A Method and Apparatus for Production Control of Stainless Steel Pipes Based on Wear Resistance Prediction

Abstract

This invention provides a method and apparatus for production control of stainless steel pipes based on wear resistance prediction, relating to the field of pipe processing technology. By inputting material engineering data and environmental data to a pre-constructed general wear prediction model, wear simulation during use is performed to obtain a degraded simulated steel pipe and detect N types of abnormal wear. Remedial process analysis is performed on these N abnormal wear types to locate N process optimization nodes and optimize production parameters. The general wear prediction model and wear-related scrap characteristics are used to verify production control and obtain production control parameters. This invention solves the technical problem in existing technologies where the wear resistance of stainless steel pipes is incompatible with the wear conditions of the usage environment, leading to the risk of premature scrapping of stainless steel pipes in special usage scenarios. It achieves the technical effect of improving the wear resistance and adaptability of stainless steel pipes to different usage scenarios, and avoiding premature scrapping of stainless steel pipes due to wear in special usage scenarios.

Description

Technical Field

[0001] This invention relates to the field of pipe processing technology, and specifically to a method and apparatus for controlling the production of stainless steel pipes based on wear resistance prediction. Background Technology

[0002] In existing steel pipe designs, some stainless steel pipes do not adequately consider the wear conditions in actual operating environments, resulting in insufficient wear resistance. For example, in the chemical industry, stainless steel pipes are frequently exposed to highly corrosive environments. If their surface coating is not sufficiently corrosion-resistant, premature corrosion and wear may occur, shortening their service life. Similarly, on offshore platforms, due to the high salinity of seawater and fluctuating load conditions, conventional stainless steel pipes may suffer rapid erosion and wear, leading to fatigue crack formation and increasing the risk of structural failure. These examples illustrate that if the wear resistance of stainless steel pipes does not match the wear characteristics of the operating environment, they may face the risk of premature scrapping in specific application scenarios, causing economic losses and safety hazards.

[0003] In summary, the existing technology has a technical problem where the wear resistance of stainless steel pipes is not compatible with the wear conditions of the usage environment, which leads to the risk of premature scrapping of stainless steel pipes due to wear in special usage scenarios. Summary of the Invention

[0004] This application provides a method and apparatus for production control of stainless steel pipes based on wear resistance prediction, which is used to address the technical problem that the wear resistance performance of stainless steel pipes is not compatible with the wear conditions of the usage environment, resulting in the risk of premature scrapping of stainless steel pipes due to wear in special usage scenarios.

[0005] In view of the above problems, this application provides a method and apparatus for production control of stainless steel pipes based on wear resistance prediction.

[0006] The first aspect of this application provides a method for production control of stainless steel pipes based on wear resistance prediction. The method includes: interactively accessing production logs to obtain material engineering data for the stainless steel pipe to be optimized, wherein the material engineering data includes material property data and surface coating characteristics; obtaining usage environment data, wherein the usage environment data includes usage environment parameters and load condition parameters, obtained by collecting data from the usage environment of the stainless steel pipe to be optimized; interactively obtaining the expected service life and wear scrapping characteristics of the stainless steel pipe to be optimized; using the expected service life as a simulation time constraint, inputting the material engineering data and usage environment data into a pre-constructed general wear prediction model to perform wear simulation during use, obtaining a degraded simulated steel pipe; using the wear scrapping characteristics to detect the degraded simulated steel pipe, obtaining N types of abnormal wear, where N is a positive integer; performing remedial process analysis on the N types of abnormal wear, and matching and locating N process optimization nodes on the production line based on the remedial process analysis results; pre-setting optimization rules to perform production parameter optimization control at the N process optimization nodes, and using the general wear prediction model and wear scrapping characteristics to perform production control verification, obtaining target production control parameters.

[0007] A second aspect of this application provides a stainless steel pipe production control device based on wear resistance prediction. The device includes: an engineering data acquisition unit for interacting with production logs to obtain material engineering data of the stainless steel pipe to be optimized, wherein the material engineering data includes material property data and surface coating characteristics; an environmental data acquisition unit for obtaining usage environment data, wherein the usage environment data includes usage environment parameters and load condition parameters, obtained by collecting data from the usage environment of the stainless steel pipe to be optimized; a design information acquisition unit for interactively obtaining the expected service life and wear and scrap characteristics of the stainless steel pipe to be optimized; and a degradation simulation execution unit for using the expected service life... The lifespan is a simulated time constraint. The material engineering data and the data of the environment to be used are input into a pre-built general wear prediction model to perform wear simulation during use, and obtain a degraded simulated steel pipe. An abnormal wear identification unit is used to detect the degraded simulated steel pipe using the wear scrap characteristics to obtain N types of abnormal wear, where N is a positive integer. An optimization node positioning unit is used to perform remedial process analysis on the N types of abnormal wear, and match and locate N process optimization nodes on the production line based on the remedial process analysis results. A production control optimization unit is used to preset optimization rules to perform production parameter optimization control on the N process optimization nodes, and use the general wear prediction model and wear scrap characteristics to perform production control verification to obtain target production control parameters.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The method provided in this application embodiment obtains material engineering data of the stainless steel pipe to be optimized through interactive production logs. The material engineering data includes material property data and surface coating characteristics. It also obtains usage environment data, including usage environment parameters and load condition parameters, acquired by collecting data from the usage environment of the stainless steel pipe to be optimized. The method interactively obtains the expected service life and wear-related scrap characteristics of the stainless steel pipe to be optimized. Using the expected service life as a simulation time constraint, the material engineering data and usage environment data are input into a pre-constructed general wear prediction model to perform wear simulation during use, obtaining a degraded simulated steel pipe. The wear-related scrap characteristics are used to detect the degraded simulated steel pipe, obtaining N types of abnormal wear, where N is a positive integer. Remedial process analysis is performed on the N types of abnormal wear, and N process optimization nodes are matched and located on the production line based on the remedial process analysis results. Pre-set optimization rules are used to optimize and control production parameters at the N process optimization nodes, and the general wear prediction model and wear-related scrap characteristics are used to verify production control and obtain target production control parameters. By predicting the wear of stainless steel pipes under different usage scenarios, production parameters can be optimized in a targeted manner. This achieves the technical effect of improving the wear resistance and adaptability of stainless steel pipes to different usage scenarios, effectively preventing stainless steel pipes from being scrapped prematurely due to wear in special usage scenarios, ensuring the reliability of stainless steel pipes within their expected service life, and reducing the safety risks caused by failure. Attached Figure Description

