Intelligent cold press forming method and die system for 9Ni steel spherical head

Abstract

The application discloses an intelligent cold-press forming method and a die system for a 9Ni steel spherical head, wherein the method realizes real-time feedback of pressure distribution and strain data through a sensor network, calls a preset springback amount prediction model in a database, calculates expected springback amounts of each region in combination with real-time data, drives a fine adjustment mechanism of a corresponding sector module of an upper die by a control system, and compensates for a reverse micro-deformation of a local die surface according to the predicted springback amount. The embodiment of the application realizes real-time closed-loop control of sensing, prediction and compensation, actively offsets springback, improves multi-dimensional linear precision of the head, realizes accurate guidance of material flow, reduces repeated forging and pressing, effectively avoids local excessive thinning, greatly reduces the number of repeated pressing, avoids work hardening and toughness loss, guarantees low-temperature impact toughness of the 9Ni steel head, greatly improves the qualification rate of one-time pressing, basically eliminates a subsequent manual correction process, and improves production efficiency.

Description

Technical Field

[0001] This disclosure relates to the field of pressure vessel manufacturing technology, and in particular to an intelligent cold pressing forming method and mold system for 9Ni steel spherical heads. Background Technology

[0002] Spherical heads are key structural components for pressure bearing in liquid cargo tanks, and their manufacturing technology directly affects the safety and service life of the equipment. Spherical heads are generally formed by cold pressing. Due to the high strength and high toughness of 9Ni steel, the elastic recovery energy is relatively large, making spherical heads prone to springback during cold forming. The complex multi-dimensional lines in the structure reduce pressing efficiency, make it difficult to control the accuracy of the lines, and result in high cold processing costs and uncontrollable cycles.

[0003] The spherical end plate of the related technology is formed by cold pressing. It is formed by repeated pressing using a four-column hydraulic press and a solid sand casting mold. Based on processing experience, repeated pressing is used to eliminate springback deformation.

[0004] The aforementioned cold-pressing method also has the following problems: First, the related technologies lack scientific springback prediction methods, and pressing parameters, such as pressure, speed, and path, are dependent on worker experience, leading to significant reliance on guesswork. Second, the method involves repeated pressing to eliminate springback, resulting in work hardening of the 9Ni steel material, reduced toughness, and impact on the low-temperature performance of the end caps. Third, the forming process lacks monitoring, making it impossible to accurately grasp the internal stress state and deformation uniformity of the billet, easily leading to localized excessive thinning or thickening. Fourth, the process adaptability is poor; process parameters need to be re-explored for different end cap specifications, resulting in high trial-and-error costs and making standardized production difficult. Summary of the Invention

[0005] This disclosure provides an intelligent cold pressing forming method and mold system for 9Ni steel spherical heads to solve or alleviate one or more technical problems in the prior art.

[0006] As a first aspect of the present disclosure, the present disclosure provides a method for intelligent cold pressing of 9Ni steel spherical heads, including:

[0007] S1: Initial pressing is performed using a first blank holder force and a first pressing speed, and the pressing depth is controlled at 75%-90% of the preset depth, so that the blank initially adheres to the mold but is not fully formed. The first blank holder force is less than the target blank holder force, and the first pressing speed is less than the preset pressing speed.

[0008] S2: Increase the blank holder force to the target blank holder force and start the main pressing program. During the process of the upper die descending to the preset depth, the control system receives real-time feedback of pressure distribution and strain data from the sensor network of the lower die. It calls the preset springback prediction model in the database, calculates the expected springback in each area in combination with the real-time data, and drives the fine-tuning mechanism of the corresponding fan-shaped module of the upper die to perform reverse micro-deformation compensation on the local surface of the die according to the predicted springback.

[0009] In some possible implementations, in step S2, the mold curvature compensation amount Determined according to the following formula: ,in, For database-based prediction of rebound amount, These are the model correction coefficients. To measure the actual pressure, For target pressure, This is the pressure deviation correction factor.

[0010] In some possible implementations, the method further includes: after mold opening, verifying the molding accuracy by optical scanning or contact measurement, and feeding the measurement data back to the database for continuous optimization.

[0011] In some possible implementations, the pressing depth is controlled at 80%-85% of the preset depth, and / or the first pressing force is 60%-70% of the target pressing force, and / or the first pressing speed is 50% of the preset pressing speed, and / or the sensor network of the lower die provides real-time feedback of pressure distribution and strain data at a frequency of not less than 100Hz.

