Steel pipe perforating process optimization method and device, electronic equipment and storage medium
Through the recursive neural network model and artificial bee colony algorithm, the perforation process parameters of steel pipes are optimized, which solves the problem of capillary quality defects caused by unreasonable perforation process parameters, and realizes high-precision capillary quality prediction and process parameter optimization, ensuring capillary quality Stability and accuracy.
Patent Information
- Application Number
- CN202411487513.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-24
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Unreasonable parameters of steel pipe perforation lead to capillary quality defects, and there is a lack of effective optimization methods to ensure capillary quality control.
The recursive neural network model is used to combine artificial bee colony algorithm to screen and optimize the importance characteristics of steel pipe perforation process parameters, predict the quality of the capillary tube, and adjust the parameters according to the preset quality requirements until the capillary tube quality meets the requirements is achieved.
It realizes high-precision capillary quality prediction and process parameter optimization, avoids product quality defects caused by unreasonable perforation process parameters, and provides more accurate guarantees for capillary quality control.
Smart Images

Figure CN119005455A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of electrical digital data processing, and in particular to a steel pipe perforation process optimization method, device, electronic equipment and storage medium. Background Art
[0002] Seamless steel pipe is one of the important materials for national construction. Two-roller oblique rolling and piercing is an efficient seamless steel pipe production process. The equipment is mainly composed of two rotating rollers, one of which is an actively rotating guide plate and the other is a passively rotating roller. This process is widely used in the first process of hot-rolled seamless steel pipe production. The tube blank becomes a rough tube after two-roller oblique rolling and piercing. Among them, the process parameters during piercing have an important influence on the quality of the rough tube.
[0003] Therefore, it is necessary to optimize and adjust the perforation process parameters in a timely manner to avoid product quality defects caused by unreasonable perforation process parameters and provide a guarantee for more accurate capillary quality control. Summary of the invention
[0004] The embodiments of the present application provide a steel pipe perforation process optimization method, device, electronic device and storage medium to solve the problem that unreasonable perforation process parameters easily lead to quality defects in the rough pipe.
[0005] In a first aspect, an embodiment of the present application provides a method for optimizing a steel pipe perforation process, comprising: Obtaining setting parameters of the steel pipe perforation process; wherein the setting parameters are the process parameters of the steel pipe perforation process that are initially set; Performing importance feature screening on the set parameters to obtain important parameters of the steel pipe perforation process; The important parameters are input into a trained recursive neural network model to obtain raw tube quality prediction data output by the recursive neural network model; wherein the recursive neural network model is trained based on process parameters of multiple tube blanks during the perforation process and raw tube quality data of each tube blank; According to the capillary quality prediction data and the preset quality requirements, the setting parameters are optimized so that the capillary quality prediction data meets the preset quality requirements.
[0006] In a possible implementation, before inputting the important parameters into the trained recursive neural network model to obtain the capillary quality prediction data output by the recursive neural network model, the method further includes: Obtaining process parameters of multiple tube blanks during the perforation process and rough tube quality data of each tube blank; The process parameters of multiple tube blanks during the perforation process are screened by importance features to obtain the important parameters of multiple tube blanks during the perforation process; wherein the important parameters are process parameters that have a strong correlation with the raw tube quality data, and the strong correlation represents that the value of the correlation is greater than a preset threshold; According to the important parameters of multiple tube blanks during the perforation process and the rough tube quality data of each tube blank, the recursive neural network model is trained to obtain a trained recursive neural network model.
[0007] In a possible implementation, the importance feature screening of the process parameters of the plurality of tube blanks during the piercing process to obtain the important parameters of the plurality of tube blanks during the piercing process includes: Calculate the value of the Pearson correlation coefficient between each process parameter and the capillary quality data; The process parameters whose Pearson correlation coefficient values are greater than the preset threshold are determined as important parameters.
[0008] In a possible implementation, the recursive neural network model includes an input layer, a hidden layer, a receiving layer and an output layer; Before training the recursive neural network model according to important parameters of the plurality of tube blanks during the perforation process and the rough tube quality data of each tube blank, the method further comprises: Based on the artificial bee colony algorithm, the model parameters of the recursive neural network model are determined; wherein the model parameters include the connection weights from the input layer to the hidden layer, the connection weights from the hidden layer to the output layer, the connection weights from the output layer to the receiving layer, the hidden layer threshold and the output layer threshold in the recursive neural network model; The recursive neural network model is trained according to the important parameters of the plurality of tube blanks during the perforation process and the rough tube quality data of each tube blank, including: The recursive neural network model is trained according to important parameters of multiple tube blanks during the perforation process, rough tube quality data of each tube blank and the model parameters.
[0009] In one possible implementation, the process parameters of the steel pipe perforation process include the upper and lower roller motor current, the upper and lower roller speed, the upper and lower roller torque, the tube billet diameter, the plug diameter, the inlet and outlet cone angles, the roller spacing, the guide plate spacing and the plug extension; the important parameters of the steel pipe perforation process include the plug diameter, the tube billet diameter, the guide plate spacing, the roller spacing and the plug extension.
