Intelligent management method and system for cast pipe production line

Abstract

The invention discloses an intelligent management method and system for a cast pipe production line, and relates to the technical field of cast pipe production, in the trial-manufacturing process, the method identifies the sequence of solidification areas under different rotation conditions by setting different rotation conditions and combining a CFD simulation technology, and analyzes the component difference and the distribution condition of solute elements in different areas. By obtaining the composition segregation coefficient Pxs, distribution of solute elements in the cast pipe production process can be optimized, the composition uniformity of the cast pipe is further improved, then the mechanical property and structural reliability of the cast pipe are enhanced, and the defect risk caused by uneven composition is reduced. In the cast pipe production process, the feeding timeliness in the molten metal solidification process is analyzed, technological data are combined, the hole defect risk coefficient Kfx is constructed, the foundation is laid for effectively reducing the occurrence rate of hole defect defects, and the production conditions of centrifugal casting are optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of cast pipe production, and particularly to an intelligent management method and system for a cast pipe production line. Background Art

[0002] The intelligent management of a cast pipe production line belongs to the field of modern manufacturing, and is particularly widely applied in the metal casting and machinery manufacturing industries. With the promotion of Industry 4.0, the application of intelligent and automated technologies in the casting process has been continuously deepened, especially for the management and optimization of cast pipe production lines. In the process of cast pipe production, the centrifugal casting technology has been widely applied due to its high efficiency and precision. In particular, the accurate analysis of the solid-liquid state in the mold and the control of the composition distribution in the solid-liquid state are crucial for improving the overall quality of the cast pipe.

[0003] Although certain progress has been made in the current intelligent management technology of cast pipe production lines, there are still certain problems and deficiencies in specific applications. During the centrifugal casting process, since the change in the rotation speed directly affects the solid-liquid state in the mold, and thus affects the composition distribution and mechanical properties of the cast pipe. Traditional casting processes mostly rely on experience and single parameter settings, and it is difficult to achieve precise control of the solid-liquid state. Especially under different rotation speeds, there are significant differences and uncertainties in the changes in the solidification zone, composition segregation phenomenon, and feeding problems. Moreover, the existing process monitoring and simulation systems fail to provide real-time feedback on the temperature field and composition distribution during the solidification process, resulting in the production of cast pipes not under suitable production conditions, thus easily causing uneven mechanical properties of the castings and untimely feeding, affecting the final quality of the cast pipe. Especially in high-demand casting products, these factors have a greater impact on the final quality of the cast pipe. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides an intelligent management method and system for a cast pipe production line, which solves the problems in the above background art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent management method for a cast pipe production line includes the following steps. S1: Extract relevant process data in the production line according to the cast pipe production line, and determine the mold making according to the casting requirements. S2: Based on the determination of the mold making, conduct conditional trial production for the cast pipe production. During the trial production process, set different rotation conditions, combine with CFD simulation, identify the order of the solidification zones under different rotation conditions, and according to the identification results under different rotation conditions, analyze the differences in the composition and the distribution of solute elements in different solidification zones to obtain the composition segregation coefficient Pxs. S3: During the solidification process of the molten metal, analyze the feeding timeliness of the cast pipe production line, and combine relevant process data to construct the pore defect risk coefficient Kfx. Use deep learning technology to construct a prediction model with minimized loss, and perform fitting to output the production selection conditions for centrifugal casting.

[0006] Preferably, the specific steps of S1 include: S11: Deploy multiple groups of sensors and data acquisition devices on the cast pipe production line to extract relevant process data in the production line in real time. The relevant process data includes the operation information of the production line equipment, process parameters, and material property information. Among them, the operation information of the production line equipment includes the furnace temperature in the cast pipe production line, the rotation speed of the centrifuge, and the current and voltage of each production device; the process parameters include the initial batching ratio of the metal material for casting the cast pipe, the initial batching content , the width of the solidification temperature range , the initial liquid volume TJ, the pouring temperature of the metal material, the cooling rate, and the supplementary volume; the material property information includes the density, solidus temperature , viscosity, and thermal conductivity of the metal material for casting the cast pipe; S12: According to the casting requirements, make the corresponding mold and obtain the geometric model of the mold. Among them, the casting requirements include the specifications and shapes of the cast pipes to be cast, and select the mold type according to the enterprise's production batch requirements, production cycle plan, and production cost control. The mold types include metal molds and sand molds.

[0007] Preferably, the specific steps of S2 include: S21: Based on the CFD simulation technology, input the mold geometric model, production line equipment operation information, and material property information in S12 into the CFD software, set the process parameters as boundary conditions, and set different rotation conditions to conduct conditional trial production for the cast pipe production. During the trial production process, identify the solidification area sequence under different rotation conditions. The specific identification steps are as follows: S211: According to the numerical simulation technology, use the finite difference method to discretize time and space, and numerically solve the heat conduction equation to obtain the temperature field T at each position in the mold under different rotation conditions; S212: Based on the temperature field T in each area of the mold, combine the material property information to obtain the solidification index at each position in the mold under different rotation conditions , specifically: ; In the formula, represents the solidification index at position r in the mold at the t-th time point, represents the unit step function. If , then = 1, otherwise = 0; represents the current temperature value at position r within the mold, represents the solidus temperature, represents the adjustment parameter, with a value range of ; represents the temperature gradient at position r within the mold.

[0008] Preferably, the specific identification steps further include: S213: Based on the solidification indexes at various positions within the mold under different rotation conditions , set a solidification threshold, and by comparing the solidification threshold with the solidification indexes at various positions under different rotation conditions , identify the order of solidification regions under different rotation conditions: S2131: If the solidification index at the corresponding position under the corresponding rotation condition exceeds the solidification threshold, it is determined that the corresponding position under the corresponding rotation condition is in a solidified state. Through statistics, the solidification region under the corresponding rotation condition is obtained, and the solidification region is marked as the pre-solidification region; S2132: If the solidification index at the corresponding position under the corresponding rotation condition does not exceed the solidification threshold, it is determined that the corresponding position under the corresponding rotation condition is not in a solidified state. Through statistics, the non-solidified region under the corresponding rotation condition is obtained, and the non-solidified region is marked as the post-solidification region.