[0012] In some possible implementations, the method further includes determining a rebound prediction model, including:

[0013] S01. Obtain the elastic parameters and yield stress ratio of 9Ni steel. Use general finite element software to input the elastic parameters and yield stress ratio to establish a material constitutive model of 9Ni steel in order to simulate and calculate the mechanical response of 9Ni steel under external load.

[0014] S02. Select baseline parameter values, design a pressing experiment, change the input process parameters, measure the output springback to form multiple sets of process parameter and output springback mapping data sets, calibrate and verify the material parameters and contact parameters in the material constitutive model. The process parameters include end cap diameter, billet thickness, target radius of curvature, pressing speed and blank holder force. The output springback includes radius of curvature springback, circumferential angle springback and height direction springback. Determine the sensitivity index based on the process parameters and output springback, and then determine the radius of curvature springback prediction model and save it in the database.

[0015] In some possible implementations, in step S02, the thickness within the experimental range is selected. ,speed Blank pressure median value , , Using this as a baseline, the logarithm of both sides of the curvature radius springback equation is transformed into a linear function: The sensitivity index is calculated by substituting the process parameters and the output springback mapping data set into the input. The sensitivity index includes the length index. Thickness index Speed ​​Index Stress index and material constants .

[0016] In some possible implementations, in step S01, the elastic parameters and yield stress ratio of 9Ni steel are obtained by uniaxial tensile test and shear test. The elastic parameters include elastic modulus, Poisson's ratio, and shear modulus. The yield stress ratio is the ratio of the measured yield stress to the reference yield stress.

[0017] As a second aspect of the present disclosure, this disclosure provides an active adjustable curvature mold system for intelligent cold pressing of 9Ni steel spherical heads, the mold system comprising:

[0018] The upper mold includes a core block and multiple radially adjustable sector modules surrounding the core block. The back of each sector module is connected to an adjustment mechanism for adjusting the local radius of curvature.

[0019] The lower mold has a distributed sensor network embedded below its working surface. The distributed sensor network includes a pressure sensor array for real-time monitoring of the pressure distribution between the blank and the mold contact surface during the forming process.

[0020] The control system receives feedback data from a distributed sensor network, runs a rebound prediction model, and outputs control commands to control the adjustment mechanism of the sector module to achieve closed-loop control.

[0021] In some possible implementations, the working surface of the front end of the upper mold's fan-shaped module and the working surface of the lower mold are coated with a low-friction coefficient coating; and / or

[0022] The lower mold body is equipped with a fiber optic strain sensor to monitor the micro-strain of the mold itself.

[0023] The technical solution of this disclosure can achieve the following beneficial effects: This solution uses a sensor network to provide real-time feedback of pressure distribution and strain data, calls a preset springback prediction model in the database, and calculates the expected springback in each region based on real-time data. The control system drives the fine-tuning mechanism of the corresponding fan-shaped module of the upper mold, and performs reverse micro-deformation compensation on the local surface of the mold according to the predicted springback. This achieves real-time closed-loop control of sensing, prediction, and compensation, actively offsetting springback and improving the multi-dimensional linear accuracy of the end cap by an order of magnitude. The contour error can be controlled within ±0.5mm, achieving near-net-shape forming. On the other hand, it achieves precise guidance of material flow, while reducing repeated forging, effectively avoiding excessive thinning in some areas, reducing thickness unevenness by more than 50%, ensuring the thickness of the thinnest point meets design requirements, significantly reducing the number of repeated pressings, avoiding work hardening and toughness loss, ensuring the low-temperature impact toughness of the 9Ni steel end cap, significantly improving the first-press pass rate, basically eliminating the subsequent manual shaping process, and improving production efficiency.

[0024] The above overview is for illustrative purposes only and is not intended to be limiting in any way. Further aspects, embodiments, and features of this disclosure will become readily apparent from the accompanying drawings and the following detailed description, in addition to the illustrative aspects, embodiments, and features described above. Attached Figure Description

[0025] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the various drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments according to this disclosure and should not be construed as limiting the scope of this disclosure.

[0026] Figure 1 This is a schematic flowchart of the intelligent cold pressing forming method for Ni steel spherical heads according to Embodiment 9 of this disclosure;

[0027] Figure 2 This is a schematic diagram of the functional relationship of the intelligent cold pressing forming method for Ni steel spherical heads in Embodiment 9 of this disclosure;

[0028] Figure 3 This is a schematic diagram of the mold system structure according to an embodiment of the present disclosure;

[0029] Figure 4 This is a schematic diagram of the upper mold in an embodiment of this disclosure.