[0010] In a possible implementation, the capillary tube quality prediction data includes the predicted outer diameter and predicted wall thickness of the capillary tube; The optimizing process of the setting parameters according to the capillary quality prediction data and the preset quality requirements includes: When the predicted outer diameter and / or predicted wall thickness of the capillary tube does not meet the preset quality requirement, adjusting the setting parameters based on the difference between the predicted outer diameter and / or predicted wall thickness of the capillary tube and the preset quality requirement to obtain new setting parameters; Based on the new set parameters, the step of screening the set parameters for importance features to obtain important parameters of the steel pipe perforation process is repeated until the rough pipe quality prediction data meets the preset quality requirements.
[0011] In a possible implementation, the recursive neural network model is trained according to important parameters of a plurality of tube blanks during the perforation process, rough tube quality data of each tube blank and the model parameters, including: Input important parameters of each tube blank during the piercing process into the recursive neural network model, and obtain the rough tube quality prediction data of each tube blank output by the recursive neural network model; According to the rough tube quality data of each tube blank, the rough tube quality prediction data of each tube blank and the loss function, the model parameters are adjusted to obtain a trained recursive neural network model.
[0012] In a second aspect, an embodiment of the present application provides a steel pipe perforation process optimization device, comprising: An acquisition module, used to acquire setting parameters of the steel pipe perforation process; wherein the setting parameters are the process parameters of the steel pipe perforation process that are initially set; A screening module is used to screen the set parameters by importance characteristics to obtain important parameters of the steel pipe perforation process; A processing module, used for inputting the important parameters into a trained recursive neural network model to obtain raw tube quality prediction data output by the recursive neural network model; wherein the recursive neural network model is trained based on process parameters of multiple tube blanks during the perforation process and raw tube quality data of each tube blank; The optimization module is used to optimize the set parameters according to the capillary quality prediction data and the preset quality requirements so that the capillary quality prediction data meets the preset quality requirements.
[0013] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method described in the first aspect or any possible implementation of the first aspect are implemented.
[0014] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the method described in the first aspect or any possible implementation method of the first aspect.
[0015] The embodiments of the present application provide a method, device, electronic device and storage medium for optimizing the steel pipe perforation process. First, the process parameters of the steel pipe perforation process that are preliminarily set are obtained, and the set parameters are screened for importance characteristics to obtain the important parameters of the steel pipe perforation process. The important parameters are then input into a trained recursive neural network model to obtain the raw tube quality prediction data output by the recursive neural network model. Finally, based on the raw tube quality prediction data and the preset quality requirements, the set parameters are optimized so that the raw tube quality data subsequently obtained based on the optimized process parameters meet the preset quality requirements. Among them, the embodiments of the present application utilize the idea of parameter optimization using a model to provide a method for optimizing the steel pipe perforation process parameters, and quickly obtain data that meets the preset quality requirements. Moreover, by screening the input items of the model for importance, the accuracy of the model output results is improved, and high-precision raw tube quality prediction can be achieved, thereby quickly and effectively optimizing the perforation process parameters and optimizing the steel pipe perforation process. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0017] Figure 1 It is a flow chart of an implementation method of a steel pipe perforation process optimization method provided in an embodiment of the present application; Figure 2 It is a flow chart of another implementation method of a steel pipe perforation process optimization method provided in an embodiment of the present application; Figure 3 It is a structural schematic diagram of a steel pipe perforation process optimization device provided in an embodiment of the present application; Figure 4 It is a schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0018] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.
[0019] In order to optimize the steel pipe perforation process and improve the perforation quality of the tube blank, the embodiment of the present application uses the idea of parameter optimization using a model, and provides a method for optimizing the steel pipe perforation process parameters to quickly obtain data that meets the preset quality requirements. Moreover, the embodiment of the present application improves the accuracy of the model output results by screening the importance of the model input items, and can achieve high-precision rough pipe quality prediction, thereby quickly and effectively correcting the perforation process parameters and optimizing the steel pipe perforation process.
[0020] In order to make the purpose, technical solutions and advantages of the present application clearer, specific embodiments will be described below in conjunction with the accompanying drawings.
[0021] See also Figure 1 , which shows a flow chart of an implementation of a steel pipe perforation process optimization method provided in an embodiment of the present application, which is described in detail as follows: In step 101, the setting parameters of the steel pipe perforation process are obtained; wherein the setting parameters are the process parameters of the steel pipe perforation process that are initially set.
[0022] Exemplarily, this embodiment obtains setting parameters of the steel pipe perforation process, wherein the setting parameters are process parameters of the steel pipe perforation process that are preliminarily set and to be optimized, wherein the steel pipe perforation process is a two-roll oblique rolling perforation process for the tube billet, and correspondingly, the process parameters may include the upper and lower roller motor current, the upper and lower roller speed, the upper and lower roller torque, the tube billet diameter, the plug diameter, the inlet and outlet cone angles, the roller spacing, the guide plate spacing, and the plug extension, etc.
[0023] In step 102, the set parameters are screened for important features to obtain important parameters of the steel pipe perforation process.
[0024] Exemplarily, this embodiment performs importance feature screening on the set parameters to obtain important parameters of the steel pipe perforation process, wherein the important parameters may refer to process parameters that have a strong correlation with the quality of the rough pipe finally obtained by the steel pipe perforation process, and the strong correlation indicates that the value of the correlation is greater than a preset threshold.
[0025] In one example, this embodiment can screen the importance characteristics of the process parameters of the two-roll oblique rolling piercing process of the tube billet based on the Pearson coefficient, and the important parameters screened are the head diameter, the tube billet diameter, the guide plate distance, the roller spacing and the head extension.