[0010] Figure 1 A schematic diagram of the production control method for stainless steel pipes based on wear resistance prediction provided in this application;

[0011] Figure 2 A schematic diagram of the process for obtaining the environmental data to be used in the stainless steel pipe production control method based on wear resistance prediction provided in this application;

[0012] Figure 3 This is a schematic diagram of the stainless steel pipe production control device based on wear resistance prediction provided in this application.

[0013] Explanation of reference numerals in the attached figures: Engineering data acquisition unit 11, Environmental data acquisition unit 12, Design information acquisition unit 13, Degradation simulation execution unit 14, Abnormal wear identification unit 15, Optimization node positioning unit 16, Production control optimization unit 17. Detailed Implementation

[0014] This application provides a method and apparatus for production control of stainless steel pipes based on wear resistance prediction. It addresses the technical problem in existing technologies where the wear resistance of stainless steel pipes is incompatible with the wear conditions of the usage environment, leading to the risk of premature scrapping of stainless steel pipes due to wear in specific application scenarios. By predicting the wear of stainless steel pipes under different usage scenarios, production parameters are optimized in a targeted manner. This improves the wear resistance and adaptability of stainless steel pipes to different application scenarios, effectively preventing premature scrapping due to wear in special application scenarios, ensuring the reliability of stainless steel pipes within their expected service life, and reducing the safety risks caused by failure.

[0015] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. It should also be noted that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of them.

[0016] Example 1, as Figure 1 As shown, this invention provides a method for controlling the production of stainless steel pipes based on wear resistance prediction, the method comprising:

[0017] A100: Interactive production logs, obtaining material engineering data for the stainless steel pipe to be optimized, wherein the material engineering data includes material property data and surface coating characteristics.

[0018] Specifically, in this embodiment, real-time interaction is performed with logs and records generated during the production process to directly obtain the material engineering data of the stainless steel pipe to be optimized. The material engineering data characterizes the metal dimension performance parameters and coating dimension performance parameters of the steel to be optimized obtained from the processing.

[0019] Specifically, the material engineering data includes, but is not limited to, the chemical composition, microstructure, mechanical properties, coefficient of thermal expansion, heat resistance, and corrosion resistance of the stainless steel pipe. The surface coating characteristics include, but are not limited to, coating material type, coating thickness, coating wear resistance, corrosion resistance, and adhesion.

[0020] A200: Obtain the usage environment data, wherein the usage environment data includes usage environment parameters and load condition parameters, and the usage environment data is obtained by collecting data on the usage environment of the stainless steel pipe to be optimized.

[0021] In one embodiment, such as Figure 2As shown, the method obtains the data of the environment to be used, wherein the data of the environment to be used includes the environmental parameters and load condition parameters. The data of the environment to be used is obtained by collecting data on the environmental conditions of the stainless steel pipe to be optimized. Step A200 of the method provided by the present invention further includes:

[0022] A210: Data is collected on the usage environment of the stainless steel pipe to be optimized to obtain multiple historical usage environments and multiple historical load conditions.

[0023] A220: Aggregate environmental indicators for the multiple historical usage environments, and split the multiple historical usage environments based on the aggregation results to obtain M historical environment datasets with M environmental indicators, where M is a positive integer.

[0024] A230: Perform extreme value extraction on the M historical environment datasets to form the usage environment parameters.

[0025] A240: Aggregate load metrics for the multiple historical load conditions, and split the multiple historical load conditions based on the aggregation results to obtain K historical load datasets of K load metrics.

[0026] A250: Perform extreme value extraction on the K historical load datasets to form the load condition parameters, where K is a positive integer.

[0027] Specifically, the purpose of this embodiment is to collect data on the stainless steel pipe to be optimized in actual use, including environmental conditions and load conditions. The specific data acquisition method is to obtain multiple historical usage environments and multiple historical load conditions of the stainless steel pipe at different time points through means such as sensors, log records or database queries.

[0028] Based on the data composition characteristics of each historical usage environment, which includes multiple environmental parameters and various environmental indicators, this implementation aggregates the environmental indicators of the multiple historical usage environments to obtain all environmental monitoring indicators for the usage environment, totaling M environmental indicators.

[0029] The multiple historical usage environments are split using M environmental indicators obtained by aggregation, resulting in M ​​historical environment datasets for the M environmental indicators, where M is a positive integer. Extreme value extraction is performed on the M historical environment datasets to form the usage environment parameters, which characterize the extreme environmental conditions of the usage environment.

[0030] Meanwhile, it should be understood that each historical load condition includes multiple load parameter values ​​of various load indicators. Therefore, this embodiment adopts the same method of obtaining usage environment parameters, aggregates the load indicators of the multiple historical load conditions, splits the multiple historical load conditions based on the aggregation result, obtains K historical load datasets of K load indicators, performs extreme value extraction on the K historical load datasets, and constitutes the load condition parameters, where K is a positive integer, and the load condition parameters are representative load conditions of the usage environment.

