Square tube forming unit online intelligent deviation rectification control method based on real-time detection feedback

Abstract

The invention discloses a square tube forming unit online intelligent deviation rectification control method based on real-time detection feedback, and relates to the technical field of intelligent welding. The method comprises the steps that the vertex offset, the forming angle deviation and the cross section diagonal deviation of a square tube are obtained; performing deviation analysis to obtain a deviation analysis result; the deviation degree of the current square tube is judged, when the deviation degree is larger than an allowable deviation threshold value, coupling influence analysis is conducted on the vertex offset, the forming angle deviation and the cross section diagonal deviation, and a prediction deviation source is obtained; and performing online optimization on the current square tube control parameters to obtain optimized square tube control parameters, and performing square tube forming process control. The technical problems that an existing square tube forming unit control method cannot capture dynamic changes in the forming process in real time, response lags, efficiency is low, complex coupling deviation sources are difficult to accurately recognize, and then the deviation correction effect is poor are solved.

Description

Technical Field

[0001] This application relates to the field of intelligent welding technology, specifically to an online intelligent correction control method for square tube forming units based on real-time detection feedback. Background Technology

[0002] With increasingly stringent precision requirements for square tube products in modern industry, especially in fields such as building structures, machinery manufacturing, and the automotive industry, the forming quality of square tubes affects the accuracy of subsequent welding and assembly, as well as the stability and safety of the overall structure. In actual production, various factors such as strip thickness fluctuations, material inhomogeneity, rolling force variations, roll wear, and installation errors and transmission clearances of the rolling mill equipment can all lead to problems such as vertex offset, forming angle deviation, and cross-sectional diagonal deviation during the forming process of square tubes.

[0003] However, traditional control methods for square tube forming machines rely heavily on empirical parameter settings or simple offline detection and manual adjustment, making it difficult to capture dynamic changes in the forming process in real time. This not only results in slow response and low efficiency, but also makes it difficult to accurately identify the sources of complex coupling deviations, leading to poor correction effects. Summary of the Invention

[0004] This application provides an online intelligent correction control method for square tube forming units based on real-time detection feedback. This solves the technical problem that existing square tube forming unit control methods cannot capture the dynamic changes in the forming process in real time, resulting in delayed response, low efficiency, and difficulty in accurately identifying the source of complex coupled deviations, thus leading to poor correction effect.

[0005] The technical solution to the above-mentioned technical problems in this application is as follows: This application provides an online intelligent deviation correction control method for square tube forming units based on real-time detection feedback, the method comprising: Obtain the vertex offset, forming angle deviation, and cross-sectional diagonal deviation of the square tube; Based on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, a deviation analysis is performed to obtain the deviation analysis results; Based on the deviation analysis results, the degree of deviation of the current square tube is determined. When the degree of deviation is greater than the allowable deviation threshold, a coupled influence analysis is performed on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation to obtain the predicted source of deviation. Based on the predicted deviation sources and the degree of deviation, the current square tube control parameters are optimized online to obtain optimized square tube control parameters for controlling the square tube forming process.

[0006] This application provides one or more technical solutions, which have at least the following technical effects or advantages: This application provides an online intelligent deviation correction control method for square tube forming units based on real-time detection feedback. First, it achieves perception of the forming state by acquiring the vertex offset, forming angle deviation, and cross-sectional diagonal deviation during the square tube forming process in real time. Second, it performs deviation analysis based on multi-dimensional deviation data to quantify the severity of each deviation. Finally, based on the predicted deviation sources and degrees, it achieves online and precise optimization of the square tube control parameters through steps such as dynamically generating control parameter item weight reorganization, screening effective adjustment vectors, and optimizing and fusing them. This enables rapid and effective deviation correction control of the forming process.