[0026] In step 103, important parameters are input into the trained recursive neural network model to obtain rough tube quality prediction data output by the recursive neural network model; wherein the recursive neural network model is trained based on process parameters of multiple tube blanks during the perforation process and rough tube quality data of each tube blank.
[0027] Exemplarily, in this embodiment, the important parameters obtained after screening are input into a pre-trained recursive neural network model, and the recursive neural network model performs calculations to obtain the capillary quality prediction data output by the recursive neural network model. The recursive neural network model may be an ELMAN neural network model.
[0028] The capillary tube quality prediction data may include the predicted outer diameter and predicted wall thickness of the capillary tube.
[0029] Optionally, the recursive neural network model is trained based on process parameters of multiple tube blanks during the perforation process and rough tube quality data of each tube blank. The recursive neural network model can also be optimized based on an artificial bee colony (ABC) algorithm.
[0030] In one example, this embodiment optimizes the recursive neural network model based on the ABC algorithm, and trains the recursive neural network model based on the process parameters of multiple tube blanks during the perforation process and the rough tube quality data of each tube blank, and finally obtains a trained recursive neural network model.
[0031] Here, this embodiment obtains process parameters of multiple tube blanks during the punching process and rough tube quality data of each tube blank; and performs importance feature screening on the process parameters of multiple tube blanks during the punching process to obtain important parameters that have a strong correlation with the rough tube quality data of multiple tube blanks during the punching process; and then trains the recursive neural network model based on the important parameters of multiple tube blanks during the punching process and the rough tube quality data of each tube blank to obtain a trained recursive neural network model.
[0032] In step 104, the setting parameters are optimized according to the capillary quality prediction data and the preset quality requirements, so that the capillary quality prediction data meets the preset quality requirements.
[0033] Illustratively, after obtaining the capillary quality prediction data output by the model, such as the predicted outer diameter and predicted wall thickness of the capillary, this embodiment compares the capillary quality prediction data with the preset capillary quality requirements, and optimizes the set parameters based on the difference between the capillary quality prediction data and the preset quality requirements so that subsequent capillary quality data meets the preset quality requirements.
[0034] In one example, if the predicted outer diameter and predicted wall thickness of the raw tube both meet the preset quality requirements, the set parameters can be used as process parameters of the steel pipe perforation process; if one or more of the predicted outer diameter and predicted wall thickness of the raw tube do not meet the preset quality requirements, the set parameters can be optimized based on the gap between the predicted outer diameter and / or predicted wall thickness of the raw tube and the preset quality requirements so that the raw tube quality prediction data meets the preset quality requirements.
[0035] The embodiment of the present application first obtains the process parameters of the steel pipe perforation process that are preliminarily set, and performs importance feature screening on the set parameters to obtain the important parameters of the steel pipe perforation process, and then inputs the important parameters into the trained recursive neural network model to obtain the rough pipe quality prediction data output by the recursive neural network model, and finally, based on the rough pipe quality prediction data and the preset quality requirements, the set parameters are optimized. Among them, this embodiment uses the idea of parameter optimization by the model to provide a method for optimizing the steel pipe perforation process parameters, and quickly obtains data that meets the preset quality requirements. Moreover, by performing importance screening on the input items of the recursive neural network model, the accuracy of the model output results is improved, and a high-precision prediction of the rough pipe quality can be achieved before the tube blank is perforated, which is helpful for controlling the rough pipe quality, and then the perforation process parameters can be quickly and effectively optimized, and the steel pipe perforation process can be optimized.
[0036] In this embodiment, before the important parameters are input into the trained recursive neural network model, model training is required first, so that the set parameters of the steel pipe perforation process can be optimized based on the trained recursive neural network model, so as to avoid product quality defects caused by unreasonable perforation process parameters and provide guarantee for more accurate raw pipe quality control. As an example, how to obtain a trained recursive neural network model is further described below.
[0037] In some embodiments, the recursive neural network model includes an input layer, a hidden layer, a receiving layer, and an output layer; before training the recursive neural network model according to important parameters of multiple tube blanks during the perforation process and the rough tube quality data of each tube blank, the method further includes: Based on the artificial bee colony algorithm, the model parameters of the recursive neural network model are determined; wherein the model parameters include the connection weights from the input layer to the hidden layer, the connection weights from the hidden layer to the output layer, the connection weights from the output layer to the receiving layer, the hidden layer threshold and the output layer threshold in the recursive neural network model.
[0038] Exemplarily, the recursive neural network model in this embodiment includes an input layer, a hidden layer, a receiving layer and an output layer. Before training the recursive neural network model according to the important parameters of multiple tube blanks during the perforation process and the rough tube quality data of each tube blank, this embodiment first performs a preliminary optimization processing on the model parameters based on the artificial bee colony algorithm to obtain the connection weights from the input layer to the hidden layer, the connection weights from the hidden layer to the output layer, the connection weights from the output layer to the receiving layer, the hidden layer threshold and the output layer threshold in the recursive neural network model.
