A control method for a hot-rolling skin pass mill skin pass pattern plate

Abstract

The present application belongs to the technical field of plate rolling, and provides a control method for a patterned plate of a hot rolling and skin passing mill train, comprising: constructing a multi-dimensional deviation index system about the pattern height, matching with a constructed index-deviation mode matching matrix, determining the height deviation mode of the patterned plate, and performing adaptability evaluation; if the adaptability is low, analyzing the adjustment mapping relationship between the rolling force of the hot rolling and skin passing mill train and the patterned plate height, calculating the rolling force adjustment amount of the hot rolling mill train in each adjustment partition; determining the adjustment influence coefficient matrix between the patterned plate adjustment partitions, calculating the passive adjustment amount of the patterned plate height of each affected partition, judging whether the patterned plate height offset adjustment of the affected partition is needed, identifying the offset adjustment partition and calculating the rolling force compensation range of the offset adjustment partition, which improves the comprehensiveness of the patterned plate height deviation identification, the accuracy of the adjustment strategy and the stability of the production process.

Description

Technical Field

[0001] This invention belongs to the field of sheet metal rolling technology, specifically a control method for leveling patterned sheets in a hot rolling leveling unit. Background Technology

[0002] In the production process of patterned steel plates in hot rolling mills, existing technologies have many specific problems in controlling the pattern height deviation, resulting in poor transverse height uniformity and a high defect rate. These problems are mainly reflected in the following aspects:

[0003] First, deviation pattern recognition is one-sided and crude. Existing technologies mostly rely on single-dimensional indicators (such as judging deviations based solely on the height difference between the edge and the middle), which cannot cover the complex deviation characteristics of the entire width of the strip. It is difficult to identify depressions or protrusions in local areas such as 1 / 4 width or 3 / 4 width, and it cannot capture asymmetrical deviation patterns such as wavy (periodic alternation of high and low) or unilateral tilt (linear gradual deviation). As a result, the problem of "the edge and middle are qualified, but local areas exceed the standard" often occurs.

[0004] Secondly, the deviation judgment criteria are rigid. In existing technologies, the height deviation modes are mostly fixed values ​​(such as the preset symmetrical edge deviation as the only judgment criterion). When the actual deviation mode (such as local concavity) does not match the set mode, the fixed standard is still used for judgment, resulting in a high misjudgment rate and the adjustment strategy being out of touch with the actual deviation requirements.

[0005] Third, the cascading effects of adjustments are not considered. Since the strip steel is a continuous whole in the transverse direction, existing technologies do not quantify the cascading effects of adjusting the rolling force in a certain zone on adjacent areas (e.g., adjusting the rolling force in the 1 / 4 width zone may lead to a passive increase or decrease in height in the 3 / 8 width zone). Due to the lack of analysis and compensation for these cascading effects, isolated adjustments often trigger new height deviations (e.g., exceeding tolerances in adjacent areas), thus exacerbating transverse inhomogeneity.

[0006] Fourth, there is a lack of offset adjustment mechanisms. Even if the adjustment causes a cascading effect, the existing technology lacks the ability to identify the affected areas that require offset adjustment and to calculate the targeted rolling force compensation range, thus leading to the continuous accumulation of cascading deviations.

[0007] Therefore, the present invention provides a method for controlling the leveling of patterned plates in a hot-rolled leveling unit. Summary of the Invention

[0008] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0009] The technical solution adopted by this invention to solve its technical problem is: a control method for leveling patterned plates in a hot-rolled leveling unit, comprising:

[0010] Step 1: By analyzing the pattern height of the patterned strip in the width direction of the strip, a multi-dimensional deviation index system for the pattern height is constructed, and this system is matched with the constructed index-deviation pattern matching matrix to determine the height deviation pattern of the patterned strip.

[0011] Step 2: Compare the height deviation mode of the patterned plate with the height deviation mode set for the patterned plate to evaluate the adaptability of the height deviation mode set for the patterned plate.

[0012] Step 3: If the adaptability is low, analyze the adjustment mapping relationship between the rolling force of the hot rolling leveling unit and the height of the patterned plate, and combine the pattern height adjustment amount of the patterned plate adjustment zone to calculate the rolling force adjustment amount of the hot rolling unit in each adjustment zone under the determined patterned plate height deviation mode.

[0013] Step 4: Determine the adjustment influence coefficient matrix between the pattern plate adjustment zones under the height deviation mode, and calculate the passive adjustment amount of the pattern height of each affected zone in combination with the rolling force adjustment amount in each adjustment zone under the height deviation mode. Determine whether the pattern height offset adjustment needs to be performed on the affected zones, and identify the offset adjustment zones in the affected zones and calculate the rolling force compensation range of the offset adjustment zones based on the judgment results.

[0014] Furthermore, the deviation indicators included in the multi-dimensional deviation index system include the maximum regional deviation, adjacent partition deviation, local target deviation, and deviation standard deviation.

[0015] The process of constructing a multi-dimensional deviation index system for pattern height is as follows:

[0016] Divide the strip into n consecutive sections along its width and calculate the pattern height for each section.

