A high-pressure aluminum hydrogen storage liner pipe forging blank multi-dimensional parameter adaptive cutting method

Abstract

This invention discloses a multi-dimensional parameter adaptive cutting method for high-pressure aluminum hydrogen storage inner tube forging blanks, relating to the field of metal forging blank cutting and processing technology. The method includes: S1. For the aluminum alloy material of the target forging blank, constructing a Johnson-Cook constitutive model and a damage failure model, coupling elastoplastic deformation solution and cutting heat conduction calculation, and establishing a three-dimensional finite element simulation model for forging blank sawing; This invention establishes a high-fidelity three-dimensional finite element simulation model for forging blank sawing by constructing a Johnson-Cook constitutive model and a damage failure model for aluminum alloy forging blank cutting, and constructing an iteratively updatable digital twin based on actual equipment operating data, achieving accurate virtual-real mapping and high-fidelity simulation of the cutting process. Through offline simulation, it traverses multi-dimensional cutting parameter combinations and pre-selects the optimal parameters based on a comprehensive evaluation of multiple indicators, significantly reducing the number of actual trial cuts and the waste of valuable aluminum alloy materials.

Description

Technical Field

[0001] This invention relates to the field of metal forging blank cutting and processing technology, specifically to a multi-dimensional parameter adaptive cutting method for high-pressure aluminum hydrogen storage inner tube forging blanks. Background Technology

[0002] The hydrogen storage inner liner is a core functional component inside a high-pressure hydrogen storage container, used for guiding hydrogen gas and homogenizing the internal pressure and flow field distribution. Its inner wall smoothness, dimensional accuracy, and microstructure directly affect the filling rate, operational sealing, and long-term safe service performance of the hydrogen storage system, making it a core component of high-pressure hydrogen storage equipment. The high-pressure aluminum hydrogen storage inner liner forging blank is a tubular blank made from high-strength, hydrogen-resistant aluminum alloy through a hot forging process. It serves as the pre-processing blank for subsequent machining and surface strengthening processes to produce the finished hydrogen storage inner liner. Its material uniformity, end-face cutting quality, and microstructure directly determine the high-pressure withstand capability and hydrogen corrosion resistance of the finished tube, making it a crucial basic blank supporting the manufacturing of equipment in fields such as high-pressure solid-state hydrogen storage and vehicle-mounted high-pressure hydrogen storage. As the pressure levels of high-pressure hydrogen storage technology continue to increase, the requirements for the machining accuracy and quality consistency of the inner liner components are constantly rising. The cutting of the forging blank, as the first step in the entire processing chain, directly affects the machining allowance setting and overall yield of subsequent processes due to its machining quality and process stability.

[0003] Adaptive cutting refers to a machining system that automatically matches and dynamically adjusts cutting process parameters based on the characteristics of the billet material, real-time machining conditions, and preset quality targets, achieving stable output of qualified machining results without relying on repeated manual trial cuts. For high-value, large-size precision forgings such as high-pressure aluminum hydrogen storage inner tube forgings, adaptive cutting can effectively reduce the loss of valuable materials and shorten the production debugging cycle while ensuring the quality and dimensional accuracy of the cut end face. It has significant engineering application value for improving the manufacturing efficiency and reducing the production cost of core components for high-pressure hydrogen storage.

[0004] Currently, the debugging of cutting parameters for high-pressure aluminum hydrogen storage inner tube forgings mainly relies on actual trial cutting. Multiple blanking tests are needed to gradually adjust process parameters such as spindle speed, feed rate, and cutting depth to find the process window that meets quality requirements. This process not only consumes a large amount of forging material and machining time, resulting in high production costs and extended production cycles, but also makes it difficult to fully capture the impact of dynamic factors such as cutting heat conduction, tool wear, and feed resistance fluctuations on cutting quality. It is impossible to achieve global optimization of multi-dimensional parameters before formal production. Furthermore, when there are fluctuations in forging material between batches, it is difficult to adjust parameters in a timely manner to adapt to the changes, easily leading to unstable cutting quality and excessive tool wear. This method cannot meet the high-precision and high-consistency machining requirements of high-pressure hydrogen storage core components. Therefore, developing a multi-dimensional parameter adaptive cutting method for high-pressure aluminum hydrogen storage inner tube forgings is of great significance. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a multi-dimensional parameter adaptive cutting method for high-pressure aluminum hydrogen storage inner tube forging blanks. This method can solve the problems in the prior art where the adjustment of cutting parameters for high-pressure aluminum hydrogen storage inner tube forging blanks depends on actual trial cutting, which is costly and inefficient, and it is difficult to adapt to material fluctuations to achieve parameter adaptive optimization. This method can improve the process adjustment efficiency and processing quality stability of forging blank cutting.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a multi-dimensional parameter adaptive cutting method for high-pressure aluminum hydrogen storage inner tube forging blanks, the method comprising the following steps: Step S1: For the aluminum alloy material of the target forging billet, construct the Johnson-Cook constitutive model and the damage failure model, couple the elastoplastic deformation solution and the cutting heat conduction calculation, establish a three-dimensional finite element simulation model of forging billet sawing, simulate the material deformation, chip fracture, cutting heat conduction and mechanical response during the sawing process, and obtain the stress field, temperature field, cutting force and surface quality prediction data under different process conditions. Step S2: Build a digital twin of the cutting process. Using the three-dimensional finite element simulation model of forging billet sawing as the virtual core, collect real-time operating data of the actual cutting equipment, such as spindle power, feed resistance, equipment vibration, and cutting zone temperature, and map them to the virtual model. Compare and verify the simulation output with the actual data, correct the model parameters and boundary conditions, and iteratively update the digital twin. Step S3: Before the new batch of forging billets is put into production, offline simulation is performed on the multi-dimensional parameter combinations of spindle speed, feed rate, cutting depth and cooling medium supply conditions based on the digital twin. The parameter combinations are evaluated and ranked with cutting surface quality, cutting efficiency and tool life as evaluation indicators, and the optimal parameter combination is pre-selected as the initial process parameters for production. Step S4: During formal production, the actual processing operation data and online quality inspection data are synchronized through the digital twin to monitor deviations in the processing status. Based on the digital twin model, parameter adjustment schemes are deduced, the optimal parameter correction values ​​are matched, and feedback is sent to the cutting control system for adjustment.