[0007] Through the above technical solutions, this application forms a closed loop of "detection-analysis-tracing-optimization-control" based on real-time detection feedback, which effectively solves the problems of slow response and difficulty in handling complex coupling deviations in traditional methods, and improves the accuracy and efficiency of square tube forming. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a flowchart illustrating the online intelligent correction control method for square tube forming units based on real-time detection feedback provided in the embodiments of this application. Detailed Implementation

[0010] This application provides an online intelligent correction control method for square tube forming units based on real-time detection feedback. This method addresses the technical problem that existing square tube forming unit control methods cannot capture the dynamic changes in the forming process in real time, resulting in delayed response, low efficiency, and difficulty in accurately identifying the sources of complex coupled deviations, thus leading to poor correction effects.

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0012] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0013] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use this application. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that this application can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid unnecessarily obscuring the description of this application. Therefore, this application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0014] Examples, such as Figure 1 As shown in the embodiments of this application, an online intelligent correction control method for square tube forming units based on real-time detection feedback is provided, including: S10: Obtain the vertex offset, forming angle deviation, and cross-sectional diagonal deviation of the square tube; In this embodiment, a high-precision visual inspection system and a laser displacement sensor group are deployed at the workstation of the square tube forming unit. The visual inspection system uses an industrial camera with a high-resolution lens to capture real-time images of the cross-section of the formed square tube. The actual coordinates of the four vertices of the square tube are extracted through an image recognition algorithm and compared with the theoretical coordinates of the vertices in the square tube forming standard to calculate the offset of each vertex in the X and Y axes, i.e., the vertex offset.

[0015] Furthermore, laser displacement sensor arrays are installed on both sides and above and below the square tube to measure the actual length of each side of the square tube's cross-section and the included angle between adjacent sides. For the forming angle deviation, the difference is obtained by subtracting the measured included angle from the specified angle in the forming angle standard, which is typically 90 degrees.

[0016] The deviation of the cross-section diagonal is determined by measuring the actual lengths of the two diagonals of the square tube cross-section, calculating the difference, and then comparing it with the allowable range of difference in the diagonal standard.

[0017] Specifically, step S10 in the method includes: Obtain square tube forming standards, wherein the square tube forming standards include vertex standards, forming angle standards, and diagonal standards; Based on the deviation of the current square tube and square tube forming standards, the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation are obtained.

[0018] In this embodiment, firstly, square tube forming standards are pre-defined based on the design requirements, material properties, and application scenarios of the square tube. The vertex standard specifies the three-dimensional coordinates of the four vertices of the square tube cross-section under ideal conditions. For example, in a coordinate system with the center of the square tube cross-section as the origin, the standard vertex coordinates might be set as (a / 2, a / 2, 0), (-a / 2, a / 2, 0), (-a / 2, -a / 2, 0), (a / 2, -a / 2, 0), where a is the theoretical side length of the square tube. The forming angle standard clarifies the ideal included angle between adjacent sides of the square tube, generally 90 degrees, and also specifies the allowable angle error range, such as ±0.5 degrees. The diagonal standard determines the theoretical lengths of the two diagonals of the square tube cross-section and the allowable length difference. For example, for a square tube with a side length of a, the theoretical diagonal length is a... The permissible difference in diagonal length is usually no more than 0.3 mm.

[0019] Secondly, after obtaining the relevant data of the current square tube, it is compared and calculated one by one with the square tube forming standard to obtain the vertex offset, forming angle deviation and cross-sectional diagonal deviation.

[0020] For example, to calculate the vertex offset, firstly, the actual coordinates of the four vertices of the current square tube extracted by the visual inspection system (x1 real, y1 real, z1 real), (x2 real, y2 real, z2 real), (x3 real, y3 real, z3 real), and (x4 real, y4 real, z4 real) are compared with the theoretical coordinates in the vertex standard (x1 theoretical, y1 theoretical, z1 theoretical), (x2 theoretical, y2 theoretical, z2 theoretical), (x3 theoretical, y3 theoretical, z3 theoretical), and (x4 theoretical, y4 theoretical, z4 theoretical). The offset components of each vertex in the X, Y, and Z directions are calculated using the formulas Δxi = xi real - xi theoretical, Δyi = yi real - yi theoretical, and Δzi = zi real - zi theoretical, where i = 1, 2, 3, 4. Then, the offset is calculated using Δi = Calculate the spatial offset of each vertex, and finally take the maximum value of the four vertex offsets as the vertex offset to measure the degree of vertex offset of the square tube.