[0031] This embodiment fits the extreme environmental conditions and exponential load conditions of the usage environment, achieving the technical effect of providing a reference for subsequent wear resistance prediction of the stainless steel pipe to be optimized.

[0032] A300: Interactively obtain the expected service life and wear and tear characteristics of the stainless steel pipe to be optimized.

[0033] In one embodiment, the wear and tear characteristics include multiple wear degree constraints for various wear types, wherein each wear degree constraint consists of wear depth, crack length, and material loss.

[0034] Specifically, in this embodiment, the expected service life of the stainless steel pipe to be optimized is obtained interactively. The expected service life is the designed service life of the stainless steel pipe to be optimized, for example, 3 years. In actual use, the wear of the stainless steel pipe should not cause the pipe to be scrapped prematurely before the pipe actually reaches its expected service life.

[0035] The wear and tear characteristics are the criteria for judging whether a pipeline is prematurely scrapped. It should be understood that pipelines suffer from various types of wear during use, including but not limited to adhesive wear, abrasive wear, corrosive wear, and fatigue wear. Each wear mechanism leads to different types of premature pipeline scrapping. Based on this, this embodiment sets an acceptable wear range or threshold for each type of wear to obtain multiple wear degree constraints. The wear degree constraints are used to assess whether the current state of the steel pipe is close to or reaches the scrapping standard.

[0036] In the wear degree constraint, wear depth refers to the dimension by which the material surface is reduced due to wear, usually expressed in a certain unit of measurement (such as millimeters).

[0037] It should be understood that the length of a crack in a material is an important indicator for assessing structural integrity. Crack propagation can lead to material failure, so crack length is one of the key factors in determining whether a material is scrapped. Based on this, the crack length is set in this embodiment.

[0038] Meanwhile, the amount of material loss can reflect the severity of wear and tear and serve as a basis for determining whether the material needs to be scrapped. Based on this, the material loss amount is set in this embodiment.

[0039] Meanwhile, this embodiment does not set actual numerical values ​​for the wear degree constraints of each wear type, but the values ​​can be set according to the specific safety requirements of the stainless steel pipe.

[0040] A400: Using the estimated service life as the simulation time constraint, the material engineering data and the environmental data to be used are input into a pre-built general wear prediction model to perform wear simulation during use, and a degradation simulation steel pipe is obtained.

[0041] In one embodiment, method step A400 provided by the present invention further includes:

[0042] A410: Pre-build a general steel pipe model and synchronize the general steel pipe model to the finite element analysis network, wherein the general steel pipe model has a variety of parametric geometric features.

[0043] A420: Perform simulation model matching on the various wear types to obtain multiple wear mechanism prediction models.

[0044] A430: Pre-construct a finite element analysis network, and fit the general steel pipe model and the multiple wear mechanism prediction models to the finite element analysis network to obtain multiple directional wear prediction models for the multiple wear types.

[0045] A440: Connect the multiple directional wear prediction models in parallel to generate the general wear prediction model.

[0046] In one embodiment, using the expected service life as a simulation time constraint, the material engineering data and the environmental data to be used are input into a pre-built general wear prediction model to perform wear simulation during use, thereby obtaining a degradation simulation steel pipe. Step A400 of the method provided by this invention further includes:

[0047] A441: Measure the actual parameters of the stainless steel pipe to be optimized, and input the actual parameters of the steel pipe into multiple directional wear prediction models of the general wear prediction model to adjust the parameters of the general steel pipe model using various parametric geometric features, thereby obtaining multiple real-time steel pipe models.

[0048] A442: A pre-built material database, wherein the material database stores multiple sample engineering data and multiple standard mechanical property data of various sample coated steel pipe materials.

[0049] A443: The material property data and surface coating characteristics are used to traverse and compare the material database to obtain the target mechanical property data.

[0050] A444: Input the target mechanical property data and the environmental data to be used into the multiple directional wear prediction models, and use the expected service life as the simulation time constraint to perform wear simulation of the multiple real-time steel pipe models to obtain multiple directional wear steel pipe models.

[0051] A445: Overlap the wear mechanisms of the multiple directional wear steel pipe models to generate the degraded simulated steel pipe.

[0052] Specifically, in this embodiment, the general wear prediction model can predict the wear and tear of stainless steel pipes under specific usage environments. The construction process of the general wear prediction model is as follows:

[0053] A general steel pipe model with various parametric geometric features can be constructed using CAD software. By adjusting the various parametric geometric features of the general steel pipe model, steel pipe models with different geometric parameters can be obtained.

[0054] Based on the physical mechanism of each wear type, an appropriate mathematical model is selected. For example, the Archard wear model is selected to predict adhesive wear, and a particle-based model is selected to predict abrasive wear. This allows for simulation model matching to be performed on the various wear types, resulting in predictive models for the various wear mechanisms.

[0055] Numerical solutions for various wear mechanism prediction models are obtained using the finite element analysis (FEA) method. These numerical solutions represent the specific wear levels of different wear types on steel pipes under various operating environments. Based on the wear levels, numerical extraction is performed to obtain wear depth, crack length, and material loss.

[0056] Multiple wear mechanism prediction models are integrated into the finite element analysis network to ensure their effectiveness in the simulation process. The general steel pipe model is also integrated into the finite element analysis network to ensure the adjustability of the steel pipe model. A total of multiple directional wear prediction models for directional prediction of various wear types are obtained. The multiple directional wear prediction models are connected in parallel to facilitate the input of environmental data and the design parameters of the stainless steel pipe to be optimized into the multiple directional wear prediction models. The multiple directional wear prediction models connected in parallel constitute the general wear prediction model.