[0009] Preferably, the specific steps of S2 further include: S22: Based on the pre-solidification region and post-solidification region identified in S21, analyze the relative contents of different melting point phases in the pre-solidification region and post-solidification region at different solidification stages within each rotation condition, so as to obtain the mass fraction of the high melting point phase in the pre-solidification region and the mass fraction of the high melting point phase in the post-solidification region , specifically: In the formula, represents the content of the high melting point phase in the pre-solidification region at solidification stage d, represents the content of the high melting point phase in the post-solidification region at solidification stage d, represents the initial ingredient content of the metal material, represents the mass fraction of the high melting point phase in the pre-solidification region at solidification stage d, represents the mass fraction of the high melting point phase in the post-solidification region at solidification stage d; S23: The mass fraction of the high melting point phase in the pre-solidification region and the mass fraction of the high melting point phase in the post-solidification region Perform a subtraction operation to obtain the mass fraction difference , specifically: S24: Based on the way of obtaining the mass fraction difference , extract the maximum mass fraction difference from multiple solidification stages under the corresponding rotation conditions as the component distribution difference index between the pre-solidified region and the post-solidified region , which is used to measure the mechanical property uniformity of different positions of the cast pipe.

[0010] Preferably, the specific steps of S2 further include: S25: Based on the pre-solidified region and the post-solidified region identified in S21, and combined with the phase diagram principle, analyze the distribution relationship of solute elements in the solid phase and the liquid phase in the mold under different rotation conditions to obtain the distribution coefficient Fx under the corresponding rotation conditions, specifically: In the formula, represents the solute concentration in the solid phase, represents the solute concentration in the liquid phase; S26: Based on the component distribution difference index between the pre-solidified region and the post-solidified region obtained in S24 and S25 and the distribution coefficient Fx under the corresponding rotation conditions, obtain the component segregation coefficient Pxs through dimensionless normalization, specifically: In the formula, and are both weight values, and the specific values are set by the user according to the situation, 0 < < 1, 0 < < 1.

[0011] Preferably, the specific steps of S3 include: S31: Based on the material handbook and database, query the solidification shrinkage rate data of the corresponding metal material, and extract the initial liquid volume TJ in the process parameters. During the solidification process of the molten metal, collect the actual supplementary molten metal volume Vb at the riser, analyze the feeding timeliness of the cast pipe production line to generate the timeliness function under different rotation conditions, specifically: In the formula, represents the solidification shrinkage volume. If the actual supplementary molten metal volume Vb < the solidification shrinkage volume , then = 1, indicating that the feeding of the current cast pipe production line is in an untimely state; if the actual supplementary molten metal volume Vb ≥ the solidification shrinkage volume , then = 0, indicating that the feeding of the current cast pipe production line is in a timely state; S32: Before the trial production, collect the supplementary data in the corresponding molds on the cast pipe production line under different rotation conditions. The supplementary data includes the riser content Hm in the corresponding molds on the cast pipe production line under different rotation conditions; S33: Obtain the pore defect risk coefficient Kfx, specifically: In the formula, represents the width of the solidification temperature range, , and are all weight values.

[0012] Preferably, the specific steps of S3 further include: S34: Take the pore defect risk coefficient Kfx and the composition segregation coefficient Pxs as the input values of the loss minimization prediction model. After dimensionless processing, fit and output the failure evaluation index DPzs under the corresponding rotation conditions, specifically: ; In the formula, and are both influence coefficients, used to adjust the influence intensity of pore defects and composition segregation on the failure evaluation. e represents the Euler number.

[0013] Preferably, the specific steps of S3 further include: S35: Based on the failure evaluation index DPzs under the corresponding rotation conditions obtained in S34, after feature extraction, extract the failure evaluation index DPzs with the smallest value, and use the rotation conditions corresponding to the failure evaluation index DPzs with the smallest value as the production conditions for centrifugal casting.

[0014] An intelligent management system for a cast pipe production line includes a preparation subsystem, a trial production subsystem, and a production management subsystem; The preparation subsystem extracts relevant process data in the production line according to the cast pipe production line, and determines the mold making according to the casting requirements; The trial production subsystem will conduct conditional trial production on the cast pipe production based on the determination of the mold making. During the trial production process, set different rotation conditions, combine CFD simulation, identify the solidification zone sequence under different rotation conditions, and analyze the composition differences and solute element distribution in different solidification zones according to the identification results under different rotation conditions to obtain the composition segregation coefficient Pxs; The production management subsystem is used to analyze the feeding timeliness of the cast pipe production line during the solidification process of the molten metal, construct the pore defect risk coefficient Kfx in combination with relevant process data, and use deep learning technology to construct a loss minimization prediction model for fitting and outputting the production selection conditions for centrifugal casting.

[0015] The present invention provides an intelligent management method and system for a cast pipe production line, having the following beneficial effects: (1) By extracting the process data in the cast pipe production line and determining the mold making according to the casting requirements, the present invention can accurately formulate molds that meet the actual production needs, reducing the losses of manufacturers and the fluctuations in cast pipe production caused by improper or mismatched mold designs. During the trial production process, by setting different rotation conditions and combining CFD simulation technology, the present invention can identify the sequence of solidification regions under different rotation conditions and analyze the composition differences and the distribution of solute elements in different regions. By obtaining the composition segregation coefficient Pxs, the distribution of solute elements in the cast pipe production process can be optimized, further improving the composition uniformity of the cast pipe, thereby enhancing the mechanical properties and structural reliability of the cast pipe and reducing the defect risk caused by uneven composition. During the cast pipe production process, by analyzing the feeding timeliness during the solidification process of the molten metal and combining process data to construct the pore defect risk coefficient Kfx, this analysis can predict potential pore defects in advance, laying a foundation for effectively reducing the occurrence rate of pore defects. Using deep learning technology, the production conditions for centrifugal casting are optimized, reducing losses and energy waste during the production process and ensuring that the cast pipe is produced under the best conditions, further improving production efficiency. In short, through the integration of advanced simulation technology, deep learning algorithms and refined production management means, the present invention realizes the intelligent optimization of the cast pipe production process.

[0016] (2) By analyzing the relative contents of different melting point phases in the pre-solidified region and the post-solidified region, the present invention can accurately identify the composition differences in each region at different solidification stages. By calculating the mass fraction of the high melting point phase based on the phase equilibrium principle, the material composition differences in different regions of the cast pipe can be accurately reflected. For example, at different solidification stages (corresponding to different temperatures), the changes in the mass fractions of the high melting point phases in the pre-solidified region and the post-solidified region can provide a scientific basis for subsequent evaluation of the mechanical property uniformity of the cast pipe. If the mass fraction difference is large, it indicates uneven composition distribution, which may lead to more impurities or brittle phases in some regions, reducing the overall mechanical properties of the cast pipe. Mechanical property optimization: The present invention helps to optimize the mechanical properties of the cast pipe by extracting the maximum mass fraction difference as the composition distribution difference index from multiple solidification stages. Cast pipes with uneven composition distribution are prone to brittle fracture or fatigue failure under external forces such as impact, affecting the service life of the product. Through the method of the present invention, the composition segregation can be further controlled, and the toughness and fatigue resistance of the cast pipe can be improved.