[0017] Among them, the maximum deviation of the region is the difference between the maximum and minimum values ​​of the pattern height of all zones;

[0018] The adjacent partition deviation is the height difference between any two adjacent partitions;

[0019] Local target deviation is the difference between the average pattern height and the target pattern height for each zone;

[0020] The standard deviation of the deviation is the standard deviation of the pattern height for all zones.

[0021] Furthermore, the height deviation pattern of the patterned plate is determined by matching the multi-dimensional deviation index contained in the multi-dimensional deviation index system of the pattern height with the index-deviation pattern matching matrix.

[0022] The index-deviation pattern matching matrix is ​​constructed by using multi-dimensional deviation indices, including each height deviation pattern, the corresponding maximum deviation of the region, the deviation of adjacent zones, the deviation of local targets, and the standard deviation of deviation.

[0023] Furthermore, the process of comparing the height deviation pattern of the patterned plate with the height deviation pattern set for the patterned plate is as follows:

[0024] Obtain the height deviation patterns of multiple patterned plates from the same batch within a historical production cycle, and compare them with the deviation patterns set for the patterned plates.

[0025] If the height deviation mode of the pattern plate is different from the deviation mode set for the pattern plate, then the deviation mode set for the pattern plate will be calibrated for mismatch.

[0026] Based on the mismatch calibration, the deviation mode calibration value and the deviation mode non-coverage value are processed and analyzed, and then multiplied to obtain the evaluation adaptation value of the height deviation mode set by the pattern plate.

[0027] If the evaluation adaptation value of the height deviation mode set by the pattern plate is greater than or equal to the evaluation adaptation threshold, it means that the adaptation of the height deviation mode set by the pattern plate is low; otherwise, the adaptation is high.

[0028] Furthermore, the deviation mode calibration value and the deviation mode non-coverage value are obtained in the following way:

[0029] The proportion of times mismatch calibration was performed on the deviation pattern set for the pattern plate was statistically analyzed to obtain the deviation pattern calibration value.

[0030] Obtain the adjustment zones of the pattern plate under the set deviation mode and compare them with the adjustment zones of the pattern plate under the height deviation mode.

[0031] If the adjustment zone of the pattern plate in the set deviation mode is different from the adjustment zone of the pattern plate in the height deviation mode, then the adjustment zone of the pattern plate in the set deviation mode will be marked as an overlay zone.

[0032] The proportion of non-covered partitions in the adjustment partitions under the pattern plate height deviation mode is statistically analyzed to obtain the proportion of non-covered partitions in the pattern plate. The proportion of non-covered partitions in all pattern plates is then averaged to obtain the non-covered value of the deviation mode.

[0033] Furthermore, the calculation process for the rolling force adjustment amount of the hot rolling mill in each adjustment zone is as follows:

[0034] The rolling force adjustment amount and the pattern height change amount of each adjustment zone are obtained during multiple rolling adjustments in the height deviation mode.

[0035] By fitting the rolling force adjustment amount and the pattern height change amount of each adjustment zone after multiple rolling adjustments using the least squares method, a mapping relationship model between the rolling force of the hot rolling leveling mill and the pattern plate adjustment height is obtained.

[0036] The pattern height adjustment amount of the adjustment zone is input into the mapping relationship model to obtain the rolling force adjustment amount of the hot rolling mill within the adjustment zone;

[0037] Among them, the adjustment amount of the pattern height of the adjustment zone is the local target deviation of the adjustment zone.

[0038] Furthermore, the process of determining the adjustment influence coefficient matrix between the pattern plate adjustment zones under the height deviation mode is as follows:

[0039] Construct an n*n adjustment influence coefficient matrix, denoted as K, based on the number of continuous partitions in the strip direction;

[0040] In the adjustment influence coefficient matrix K, row index i represents the affected partition, column index j represents the adjusted partition, and matrix element K[i][j] represents the change in rolling force of partition j and pattern height of partition i, i. ,in, This indicates the rolling force adjustment amount for partition j. This indicates the change in pattern height within partition i.

[0041] Furthermore, the matrix element K[i][j] is obtained as follows:

[0042] Obtain the rolling force adjustment amount and the pattern height change amount of the affected partition during multiple adjustments, and perform fitting. Then, solve for K[i][j] using the least squares method, i.e.: .

[0043] Furthermore, the process of determining whether pattern height offset adjustment is needed for the affected partitions, and identifying the offset adjustment partitions among the affected partitions based on the determination result, is as follows:

[0044] The passive adjustment of the pattern height of the affected zone is calculated by multiplying the rolling force adjustment amount within the adjustment zone with the matrix element K[i][j] in the adjustment influence coefficient matrix K.

[0045] The passive pattern height of the affected partition is calculated by summing the passive pattern height adjustment of the affected partition with the current pattern height of the affected partition.

[0046] If the difference between the passive pattern height and the target pattern height of the affected partition is greater than the local target deviation threshold, the affected partition will be marked as an offset adjustment partition.

[0047] Furthermore, the process of calculating the rolling force compensation range of the offset adjustment zone is as follows:

[0048] The difference between the passive pattern height of the affected zone and the target pattern height is used as the maximum pattern height compensation amount to offset the adjustment zone.