[0021] S20: Based on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, perform deviation analysis and obtain the deviation analysis results; In this embodiment of the application, deviation analysis is performed on the vertex offset, forming angle deviation, and cross-sectional diagonal deviation. That is, the deviations are quantitatively evaluated and comprehensively judged. Specifically, the cooperative deviation coefficients between the three are calculated, and the mutual influence relationship and cooperative change trend between the deviations are analyzed as the deviation analysis results to quantify the overall deviation degree of the current square tube.

[0022] Specifically, step S20 in the method includes: Calculate the cooperative deviation coefficient between the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation; Based on the aforementioned collaborative deviation coefficient, the mutual influence relationship and collaborative change trend among various deviations are analyzed to obtain deviation analysis results, wherein the deviation analysis results include deviation severity and deviation correlation.

[0023] In this embodiment, the three deviation values ​​are first defined as vectors in three-dimensional space, serving as deviation vectors. Then, a baseline collaborative pattern library is established, and common deviation collaborative patterns are obtained from historical normal or typical fault data through cluster analysis. Next, the similarity is calculated as the collaborative deviation coefficient, i.e., the cosine similarity between the current deviation vector and each pattern vector is calculated. The cosine similarity measures the consistency of direction using the cosine of the angle between the two vectors, with a value range of [-1, 1]. A larger value indicates a more similar collaborative pattern.

[0024] For example, the formula for calculating the cooperative deviation coefficient is: cosθ=(V1·V2) / (||V1||×||V2||), where V1 is the current deviation vector, V2 is a certain pattern vector in the baseline cooperative pattern library, "·" represents the vector dot product, and ||V1|| and ||V2|| represent the magnitudes of vectors V1 and V2, respectively.

[0025] Secondly, the calculated current deviation vector is compared with the cosine similarity of each mode vector, and the cooperative deviation coefficient corresponding to the mode with the highest similarity is selected as the final cooperative deviation coefficient. If the cooperative deviation coefficient is close to 1, it indicates that the cooperative mode of the current deviations is highly consistent with the baseline mode; if it is close to 0 or negative, it indicates that the current cooperative mode differs significantly from the baseline mode or shows an opposite trend.

[0026] Furthermore, based on this collaborative deviation coefficient, and combined with the absolute value of each deviation, the severity level of the deviation is determined. For example, the vertex offset, forming angle deviation, and cross-sectional diagonal deviation are compared with their respective allowable deviation thresholds. If a certain deviation exceeds its threshold, the deviation is determined to be severe. If multiple deviations exceed the threshold at the same time and the collaborative deviation coefficient is high, the overall deviation is determined to be of high severity.

[0027] Simultaneously, analyzing the correlation between various deviations reflects the degree of mutual influence between different deviations. For example, when the vertex offset increases, it may lead to changes in the forming angle deviation and the cross-sectional diagonal deviation. The degree of correlation can be quantified by the magnitude of the co-correlation deviation coefficient. If the co-correlation deviation coefficient is positive and large, it indicates that the various deviations are positively correlated, that is, an increase in one deviation will lead to an increase in other deviations in the same direction; if it is negative and the absolute value is large, it indicates a negative correlation, that is, an increase in one deviation will lead to an inverse change in other deviations; if it is close to 0, it indicates that the correlation between the various deviations is weak and may be caused by independent factors.

[0028] S30: Based on the deviation analysis results, determine the degree of deviation of the current square tube. When the degree of deviation is greater than the allowable deviation threshold, perform a coupled influence analysis on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation to obtain the predicted source of deviation. In this embodiment, the determination of the degree of deviation requires comprehensive consideration of both the severity level and the correlation of the deviation. Specifically, the absolute values ​​of each deviation in the deviation analysis results are compared with the corresponding allowable deviation thresholds. If at least one deviation exceeds its threshold, and the overall deviation severity level reflected by the synergistic deviation coefficient reaches the preset "significant deviation" standard, then the current degree of deviation is determined to be greater than the allowable deviation threshold. At this time, coupling effect analysis is initiated to trace and predict the source of the deviation.