[0017] (3) By comparing based on solidification shrinkage rate data and the actual volume of supplementary liquid metal, analyze the timeliness of feeding in the cast pipe production line, so as to ensure that the cast pipe can be supplemented with liquid metal in a timely manner during the solidification process. By defining a timeliness function, it is possible to effectively determine whether the cast pipe production line is in a timely feeding state. By collecting the riser content in the mold under different rotation conditions and obtaining the porosity risk coefficient based on dimensionless processing, the present invention can quantify the porosity risk that may occur during the cast pipe production process. This risk coefficient helps to identify potential defects in the production process and provides an effective adjustment direction. Generally speaking, through precise analysis of feeding timeliness, porosity risk assessment and failure prediction, the present invention optimizes the process parameters of cast pipe production, reduces the occurrence of defects, and improves the mechanical properties and stability of the cast pipe. Description of the Drawings

[0018] Figure 1 It is a schematic flow chart of an intelligent management method for a cast pipe production line of the present invention; Figure 2 It is a partial cross-sectional view of the mold of the present invention; Figure 3 It is a distribution diagram of the solidification status in the mold under different rotation conditions of the present invention; Figure 4 It is a block diagram of an intelligent management system for a cast pipe production line of the present invention. Detailed Embodiments

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0020] Embodiment 1 Please refer to Figure 1 , the present invention provides an intelligent management method for a cast pipe production line, including the following steps, S1: According to the cast pipe production line, extract relevant process data in the production line, and determine the mold making according to the casting requirements; S2: Based on the determination of the mold making, conduct conditional trial production for the cast pipe production. During the trial production process, set different rotation conditions (different rotation speeds), combine CFD simulation, identify the sequence of solidification regions under different rotation conditions, and analyze the compositional differences and solute element distribution in different solidification regions according to the identification results under different rotation conditions to obtain the composition segregation coefficient Pxs; S3: During the solidification process of the molten metal, analyze the feeding timeliness of the cast pipe production line, and combine relevant process data to construct the pore defect risk coefficient Kfx. Use deep learning technology to construct a loss-minimization prediction model and perform fitting to output the production selection conditions for centrifugal casting.

[0021] In this embodiment, by extracting and analyzing the process data in the production line in real time and combining CFD simulation technology, it is possible to accurately identify the sequence of solidification regions under different rotation conditions, thereby providing a scientific basis for the mold preparation and rotation condition setting in the cast pipe production process. This method effectively avoids the production instability under traditional empirical or single process conditions. Through precise simulation and data-driven approach, it significantly improves the quality and production efficiency of cast pipes. During the solidification process of cast pipes, based on the identification of solidification regions under different rotation conditions, it is possible to deeply analyze the distribution differences of solute elements in different solidification regions, and then obtain the composition segregation coefficient Pxs. By analyzing the composition segregation coefficient, the problem of uneven composition caused by the change of solid-liquid state in the cast pipe production process is effectively solved, thereby improving the mechanical properties and reliability of cast pipes and reducing the defects caused by uneven composition. During the solidification process, the present invention constructs the pore defect risk coefficient Kfx by analyzing the feeding timeliness of the cast pipe production line and combining relevant process data. Combining deep learning technology, through the optimization and fitting of the loss-minimization prediction model, it is possible to accurately predict the production conditions in the trial production stage, reduce defects such as pore defects and cracks caused by untimely feeding, and reduce the quality fluctuations and resource waste in cast pipe production. In short, this method realizes the whole-process intelligent management from mold design, rotation condition optimization to feeding timeliness analysis by precisely controlling each link in the cast pipe production process, effectively improves the quality stability, production efficiency and resource utilization rate of cast pipe products, and has broad application prospects and economic value.

[0022] Embodiment 2 Please refer to Figure 1 、 Figure 2 and Figure 3 , specifically: The specific steps of S1 include: S11: Deploy multiple groups of sensors and data acquisition devices on the cast pipe production line to extract relevant process data in the production line in real time. The relevant process data includes the operation information of production line equipment, process parameters and material property information. Among them, the operation information of production line equipment includes the furnace temperature in the cast pipe production line, the rotation speed of the centrifuge, and the current and voltage of each production device; the process parameters include the initial batching ratio of the metal material for casting the cast pipe, the initial batching content of the metal material , the width of the solidification temperature range , the initial liquid volume TJ, the pouring temperature of the metal material, the cooling rate and the supplementary volume; the material property information includes the density of the metal material for casting the cast pipe, the solidus temperature , viscosity and thermal conductivity; Among them, based on the batching plan before casting and the known raw material composition, the initial batching content of the metal material is directly obtained , which is usually based on the design requirements and target performance of the material.

[0023] Differential scanning calorimetry or differential thermal analysis can be used to measure the change in heat flow of the alloy at different temperatures during the cooling process, so as to obtain the solidification temperature range (including the solidification start temperature and the solidification end temperature). The width of the solidification temperature range usually refers to the temperature range from the liquid state to the completely solid state, and is usually determined by measuring the solidus temperature (the temperature at which solidification begins in the liquid alloy) and the liquidus temperature (the temperature at which the liquid phase is completely solidified) during the cooling process of the liquid alloy.

[0024] Use physical experiments to measure the density of the liquid metal, and combine the mass of the liquid metal at a known temperature to obtain the liquid volume.

[0025] The solidus temperature can be obtained by referring to the phase diagram. The alloy phase diagram usually provides the solidus line (the temperature at which solidification begins) at different alloy compositions and temperatures.

[0026] S12: According to the casting requirements, make the corresponding mold and obtain the geometric model of the mold. Among them, the casting requirements include the specifications and shapes of the cast pipes to be cast, and the mold type is selected according to the production batch requirements, production cycle plan and production cost control of the enterprise. The mold types include metal molds and sand molds.