[0007] Furthermore, step S1, in establishing the three-dimensional finite element simulation model of forging billet sawing, includes the following steps: Quasi-static tensile tests and Hopkinson bar tests were conducted on the target aluminum alloy material to collect material stress-strain data under different strain rates and temperatures. The constitutive parameters of the Johnson-Cook constitutive model were obtained by data fitting. Fracture tests were conducted on aluminum alloy samples under different stress triaxiality conditions. Data on the relationship between the critical strain for material failure and stress triaxiality, strain rate, and temperature were collected. Various failure parameters of the damage failure model were calibrated. The cumulative damage factor of the material during the cutting process was calculated using the following formula: ,in As a cumulative damage factor, Total loading steps For the first The equivalent plastic strain increment of the loading step, For the first The critical strain for material failure corresponding to the loading step. This is the shear fracture influence coefficient. For the first Equivalent shear stress during the loading step, For the first Equivalent stress and shear fracture influence coefficient during loading step The values ​​were determined by shear fracture tests of aluminum alloys under different stress triaxiality conditions, and the values ​​were matched to the shear fracture sensitivity characteristics of the target aluminum alloy. A three-dimensional geometric model of the forging billet and sawing tool was constructed using an adaptive mesh generation algorithm. Contact boundary conditions, thermal boundary conditions, and load conditions were set, and an elastoplastic mechanics solver and a heat conduction solver were coupled to complete the construction of a three-dimensional finite element simulation model for forging billet sawing.

[0008] Furthermore, step S2, when iteratively updating the digital twin, includes the following steps: Various sensors are installed at the spindle output end, feed actuator, tool holder mounting base and corresponding positions in the cutting area of ​​the actual cutting equipment to collect multi-source real-time operating data such as spindle power, feed resistance, tool holder vibration acceleration and cutting zone temperature during the cutting process. The real-time operation data collected is synchronously transmitted to the virtual simulation terminal through the industrial data transmission unit. The operation data is mapped to the corresponding input port of the three-dimensional finite element simulation model of forging billet sawing according to a unified timestamp, thus completing the time sequence alignment of virtual and real data. The corresponding physical quantity values ​​output by the simulation model are extracted and the deviation is calculated with the actual collected data. When the deviation exceeds the preset threshold, the material performance parameters, interface friction coefficient and thermal conductivity coefficient in the model are corrected in reverse to complete one iteration update of the model.

[0009] Furthermore, step S3, when pre-selecting the optimal parameter combination, includes the following steps: Select process parameters in four dimensions: spindle speed, feed rate, cutting depth, and cooling medium supply conditions. Set the value range and value step size for each parameter to generate a full factor parameter combination matrix. Among them, the cooling medium supply conditions are adjusted with the cooling medium flow rate as the core parameter. The parameters are input into the digital twin in sequence for offline simulation calculation, and the quantitative simulation results of the cutting surface roughness, material removal rate and tool wear corresponding to each set of parameters are output. The weighting coefficients for three evaluation indicators—cutting surface quality, cutting efficiency, and tool life—are determined using the analytic hierarchy process (AHP). The comprehensive performance evaluation value of each parameter combination is calculated using the following formula: ,in This is the comprehensive performance evaluation value of the parameter combination. This is a quantitative value for the quality of the cut surface. To quantify the cutting efficiency, This is a quantified value for tool life. , , These are the weighting coefficients for cutting surface quality, cutting efficiency, and tool life, respectively. , , A judgment matrix is ​​constructed using the analytic hierarchy process (AHP). After consistency testing, the matrix is ​​calculated and determined. The sum of the three weight coefficients is 1. The comprehensive performance evaluation values ​​of each parameter combination are ranked, and the parameter combination with the highest ranking is selected as the optimal parameter combination.

[0010] Furthermore, the three-dimensional finite element simulation model of forging billet sawing in step S1 adopts a Lagrange mesh format, the mesh in the cutting area is locally refined, the mesh element type adopts an eight-node hexahedral reduced integral element, and the cutting heat conduction calculation includes two types of heat sources: heat generated by plastic deformation and heat generated by interface friction. The calculation of heat generated by plastic deformation is based on the conversion coefficient of plastic deformation work to heat, and the calculation of heat generated by interface friction is based on the friction work and heat distribution coefficient of the interface between the tool and the forging billet. The heat generated by the two types of heat sources is transferred to the forging billet and the tool respectively according to the set distribution ratio, and the heat conduction process follows Fourier's law of heat conduction.