[0030] Reference numerals: Upper mold 10, core block 11, adjusting fan-shaped module 12, fine-tuning ejector mechanism 13, low friction coefficient coating 14;

[0031] Lower mold 20, distributed sensor network 21, fiber optic strain sensor 22;

[0032] Control system 30. Detailed Implementation

[0033] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of this disclosure. Therefore, the drawings and description are to be considered exemplary in nature and not restrictive.

[0034] Reference Figure 1 As shown in the embodiments of this disclosure, a method for intelligent cold pressing of 9Ni steel spherical heads is provided, including:

[0035] S1: Initial pressing is performed using a first blank holder force and a first pressing speed, and the pressing depth is controlled at 75%-90% of the preset depth, so that the blank initially adheres to the mold but is not fully formed. The first blank holder force is less than the target blank holder force, and the first pressing speed is less than the preset pressing speed.

[0036] S2: Increase the blank holder force to the target blank holder force and start the main pressing program. During the process of the upper die descending to the preset depth, the control system receives real-time feedback of pressure distribution and strain data from the sensor network of the lower die. It calls the preset springback prediction model in the database, calculates the expected springback in each area in combination with the real-time data, and drives the fine-tuning mechanism of the corresponding fan-shaped module of the upper die to perform reverse micro-deformation compensation on the local surface of the die according to the predicted springback.

[0037] For example, the method disclosed herein employs a two-step molding strategy, combined with real-time feedback to achieve precise control.

[0038] In step S1, preforming and stress release are employed. The pressing depth during the preforming stage is controlled at 80%-85% of the preset depth, and the first pressing speed is 50% of the preset pressing speed. For example, initial pressing can be performed using a first blank holder force of 60% or 70% of the target blank holder force, the first pressing speed can be controlled at half the preset pressing speed, and the pressing depth can be controlled at 80% or 85% of the final preset depth, allowing the billet to initially adhere to the mold but not fully form. This setup allows the material to undergo partial plastic deformation, consuming some of the elastic stored energy, releasing the initial residual stress in the billet, such as rolling residual stress, and then acquiring the actual deformation resistance distribution of the billet through a sensor network, providing measured data for the compensation parameter correction in step S2.

[0039] After the initial pressing and preforming are completed, the pressure is held for 5-10 seconds to allow the stress to fully relax. At the same time, the sensor network continuously collects pressure data to obtain the pressure decay curve and evaluate the stress relaxation characteristics of the material.

[0040] In step S2, dynamic fine pressing and real-time compensation are performed. The first blank holder force is increased to the target blank holder force, and the sensor network of the lower die provides real-time feedback of pressure distribution and strain data at a frequency of not less than 100Hz, thus initiating the main pressing program. When the upper die descends to the fine pressing stage (the last 15-20% of the stroke), the system initiates closed-loop control.

[0041] The sensor network of the lower mold provides real-time feedback of pressure distribution and strain data at a frequency of no less than 100Hz. Upon receiving the feedback, the control system immediately calls upon the springback prediction model in the database and calculates the expected springback amount for each region based on the real-time data. Subsequently, it drives the fine-tuning mechanism of the corresponding sector module of the upper mold to perform reverse micro-deformation compensation on the local surface of the mold according to the predicted springback amount.

[0042] After precision pressing, the pressure is held for another 10-15 seconds to achieve final stress homogenization. After mold opening, the molding accuracy is verified by 3D optical scanning or contact measurement, and the measurement data is fed back to the database for continuous model optimization.

[0043] The method of this disclosure uses a springback prediction model to predict springback. Data is collected via a sensor network in the lower mold, and the control system receives this data and calls the springback prediction model from the database. After calculating the expected springback amount for each region using real-time data, the system controls the fine-tuning mechanism of the corresponding fan-shaped module in the upper mold to perform reverse micro-deformation compensation on the local surface of the mold based on the predicted springback amount. This allows for real-time closed-loop control, actively offsetting springback and improving the multi-dimensional linear accuracy of the end cap, such as contour and curvature consistency, by an order of magnitude. The contour error can be controlled within ±0.5mm, achieving near-net-shape forming.