[0039] Optionally, this embodiment determines the structure of the recursive neural network model, including the number of network layers and the number of neurons in each layer, and initializes the connection weights and thresholds of the recursive neural network model, sets the initial weights and thresholds of T groups of recursive neural network models; determines the size of the bee colony and the ratio of the leading bees to the observation bees, wherein the size of the bee colony is 100, the ratio of the leading bees to the observation bees is 1:1, and sets the initial nectar source; near the initial nectar source, the leading bee searches for a new nectar source, compares the new nectar source with the initial nectar source, and selects the nectar source with a better fitness function value to enter the next step; compares the current nectar source with the new nectar source searched near the nectar source, and selects the one with a larger fitness function value to enter the next step; determines whether a scout bee appears; determines whether the set termination condition is met. After the termination condition is met, the connection weights from the input layer to the hidden layer, the connection weights from the hidden layer to the output layer, the connection weights from the output layer to the receiving layer, the hidden layer threshold, and the output layer threshold in the recursive neural network model are obtained.
[0040] In one example, the structure of the recursive neural network model is IHO, where I=5, H=10, O=2, I, H, and O are the number of neurons in the input layer, hidden layer, and output layer of the model, respectively. The formula for calculating the number of neurons in the hidden layer is as follows:
[0041] in, is a value in the interval [1, 10].
[0042] In an example, the fitness evaluation function is as follows:
[0043] in is the mean square error of the ith solution, under ideal conditions is 0.
[0044] Before training the recursive neural network model, this embodiment first optimizes the model parameters of the recursive neural network model through an artificial bee colony algorithm, and uses the obtained optimal solution as the initial model parameters for subsequent recursive neural network model training, thereby improving the prediction accuracy of the recursive neural network model for capillary quality data, saving model training time, and reducing the consumption of computing resources during model training.
[0045] In some embodiments, before inputting the important parameters into the trained recursive neural network model to obtain the capillary quality prediction data output by the recursive neural network model, the method further includes: The process parameters of multiple tube blanks during the piercing process and the rough tube quality data of each tube blank are obtained.
[0046] Importance features of process parameters of multiple tube billets during the piercing process are screened to obtain important parameters of multiple tube billets during the piercing process; wherein the important parameters are process parameters that have a strong correlation with the rough tube quality data, and the strong correlation indicates that the value of the correlation is greater than a preset threshold.
[0047] According to the important parameters of multiple tube blanks during the perforation process and the rough tube quality data of each tube blank, the recursive neural network model is trained to obtain a trained recursive neural network model.
[0048] Among them, the importance characteristics of the process parameters of multiple tube blanks during the piercing process are screened to obtain the important parameters of multiple tube blanks during the piercing process, including: Calculate the value of the Pearson correlation coefficient between each process parameter and the capillary quality data.
[0049] Process parameters whose Pearson correlation coefficient values are greater than a preset threshold are determined as important parameters.
[0050] Exemplarily, this embodiment acquires multiple sets of data, each set of data includes process parameters of the tube billet during the perforation process and rough tube quality data obtained after the perforation, and the multiple sets of data serve as data sources for subsequent model training.
[0051] Among them, the process parameters of the tube billet during the piercing process, that is, the specific values of the process parameters set by the piercer when the tube billet is subjected to two-roll oblique rolling piercing treatment, include: the upper and lower roller motor current, the upper and lower roller speed, the upper and lower roller torque, the tube billet diameter, the plug diameter, the inlet and outlet cone angles, the roller spacing, the guide plate spacing and the plug extension amount; the rough tube quality data includes the rough tube outer diameter and rough tube wall thickness obtained for each tube billet after the piercing treatment.
[0052] In this embodiment, importance feature screening is performed on the process parameters of each tube blank during the piercing process, so as to screen out important process parameters of each tube blank that have a strong correlation with the rough tube quality data during the piercing process.
[0053] In a feasible implementation manner, before screening the importance features of process parameters of multiple tube blanks during the piercing process, the data are preprocessed, including outlier processing and standardization processing.
[0054] In one example, this embodiment uses the Pauta criterion to eliminate outliers and noise in the data and restore the original information in the data. , the value of the process parameter during the perforation of tube blank i is used It means that the average value of the process parameter setting value of multiple tube blanks during the perforation process is , the calculation process of abnormal data elimination is as follows:
[0055]
[0056]
[0057] in, is the standard deviation of the data, and n is the total number of tube blanks.
[0058] Data with a difference from the mean greater than 3 standard deviations were removed as outliers.
[0059] In an example, in order to eliminate the impact of differences in dimensions and value ranges among parameter variables, data standardization is required. The standardization method is as follows:
[0060] Among them, for each process parameter , Refers to each The maximum data in Refers to each The minimum data in . for Normalized data.
[0061] After preprocessing, the value of the Pearson correlation coefficient between each process parameter and the capillary quality data is calculated, and the process parameters whose values of the Pearson correlation coefficient are greater than a preset threshold are determined as important parameters.
[0062] In one example, for each process parameter , calculate the average value of the process parameter setting value of multiple tube blanks during the perforation process , and calculate the average value of the rough tube quality data obtained after the perforation treatment of multiple tube blanks , where, when calculating the Pearson correlation coefficient, It can be two values, namely, the average value of the outer diameter of the capillary tube and the average value of the wall thickness of the capillary tube; It can also be replaced by a numerical representation obtained by weighted calculation of the average value of the capillary outer diameter and the average value of the capillary wall thickness.
[0063] For tube blank i , the corresponding capillary quality data is , where capillary quality data It can include the outer diameter of the capillary and the wall thickness of the capillary, or it can be represented by the weighted calculated value of the outer diameter of the capillary and the wall thickness of the capillary; for the selected process parameters , tube blank i The value of this process parameter is used in the perforation process. express, i From 1 to n. n is the total number of tube blanks.
