurately according to the residual degree of the iron oxide film on the surface of the hot-rolled plate and the corrosion degree of the corrosion area, so as to avoid the subsequent hot-rolled plates from being under-pickled or over-pickled, thereby improving the strength and toughness of the bainitic steel products produced subsequently. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 A flow chart of a 1000MPa grade bainitic steel cold rolling production method provided in one embodiment of the present application;

[0052] Figure 2 Schematic diagram of a method for obtaining the pickling time of a plate to be processed. DETAILED DESCRIPTION

[0053] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the 1000MPa grade bainite steel cold rolling production method proposed in the present application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0054] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0055] A 1000MPa grade bainite steel cold rolling production method embodiment:

[0056] The specific scheme of a 1000MPa grade bainitic steel cold rolling production method provided in the present application is described in detail below with reference to the accompanying drawings.

[0057] The production steps of the 1000MPa grade bainitic steel in the present application are: smelting→continuous casting→hot rolling→pickling→cold rolling→continuous annealing→pickling again→nickel flashing.

[0058] See also Figure 1 , which shows a flow chart of a 1000MPa grade bainite steel cold rolling production method provided by an embodiment of the present application, the method comprising the following steps:

[0059] Step S001, smelting and continuous casting to form ingots according to set composition.

[0060] In this embodiment, the composition of the produced ingot includes, by weight percentage, the following: C: 0.19%, Si: 1.50%, Mn: 2.10%, Als: 0.02%, P: 0.02%, S≤0.005%, N≤0.005%, and the remainder is Fe and other inevitable impurities.

[0061] In another embodiment of the present application, the composition of the produced ingot comprises, by weight percentage, the following: C: 0.23%, Si: 2.00%, Mn: 2.50%, Als: 0.06%, P: 0.02%, S≤0.005%, N≤0.005%, and the remainder is Fe and other inevitable impurities.

[0062] Step S002, hot rolling the ingot obtained by smelting and continuous casting.

[0063] The ingot obtained in the above steps is subjected to hot rolling treatment, wherein the heating temperature of the ingot is 1230-1280°C and the insulation time is more than 200 minutes; the hot rolling start temperature is >1080°C and the final rolling temperature is 930-970°C; the coiling temperature is 540-580°C, the hot rolling is started with two-stage descaling, and the descaling water pressure is ≥18MPa.

[0064] Eight examples of hot rolling related parameters in this application are shown in Table 1.

[0065] Table 1 Hot rolling process parameters of the embodiment

[0066]

[0067] Step S003, obtaining the pickling time of the next hot-rolled plate based on the characteristics of the previous pickled hot-rolled plate, and pickling the hot-rolled plate based on the obtained pickling time.

[0068] S1, record the last pickled hot-rolled plate as the plate to be processed, and obtain the pickling time and three-dimensional point cloud data of the plate to be processed.

[0069] The ingot after hot rolling treatment is recorded as hot rolled plate, the hot rolled plate is pickled, and the hot rolled plate after pickling is recorded as plate to be treated.

[0070] In this embodiment, the pickling time is ≥2 min, the pickling medium is 15% dilute hydrochloric acid, the acid solution temperature is 75°C, and the surface residual oil after pickling is required to be 100 mg / m2 (single side) and the surface residual iron is required to be 20 mg / m2 (single side).

[0071] In another embodiment of the present application, the pickling time is ≥2 min, the pickling medium is 20% dilute hydrochloric acid, the acid solution temperature is 85°C, and the surface residual oil after pickling treatment is required to be 200 mg / m2 (single side) and the surface residual iron is 45 mg / m2 (single side).

[0072] For the obtained plate to be processed, the pickling time t(P) during the pickling treatment is obtained, and the three-dimensional point cloud data of the surface morphology of the plate to be processed is collected using a three-dimensional topography instrument to obtain the three-dimensional point cloud data A of the plate to be processed, wherein the XOY plane in the three-dimensional point cloud data A is the plane where the bottom of the plate to be processed is located, and the Z axis is the direction perpendicular to the surface of the plate to be processed.

