Aluminum alloy precision forging forming process parameter self-adaptive optimization system and method

Abstract

The invention discloses a self-adaptive optimization system and method for aluminum alloy precision forging forming process parameters, and relates to the technical field of metal plastic forming. According to the system, a multi-point array type active flexible die holder is arranged below a lower die, and a dynamic actuating unit formed by connecting a magneto-rheological damper and a piezoelectric stack in series is integrated in the multi-point array type active flexible die holder; the acoustic-mechanical coupling control unit controls all the actuating units to generate a static prestress field, ultrasonic guided waves are guided to be focused to the metal flowing front edge through the acoustic elastic effect, normal sliding, interface bonding and rheological locking states are recognized based on broadband impedance spectrum inversion data of echo signals, and the metal flowing front edge is determined according to the bonding state. The system executes high-frequency micro-motion unbonding under high-rigidity support; and for the locking state, trace retraction compliance compensation under low-rigidity support is executed. Dynamic scanning of the internal state of the mold can be achieved without moving a probe, and the forming quality is remarkably improved by actively adjusting boundary conditions.

Description

Technical Field

[0001] This invention relates to the field of metal plastic forming technology, specifically to an adaptive optimization system and method for aluminum alloy precision forging forming process parameters. Background Technology

[0002] Precision forging technology for aluminum alloys has been widely used in high-end equipment fields such as aerospace, rail transportation, and lightweight automotive manufacturing due to its ability to produce complex structural parts with high specific strength and excellent mechanical properties. In current industrial production, to ensure the forming accuracy and internal quality of forgings, finite element numerical simulation technology is typically used to pre-design and verify the mold structure and process parameters. During actual processing, displacement and pressure sensors installed on the forging press mainly rely on collecting macroscopic load-displacement curves, or fixed acoustic sensors installed at specific locations outside the mold to monitor signal changes during the forming process. This indirectly determines the filling state inside the mold and the deformation behavior of the billet, and adjusts basic process parameters such as forging speed and holding time accordingly.

[0003] However, existing monitoring methods have significant limitations when dealing with precision forging tasks with complex cavity structures. Because the position of the metal flow front changes in real time during aluminum alloy forging, and the irregular curved surface structure of complex molds hinders signal propagation, traditional sensors fixed on the outer surface of the mold cannot flexibly adjust their detection path according to the forming process, making it difficult to accurately guide the detection beam to the dynamically changing critical deformation area. This spatial misalignment between the fixed sensor position and the dynamic migration of the metal flow area prevents the system from acquiring real-time information on the actual contact state and internal stress distribution of the metal flow front. This results in a large monitoring blind spot inside the mold, making it difficult to achieve precise dynamic optimization of forming process parameters based on changes in local rheological states. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an adaptive optimization system and method for aluminum alloy precision forging process parameters, solving the problem of dynamic detection blind spots in the forming state inside complex molds.

[0005] To achieve the above objectives, the present invention provides an adaptive optimization system for aluminum alloy precision forging process parameters. This system includes a precision forging press body, an upper die assembly, a lower die assembly, a multi-point array active compliant die base, an ultrasonic guided wave transceiver array, and an acoustic-mechanical coupling control unit. The upper die assembly is connected to the slider component of the precision forging press body, and the lower die assembly is correspondingly positioned below the upper die assembly. The multi-point array active compliant die base is installed between the base worktable of the precision forging press body and the lower die assembly, serving as a support base for the lower die assembly. The multi-point array active compliant die base integrates several independently controlled dynamic actuation units, which are distributed across various stress-bearing areas covering the bottom of the lower die assembly. The ultrasonic guided wave transceiver array is positioned on the outer wall surface or the bottom non-working surface of the lower die assembly, used to excite ultrasonic guided waves into the structure of the lower die assembly and collect the reflected sound wave signals. The acoustic-mechanical coupling control unit is electrically connected to the main body of the precision forging press, the various dynamic actuation units in the multi-point array active compliant mold base, and the ultrasonic guided wave transceiver array.