[0064] The Pearson correlation coefficient is calculated as follows:
[0065] in, and are two variables; For variables The mean of For variables The mean value; n is the total number of tube blanks, for and The correlation coefficient between .
[0066] Optionally, if the calculated If the value is greater than the preset threshold, and There is a strong correlation, that is It is an important parameter of tube blank i.
[0067] In one example, a recursive neural network model is trained according to important parameters of multiple tube blanks during the perforation process and rough tube quality data corresponding to each tube blank to obtain a trained recursive neural network model, including: The important parameters of each tube blank during the piercing process are input into the recursive neural network model, and the rough tube quality prediction data of each tube blank output by the recursive neural network model is obtained.
[0068] According to the rough tube quality data of each tube blank, the rough tube quality prediction data of each tube blank and the loss function, the model parameters are adjusted to obtain a trained recursive neural network model.
[0069] This embodiment further trains and optimizes the model parameters of the recursive neural network model obtained by the artificial bee colony algorithm in the above embodiment based on the important parameters of each tube blank after pretreatment and the quality data of the rough tube, so as to obtain a trained recursive neural network model, thereby further improving the prediction accuracy of the model.
[0070] The above embodiment introduces how to obtain a trained recursive neural network model. The following further introduces the preprocessing of setting parameters.
[0071] In some embodiments, before screening the set parameters for important features to obtain important parameters of the steel pipe perforation process, the process further includes: The set parameters are standardized; the standardization is used to make the data dimensionless and eliminate the differences in the value range.
[0072] Exemplarily, before performing importance feature screening on the setting parameters, this embodiment preprocesses the acquired setting parameters, including standardization processing, where the standardization processing is used to perform dimensionless processing on the data and eliminate the difference in the value range.
[0073] In an example, in order to eliminate the impact of differences in dimensions and value ranges among parameter variables, data standardization is required. The standardization method is as follows:
[0074] In the formula, Refers to the maximum data in the set parameters. Refers to the minimum data in the setting parameters. for Normalized data.
[0075] This embodiment improves the accuracy of subsequent capillary quality data prediction based on the set parameters by standardizing the set parameters, and eliminates the influence of the differences in dimensions and value ranges between the parameter variables on data screening and identification.
[0076] In some embodiments, after the setting parameters are standardized, the setting parameters are screened for important features to obtain important parameters of the steel pipe perforation process.
[0077] Exemplarily, this embodiment performs importance feature screening on the set parameters to obtain important parameters of the steel pipe perforation process, wherein the important parameters may refer to process parameters that have a strong correlation with the quality of the rough pipe finally obtained by the steel pipe perforation process. The strong correlation indicates that the value of the correlation is greater than a preset threshold.
[0078] In a feasible implementation manner, the set parameters may be screened based on the important parameters of each tube blank during the piercing process screened in the aforementioned embodiments to obtain the important parameters of the steel tube piercing process.
[0079] In one example, the intersection or union of the types of important parameters of each tube blank screened in the aforementioned step during the piercing process is taken to obtain the types of important parameters of the steel tube piercing process, and then determine the important parameters of the steel tube piercing process.
[0080] In one example, the important parameters screened by this embodiment are the plug diameter, the tube diameter, the guide plate distance, the roller spacing and the plug extension.
[0081] In one example, this embodiment predicts the capillary quality data based on the selected important parameters and the trained recursive neural network model. Prior to this, this embodiment sets up multiple groups of experimental controls to verify the feasibility of the method. The capillary quality prediction data output by the recursive neural network model is compared with the actual capillary quality data, and the correlation coefficient is used to R 2 The prediction accuracy was evaluated by , and the prediction accuracy of the method for capillary wall thickness reached 0.90, and the prediction accuracy for capillary outer diameter reached 0.88, which ensured the feasibility of the method.
[0082] This embodiment performs importance screening of setting parameters based on important parameters after screening of multiple tube blanks, and then predicts the rough tube quality based on the important parameters, which further improves the prediction accuracy, speeds up the solution of the model, and saves computing resources.
[0083] In some embodiments, the capillary quality prediction data includes a predicted outer diameter and a predicted wall thickness of the capillary; According to the predicted quality data of the capillary tube and the preset quality requirements, the set parameters are optimized, including: When the predicted outer diameter and / or predicted wall thickness of the capillary tube does not meet the preset quality requirements, the setting parameters are adjusted based on the difference between the predicted outer diameter and / or predicted wall thickness of the capillary tube and the preset quality requirements to obtain new setting parameters.
[0084] Based on the new set parameters, the steps of screening the importance features of the set parameters to obtain the important parameters of the steel pipe perforation process are repeated until the rough pipe quality prediction data meets the preset quality requirements.
[0085] Illustratively, after obtaining the capillary quality prediction data, this embodiment compares it with the preset quality requirements, and optimizes the set parameters according to the difference between the capillary quality prediction data and the preset quality requirements so that subsequent capillary quality data meets the preset quality requirements.