[0011] Furthermore, in step S2, the spindle power data is directly acquired by a power sensor, the feed resistance data is acquired by a three-dimensional force sensor installed on the feed mechanism, the equipment vibration data is acquired by a three-axis accelerometer, and the cutting zone temperature data is acquired by an infrared pyrometer. The various types of raw data are first processed by filtering and noise reduction and outlier removal. The filtering and noise reduction adopts the Butterworth low-pass filtering algorithm, and the outlier removal adopts the Laida criterion. The processed data is time-aligned with the trigger pulse of the equipment spindle as a unified time reference, and then mapped to the virtual model.

[0012] Furthermore, in step S3, the surface quality evaluation index corresponds to the surface roughness value of the cutting end face, which is calculated from the end face contour height data output by the simulation. The cutting efficiency evaluation index corresponds to the material removal volume per unit time, which is calculated from the product of the feed rate and the cutting cross-sectional area. The tool life evaluation index corresponds to the tool flank wear width after a single pass, which is calculated from the integral of the tool interface wear amount output by the simulation. The values ​​of the three evaluation indices are directly output from the simulation results of the digital twin, and the values ​​of each index are all dimensionless normalized before participating in the comprehensive performance evaluation.

[0013] Furthermore, the online quality inspection data in step S4 is acquired by a vision inspection unit installed on the discharge side of the cutting equipment. The vision inspection unit acquires a two-dimensional image of the cut end face and extracts quality parameters such as surface roughness, cut perpendicularity, and cut burr height of the end face through an image processing algorithm. The image processing algorithm includes three processing steps: image preprocessing, edge extraction, and feature calculation. Image preprocessing includes grayscale transformation and image enhancement operations. Edge extraction adopts a sub-pixel edge detection algorithm. Feature calculation is based on the extracted end face contour data for quantitative calculation. The acquisition cycle of the vision inspection unit is synchronized with the unloading cycle of the cutting equipment. A set of end face quality data is acquired for each cut completed.

[0014] Furthermore, in step S4, the parameter adjustment scheme is derived based on the current operating parameters. Multiple candidate parameter combinations are generated according to a set adjustment step size. The parameter adjustment step size is set according to the accuracy of each parameter value. The spindle speed adjustment step size corresponds to the spindle speed adjustment accuracy, the feed speed adjustment step size corresponds to the feed mechanism speed adjustment accuracy, the cutting depth adjustment step size corresponds to the tool holder feed positioning accuracy, and the cooling medium supply condition adjustment step size corresponds to the cooling valve flow rate adjustment accuracy. The generation of candidate parameter combinations follows a priority order: spindle speed adjustment first, feed speed second, cutting depth third, and cooling medium supply condition last. Each candidate parameter combination is input into a digital twin for rapid simulation verification, and the corresponding quality prediction result is output. The target cost function value of each candidate scheme is calculated using the following formula: ,in Adjust the objective cost function value of the candidate parameter adjustment scheme. This represents the predicted cutting quality value corresponding to the candidate solution. The preset cutting quality target value, Adjust the penalty coefficient for the parameters. For the first The adjustment weighting coefficient of each process parameter, For the first The adjustment amount of a process parameter relative to the current operating parameters, and the parameter adjustment penalty coefficient. The adjustment weighting coefficients for each parameter are set according to the stability requirements of equipment operation and are used to balance quality deviation and parameter fluctuation amplitude. The optimal parameter correction value is determined by calculating the sensitivity of the influence of each process parameter on the cutting quality. The higher the sensitivity, the greater the corresponding weight. The candidate parameter combination with the smallest objective cost function value is selected as the optimal parameter correction value.

[0015] Furthermore, before step S3 is executed, material sampling inspection is performed on the new batch of forging blanks. The sampling ratio for material sampling inspection is set according to the batch size of the forging blank batch. The number of samples for each batch of forging blanks is not less than three. The inspection items include chemical composition spectral analysis, Brinell hardness test, and room temperature tensile test. The average value of the parameters obtained from the inspection is used as the representative material parameters of the batch of forging blanks. The material parameters obtained from the inspection are input into the digital twin to update the material constitutive parameters and damage failure parameters in the three-dimensional finite element simulation model of forging blank sawing. After the material parameters are updated, offline simulation traversal and optimization operation of multi-dimensional parameter combination is performed.

[0016] Compared with existing technologies, this multi-dimensional parameter adaptive cutting method for high-pressure aluminum hydrogen storage inner tube forging billets has the following advantages: This invention constructs a Johnson-Cook constitutive model and a damage and failure model for cutting aluminum alloy forging billets, establishes a high-fidelity three-dimensional finite element simulation model for forging billet sawing, and builds an iteratively updatable digital twin based on actual equipment operating data. This achieves accurate virtual-real mapping and high-fidelity simulation of the cutting process. Through offline simulation, it traverses multi-dimensional cutting parameter combinations and pre-selects the optimal parameters based on comprehensive evaluation of multiple indicators, significantly reducing the number of actual trial cuts and the waste of valuable aluminum alloy materials. Through real-time extrapolation and dynamic parameter adjustment of the twin during the production process, it adapts to fluctuations in forging billet material and changes in working conditions, achieving adaptive optimization of cutting parameters, stably ensuring cutting quality and processing efficiency, extending tool life, and meeting the high-precision and high-consistency processing requirements of high-pressure hydrogen storage inner liner forging billets.