[0049] The difference between the passive pattern height and the target pattern height of the affected zone is calculated by subtracting the local target deviation threshold to obtain the minimum pattern height compensation amount for the offset adjustment zone.

[0050] The minimum and maximum pattern height compensation amounts for the offset adjustment zone are input into the mapping model to obtain the rolling force compensation range of the offset adjustment zone.

[0051] The beneficial effects of this invention are as follows: By constructing a multi-dimensional deviation index system for the pattern height of a patterned strip by uniformly dividing the strip width into continuous zones, and matching this system with the constructed index-deviation pattern matching matrix, the height deviation pattern of the patterned strip is determined. The height deviation pattern of the patterned strip is then compared with the height deviation pattern set for the patterned strip to evaluate its suitability. If the suitability is low, the adjustment mapping relationship between the rolling force of the hot rolling mill and the pattern height is analyzed, and combined with the pattern height adjustment amount of the adjustment zones, the rolling force adjustment amount of the hot rolling mill in each adjustment zone under the determined pattern height deviation pattern is calculated. Finally, the adjustment influence coefficient matrix between the patterned strip adjustment zones under the height deviation pattern is determined, and combined with the rolling force adjustment in each adjustment zone under the height deviation pattern, the adjustment is further refined. This invention calculates the passive adjustment amount of the pattern height in each affected zone, determines whether pattern height offset adjustment is needed for the affected zone, and identifies offset adjustment zones and calculates the rolling force compensation range for these zones based on the judgment results. The technical solution achieves accurate identification of pattern plate height deviation patterns by constructing a multi-dimensional deviation index system and an index-deviation pattern matching matrix. It dynamically optimizes deviation judgment criteria by evaluating the adaptability of the set deviation patterns, avoiding misjudgments caused by rigid deviation patterns. Combining the mapping relationship between rolling force and adjustment height, it accurately calculates the rolling force adjustment amount for each zone and quantifies the associated effects of rolling force adjustment using an adjustment influence coefficient matrix. By identifying offset adjustment zones and calculating the rolling force compensation range, it effectively avoids new pattern height deviations caused by isolated rolling force adjustments. The overall technical effect is to improve the comprehensiveness of pattern plate height deviation identification, the accuracy of adjustment strategies, and the stability of the production process, ultimately improving the uniformity of the transverse height of the pattern plate and reducing the quality defect rate caused by improper pattern height deviation adjustment. Attached Figure Description

[0052] The invention will now be further described with reference to the accompanying drawings.

[0053] Figure 1 This is a flowchart illustrating the steps of a control method for leveling patterned plates using a hot-rolled leveling mill, as described in an embodiment of the present invention.

[0054] Figure 2 This is a logic diagram of a control method for leveling patterned plates in a hot-rolled leveling unit according to an embodiment of the present invention. Detailed Implementation

[0055] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0056] Please see Figures 1-2 As shown in the embodiment of the present invention, a method for controlling the leveling of patterned plates in a hot-rolled leveling unit includes the following steps:

[0057] Step 1: By analyzing the pattern height of the patterned strip in the width direction of the strip, a multi-dimensional deviation index system for the pattern height is constructed, and this system is matched with the constructed index-deviation pattern matching matrix to determine the height deviation pattern of the patterned strip.

[0058] In step one, the deviation indicators included in the multi-dimensional deviation indicator system include: maximum regional deviation. Adjacent partition deviation ΔH 邻 Local target deviation ΔH 目 And the standard deviation σ;

[0059] In step one, the process of constructing a multi-dimensional deviation index system for pattern height is as follows:

[0060] Divide the strip width direction evenly into n consecutive sections (e.g., n=10-20, adjusted according to the strip width, each section width 50-100mm), and label them as Z1 (left edge), Z2, ..., Z... n (Right edge), where Z n This indicates the nth partition along the width of the strip.

[0061] Based on the sampling points of the laser profilometer every 10mm (let the total number of sampling points be M, M=W / 0.01, where W is the length of the patterned plate in the width direction of the strip), calculate the value of each zone. Pattern height (Mean height of the pattern at all sampling points within the partition);

[0062] According to each partition Pattern height The maximum deviation in the region was calculated. Adjacent partition deviation ΔH邻 Local target deviation ΔH 目 And the standard deviation σ;

[0063] Among them, the maximum deviation in the region For all partition pattern heights The difference between the maximum and minimum values ​​in the range;

[0064] Adjacent partition deviation ΔH 邻 For any adjacent partition With Z i+1 The height difference;

[0065] Local target deviation ΔH 目 Average pattern height for each zone The difference between the height H of the target pattern and the height H of the target pattern;

[0066] The standard deviation σ is the pattern height of all zones. Standard deviation;

[0067] By mapping the multi-dimensional deviation indicators of pattern height to their corresponding preset threshold values, a multi-dimensional deviation indicator system is obtained, as shown in Table 1 below:

[0068] Table 1: Multi-dimensional deviation indicators and corresponding preset indicator thresholds;

[0069]

[0070] In step one, the process of constructing the indicator-deviation pattern matching matrix based on the multi-dimensional deviation indicator system is as follows:

[0071] Based on the multi-dimensional deviation indices corresponding to different height deviation patterns of the patterned plate, an index-deviation pattern matching matrix is ​​constructed. Specifically:

[0072] (1) Symmetrical edge low pattern:

[0073] Features: Two-sided side division ΔH 目 Negative (< -0.15mm), middle section ΔH 目 The gradient ΔH between adjacent partitions is close to 0. 邻 It gradually decreases from the edge to the middle (without steep abrupt changes).

