Anti-rebound safety control method and platform based on pipe bending machine

Abstract

The invention provides an anti-springback safety control method and platform based on a pipe bending machine, and relates to the technical field of metal machining, and the method comprises the steps that pipe specification information of a target pipe and pipe bending machining parameters are collected; pipe heating pretreatment parameters are optimized, and optimized pretreatment parameters are obtained; elbow springback prediction is carried out, elbow springback amount parameters are obtained, and turning processing parameters are generated; carrying out pipe bending treatment, collecting a pipe processing image sequence, carrying out clamp stability analysis to obtain clamp stability parameters, and carrying out springback risk analysis to obtain springback risk parameters; and performing safe stop control judgment to obtain a judgment result, and performing control. The technical problem that in the prior art, due to the fact that the rebound characteristic of the pipe is difficult to accurately predict, the rebound danger degree of pipe bending treatment is high can be solved, the technical target of accurate prediction of bend rebound is achieved, and the technical effect of improving the pipe machining quality is achieved.

Description

Technical Field

[0001] The present application relates to the field of metal processing technology, and in particular to an anti-rebound safety control method and platform based on a pipe bending machine. Background Art

[0002] In modern manufacturing, pipe bending technology is widely used in aerospace, automobile manufacturing, construction engineering and other fields. Pipe bending technology is used to perform high-precision bending operations on metal pipes to meet specific design requirements.

[0003] Currently, existing pipe bending technologies have limited effectiveness in dealing with the springback problem. Due to the elastic properties of metal materials, after bending, the pipe often rebounds to a certain extent, resulting in the bending angle and shape not meeting the design requirements. Traditional pipe bending machines usually adjust the bending angle through empirical values ​​to compensate for the springback phenomenon, but this method has large uncertainties and errors. Once the springback cannot be accurately predicted, the accuracy and quality of the final product will be seriously affected, and may even result in unqualified products. In addition, the springback characteristics of pipes of different materials and specifications vary, and traditional methods are difficult to accurately predict and adjust, further limiting production efficiency and product consistency.

[0004] In summary, the prior art has a technical problem that the rebound characteristics of the pipe are difficult to accurately predict, resulting in a high degree of rebound risk in the pipe bending process, which further affects the quality of the pipe. Summary of the invention

[0005] The purpose of this application is to provide an anti-rebound safety control method and platform based on a pipe bending machine, so as to solve the technical problem in the prior art that the rebound characteristics of the pipe are difficult to accurately predict, resulting in a high degree of rebound risk in the pipe bending process, which further affects the quality of the pipe.

[0006] In view of the above problems, the present application provides an anti-rebound safety control method and platform based on a pipe bending machine.

[0007] In the first aspect, the present application provides an anti-rebound safety control method based on a pipe bending machine, which is implemented through an anti-rebound safety control platform based on a pipe bending machine, including: collecting pipe specification information of a target pipe to be bent by a pipe bending machine, and pipe bending processing parameters of the target pipe to be bent; optimizing pipe heating pretreatment parameters according to the pipe specification information and pipe bending processing parameters to obtain optimized pretreatment parameters, wherein the optimization is performed by minimizing the change in surface performance of the pipe after pretreatment and minimizing the springback of the pipe after pretreatment; based on The optimized preprocessing parameters are combined with the pipe specification information and the pipe bending processing parameters to predict the springback of the pipe bending, obtain the pipe bending springback amount parameters, and generate the bending processing parameters; the bending processing parameters are used to bend the target pipe, and the pipe processing image sequence is collected during the bending process to perform a fixture stability analysis to obtain the fixture stability parameters, and the springback risk analysis is performed in combination with the pipe bending springback amount parameters to obtain the springback risk parameters; according to the springback risk parameters, a safety stop control judgment is performed to obtain a judgment result and control is performed.

[0008] In the second aspect, the present application also provides an anti-rebound safety control platform based on a pipe bending machine, which is used to execute the anti-rebound safety control method based on a pipe bending machine as described in the first aspect, including: a pipe bending processing parameter acquisition module, the pipe bending processing parameter acquisition module is used to collect the pipe specification information of the target pipe to be bent by the pipe bending machine, and the pipe bending processing parameters of the target pipe to be bent; an optimized pretreatment parameter acquisition module, the optimized pretreatment parameter acquisition module is used to optimize the pipe heating pretreatment parameters according to the pipe specification information and the pipe bending processing parameters, and obtain the optimized pretreatment parameters, wherein the optimization is performed by minimizing the change in surface performance of the pipe after pretreatment and minimizing the springback amount of the pipe after pretreatment; the pipe bending springback amount parameter acquisition module Module, the bending springback parameter acquisition module is used to predict the bending springback based on the optimized preprocessing parameters combined with the pipe specification information and the bending processing parameters, obtain the bending springback parameters, and generate the bending processing parameters; the rebound risk parameter acquisition module, the rebound risk parameter acquisition module is used to use the bending processing parameters to bend the target pipe, and collect the pipe processing image sequence during the bending process, perform fixture stability analysis, obtain fixture stability parameters, perform rebound risk analysis in combination with the bending springback parameters, and obtain rebound risk parameters; the discrimination result acquisition module, the discrimination result acquisition module is used to perform safety stop control discrimination according to the rebound risk parameters, obtain discrimination results, and perform control.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages: The method collects the pipe specification information of the target pipe to be bent by the pipe bending machine and the pipe bending processing parameters of the target pipe to be bent; optimizes the pipe heating pretreatment parameters according to the pipe specification information and the pipe bending processing parameters to obtain the optimized pretreatment parameters, wherein the optimization is performed by minimizing the change in the surface performance of the pipe after pretreatment and minimizing the springback amount of the pipe after pretreatment; predicts the springback of the pipe based on the optimized pretreatment parameters combined with the pipe specification information and the pipe bending processing parameters to obtain the springback amount parameters and generate the bending processing parameters; uses the bending processing parameters to bend the target pipe, collects the pipe processing image sequence during the bending process, performs the fixture stability analysis to obtain the fixture stability parameters, and performs the springback risk analysis combined with the pipe springback amount parameters to obtain the springback risk parameters; performs the safety stop control judgment according to the springback risk parameters to obtain the judgment result, and performs the control to achieve the technical goal of accurately predicting the springback of the pipe, and achieves the technical effect of improving the pipe processing quality and reducing safety hazards.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 This is a flow chart of the anti-rebound safety control method based on a pipe bending machine in this application; Figure 2 This is a schematic diagram of the structure of the anti-rebound safety control platform based on the pipe bending machine in this application.