[0044] This method combines gradient blank holder force with segmented adjustable dies to achieve precise guidance of material flow. It also reduces repeated forging, effectively preventing excessive local thinning and reducing thickness unevenness by more than 50%, ensuring that even the thinnest point meets design requirements. This method significantly reduces the number of rolling passes from 5-8 times in traditional methods to 1-2 times, avoiding work hardening and toughness loss, and ensuring that the low-temperature impact toughness of the 9Ni steel head meets specifications.

[0045] This method significantly improves the first-pass yield from 60% to over 95%, essentially eliminating the need for subsequent manual shaping processes. Production efficiency is expected to increase by over 40%, and the processing cycle is shortened by 30%, greatly enhancing production efficiency. Furthermore, the database supports rapid process generation for different diameters, thicknesses, and curvatures. The same equipment can be adapted to various head production processes through software adjustments, demonstrating strong flexible manufacturing capabilities and reducing changeover time from several days to several hours.

[0046] This solution utilizes a sensor network to provide real-time feedback of pressure distribution and strain data. It then calls upon a pre-set springback prediction model from the database and calculates the expected springback amount for each region based on real-time data. The control system drives the fine-tuning mechanism of the corresponding sector module in the upper mold, compensating for local surface deformation based on the predicted springback amount. This achieves real-time closed-loop control of sensing, prediction, and compensation, actively counteracting springback and improving the multi-dimensional linear accuracy of the end cap by an order of magnitude. The contour error can be controlled within ±0.5mm, achieving near-net-shape forming. Furthermore, it enables precise guidance of material flow, reduces repeated forging, effectively avoids excessive thinning in certain areas, reduces thickness unevenness by more than 50%, ensures the thinnest point meets design requirements, significantly reduces the number of repeated pressings, avoids work hardening and toughness loss, guarantees the low-temperature impact toughness of the 9Ni steel end cap, and significantly improves the first-press pass rate. It essentially eliminates subsequent manual shaping processes, improving production efficiency.

[0047] In one disclosed embodiment, in step S2, the mold curvature compensation amount Determined according to the following formula: ,in, For database-based prediction of rebound amount, These are the model correction coefficients. To measure the actual pressure, For target pressure, This is the pressure deviation correction factor.

[0048] For example, Initial value 1.0, adaptively adjusted based on actual results. Pressure deviation correction factor. The determination can be achieved by designing two sets of pressure tests, keeping other parameters constant, using two sets of significantly different pressures. and The compression was performed, and the corresponding two sets of actual rebound amounts were measured, which represent the required compensation amount. and Then the pressure deviation correction factor can be determined. Model correction coefficients The system needs to be updated and iterated after multiple actual pressing tests. After each pressing test, the actual rebound amount is measured and substituted into the formula to calculate the coefficient. Then use the exponentially weighted moving average formula Perform a smooth update. This is the smoothing coefficient, usually taken as 0.3.

[0049] In one embodiment, the method further includes: after mold opening, verifying the molding accuracy through 3D optical scanning or contact measurement, and feeding the measurement data back to the database for continuous optimization. This allows for continuous optimization, as the measured data from each molding process can be automatically fed back to the database, enabling the model to self-improve and process knowledge to accumulate, forming a positive cycle of data, model, and process.

[0050] In some possible implementations, the method also includes determining a rebound prediction model, including:

[0051] S01. Obtain the elastic parameters and yield stress ratio of 9Ni steel. Use general finite element software to input the elastic parameters and yield stress ratio to establish a material constitutive model of 9Ni steel in order to simulate and calculate the mechanical response of 9Ni steel under external load.

[0052] S02. Select baseline parameter values, design a pressing experiment, change the input process parameters, measure the output springback to form multiple sets of process parameter and output springback mapping data sets, calibrate and verify the material parameters and contact parameters in the material constitutive model. The process parameters include end cap diameter, billet thickness, target radius of curvature, pressing speed and blank holder force. The output springback includes radius of curvature springback, circumferential angle springback and height direction springback. Determine the sensitivity index based on the process parameters and output springback, and then determine the radius of curvature springback prediction model and save it in the database.

[0053] For example, in step S01, the elastic parameters and yield stress ratio of 9Ni steel are obtained through uniaxial tensile test and shear test. The elastic parameters include elastic modulus, Poisson's ratio, and shear modulus. The yield stress ratio is the ratio of the measured yield stress to the reference yield stress.