[0027] The specific steps of S2 include: S21: Based on the CFD simulation technology, input the mold geometric model, production line equipment operation information and material property information in S12 into the CFD software, set the process parameters as boundary conditions, and set different rotation conditions to conduct conditional trial production of the cast pipe production. During the trial production process, identify the solidification area sequence under different rotation conditions. The specific identification steps are as follows: S211: According to the numerical simulation technology, use the finite difference method to discretize time and space, and numerically solve the heat conduction equation to obtain the temperature field T (after high-temperature melting) at each position in the mold under different rotation conditions, that is, the temperature distribution at any position in the casting; S212: Based on the temperature field T in each area of the mold, combined with the material property information, obtain the solidification index at each position in the mold under different rotation conditions , specifically: ; In the formula, represents the solidification index at the position r in the mold at the t-th time point, represents the unit step function. If , then = 1, otherwise = 0; represents the current temperature value at position r in the mold, represents the solidus temperature, i.e., the solidification temperature of the material (the temperature at which the liquid turns into a solid), represents the adjustment parameter, with a value range of , which is used to describe the influence of the temperature gradient on the solidification rate; represents the temperature gradient at position r in the mold, reflecting the rate of change of temperature with spatial position. r represents the position and t represents the time point; Specifically, please refer to Figure 3 , Figure 3 The pilot test lists multiple sets of rotation conditions and multiple position points. Combining with the solidification threshold, it clearly shows the solidification indexes at the corresponding positions under different rotation conditions and whether they are above the solidification threshold; Among them, the temperature gradient at position r in the mold The specific acquisition method is as follows: Based on the central difference method in the discretization method: ; In the formula, , and are the step sizes of the spatial grid respectively, that is, the distances between adjacent points in different directions, which determine the accuracy of spatial discretization; for example, if we use the grid step size in the x direction, then the spatial distribution of the calculation points is discrete, and the step size represents the distance from one grid point to the adjacent grid point. , , , , and These are the temperature values of adjacent grid points respectively. and represent the temperature values after the current point r moves one step forward and backward in the x direction; and represent the temperature values in the y direction; and represent the temperature values in the z direction; The central difference method approximates the derivative by taking points on both sides of the current point. In this way, it can better capture the change trend of the function (here is the temperature distribution) at this point.

[0028] The specific identification steps also include: S213: Based on the solidification indexes at each position in the mold under different rotation conditions , a solidification threshold is set, and by comparing the solidification threshold with the solidification indexes at each position under different rotation conditions , the order of solidification regions under different rotation conditions is identified: S2131: If the solidification index at the corresponding position under the corresponding rotation condition exceeds the solidification threshold, it is determined that the corresponding position under the corresponding rotation condition is in a solidified state. Through statistics, the solidification region under the corresponding rotation condition is obtained, and the solidification region is marked as the pre-solidification region; S2132: If the solidification index at the corresponding position under the corresponding rotation condition does not exceed the solidification threshold, it is determined that the corresponding position under the corresponding rotation condition is not in a solidified state. Through statistics, the non-solidified region under the corresponding rotation condition is obtained, and the non-solidified region is marked as the post-solidification region.

[0029] In this embodiment, in step S12, a suitable mold is selected according to the casting requirements and production batch requirements, and the geometric model of the mold is obtained. This process combines the production cycle and cost control. By selecting a metal mold or a sand mold, the production efficiency can be maximized while meeting the production requirements. For example, metal molds are usually suitable for large-scale production, which can improve production efficiency and reduce the production cost of single-piece cast pipes, while sand molds are suitable for small-batch and high-complexity cast pipe production. The selection of such molds optimizes the production process, makes the production more flexible, and can adjust the production mode according to the requirements. In step S2, using CFD simulation technology and numerical solution methods, combined with the mold geometric model, production data, and material property information, a conditional trial production of cast pipe production is carried out. By identifying the order of solidification regions under different rotation conditions, the temperature distribution and composition segregation during the solidification process can be predicted and controlled more accurately.

[0030] Suppose that in a certain cast pipe production, the rotation speed is set to a low-speed condition. During the trial production process, it is identified by using CFD simulation technology that the temperatures of some positions of the cast pipe mold are relatively low and the solidification process is slow. The solidification indexes of these positions are obtained through numerical solution methods, and these regions are determined as the post-solidification regions. At this time, by adjusting the rotation speed, the order of solidification regions can be changed to make the temperature gradient more uniform, thereby optimizing the solidification process and reducing the segregation of solute elements. For example, if the solute elements at some positions segregate during the solidification process, it may lead to uneven performance of the cast pipe and affect subsequent use and processing. By setting the solidification threshold and comparing it with the simulation results, the present invention can identify the regional changes during the solidification process according to different rotation conditions, so as to manage the production line in a timely manner and ensure the quality of the cast pipe. The accurate identification of the solidification region, especially the accurate judgment of the pre-solidification region and the post-solidification region, can avoid defects (such as pores, cracks, etc.) caused by uneven solidification. These optimization processes effectively improve the structural integrity and mechanical properties of the cast pipe through intelligent analysis and prediction.

[0031] Example 3 Please refer to Figure 1 , specifically: The specific steps of S2 also include: S22: Based on the first-solidified region and the last-solidified region identified in S21, analyze the relative contents of different melting-point phases in the first-solidified region and the last-solidified region at different solidification stages (corresponding to different temperatures) within each rotation condition, so as to obtain the mass fraction of the high-melting-point phase in the first-solidified region and the mass fraction of the high-melting-point phase in the last-solidified region , specifically: In the formula, represents the content of the high-melting-point phase in the first-solidified region at the solidification stage d, represents the content of the high-melting-point phase in the last-solidified region at the solidification stage d, represents the initial ingredient content of the metal material, represents the mass fraction of the high-melting-point phase in the first-solidified region at the solidification stage d, represents the mass fraction of the high-melting-point phase in the last-solidified region at the solidification stage d; Among them, the melting point can be monitored and obtained by methods such as differential scanning calorimetry, differential thermal analysis, melting point method, etc. Here, the high-melting-point phase refers to the substance or phase with a relatively high melting point in a multiphase system. In a metal alloy or other solid-liquid phase change system, different phases of the material will transform at different temperatures, and some of these phases have relatively high melting points, which are usually called high-melting-point phases. Therefore, the high-melting-point phase is a phase with a relatively high melting point.

[0032] It should be noted that: The mass fraction reflects the respective proportions of the melting-point phases in the corresponding solidification region. Based on the phase equilibrium principle, it shows that at a given temperature, the relative contents of the two phases in the metal material depend on the overall composition of the metal material and the respective equilibrium compositions of the two phases (the content of the corresponding melting-point phase in the corresponding solidification region at a certain temperature).