[0017] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0019] Figure 1 A flowchart of a multi-dimensional parameter adaptive cutting method for a high-pressure aluminum hydrogen storage inner tube forging blank; Figure 2 This is a flowchart of the process for creating a three-dimensional finite element simulation model of forging billet sawing; Figure 3 This is a flowchart of the process of iteratively updating the digital twin. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] This invention provides a multi-dimensional parameter adaptive cutting method for high-pressure aluminum hydrogen storage inner tube forgings. Based on the constitutive and damage models of aluminum alloy materials, a high-fidelity three-dimensional finite element simulation model for sawing the forging is constructed. Combined with actual equipment operating data, an iteratively updatable digital twin of the cutting process is built. Through offline simulation, multi-dimensional process parameter combinations are traversed, and the optimal initial parameters are pre-selected through comprehensive evaluation of multiple indicators. During production, the processing status is monitored in real time through the twin, and parameters are dynamically deduced and adjusted, achieving adaptive optimization of parameters in the forging cutting process. The method comprehensively covers four core stages: simulation modeling, twin construction, offline optimization, and online control, forming a complete closed-loop processing control system. A detailed description is provided below with reference to specific embodiments.

[0022] This embodiment applies to the sawing and blanking process of the inner liner tube forging blank in a high-pressure hydrogen storage equipment manufacturing production line. The processing object is a tubular forging blank made of high-strength hydrogen-resistant aluminum alloy, with an outer diameter of 600mm, a wall thickness of 20mm, and a single blank length of 1000mm. The processing equipment is a CNC metal band saw. The processing objective is to obtain blanking sections of forging blanks with uniform end face quality and stable dimensional accuracy, while also considering processing efficiency and tool wear control. The overall processing control flow is as follows: Figure 1 As shown, a three-dimensional finite element simulation model of the forging billet sawing process is first constructed, and then the corresponding digital twin is built and iterative verification is completed. Before the new batch of forging billets is put into production, the parameters are optimized through offline simulation. During the formal production process, the working condition data is synchronized in real time and the process parameters are dynamically adjusted until the entire batch of forging billets is cut.

[0023] The process of constructing a three-dimensional finite element simulation model for forging billet sawing is as follows: Figure 2 As shown, mechanical property tests were first conducted on the target aluminum alloy material to obtain the basic parameters required for constructing the constitutive model and damage model. Quasi-static tensile tests and Hopkinson bar tests were performed on the target aluminum alloy material, with the strain rate set to 0.001 s-1 for the quasi-static tensile tests. -1 0.01s-1 0.1s -1 Three levels, with strain rates covering 100 s-100 in the Hopkinson bar test. -1 up to 5000s -1 Within a range, three parallel specimens were set up under each strain rate condition, and the test temperatures were set to three levels: 25℃, 200℃, and 400℃. Material stress-strain data were collected under different strain rates and temperatures. The test data were fitted using the nonlinear least squares method to obtain the constitutive parameters of the Johnson-Cook constitutive model.

[0024] Subsequently, fracture performance tests were conducted on the aluminum alloy material. Standard specimens with different notch shapes were designed, and different stress triaxiality levels were obtained by changing the notch curvature radius. Tensile fracture tests were carried out under different stress triaxiality conditions, and data on the correspondence between the material's critical failure strain and stress triaxiality, strain rate, and temperature were collected to calibrate various failure parameters of the damage failure model. In the specific implementation of this embodiment, the cumulative damage factor of the material during the cutting process is calculated using the following formula: ,in As a cumulative damage factor, For the first The equivalent plastic strain increment of the loading step, For the first The critical strain for material failure corresponding to the loading step. This is the shear fracture influence coefficient. For the first Equivalent shear stress during the loading step, For the first Equivalent stress of the loading step.

[0025] Shear fracture influence coefficient The shear fracture coefficient was determined through shear fracture tests of aluminum alloys under different stress triaxialities. Specifically, the stress triaxiality was defined as the ratio of hydrostatic pressure to equivalent stress. Three typical stress triaxiality levels (0, 0.33, and 0.66) were obtained by designing smooth cylindrical specimens, single-notch tensile specimens, and double-notch shear specimens, respectively. Five repeated fracture tests were conducted at each stress triaxiality level, and the critical strain value for failure was recorded for each test. A linear fit was performed with stress triaxiality as the abscissa and the reciprocal of the critical strain as the ordinate. The slope of the fitted line is the shear fracture influence coefficient. The value of is ultimately obtained The value matches the shear fracture sensitivity characteristics of the target aluminum alloy, which can accurately reflect the promoting effect of shearing on material fracture during sawing.

[0026] After material parameter calibration, a three-dimensional geometric model of the forging billet and sawing tool was constructed using an adaptive mesh generation algorithm. The tool model adopted the actual three-dimensional contour with serrated teeth, and the forging billet model was constructed according to the dimensional parameters of the actual tubular forging billet. The mesh in the cutting contact area was locally refined, and the mesh element type adopted was an eight-node hexahedral reduced integral element. The non-cutting area adopted a relatively sparse mesh to control the computational scale. Then, contact boundary conditions, thermal boundary conditions, and load conditions were set. The contact interface adopted a Coulomb friction model, the thermal boundary adopted convection heat dissipation conditions, and the load conditions were set according to the actual sawing feed rate and spindle speed.