[0057] Based on the general wear prediction model, this embodiment uses the general wear prediction model to perform wear simulation of the stainless steel pipe to be optimized during use, and obtains a degraded simulated steel pipe. The degraded simulated steel pipe is a simulation representation of the surface wear of the stainless steel pipe to be optimized after its expected service life, caused by factors such as wear, corrosion, and fatigue in actual use.

[0058] The specific implementation method for obtaining the degraded simulated steel pipe in this embodiment is as follows:

[0059] The measured parameters of the stainless steel pipe to be optimized are obtained, including but not limited to diameter and wall thickness. These measured parameters are then input into multiple directional wear prediction models of the general wear prediction model to adjust various parametric geometric features of the general steel pipe model, resulting in multiple real-time steel pipe models. These multiple real-time steel pipe models are completely identical and accurately represent the stainless steel pipe to be optimized. The purpose of performing the above process is to obtain multiple real-time steel pipe models that conform to the actual steel pipe, providing an accurate geometric basis for subsequent simulations.

[0060] Specifically, the material engineering data includes, but is not limited to, the chemical composition, microstructure, mechanical properties, coefficient of thermal expansion, heat resistance, and corrosion resistance of the stainless steel pipe. The surface coating characteristics include, but are not limited to, coating material type, coating thickness, coating wear resistance, corrosion resistance, and adhesion.

[0061] Specifically, it should be understood that in practical applications, not all material property data may be readily available. Therefore, this embodiment pre-constructs a material database, which stores multiple sample engineering data and multiple standard mechanical property data for various coated steel pipe materials. The material property data and surface coating characteristics are used to iterate and compare against the material database to obtain the standard mechanical property data with the highest overlap with the surface coating characteristics and material property data. This standard mechanical property data is then used to complete the mechanical performance data of the stainless steel pipe to be optimized in this embodiment, thus obtaining the target mechanical property data. This embodiment constructs a material database for data completion and comparison, achieving the technical effect of obtaining mechanical property data that better matches the stainless steel pipe to be optimized.

[0062] The target mechanical property data obtained by comparing with the material database is input into the directional wear prediction model, which provides the model with detailed characteristics of the steel pipe material, such as hardness, elastic modulus, yield strength, etc. At the same time, the environmental data of the service environment that affects the wear process of the steel pipe, including temperature, humidity, chemical media, etc., are also input. The expected service life is used as the simulation time range to ensure that the simulation covers the wear of the steel pipe throughout the entire expected service life.

[0063] In the multiple directional wear prediction models, wear simulation is performed on each real-time adjusted steel pipe model to take into account the influence of different geometric parameters and material properties on wear. Finally, multiple directional wear steel pipe models that more accurately reflect the wear condition of the stainless steel pipe to be optimized are obtained. Each steel pipe model represents the wear behavior of different wear types under specific conditions.

[0064] It should be understood that in practical applications, the stainless steel pipe to be optimized may be affected by multiple wear mechanisms at the same time, such as frictional wear and corrosion wear. In this implementation, wear mechanism overlap is used to simulate the combined effect of multiple wear mechanisms in actual use.

[0065] Specifically, this embodiment generates a comprehensive structural model of a degradation simulation steel pipe by analyzing and integrating the results of multiple directional wear models. The degradation simulation steel pipe reflects the cumulative effect of all types of wear that the steel pipe may experience during actual use.

[0066] This embodiment simulates the wear of steel pipes by comprehensively considering measured parameters, material property data, and environmental data. It achieves the technical effect of quickly obtaining a degradation simulation steel pipe that can more accurately reflect the wear under actual use conditions, providing a reference for subsequent steel pipe production control.

[0067] A500: The wear and scrapping characteristics are used to detect the degraded simulated steel pipe to obtain N types of abnormal wear, where N is a positive integer.

[0068] A600: Perform remedial process analysis on the N types of abnormal wear, and match and locate N process optimization nodes on the production line based on the remedial process analysis results.

[0069] Specifically, in this embodiment, by using wear and scrap characteristics, the degraded simulated steel pipe is systematically detected to identify any possible abnormal wear conditions. The simulated degraded steel pipe is compared with predefined wear and scrap characteristics, which may include the maximum allowable wear depth, crack length, and material loss. Finally, N abnormal wear conditions that do not meet the corresponding wear and scrap characteristics are identified. In other words, the stainless steel pipe to be optimized will be scrapped prematurely due to these N abnormal wear conditions before reaching its expected service life.

[0070] Based on this, this embodiment performs remedial process analysis on the detected abnormal wear to determine how and by what remedial measures can these wear problems be mitigated or prevented, such as changing materials or adjusting coating processes.

[0071] Based on the results of the remedial process analysis, N process optimization nodes are identified on the production line. These optimization nodes may involve key production steps such as raw material selection, forming process, heat treatment, and coating application. Adjusting the production parameters of these N optimization nodes can optimize and improve the quality and performance of the steel pipe, so that the wear degree of N abnormal wear will not cause premature scrapping of the stainless steel pipe to be optimized before reaching its expected service life.

[0072] A700: The preset optimization rules are used to optimize and control production parameters at the N process optimization nodes, and the general wear prediction model and wear scrapping characteristics are used to verify the production control and obtain the target production control parameters.

[0073] In one embodiment, a preset optimization rule is used to optimize and control production parameters at the N process optimization nodes, and the general wear prediction model and wear and scrap characteristics are used to verify the production control and obtain the target production control parameters. Step A700 of the method provided by this invention further includes:

[0074] A710: The preset optimization rules include optimization vector, optimization parameter tuning step size, and optimization parameter tuning variables.

[0075] A720: Interactively obtains multiple sample production control parameters, multiple sample steel pipe geometric parameters, and multiple sample steel pipe engineering data.

[0076] A730: An engineering data prediction model is generated by training multiple sample production control parameters, multiple sample steel pipe geometric parameters, and multiple sample steel pipe engineering data.