[0033] By analyzing the relative contents of the melting-point phases in the first-solidified region and the last-solidified region at different solidification stages (corresponding to different temperatures), the composition difference can be reflected from the perspective of phase composition. During the solidification process, as the temperature decreases, according to the lever rule, the content changes of the high-melting-point phase in the first-solidified region and the last-solidified region can be known, thereby indirectly reflecting the composition difference.

[0034] S23: Subtract the mass fraction of the high-melting-point phase in the first-solidified region and the mass fraction of the high-melting-point phase in the last-solidified region to obtain the mass fraction difference , specifically: S24: Based on quality score difference The method of obtaining the maximum mass fraction difference from multiple solidification stages under the corresponding rotation conditions is to extract the maximum mass fraction difference from multiple solidification stages. As an indicator of the difference in composition distribution between the first solidification area and the later solidification area , used to measure the uniformity of mechanical properties at different positions of the cast pipe.

[0035] It should be noted that uneven distribution of components may cause more impurities or brittle phases to exist in certain areas. When subjected to dynamic loads such as impact, these areas are prone to brittle fracture, reducing the toughness and fatigue resistance of the casting.

[0036] In this embodiment, by analyzing the relative contents of different melting point phases in the first solidification area and the later solidification area, this method can further evaluate the distribution of the high melting point phase in the cast pipe at different solidification stages. This analysis helps to deeply understand the differences in the composition of the cast pipe, especially the changes in the mass fraction of the high melting point phase under different temperature conditions, thereby optimizing the production process and reducing segregation. The present invention obtains the mass fraction difference and then quantifies the difference in composition distribution between different positions of the cast pipe. This indicator can further reflect the uniformity of the composition in the cast pipe, avoid inconsistent mechanical properties due to uneven composition, especially when subjected to dynamic loads, avoid brittle fracture and fatigue failure.

[0037] Example: Assume that in a cast pipe production, after CFD simulation analysis, under certain rotation conditions, the mass fraction of the high melting point phase in the first solidification area at a certain temperature stage is 0.65, and the mass fraction of the high melting point phase in the later solidification area at the same temperature stage is 0.35. Through subtraction, the mass fraction difference is 0.30, which indicates that there is a certain difference in the composition distribution of the cast pipe, which may cause the mechanical properties of the first solidification area to be better than those of the later solidification area. Further extracting the maximum mass fraction difference in multiple solidification stages as a composition difference index can timely optimize the rotation conditions, reduce the brittle phase problem caused by uneven composition in the cast pipe, and ultimately improve the mechanical properties and service life of the cast pipe.

[0038] Example 4 Please refer to Figure 1 Specifically: S2 includes the following specific steps: S25: Based on the first solidification area and the last solidification area identified in S21, and combined with the phase diagram principle, the distribution relationship of the solute elements in the solid phase and the liquid phase in the mold under different rotation conditions is analyzed to obtain the distribution coefficient Fx under the corresponding rotation conditions, specifically: In the formula, represents the solute concentration in the solid phase, represents the solute concentration in the liquid phase; When < 1, it indicates that the solubility of the solute in the liquid phase is higher than that in the solid phase. As the solid phase precipitates, the solute concentration in the liquid phase will gradually increase. For example, during the solidification of some aluminum alloys, the < 1. As solidification proceeds from the outside to the inside, the copper element will be continuously excluded to the center of the casting, resulting in an increase in the copper element concentration at the center. When solidifying, copper will gradually accumulate in the liquid phase, which belongs to positive segregation; Inverse segregation: Contrary to positive segregation, the solute concentration on the surface or outer layer of the casting is higher than that in the central part. This situation is usually due to the temperature gradient, shrinkage stress, etc. inside and outside the casting during the solidification process, which causes the flow of the liquid phase, bringing the solute elements to the surface of the casting.

[0039] Among them, the phase diagram principle is a graphical tool used to represent the phase state (such as solid state, liquid state, gaseous state, etc.) distribution and change rules of substances under different conditions (such as temperature, pressure, composition, etc.). It is a chart that describes the thermodynamic properties, phase changes and their mutual relationships of multi-component substances or multi-phase substance systems.

[0040] The solute concentration in the solid phase can be measured by X-ray diffraction (XRD) method; The solute concentration in the liquid phase can be measured by thermal analysis methods, such as differential scanning calorimetry (DSC) or thermogravimetric analysis (TGA), etc.

[0041] It should be noted that: General solidification process: During the solidification of a binary alloy, as the temperature decreases, when the solidus temperature is reached, the alloy begins to transform from the liquid state to the solid state. The region where solidification first starts will form the solid phase, and these solid phases usually appear in the form of crystal nuclei and then gradually grow. At this stage, the region where solidification first occurs can indeed be regarded as the solid phase, which has the crystal structure and physical properties of the solid phase and is significantly different from the liquid phase that has not yet solidified.

[0042] Initial stage of solidification: In the initial stage of solidification, the region where solidification first occurs is mainly some tiny solid-phase crystal nuclei. These crystal nuclei form in the liquid alloy and, as time and temperature change, continuously absorb atoms from the surrounding liquid alloy and gradually grow. At this time, the region where solidification first occurs is mainly in the solid phase, but it is still surrounded by a large amount of liquid phase. There is an obvious phase interface between the solid phase and the liquid phase, and there is continuous material and energy exchange between the solid phase and the liquid phase.

[0043] Middle stage of solidification: As the solidification process progresses, the solid phase in the region that solidifies first continuously increases. Each solid-phase crystal nucleus gradually grows and connects with each other to form a continuous solid-phase framework. However, at this time, there is still some liquid phase between the solid-phase frameworks, and these liquid phases will gradually be squeezed into the gaps between the solid phases as the solidification continues. In this stage, a solid-phase-dominated structure has been formed in the region that solidifies first as a whole, but it still contains liquid phases that have not completely solidified, belonging to the state of coexistence of liquid and solid phases, but the solid phase has taken the dominant position.

[0044] Late stage of solidification: In the late stage of solidification, most of the liquid alloy has been transformed into the solid phase, and only a small amount of liquid phase exists at positions such as grain boundaries of the solid phase. At this time, the region that solidifies first has basically completely become the solid phase, and only a very small amount of residual liquid phase may solidify last. However, due to factors such as compositional segregation during the solidification process, there may be certain differences in composition and microstructure between the region that solidifies first and the region that solidifies later. However, from the perspective of phases, the region that solidifies first is basically the solid phase.