[0027] The heat transfer calculation for cutting includes two types of heat sources: heat generated by plastic deformation and heat generated by interfacial friction. The calculation of heat generated by plastic deformation is based on the conversion coefficient of the work done during plastic deformation to heat, while the calculation of heat generated by interfacial friction is based on the frictional work and heat distribution coefficient at the interface between the tool and the forging blank. The heat generated by both types of heat sources is transferred to the forging blank and the tool respectively according to a set distribution ratio, and the heat transfer process follows Fourier's law of heat conduction. A coupled elastoplastic mechanics solver and a heat transfer solver are used to build a three-dimensional finite element simulation model of forging blank sawing.

[0028] The process of building a digital twin of the cutting process and completing iterative updates is as follows: Figure 3 As shown, various sensors are first installed at corresponding positions on the actual cutting equipment. A power sensor is installed at the spindle output end to collect spindle power data, a three-dimensional force sensor is installed on the feed actuator to collect feed resistance data, a three-axis accelerometer is installed on the tool holder mounting base to collect equipment vibration data, and an infrared pyrometer is installed on the side of the cutting area to collect cutting area temperature data. The sampling frequency of all sensors is uniformly set to 1000Hz to ensure that dynamic fluctuation signals during the cutting process can be captured.

[0029] The collected raw data first undergoes filtering, noise reduction, and outlier removal. Butterworth low-pass filtering is used for noise reduction, and the Laida criterion is used for outlier removal. The processed data then undergoes time-series alignment and mapping. The processed real-time operating data is synchronously transmitted to the virtual simulation terminal via an industrial data transmission unit. Data transmission uses the industrial Ethernet protocol, with transmission latency controlled within 10ms. The operating data is mapped to the corresponding input ports of the three-dimensional finite element simulation model of forging billet sawing according to a unified timestamp. The timing alignment of virtual and real data is completed using the trigger pulse of the equipment spindle as a unified time reference, ensuring that physical quantities at the same moment can accurately correspond.

[0030] The corresponding physical quantity values ​​output by the simulation model are extracted and the deviation is calculated with the actual collected data. The relative deviation values ​​of three core physical quantities, namely spindle power, feed resistance, and cutting zone temperature, are calculated respectively. When the deviation of any physical quantity exceeds the preset threshold, the model parameter correction process is initiated to correct the material performance parameters, interface friction coefficient and thermal conductivity coefficient in the model in reverse. The correction process adopts the gradient descent method and iteratively adjusts the parameter values ​​with the goal of minimizing the deviation, completing one iteration update of the model. After multiple iterations, the deviation between the model output and the actual data stabilizes within the allowable range, resulting in a digital twin that highly matches the actual processing scenario.

[0031] Before a new batch of forging billets is put into formal production, material sampling and testing are carried out on the batch of forging billets. The sampling ratio for material sampling and testing is set according to the batch size of the forging billet batch. The number of samples for each batch of forging billets is no less than three pieces. The testing items include chemical composition spectral analysis, Brinell hardness testing, and room temperature tensile testing. The average value of the parameters obtained from the tests is used as the representative material parameters of the batch of forging billets. The material parameters obtained from the tests are input into the digital twin to update the material constitutive parameters and damage and failure parameters in the three-dimensional finite element simulation model of forging billet sawing. After the material parameters are updated, offline simulation traversal and optimization operations of multi-dimensional parameter combinations are performed.

[0032] Four process parameters were selected: spindle speed, feed rate, cutting depth, and cooling medium supply conditions. Among them, the cooling medium supply conditions were adjusted with the cooling medium flow rate as the core parameter. The value range and step size of each parameter were set. The spindle speed range was 20 m / min to 60 m / min with a step size of 5 m / min. The feed rate range was 10 mm / s to 30 mm / s with a step size of 2 mm / s. The cutting depth range was 5 mm to 15 mm with a step size of 2 mm. The cooling medium flow rate range was 5 L / min to 15 L / min with a step size of 2 L / min. A parameter combination matrix consisting of all parameter combinations was generated according to the full factorial experimental design.

[0033] Each parameter combination is sequentially input into the digital twin for offline simulation calculation. The simulation duration for each parameter combination covers a complete single sawing pass. Quantitative simulation results for the cutting surface roughness, material removal rate, and tool wear corresponding to each parameter combination are output. The cutting surface quality evaluation index corresponds to the surface roughness value of the cutting end face, which is calculated from the end face contour height data output by the simulation. The cutting efficiency evaluation index corresponds to the material removal volume per unit time, which is calculated by multiplying the feed rate and the cutting cross-sectional area. The tool life evaluation index corresponds to the tool flank wear width after a single pass, which is calculated by integrating the tool interface wear amount output by the simulation. The values ​​of all three evaluation indices are directly output from the simulation results of the digital twin. The values ​​of each index are dimensionless normalized before being used in the comprehensive performance evaluation.