[0013] Description of reference numerals: The module includes a pipe bending processing parameter acquisition module 11, an optimization preprocessing parameter acquisition module 12, a pipe bending rebound parameter acquisition module 13, a rebound risk parameter acquisition module 14, and a judgment result acquisition module 15. DETAILED DESCRIPTION

[0014] This application provides a rebound prevention safety control method and platform based on a pipe bending machine, which solves the technical problem in the prior art that the rebound characteristics of pipes are difficult to accurately predict, resulting in a high rebound risk in pipe bending, which further affects the quality of pipes. The technical goal of accurately predicting the rebound of pipe bends is achieved, achieving the technical effect of improving pipe processing quality and reducing safety hazards.

[0015] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0016] For example, please refer to the attached Figure 1 The present application provides an anti-rebound safety control method based on a pipe bending machine, which is applied to an anti-rebound safety control platform based on a pipe bending machine, and specifically includes the following steps: Step 1: Collect the pipe specification information of the target pipe to be bent by the pipe bending machine, and the bending processing parameters of the target pipe to be bent.

[0017] Specifically, due to the elasticity of the pipe, the pipe may rebound during bending and cause harm to the operator. Therefore, by collecting the specification information of the target pipe to be bent by the pipe bender, and then obtaining the specific bending processing parameters of the target pipe to be bent, the potential rebound during the pipe bending process can be accurately controlled. For example, the bending processing parameters include the bending angle, etc. Then, the operation of the pipe bender is adjusted using the pipe specification information and the bending processing parameters to ensure that the risk of rebound is minimized during the bending process, thereby improving the safety and accuracy of the processing.

[0018] Step 2: According to the pipe specification information and the pipe bending processing parameters, the pipe heating pretreatment parameters are optimized to obtain the optimized pretreatment parameters, wherein the optimization is performed to minimize the change in surface performance of the pipe after pretreatment and minimize the springback of the pipe after pretreatment.

[0019] Specifically, according to the pipe specification information and pipe bending processing parameters, the processing quality is improved by optimizing the heating pretreatment parameters of the pipe. In the optimization process, heating pretreatment of the pipe can reduce springback, but excessive heating may cause performance changes. With the goal of minimizing the change in surface performance of the pipe after pretreatment and minimizing the springback of the pipe, the temperature and time of the heating pretreatment are optimized. That is, when determining the optimal heating pretreatment parameters, priority is given to reducing the change in surface quality and reducing the springback effect after pipe bending. That is, according to the pipe specification information and pipe bending processing parameters, the surface performance change and pipe springback after heating pretreatment are predicted, so as to ensure that the processed pipe is closer to the design requirements in shape and surface characteristics, and ultimately improve the consistency and quality of the product. For example, the change in surface performance of the pipe after pretreatment and the minimization of the springback of the pipe after pretreatment are negatively correlated with the fitness, that is, the smaller the change in surface performance of the pipe after pretreatment and the minimization of the springback of the pipe after pretreatment, the greater the fitness, and vice versa. Among them, the optimization of the pipe heating pretreatment parameters can be obtained according to the parameter range setting of the pipe normalizing and other pretreatments.

[0020] Step 3: Based on the optimized preprocessing parameters combined with the pipe specification information and the pipe bending processing parameters, the pipe bending springback prediction is performed to obtain the pipe bending springback amount parameters and generate the bending processing parameters.

[0021] Specifically, overbending refers to bending the tube slightly beyond the target angle during the bending process, and offsetting the impact of springback by "overbending". For example, if the target angle is 90 degrees, the actual bend may reach 92 or 93 degrees to offset the springback. Based on the optimized pre-processing parameters, combined with the tube specification information and the tube bending processing parameters, the tube bending springback prediction is performed to predict the springback parameters after the tube is bent. Subsequently, based on the predicted springback parameters, the corresponding overbending processing parameters are generated to reduce the impact of springback on the quality of the final product, thereby improving processing accuracy and product consistency.

[0022] Step 4: Using the bending processing parameters, bend the target pipe, collect the pipe processing image sequence during the bending process, perform fixture stability analysis, obtain fixture stability parameters, and perform rebound risk analysis in combination with the bending springback parameters to obtain rebound risk parameters.

[0023] Specifically, since springback is impossible to avoid completely, when bending the target pipe, the determined bending processing parameters are used, and the image sequence of the pipe is collected during the processing to perform fixture stability analysis, thereby obtaining the fixture stability parameters, which are then used to reduce the risk of springback. Subsequently, the fixture stability parameters are combined with the previously obtained bending springback parameters to perform springback risk analysis, evaluate the springback risk of the pipe during the bending process, and finally obtain the springback risk parameters, which helps to fully understand the potential risks in the bending process and improve processing accuracy and safety. Among them, the worse the fixture stability parameters and the larger the bending springback parameters, the larger the springback risk parameters, the greater the risk, and vice versa, the smaller the risk.

[0024] Step 5: According to the rebound risk parameter, perform safety stop control judgment, obtain the judgment result, and perform control.

[0025] Specifically, by analyzing the rebound hazard parameters, the safety of the current operation is evaluated to determine whether there are potential risks. Then, a judgment is made based on the evaluation results to decide whether the operation needs to be stopped immediately to prevent accidents. Finally, corresponding control measures are implemented based on the judgment results to ensure the safety and stability of the processing process, thereby effectively preventing possible operational risks and improving the safety of overall production. For example, it is determined whether it is greater than a threshold. If so, the bending is stopped, and the bending machine is controlled to slowly return to a state where no bending is performed to avoid rebound, thereby preventing rebound from injuring the operator, and adjusting the fixture and other adjustments.