[0006] In terms of the specific actuator construction, each dynamic actuation unit consists of a magnetorheological damper and a piezoelectric stack actuator connected in series along the axial direction. The fixed end of the magnetorheological damper is fixedly connected to the base plate of the multi-point array active compliant mold base, and its piston rod output end is fixedly connected to the bottom surface of the piezoelectric stack actuator. The top output end of the piezoelectric stack actuator is fixedly connected to the bottom surface of the lower mold assembly through a ball joint connection assembly. The magnetorheological damper uses excitation current to adjust its axial support stiffness and damping coefficient, while the piezoelectric stack actuator is configured to generate the static preload displacement required to construct the prestressed field or to generate high-frequency mechanical vibration.

[0007] This invention employs an acoustoelastic stress field control mechanism to focus the ultrasonic detection beam. The acoustic-mechanical coupling control unit calculates and controls differentiated static elongation displacements of each dynamic actuation unit based on the acoustoelastic effect—the correlation between sound wave propagation velocity and internal stress state within the material. This operation creates a non-uniform static prestress field within the lower mold assembly, causing a gradient distribution of sound wave propagation velocity within the lower mold assembly material, forcing refraction and deflection of the ultrasonic guided wave propagation path. By adjusting the distribution of the prestress field in real time, the system guides the main beam energy of the ultrasonic guided wave to converge on the target detection area at the mold-to-workpiece interface, and dynamically scans by following the metal flow front.

[0008] For the identification and control of interface states, the system applies broadband impedance spectrum inversion logic. The acoustic-mechanical coupling control unit processes the echo signals acquired by the ultrasonic guided wave transceiver array and calculates the complex acoustic impedance spectrum of the contact interface based on one-dimensional acoustic transmission line theory. The real acoustic impedance component characterizes the microscopic contact area and density of the interface, while the rate of change of the imaginary acoustic impedance component characterizes the degree of work hardening of the material. Based on the characteristic combination of these physical quantities, the system determines the mechanical state of the contact interface as either a normal sliding state, an interface bonding state, or a rheologically locked state.

[0009] The system performs graded adaptive control based on the determined contact mechanical state: when the interface is determined to be in a bonded state, the system instructs the magnetorheological damper to enter a high-damping rigid support state, and controls the piezoelectric stack actuator to superimpose high-frequency axial mechanical vibration while maintaining static preload displacement. The frequency of this high-frequency vibration is set to the frequency corresponding to the minimum value of the real acoustic impedance component, aiming to reduce the interface friction coefficient and assist in local debonding. When the rheological lock-in state is determined, the system reduces the excitation current of the magnetorheological damper to reduce the support stiffness, and drives the piezoelectric stack actuator to perform a slight retraction, relieving the hydrostatic pressure in the lock-in area through local compliant yielding and improving metal fluidity.

[0010] The second aspect of the present invention provides an adaptive optimization method for the forming process parameters of precision forging of aluminum alloys, which is executed by the aforementioned acoustic-mechanical coupling control unit.

[0011] In the initial forming stage, this method determines the target detection area based on the geometric features of the lower mold assembly and simulation data, calculates the corresponding stress field distribution parameters, drives the dynamic actuation unit to construct a static prestress field, and thus guides the ultrasonic guided wave to the target area.

[0012] During the forming process, the method maintains the existence of a static prestress field, continuously collects reflected acoustic wave signals, and calculates complex acoustic impedance spectrum data containing real acoustic impedance components and imaginary acoustic impedance components, thereby judging the contact mechanical state in real time.

[0013] Based on the state determination results, the method implements closed-loop adjustment of the dynamic actuation unit: if it is in the interface bonding state, high-frequency vibration is applied under rigid support; if it is in the rheological lock-in state, the stiffness is reduced and a small amount of shrinkage compensation is implemented.

[0014] Furthermore, during the downward movement of the slider component, the method updates the coordinates of the target detection area based on the real-time detected changes in acoustic impedance and adjusts the static prestress field distribution accordingly to ensure that the ultrasonic waveguide focal point always follows the metal flow front.