[0034] In the specific implementation of this embodiment, the comprehensive performance evaluation value of each parameter combination is calculated according to the following formula: ,in This is the comprehensive performance evaluation value of the parameter combination. This is a quantitative value for the quality of the cut surface. To quantify the cutting efficiency, This is a quantified value for tool life. , , These are the weighting coefficients corresponding to cutting surface quality, cutting efficiency, and tool life, respectively.

[0035] Weighting coefficient , , A judgment matrix was constructed using the Analytic Hierarchy Process (AHP), and its final value was determined after a consistency check. The specific calculation process was as follows: First, an AHP structure was constructed. The target layer represented the comprehensive performance of the cutting process, while the criterion layer included three indicators: cutting surface quality, cutting efficiency, and tool life. Five engineers and simulation technicians with over 10 years of experience in forging blank cutting processes were invited to pairwise score the relative importance of the three indicators using a 1-9 scale, thus constructing a third-order judgment matrix. The largest eigenvalue and corresponding eigenvector of the judgment matrix were calculated. Simultaneously, the ratio of the consistency index to the random consistency index (i.e., the consistency ratio) was calculated. When the consistency ratio was less than 0.1, the judgment matrix was considered to have satisfactory consistency. In this case, the three components of the eigenvector corresponded to the weight coefficients of the three indicators, and the sum of the three weight coefficients was 1. The comprehensive performance evaluation values ​​of each parameter combination were ranked, and the parameter combination with the highest ranking was selected as the optimal parameter combination, serving as the initial process parameters for the actual production of this batch of forging blanks.

[0036] During the formal cutting production process, the optimal parameter combination obtained from the selection is loaded into the cutting equipment control system, and the equipment is started to begin the forging billet cutting operation. A vision inspection unit installed on the discharge side of the cutting equipment synchronously collects quality data of the cut end face. The acquisition cycle of the vision inspection unit is synchronized with the unloading cycle of the cutting equipment, and a set of end face quality data is collected after each cut. The vision inspection unit acquires a two-dimensional image of the cut end face and extracts quality parameters such as surface roughness, cut perpendicularity, and cut burr height through image processing algorithms. The image processing algorithm includes three processing stages: image preprocessing, edge extraction, and feature calculation. Image preprocessing includes grayscale transformation and image enhancement operations. Edge extraction uses a sub-pixel edge detection algorithm, and feature calculation is based on the extracted end face contour data for quantization calculation. The obtained quality parameters are transmitted to the digital twin in real time.

[0037] By synchronizing actual machining operation data with online quality inspection data through a digital twin, deviations between cutting status and quality indicators are continuously monitored. When factors such as fluctuations in forging billet material or normal tool wear cause the machining status to deviate from the preset quality target, an adaptive parameter adjustment process is initiated. The parameter adjustment scheme is derived based on the current operating parameters, generating multiple sets of candidate adjustment parameter combinations according to the set adjustment step size. The parameter adjustment step size is set according to the accuracy of each parameter value. The adjustment step size of the spindle speed corresponds to the speed adjustment accuracy of the equipment spindle, the adjustment step size of the feed rate corresponds to the speed adjustment accuracy of the feed mechanism, the adjustment step size of the cutting depth corresponds to the feed positioning accuracy of the tool holder, and the adjustment step size of the cooling medium supply condition corresponds to the flow rate adjustment accuracy of the cooling valve. The generation of candidate parameter combinations follows the priority order of spindle speed adjustment first, feed rate second, cutting depth third, and cooling medium supply condition last. Parameters are adjusted sequentially according to priority to generate different candidate combinations.

[0038] In the specific implementation of this embodiment, the target cost function value of each candidate solution is calculated according to the following formula: ,in Adjust the objective cost function value of the candidate parameter adjustment scheme. This represents the predicted cutting quality value corresponding to the candidate solution. The preset cutting quality target value, Adjust the penalty coefficient for the parameters. For the first The adjustment weighting coefficient of each process parameter, For the first The adjustment amount of each process parameter relative to the current operating parameters.

[0039] Parameter adjustment penalty coefficient The parameter adjustment penalty coefficient is set based on the stability requirements of equipment operation and is used to balance quality deviation and parameter fluctuation. Specifically, it is determined by statistically analyzing the equipment failure rate and quality fluctuation coefficient corresponding to different parameter fluctuation amplitudes during historical production processes. Using the constraint of achieving equipment operational stability, the optimal balance coefficient between quality deviation and parameter fluctuation is fitted, which is the parameter adjustment penalty coefficient. The values ​​of . The adjustment weight coefficients for each parameter. The sensitivity of each process parameter on the cutting quality is determined by calculation. Specifically, the calculation process involves keeping the other three parameters constant while changing the unit adjustment step size of one parameter. The corresponding change in cutting quality is calculated using digital twin simulation. The ratio of the quality change to the parameter change is the sensitivity value of that parameter; higher sensitivity corresponds to greater weight. After normalizing the sensitivity values ​​of the four parameters, the adjustment weight coefficient for each parameter is obtained. .

[0040] Each candidate parameter combination is input into a digital twin for rapid simulation verification, outputting the corresponding quality prediction results and calculating the corresponding objective cost function value. The candidate parameter combination with the smallest objective cost function value is selected as the optimal parameter correction value. The corrected parameters are fed back to the cutting control system for adjustment. After adjustment, the cutting operation continues. The above status monitoring and parameter adjustment process is repeated until the entire forging billet is cut, and then the next forging billet is cut, until the entire batch of forging billets is processed.