[0026] The anti-rebound safety control method based on a pipe bending machine is applied to an anti-rebound safety control platform based on a pipe bending machine, which can achieve the technical goal of accurately predicting the springback of bent pipes, and achieve the technical effects of improving pipe processing quality and reducing safety hazards.

[0027] Furthermore, the present application also includes: The material information and size information of the target pipe to be bent by the pipe bending machine are collected as pipe specification information; and the standard bending specification parameters of the target pipe bending processing are collected as bending processing parameters.

[0028] Specifically, the material information and size information of the target pipe to be bent by the pipe bending machine are collected as pipe specification information. For example, the material information includes the material strength of the pipe, and the size information includes the diameter, wall thickness, etc. This provides reliable basic data for subsequent pipe bending operations and ensures that the various parameter settings during the pipe bending process are more scientific and reasonable.

[0029] Next, the standard bending specification parameters of the target pipe bending processing are collected as the bending processing parameters, such as the bending angle, radius and length of the bending part, etc., to obtain the final shape and size of the bent pipe, so as to facilitate the precise control of the pipe bending machine during the processing process to ensure that the final product meets the design requirements.

[0030] By ensuring the accuracy and stability of pipe bending, the processing quality is improved and the risks in processing are reduced.

[0031] Furthermore, the present application also includes: Obtain a pretreatment parameter space for the heating pretreatment of the pipe; randomly select a first pretreatment parameter in the pretreatment parameter space, combine the pipe specification information and the pipe bending processing parameters, perform pretreatment performance prediction, and obtain a first surface performance change and a first pipe springback; based on the first surface performance change and the first pipe springback, calculate and obtain a first pretreatment fitness, as shown in the following formula: ; Among them, PBF is the pre-processing fitness, and is the weight, is the first surface property change, To preset the surface performance change, is the springback of the pipe, The springback amount of the preset pipe is set; the pipe heating pretreatment parameters are continuously optimized in the pretreatment parameter space until the convergence requirements are met, and the pretreatment parameters with the maximum pretreatment fitness are output to obtain the optimized pretreatment parameters.

[0032] Specifically, the pretreatment parameter space for heating pretreatment of the pipe is obtained to determine how to heat the pipe before the pipe bending process, including identifying and collecting various factors and conditions that affect the heating pretreatment, such as the heating temperature, time, heating speed, etc. that determine the mechanical properties and surface quality of the pipe in subsequent processing, comprehensively covering possible pretreatment conditions, and providing a basic framework for further optimization. Exemplarily, the pretreatment parameter space includes temperature parameters for heating pretreatment, such as 200°C-600°C.

[0033] After obtaining the pretreatment parameter space, a first pretreatment parameter is randomly selected from it, and the pretreatment performance is predicted in combination with the specification information of the pipe and the bending processing parameters. The performance change of the pipe surface and the rebound amount after bending under different pretreatment conditions are simulated and evaluated, which can help understand the effectiveness of the pretreatment, especially the impact on the quality of the bending, thereby providing data support for the next step of the optimization process.

[0034] Next, based on the predicted performance change of the pipe surface and the rebound after bending, that is, the first surface performance change and the first pipe rebound, the first pretreatment fitness is calculated to evaluate the pros and cons of each pretreatment parameter, providing a clear evaluation standard for further optimization.

[0035] Among them, the pretreatment fitness PBF is used to quantify the effect of the first pretreatment parameter in the heating pretreatment, indicating the influence of the first pretreatment parameter on the quality and accuracy during the entire pipe bending process. The higher the pretreatment fitness PBF, the better the optimization processing effect, and vice versa.

[0036] Represents the importance of surface property changes in calculating pretreatment fitness, Represents the importance of pipe springback in calculating pretreatment fitness. and The setting of is adjusted by those skilled in the art according to the specific application scenario and actual processing requirements, that is, the most appropriate value is determined according to experiments and simulations, or and To make a decision based on importance, if the stability of the surface performance variation is more important, then the corresponding weight will be larger, and vice versa. If the stability of the surface performance variation is extremely important, set =0.7, and = 0.3, or =0.6, and = 0.4. If both are equally important, set =0.5, and =0.5. If the control of pipe springback is more important, set =0.4, and = 0.6, or =0.3, and = 0.7. The adjustment of weights makes the calculation of PBF more flexible and can be optimized according to different processing objectives.

[0037] First surface property change It refers to the change in the surface of the pipe during the pretreatment process after the first pretreatment parameter is selected, reflecting the influence of the surface quality of the pipe before and after the bending. Smaller surface performance change usually means better surface quality maintenance, and vice versa.

[0038] Preset surface property variation It refers to the ideal or allowable range of surface performance changes set before pretreatment, such as the surface hardness change. The smaller the better, to avoid pretreatment affecting the surface performance of the pipe. Preset surface performance change The parameters are obtained based on historical data or processing standard settings, or can be customized by technicians in this field according to actual conditions, as a benchmark for measuring the actual surface performance change. By comparing the actual change with the preset value, it can be evaluated whether the current pretreatment parameters have reached the expected quality standard.

[0039] Pipe springback It refers to the elastic recovery degree of the pipe after bending. The smaller it is, the closer the bent shape is to the expected target shape, otherwise, the greater the difference.

[0040] Preset pipe springback It refers to the allowable springback range of the pre-treated pipe during bending under ideal conditions. Preset pipe springback It is set according to processing requirements and standards, or can be customized by technicians in the field according to actual conditions to evaluate the actual springback effect. Assuming that according to processing requirements and industry standards, the preset springback of the pipe is set to ±1.5mm, it means that under ideal conditions, the springback of the pre-treated pipe after bending is controlled within the range of +1.5mm to -1.5mm. By comparing the actual springback with the preset value, it can be determined whether the current pre-processing parameters can effectively control the springback, thereby ensuring the processing accuracy.