[0077] A740: Interactively obtain the original production control parameters of the stainless steel pipe to be optimized.

[0078] A750: Using the optimization vector and optimization step size as constraints, and starting from the original production control parameters, perform processing parameter control optimization on the N process optimization nodes, and use the engineering data prediction model and general wear prediction model to perform wear simulation in the usage process until the target production control parameters that result in zero abnormal wear are obtained.

[0079] In one embodiment, using the optimization vector and optimization parameter tuning step size as constraints, and starting from the original production control parameters, the processing parameter tuning optimization is performed on the N process optimization nodes. The engineering data prediction model and the general wear prediction model are used to perform wear simulation during use until the target production control parameters that reduce abnormal wear to zero are obtained. Step A750 of the method provided by this invention further includes:

[0080] A751: Using the optimization vector and optimization parameter tuning step size as constraints, process parameters are adjusted for the N process optimization nodes to obtain N first update control parameters, and the original production control parameters are updated using the N first update control parameters to obtain the first production control parameters.

[0081] A752: Input the first production control parameter into the engineering data prediction model to obtain the first prediction geometric parameter and the first prediction engineering data.

[0082] A753: The first predicted geometric parameters are used to adjust the various parametric geometric features of the general wear prediction model for the general steel pipe model, and the expected service life is used as the simulation time constraint. The first predicted engineering data and the environmental data to be used are input into the general wear prediction model to perform wear simulation during use, and the first predicted simulated steel pipe is obtained.

[0083] A754: The first predicted simulated steel pipe is detected using the wear and scrap characteristics mentioned above to obtain E types of abnormal wear, where E is a positive integer less than or equal to N.

[0084] A755: The wear repair index is calculated based on the E types of abnormal wear and the N types of abnormal wear, and the optimization step size update coefficient is calculated based on the wear repair index and the optimization parameter adjustment variable.

[0085] A756: The optimization step size is adjusted by using the optimization step size update coefficient to obtain the first update step size.

[0086] A757: Using the optimization vector and the first update step size as constraints, process parameter adjustment is performed on the N process optimization nodes to obtain N second update control parameters.

[0087] A758: Based on the N second update control parameters, engineering data prediction model and general wear prediction model, perform wear simulation during use to obtain the second predicted simulated steel pipe.

[0088] A759: The second predicted simulated steel pipe is detected using the wear and scrap characteristics mentioned above to obtain G types of abnormal wear, where G is a positive integer less than or equal to N.

[0089] A7510: And so on, until the abnormal wear detected by the wear and scrap characteristics is reduced to zero, and the target production control parameters are obtained.

[0090] Specifically, in this embodiment, an engineering data prediction model is pre-constructed. This model can directly predict the material engineering data and geometric parameters of the steel pipes produced based on the production control parameters of the production process.

[0091] The method for constructing the engineering data prediction model is as follows:

[0092] The system interactively obtains multiple sample production control parameters, multiple sample steel pipe geometric parameters, and multiple sample steel pipe engineering data from the same production line as the stainless steel pipe to be produced. Each sample production control parameter includes control parameters for multiple production nodes (process steps) throughout the entire steel pipe production process, and each sample steel pipe engineering data includes sample material property data and sample surface coating characteristics.

[0093] The engineering data prediction model is constructed based on a backpropagation neural network. The input data of the engineering data prediction model is the sample production control parameters, and the output results include the sample steel pipe geometric parameters and sample steel pipe engineering data.

[0094] This embodiment employs the conventional training method of backpropagation neural networks, using the multiple sample production control parameters, multiple sample steel pipe geometric parameters, and multiple sample steel pipe engineering data as training data to train and generate an engineering data prediction model. The value of constructing the engineering data prediction model in this embodiment lies in quickly fitting and outputting the geometric parameters and engineering data of the produced steel pipe after obtaining the production control parameters. The specific use of the engineering data prediction model will be described in detail in the subsequent description.

[0095] In this embodiment, the preset optimization rules define the key elements in the optimization process, including the optimization vector, the optimization parameter tuning step size, and the optimization parameter tuning variable. The optimization vector is a parameter that guides the optimization direction. The optimization parameter tuning step size is the adjustable amount of the equipment control parameter in the production node in a single operation. For example, the optimization parameter tuning step size is 1 / 20 of the adjustable range of the equipment control parameter. The optimization parameter tuning variable is the variable amount of the optimization parameter tuning step size.

[0096] The original production control parameters of the stainless steel pipe to be optimized are obtained interactively. The original production control parameters are the control parameters of the stainless steel pipe to be optimized at multiple production nodes (process steps) in the entire production process.

[0097] Using the optimization vector and optimization parameter tuning step size as constraints, the processing parameters of the N process optimization nodes are adjusted to obtain N first update control parameters, and the original production control parameters are updated using the N first update control parameters to obtain the first production control parameters.

[0098] The first production control parameter is input into the engineering data prediction model to obtain the first predicted geometric parameter and the first predicted engineering data. The first predicted engineering data is then input into the material database to update the mechanical properties and obtain the first completed engineering data.

[0099] Using the same method as described above for obtaining the degraded simulated steel pipe, the first predicted geometric parameters are used to adjust the various parametric geometric features of the general wear prediction model for the general steel pipe model. With the expected service life as the simulation time constraint, the first predicted engineering data and the environmental data to be used are input into the general wear prediction model to perform wear simulation during use, thereby obtaining the first predicted simulated steel pipe.

[0100] By employing the method for determining that the N types of abnormal wear are the same, and using the wear scrapping characteristics to detect the first predicted simulated steel pipe, E types of abnormal wear are obtained, where E is a positive integer less than or equal to N.

[0101] Wear types are counted and extracted based on the E types of abnormal wear and N types of abnormal wear. Then, the wear repair index is calculated by using E as the numerator and E+N as the denominator.