[0054] Step S01 establishes an accurate constitutive model of 9Ni steel, which needs to consider elastic changes and recovery yield. Elastic parameters of 9Ni steel, such as elastic modulus, Poisson's ratio, shear modulus, and yield stress ratio (i.e., measured yield stress and reference yield stress), are obtained through material mechanical property tests, including uniaxial tensile and shear tests. Using general-purpose finite element software such as ABAQUS or ANSYS, the material parameters obtained from the tests are input to establish the constitutive model of 9Ni steel, simulating and calculating the mechanical response of 9Ni steel under external loads.

[0055] A springback process mapping database was constructed. By combining finite element simulations of crystal plasticity with experimental data from typical operating conditions, a database of "process parameters - springback amount" mapping relationships was established. The database covers process parameters including: head diameter... billet thickness Target radius of curvature Suppression speed Blankholding force Conduct single-point, single-path pressing experiments on simplified parts (such as standard hemispherical specimens) to measure the actual springback amount, calibrate and verify the material and contact parameters in the simulation model. Complete the sensitivity analysis of the springback principal factors (radius of curvature, thickness, pressing speed, and blank holder force) to determine the influence weight of each parameter on the springback amount, providing a basis for subsequent compensation strategies.

[0056] In step S02, the thickness within the experimental range is selected. ,speed Blank pressure median value , , Using this as a baseline, the logarithm of both sides of the curvature radius springback equation is transformed into a linear function: The sensitivity index is calculated by substituting the process parameters and the output springback mapping data set into the input. The sensitivity index includes the length index. Thickness index Speed ​​Index Stress index and material constants .

[0057] For example, the selection of the reference value is relatively flexible, and it is usually taken as the thickness within the experimental range. ,speed Blank pressure The median value is used as the benchmark value. Once selected, it needs to be fixed, and the corresponding sensitivity index is calculated based on this benchmark value. The choice of benchmark value is not limited to this and can be set according to actual requirements.

[0058] Reference Figure 2 For example, the input parameter space is a set of process parameters in a 5-dimensional vector. , The diameter of the end cap is in mm. The blank thickness is in mm; The target radius of curvature is mm; The pressing speed is mm / s; The blank holder force is kN.

[0059] Output springback is a three-dimensional vector ,in, The springback amount is the radius of curvature. This refers to the circumferential angle rebound amount. This represents the rebound amount in the height direction. Among them, and There is a mapping relationship between them. The mapping relationship between the two can be constructed by fitting crystal plasticity finite element simulation and calibration experimental data.

[0060] After selecting the baseline parameter values, a compression experiment is designed to systematically change the input parameters. This covers the required process range and observes and records the springback amount, forming multiple sets of process parameter and springback amount mapping data sets. The logarithm of both sides of the springback amount equation is then taken to transform it into a linear function: The sensitivity index can be obtained by substituting the previously recorded parameters and rebound data into the function. , , , and . This is a material constant, which can be calibrated through crystal plasticity simulation.

[0061] For example, the method disclosed herein can be divided into an input layer, an analysis layer, a decision layer, and an execution layer. The input layer is a combination of process parameters in a 5-dimensional vector. The analysis layer comprises finite element simulation analysis of crystal plasticity, microstructure analysis, and sensitivity analysis. The decision layer consists of the springback prediction model, compensation algorithm calculation, and process parameter generation. The input layer, analysis layer, and decision layer together form a springback process mapping database.

[0062] The main structure of the execution layer consists of a hydraulic master cylinder, a sector-shaped fine-tuning module, and a blank holder force control module. The execution layer can provide real-time sensing feedback and data feed back to optimize the database. The molding accuracy is verified through 3D optical scanning or contact measurement, and the measurement data is fed back to the database for continuous model optimization.

[0063] In one embodiment, the database adopts a structured storage format, supports fast querying and interpolation calculations, and can be deployed on a local server or in the cloud for multiple sets of finished equipment to share and call.

[0064] The method of this disclosure starts from the microstructure of 9Ni steel (laminate martensite + reversed austenite) and establishes a constitutive model of the material through material mechanical property experiments, which is the basis for accurate springback prediction. Furthermore, by combining simulation and experiment, a structured database covering a multi-parameter space is established, supporting real-time querying and interpolation.

[0065] By analyzing the microscopic mechanisms, the physical basis of the prediction model is formed, and rapid prediction is achieved through a database. By combining the two-step process of preforming and finishing rolling, measured correction data is obtained. By designing a closed-loop algorithm, the interconnected data form a complete technical system, transforming prediction into actual control.