[0041] For example, setting =0.5, and =0.5, the preset tube rebound amount is 1.5mm, the first tube rebound amount is 3mm, the first surface performance change is the hardness change of the surface after the tube pretreatment, the smaller the better, to avoid affecting the surface performance of the tube, for example, 20HRC, the preset surface performance change amount is 10HRC, then substitute the above formula to calculate the pretreatment fitness to be -1. Further, for example, when the pretreatment parameters are adjusted so that the first tube rebound amount is reduced to 1.5mm and the first surface performance change amount is reduced to 10HRC, the surface performance change amount of the tube under the pretreatment parameters is reduced, and the tube rebound amount is reduced. Due to the first processing parameters, the calculated pretreatment fitness of the pretreatment parameters is 0, which is greater than -1, which conforms to this rule.

[0042] According to the results of pretreatment fitness, the pipe heating pretreatment parameters are optimized in the pretreatment parameter space until the convergence requirements are met. Finally, the pretreatment parameters with the largest pretreatment fitness are output and the best pretreatment parameters are found to make the pipe achieve the best surface performance and the smallest springback after bending, thereby improving the processing quality and product reliability.

[0043] By obtaining the pretreatment parameter space, performing performance prediction and fitness calculation, parameter optimization is finally achieved, ensuring the best pretreatment effect of the pipe in the pipe bending process, which helps to achieve high-quality processing results and more stable production control.

[0044] Furthermore, the present application also includes: According to the data records of heating pretreatment in historical time, a sample pretreatment parameter set, a sample pipe specification information set, and a sample pipe bending processing parameter set are collected, and according to the pipe test data under different sample pretreatment parameters, sample pipe specification information, and sample pipe bending processing parameters, a sample surface performance change amount set and a sample pipe rebound amount set are annotated and obtained; the sample pretreatment parameter set, sample pipe specification information set, and sample pipe bending processing parameter set are used as inputs, and the sample surface performance change amount set and the sample pipe rebound amount set are used as outputs to train a pretreatment performance predictor; the pretreatment performance predictor is used to predict the pretreatment performance of the first pretreatment parameters, pipe specification information, and pipe bending processing parameters to obtain a first surface performance change amount and a first pipe rebound amount.

[0045] Specifically, in the data of heating pretreatment recorded in historical time, a set of sample pretreatment parameters, a set of sample pipe specification information, and a set of sample bending processing parameters are collected, and according to the pipe test data under different sample pretreatment parameters, sample pipe specification information, and sample bending processing parameters, a set of sample surface performance change quantities and a set of sample pipe rebound quantities are marked and obtained.

[0046] Next, by extracting the pipe test data of multiple sample sets, the performance changes and rebound amounts of the pipe surface are recorded and annotated with the corresponding sample parameters, thereby forming a sample surface performance change set and a sample pipe rebound amount set, providing a rich data foundation for subsequent prediction model training.

[0047] Then, a pretreatment performance predictor is trained by using a set of sample pretreatment parameters, a set of sample pipe specification information, and a set of sample pipe bending processing parameters as inputs, and a set of sample surface performance variation and a set of sample pipe springback as outputs.

[0048] Using the collected sample data set, a pre-processing performance predictor is trained through a machine learning algorithm. The pre-processing performance predictor is implemented by constructing a data analysis network layer that can be continuously iterated and optimized through a BP neural network. The data analysis network layer uses the training data set for learning through supervised training in machine learning. According to the preset data division rules, the input sample pre-processing parameter set, sample pipe specification information set, and sample pipe bending processing parameter set are divided into 85% sample training set and 15% sample verification set. This ratio can be customized by technicians in this field according to actual conditions. The data analysis network layer is supervised and trained through the sample training set until the output result of the data analysis network layer tends to converge. Then, the sample verification set is used to verify the accuracy of the output result of the network layer to obtain a preset verification accuracy index, such as 95%. The verification accuracy index can be customized by technicians according to actual conditions. When the accuracy of the output result of the data analysis network layer is greater than or equal to the preset verification accuracy index, the final data analysis network layer is obtained and added to the pre-processing performance predictor. The pretreatment performance predictor can predict the corresponding surface performance change and pipe springback according to different pretreatment parameters, pipe specifications and pipe bending processing parameters by learning the relationship between input and output. The trained predictor provides the necessary tools for the prediction of subsequent steps and can perform effective performance prediction in practical applications.

[0049] Next, the trained pretreatment performance predictor is applied to the actual situation. By inputting the first pretreatment parameters, pipe specification information and pipe bending processing parameters, the pretreatment performance predictor will output the corresponding first surface performance change and the first pipe springback. The prediction results can help evaluate the pretreatment effect, thereby guiding the parameter adjustment of the pipe bending process to ensure that the final product meets the quality standards.

[0050] By training the pretreatment performance predictor, we finally achieved accurate prediction of the actual pretreatment effect, which not only improved the understanding and control of pretreatment parameters, but also provided a scientific basis for optimizing the bending process and ensuring the stability and consistency of the processing quality.

[0051] Furthermore, the present application also includes: A sample preprocessing parameter set, a sample pipe specification information set and a sample pipe bending processing parameter set are collected, and according to the springback amount of the sample pipe after bending processing, a sample pipe bending springback parameter set is marked and obtained; the sample preprocessing parameter set, the sample pipe specification information set, the sample pipe bending processing parameter set and the sample pipe bending springback parameter set are used to train a pipe bending springback predictor, and the pipe bending springback prediction is performed on the optimized preprocessing parameters, the pipe specification information and the pipe bending processing parameters to obtain the pipe bending springback parameters; according to the pipe bending springback parameters, the pipe bending processing parameters are classified and obtained, wherein the pipe bending springback parameters are input into a bending processing mapping table of the same type of pipes for mapping and classification, and the bending processing mapping table is constructed based on mapping of multiple sample pipe bending springback parameters and multiple sample bending processing parameters; the bending processing parameters and pipe bending processing parameters are superimposed to generate bending processing parameters.

[0052] Specifically, a sample preprocessing parameter set, a sample pipe specification information set and a sample pipe bending processing parameter set are collected, and according to the springback amount of the sample pipe after bending, the springback amount of the sample pipe after bending is associated with the corresponding sample preprocessing parameters, sample pipe specification information and sample pipe bending processing parameters to obtain the sample pipe bending springback parameter set.