[0045] S26: Based on the compositional distribution difference index between the region that solidifies first and the region that solidifies later obtained in S24 and S25 and the distribution coefficient Fx under the corresponding rotation conditions, the compositional segregation coefficient Pxs is obtained through dimensionlessization, specifically as follows: In the formula, and are both weight values, and the specific values are set by the user according to the situation, 0 < < 1, 0 < < 1.

[0046] In this embodiment, by analyzing the distribution relationship of solute elements in the solid phase and liquid phase in the region that solidifies first and the region that solidifies later in combination with the phase diagram principle, the segregation behavior of solute elements in the casting under different rotation conditions can be accurately identified. By calculating the distribution coefficient Fx, the concentration changes of solute elements in the solid phase and liquid phase can be reflected, so as to master the distribution of solutes during the solidification process. For example, during the solidification of aluminum alloy, the distribution coefficient Fx of copper element may be less than 1, resulting in the copper element being gradually repelled from the outside to the center of the casting, forming a positive segregation phenomenon. Scientific identification of segregation phenomenon: By calculating the distribution coefficient Fx, the positive segregation and inverse segregation phenomena in the cast pipe can be identified. The accurate identification of this process can help adjust the casting process, reduce the influence of the segregation phenomenon, and thus improve the quality and uniformity of the cast pipe. Through the compositional segregation coefficient Pxs obtained through dimensionlessization, this method can quantify the degree of compositional segregation in the cast pipe, prepare for helping to optimize the process conditions, and reducing the mechanical property problems caused by compositional segregation.

[0047] Example 5 Please refer to Figure 1, specifically: The specific steps of S3 include: S31: Query the solidification shrinkage rate data of the corresponding metal material based on the material manual and database, extract the initial liquid volume TJ in the process parameters. During the solidification of the molten metal, collect the actual supplementary liquid metal volume Vb at the riser, analyze the feeding timeliness of the cast pipe production line, and generate the timeliness function under different rotation conditions , specifically: In the formula, represents the solidification shrinkage volume. If the actual supplementary liquid metal volume Vb < solidification shrinkage volume , then = 1, indicating that the feeding of the current cast pipe production line is in an untimely state; if the actual supplementary liquid metal volume Vb ≥ solidification shrinkage volume , then = 0, indicating that the feeding of the current cast pipe production line is in a timely state.

[0048] It should be noted that: The solidification shrinkage rate data refers to the solidification shrinkage rate of the casting, which can also be expressed by the formula: casting solidification shrinkage rate = (TJ - gj) / TJ; where, TJ represents the initial liquid volume, and gj represents the volume of the solid metal after solidification; Among them, the difference between the initial liquid volume TJ and the volume of the solid metal gj after solidification is expressed as the solidification shrinkage volume ; Specifically, the material manual includes but is not limited to "Mechanical Design Manual", "Casting Manual" and "Metal Material Manual", and can also be through the CNKI database, Web of Science database and Materials Project database; Among them, "Mechanical Design Manual": It is a comprehensive reference book in the field of mechanical design, which contains the performance data of various mechanical engineering materials. In the part of the casting performance of the materials, it contains relevant data such as the solidification shrinkage rate of common metal materials, and can provide reference for mechanical design and casting process design.

[0049] "Casting Manual": This set of manuals has a detailed introduction to casting processes and materials, and has a special chapter discussing the characteristics of various casting metal materials, including the solidification shrinkage rate data of various casting metal materials; "Metal Material Manual": It systematically introduces the performance, characteristics, etc. of various metal materials. In addition to the basic mechanical properties, it also involves thermal physical property data such as the shrinkage rate of materials during solidification, which is convenient for material selection and process formulation.

[0050] CNKI Database: It has collected a large number of literature resources such as academic journals and dissertations. By searching for the keyword "solidification shrinkage rate of common metal materials", many relevant research papers can be found, including experimental data and research analysis on the solidification shrinkage rate of specific metal materials.

[0051] Web of Science Database: It is an authoritative academic database globally, covering high-quality literature in many disciplinary fields, and can be used to find cutting-edge research results and data on the solidification shrinkage rate of metal materials internationally.

[0052] Materials Project Database: It is a database focusing on materials science, providing a large amount of computational and experimental data on materials, including thermodynamic and physical property data such as the solidification shrinkage rate of metal materials. Users can obtain the required information through search and screening functions.

[0053] S32: Before trial production, collect supplementary data in the corresponding molds on the cast pipe production line under different rotation conditions. The supplementary data includes the riser content Hm in the corresponding molds on the cast pipe production line under different rotation conditions. S33: After dimensionless processing, obtain the pore defect risk coefficient Kfx, specifically: In the formula, represents the width of the solidification temperature range, , and are all weight values, and the specific values are set by users according to the situation.

[0054] Among them, the width of the solidification temperature range refers to the temperature difference between the start of solidification and complete solidification of the liquid metal. It describes the transition range of different components or different phases during the solidification process of the alloy. The width of the solidification temperature range is an important indicator to measure the stability of the metal solidification process and the uniformity of the alloy composition.

[0055] Alloys with a relatively wide solidification temperature range are more likely to produce shrinkage porosity. In this temperature range, the coexistence time of liquid metal and solid metal is longer, and the feeding channels of the liquid metal are easily blocked by the solid phase that solidifies first, resulting in insufficient liquid metal supply for the later solidified part, thus generating shrinkage porosity.

[0056] The riser is the liquid storage part used to compensate for the solidification shrinkage of the casting. The content in the riser directly affects its feeding effect. If the riser capacity is insufficient, it is difficult to effectively provide liquid metal feeding for the casting, resulting in shrinkage porosity in the casting.

[0057] The specific steps of S3 also include: S34: Take the pore defect risk coefficient Kfx and the composition segregation coefficient Pxs as the input values of the loss minimization prediction model. After dimensionless processing, fit and output the failure evaluation index DPzs under the corresponding rotation conditions, specifically: ; In the formula, and are both influence coefficients, used to adjust the influence intensity of pore defects and composition segregation on failure evaluation. e represents the Euler number, with a value of approximately 2.71828.

[0058] The specific steps of S3 also include: S35: Based on the failure evaluation index DPzs under the corresponding rotation conditions obtained in S34, after feature extraction, extract the failure evaluation index DPzs with the smallest value, and use the rotation conditions corresponding to the failure evaluation index DPzs with the smallest value as the production conditions for centrifugal casting.