[0066] Because spherical heads are generally formed by cold pressing, the high strength and toughness of 9Ni steel result in a large elastic recovery energy, making them prone to springback during cold forming. The complex multi-dimensional lines of the structure reduce pressing efficiency, make line precision control difficult, and lead to high cold processing costs and uncontrollable cycles. Existing spherical head plates are formed by cold pressing using a four-column hydraulic press and a solid sand casting mold, involving repeated pressing. Based on processing experience, repeated pressing eliminates springback deformation. The mold system of this technology has the following main defects: First, rigid molds cannot adapt to springback: traditional solid sand casting molds have fixed surfaces and cannot be dynamically adjusted according to the material's springback characteristics, requiring manual judgment of springback amount and repeated pressing after each pressing. Second, lack of real-time monitoring: existing molds do not integrate sensing systems, making it impossible to obtain stress and strain data during the forming process, resulting in a "blind pressing" state. Third, high frictional resistance: the working surface of the mold is in direct contact with the high-strength 9Ni steel billet, resulting in a high coefficient of friction, leading to uneven material flow, exacerbating thickness unevenness and localized stress concentration. Fourth, poor versatility: a set of molds can only correspond to specific specifications of end caps, resulting in high changeover costs and making it unable to meet the flexible manufacturing needs of multiple varieties and small batches.

[0067] In related technologies, the mold body is still a rigid structure, which cannot achieve active adjustment of the shape.

[0068] This invention provides a mold system capable of actively adapting to the springback characteristics of cold-pressing 9Ni steel spherical heads and achieving real-time surface compensation, thereby reducing the number of repeated pressing operations, improving molding accuracy and efficiency, and reducing processing cycle and cost. The mold system, through its "actively adjustable" and "intelligent feedback" characteristics, achieves dynamic adjustment of the mold surface, actively counteracting material springback and achieving near-net-shape forming results.

[0069] Reference Figure 3 and Figure 4 As shown, an active adjustable curvature mold system is used for intelligent cold pressing of 9Ni steel spherical heads. The mold system includes an upper mold 10, a lower mold 20, and a control system 30.

[0070] The upper die 10 is a segmented, adjustable curvature punch. The upper die body adopts a modular design, comprising a core block 11 and multiple radially adjustable sector modules 12 surrounding the core block 11. The back of each sector module 12 is connected to an adjustment mechanism for adjusting the local radius of curvature. The central core block 11 is located at the geometric center of the die, serving as a fixed structure and providing a reference positioning. The sector modules 12 are evenly distributed radially, with the number determined by the diameter of the end cap, typically 6-12 sector modules.

[0071] Each sector module 12 is connected to a high-precision electric or hydraulic fine-tuning push rod mechanism 13 on its back. This mechanism can independently and in real time fine-tune its local radius of curvature according to control commands, with an adjustment stroke typically within ±5mm and an adjustment accuracy of up to 0.01mm. The sector modules are connected by a flexible sealed connection, which ensures the continuity of the profile while allowing each module to move independently.

[0072] The front working surface of the sector module 12 is designed according to the curvature of the target head, but its initial surface has a springback compensation allowance. During the forming process, the fine-tuning push rod mechanism dynamically adjusts the extension position of each sector module according to real-time feedback data, thereby changing the local surface curvature and realizing active adaptation to the springback trend of different areas.

[0073] The upper mold, composed of a central core block 11 and multiple radially adjustable sector modules 12, achieves real-time adjustment of local curvature through an independent fine-tuning push rod mechanism, which is an important structure for realizing this method.

[0074] The lower mold 20 is a concave mold with an embedded sensor network. A distributed sensor network 21 is embedded below the working surface of the lower mold 20. The distributed sensor network 21 includes a pressure sensor array for real-time monitoring of the pressure distribution between the blank and the mold contact surface during the molding process. The distributed sensor network 21 uses thin-film pressure sensors or piezoresistive sensors, uniformly distributed along the mold surface at a density of 100 cm² / cm². 2 At least one measuring point is used to monitor the pressure distribution on the contact surface between the blank and the mold during the molding process in real time.

[0075] A fiber optic strain sensor 22 is embedded inside the lower die 20 mold body to monitor the micro-strain of the die itself, indirectly reflecting the deformation state of the billet. The distributed pressure sensor array and fiber optic strain sensor 22 in the lower die are necessary components for achieving real-time feedback.