[0053] Next, using the labeled sample bending springback parameter set, the bending springback predictor is trained through the machine learning algorithm. It can predict the corresponding bending springback parameters based on the input optimization pretreatment parameters, pipe specification information and bending processing parameters, which helps to adjust the processing parameters in advance. Among them, using the labeled sample bending springback parameter set, the neural network algorithm is used to train the bending springback predictor. The bending springback predictor can accurately predict the corresponding bending springback parameters based on the input optimization pretreatment parameters, pipe specification information and bending processing parameters, which helps to adjust the processing parameters in advance to optimize the production process. In the specific training process, a data analysis network layer is constructed. The data analysis network layer refers to the neural network model that is continuously iteratively optimized in machine learning. The data analysis network layer is obtained on the training data set through supervised training. According to the preset data division rules, the sample bending springback parameter set is divided into 90% sample training set and 10% sample verification set. This ratio can be customized by technicians according to actual needs. Subsequently, the data analysis network layer is supervised and trained using the sample training set. When the output result of the data analysis network layer tends to converge, the output result accuracy of the data analysis network layer is verified through the sample verification set to obtain a preset verification accuracy index, such as 90%. The verification accuracy index can be customized by the technician according to the actual situation. When the output result accuracy of the data analysis network layer is greater than or equal to the preset verification accuracy index, the final data analysis network layer is obtained and added to the pipe bending springback predictor.

[0054] Then, the predicted bending springback parameters are used to query and match the pre-established bending processing mapping table, that is, to classify the appropriate bending processing parameters according to the bending springback parameter mapping. The bending processing mapping table is constructed based on multiple sample bending springback parameters and multiple sample bending processing parameter mappings, and can classify the bending springback parameters according to the corresponding bending processing parameters, so as to eliminate the generated bending springback parameters through bending processing, improve the accuracy of bending processing, and make the pipe after bending meet the standard bending specification parameters under the ideal processing standard, thereby providing a basis for subsequent actual production.

[0055] Finally, based on the obtained bending processing parameters, the original bending processing parameters are superimposed and combined to finally generate the bending processing parameters, which are used in the actual bending production process to ensure that the processing results meet the design requirements, reduce the errors caused by springback, and improve the processing accuracy and quality.

[0056] By generating bending process parameters for production, springback control during tube bending is optimized, thereby improving the quality and consistency of the final product.

[0057] Furthermore, the present application also includes: The target pipe is bent by adopting the bending processing parameters, and a pipe processing image sequence is collected during the bending process, wherein each pipe processing image includes an image of a pipe fixing fixture; a set of sample pipe processing image sequences is collected, and the loose amplitude of the fixture in each sample pipe processing image sequence is marked to obtain a set of sample fixture stability parameters, wherein the size of the sample fixture stability parameter is negatively correlated with the size of the loose amplitude of the fixture in the sample pipe processing image sequence; a fixture stability identifier is trained based on a convolutional neural network by adopting the sample pipe processing image sequence set and the sample fixture stability parameter set; and the fixture stability identifier is used to perform fixture stability analysis on the pipe processing image sequence to obtain a fixture stability parameter.

[0058] Specifically, the actual bending operation is performed on the target tube through the adjusted bending processing parameters. During this process, the image sequence in the process is continuously collected, with special attention paid to the state of the tube fixing fixture contained in each tube processing image, which provides key visual data for the subsequent fixture stability analysis and helps to determine whether the fixture remains stable during the processing process.

[0059] Next, we collected a set of sample pipe processing image sequences, analyzed the looseness of the fixtures in each sample pipe processing image sequence, and annotated the samples according to the looseness amplitude. The annotated results formed a set of sample fixture stability parameters, where the fixture stability parameter is inversely proportional to the looseness amplitude. For example, the smaller the looseness amplitude, the higher the stability parameter, and vice versa. This laid the foundation for the automatic identification and analysis of fixture stability, ensuring the accuracy and relevance of the data.

[0060] Then, the sample pipe processing image sequence set and the sample fixture stability parameter set are used to train a fixture stability identifier based on a convolutional neural network.

[0061] The convolutional neural network is trained using the labeled sample pipe processing image sequence set and the corresponding sample fixture stability parameter set. Convolutional neural networks are good at processing image data, so they can learn the subtle features in the image. The convolutional neural network associates the loose amplitude characteristics with the stability of the fixture, and generates a fixture stability identifier that can identify the stability state of the fixture in the image, providing an automated tool for subsequent stability analysis.

[0062] Finally, the fixture stability identifier is used to perform fixture stability analysis on the pipe processing image sequence to obtain fixture stability parameters.

[0063] The trained fixture stability identifier is used in the actual collected pipe processing image sequence. By analyzing the state of the fixture in the pipe processing image sequence, the corresponding fixture stability parameters are generated. For example, the stability parameters are divided into different levels as preset stability parameter standards, such as high stability (stability parameter range 1-0.8), medium stability (stability parameter range 0.8-0.6) and low stability (stability parameter range 0.6-0), ensuring the accuracy and relevance of the data. Specifically, the sample pipe processing image sequence set is preprocessed to enhance the characteristics of the fixture in the image. Secondly, the image recognition technology is used to detect the position and state of the fixture in the image. Then, the looseness of the fixture is calculated and marked according to the preset stability parameter standard, which reflects the stability of the fixture in the entire pipe bending process and helps to determine whether the processing parameters need to be further adjusted to ensure the stability of the fixture and the accuracy of the processing.

[0064] It provides the necessary support by ensuring the stability of the fixture during the tube bending process, thereby improving the processing quality and product consistency.

[0065] Furthermore, the present application also includes: According to the elbow rebound record data, a sample elbow rebound parameter set and a sample rebound risk parameter set are collected, and according to the rebound risk degree under different sample elbow rebound parameters and sample rebound risk parameters, the sample rebound risk parameter set is labeled and obtained; based on the decision tree, the sample elbow rebound parameter set and the sample rebound risk parameter set are used as classification inputs, and the sample rebound risk parameter set is used as classification output to construct a rebound risk classifier; the rebound risk classifier is used to classify the fixture stability parameters and elbow rebound parameters to obtain the rebound risk parameters.