[0073] At this point, the pickling time and three-dimensional point cloud data corresponding to the plate to be processed are obtained.

[0074] S2, clustering the three-dimensional point cloud data of the plate to be processed to obtain a number of supervoxels, obtaining the protruding and concave volumes of the supervoxels based on the height of the three-dimensional point cloud of the hot-rolled plate after normal pickling, and obtaining the surface concave-convex degree of the supervoxels based on the sum of the volumes.

[0075] Under normal circumstances, most areas of the plate to be treated after pickling are usually normal areas. The surface of the normal area is usually flat and smooth, and there is basically no loose porous morphology. This means that the height of these normal areas varies little in the direction perpendicular to the plate surface, so that these normal areas usually have the same height.

[0076] Firstly, a hot-rolled plate with normal standard pickling is selected as the standard plate, and the three-dimensional point cloud data of the standard plate is obtained. The median of the coordinate values ​​of all data points in the three-dimensional point cloud data of the standard plate is obtained, and the median of the height is used as the standard height, which is used to characterize the height of the corresponding data point in the normal area of ​​the surface of the hot-rolled plate after normal pickling in the three-dimensional point cloud data.

[0077] During the pickling process of hot-rolled plates, if the hot-rolled plates are under-pickled, the iron oxide film formed during the hot rolling process of the plates to be treated will remain on the surface, and the iron oxide film has the properties of looseness, porosity and cracks; if the hot-rolled plates are over-pickled, the surface of the plates to be treated will be corroded; and the surface area of ​​the plates to be treated where there is an iron oxide film and corrosion usually has a more uneven morphology than the normal surface area.

[0078] Based on the above analysis, the super volume clustering point cloud segmentation algorithm is used to segment all data points in the three-dimensional point cloud data of the plate to be processed, and multiple supervoxels of the three-dimensional point cloud data are obtained, and each supervoxel corresponds to a surface area of ​​the plate to be processed; in this embodiment, the parameters of the supervoxel resolution, seed resolution, color importance value, space importance value, and normal importance value of the super volume clustering point cloud algorithm are set to 0.008, 0.1, 0.2, 0.4, and 1.0 respectively. The super volume clustering point cloud segmentation algorithm is a well-known technology, and the specific process will not be repeated.

[0079] Taking the i-th supervoxel A(i) of the three-dimensional point cloud data as an example, the coordinate values ​​of all data points in the supervoxel A(i) are obtained, and all data points in the supervoxel A(i) are divided into two parts based on the standard height, which are respectively recorded as convex point cloud data A1(i) and concave point cloud data A2(i), which are used to characterize the point cloud data corresponding to all convex areas and concave areas in the surface area of ​​the plate to be processed corresponding to the supervoxel A(i) in the three-dimensional point cloud data A.

[0080] The irregular point cloud volume calculation method based on the Alpha-shape algorithm (surface reconstruction algorithm) calculates the volume of the space enclosed by the convex point cloud data A1(i) and the concave point cloud data A2(i) in each supervoxel and the plane corresponding to the standard height, and the volumes calculated from the convex point cloud data A1(i) and the concave point cloud data A2(i) are respectively recorded as the convex volume V1(i) and the concave volume V2(i) of the supervoxel A(i), which are used to characterize the volumes of all convex areas and concave areas in the surface area of ​​the hot-rolled plate corresponding to the supervoxel A(i). The irregular point cloud volume calculation method based on the Alpha-shape algorithm is a well-known technology, and the specific process will not be repeated here.

[0081] The sum of the protruding volume V1(i) and the concave volume V2(i) is recorded as the surface concavity S(i) of the supervoxel A(i), which is used to characterize the concavity of the hot-rolled plate surface area corresponding to the supervoxel A(i). The larger the sum is, the greater the concavity is, that is, the greater the surface concavity S(i).

[0082] At this point, the surface concavity and convexity of each supervoxel in the plate to be processed is obtained.

[0083] S3, filtering supervoxels based on the surface concavity, obtaining the fitted surface and boundary data points of the filtered supervoxels, and obtaining the consistent surface feature coefficients of the supervoxels according to the similarity of the vectors obtained on the fitted surface of all boundary data points in the neighborhood of the boundary data points.