[0029] Specifically, step S30 in the method includes: Based on the deviation analysis results, the degree of deviation is determined, wherein the degree of deviation includes the severity of the deviation and the degree of correlation of the deviation; When any of the deviations exceeds the allowable deviation threshold, a coupled influence analysis is performed on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation to obtain the predicted deviation source.

[0030] In this embodiment, the severity of the deviation is first determined by comparing the actual measured values ​​of each deviation with their respective allowable deviation thresholds. For example, the allowable threshold for vertex offset is set to ±0.2 mm, the allowable threshold for forming angle deviation is ±0.5 degrees, and the allowable threshold for cross-sectional diagonal deviation is ±0.3 mm. If the absolute value of a deviation exceeds its corresponding allowable threshold, the severity of that deviation is determined to be "high"; if it is within the threshold range but close to the threshold, such as reaching 80% or more of the threshold, it is determined to be "medium"; if it is far below the threshold, it is determined to be "low". The degree of correlation between deviations is mainly evaluated based on the cooperative deviation coefficient calculated in step S20. The larger the absolute value of the coefficient, the higher the degree of correlation between the various deviations, and vice versa.

[0031] Secondly, when any of the severity levels of the deviation reaches the "high" level, or when the absolute value of the co-deviation coefficient of the deviation correlation degree exceeds the preset correlation threshold, such as 0.7, the current deviation level is determined to be greater than the allowable deviation threshold, and a coupling effect analysis is performed. The coupling effect analysis reflects the inherent connection between the various deviations and the reasons for the occurrence of the deviation.

[0032] Specifically, a coupled influence analysis is performed on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation to obtain the sources of predicted deviations, including: Construct a sample square tube parameter set containing multiple historical parameters. Each parameter set contains a set of control parameters of the square tube forming machine and its corresponding comprehensive deviation index. The comprehensive deviation index includes historical vertex offset, historical forming angle deviation and historical cross-sectional diagonal deviation. The control parameters include multiple control parameter items. Based on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, a similar sample parameter set is obtained by searching the sample square tube parameter set. Based on the set of similar sample parameters, the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, a coupled influence analysis is performed to obtain the source of the prediction deviation.

[0033] In this embodiment, firstly, production data with the same or similar model, material specifications, and process conditions as the currently produced square tubes are selected from the historical production database of the square tube forming unit to construct a sample set of square tube parameters. Each sample parameter set includes various control parameters of the square tube forming unit at a specific production moment, such as the horizontal and vertical positions of the rolls, rolling speed, rolling force, strip inlet tension, outlet tension, raw strip thickness deviation, and equipment operating temperature, as well as comprehensive deviation indicators such as the historical vertex offset, historical forming angle deviation, and historical cross-sectional diagonal deviation detected in the corresponding produced square tubes. Through data cleaning and standardization, the accuracy and consistency of the data are ensured, for example, by unifying the dimensions of the control parameters and normalizing the deviation indicators.

[0034] Secondly, based on the current square tube's vertex offset, forming angle deviation, and cross-sectional diagonal deviation, a similarity search is performed within the sample square tube parameter set. The deviation Euclidean distance algorithm is used to calculate the similarity between the current deviation combination and each historical comprehensive deviation index in the sample parameter set. A similarity threshold is set, and sample parameter sets with similarities higher than this threshold are selected to form a similar sample parameter set. For example, if the current deviation combination is Δx=0.15mm, Δθ=0.3 degrees, and Δd=0.2mm, the 20 samples with the smallest deviation Euclidean distances to each historical sample, or samples with distances less than 0.1, are selected as the similar sample parameter set.