[0015] This invention provides an adaptive optimization system and method for precision forging process parameters of aluminum alloys. It offers the following advantages: 1. This invention constructs a non-uniform static prestress field inside the lower mold assembly by controlling a multi-point array dynamic actuation unit, and uses the acoustic elastic effect to regulate the sound velocity gradient distribution inside the mold material, driving the ultrasonic guided wave beam to undergo controlled deflection and focus on the metal flow front. This solves the problem that fixed sensors in complex cavity structures cannot detect key deformation areas, and realizes real-time dynamic scanning and blind-spot-free perception of the forming state inside the mold without the need for mechanically moving the probe.

[0016] 2. This invention decouples complex acoustic impedance data into real acoustic impedance components characterizing the microscopic contact state and imaginary acoustic impedance components characterizing the degree of work hardening by performing broadband impedance spectrum inversion analysis on the reflected signals of the contact interface. Based on the differences in physical characteristics, it accurately determines three mechanical states: normal sliding, interfacial adhesion, and rheological lock-in. This overcomes the defect that a single signal feature cannot analyze complex rheological behavior and achieves accurate diagnosis of interfacial friction conditions and material flow state during the precision forging of aluminum alloys.

[0017] 3. This invention employs a composite actuation structure consisting of a magnetorheological damper and a piezoelectric stacked actuator connected in series. It implements differentiated adaptive control strategies for different process states. During interface bonding, high-stiffness support is applied in conjunction with high-frequency micro-motion to reduce the friction coefficient. During rheological lock-up, the support is adjusted to low stiffness and combined with a small amount of retraction to provide compliant clearance space. This changes the boundary conditions of the traditional passive mold fixation and realizes active optimization and control of local forming resistance and metal flow behavior. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall structure of the present invention. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see the appendix Figure 1 The present invention provides an adaptive optimization system for aluminum alloy precision forging forming process parameters, including: a precision forging press body 10, an upper mold assembly 20, a lower mold assembly 30, a multi-point array type active compliant mold base 40, an ultrasonic guided wave transceiver array 50, and an acoustic-mechanical coupling control unit 60.

[0021] Before this invention is put into formal operation, key control parameters need to be calibrated through the following steps to ensure the accuracy of state determination: 1. Ideal flow hardening rate ( Calibration: A three-dimensional finite element model (FEM) of the lower mold assembly and standard aluminum alloy billet is established, and the boundary conditions are set to "perfectly ideal lubrication" (friction coefficient μ=0). The forming process is simulated, and the curve of the nodal acoustic impedance at the mold-billet contact interface as a function of the reduction is extracted. The slope of the tangent to this curve is taken as the ideal flow hardening rate. The baseline curve.

[0022] 2. Adhesion threshold ( Calibration: Critical failure tests were conducted using standard specimens. Forging was performed under dry friction conditions without lubricant. When the acoustic monitoring signal showed a sudden change in the interfacial contact stiffness (the change rate exceeded 15%) or primary roughening defects appeared on the specimen surface, the interfacial contact density coefficient at this point was recorded and set as the adhesion threshold. .

[0023] 3. Flatness threshold ( The settings are as follows: The background noise impedance spectrum (approximately white noise, flatness close to 1) under no-load conditions and the resonant impedance spectrum under well-coupled conditions (flatness typically less than 0.6) are collected. The weighted average of the two (e.g., using 0.8 as the dividing line) is set as the flatness threshold. It is used to distinguish whether there is random stick-slip vibration.

[0024] The main body 10 of the precision forging press forms the basic support frame of the adaptive control system for the rheological boundary of aluminum alloy forging based on acoustoelastic stress field focusing detection. The main body 10 includes a hydraulic drive mechanism or servo drive mechanism for driving the slide component up and down, and a base worktable for supporting the die component. The upper die assembly 20 is rigidly connected to the lower end of the slide component of the precision forging press 10 by bolts, and the upper die assembly 20 reciprocates linearly with the slide component. The lower die assembly 30 is correspondingly located directly below the upper die assembly 20. The cavity space formed when the upper die assembly 20 and the lower die assembly 30 are closed is used to accommodate and form the aluminum alloy billet.