[0084] Since the normal surface area of ​​the hot-rolled plate after pickling usually has a smaller surface unevenness, the surface unevenness of all supervoxels in the three-dimensional point cloud data of the plate to be processed is used as the input of the maximum inter-class variance algorithm, and a threshold S is output. The set consisting of all supervoxels in the supervoxel whose surface unevenness is greater than the threshold S is recorded as the abnormal supervoxel set of the plate to be processed, which is used to characterize the set consisting of all supervoxels corresponding to all abnormal areas outside the normal surface area in the plate to be processed in the three-dimensional point cloud data. The maximum inter-class variance algorithm is a well-known technology, and the specific process will not be repeated here.

[0085] Since the abnormal surface area on the surface of the plate to be treated usually includes not only the residual iron oxide film area and the corroded area, but may also include the indentation and pit area generated during the hot rolling process of the hot-rolled plate, in order to reduce the impact of the indentation and pit area on the surface of the plate to be treated caused by the hot rolling process on the subsequent evaluation of the residual degree of the iron oxide film on the surface of the plate to be treated and the corrosion degree of the corrosion area, the following treatment is performed.

[0086] The formation of iron oxide film areas and corrosion areas in the plate to be treated is more affected by chemical factors. These processes are often more random, resulting in stronger randomness in the surface features of each position in these areas. In contrast, the indentation and pit areas in the plate to be treated are usually formed by mechanical forces, such as roller pressure or foreign matter intrusion. These mechanical actions often leave regular marks on the surface of the plate to be treated, making each boundary point in the boundary of the indentation and pit area have more consistent surface features than the boundary points near it.

[0087] For each supervoxel in the abnormal supervoxel set, all data points in the supervoxel are fitted using a surface fitting algorithm based on NURBS (non-uniform rational splines) to obtain a fitting surface for each supervoxel. The surface fitting algorithm based on NURBS is a well-known technology and the specific process will not be repeated here.

[0088] The AC point cloud boundary contour extraction algorithm is used to extract all boundary data points in the three-dimensional point cloud data, and all boundary data points in each supervoxel in the abnormal supervoxel set are combined into a boundary data point set of each supervoxel. The AC point cloud boundary contour extraction algorithm is a well-known technology and the specific process will not be repeated here.

[0089] For any boundary data point in the boundary data point set, the Gaussian curvature and the mean curvature of the boundary data point in the corresponding fitting surface are calculated respectively, wherein the calculation of the Gaussian curvature and the mean curvature is a well-known technology and the specific process is not repeated here. The vector composed of the obtained Gaussian curvature and the mean curvature is recorded as the local surface feature vector of the boundary data point, which is used to characterize the surface characteristics of the local area where the boundary data point is located in the fitting surface.

[0090] The M boundary data points with the smallest distance from the boundary data point are obtained and recorded as neighborhood data points, wherein the M neighborhood data points include this boundary data point. In this embodiment, the value of M is 5, and the method for calculating the distance between the boundary data points is the Euclidean distance. For the M neighborhood data points, the cosine similarity between the local surface feature vectors of the neighborhood data points is calculated. The calculation of the cosine similarity is a well-known technology, and the specific process is not repeated here. The mean of the cosine similarities between the local surface feature vectors of all neighborhood data points is used as the surface feature consistency of the boundary data point. It is used to characterize the consistency of the surface features of each position in the local area where the surface position of the board to be processed corresponding to this boundary data point is located. The larger the mean, the greater the surface feature consistency.

[0091] The mean of the surface feature consistency of all boundary data points in the boundary data point set is normalized, and the normalized result is recorded as the consistent surface feature coefficient of the supervoxel, which is used to characterize the degree to which the boundary points in the boundary of the surface area of ​​the plate to be processed corresponding to the supervoxel have consistent surface features. The larger the mean, the greater the degree of consistent surface features, that is, the larger the consistent surface feature coefficient. The normalization method used in this embodiment is the maximum and minimum value normalization method.