[0035] Finally, a coupling effect analysis was conducted on the parameter set of similar samples. On the one hand, the correlation between each control parameter item and the comprehensive deviation index in the parameter set of similar samples was analyzed. By calculating the Pearson correlation coefficient or Spearman's rank correlation coefficient, control parameters that significantly affect the deviation were identified. For example, the correlation coefficient between the horizontal position of the roll and the vertex offset was found to be 0.8, and the correlation coefficient between the rolling force and the forming angle deviation was 0.75, indicating a strong correlation between the control parameters and the deviation. On the other hand, combined with historical fault records and process knowledge, the range and trend of the values ​​of the significantly correlated control parameters in the parameter set of similar samples when deviations occur were analyzed. If, in similar samples, when the horizontal position of the roll exceeds the standard range of ±0.1 mm, the vertex offset and the diagonal deviation of the cross section both increase significantly, and historical records show that this situation is often caused by the loosening of the roll positioning mechanism, then the abnormal horizontal position of the roll can be considered as one of the sources of predicted deviation.

[0036] By combining the results of correlation analysis and historical experience, the main sources of predicted deviations that cause the current deviations in square tubes were finally identified, such as the horizontal position of the rolls, excessive fluctuations in strip tension, and uneven thickness of raw materials.

[0037] Specifically, based on the set of similar sample parameters and the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, a coupled influence analysis is performed to obtain the sources of prediction deviation, including: Calculate the mean value of each control parameter item in the similar sample parameter set, and calculate the deviation value from the mean value of each control parameter item in the sample square tube parameter group set to obtain the degree of deviation of the control parameter item; The control parameter items with the largest deviations are selected as the sources of prediction deviation.

[0038] In this embodiment, firstly, the mean value of each control parameter item in the similar sample parameter set is calculated. For example, for the horizontal position of the roll, the arithmetic mean of the horizontal position data of all samples in the similar sample parameter set is taken to obtain the mean value of the horizontal position of the roll. Similarly, the mean values ​​of all control parameter items in the similar sample parameter set are calculated respectively.

[0039] Then, from the sample square tube parameter set, the overall mean of each control parameter item is calculated, which is the average level of that control parameter item across all historical samples. Next, the difference between the mean of each control parameter item in the similar sample parameter set and the corresponding overall mean of the control parameter item in the sample square tube parameter set is calculated; this difference is used as the degree of deviation of the control parameter item.

[0040] For example, the mean of the horizontal position of the roll in similar samples is 5.2 mm, while the overall mean of the horizontal position of the roll in the sample square tube parameter group set is 5.0 mm, so the deviation of the horizontal position of the roll is 0.2 mm.

[0041] Finally, the deviation levels of all control parameters are compared, and the control parameter with the largest deviation is selected. This control parameter is identified as the most likely source of the current square tube deviation because its value differs most significantly from the average level during normal production under similar deviation conditions. For example, if the calculated deviation of strip tension is 15N, which is much greater than the deviation levels of other control parameters, then abnormal strip tension is identified as the primary source of the predicted deviation.

[0042] S40: Based on the predicted deviation source and the degree of deviation, optimize the current square tube control parameters online to obtain optimized square tube control parameters and control the square tube forming process.

[0043] In this embodiment, online optimization is performed based on the predicted source and degree of deviation. By adjusting the control parameters of the forming unit and correcting the initial adjustment direction and initial adjustment range, the various deviations of the square tube are controlled within the allowable range, ensuring the forming quality.

[0044] Specifically, step S40 in the method includes: Based on the sources and degrees of the predicted deviations, the control parameter item weighting is reorganized. Based on the control parameter project weight reorganization and the current square tube control parameters, multiple initial adjustment directions and initial adjustment ranges are obtained; Based on process constraint rules, the feasibility of multiple initial adjustment directions and multiple initial adjustment magnitudes is modified to obtain an effective adjustment vector; Based on the effective adjustment vector, the current square tube control parameters are optimized online to obtain optimized square tube control parameters for controlling the square tube forming process.

[0045] In this embodiment, firstly, based on the source and degree of the prediction deviation, each control parameter item is assigned a corresponding weight, thus constructing a weighted reorganization of the control parameter items. The magnitude of the weight reflects the degree of influence of the control parameter item on the current deviation and the priority of its adjustment.

[0046] Specifically, the control parameter with the greatest degree of deviation from the sources of prediction bias is assigned the highest weight, for example, 0.6; the control parameter with the second greatest degree of deviation is assigned the second highest weight, for example, 0.3; and other potentially relevant control parameters are assigned lower weights, for example, 0.1. The total weights are 1 to ensure the normalization of the weight groups.