[0102] Determine whether the wear repair index is less than the optimization parameter tuning variable. If it is less, subtract the wear repair index from 1 as the optimization step size update coefficient. Multiply the optimization step size update coefficient by the optimization parameter tuning step size as the first update step size.

[0103] Conversely, if the wear repair index is greater than or equal to the optimization parameter tuning variable, then the wear repair index is 1, meaning that the optimization parameter tuning step size is not adjusted.

[0104] Using the optimization vector and the first update step size as constraints, the processing parameters of the N process optimization nodes are adjusted to obtain N second update control parameters. Based on the N second update control parameters, the engineering data prediction model and the general wear prediction model, the wear simulation of the usage process is performed to obtain the second predicted simulated steel pipe. The wear scrapping characteristics are used to detect the second predicted simulated steel pipe to obtain G types of abnormal wear, where G is a positive integer less than or equal to N.

[0105] Next, the step size of the optimization parameter adjustment is continuously adjusted based on the abnormal wear of the two predicted simulated steel pipes until the abnormal wear of the predicted simulated steel pipe obtained by the wear scrapping feature detection is zero. The N updated control parameters obtained at this time are used to update the original production control parameters to obtain the production control parameters, which are used as the target production control parameters.

[0106] This embodiment achieves the technical effect of quickly obtaining steel pipe processing parameters that can eliminate abnormal wear that may lead to premature scrapping by using engineering data prediction models and general wear prediction models to guide the optimization of production parameters in an automated iterative optimization process.

[0107] This embodiment predicts the wear of stainless steel pipes under different usage scenarios and optimizes production parameters accordingly. This improves the wear resistance and adaptability of stainless steel pipes to different usage scenarios, effectively prevents them from being scrapped prematurely due to wear in special usage scenarios, ensures their reliability within the expected service life, and reduces the safety risks caused by failure.

[0108] Example 2 is based on the same inventive concept as the stainless steel pipe production control method based on wear resistance prediction in the previous examples, such as... Figure 3 As shown, the present invention provides a stainless steel pipe production control device based on wear resistance prediction, wherein the device includes:

[0109] The engineering data acquisition unit 11 is used to interact with the production log and obtain the material engineering data of the stainless steel pipe to be optimized, wherein the material engineering data includes material property data and surface coating characteristics.

[0110] The environmental data acquisition unit 12 is used to obtain the environmental data to be used, wherein the environmental data to be used includes usage environment parameters and load condition parameters, and the environmental data to be used is obtained by collecting data on the usage environment of the stainless steel pipe to be optimized.

[0111] The design information acquisition unit 13 is used to interactively obtain the expected service life and wear and scrap characteristics of the stainless steel pipe to be optimized.

[0112] The degradation simulation execution unit 14 is used to input the material engineering data and the environmental data to be used into a pre-built general wear prediction model with the expected service life as the simulation time constraint to perform wear simulation during use and obtain a degradation simulation steel pipe.

[0113] The abnormal wear identification unit 15 is used to detect the degraded simulated steel pipe using the wear and scrap characteristics to obtain N types of abnormal wear, where N is a positive integer.

[0114] The optimized node positioning unit 16 is used to perform remedial process analysis on the N types of abnormal wear, and to match and position N process optimization nodes on the production line based on the remedial process analysis results.

[0115] The production control optimization unit 17 is used to preset optimization rules to perform production parameter optimization and control at the N process optimization nodes, and to use the general wear prediction model and wear scrapping characteristics to verify the production control and obtain the target production control parameters.

[0116] In one embodiment, the environmental data acquisition unit 12 further includes:

[0117] Data is collected on the usage environment of the stainless steel pipe to be optimized to obtain multiple historical usage environments and multiple historical load conditions. Environmental indicators are aggregated for the multiple historical usage environments, and the multiple historical usage environments are split based on the aggregation results to obtain M historical environment datasets with M environmental indicators, where M is a positive integer. Extreme value extraction is performed on the M historical environment datasets to form the usage environment parameters. Load indicators are aggregated for the multiple historical load conditions, and the multiple historical load conditions are split based on the aggregation results to obtain K historical load datasets with K load indicators. Extreme value extraction is performed on the K historical load datasets to form the load condition parameters, where K is a positive integer.

[0118] In one embodiment, the production control optimization unit 17 further includes:

[0119] The wear and tear characteristics include multiple wear degree constraints for various wear types, wherein each wear degree constraint consists of wear depth, crack length and material loss.

[0120] In one embodiment, the degradation simulation execution unit 14 further includes:

[0121] A general steel pipe model is pre-constructed and synchronized to a finite element analysis network. The general steel pipe model has multiple parametric geometric features. Simulation model matching is performed on the multiple wear types to obtain multiple wear mechanism prediction models. A finite element analysis network is pre-constructed, and the general steel pipe model and the multiple wear mechanism prediction models are respectively fitted to the finite element analysis network to obtain multiple directional wear prediction models for the multiple wear types. The multiple directional wear prediction models are connected in parallel to generate the general wear prediction model.

[0122] In one embodiment, the degradation simulation execution unit 14 further includes:

[0123] The measured parameters of the stainless steel pipe to be optimized are obtained by measuring and inputting the measured parameters into multiple directional wear prediction models of the general wear prediction model to adjust the parameterized geometric features of the general steel pipe model, thereby obtaining multiple real-time steel pipe models. A material database is pre-constructed, which stores multiple sample engineering data and multiple standard mechanical property data of various sample coated steel pipe materials. The material property data and surface coating characteristics are used to traverse and compare the material database to obtain target mechanical property data. The target mechanical property data and the data of the environment to be used are input into the multiple directional wear prediction models, and the wear simulation of the multiple real-time steel pipe models is performed with the expected service life as the simulation time constraint to obtain multiple directional wear steel pipe models. The wear mechanism is overlapped among the multiple directional wear steel pipe models to generate the degradation simulation steel pipe.