[0092] At this point, the consistent surface characteristic coefficients of each supervoxel are obtained.

[0093] S4, for each boundary data point, obtain its normal vector, and obtain the surface sharpness of the supervoxel according to the similarity between the normal vectors of the boundary data point and the adjacent data points.

[0094] Since the indentations and pit areas on the surface of the plate to be treated are usually formed by the pressure of the rolling roller or the intrusion of foreign matter, which usually leads to sharp deformation of the surface of the hot-rolled plate, the boundary points of the indentation and pit areas have sharper surface changes, while the iron oxide film area and the corrosion area on the surface of the plate to be treated are usually areas formed by gradual erosion caused by the chemical action of pickling, so that the surface changes at various positions in the iron oxide film area and the corrosion area are relatively smooth.

[0095] For each boundary data point, a circle with a radius of r is formed with the boundary data point as the center, and the data points in the supervoxel corresponding to the boundary data point in the circle are recorded as adjacent points. In this embodiment, the empirical value of r is 2. The normal vector of the fitting surface corresponding to the supervoxel at the boundary data point and each adjacent point is calculated. The method of obtaining the vector of the point based on the surface is common knowledge and is not described in this embodiment.

[0096] The cosine similarity between the normal vector of the boundary data point and the normal vector of each of its neighboring points is calculated, and the local sharpness of each boundary data point is obtained based on all cosine similarities.

[0097] Local sharpness is negatively correlated with all cosine similarities.

[0098] It should be noted that negative correlation means that when one variable increases, the other variable decreases accordingly, and the two variables change in opposite directions. When one variable changes from large to small or from small to large, the other variable also changes from small to large or from large to small. The specific relationship is determined by actual application and this application does not impose any special restrictions.

[0099] Preferably, in this embodiment, all cosine similarities are accumulated and the reciprocal is taken, and the reciprocal is used as the local sharpness of the boundary data point.

[0100] The local sharpness is used to characterize the sharpness of the surface change in the surface area where the boundary data points are located. In the surface area, the less close the surface change direction of the surface position of the plate to be processed is to the adjacent position, that is, the smaller the mean, the sharper the surface change of the surface position of the hot-rolled plate in the surface area, that is, the greater the local sharpness.

[0101] The normalized result of the mean of the local sharpness of all boundary data points in the boundary data point set is recorded as the surface sharpness of the supervoxel, which is used to characterize the degree to which the boundary points in the boundary of the surface area of ​​the plate to be processed corresponding to the supervoxel have sharp surface changes. The larger the mean of the local sharpness, the greater the surface sharpness.

[0102] At this point, the surface sharpness of each supervoxel is obtained.

[0103] S5, obtaining regional characteristic values ​​based on the consistent surface characteristic coefficients and surface sharpness of the supervoxels; screening supervoxels according to the regional characteristic values, and dividing the supervoxels into corroded supervoxels and oxidized supervoxels in combination with the protruding volumes and recessed volumes of the supervoxels; obtaining the pickling coefficient of the board to be processed according to the difference in surface concave-convexity between the corroded supervoxels and the oxidized supervoxels.

[0104] The regional characteristic value of the supervoxel is obtained according to the consistent surface characteristic coefficient and the surface sharpness of the supervoxel, so as to evaluate the possibility that the region corresponding to the supervoxel has the characteristics of the iron oxide film region and the corrosion region.

[0105] The regional eigenvalue is negatively correlated with the consistent surface characteristic coefficient and the surface sharpness.

[0106] Preferably, in this embodiment, the expression of the regional characteristic value is:

[0107] , where h1(j) and h2(j) represent the consistent surface characteristic coefficient and surface sharpness of the jth supervoxel respectively; is a parameter adjustment constant. In order to prevent the denominator from being 0, The empirical value is taken as 0.01; norm() is the Min-Max normalization function, and U(j) represents the regional eigenvalue of the jth supervoxel.

[0108] The more consistent the surface features are, the larger the consistent surface feature coefficient is, and the sharper the surface change features are, the greater the surface sharpness is, the less the surface area to be treated has the features of the iron oxide film area and the corrosion area on the surface of the hot-rolled plate, that is, the smaller the regional feature value is.