[0074] Example: If Central If the gradient is gentle, it is determined to be a low-gradient pattern at the symmetrical edge.

[0075] (2) Single-sided tilt mode:

[0076] Features: One side edge (e.g.: )ΔH目 It is negative (severely low), on the other side (Z) n )ΔH 目 It is positive (too high), from To Z n of It increases linearly, with the gradient ΔH between adjacent partitions. 邻 Uniform (e.g., all are 0.2mm).

[0077] Example: , As the partition number increases, ΔH 邻 ≈0.2mm, determined to be a unilateral tilt mode;

[0078] (3) Local depression pattern:

[0079] Features: A partition that is neither a border nor a center (e.g., )ΔH 目 It is negative (< -0.15mm), and its two sides are divided into zones. ΔH 目 Gradients close to 0 or positive, adjacent partition gradients >0.3mm (steep abrupt change).

[0080] Example: (5.1mm, ΔH) 目 =+0.1), (Steep) is determined to be a local depression pattern.

[0081] (4) Wave-shaped deviation pattern:

[0082] Feature: ΔH 目 Alternating between positive and negative (e.g.) ), adjacent partition gradient ΔH 邻 Repeatedly switching between positive and negative, and >0.4mm, standard deviation σ>0.1mm (poor overall uniformity).

[0083] Example: In the table above, , =0.8mm, σ=0.28mm, judged as a wavy deviation pattern;

[0084] As shown in Table 2 below, the constructed index-deviation pattern matching matrix is ​​as follows:

[0085] Table 2: Height Deviation Patterns and Corresponding Multi-Dimensional Deviation Indicators;

[0086]

[0087] The multi-dimensional deviation indexes contained in the multi-dimensional deviation index system of pattern height are matched with the index-deviation pattern matching matrix to determine the height deviation pattern of the pattern plate.

[0088] Understandably, the purpose of step one is:

[0089] Constructing a multi-dimensional deviation index system and an index-deviation pattern matching matrix is ​​beneficial for judging the height deviation pattern of patterned plates, and facilitates accurate monitoring and evaluation of height deviation when the hot rolling mill is leveling patterned plates.

[0090] Determining the height deviation pattern of the patterned plate is beneficial for subsequent comparison with the height deviation pattern set on the patterned plate, thereby determining whether the current height deviation pattern of the patterned plate is suitable, and thus making effective and timely adjustments to the height deviation pattern of the patterned plate.

[0091] The height deviation pattern of the patterned plate can be determined so that effective and timely adjustments can be made when the current height deviation pattern of the patterned plate is not suitable. This is beneficial for accurate rolling adjustment of each patterned plate area when the hot press unit flattens the patterned plate in the subsequent process.

[0092] Step 2: Compare the height deviation mode of the patterned plate with the height deviation mode set for the patterned plate to evaluate the adaptability of the height deviation mode set for the patterned plate.

[0093] In step two, the height deviation mode set for the patterned plate is either the symmetrical edge low mode or the symmetrical edge high mode. Specifically, the four-roll mill usually only uses the height difference between the edge and the middle as the control basis. This control basis is usually suitable for the symmetrical edge low mode or the symmetrical edge high mode, and usually ignores deviation modes such as wavy deviation mode and local concave deviation mode.

[0094] In step two, the process of comparing the height deviation pattern of the patterned plate with the height deviation pattern set for the patterned plate is as follows:

[0095] Obtain the height deviation patterns of multiple patterned plates from the same batch within a historical production cycle, and compare them with the deviation patterns set for the patterned plates.

[0096] If the height deviation mode of the pattern plate is different from the deviation mode set for the pattern plate, then the deviation mode set for the pattern plate will be calibrated for mismatch.

[0097] If the height deviation mode of the pattern plate is the same as the deviation mode set for the pattern plate, no operation will be performed.

[0098] The proportion of times mismatch calibration was performed on the deviation pattern set for the pattern plate was statistically analyzed to obtain the deviation pattern calibration value.

[0099] Based on the deviation mode set by the mismatched calibrated pattern plate, the adjustment zone of the pattern plate under the set deviation mode is obtained and compared with the adjustment zone of the pattern plate under the height deviation mode.

[0100] If the adjustment zone of the pattern plate in the set deviation mode is different from the adjustment zone of the pattern plate in the height deviation mode, then the adjustment zone of the pattern plate in the set deviation mode will be marked as an overlay zone.

[0101] If the adjustment zone of the pattern plate in the set deviation mode is the same as the adjustment zone of the pattern plate in the height deviation mode, then the adjustment zone of the pattern plate in the set deviation mode is marked as the coverage zone.