[0047] Secondly, multiple initial adjustment directions and initial adjustment ranges are obtained. Based on the actual values ​​of the current square tube control parameters, combined with the weights of each parameter in the control parameter weighting, and the direction and magnitude of each deviation in the deviation analysis results, the adjustment direction and range are initially determined.

[0048] For example, if the predicted deviation originates from the horizontal position offset of the roll, with a weight of 0.6 and an adjustment coefficient set to 1.2 based on process experience, and the current vertex offset is 0.15mm in the positive direction, historical data shows that adjusting the horizontal position of the roll in the negative direction can reduce the vertex offset. Therefore, the initial adjustment direction is set to the negative direction of the roll's horizontal position, and the initial adjustment amplitude is calculated based on the degree of deviation and its weight, such as 0.15mm × 0.6 × 1.2 = 0.108mm. The strip tension fluctuation weight is 0.3. If the forming angle deviation is 0.3 degrees in the positive direction, and an increase in strip tension can cause the forming angle deviation to change in the negative direction, then the initial adjustment direction is the direction of increased strip tension, and the initial adjustment amplitude is 0.3 degrees × 0.3 × adjustment coefficient 1.0 = 0.09N.

[0049] Similarly, for each weighted control parameter item, a corresponding initial adjustment direction and initial adjustment magnitude are generated, forming multiple initial adjustment schemes.

[0050] Secondly, based on process constraints, multiple initial adjustment directions and magnitudes are modified for feasibility to ensure that the adjustment schemes meet actual production conditions and equipment safety operation requirements. Process constraints include equipment physical limits, allowable ranges of process parameters, and operational safety regulations. For example, the adjustment range of the roll's horizontal position is limited by its mechanical stroke, and the maximum adjustment cannot exceed ±5mm; the strip tension adjustment must be within the allowable range of the equipment motor's output power and cannot be lower than the minimum tension value required to ensure stable strip transmission, such as not lower than 200N; the rolling speed adjustment must consider the matching of preceding and following processes to avoid steel piling or pulling. Initial adjustment directions or magnitudes that exceed the process constraints are corrected.

[0051] Finally, the current square tube control parameters are optimized online. The effective adjustment vector, after feasibility correction, is superimposed onto the actual values ​​of the current square tube control parameters to obtain the optimized square tube control parameters. All optimized control parameters are then sent to the control system of the square tube forming unit to update the unit's operating parameters in real time, achieving online closed-loop control of the square tube forming process.

[0052] Among them, based on the effective adjustment vector, the control parameters of the square tube are optimized online, including: If the parameter optimization regions pointed to by multiple effective adjustment vectors tend to the same optimization region, the same optimization region is determined to be a potential optimization concentration region; When the potential optimization concentration area exists, multiple effective adjustment vectors pointing to the potential optimization concentration area are fused to generate a fused dominant adjustment vector, and the current square tube control parameters are updated and optimized using the dominant adjustment vector. When there is no potential optimization concentration area or the correction effect of the current square tube control parameters after updating and optimizing with the dominant adjustment vector is less than the correction amplitude threshold, multiple new adjustment vectors are generated based on the divergence of each effective adjustment vector, and the optimization trend evaluation and vector fusion are re-performed for iterative optimization.

[0053] In this embodiment, firstly, cluster analysis is performed on the parameter optimization regions pointed to by multiple effective adjustment vectors. For example, the K-means clustering algorithm is used to divide the adjustment vectors that are close to each other in the parameter space into the same cluster. The region with the most vectors in the cluster is determined as the potential optimization concentration region. If the clustering results show that more than 60% of the effective adjustment vectors fall into the same region, then the region is confirmed as the potential optimization concentration region, indicating that most adjustment directions point to this region and the optimization objectives have a high degree of consistency.