[0086] In one example, if the predicted outer diameter and predicted wall thickness of the raw tube both meet the preset quality requirements, the set parameters can be used as process parameters of the steel tube perforation process; if one or more of the predicted outer diameter and predicted wall thickness of the raw tube do not meet the preset quality requirements, the set parameters can be optimized based on the gap between the predicted outer diameter and / or predicted wall thickness of the raw tube and the preset quality requirements to obtain new set parameters, and based on the new set parameters, the steps of screening the importance features of the set parameters in the above embodiment to obtain the important parameters of the steel tube perforation process are repeated until the raw tube quality prediction data meets the preset quality requirements.
[0087] This embodiment uses the rough pipe quality prediction data to assist technical personnel in quickly optimizing the process parameters of the steel pipe perforation process, which can optimize the steel pipe perforation process and improve the production stability of the equipment.
[0088] Figure 2 The following is a flowchart of the steel pipe perforation process optimization method provided in the embodiment of the present application, which is described in detail as follows: In step 201, process parameters of a plurality of tube blanks during the perforation process and rough tube quality data of each tube blank are obtained.
[0089] In step 202, the process parameters of multiple tube blanks during the piercing process are screened for importance features to obtain important parameters of each tube blank during the piercing process; wherein the important parameters are process parameters that have a strong correlation with the rough tube quality data, and the strong correlation indicates that the value of the correlation is greater than a preset threshold.
[0090] In one example, step 202 includes the following steps: Calculate the value of the Pearson correlation coefficient between each process parameter and the capillary quality data.
[0091] Process parameters whose Pearson correlation coefficient values are greater than a preset threshold are determined as important parameters.
[0092] In step 203, based on the artificial bee colony algorithm, the model parameters of the recursive neural network model are determined; the recursive neural network model includes an input layer, a hidden layer, a receiving layer and an output layer; wherein the model parameters include the connection weights from the input layer to the hidden layer, the connection weights from the hidden layer to the output layer, the connection weights from the output layer to the receiving layer, the hidden layer threshold and the output layer threshold in the recursive neural network model.
[0093] In step 204, the recursive neural network model is trained according to the important parameters of each tube blank during the perforation process, the rough tube quality data and model parameters corresponding to each tube blank to obtain a trained recursive neural network model.
[0094] In one example, step 204 includes the following steps: The important parameters of each tube blank during the piercing process are input into the recursive neural network model, and the rough tube quality prediction data of each tube blank output by the recursive neural network model is obtained.
[0095] According to the rough tube quality data of each tube blank, the rough tube quality prediction data of each tube blank and the loss function, the model parameters are adjusted to obtain a trained recursive neural network model.
[0096] In step 205, the setting parameters of the steel pipe perforation process are obtained; wherein the setting parameters are the process parameters of the steel pipe perforation process that are initially set.
[0097] In step 206, the set parameters are screened for important features to obtain important parameters of the steel pipe perforation process.
[0098] In one example, before screening, the set parameters are standardized; wherein the standardization is used to make the data dimensionless and eliminate the difference in the value range.
[0099] In one example, the process parameters of the steel pipe piercing process include the upper and lower roller motor current, the upper and lower roller speed, the upper and lower roller torque, the tube billet diameter, the plug diameter, the inlet and outlet cone angles, the roller spacing, the guide plate spacing and the plug extension; the important parameters of the steel pipe piercing process include the plug diameter, the tube billet diameter, the guide plate spacing, the roller spacing and the plug extension.
[0100] In step 207, important parameters are input into the trained recursive neural network model to obtain rough tube quality prediction data output by the recursive neural network model; wherein the recursive neural network model is trained based on process parameters of multiple tube blanks during the perforation process and rough tube quality data of each tube blank.
[0101] In step 208, the setting parameters are optimized according to the capillary quality prediction data and the preset quality requirements, so that the capillary quality prediction data meets the preset quality requirements.
[0102] In one example, step 208 includes the following steps: When the predicted outer diameter and / or predicted wall thickness of the capillary tube does not meet the preset quality requirements, the setting parameters are adjusted based on the difference between the predicted outer diameter and / or predicted wall thickness of the capillary tube and the preset quality requirements to obtain new setting parameters.
[0103] Based on the new set parameters, the steps of screening the importance features of the set parameters to obtain the important parameters of the steel pipe perforation process are repeated until the rough pipe quality prediction data meets the preset quality requirements.
[0104] The detailed implementation process of the above steps 201 to 208 can refer to the relevant descriptions of the above embodiments, which will not be repeated here. The execution order of the above steps 201 to 208 can be set according to actual needs and is not specifically limited here.
[0105] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0106] The following is an embodiment of the device of the present application. For details not described in detail, please refer to the corresponding method embodiment described above.
[0107] Figure 3 The structural schematic diagram of the steel pipe perforation process optimization device provided in the embodiment of the present application is shown. For the convenience of explanation, only the part related to the embodiment of the present application is shown, which is described in detail as follows: like Figure 3 As shown, the steel pipe perforation process optimization device 3 includes: The acquisition module 301 is used to acquire the setting parameters of the steel pipe perforation process; wherein the setting parameters are the process parameters of the steel pipe perforation process that are initially set.
[0108] The screening module 302 is used to screen the set parameters for important features and obtain important parameters of the steel pipe perforation process.
[0109] The processing module 303 is used to input important parameters into the trained recursive neural network model to obtain the rough tube quality prediction data output by the recursive neural network model; wherein the recursive neural network model is trained based on the process parameters of multiple tube blanks during the perforation process and the rough tube quality data of each tube blank.