[0102] The proportion of non-covered partitions in the adjustment partitions under the pattern plate height deviation mode is statistically analyzed to obtain the proportion of non-covered partitions in the pattern plate. The proportion of non-covered partitions in all pattern plates is averaged to obtain the non-covered value of the deviation mode.

[0103] The deviation mode calibration value and the deviation mode non-coverage value are multiplied to obtain the evaluation adaptation value of the height deviation mode set for the patterned plate.

[0104] In step two, the adjustment zones of the pattern plate in the set deviation mode and in the height deviation mode are determined by the core features of the height deviation mode.

[0105] For example, the core features of the local concave deviation pattern are: there are 1-3 consecutive partitions whose heights are out of tolerance, and the height difference between them and the adjacent non-out-of-tolerance partitions exceeds the gradient threshold.

[0106] The method for adjusting partitions is as follows:

[0107] Filter out out-of-tolerance partitions: Traverse all partitions and find the pattern height. The partitions that do not fall within the normal range of pattern height (H - pattern height error, H + pattern height error) are denoted as S={Z}. x Z x+1 ......Z y}, where Z x Zy represents the x-th partition, and Zy represents the y-th partition, where x and y are the partition numbers, respectively.

[0108] Determine the consecutive quantities; if y-x+1≤3, then calculate the pattern height gradient difference.

[0109] Left adjacent partition (if x > 1): Calculate |H x -H x-1 | If it is greater than the adjacent partition deviation threshold;

[0110] Right adjacent partition (if y < n): Calculate |H y -H y-1 | If it is greater than the adjacent partition deviation threshold;

[0111] If at least one side of the gradient difference exceeds the limit, it is considered to have a significant height difference with the surrounding area, and an adjustment zone is determined, specifically as follows:

[0112] If it is a depression <H− Pattern height error: Adjustment area is S (pattern height needs to be increased);

[0113] If it is a protrusion >H+ Pattern height error: Adjustment area is S (pattern height needs to be reduced);

[0114] The core characteristics of the wave-shaped deviation pattern are: the height deviation exhibits a periodic alternation of "positive out-of-tolerance → negative out-of-tolerance → positive out-of-tolerance" along the width direction, and the out-of-tolerance area is continuous within the period and the period length is stable (e.g., every 3-5 partitions constitute one period).

[0115] The method for adjusting partitions is as follows:

[0116] Calculate ΔH for each partition 目 The sequence {ΔH} is obtained. 目1 ,ΔH 目2 ,...,ΔH 目n}, where ΔH 目n This represents the deviation value for the nth partition;

[0117] Statistical analysis of two adjacent positive deviation peaks (ΔH) 目i > The number of interval partitions (period T) for pattern height error, where ΔH 目i This represents the local target deviation of the i-th partition. If the fluctuation of 3 consecutive periods T is ≤ 1 partition (e.g., T = 4, 4, 5), it is considered periodically stable, and the peak and trough are located.

[0118] Peak region: ΔH 目i > Continuous partitioning of pattern height error (positive out-of-tolerance);

[0119] Trough region: ΔH 目i <− Continuous partitioning of pattern height error (negative out-of-tolerance);

[0120] All peak areas (where height needs to be reduced) and trough areas (where height needs to be increased) are adjustment zones;

[0121] In step two, the process of evaluating the adaptability of the height deviation pattern set for the patterned plate is as follows:

[0122] Compare the evaluation adaptation value of the height deviation mode set for the pattern plate with the evaluation adaptation threshold.

[0123] If the evaluation adaptation value of the height deviation mode set by the pattern plate is greater than or equal to the evaluation adaptation threshold, it means that the adaptation of the height deviation mode set by the pattern plate is low.

[0124] If the evaluation adaptation value of the height deviation mode set by the pattern plate is less than the evaluation adaptation threshold, it means that the adaptation of the height deviation mode set by the pattern plate is high, and no operation is performed.

[0125] Understandably, the assessment fit value is calculated using the deviation mode calibration value and the deviation mode non-coverage value. The deviation mode calibration value reflects the proportion of mismatches between the height deviation mode and the set deviation mode in the pattern plate. The higher the proportion of mismatches, the lower the fit of the deviation mode set for the pattern plate. Similarly, the deviation mode non-coverage value reflects the coverage difference between the adjustment zones involved in the two deviation modes when the height deviation mode and the set deviation mode do not match. The greater the coverage difference, the lower the fit of the deviation mode set for the pattern plate.

[0126] Step 3: If the adaptability is low, analyze the adjustment mapping relationship between the rolling force of the hot rolling leveling unit and the height of the patterned plate, and combine the adjustment amount of the patterned plate height in the adjustment zone to calculate the rolling force adjustment amount of the hot rolling unit in each adjustment zone under the determined patterned plate height deviation mode.

[0127] In step three, the calculation process for the rolling force adjustment in each adjustment zone of the hot rolling mill is as follows:

[0128] Obtain historical rolling adjustment data of patterned steel plate, which includes the rolling force adjustment amount of each adjustment zone under the height deviation mode and the change amount of pattern height in each adjustment zone.