[0109] The regional eigenvalue of each supervoxel in the abnormal supervoxel set is calculated, and the eigenvalue threshold is set. The set consisting of all supervoxels in the abnormal supervoxel set whose regional eigenvalue is greater than the eigenvalue threshold is recorded as the first supervoxel set of the plate to be processed, which is used to characterize the set consisting of supervoxels corresponding to the iron oxide film area and the corrosion area on the surface of the plate to be processed in the three-dimensional point cloud data. In this embodiment, the selected eigenvalue threshold is 0.8.

[0110] Since the corrosion area on the surface of the plate to be treated usually has a larger surface depression degree than the iron oxide film area, for each supervoxel in the first supervoxel set, the ratio of the protruding volume to the concave volume in the supervoxel is recorded as the concave-convex characteristic value of the supervoxel, which is used to evaluate the possibility that the surface area of ​​the plate to be treated corresponding to the supervoxel is an iron oxide film area or a corrosion area. The greater the depression degree in the surface area of ​​the plate to be treated compared to the protrusion degree, that is, the smaller the ratio, the more likely the surface area of ​​the plate to be treated is a corrosion area, that is, the larger the concave-convex characteristic value; conversely, the smaller the concave-convex characteristic value.

[0111] The concave and convex eigenvalues ​​of all supervoxels in the first supervoxel set are used as the input of the maximum inter-class variance algorithm, and the concave and convex eigenvalue threshold is output. The supervoxels whose concave and convex eigenvalues ​​are less than the concave and convex eigenvalue threshold are recorded as oxidized supervoxels, which are used to characterize the supervoxels corresponding to the iron oxide film area on the surface of the board to be treated, and the remaining supervoxels in all supervoxels are recorded as corroded supervoxels, which are used to characterize the supervoxels corresponding to the corroded area on the surface of the board to be treated.

[0112] The surface concave-convex degrees of all oxidized supervoxels in the first supervoxel set are summed to obtain a first oxidized concave-convex value of the first supervoxel, and the surface concave-convex degrees of all corroded supervoxels in the first supervoxel set are summed to obtain a first corroded concave-convex value of the first supervoxel; and the pickling coefficient of the to-be-processed board is obtained based on the first oxidized concave-convex value and the first corroded concave-convex value;

[0113] The pickling coefficient of the plate to be processed is positively correlated with the first oxidation concave-convex value, and negatively correlated with the first corrosion concave-convex value.

[0114] It should be noted that positive correlation means that when one variable increases, the other variable also increases, and the two variables change in the same direction. When one variable changes from large to small or from small to large, the other variable also changes from large to small or from small to large. The specific relationship is determined by actual application and this application does not impose any special restrictions.

[0115] Preferably, in this embodiment, the expression of the pickling coefficient is:

[0116] , w1 represents the first oxidation concave-convex value of the first supervoxel set; w2 represents the first corrosion concave-convex value of the first supervoxel set; exp() is an exponential function, Indicates the pickling coefficient of the plate to be treated.

[0117] The greater the residual degree of the iron oxide film on the surface of the plate to be treated, the smaller the degree of corrosion on the surface of the plate to be treated. In order to effectively remove the iron oxide film on the surface of the next hot-rolled plate belonging to the same batch as the plate to be treated, the larger the pickling coefficient should be, so as to increase the pickling time of the next hot-rolled plate when it is pickled, and vice versa, reduce the pickling time of the subsequent hot-rolled plate when it is pickled.

[0118] At this point, the pickling coefficient of the plate to be treated is obtained.

[0119] S6, obtaining the pickling time of the next hot-rolled plate based on the pickling coefficient and pickling time of the plate to be processed.