[0110] The optimization module 304 is used to optimize the set parameters according to the capillary quality prediction data and the preset quality requirements, so that the capillary quality prediction data meets the preset quality requirements.
[0111] In a possible implementation, before the processing module 303, the apparatus 3 further includes a training module, and the training module is used to: The process parameters of multiple tube blanks during the piercing process and the rough tube quality data of each tube blank are obtained; the process parameters of the multiple tube blanks during the piercing process are screened by importance features to obtain the important parameters of the multiple tube blanks during the piercing process; wherein the important parameters are process parameters that have a strong correlation with the rough tube quality data, and the strong correlation indicates that the value of the correlation is greater than a preset threshold; according to the important parameters of the multiple tube blanks during the piercing process and the rough tube quality data of each tube blank, a recursive neural network model is trained to obtain a trained recursive neural network model.
[0112] In a possible implementation, the training module is specifically used to: Calculate the value of the Pearson correlation coefficient between each process parameter and the capillary quality data.
[0113] Process parameters whose Pearson correlation coefficient values are greater than a preset threshold are determined as important parameters.
[0114] In a possible implementation, the recursive neural network model includes an input layer, a hidden layer, a receiving layer and an output layer; before the training module, the device 3 also includes: A model optimization module, used to determine the model parameters of the recursive neural network model based on the artificial bee colony algorithm; wherein the model parameters include the connection weights from the input layer to the hidden layer, the connection weights from the hidden layer to the output layer, the connection weights from the output layer to the receiving layer, the hidden layer threshold and the output layer threshold in the recursive neural network model; The training module is used to train the recursive neural network model according to important parameters of multiple tube blanks during the perforation process, rough tube quality data of each tube blank and model parameters.
[0115] In one possible implementation, the process parameters of the steel pipe perforation process include the upper and lower roller motor current, the upper and lower roller speed, the upper and lower roller torque, the tube billet diameter, the plug diameter, the inlet and outlet cone angles, the roller spacing, the guide plate spacing and the plug extension; the important parameters of the steel pipe perforation process include the plug diameter, the tube billet diameter, the guide plate spacing, the roller spacing and the plug extension.
[0116] In a possible implementation, the capillary quality prediction data includes the predicted outer diameter and predicted wall thickness of the capillary; the optimization module 304 is specifically used for: When the predicted outer diameter and / or predicted wall thickness of the capillary tube does not meet the preset quality requirements, the setting parameters are adjusted based on the difference between the predicted outer diameter and / or predicted wall thickness of the capillary tube and the preset quality requirements to obtain new setting parameters.
[0117] Based on the new set parameters, the steps of screening the importance features of the set parameters to obtain the important parameters of the steel pipe perforation process are repeated until the rough pipe quality prediction data meets the preset quality requirements.
[0118] In a possible implementation, the training module is specifically used to: The important parameters of each tube blank during the piercing process are input into the recursive neural network model, and the rough tube quality prediction data of each tube blank output by the recursive neural network model is obtained.
[0119] According to the rough tube quality data of each tube blank, the rough tube quality prediction data of each tube blank and the loss function, the model parameters are adjusted to obtain a trained recursive neural network model.
[0120] Figure 4 Schematic diagram of an electronic device provided in an embodiment of the present application. Figure 4 As shown, the electronic device 4 of this embodiment includes: a processor 40 and a memory 41. The memory 41 is used to store a computer program 42, and the processor 40 is used to call and run the computer program 42 stored in the memory 41 to execute the steps in the above-mentioned various steel pipe perforation process optimization method embodiments, such as Figure 1 Alternatively, the processor 40 is used to call and run the computer program 42 stored in the memory 41 to implement the functions of each module / unit in the above-mentioned device embodiments, such as Figure 3 Functionality of modules / units 301 to 304 is shown.
[0121] Exemplarily, the computer program 42 may be divided into one or more modules / units, one or more modules / units are stored in the memory 41 and executed by the processor 40 to complete the present application. One or more modules / units may be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program 42 in the electronic device 4. For example, the computer program 42 may be divided into Figure 3 Modules / units 301 to 304 are shown.
[0122] The electronic device 4 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will appreciate that Figure 4 It is only an example of the electronic device 4 and does not constitute a limitation of the electronic device 4. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.
[0123] The processor 40 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0124] The memory 41 may be an internal storage unit of the electronic device 4, such as a hard disk or memory of the electronic device 4. The memory 41 may also be an external storage device of the electronic device 4, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the electronic device 4. The memory 41 is used to store computer programs and other programs and data required by the electronic device. The memory 41 may also be used to temporarily store data that has been output or is to be output.
[0125] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0126] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0127] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0128] In the embodiments provided in the present application, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0129] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0130] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0131] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can be executed by the processor to implement the steps of the above-mentioned various steel pipe perforation process optimization method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.
[0132] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A steel pipe perforation process optimization method, characterized in that: The method comprises: Obtaining setting parameters of the steel pipe perforation process; wherein the setting parameters are the process parameters of the steel pipe perforation process that are initially set; Performing importance feature screening on the set parameters to obtain important parameters of the steel pipe perforation process; The important parameters are input into a trained recursive neural network model to obtain raw tube quality prediction data output by the recursive neural network model; wherein the recursive neural network model is trained based on process parameters of multiple tube blanks during the perforation process and raw tube quality data of each tube blank; According to the capillary quality prediction data and the preset quality requirements, the setting parameters are optimized so that the capillary quality prediction data meets the preset quality requirements.