[0059] In this embodiment, based on the comparison of the volume of supplementary liquid metal at the riser and the solidification shrinkage volume actually collected, the present invention can monitor the feeding timeliness of the cast pipe production line in real time. By generating a timeliness function, it can quickly identify and adjust the production conditions when the feeding is not timely, thereby effectively avoiding defects such as pore defects caused by insufficient feeding. This optimization measure helps to ensure the quality stability of the castings during the cast pipe production process and avoid unnecessary quality losses. By constructing a pore defect risk coefficient, the present invention can quantify the risk of pore defects during the cast pipe production process, provide targeted data support, help production personnel adjust the rotation conditions and the supplementary amount, and reduce the probability of pore defects occurring. This risk coefficient combined with the analysis results after dimensionless processing can effectively guide the optimization of the casting process and reduce production failures caused by pore defects. Failure evaluation and production condition optimization: By constructing a loss minimization prediction model and combining the input data of the pore defect risk coefficient and the composition segregation coefficient, it is possible to evaluate the failure risk under different rotation conditions. According to the obtained failure evaluation index, the present invention helps users identify relatively optimal production conditions. By extracting the failure evaluation index with the smallest value and combining the prediction results of deep learning technology, it is possible to optimize the production conditions of centrifugal casting and improve production efficiency and the quality of castings.

[0060] Specifically, assume that during the production process of cast pipes, it is found through real-time monitoring that the actual volume of supplementary liquid metal at the riser is less than the solidification shrinkage volume. This indicates that the feeding of the current cast pipe production line is not timely. To avoid porosity due to untimely replenishment, the production system will adjust the replenishment speed according to the timeliness function. After further calculating the porosity risk coefficient, it is found that the value is relatively high, indicating a relatively high porosity risk under the current rotation conditions. Then, the failure assessment index output by fitting with the deep learning model shows that the failure risk is the lowest under a certain rotation condition. The system will automatically select this condition as the optimal production condition. In this way, the quality of the entire production process is effectively improved, and at the same time, rework or waste caused by process problems is reduced.

[0061] Example 6 Please refer to Figure 4 , specifically: An intelligent management system for a cast pipe production line, including a preparation subsystem, a trial production subsystem, and a production management subsystem; The preparation subsystem extracts relevant process data within the production line according to the cast pipe production line, and determines the mold making according to the casting requirements; The trial production subsystem will conduct conditional trial production of cast pipe production based on the determination of the mold making. During the trial production process, different rotation conditions are set, combined with CFD simulation, to identify the solidification zone sequence under different rotation conditions, and according to the identification results under different rotation conditions, analyze the compositional differences and solute element distribution in different solidification zones to obtain the composition segregation coefficient Pxs; The production management subsystem is used to analyze the feeding timeliness of the cast pipe production line during the solidification process of the molten metal, and combine relevant process data to construct a porosity risk coefficient Kfx, use deep learning technology to construct a loss minimization prediction model, and perform fitting to output the production selection conditions for centrifugal casting.

[0062] During the operation of this system, the main function of the preparation subsystem is to extract relevant process data within the cast pipe production line and determine the mold according to casting requirements. This subsystem provides data support for the subsequent casting process by collecting the process parameters of the production line. The trial production subsystem conducts conditional trial production based on the determined mold conditions. During the trial production process, the system sets different rotation conditions and, through computational fluid dynamics (CFD) simulation, identifies the solidification zone sequence of the cast pipe production line under different rotation conditions. By simulating the solidification process under different rotation conditions, the trial production subsystem can analyze the compositional differences and solute element distribution in different solidification zones of the cast pipe. These data are used to obtain the composition segregation coefficient, which is crucial for subsequent production optimization. By accurately identifying the solidification zone sequence and composition distribution, the trial production subsystem provides accurate process data support for the production management subsystem, helping to improve the quality of cast pipes. The production management subsystem is responsible for analyzing the timeliness of feeding during the solidification process of the molten metal and constructing a pore defect risk coefficient in combination with relevant process data. By monitoring the changes in the volume of supplementary liquid metal and the volume of solidification shrinkage in real time, the production management subsystem determines whether the feeding is timely and promptly discovers potential production problems. Subsequently, the production check conditions for centrifugal casting are fitted and output. These optimized production conditions can effectively reduce the risks of cast pipe defects and failures and improve production efficiency.

[0063] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent management method for a cast pipe production line, characterized in that: The following steps are included: S1: According to the casting pipe production line, extract the relevant process data in the production line, and determine the mold making according to the casting requirements; S2: Based on the determination of the mold making, conditional trial production is carried out for the cast pipe production. During the trial production process, different rotation conditions are set, and combined with CFD simulation, the sequence of solidification zones under different rotation conditions is identified. Based on the identification results under different rotation conditions, the differences in composition and the distribution of solute elements in different solidification zones are analyzed to obtain the composition segregation coefficient Pxs; S3: During the solidification process of the molten metal, the timeliness of the shrinkage compensation of the casting pipe production line is analyzed, and the hole defect risk coefficient Kfx is constructed in combination with relevant process data. The loss minimization prediction model is constructed using deep learning technology to fit and output the production selection conditions of centrifugal casting.

2. According to claim 1, a method for intelligent management of a cast pipe production line is characterized in that: The specific steps of S1 include: S11: On the cast pipe production line, multiple sets of sensors and data acquisition equipment are deployed to extract relevant process data in the production line in real time. The relevant process data include production line equipment operation information, process parameters and material property information. The production line equipment operation information includes the furnace temperature, centrifuge rotation speed and current and voltage of each production equipment in the cast pipe production line; the process parameters include the initial batching ratio of metal materials for casting cast pipes, the initial batching content of metal materials ... , Solidification temperature range width , initial liquid volume TJ, metal material pouring temperature, cooling rate and supplementary volume; material property information includes the density of the metal material of the cast pipe, solidus temperature , viscosity and thermal conductivity; S12: According to the casting requirements, a corresponding mold is made and a mold geometry model is obtained, wherein the casting requirements include the specifications and shapes of the cast pipes to be cast, and the mold type is selected according to the production batch requirements, production cycle planning and production cost control of the enterprise, and the mold types include metal molds and sand molds.

3. The intelligent management method of a cast pipe production line according to claim 2, characterized in that: The specific steps of S2 include: S21: Based on CFD simulation technology, the mold geometry model, production line equipment operation information and material property information in S12 are input into the CFD software, the process parameters are set as boundary conditions, and different rotation conditions are set to conduct conditional trial production of cast pipes. During the trial production process, the sequence of solidification areas under different rotation conditions is identified. The specific identification steps are as follows: S211: Based on numerical simulation technology, the finite difference method is used to discretize time and space, and the heat conduction equation is numerically solved to obtain the temperature field T at each position in the mold under different rotation conditions; S212: Based on the temperature field T of each area in the mold and combined with the material property information, obtain the solidification index at each position in the mold under different rotation conditions , specifically: ； In the formula, It represents the solidification index at the position r in the mold at the tth time point, represents a unit step function, if ,but =1, otherwise =0; Indicates the current temperature value at position r in the mold, represents the solidus temperature, Indicates adjustment parameters, value range ; represents the temperature gradient at position r in the mold.