[0054] Secondly, when a potential optimization concentration zone exists, all valid adjustment vectors pointing to that zone are fused. The fusion method uses a weighted average, with the weights corresponding to each adjustment vector as coefficients. The adjustment amounts of each control parameter in the vector are weighted and summed to obtain the fused dominant adjustment vector. The weights are determined based on the degree of deviation and correlation. For example, among the three valid adjustment vectors pointing to the potential optimization concentration zone, the roll horizontal position adjustment amounts are -0.1mm, -0.12mm, and -0.09mm, with corresponding weights of 0.5, 0.3, and 0.2, respectively. Then, the fused roll horizontal position adjustment amount is (-0.1×0.5) + (-0.12×0.3) + (-0.09×0.2) = -0.104mm. This process is repeated to obtain all parameters of the dominant adjustment vector. This dominant adjustment vector is then directly applied to the current square tube control parameters to achieve rapid parameter optimization.

[0055] Furthermore, if there is no potential optimization concentration area, that is, the effective adjustment vectors are scattered, or after updating with the dominant adjustment vector, the real-time detection feedback shows that the correction amplitude of the square tube deviation has not reached the preset threshold, such as the deviation value only decreasing by 20%, which is lower than the correction amplitude threshold of 30%, then the iterative optimization stage is entered.

[0056] At this point, based on the divergence of the current effective adjustment vectors, if the divergence is high (e.g., greater than a preset divergence threshold), a new adjustment vector is generated by performing a mutation operation in the parameter space of the effective adjustment vectors, such as randomly perturbing the adjustment amplitude by ±10%. If the divergence is low, a new adjustment vector is generated by cross-combining the parameters of different effective adjustment vectors. After the new adjustment vector is generated, the optimization trend is re-evaluated, i.e., it is determined whether a new potential optimization concentration area has been formed and fused with the vector. The above process is repeated until the deviation correction amplitude reaches the threshold or the number of iterations reaches the upper limit, such as 5 iterations, to ensure that the final optimized square tube control parameters can effectively reduce the forming deviation.

[0057] In summary, compared with existing technologies, this application constructs a set of sample square tube parameter groups and performs similarity retrieval to locate the historical production data most similar to the current deviation. Then, through coupling influence analysis, it identifies the most likely source of prediction deviation and its degree from complex multi-parameter interactions, overcoming the limitations of traditional deviation correction methods that rely too heavily on operator experience and are unable to cope with the coupling influence of multiple factors.

[0058] In summary, the embodiments of this application have at least the following technical effects: This application provides an online intelligent deviation correction control method for square tube forming units based on real-time detection feedback. First, it achieves perception of the forming state by acquiring the vertex offset, forming angle deviation, and cross-sectional diagonal deviation during the square tube forming process in real time. Second, it performs deviation analysis based on multi-dimensional deviation data to quantify the severity of each deviation. Finally, based on the predicted deviation sources and degrees, it achieves online and precise optimization of the square tube control parameters through steps such as dynamically generating control parameter item weight reorganization, screening effective adjustment vectors, and optimizing and fusing them. This enables rapid and effective deviation correction control of the forming process.

[0059] Through the above technical solutions, this application forms a closed loop of "detection-analysis-tracing-optimization-control" based on real-time detection feedback, which effectively solves the problems of slow response and difficulty in handling complex coupling deviations in traditional methods, and improves the accuracy and efficiency of square tube forming.

[0060] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0061] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0062] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. An online intelligent deviation correction control method for square tube forming units based on real-time detection feedback, characterized in that, include: Obtain the vertex offset, forming angle deviation, and cross-sectional diagonal deviation of the square tube; Based on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, a deviation analysis is performed to obtain the deviation analysis results; Based on the deviation analysis results, the degree of deviation of the current square tube is determined. When the degree of deviation is greater than the allowable deviation threshold, a coupled influence analysis is performed on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation to obtain the predicted source of deviation. Based on the predicted deviation sources and the degree of deviation, the current square tube control parameters are optimized online to obtain optimized square tube control parameters for controlling the square tube forming process.

2. The online intelligent correction control method for square tube forming units based on real-time detection feedback as described in claim 1, characterized in that, Obtain the vertex offset, forming angle deviation, and cross-sectional diagonal deviation of the square tube, including: Obtain square tube forming standards, wherein the square tube forming standards include vertex standards, forming angle standards, and diagonal standards; Based on the deviation of the current square tube and square tube forming standards, the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation are obtained.