[0066] Specifically, by analyzing the sample bend springback parameter set and the sample springback risk parameter set, the springback risk degree of the sample bend springback parameter and the sample springback risk parameter are collected and sorted out from the bend springback record data. Then, according to the performance of the springback risk degree in actual operation, the sample springback risk parameter set corresponding to the springback risk degree is marked to provide accurate reference data for subsequent classification and analysis. For example, a sample springback risk parameter set corresponding to the springback risk degree is formed, including dividing the springback risk degree into different levels, such as low risk (risk degree 1-3), medium risk (risk degree 4-6) and high risk (risk degree 7-10). Specifically, the sample bend springback parameters are statistically analyzed to determine the distribution and trend of the springback amount. Secondly, the springback risk degree of each sample is evaluated in combination with the sample springback risk parameters. Then, the springback parameters of each sample bend are marked according to the predetermined risk level.

[0067] Then, using the decision tree algorithm, by inputting the sample bending springback parameter set and the sample springback hazard parameter set, classification training is performed, and finally a springback hazard classifier is constructed, which can effectively classify and predict the springback hazard according to the input parameters, thereby providing a reliable tool for risk assessment in actual production. Specifically, the training set and the validation set are obtained by using the sample bending springback parameter set and the sample springback hazard parameter set. The decision tree is constructed through the training set, and the optimal features and segmentation points are found through recursive segmentation. Then, the decision tree is pruned to avoid overfitting. Finally, the validation set is used to evaluate the performance of the decision tree and make necessary adjustments.

[0068] Next, after completing the decision tree training, the springback hazard classifier constructed earlier is used to classify the fixture stability parameters and pipe bending springback parameters collected during the actual production process to obtain the springback hazard parameters, which reflect the potential risk level in the current pipe bending operation and help operators predict the hazard of springback.

[0069] By ensuring that potential hazards can be identified and addressed in a timely manner during the tube bending process, the safety and accuracy of the operation are improved.

[0070] In summary, the anti-rebound safety control method based on the pipe bending machine provided in this application has the following technical effects: The method collects the pipe specification information of the target pipe to be bent by the pipe bending machine and the pipe bending processing parameters of the target pipe to be bent; optimizes the pipe heating pretreatment parameters according to the pipe specification information and the pipe bending processing parameters to obtain the optimized pretreatment parameters, wherein the optimization is performed by minimizing the change in the surface performance of the pipe after pretreatment and minimizing the springback amount of the pipe after pretreatment; predicts the springback of the pipe based on the optimized pretreatment parameters combined with the pipe specification information and the pipe bending processing parameters to obtain the springback amount parameters and generate the bending processing parameters; uses the bending processing parameters to bend the target pipe, collects the pipe processing image sequence during the bending process, performs the fixture stability analysis to obtain the fixture stability parameters, and performs the springback risk analysis combined with the pipe springback amount parameters to obtain the springback risk parameters; performs the safety stop control judgment according to the springback risk parameters to obtain the judgment result, and performs the control to achieve the technical goal of accurately predicting the springback of the pipe, and achieves the technical effect of improving the pipe processing quality and reducing safety hazards.

[0071] Embodiment 2, based on the anti-rebound safety control method based on the pipe bending machine in the previous embodiment, the same inventive concept, this application also provides an anti-rebound safety control platform based on the pipe bending machine, please refer to the attached Figure 2 ,include: A pipe bending processing parameter acquisition module 11, the pipe bending processing parameter acquisition module 11 is used to collect the pipe specification information of the target pipe to be bent by the pipe bending machine, and the pipe bending processing parameters of the target pipe to be bent; an optimization pretreatment parameter acquisition module 12, the optimization pretreatment parameter acquisition module 12 is used to optimize the pipe heating pretreatment parameters according to the pipe specification information and the pipe bending processing parameters, and obtain the optimized pretreatment parameters, wherein the optimization is performed by minimizing the change in the surface performance of the pipe after pretreatment and minimizing the springback amount of the pipe after pretreatment; a pipe bending springback amount parameter acquisition module 13, the pipe bending springback amount parameter acquisition module 13 is used to optimize the pipe heating pretreatment parameters based on the optimization pretreatment parameters. The parameters are combined with the pipe specification information and the pipe bending processing parameters to predict the springback of the pipe bending, obtain the pipe bending springback parameters, and generate the bending processing parameters; the springback risk parameter acquisition module 14 is used to adopt the bending processing parameters to bend the target pipe, and collect the pipe processing image sequence during the pipe bending process, perform the fixture stability analysis, obtain the fixture stability parameters, and perform the springback risk analysis in combination with the pipe bending springback parameters to obtain the springback risk parameters; the judgment result acquisition module 15 is used to perform safety stop control judgment according to the springback risk parameters, obtain the judgment result, and perform control.

[0072] Furthermore, the pipe bending processing parameter acquisition module 11 in the anti-rebound safety control platform based on the pipe bending machine is also used for: The material information and size information of the target pipe to be bent by the pipe bending machine are collected as pipe specification information; and the standard bending specification parameters of the target pipe bending processing are collected as bending processing parameters.

[0073] Further, the optimization preprocessing parameter acquisition module 12 in the anti-rebound safety control platform based on the pipe bending machine is also used for: Obtain a pretreatment parameter space for the heating pretreatment of the pipe; randomly select a first pretreatment parameter in the pretreatment parameter space, combine the pipe specification information and the pipe bending processing parameters, perform pretreatment performance prediction, and obtain a first surface performance change and a first pipe springback; based on the first surface performance change and the first pipe springback, calculate and obtain a first pretreatment fitness, as shown in the following formula: ; Among them, PBF is the pre-processing fitness, and is the weight, is the first surface property change, To preset the surface performance change, is the springback of the pipe, The springback amount of the preset pipe is set; the pipe heating pretreatment parameters are continuously optimized in the pretreatment parameter space until the convergence requirements are met, and the pretreatment parameters with the maximum pretreatment fitness are output to obtain the optimized pretreatment parameters.