[0041] In summary, through the complete implementation of this embodiment, offline pre-optimization and online adaptive adjustment of cutting parameters for high-pressure aluminum hydrogen storage inner liner tube forgings can be achieved based on digital twins. Optimal process parameters suitable for material characteristics can be obtained without relying on a large number of actual trial cuts, effectively reducing the loss of valuable aluminum alloy materials and the time required for process debugging. During production, the system can adapt to changes in working conditions caused by material fluctuations and tool wear in real time, stably maintaining the quality and dimensional accuracy of the cut end face. Simultaneously, it can reasonably control the tool wear rate, extend tool life, and comprehensively improve the processing efficiency and quality consistency of the forging process, meeting the high-precision and high-stability processing requirements of high-pressure hydrogen storage inner liner tube forgings.

[0042] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A multi-dimensional parameter adaptive cutting method for high-pressure aluminum hydrogen storage inner tube forging billets, characterized in that, The method includes the following steps: Step S1: For the aluminum alloy material of the target forging billet, construct the Johnson-Cook constitutive model and the damage failure model, couple the elastoplastic deformation solution and the cutting heat conduction calculation, establish a three-dimensional finite element simulation model of forging billet sawing, simulate the material deformation, chip fracture, cutting heat conduction and mechanical response during the sawing process, and obtain the stress field, temperature field, cutting force and surface quality prediction data under different process conditions. Step S2: Build a digital twin of the cutting process. Using the three-dimensional finite element simulation model of forging billet sawing as the virtual core, collect real-time operating data of the actual cutting equipment, such as spindle power, feed resistance, equipment vibration, and cutting zone temperature, and map them to the virtual model. Compare and verify the simulation output with the actual data, correct the model parameters and boundary conditions, and iteratively update the digital twin. Step S3: Before the new batch of forging billets is put into production, offline simulation is performed on the multi-dimensional parameter combinations of spindle speed, feed rate, cutting depth and cooling medium supply conditions based on the digital twin. The parameter combinations are evaluated and ranked with cutting surface quality, cutting efficiency and tool life as evaluation indicators, and the optimal parameter combination is pre-selected as the initial process parameters for production. Step S4: During formal production, the actual processing operation data and online quality inspection data are synchronized through the digital twin to monitor deviations in the processing status. Based on the digital twin model, parameter adjustment schemes are deduced, the optimal parameter correction values ​​are matched, and feedback is sent to the cutting control system for adjustment.

2. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, Step S1, when establishing a three-dimensional finite element simulation model for forging billet sawing, includes the following steps: Quasi-static tensile tests and Hopkinson bar tests were conducted on the target aluminum alloy material to collect material stress-strain data under different strain rates and temperatures. The constitutive parameters of the Johnson-Cook constitutive model were obtained by data fitting. Fracture tests were conducted on aluminum alloy samples under different stress triaxiality conditions. Data on the relationship between the critical strain for material failure and stress triaxiality, strain rate, and temperature were collected. Various failure parameters of the damage failure model were calibrated. The cumulative damage factor of the material during the cutting process was calculated using the following formula: ,in As a cumulative damage factor, For the first The equivalent plastic strain increment of the loading step, For the first The critical strain for material failure corresponding to the loading step. This is the shear fracture influence coefficient. For the first Equivalent shear stress during the loading step, For the first Equivalent stress during the loading step; A three-dimensional geometric model of the forging billet and sawing tool was constructed using an adaptive mesh generation algorithm. Contact boundary conditions, thermal boundary conditions, and load conditions were set, and an elastoplastic mechanics solver and a heat conduction solver were coupled to complete the construction of a three-dimensional finite element simulation model for forging billet sawing.

3. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, Step S2, when iteratively updating the digital twin, includes the following steps: Various sensors are installed at the spindle output end, feed actuator, tool holder mounting base and corresponding positions in the cutting area of ​​the actual cutting equipment to collect multi-source real-time operating data such as spindle power, feed resistance, tool holder vibration acceleration and cutting zone temperature during the cutting process. The real-time operation data collected is synchronously transmitted to the virtual simulation terminal through the industrial data transmission unit. The operation data is mapped to the corresponding input port of the three-dimensional finite element simulation model of forging billet sawing according to a unified timestamp, thus completing the time sequence alignment of virtual and real data. The corresponding physical quantity values ​​output by the simulation model are extracted and the deviation is calculated with the actual collected data. When the deviation exceeds the preset threshold, the material performance parameters, interface friction coefficient and thermal conductivity coefficient in the model are corrected in reverse to complete one iteration update of the model.

4. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, Step S3, when pre-selecting the optimal parameter combination, includes the following steps: Select process parameters in four dimensions: spindle speed, feed rate, cutting depth, and cooling medium supply conditions. Set the value range and value step size for each parameter to generate a full factor parameter combination matrix. Among them, the cooling medium supply conditions are adjusted with the cooling medium flow rate as the core parameter. The parameters are input into the digital twin in sequence for offline simulation calculation, and the quantitative simulation results of the cutting surface roughness, material removal rate and tool wear corresponding to each set of parameters are output. The weighting coefficients for three evaluation indicators—cutting surface quality, cutting efficiency, and tool life—are determined using the analytic hierarchy process (AHP). The comprehensive performance evaluation value of each parameter combination is calculated using the following formula: ,in This is the comprehensive performance evaluation value of the parameter combination. This is a quantitative value for the quality of the cut surface. To quantify the cutting efficiency, This is a quantified value for tool life. , , The weighting coefficients corresponding to cutting surface quality, cutting efficiency, and tool life are used to rank the comprehensive performance evaluation values ​​of each parameter combination, and the parameter combination with the highest ranking is selected as the optimal parameter combination.

5. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, The three-dimensional finite element simulation model of forging billet sawing in step S1 adopts the Lagrange mesh format, the mesh in the cutting area is locally refined, and the mesh element type adopts the eight-node hexahedral reduced integral element. The cutting heat conduction calculation includes two types of heat sources: heat generated by plastic deformation and heat generated by interface friction. The calculation of heat generated by plastic deformation is based on the conversion coefficient of plastic deformation work to heat, and the calculation of heat generated by interface friction is based on the friction work and heat distribution coefficient of the interface between the tool and the forging billet. The heat generated by the two types of heat sources is transferred to the forging billet and the tool respectively according to the set distribution ratio. The heat conduction process follows Fourier's law of heat conduction.

6. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, In step S2, the spindle power data is directly acquired by a power sensor, the feed resistance data is acquired by a three-dimensional force sensor installed on the feed mechanism, the equipment vibration data is acquired by a three-axis accelerometer, and the cutting zone temperature data is acquired by an infrared pyrometer. The acquired raw data are first processed by filtering and noise reduction and outlier removal. The filtering and noise reduction adopts the Butterworth low-pass filtering algorithm, and the outlier removal adopts the Laida criterion. The processed data is time-aligned with the trigger pulse of the equipment spindle as a unified time reference, and then mapped to the virtual model.

7. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, The surface quality evaluation index in step S3 corresponds to the surface roughness value of the cutting end face. The surface roughness value is calculated from the end face contour height data output by the simulation. The cutting efficiency evaluation index corresponds to the material removal volume per unit time. The material removal volume is calculated by multiplying the feed rate and the cutting cross-sectional area. The tool life evaluation index corresponds to the tool flank wear width after a single pass. The tool flank wear width is calculated by integrating the tool interface wear amount output by the simulation. The values ​​of the three evaluation indices are directly output from the simulation results of the digital twin. The values ​​of each index are all dimensionless normalized before participating in the comprehensive performance evaluation.

8. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, The online quality inspection data in step S4 is acquired by a vision inspection unit installed on the discharge side of the cutting equipment. The vision inspection unit acquires a two-dimensional image of the cut end face and extracts quality parameters such as surface roughness, cut perpendicularity, and cut burr height through an image processing algorithm. The image processing algorithm includes three processing steps: image preprocessing, edge extraction, and feature calculation. Image preprocessing includes grayscale transformation and image enhancement operations. Edge extraction uses a sub-pixel edge detection algorithm. Feature calculation is based on the extracted end face contour data and is quantized. The acquisition cycle of the vision inspection unit is synchronized with the unloading cycle of the cutting equipment. A set of end face quality data is acquired for each cut.

9. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, In step S4, the parameter adjustment scheme is derived based on the current operating parameters. Multiple candidate parameter combinations are generated according to the set adjustment step size. The parameter adjustment step size is set according to the accuracy of each parameter value. The spindle speed adjustment step size corresponds to the spindle speed adjustment accuracy, the feed speed adjustment step size corresponds to the feed mechanism speed adjustment accuracy, the cutting depth adjustment step size corresponds to the tool holder feed positioning accuracy, and the cooling medium supply condition adjustment step size corresponds to the cooling valve flow rate adjustment accuracy. The generation of candidate parameter combinations follows the priority order: spindle speed adjustment first, feed speed second, cutting depth third, and cooling medium supply condition last. Each candidate parameter combination is input into a digital twin for rapid simulation verification, and the corresponding quality prediction result is output. The target cost function value of each candidate scheme is calculated using the following formula: ,in Adjust the objective cost function value of the candidate parameter adjustment scheme. This represents the predicted cutting quality value corresponding to the candidate solution. The preset cutting quality target value, Adjust the penalty coefficient for the parameters. For the first The adjustment weighting coefficient of each process parameter, For the first The adjustment amount of each process parameter relative to the current operating parameters is used to select the candidate parameter combination with the smallest objective cost function value as the optimal parameter correction value.

10. The method for adaptive cutting of high-pressure aluminum hydrogen storage inner tube forging billets using multi-dimensional parameters according to claim 1, characterized in that, Before step S3 is executed, material sampling inspection is performed on the new batch of forging blanks. The sampling ratio for material sampling inspection is set according to the batch size of the forging blank batch. The number of samples for each batch of forging blanks is not less than three. The inspection items include chemical composition spectral analysis, Brinell hardness test, and room temperature tensile test. The average value of the parameters obtained from the inspection is used as the representative material parameters of the batch of forging blanks. The material parameters obtained from the inspection are input into the digital twin to update the material constitutive parameters and damage failure parameters in the three-dimensional finite element simulation model of forging blank sawing. After the material parameters are updated, the offline simulation traversal and optimization operation of multi-dimensional parameter combination is performed.

Images