3. The online intelligent deviation correction control method for square tube forming units based on real-time detection feedback as described in claim 1, characterized in that, Based on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, a deviation analysis is performed to obtain the deviation analysis results, including: Calculate the cooperative deviation coefficient between the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation; Based on the aforementioned collaborative deviation coefficient, the mutual influence relationship and collaborative change trend among various deviations are analyzed to obtain deviation analysis results, wherein the deviation analysis results include deviation severity and deviation correlation.

4. The online intelligent correction control method for square tube forming units based on real-time detection feedback as described in claim 1, characterized in that, Based on the deviation analysis results, the degree of deviation of the square tube is determined. When the degree of deviation exceeds the allowable deviation threshold, a coupled influence analysis is performed on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation to obtain the predicted sources of deviation, including: Based on the deviation analysis results, the degree of deviation is determined, wherein the degree of deviation includes the severity of the deviation and the degree of correlation of the deviation; When any of the deviations exceeds the allowable deviation threshold, a coupled influence analysis is performed on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation to obtain the predicted deviation source.

5. The online intelligent correction control method for square tube forming units based on real-time detection feedback as described in claim 4, characterized in that, A coupled influence analysis is performed on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation to obtain the sources of predicted deviation, including: Construct a sample square tube parameter set containing multiple historical parameters. Each parameter set contains a set of control parameters of the square tube forming machine and its corresponding comprehensive deviation index. The comprehensive deviation index includes historical vertex offset, historical forming angle deviation and historical cross-sectional diagonal deviation. The control parameters include multiple control parameter items. Based on the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, a similar sample parameter set is obtained by searching the sample square tube parameter set. Based on the set of similar sample parameters, the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, a coupled influence analysis is performed to obtain the source of the prediction deviation.

6. The online intelligent correction control method for square tube forming units based on real-time detection feedback as described in claim 5, characterized in that, Based on the set of similar sample parameters and the coupled influence analysis of the vertex offset, the forming angle deviation, and the cross-sectional diagonal deviation, the sources of prediction deviation are obtained, including: Calculate the mean value of each control parameter item in the similar sample parameter set, and calculate the deviation value from the mean value of each control parameter item in the sample square tube parameter group set to obtain the degree of deviation of the control parameter item; The control parameter items with the largest deviations are selected as the sources of prediction deviation.

7. The online intelligent deviation correction control method for square tube forming units based on real-time detection feedback as described in claim 1, characterized in that, Based on the predicted deviation sources and the degree of deviation, the control parameters of the square tube are optimized online to obtain optimized square tube control parameters, and the square tube forming process is controlled, including: Based on the sources and degrees of the predicted deviations, the control parameter item weighting is reorganized. Based on the control parameter project weight reorganization and the current square tube control parameters, multiple initial adjustment directions and initial adjustment ranges are obtained; Based on process constraint rules, the feasibility of multiple initial adjustment directions and multiple initial adjustment magnitudes is modified to obtain an effective adjustment vector; Based on the effective adjustment vector, the current square tube control parameters are optimized online to obtain optimized square tube control parameters for controlling the square tube forming process.

8. The online intelligent correction control method for square tube forming units based on real-time detection feedback as described in claim 7, characterized in that, Based on the effective adjustment vector, the square tube control parameters are optimized online, including: If the parameter optimization regions pointed to by multiple effective adjustment vectors tend to the same optimization region, the same optimization region is determined to be a potential optimization concentration region; When the potential optimization concentration area exists, multiple effective adjustment vectors pointing to the potential optimization concentration area are fused to generate a fused dominant adjustment vector, and the current square tube control parameters are updated and optimized using the dominant adjustment vector. When there is no potential optimization concentration area or the correction effect of the current square tube control parameters after updating and optimizing with the dominant adjustment vector is less than the correction amplitude threshold, multiple new adjustment vectors are generated based on the divergence of each effective adjustment vector, and the optimization trend evaluation and vector fusion are re-performed for iterative optimization.

Images