[0120] The pickling time of the next hot-rolled plate is calculated based on the pickling coefficient and pickling time of the plate to be treated. The expression is:

[0121] , Indicates the pickling time of the plate to be treated, It represents the parameter for adjusting the pickling time. Its range is [0,60] obtained through experiments. In this embodiment, the value is 20. is the rounding function, Indicates the pickling coefficient of the plate to be treated, Represents the normalization function, in which the normalization is performed in combination with the pickling coefficient in the previous calculation process. Indicates the pickling time of the next hot-rolled plate to be processed.

[0122] The next hot-rolled plate is pickled by the pickling time of the next hot-rolled plate, and the next hot-rolled plate is calculated based on the pickling time. After the hot-rolled plate is pickled, the hot-rolled plate is cold-rolled. The schematic diagram of obtaining the pickling time is as follows: Figure 2 shown.

[0123] At this point, the pickling time of the hot rolled plate is completed.

[0124] Step S004, cold rolling the hot-rolled sheet after pickling to obtain a cold-rolled sheet.

[0125] The hot-rolled plate after pickling in the above step is cold-rolled, the straightening elongation is 0.5-1%, and the cold rolling reduction ratio is controlled between 30% and 40%. Eight examples of cold rolling process parameters are shown in Table 2.

[0126] Table 2: Cold rolling process parameters

[0127]

[0128] Step S005, annealing the cold-rolled sheet.

[0129] After cold rolling, the cold rolled sheet is annealed at a heating temperature of 850-880°C, a holding time of 120-150s, slowly cooled to 730-760°C, quickly cooled to 220-240°C, tempered at a tempering temperature of 350-370°C, a holding time of 380s-510s, and finally air-cooled to room temperature. 8 examples of continuous annealing process parameters are shown in Table 3.

[0130] Table 3: Continuous annealing process parameters

[0131]

[0132] Step S006, pickling the annealed cold-rolled sheet again to obtain a rigid body.

[0133] After continuous annealing, short-process pickling is used to improve the surface quality, the pickling concentration is 12-16% dilute hydrochloric acid, the pickling temperature is 60-70°C, and the pickling time is ≥10s. 8 examples of pickling process parameters after continuous annealing are shown in Table 4.

[0134] Table 4: Pickling process parameters after continuous annealing

[0135]

[0136] Step S007, electroplating a nickel layer on the rigid body after short-process pickling, and completing the cold rolling production of bainitic steel.

[0137] After a short-process pickling, a nickel layer is electrochemically plated on the surface to improve the surface quality. The thickness of the nickel layer is 15-20 mg / m².

[0138] At this point, the cold rolling production of a 1000MPa grade bainitic steel has been completed. This application realizes the industrial production of high surface grade bainitic steel with mechanical properties meeting YS: 600-800MPa, TS: ≥980, EL: ≥20, and has been applied to the industrial mass production of 1000MPa grade high surface grade cold rolled advanced bainitic steel, with good application effect.

[0139] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0140] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A 1000MPa grade bainitic steel cold rolling production method, characterized in that: The method comprises the following steps: Smelting and continuous casting into ingots according to the set composition; The obtained ingot is hot-rolled to obtain a hot-rolled plate; The pickling time of the next hot-rolled plate is obtained based on the characteristics of the hot-rolled plate after the previous pickling, and the hot-rolled plate is pickled based on the obtained pickling time. The method for obtaining the pickling time is as follows: S1, record the last pickled hot-rolled plate as the plate to be processed, and obtain the pickling time and three-dimensional point cloud data of the plate to be processed; S2, clustering the three-dimensional point cloud data of the plate to be processed to obtain a number of supervoxels, obtaining the protruding and concave volumes of the supervoxels based on the height of the three-dimensional point cloud of the hot-rolled plate after normal pickling, and obtaining the surface concave-convex degree of the supervoxels based on the sum of the volumes; S3, screening supervoxels based on the surface concavity and convexity, obtaining the fitted surface and boundary data points of the screened supervoxels, and obtaining the consistent surface feature coefficient of the supervoxels according to the similarity of the vectors obtained on the fitted surface of all boundary data points in the neighborhood of the boundary data points; S4, obtaining the normal vector of each boundary data point, and obtaining the surface sharpness of the supervoxel according to the similarity of the normal vectors of the boundary data point and the adjacent data points; S5, obtaining a regional characteristic value based on the consistent surface characteristic coefficient and surface sharpness of the supervoxel; selecting the supervoxel according to the regional characteristic value, and dividing the supervoxel into a corroded supervoxel and an oxidized supervoxel in combination with the protruding volume and the concave volume of the supervoxel; obtaining the pickling coefficient of the to-be-processed board according to the difference in the degree of surface concavity and convexity between the corroded supervoxel and the oxidized supervoxel; S6, obtaining the pickling time of the next hot-rolled plate based on the pickling coefficient and pickling time of the plate to be processed; The pickled hot-rolled sheet is cold-rolled to obtain a cold-rolled sheet; Annealing the cold rolled sheet; The annealed cold-rolled sheet is pickled again to obtain a rigid body; The steel body is pickled in a short process and then electroplated with a nickel layer to complete the cold rolling production of bainitic steel.