[0129] By fitting the rolling force adjustment amount and the pattern height change amount of each adjustment zone after multiple rolling adjustments using the least squares method, a mapping relationship model between the rolling force of the hot rolling leveling mill and the pattern plate adjustment height is obtained.

[0130] For example, suppose the rolling force adjustments for each adjustment zone are {yt1, yt2, ..., ytz}, where ytz represents the rolling force adjustment for the z-th adjustment zone, and the pattern height changes for each adjustment zone are {gd1, gd2, ..., gdz}, where gdz represents the pattern height change for the z-th adjustment zone. Then, the mapping relationship model obtained after least squares fitting is yti=k*gdi+b, where k is the slope, b is the intercept, and i represents the adjustment zone number.

[0131] The pattern height adjustment amount of the adjustment zone is input into the mapping relationship model to obtain the rolling force adjustment amount of the hot rolling mill within the adjustment zone;

[0132] Among them, the adjustment amount of the pattern height of the adjustment zone is the local target deviation ΔH of the adjustment zone. 目 ;

[0133] It should be noted that the height deviation pattern of the patterned plate has been determined through a multi-dimensional deviation index system and an index-deviation pattern matching matrix.

[0134] Step 4: Determine the adjustment influence coefficient matrix between the pattern plate adjustment zones under the height deviation mode, and calculate the passive adjustment amount of the pattern height of each affected zone in combination with the rolling force adjustment amount in each adjustment zone under the height deviation mode. Determine whether the pattern height offset adjustment needs to be performed on the affected zones, and identify the offset adjustment zone in the affected zones and calculate the rolling force compensation range of the offset adjustment zone based on the judgment result.

[0135] In step four, the process of determining the adjustment influence coefficient matrix between the pattern plate adjustment zones under the height deviation mode is as follows:

[0136] Construct an n*n adjustment influence coefficient matrix based on the number of continuous partitions in the strip direction (n is the number of transverse partitions in the strip, such as a 10×10 matrix if there are 10 partitions), denoted as K;

[0137] In the adjustment influence coefficient matrix K, the row index i represents the affected partition number (i.e., the partition whose height may change), the column index j represents the adjusted partition, and the matrix element K[i][j] in the adjustment influence coefficient matrix K represents the change in pattern height (mm) of partition i when the rolling force of partition j is 1kN. ( This indicates the rolling force adjustment amount for partition j. (This represents the change in pattern height in partition i).

[0138] The matrix element K[i][j] is obtained as follows:

[0139] Based on any one adjusted partition and any corresponding affected partition;

[0140] Obtain the rolling force adjustment amount and the pattern height change amount of the affected zones during multiple adjustments, and perform fitting, i.e.: (Error constant);

[0141] The least squares method is used to solve for K[i][j], i.e. ;

[0142] For example: by adjusting the rolling force in region Z3 (j=3), the relationship between ΔF3 and Δhi in each region is obtained, and K[1][3] (Z1 is affected by Z3), K[2][3] (Z2 is affected by Z3) ... K

[10] [3] (Z10 is affected by Z3) is fitted.

[0143] In step four, the process of determining whether pattern height offset adjustment is needed for the affected partitions, and identifying the offset adjustment partitions among the affected partitions based on the determination result, is as follows:

[0144] The passive adjustment of the pattern height of the affected zone is calculated by multiplying the rolling force adjustment amount within the adjustment zone with the matrix element K[i][j] in the adjustment influence coefficient matrix K.

[0145] For example, K[2][3]=0.02mm / kN (for every 1kN increase in Z3, the height of Z2 increases by 0.02mm). If Z3 is increased by 20kN, then Z2 will increase by 0.02×20=0.4mm.

[0146] The passive pattern height of the affected partition is calculated by summing the passive pattern height adjustment of the affected partition with the current pattern height of the affected partition.

[0147] It is understandable that the passive pattern height of the affected zone refers to the pattern height reached by the affected zone after being affected by the adjustment of the rolling force of the adjustment zone;

[0148] If the difference between the passive pattern height and the target pattern height of the affected zone is greater than the local target deviation threshold, it indicates that after the rolling force is adjusted in the affected zone, the pattern height of the affected zone is affected, resulting in a deviation (local target deviation ΔH) from the target pattern height. 目 If the pattern height is too large and does not meet the requirements, then pattern height offset adjustment is required. The affected area will be marked as offset adjustment area.

[0149] If the difference between the passive pattern height and the target pattern height of the affected zone is less than or equal to the local target deviation threshold, it means that after the rolling force is adjusted in the affected zone, although the pattern height of the affected zone is affected, the deviation from the target pattern height (local target deviation ΔH) remains within acceptable limits. 目 The value is relatively small and meets the requirements, therefore no action is taken.

[0150] In step four, the process of calculating the rolling force compensation range of the offset adjustment zone is as follows:

[0151] The difference between the passive pattern height of the affected zone and the target pattern height is used as the maximum pattern height compensation amount to offset the adjustment zone.