[0025] A multi-point array type active compliant die holder 40 is installed between the base worktable of the precision forging press body 10 and the lower die assembly 30, serving as a support base for the lower die assembly 30. The multi-point array type active compliant die holder 40 integrates several independently controlled dynamic actuation units 41. These dynamic actuation units 41 are arranged in a matrix or concentric circle array on the horizontal projection section of the multi-point array type active compliant die holder 40, and their distribution covers all stress-bearing areas at the bottom of the lower die assembly 30.

[0026] Each dynamic actuation unit 41 is mechanically composed of a magnetorheological damper 411 and a piezoelectric stacked actuator 412 connected in series along the axial direction. The fixed end of the magnetorheological damper 411 is mechanically fixedly connected to the base plate of the multi-point array active compliant mold base 40, the piston rod output end of the magnetorheological damper 411 is mechanically fixedly connected to the bottom surface of the piezoelectric stacked actuator 412, and the top output end of the piezoelectric stacked actuator 412 is mechanically fixedly connected to the non-working bottom surface of the lower mold assembly 30, forming a rigid force transmission chain capable of transmitting tensile and compressive loads. The magnetorheological damper 411 is used to adjust the axial support stiffness and damping coefficient of the magnetorheological damper 411 by changing the excitation current, and the piezoelectric stacked actuator 412 is used to generate static preload displacement to construct a prestress field or generate mechanical vibration.

[0027] An ultrasonic guided wave transceiver array 50 is disposed on the outer wall surface or the bottom non-working surface of the lower mold assembly 30. The ultrasonic guided wave transceiver array 50 includes several transmitting transducers 51 and receiving transducers 52. The transmitting transducers 51 and receiving transducers 52 are connected to the solid material of the lower mold assembly 30 via an acoustic coupling medium or a hard brazing layer. The transmitting transducers 51 are used to excite ultrasonic guided waves into the internal structure of the lower mold assembly 30. The receiving transducers 52 are used to collect the acoustic wave signals reflected back from the interface between the lower mold assembly 30 and the aluminum alloy blank.

[0028] The acoustic-mechanical coupling control unit 60 is electrically connected via shielded cables to the main body 10 of the precision forging press, the various dynamic actuation units 41 within the multi-point array active compliant mold base 40, and the ultrasonic guided wave transceiver array 50. The acoustic-mechanical coupling control unit 60 includes a multi-channel signal generator, a data acquisition card, a power amplifier, and a digital signal processor. The acoustic-mechanical coupling control unit 60 is used to output excitation signals to the transmitting transducer 51, receive sensing signals from the receiving transducer 52, and output drive voltage or drive current commands to the magnetorheological damper 411 and the piezoelectric stack actuator 412 to adjust the boundary support state of the lower mold assembly 30.

[0029] An adaptive optimization method for aluminum alloy precision forging process parameters is implemented by the acoustic-mechanical coupling control unit 60. The adaptive optimization method for aluminum alloy precision forging process parameters includes: Step S100: In the initial forming stage where the aluminum alloy billet is placed into the lower mold assembly 30 and the upper mold assembly 20 descends to contact the aluminum alloy billet, the acoustic-mechanical coupling control unit 60 determines the target detection area on the cavity surface of the lower mold assembly 30 based on the inner contour geometry data of the lower mold assembly 30 and the preset finite element simulation data of the metal flow front position. The acoustic-mechanical coupling control unit 60 calculates the stress field distribution parameters required to focus the ultrasonic guided wave onto the target detection area based on the acoustoelastic effect. The acoustic-mechanical coupling control unit 60 sends a first control command to each dynamic actuation unit 41 within the multi-point array active compliant mold base 40. The first control command drives the piezoelectric stacked actuators 412 in each dynamic actuation unit 41 to generate independent static elongation displacements. The static elongation displacements generated by each piezoelectric stacked actuator 412 construct a non-uniform static prestress field within the solid material of the lower mold assembly 30. The static prestress field changes the sound wave propagation velocity distribution within the lower mold assembly 30, causing the propagation path of the ultrasonic guided wave to be refracted and deflected towards the target detection area.