[0124] In one embodiment, the production control optimization unit 17 further includes:

[0125] The preset optimization rules include an optimization vector, an optimization parameter adjustment step size, and optimization parameter adjustment variables; multiple sample production control parameters, multiple sample steel pipe geometric parameters, and multiple sample steel pipe engineering data are interactively obtained; an engineering data prediction model is trained using the multiple sample production control parameters, multiple sample steel pipe geometric parameters, and multiple sample steel pipe engineering data; the original production control parameters of the stainless steel pipe to be optimized are interactively obtained; with the optimization vector and optimization parameter adjustment step size as constraints, and with the original production control parameters as the starting point, the processing parameter control optimization is performed on the N process optimization nodes; the wear simulation during use is performed using the engineering data prediction model and the general wear prediction model until the target production control parameters at which abnormal wear returns to zero are obtained.

[0126] In one embodiment, the production control optimization unit 17 further includes:

[0127] Using the optimization vector and optimization parameter tuning step size as constraints, processing parameters are adjusted for the N process optimization nodes to obtain N first update control parameters. These N first update control parameters are then used to update the original production control parameters, resulting in first production control parameters. The first production control parameters are input into the engineering data prediction model to obtain first predicted geometric parameters and first predicted engineering data. The first predicted geometric parameters are used to adjust the parameters of the general wear prediction model using various parametric geometric features of the general steel pipe model. With the expected service life as the simulation time constraint, the first predicted engineering data and the environment data to be used are input into the general wear prediction model to perform wear simulation during use, resulting in a first predicted simulated steel pipe. The wear and scrapping characteristics are used to detect the first predicted simulated steel pipe, obtaining E types of abnormal wear, where E is... A positive integer less than or equal to N; calculate the wear repair index based on the E types of abnormal wear and N types of abnormal wear, and calculate the optimization step size update coefficient based on the wear repair index and the optimization parameter adjustment variable; adjust the optimization parameter adjustment step size using the optimization step size update coefficient to obtain the first update step size; use the optimization vector and the first update step size as constraints to perform processing parameter control on the N process optimization nodes to obtain N second update control parameters; perform usage process wear simulation based on the N second update control parameters, engineering data prediction model and general wear prediction model to obtain the second predicted simulated steel pipe; use the wear scrapping feature to detect the second predicted simulated steel pipe to obtain G types of abnormal wear, where G is a positive integer less than or equal to N; and so on, until the abnormal wear obtained by using the wear scrapping feature detection returns to zero, to obtain the target production control parameter.

[0128] In summary, any of the methods or steps described above can be stored as computer instructions or programs in various types of computer memory, and the computer instructions or programs can be recognized by various types of computer processors to implement any of the above methods or steps.

[0129] Based on the above specific embodiments of the present invention, any improvements and modifications made to the present invention by those skilled in the art without departing from the principle of the present invention shall fall within the patent protection scope of the present invention.

Claims

1. A method for production control of stainless steel pipes based on wear resistance prediction, characterized in that, The method includes: Interact with production logs to obtain material engineering data for the stainless steel pipe to be optimized, wherein the material engineering data includes material property data and surface coating characteristics; Obtain the usage environment data, wherein the usage environment data includes usage environment parameters and load condition parameters, and the usage environment data is obtained by collecting data on the usage environment of the stainless steel pipe to be optimized; The expected service life and wear and scrap characteristics of the stainless steel pipe to be optimized are obtained interactively. Using the expected service life as the simulation time constraint, the material engineering data and the data of the environment to be used are input into a pre-built general wear prediction model to perform wear simulation during use, thereby obtaining a degraded simulated steel pipe; The wear and scrapping characteristics are used to detect the degraded simulated steel pipe, and N types of abnormal wear are obtained, where N is a positive integer; Remedial process analysis is performed on the N types of abnormal wear, and N process optimization nodes are matched and located on the production line based on the results of the remedial process analysis. The preset optimization rules are used to optimize and control production parameters at the N process optimization nodes, and the general wear prediction model and wear scrapping characteristics are used to verify the production control and obtain the target production control parameters. The preset optimization rules are used to optimize and control production parameters at the N process optimization nodes, and the general wear prediction model and wear and scrap characteristics are used to verify the production control, thereby obtaining the target production control parameters, including: The preset optimization rules include the optimization vector, the optimization parameter tuning step size, and the optimization parameter tuning variables; Interactively obtain multiple sample production control parameters, multiple sample steel pipe geometric parameters, and multiple sample steel pipe engineering data; An engineering data prediction model is trained using the multiple sample production control parameters, multiple sample steel pipe geometric parameters, and multiple sample steel pipe engineering data. The original production control parameters of the stainless steel pipe to be optimized are obtained interactively. Using the optimization vector and optimization parameter tuning step size as constraints, and starting from the original production control parameters, the processing parameter control optimization is performed on the N process optimization nodes. The engineering data prediction model and the general wear prediction model are used to perform wear simulation in the usage process until the target production control parameters that result in zero abnormal wear are obtained. Using the optimization vector and optimization step size as constraints, and starting from the original production control parameters, the processing parameters are optimized for the N process optimization nodes. The engineering data prediction model and the general wear prediction model are used to perform wear simulation during use, until the target production control parameters for zeroing abnormal wear are obtained, including: Using the optimization vector and the optimization parameter tuning step size as constraints, the processing parameters of the N process optimization nodes are adjusted to obtain N first updated control parameters, and the original production control parameters are updated using the N first updated control parameters to obtain the first production control parameters. The first production control parameter is input into the engineering data prediction model to obtain the first prediction geometric parameter and the first prediction engineering data. The first predicted geometric parameters are used to adjust the parameters of the general wear prediction model for various parametric geometric features of the general steel pipe model. The expected service life is used as the simulation time constraint. The first predicted engineering data and the environmental data to be used are input into the general wear prediction model to perform wear simulation during use, and the first predicted simulated steel pipe is obtained. The wear and scrapping characteristics are used to detect the first predicted simulated steel pipe to obtain E types of abnormal wear, where E is a positive integer less than or equal to N; The wear repair index is calculated based on the E types of abnormal wear and the N types of abnormal wear, and the optimization step size update coefficient is calculated based on the wear repair index and the optimization parameter adjustment variable. The optimization step size is adjusted by using the optimization step size update coefficient to obtain the first update step size; Using the optimization vector and the first update step size as constraints, the processing parameters of the N process optimization nodes are adjusted to obtain N second update control parameters; Based on the N second updated control parameters, the engineering data prediction model and the general wear prediction model, a wear simulation of the usage process is performed to obtain the second predicted simulated steel pipe; The wear and scrapping characteristics are used to detect the second predicted simulated steel pipe to obtain G types of abnormal wear, where G is a positive integer less than or equal to N; This process continues until the abnormal wear detected by the wear and tear feature is reduced to zero, thus obtaining the target production control parameters.