[0152] The difference between the passive pattern height and the target pattern height of the affected zone is calculated by subtracting the local target deviation threshold to obtain the minimum pattern height compensation amount for the offset adjustment zone.

[0153] It is understandable that the maximum pattern height compensation is to keep the pattern height of the affected area consistent with the target pattern height, while the minimum pattern height compensation is to keep the pattern height of the affected area within the local target deviation range from the target pattern height.

[0154] The minimum and maximum pattern height compensation amounts of the offset adjustment zone are input into the mapping relationship model to obtain the rolling force compensation range of the offset adjustment zone.

[0155] It should be noted that the effect of the offset adjustment zone on the pattern height is opposite when the rolling force is compensated. For example, if the pattern height increases due to the impact of the offset adjustment zone, the pattern height of the offset adjustment zone will decrease when the rolling force is compensated, thereby offsetting the effect.

[0156] The technical solution of this invention is as follows: A multi-dimensional deviation index system for pattern height is constructed by uniformly dividing the patterned strip width into continuous zones based on the pattern height. This system is then matched with the constructed index-deviation pattern matching matrix to determine the pattern height deviation pattern. The pattern height deviation pattern is compared with the pattern height deviation pattern set for the patterned strip to evaluate its suitability. If the suitability is low, the adjustment mapping relationship between the rolling force of the hot rolling mill and the pattern height is analyzed. Combined with the pattern height adjustment amount of the adjustment zones, the rolling force adjustment amount of the hot rolling mill in each adjustment zone under the determined pattern height deviation pattern is calculated. Finally, the adjustment influence coefficient matrix between the patterned strip adjustment zones under the height deviation pattern is determined, and the rolling force in each adjustment zone under the height deviation pattern is combined with the rolling force... The adjustment amount is calculated to passively adjust the pattern height of each affected zone, determining whether pattern height offset adjustment is needed for the affected zone. Based on the determination result, offset adjustment zones are identified within the affected zones, and the rolling force compensation range for the offset adjustment zones is calculated. This invention achieves accurate identification of pattern plate height deviation patterns by constructing a multi-dimensional deviation index system and an index-deviation pattern matching matrix. By evaluating the adaptability of the set deviation patterns and dynamically optimizing the deviation judgment criteria, misjudgments caused by rigid deviation patterns are avoided. Combining the mapping relationship between rolling force and adjustment height, the rolling force adjustment amount for each zone is accurately calculated, and the cascading effects of rolling force adjustment are quantified using an adjustment influence coefficient matrix. By identifying offset adjustment zones and calculating the rolling force compensation range, new pattern height deviations caused by isolated rolling force adjustments are effectively avoided. The overall technical effect is to improve the comprehensiveness of pattern plate height deviation identification, the accuracy of adjustment strategies, and the stability of the production process, ultimately improving the uniformity of the transverse height of the pattern plate and reducing the quality defect rate caused by improper pattern height deviation adjustment.

[0157] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method of controlling a patterned skin pass mill of a hot skin pass mill train, characterized by: Comprise: Step one: by the pattern plate strip steel width direction of the uniform division of the continuous partition pattern height, the construction about the pattern height of multi-dimensional deviation index system, and with the constructed index-deviation mode matching matrix matching, determine the height deviation mode of the pattern plate; Step two: compare the height deviation mode of the pattern plate with the height deviation mode set by the pattern plate, and evaluate the adaptability of the height deviation mode set by the pattern plate; Step three: if the adaptability is low, analyze the adjustment mapping relationship between the rolling force of the hot rolling and flattening unit and the pattern plate height, and combine the pattern height adjustment amount of the pattern plate adjustment partition to calculate the rolling force adjustment amount of the hot rolling unit in each adjustment partition under the determined pattern plate height deviation mode; Step four: determine the adjustment influence coefficient matrix between the adjustment partitions of the pattern plate under the height deviation mode, and combine the rolling force adjustment amount in each adjustment partition under the height deviation mode to calculate the passive adjustment amount of the pattern height of each affected partition, judge whether it is necessary to make pattern height offset adjustment in the affected partition, and identify the offset adjustment partition and calculate the rolling force compensation range of the offset adjustment partition in the affected partition according to the judgment result.

2. The control method of claim 1, wherein: the deviation indexes in the multi-dimensional deviation index system include regional maximum deviation, adjacent partition deviation, local target deviation and deviation standard deviation; the process of constructing the multi-dimensional deviation index system about the pattern height is: dividing the strip steel width direction into n continuous partitions, and calculating the pattern height of each partition; wherein the regional maximum deviation is the difference between the maximum and minimum values of the pattern height of all partitions; the adjacent partition deviation is the height difference of any adjacent partitions; the local target deviation is the difference between the average pattern height of each partition and the target pattern height; the deviation standard deviation is the standard deviation of the pattern height of all partitions.

3. The control method of claim 1, wherein: the height deviation mode of the pattern plate is determined by matching the multi-dimensional deviation indexes in the multi-dimensional deviation index system of the pattern height with the index-deviation mode matching matrix; wherein the index-deviation mode matching matrix is constructed by each height deviation mode and the corresponding regional maximum deviation, adjacent partition deviation, local target deviation and deviation standard deviation.