[0030] Step S200: While maintaining the static prestress field, the acoustic-mechanical coupling control unit 60 sends a broadband excitation pulse signal to the transmitting transducer 51 in the ultrasonic guided wave transceiver array 50. The transmitting transducer 51 generates ultrasonic guided waves and injects them into the lower mold assembly 30. The ultrasonic guided waves are guided and focused to the target detection area by the static prestress field. The acoustic-mechanical coupling control unit 60 controls the receiving transducer 52 to collect the echo signal reflected from the contact interface between the lower mold assembly 30 and the aluminum alloy billet. The acoustic-mechanical coupling control unit 60 performs fast Fourier transform processing on the echo signal and calculates the complex acoustic impedance spectrum data of the contact interface. The complex acoustic impedance spectrum data includes real acoustic impedance component data and imaginary acoustic impedance component data.

[0031] Step S300: The acoustic-mechanical coupling control unit 60 performs feature analysis on the complex acoustic impedance spectrum data. The acoustic-mechanical coupling control unit 60 monitors the amplitude changes of the acoustic impedance component data in a specific frequency band and the fluctuation characteristics of the acoustic impedance component data. Based on the analysis results, the acoustic-mechanical coupling control unit 60 determines the current contact mechanical state of the target detection area. The contact mechanical state includes normal sliding state, interface bonding state, and rheological locking state.

[0032] Step S400: Based on the contact mechanical state determined in step S300, the acoustic-mechanical coupling control unit 60 sends a second control command to each dynamic actuation unit 41 within the multi-point array active compliant mold base 40. When the interface bonding state is determined, the second control command drives the piezoelectric stacked actuator 412 in the corresponding area to superimpose high-frequency axial mechanical vibration while maintaining static elongation displacement. This utilizes the ultrasonic friction reduction effect to reduce the friction coefficient of the contact interface and restore the surface flow of the aluminum alloy billet. When the rheological lock-up state is determined, the second control command adjusts the excitation current of the magnetorheological damper 411 in the corresponding area to reduce the axial support stiffness and damping coefficient of the magnetorheological damper 411, and drives the piezoelectric stacked actuator 412 to generate a slight retraction displacement, thereby implementing local compliant yielding to the lower mold assembly 30.

[0033] The specific value of the "micro-retraction displacement" mentioned here is set according to the dimensional tolerance zone of the forging to be processed. Preferably, the range of the micro-retraction displacement is 10μm to 50μm. The basis for setting this displacement is that its value is sufficient to break the local high-pressure bonding nodes formed at the interface (breaking the rheological lock-in), while it is much smaller than the dimensional tolerances usually required for aluminum alloy precision forgings (e.g., ±200μm). Therefore, it can improve the lubrication condition while ensuring that the dimensional accuracy of the final formed part meets the design requirements.

[0034] Step S500: As the slider component of the precision forging press body 10 continues to descend, the acoustic-mechanical coupling control unit 60 updates the position coordinates of the target detection area based on the real-time acquired acoustic impedance change position. The acoustic-mechanical coupling control unit 60 recalculates the required stress field distribution parameters and adjusts the static elongation displacement of each piezoelectric stack actuator 412 so that the focal point of the ultrasonic guide wave moves with the flow front of the aluminum alloy billet. The acoustic-mechanical coupling control unit 60 executes steps S200 to S400 in a loop until the forging stroke ends.

[0035] The specific structure of the multi-point array type active compliant mold base 40 includes: a rigid base plate 401, an outer guide frame 402, several sets of dynamic actuation units 41, a position detection sensor 42, and a load detection sensor 43.

[0036] The multi-point array type active compliant die holder 40 is fixed to the base worktable of the precision forging machine body 10 via a rigid base plate 401 using T-slot bolts. The outer guide frame 402 is vertically welded to the edge of the rigid base plate 401, or the outer guide frame 402 and the rigid base plate 401 are integrally cast. The inner sidewall of the outer guide frame 402 is machined with a precision guide surface, which is used to position and constrain the lower die assembly 30 in the horizontal direction, limiting the lateral displacement of the lower die assembly 30 during the forging process.