[0076] The sensor data acquisition frequency is no less than 100Hz, and the data is transmitted to the control system via a high-speed data bus (such as EtherCAT industrial Ethernet) to provide real-time data support for rebound prediction and compensation.

[0077] The control system 30 receives feedback data from the distributed sensor network, runs a springback prediction model, and outputs control commands to control the adjustment mechanism of the sector modules to achieve closed-loop control. The mold system is equipped with a dedicated control system that receives sensor network data, runs a springback prediction algorithm, and outputs control commands to the fine-tuning ejector mechanisms of each sector module to achieve closed-loop control. The control system has a reserved communication interface with a host computer and can receive preset compensation curves.

[0078] For example, the working surface of the front end of the fan-shaped module 12 of the upper mold 10 and the working surface of the lower mold 20 are coated with a low-friction coefficient coating 14. For instance, the working surfaces of the mold (including the front end of the upper mold fan-shaped module and the lower mold surface) are coated with a low-friction coefficient coating such as a nanocomposite coating. This coating possesses the following characteristics: ultra-high hardness to ensure mold life; low coefficient of friction (…). <0.15 (compared to 9Ni steel), significantly reducing frictional resistance; good toughness, adapting to minor mold deformations without cracking. The coating thickness is controlled at 2-5μm, effectively reducing friction without affecting mold surface accuracy. Such low-friction coefficient coatings, such as nanocomposite coatings, can reduce friction and improve material flow uniformity.

[0079] The sensor network, control system and the fine-tuning actuator of the upper mold, namely the fan-shaped module, are organically combined. The segmented and adjustable fan-shaped module realizes active compensation of the surface. The sensor network acquires real-time data. The low friction coefficient coating reduces the complexity of springback caused by uneven material flow. All components are combined and work together through closed-loop control.

[0080] This mold system, by actively adjusting the profile settings of the upper mold and offsetting springback in real time, improves the multi-dimensional linear accuracy (contour consistency, curvature consistency) of the end cap by an order of magnitude, achieving near-net-shape forming and significantly increasing the first-pass yield. Furthermore, the cold-pressing method using this mold system essentially eliminates repeated pressing and subsequent manual straightening processes, with production efficiency expected to increase by over 40% and the processing cycle remaining controllable.

[0081] This mold system enables flexible manufacturing. The same mold system can be adjusted by software to adapt to the production of heads with different diameters and thicknesses, greatly shortening changeover time and adapting to multi-variety, small-batch production modes.

[0082] This mold system employs a low-friction coating to reduce mold wear, while its modular structure allows for individual module replacement in case of localized damage, reducing maintenance costs and extending mold life. Data collected by the sensor network can be used for quality traceability and process optimization, driving the intelligent and digital transformation of the molding process.

[0083] Other components of the above-described intelligent cold pressing forming method and mold system for 9Ni steel spherical heads can be adopted from various technical solutions now and in the future known to those skilled in the art, and will not be described in detail here.

[0084] In the description of this specification, it should be understood that the terms "center," "longitudinal," "transverse," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this disclosure and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this disclosure.

[0085] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this disclosure, "multiple" means two or more, unless otherwise explicitly specified.

[0086] In this disclosure, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a communication connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this disclosure according to the specific circumstances.

[0087] In this disclosure, unless otherwise expressly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature being directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature being directly above or diagonally above the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0088] The foregoing disclosure provides many different implementations or examples for carrying out different structures of this disclosure. To simplify this disclosure, the components and arrangements of specific examples are described above. Of course, these are merely examples and are not intended to limit this disclosure. Furthermore, reference numerals and / or reference letters may be repeated in different examples; such repetition is for simplification and clarity and does not in itself indicate a relationship between the various implementations and / or arrangements discussed.

[0089] The above are merely specific embodiments of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in this disclosure, and these should all be included within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the scope of the claims.

Claims

1. A method for intelligent cold pressing of 9Ni steel spherical heads, characterized in that, include: S1: Initial pressing is performed using a first blank holder force and a first pressing speed, and the pressing depth is controlled at 75%-90% of the preset depth, so that the blank initially adheres to the mold but is not fully formed. The first blank holder force is less than the target blank holder force, and the first pressing speed is less than the preset pressing speed. S2: Increase the blank holder force to the target blank holder force and start the main pressing program. During the process of the upper die descending to the preset depth, the control system receives real-time feedback of pressure distribution and strain data from the sensor network of the lower die. It calls the preset springback prediction model in the database, calculates the expected springback in each area in combination with the real-time data, and drives the fine-tuning mechanism of the corresponding fan-shaped module of the upper die to perform reverse micro-deformation compensation on the local surface of the die according to the predicted springback.