4. The control method of claim 1, wherein: the process of comparing the height deviation mode of the pattern plate with the height deviation mode set by the pattern plate is: obtaining the height deviation modes of multiple pattern plates of the same batch in the historical production period, and comparing them with the deviation mode set by the pattern plate; if the height deviation mode of the pattern plate is different from the deviation mode set by the pattern plate, then the deviation mode set by the pattern plate is not matched; based on the non-matching calibration, the deviation mode calibration value and the deviation mode non-coverage value are obtained by processing and analysis, and the product processing obtains the evaluation adaptation value of the height deviation mode set by the pattern plate; ​ If the evaluation adaptation value of the height deviation mode set by the pattern plate is greater than or equal to the evaluation adaptation threshold value, it indicates that the adaptability of the height deviation mode set by the pattern plate is low, otherwise, the adaptability is high.

5. The control method of claim 4, wherein the control method further comprises: obtaining the height deviation mode of the pattern plate; and determining the adjustment influence coefficient matrix of the adjustment sub-zones of the pattern plate under the height deviation mode. The deviation mode calibration value and the deviation mode non-coverage value are obtained in the following manner: The number of times of mismatched calibration of the deviation mode set by the pattern plate is counted to obtain the deviation mode calibration value. The adjustment sub-zone of the pattern plate under the set deviation mode is obtained and compared with the adjustment sub-zone of the pattern plate under the height deviation mode. If the adjustment sub-zone of the pattern plate under the set deviation mode is different from the adjustment sub-zone of the pattern plate under the height deviation mode, the adjustment sub-zone of the pattern plate under the set deviation mode is marked as a non-coverage sub-zone. The number proportion of the non-coverage sub-zone in the adjustment sub-zone of the pattern plate under the height deviation mode is counted to obtain the number proportion of the non-coverage sub-zone of the pattern plate, and the number proportions of the non-coverage sub-zones of all the pattern plates are processed by mean value to obtain the deviation mode non-coverage value.

6. The control method of claim 5, wherein the control method further comprises: obtaining the rolling force adjustment amount of the hot rolling mill in each adjustment sub-zone. The rolling force adjustment amount of each adjustment sub-zone under the height deviation mode and the pattern height variation amount of each adjustment sub-zone are obtained. The rolling force adjustment amount of each adjustment sub-zone under the height deviation mode and the pattern height variation amount of each adjustment sub-zone are fitted by the least square method to obtain a mapping relationship model between the rolling force of the hot rolling mill and the adjustment height of the pattern plate. The pattern height adjustment amount of the adjustment sub-zone is input into the mapping relationship model to obtain the rolling force adjustment amount of the hot rolling mill in the adjustment sub-zone.

7. The control method of claim 5, wherein the control method further comprises: determining the adjustment influence coefficient matrix of the adjustment sub-zones of the pattern plate under the height deviation mode. An n*n adjustment influence coefficient matrix K is constructed according to the number of the strip direction continuous sub-zones.

8. The control method of claim 7, wherein the matrix element K[i][j] is obtained in the following manner:

9. The control method of claim 8, wherein the control method further comprises: determining whether the pattern height offset adjustment of the affected sub-zone is needed, and identifying the offset adjustment sub-zone in the affected sub-zone according to the determination result. The rolling force adjustment amount in the adjustment sub-zone is multiplied by the matrix element K[i][j] in the adjustment influence coefficient matrix K to calculate the passive pattern height adjustment amount of the affected sub-zone. In the adjustment influence coefficient matrix K, the row index i contained therein represents the affected partition, the column index j represents the adjustment partition, and the matrix element K[i][j] contained in the adjustment influence coefficient matrix K represents the change amount of the pattern height of the i partition when the rolling force of the j partition is adjusted, that is, wherein, represents the adjustment amount of the rolling force of the j partition, represents the change amount of the pattern height of the i partition. The passive pattern height of the affected sub-zone is obtained by summing the passive pattern height adjustment amount of the affected sub-zone and the current pattern height of the affected sub-zone. ​ The rolling force adjustment amount of the adjustment subzone and the pattern height change amount of the affected subzone during multiple adjustments are obtained and fitted, and K[i][j] is solved by using the least square method, that is: . ​ ​ ​ ​ If the difference between the passive profile height of the affected partition and the target profile height is greater than the local target deviation threshold, the affected partition is marked as a cancellation adjustment partition.

10. The control method of the temper pattern plate of the hot skin pass mill set according to claim 9, characterized in that: The process of calculating the rolling force compensation range of the cancellation adjustment partition is: The difference between the passive profile height of the affected partition and the target profile height is taken as the maximum profile height compensation amount of the cancellation adjustment partition; The difference between the passive profile height of the affected partition and the target profile height is subtracted from the local target deviation threshold to obtain the minimum profile height compensation amount of the cancellation adjustment partition; The minimum profile height compensation amount and the maximum profile height compensation amount of the cancellation adjustment partition are respectively input into the mapping relationship model to obtain the rolling force compensation range of the cancellation adjustment partition.

Images