4. The intelligent management method of a casting pipe production line according to claim 3 is characterized in that: The specific identification steps also include: S213: Solidification index at various positions in the mold under different rotation conditions , set the coagulation threshold, and compare the coagulation threshold with the coagulation index at each position under different rotation conditions. Comparison was performed to identify the sequence of solidification zones under different rotation conditions: S2131: If the solidification index at the corresponding position under the corresponding rotation condition If the solidification threshold is exceeded, it is determined that the corresponding position under the corresponding rotation condition is in a solidification state, and statistics are performed to obtain the solidification area under the corresponding rotation condition, and the solidification area is marked as the first solidification area; S2132: If the solidification index at the corresponding position under the corresponding rotation condition If the solidification threshold is not exceeded, it is determined that the corresponding position under the corresponding rotation condition is not in a solidification state. After statistics, the unsolidified area under the corresponding rotation condition is obtained, and the unsolidified area is marked as a post-solidification area.

5. The intelligent management method of a cast pipe production line according to claim 4, characterized in that: The specific steps of S2 also include: S22: Based on the first solidification region and the later solidification region identified in S21, the relative contents of different melting point phases in the first solidification region and the later solidification region are analyzed at different solidification stages under each rotation condition to obtain the mass fraction of the high melting point phase in the first solidification region and the mass fraction of the high melting point phase in the post-solidification region , specifically: In the formula, It represents the content of high melting point phase in the first solidification region at solidification stage d. It represents the content of high melting point phase in the post-solidification region at solidification stage d, Indicates the initial content of metal materials. represents the mass fraction of the high melting point phase in the first solidification region at solidification stage d, It represents the mass fraction of the high melting point phase in the post-solidification region at the solidification stage d; S23: The mass fraction of the high melting point phase in the first solidification region and the mass fraction of the high melting point phase in the post-solidification region Perform subtraction to obtain the mass score difference , specifically: S24: Based on quality score difference The method of obtaining the maximum mass fraction difference from multiple solidification stages under the corresponding rotation conditions is to extract the maximum mass fraction difference from multiple solidification stages. As an indicator of the difference in composition distribution between the first solidification area and the later solidification area , used to measure the uniformity of mechanical properties at different positions of the cast pipe.

6. The intelligent management method of a cast pipe production line according to claim 5, characterized in that: The specific steps of S2 also include: S25: Based on the first solidification area and the last solidification area identified in S21, and combined with the phase diagram principle, the distribution relationship of the solute elements in the solid phase and the liquid phase in the mold under different rotation conditions is analyzed to obtain the distribution coefficient Fx under the corresponding rotation conditions, specifically: In the formula, represents the solute concentration in the solid phase, It represents the solute concentration in the liquid phase; S26: The composition distribution difference index between the first solidification area and the last solidification area obtained in S24 and S25 And the distribution coefficient Fx under the corresponding rotation conditions, after dimensionless conversion, the component segregation coefficient Pxs is obtained, which is specifically: In the formula, and They are all weight values, and the specific values ​​are set by the user according to the situation.

7. The intelligent management method of a casting pipe production line according to claim 6, characterized in that: The specific steps of S3 include: S31: Based on the material manual and database, query the solidification shrinkage rate data of the corresponding metal material, and extract the initial liquid volume TJ in the process parameters. During the solidification process of the molten metal, collect the actual replenishment liquid metal volume Vb at the riser, analyze the timeliness of the shrinkage feeding of the casting pipe production line, and generate the timeliness function under different rotation conditions. , specifically: In the formula, Indicates the solidification shrinkage volume. If the actual replenished liquid metal volume Vb < solidification shrinkage volume ,but =1, indicating that the current casting pipe production line is not in a timely state of shrinkage compensation; if the actual replenishment of liquid metal volume Vb ≥ solidification shrinkage volume ,but =0, indicating that the current shrinkage compensation of the cast pipe production line is in a timely state; S32: before trial production, collecting supplementary data in the corresponding mold on the casting pipe production line under different rotation conditions, the supplementary data including the riser content Hm in the corresponding mold on the casting pipe production line under different rotation conditions; S33: Obtain the hole defect risk factor Kfx, specifically: In the formula, Indicates the width of the solidification temperature range, , and All are weight values.

8. The intelligent management method of a cast pipe production line according to claim 7, characterized in that: The specific steps of S3 also include: S34: The hole defect risk coefficient Kfx and the component segregation coefficient Pxs are used as the input values ​​of the loss minimization prediction model. After dimensionless processing, the failure assessment index DPzs under the corresponding rotation conditions is fitted and output, which is specifically: ； In the formula, and They are all influence coefficients, which are used to adjust the influence of pore defects and component segregation on failure assessment. e is expressed as the Euler number.

9. The intelligent management method of a casting pipe production line according to claim 8, characterized in that: The specific steps of S3 also include: S35: Based on the failure assessment index DPzs under the corresponding rotation conditions obtained in S34, after feature extraction, the failure assessment index DPzs with the smallest value is extracted, and the rotation condition corresponding to the failure assessment index DPzs with the smallest value is used as the production condition of centrifugal casting.

10. An intelligent management system for a cast pipe production line, used to implement an intelligent management method for a cast pipe production line according to any one of claims 1 to 9, characterized in that: Including preparation subsystem, trial production subsystem and production management subsystem; The preparation subsystem extracts relevant process data from the pipe casting production line and determines the mold making according to the casting requirements; The trial production subsystem will conduct conditional trial production of cast pipes based on the determination of the mold making. During the trial production process, different rotation conditions are set, and combined with CFD simulation, the sequence of solidification areas under different rotation conditions is identified. Based on the identification results under different rotation conditions, the differences in composition and the distribution of solute elements in different solidification areas are analyzed to obtain the composition segregation coefficient Pxs. The production management subsystem is used to analyze the timeliness of shrinkage feeding of the casting pipe production line during the solidification process of the molten metal, and to construct the hole defect risk coefficient Kfx in combination with relevant process data. The deep learning technology is used to build a loss minimization prediction model to fit and output the production selection conditions for centrifugal casting.

Images