[0074] Further, the optimization preprocessing parameter acquisition module 12 in the anti-rebound safety control platform based on the pipe bending machine is also used for: According to the data records of heating pretreatment in historical time, a sample pretreatment parameter set, a sample pipe specification information set, and a sample pipe bending processing parameter set are collected, and according to the pipe test data under different sample pretreatment parameters, sample pipe specification information, and sample pipe bending processing parameters, a sample surface performance change amount set and a sample pipe rebound amount set are annotated and obtained; the sample pretreatment parameter set, sample pipe specification information set, and sample pipe bending processing parameter set are used as inputs, and the sample surface performance change amount set and the sample pipe rebound amount set are used as outputs to train a pretreatment performance predictor; the pretreatment performance predictor is used to predict the pretreatment performance of the first pretreatment parameters, pipe specification information, and pipe bending processing parameters to obtain a first surface performance change amount and a first pipe rebound amount.

[0075] Further, the pipe bending rebound parameter acquisition module 13 in the pipe bending machine-based anti-rebound safety control platform is also used for: A sample preprocessing parameter set, a sample pipe specification information set and a sample pipe bending processing parameter set are collected, and according to the springback amount of the sample pipe after bending processing, a sample pipe bending springback parameter set is marked and obtained; the sample preprocessing parameter set, the sample pipe specification information set, the sample pipe bending processing parameter set and the sample pipe bending springback parameter set are used to train a pipe bending springback predictor, and the pipe bending springback prediction is performed on the optimized preprocessing parameters, the pipe specification information and the pipe bending processing parameters to obtain the pipe bending springback parameters; according to the pipe bending springback parameters, the pipe bending processing parameters are classified and obtained, wherein the pipe bending springback parameters are input into a bending processing mapping table of the same type of pipes for mapping and classification, and the bending processing mapping table is constructed based on mapping of multiple sample pipe bending springback parameters and multiple sample bending processing parameters; the bending processing parameters and pipe bending processing parameters are superimposed to generate bending processing parameters.

[0076] Furthermore, the rebound risk parameter acquisition module 14 in the anti-rebound safety control platform based on the pipe bending machine is also used for: The target pipe is bent by adopting the bending processing parameters, and a pipe processing image sequence is collected during the bending process, wherein each pipe processing image includes an image of a pipe fixing fixture; a set of sample pipe processing image sequences is collected, and the loose amplitude of the fixture in each sample pipe processing image sequence is marked to obtain a set of sample fixture stability parameters, wherein the size of the sample fixture stability parameter is negatively correlated with the size of the loose amplitude of the fixture in the sample pipe processing image sequence; a fixture stability identifier is trained based on a convolutional neural network by adopting the sample pipe processing image sequence set and the sample fixture stability parameter set; and the fixture stability identifier is used to perform fixture stability analysis on the pipe processing image sequence to obtain a fixture stability parameter.

[0077] Furthermore, the rebound risk parameter acquisition module 14 in the anti-rebound safety control platform based on the pipe bending machine is also used for: According to the elbow rebound record data, a sample elbow rebound parameter set and a sample rebound risk parameter set are collected, and according to the rebound risk degree under different sample elbow rebound parameters and sample rebound risk parameters, the sample rebound risk parameter set is labeled and obtained; based on the decision tree, the sample elbow rebound parameter set and the sample rebound risk parameter set are used as classification inputs, and the sample rebound risk parameter set is used as classification output to construct a rebound risk classifier; the rebound risk classifier is used to classify the fixture stability parameters and elbow rebound parameters to obtain the rebound risk parameters.

[0078] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The anti-rebound safety control method based on a pipe bender and the specific examples in the aforementioned embodiment 1 are also applicable to the anti-rebound safety control platform based on a pipe bender in this embodiment. Through the aforementioned detailed description of the anti-rebound safety control method based on a pipe bender, those skilled in the art can clearly know the anti-rebound safety control platform based on a pipe bender in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the platform disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0079] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0080] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. Anti-rebound safety control method based on pipe bending machine, characterized in that: The anti-rebound safety control method based on a pipe bending machine is applied to an anti-rebound safety control platform based on a pipe bending machine, and the anti-rebound safety control platform based on a pipe bending machine is connected to the pipe bending machine, and includes: Collecting pipe specification information of a target pipe to be bent by a pipe bending machine, and pipe bending processing parameters of the target pipe to be bent; Optimizing the pipe heating pretreatment parameters according to the pipe specification information and the pipe bending processing parameters to obtain optimized pretreatment parameters, wherein the optimization is performed by minimizing the change in the surface properties of the pipe after pretreatment and minimizing the springback of the pipe after pretreatment; Based on the optimized preprocessing parameters combined with the pipe specification information and the pipe bending processing parameters, the pipe bending springback prediction is performed to obtain the pipe bending springback amount parameters and generate the bending processing parameters; The target pipe is bent by using the bending processing parameters, and a pipe processing image sequence is collected during the bending process to perform a fixture stability analysis to obtain a fixture stability parameter, and a springback risk analysis is performed in combination with the bending springback amount parameter to obtain a springback risk parameter; According to the rebound risk parameter, a safety stop control judgment is performed, a judgment result is obtained, and control is performed.

2. The anti-rebound safety control method based on a pipe bending machine according to claim 1 is characterized in that: Collecting the pipe specification information of the target pipe to be bent by the pipe bending machine, and the pipe bending processing parameters of the target pipe to be bent, including: Collecting material information and size information of the target pipe to be bent by the pipe bending machine as pipe specification information; The standard bending specification parameters of the target pipe bending process are collected as bending process parameters.