2. The stainless steel pipe production control method based on wear resistance prediction as described in claim 1, characterized in that, Obtaining the usage environment data, wherein the usage environment data includes usage environment parameters and load condition parameters, the usage environment data is obtained by collecting data on the usage environment of the stainless steel pipe to be optimized, the method further includes: Data was collected on the usage environment of the stainless steel pipe to be optimized to obtain multiple historical usage environments and multiple historical load conditions. The environmental indicators of the multiple historical usage environments are aggregated, and the multiple historical usage environments are split based on the aggregation results to obtain M historical environment datasets with M environmental indicators, where M is a positive integer; Extreme value extraction is performed on the M historical environmental datasets to form the usage environment parameters; The load metrics of the multiple historical load conditions are aggregated, and the multiple historical load conditions are split based on the aggregation results to obtain K historical load datasets of K load metrics. Extreme value extraction is performed on the K historical load datasets to form the load condition parameters, where K is a positive integer.

3. The stainless steel pipe production control method based on wear resistance prediction as described in claim 1, characterized in that, The wear and tear characteristics include multiple wear degree constraints for various wear types, wherein each wear degree constraint consists of wear depth, crack length and material loss.

4. The stainless steel pipe production control method based on wear resistance prediction as described in claim 3, characterized in that, The method further includes: A general steel pipe model is pre-constructed and synchronized to the finite element analysis network. The general steel pipe model has multiple parametric geometric features. Simulation model matching is performed on the various wear types to obtain multiple wear mechanism prediction models; A finite element analysis network is pre-constructed, and the general steel pipe model and the multiple wear mechanism prediction models are respectively fitted to the finite element analysis network to obtain multiple directional wear prediction models for the multiple wear types. The multiple directional wear prediction models are connected in parallel to generate the general wear prediction model.

5. The stainless steel pipe production control method based on wear resistance prediction as described in claim 4, characterized in that, Using the estimated service life as a simulation time constraint, the material engineering data and the environmental data to be used are input into a pre-built general wear prediction model to perform wear simulation during use, thereby obtaining a degradation simulation steel pipe. The method further includes: The measured parameters of the stainless steel pipe to be optimized are obtained by measuring and inputting the measured parameters into multiple directional wear prediction models of the general wear prediction model to adjust the parameters of the general steel pipe model by various parametric geometric features, thereby obtaining multiple real-time steel pipe models; A pre-built material database is provided, wherein the material database stores multiple sample engineering data and multiple standard mechanical property data of various sample coated steel pipe materials; The target mechanical property data is obtained by traversing and comparing the material property data and surface coating characteristics with the material database. The target mechanical property data and the environmental data to be used are input into the multiple directional wear prediction models. The wear simulation of the multiple real-time steel pipe models is carried out with the expected service life as the simulation time constraint to obtain multiple directional wear steel pipe models. The wear mechanism of the multiple directional wear steel pipe models is overlapped to generate the degradation simulation steel pipe.

6. A stainless steel pipe production control device based on wear resistance prediction, characterized in that, The apparatus for implementing the stainless steel pipe production control method based on wear resistance prediction as described in any one of claims 1 to 5, the apparatus comprising: An engineering data acquisition unit is used to interact with production logs and obtain material engineering data of the stainless steel pipe to be optimized, wherein the material engineering data includes material property data and surface coating characteristics; An environmental data acquisition unit is used to obtain environmental data to be used, wherein the environmental data to be used includes usage environment parameters and load condition parameters, and the environmental data to be used is obtained by collecting data on the usage environment of the stainless steel pipe to be optimized. The design information acquisition unit is used to interactively obtain the expected service life and wear and scrap characteristics of the stainless steel pipe to be optimized; The degradation simulation execution unit is used to input the material engineering data and the data of the environment to be used into a pre-built general wear prediction model to perform wear simulation during use, with the expected service life as the simulation time constraint, and to obtain the degradation simulation steel pipe. An abnormal wear identification unit is used to detect the degraded simulated steel pipe using the wear and scrapping characteristics, and obtain N types of abnormal wear, where N is a positive integer; An optimized node positioning unit is used to perform remedial process analysis on the N types of abnormal wear, and to match and position N process optimization nodes on the production line based on the remedial process analysis results. The production control optimization unit is used to perform production parameter optimization and control at the N process optimization nodes by preset optimization rules, and to perform production control verification using the general wear prediction model and wear scrap characteristics to obtain target production control parameters.

Images