2. The intelligent cold pressing forming method for 9Ni steel spherical heads according to claim 1, characterized in that: After the initial pressing and preforming are completed, the pressure is held for 5-10 seconds to allow the stress to fully relax, and the sensor network continuously collects pressure data to obtain the pressure decay curve; and / or, after the main pressing program completes the fine pressing, the pressure is held for 10-15 seconds.

3. The intelligent cold pressing forming method for 9Ni steel spherical heads according to claim 1, characterized in that: In step S2, the mold curvature compensation amount Determined according to the following formula: ,in, For database-based prediction of rebound amount, These are the model correction coefficients. To measure the actual pressure, For target pressure, This is the pressure deviation correction factor.

4. The intelligent cold pressing forming method for 9Ni steel spherical heads according to claim 1, characterized in that: The method further includes: after mold opening, verifying the molding accuracy through optical scanning or contact measurement, and feeding the measurement data back to the database for continuous optimization.

5. The intelligent cold pressing forming method for 9Ni steel spherical heads according to claim 1, characterized in that: The pressing depth is controlled at 80%-85% of the preset depth, and / or the first pressing force is 60%-70% of the target pressing force, and / or the first pressing speed is 50% of the preset pressing speed, and / or the sensor network of the lower die provides real-time feedback of pressure distribution and strain data at a frequency of not less than 100Hz.

6. The intelligent cold pressing forming method for 9Ni steel spherical heads according to claim 1, characterized in that: The method further includes determining a rebound prediction model, including: S01. Obtain the elastic parameters and yield stress ratio of 9Ni steel. Use general finite element software to input the elastic parameters and yield stress ratio to establish a material constitutive model of 9Ni steel in order to simulate and calculate the mechanical response of 9Ni steel under external load. S02. Select baseline parameter values, design a pressing experiment, change the input process parameters, measure the output springback to form multiple sets of process parameter and output springback mapping data sets, calibrate and verify the material parameters and contact parameters in the material constitutive model. The process parameters include end cap diameter, billet thickness, target radius of curvature, pressing speed and blank holder force. The output springback includes radius of curvature springback, circumferential angle springback and height direction springback. Determine the sensitivity index based on the process parameters and output springback, and then determine the radius of curvature springback prediction model and save it in the database.

7. The intelligent cold pressing forming method for 9Ni steel spherical heads according to claim 6, characterized in that: In step S02, the thickness within the experimental range is selected. ,speed Blank pressure median value , , Using this as a baseline, the logarithm of both sides of the curvature radius springback equation is transformed into a linear function: The sensitivity index is calculated by substituting the process parameters and the output springback mapping data set into the input. The sensitivity index includes the length index. Thickness index Speed ​​Index Stress index and material constants .

8. The intelligent cold pressing forming method for 9Ni steel spherical heads according to claim 6, characterized in that: In step S01, the elastic parameters and yield stress ratio of 9Ni steel are obtained through uniaxial tensile test and shear test. The elastic parameters include elastic modulus, Poisson's ratio, and shear modulus. The yield stress ratio is the ratio of the measured yield stress to the reference yield stress.

9. An active adjustable curvature mold system for intelligent cold pressing of 9Ni steel spherical heads, characterized in that, The mold system includes: The upper mold includes a core block and multiple radially adjustable sector modules surrounding the core block. The back of each sector module is connected to an adjustment mechanism for adjusting the local radius of curvature. The lower mold has a distributed sensor network embedded below its working surface. The distributed sensor network includes a pressure sensor array for real-time monitoring of the pressure distribution between the blank and the mold contact surface during the forming process. The control system receives feedback data from a distributed sensor network, runs a rebound prediction model, and outputs control commands to control the adjustment mechanism of the sector module to achieve closed-loop control.

10. The active adjustable curvature mold system according to claim 9, characterized in that, The working surface of the front end of the upper mold's fan-shaped module and the working surface of the lower mold are coated with a low-friction coefficient coating. and / or The lower mold body is equipped with a fiber optic strain sensor to monitor the micro-strain of the mold itself.

Images