3. The anti-rebound safety control method based on a pipe bending machine according to claim 1, characterized in that: According to the pipe specification information and pipe bending processing parameters, the pipe heating pretreatment parameters are optimized to obtain the optimized pretreatment parameters, including: Obtain pretreatment parameter space for pipe heating pretreatment; Randomly select a first pretreatment parameter in the pretreatment parameter space, combine the pipe specification information and the pipe bending processing parameters, perform pretreatment performance prediction, and obtain a first surface performance change and a first pipe springback; Based on the first surface property change and the first tube springback, the first pretreatment fitness is calculated as follows: ； Among them, PBF is the pre-processing fitness, and is the weight, is the first surface property change, To preset the surface performance change, is the springback of the pipe, To preset the springback amount of the pipe; Continue to optimize the pipe heating pretreatment parameters in the pretreatment parameter space until the convergence requirement is met, output the pretreatment parameters with the maximum pretreatment fitness, and obtain the optimized pretreatment parameters.

4. The anti-rebound safety control method based on a pipe bending machine according to claim 3 is characterized in that: Randomly selecting a first pretreatment parameter in the pretreatment parameter space, combining the pipe specification information and the pipe bending processing parameters, performing pretreatment performance prediction, and obtaining a first surface performance change and a first pipe springback, including: According to the data records of heating pretreatment in the historical time, the sample pretreatment parameter set, the sample pipe specification information set, and the sample pipe bending processing parameter set are collected, and according to the pipe test data under different sample pretreatment parameters, sample pipe specification information and sample pipe bending processing parameters, the sample surface performance change amount set and the sample pipe rebound amount set are marked and obtained; The sample preprocessing parameter set, the sample pipe specification information set, and the sample pipe bending processing parameter set are used as inputs, and the sample surface performance variation set and the sample pipe springback amount set are used as outputs to train a preprocessing performance predictor; The pretreatment performance predictor is used to predict the pretreatment performance of the first pretreatment parameters, pipe specification information and pipe bending processing parameters to obtain a first surface performance change and a first pipe springback.

5. The anti-rebound safety control method based on a pipe bending machine according to claim 1, characterized in that: Based on the optimized preprocessing parameters combined with the pipe specification information and the pipe bending processing parameters, the pipe bending springback prediction is performed to obtain the pipe bending springback amount parameters and generate the bending processing parameters, including: Collect a sample preprocessing parameter set, a sample pipe material specification information set and a sample pipe bending processing parameter set, and mark and obtain a sample pipe bending springback parameter set according to the springback amount of the sample pipe after the bending process; The sample preprocessing parameter set, the sample pipe specification information set, the sample pipe bending processing parameter set and the sample pipe bending springback parameter set are used to train a pipe bending springback predictor, and the pipe bending springback prediction is performed on the optimized preprocessing parameters, the pipe specification information and the pipe bending processing parameters to obtain the pipe bending springback parameters; According to the bend springback parameter, the bending processing parameter is obtained by classification, wherein the bend springback parameter is input into a bending processing mapping table of the same type of pipes for mapping classification, and the bending processing mapping table is constructed based on mapping of multiple sample bend springback parameters and multiple sample bending processing parameters; The bending processing parameters and the pipe bending processing parameters are superimposed to generate the bending processing parameters.

6. The anti-rebound safety control method based on a pipe bending machine according to claim 1, characterized in that: The target pipe is bent by using the bending processing parameters, and a pipe processing image sequence is collected during the bending process to perform a fixture stability analysis to obtain a fixture stability parameter, including: Using the bending processing parameters, the target pipe is bent, and a pipe processing image sequence is collected during the bending process, wherein each pipe processing image includes an image of a pipe fixing fixture; A set of sample pipe processing image sequences is collected, and the loose amplitude of the clamps in each sample pipe processing image sequence is marked to obtain a set of sample clamp stability parameters, wherein the magnitude of the sample clamp stability parameter is negatively correlated with the magnitude of the loose amplitude of the clamps in the sample pipe processing image sequence; Using the sample tube processing image sequence set and the sample fixture stability parameter set, a fixture stability identifier is trained based on a convolutional neural network; The fixture stability identifier is used to perform fixture stability analysis on the pipe processing image sequence to obtain fixture stability parameters.

7. The anti-rebound safety control method based on a pipe bending machine according to claim 1, characterized in that: The springback risk analysis is performed in combination with the springback amount parameter of the bent pipe to obtain the springback risk parameters, including: According to the elbow springback record data, a set of sample elbow springback quantity parameters and a set of sample springback risk parameters are collected, and according to the springback risk degree under different sample elbow springback quantity parameters and sample springback risk parameters, a set of sample springback risk parameters is obtained by marking; Based on the decision tree, the sample bending pipe springback parameter set and the sample springback risk parameter set are used as classification inputs, and the sample springback risk parameter set is used as classification output to construct a springback risk classifier; The springback risk classifier is used to classify the fixture stability parameter and the pipe bending springback parameter to obtain the springback risk parameter. 8.Anti-rebound safety control platform based on pipe bending machine, characterized in that: The steps for implementing the anti-rebound safety control method based on a pipe bending machine as described in any one of claims 1 to 7 include: A pipe bending processing parameter acquisition module, the pipe bending processing parameter acquisition module is used to collect pipe specification information of a target pipe to be bent by a pipe bending machine, and pipe bending processing parameters of the target pipe to be bent; An optimized pretreatment parameter acquisition module, which is used to optimize the pipe heating pretreatment parameters according to the pipe specification information and the pipe bending processing parameters to obtain optimized pretreatment parameters, wherein the optimization is performed by minimizing the change in the surface performance of the pipe after pretreatment and minimizing the springback of the pipe after pretreatment; A pipe bending springback parameter acquisition module, which is used to predict pipe bending springback based on the optimized preprocessing parameters combined with the pipe specification information and pipe bending processing parameters, obtain pipe bending springback parameters, and generate bending processing parameters; A springback risk parameter acquisition module, the springback risk parameter acquisition module is used to use the bending processing parameters to bend the target pipe, collect a pipe processing image sequence during the bending process, perform a fixture stability analysis, obtain a fixture stability parameter, and perform a springback risk analysis in combination with the bending springback amount parameter to obtain a springback risk parameter; A discrimination result obtaining module is used to perform safety stop control discrimination according to the rebound risk parameter, obtain a discrimination result, and perform control.

Images