2. A 1000MPa grade bainitic steel cold rolling production method as claimed in claim 1, characterized in that: The chemical composition of the ingot comprises, by mass percentage, C: 0.19% to 0.23%, Si: 1.50% to 2.00%, Mn: 2.10% to 2.50%, Als: 0.02% to 0.06%, P: 0.02%, S≤0.005%, N≤0.005%, and the remainder is Fe and other inevitable impurities.

3. A 1000MPa grade bainitic steel cold rolling production method as claimed in claim 1, characterized in that: During the hot rolling treatment, the heating temperature of the ingot is 1230-1280°C and the insulation time is more than 200 minutes; the hot rolling start temperature is greater than 1080°C and the final rolling temperature is 930-970°C; the coiling temperature is 540-580°C, the hot rolling is started for two-stage descaling, and the descaling water pressure is ≥18MPa.

4. A 1000MPa grade bainitic steel cold rolling production method as claimed in claim 1, characterized in that: The method of clustering the three-dimensional point cloud data of the plate to be processed to obtain a plurality of supervoxels, obtaining the protruding and recessed volumes of the supervoxels based on the height of the three-dimensional point cloud of the hot-rolled plate of normal pickling, and obtaining the surface concave-convex degree of the supervoxels based on the sum of the volumes is as follows: The three-dimensional point cloud data of the plate to be processed is divided into a plurality of supervoxels by using a supervoxel clustering point cloud segmentation algorithm; The median of the coordinate values ​​of all data points in the three-dimensional point cloud data of the hot-rolled plate of normal pickling is used as the standard height; The data points in the supervoxel are divided into two parts based on the standard height, which are recorded as convex point cloud data and concave point cloud data; The volumes of the convex area and the concave area enclosed by the convex point cloud data and the concave point cloud data and the standard height are obtained through the surface reconstruction algorithm, and the sum of the volumes of the convex area and the concave area is used as the surface concaveness of each supervoxel.

5. A 1000MPa grade bainitic steel cold rolling production method as claimed in claim 1, characterized in that: The method of screening supervoxels based on the surface concavity, obtaining the fitted surface and boundary data points of the screened supervoxels, and obtaining the consistent surface characteristic coefficient of the supervoxels according to the similarity of the vectors obtained on the fitted surface of all boundary data points in the neighborhood of the boundary data points is: All surface concavity and convexity levels are used as the input of the maximum inter-class variance to obtain the partition threshold, and the set of supervoxels greater than the partition threshold is recorded as an abnormal supervoxel set; For the supervoxels of the abnormal supervoxel set, a surface fitting algorithm is used to obtain a fitting surface; The boundary data points are obtained by using a contour extraction algorithm, and the Gaussian curvature and the average curvature of the boundary data points in the fitting surface are obtained, and the vector composed of the obtained Gaussian curvature and the average curvature is recorded as the local surface feature vector of the boundary data points; For any boundary pixel point, a neighborhood is constructed with it as the center, and a preset number of boundary data points closest to it are obtained as neighborhood data points. The cosine similarity of the local surface feature vectors between all neighborhood data points is obtained, and the mean of all cosine similarities is taken as the surface feature consistency of the boundary data points. The consistent degrees of all surface features in the supervoxel are averaged to obtain the consistent surface feature coefficient of the supervoxel.