2. The steel pipe perforation process optimization method according to claim 1, characterized in that: Before inputting the important parameters into the trained recursive neural network model to obtain the capillary quality prediction data output by the recursive neural network model, the method further includes: Obtaining process parameters of multiple tube blanks during the perforation process and rough tube quality data of each tube blank; The process parameters of multiple tube blanks during the perforation process are screened by importance features to obtain the important parameters of multiple tube blanks during the perforation process; wherein the important parameters are process parameters that have a strong correlation with the raw tube quality data, and the strong correlation represents that the value of the correlation is greater than a preset threshold; According to the important parameters of multiple tube blanks during the perforation process and the rough tube quality data of each tube blank, the recursive neural network model is trained to obtain a trained recursive neural network model.
3. The steel pipe perforation process optimization method according to claim 2, characterized in that: The important feature screening of the process parameters of the plurality of tube blanks during the piercing process is performed to obtain the important parameters of the plurality of tube blanks during the piercing process, including: Calculate the value of the Pearson correlation coefficient between each process parameter and the capillary quality data; The process parameters whose Pearson correlation coefficient values are greater than the preset threshold are determined as important parameters.
4. The steel pipe perforation process optimization method according to claim 2, characterized in that: The recursive neural network model includes an input layer, a hidden layer, a receiving layer and an output layer; Before training the recursive neural network model according to important parameters of the plurality of tube blanks during the perforation process and the rough tube quality data of each tube blank, the method further comprises: Based on the artificial bee colony algorithm, the model parameters of the recursive neural network model are determined; wherein the model parameters include the connection weights from the input layer to the hidden layer, the connection weights from the hidden layer to the output layer, the connection weights from the output layer to the receiving layer, the hidden layer threshold and the output layer threshold in the recursive neural network model; The recursive neural network model is trained according to the important parameters of the plurality of tube blanks during the perforation process and the rough tube quality data of each tube blank, including: The recursive neural network model is trained according to important parameters of multiple tube blanks during the perforation process, rough tube quality data of each tube blank and the model parameters.
5. The steel pipe perforation process optimization method according to any one of claims 1 to 4, characterized in that: The process parameters of the steel pipe piercing process include the upper and lower roller motor current, the upper and lower roller speed, the upper and lower roller torque, the tube billet diameter, the plug diameter, the inlet and outlet cone angles, the roller spacing, the guide plate spacing and the plug extension; the important parameters of the steel pipe piercing process include the plug diameter, the tube billet diameter, the guide plate spacing, the roller spacing and the plug extension.
6. The steel pipe perforation process optimization method according to any one of claims 1 to 4, characterized in that: The capillary tube quality prediction data includes the predicted outer diameter and predicted wall thickness of the capillary tube; The optimizing process of the setting parameters according to the capillary quality prediction data and the preset quality requirements includes: When the predicted outer diameter and / or predicted wall thickness of the capillary tube does not meet the preset quality requirement, adjusting the setting parameters based on the difference between the predicted outer diameter and / or predicted wall thickness of the capillary tube and the preset quality requirement to obtain new setting parameters; Based on the new set parameters, the step of screening the set parameters for importance features to obtain important parameters of the steel pipe perforation process is repeated until the rough pipe quality prediction data meets the preset quality requirements.
7. The steel pipe perforation process optimization method according to claim 4, characterized in that: The recursive neural network model is trained according to the important parameters of the plurality of tube blanks during the perforation process, the rough tube quality data of each tube blank and the model parameters, including: Input important parameters of each tube blank during the piercing process into the recursive neural network model, and obtain the rough tube quality prediction data of each tube blank output by the recursive neural network model; According to the rough tube quality data of each tube blank, the rough tube quality prediction data of each tube blank and the loss function, the model parameters are adjusted to obtain a trained recursive neural network model.
8. A steel pipe perforation process optimization device, characterized in that: include: An acquisition module, used to acquire setting parameters of the steel pipe perforation process; wherein the setting parameters are the process parameters of the steel pipe perforation process that are initially set; A screening module is used to screen the set parameters by importance characteristics to obtain important parameters of the steel pipe perforation process; A processing module, used for inputting the important parameters into a trained recursive neural network model to obtain raw tube quality prediction data output by the recursive neural network model; wherein the recursive neural network model is trained based on process parameters of multiple tube blanks during the perforation process and raw tube quality data of each tube blank; The optimization module is used to optimize the set parameters according to the capillary quality prediction data and the preset quality requirements so that the capillary quality prediction data meets the preset quality requirements.
9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute the steel pipe perforation process optimization method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the steel pipe perforation process optimization method as described in any one of claims 1 to 7 are implemented.
Citation Information
Patent Citations
Matlab-based seamless steel tube perforation process model parameter prediction system
CN113486587A
Building component quality prediction method and system based on MIV-BP
CN116128165A
Two-roller cross piercing parameter optimization method based on mechanism data hybrid drive
CN117055487A
Cited By
Perforating parameter setting method and system for cross piercing and tubular billet machining method
CN120421334A
Self-adaptive optimization method and system for pipe blank perforation process parameters
CN120571866A
Oblique piercing quality prediction method and system and tubular billet processing method
CN120670811A
Metal material production process optimization method based on Pearson correlation analysis
CN121303425A