6. A 1000MPa grade bainitic steel cold rolling production method as claimed in claim 1, characterized in that: The method of obtaining the normal vector of each boundary data point and obtaining the surface sharpness of the supervoxel according to the similarity of the normal vectors of the boundary data point and the adjacent data points is: A circle of preset size is constructed with the boundary data point as the center, the data points in the circle are recorded as adjacent points, and the normal vectors of the fitting surface corresponding to the supervoxel at the boundary data point and each adjacent point are obtained; Calculate the cosine similarity between the normal vector of the boundary data point and the normal vector of each of its adjacent points, and obtain the local sharpness of each boundary data point based on all cosine similarities; Local sharpness is negatively correlated with all cosine similarities; The local sharpness of all boundary data points in the supervoxel is normalized by taking the mean value to obtain the surface sharpness of the supervoxel.

7. A 1000MPa grade bainitic steel cold rolling production method as claimed in claim 5, characterized in that: The method of obtaining regional characteristic values ​​based on the consistent surface characteristic coefficient and surface sharpness of supervoxels; selecting supervoxels according to the regional characteristic values, and dividing supervoxels into corroded supervoxels and oxidized supervoxels in combination with the protruding volume and the concave volume of the supervoxels; and obtaining the pickling coefficient of the to-be-processed board according to the difference in the surface concave-convexity degree of the corroded supervoxels and the oxidized supervoxels is as follows: The regional eigenvalue is negatively correlated with the consistent surface characteristic coefficient and the surface sharpness; A characteristic value threshold is preset, and a set consisting of all supervoxels whose regional characteristic values ​​in the abnormal supervoxel set are greater than the characteristic value threshold is recorded as a first supervoxel set of the plate to be processed; For each supervoxel in the first supervoxel set, the ratio of the convex volume to the concave volume in the supervoxel is recorded as the concave-convex feature value of the supervoxel; The concave-convex eigenvalues ​​of all supervoxels in the first supervoxel set are used as the input of the maximum inter-class variance algorithm, and the concave-convex eigenvalue threshold is output, and the supervoxels whose concave-convex eigenvalues ​​are less than the concave-convex eigenvalue threshold are recorded as oxidized supervoxels, and the remaining supervoxels in all supervoxels are recorded as eroded supervoxels; The surface concave-convex degrees of all oxidized supervoxels in the first supervoxel set are summed to obtain a first oxidized concave-convex value of the first supervoxel, and the surface concave-convex degrees of all corroded supervoxels in the first supervoxel set are summed to obtain a first corroded concave-convex value of the first supervoxel; and the pickling coefficient of the to-be-processed board is obtained based on the first oxidized concave-convex value and the first corroded concave-convex value; The pickling coefficient of the plate to be processed is positively correlated with the first oxidation concave-convex value, and negatively correlated with the first corrosion concave-convex value.

8. A 1000MPa grade bainitic steel cold rolling production method as claimed in claim 1, characterized in that: The method for obtaining the pickling time of the next hot-rolled plate based on the pickling coefficient and pickling time of the plate to be processed is: , Indicates the pickling time of the board to be treated, Indicates the preset parameters for adjusting the pickling time. is the rounding function, Indicates the pickling coefficient of the plate to be treated, represents the normalization function, Indicates the pickling time of the next hot-rolled plate to be processed.

9. A 1000MPa grade bainite steel cold rolling production method as claimed in claim 1, characterized in that: During the annealing treatment, the heating temperature is 850-880° C., the holding time is 120-150 seconds, the temperature is slowly cooled to 730-760° C., and the temperature is quickly cooled to 220-240° C., and tempered at 350-370° C. for 380 seconds to 510 seconds.

10. A 1000MPa grade bainite steel cold rolling production method as claimed in claim 1, characterized in that: In the step of pickling again to obtain the rigid body, the pickling concentration is 12-16% dilute hydrochloric acid, the pickling temperature is 60-70° C., and the pickling time is ≥10s.

Images