[0037] Several sets of dynamic actuation units 41 are distributed on the upper surface of the rigid base plate 401 according to a preset coordinate matrix. The spatial position coordinates of each dynamic actuation unit 41 correspond to a specific stress area at the bottom of the lower mold assembly 30. The number and distribution density of the dynamic actuation units 41 are determined according to the size of the lower mold assembly 30 and the forming load distribution characteristics. Each dynamic actuation unit 41 includes a vertically arranged magnetorheological damper 411 and a piezoelectric stacked actuator 412.

[0038] The bottom of the cylinder of the magnetorheological damper 411 is mechanically fixed to the rigid base plate 401 by a flange. The internal cavity of the magnetorheological damper 411 is filled with magnetorheological fluid. The piston assembly of the magnetorheological damper 411 integrates an excitation coil. The excitation coil controls the shear yield stress of the magnetorheological fluid by changing the magnetic field strength generated by the input current, thereby adjusting the damping force and equivalent support stiffness of the magnetorheological damper 411 along the axial direction.

[0039] The piezoelectric stack actuator 412 is coaxially mounted above the magnetorheological damper 411. A connecting block 413 is located at the top of the piston rod of the magnetorheological damper 411. The bottom surface of the piezoelectric stack actuator 412 is fixed to the upper surface of the connecting block 413 by a high-strength insulating layer or mechanical clamps. The piezoelectric stack actuator 412 adopts an encapsulated structure. A preload spring is installed inside the encapsulated shell of the piezoelectric stack actuator 412. The preload spring applies axial preload to the piezoelectric ceramic sheet to prevent breakage under tensile load.

[0040] The piezoelectric stack actuator 412 has a ball joint connection assembly 414 at its top output end. The lower end of the ball joint connection assembly 414 is mechanically fixedly connected to the piezoelectric stack actuator 412, and the upper end of the ball joint connection assembly 414 is mechanically fixedly connected to the bottom surface of the lower mold assembly 30 via threads or a flange. The ball joint connection assembly 414 is used to accommodate the small angular deviation between the bottom surface of the lower mold assembly 30 and the dynamic actuation unit 41 while transmitting axial tensile and compressive loads. During the system initialization phase, the piezoelectric stack actuator 412 outputs a static elongation and applies a lifting force to the corresponding area of ​​the lower mold assembly 30 through the ball joint connection assembly 414, generating a prestress distribution for sound field focusing within the solid material of the lower mold assembly 30.

[0041] The position detection sensor 42 is installed between the rigid base plate 401 and the transition connection block 413, or between the rigid base plate 401 and the ball joint connection assembly 414. The position detection sensor 42 is used to measure the total axial displacement of the dynamic actuation unit 41 in real time. The load detection sensor 43 is connected in series between the piezoelectric stacked actuator 412 and the transition connection block 413, or at the bottom of the magnetorheological damper 411. The load detection sensor 43 is used to monitor the vertical forging load borne by the dynamic actuation unit 41 in real time.

[0042] The multi-point array type active compliant mold base 40 also includes a cable busbar 403 disposed on a rigid base plate 401. The control cable of the magnetorheological damper 411, the drive cable of the piezoelectric stacked actuator 412, and the signal cables of the position detection sensor 42 and the load detection sensor 43 are all merged into the cable busbar 403 and uniformly connected to the acoustic-mechanical coupling control unit 60. A circulating cooling channel 404 is provided inside the rigid base plate 401. The circulating cooling channel 404 is used to remove the heat generated by the magnetorheological damper 411 during operation and the heat generated by the high-frequency vibration of the piezoelectric stacked actuator 412 through the circulating cooling medium, so as to maintain the stable operating temperature of the dynamic actuation unit 41.

[0043] The specific layout and connection method of the ultrasonic guided wave transceiver array 50 include: the ultrasonic guided wave transceiver array 50 is disposed on the outer sidewall surface of the lower mold assembly 30. The ultrasonic guided wave transceiver array 50 avoids the stress area at the bottom of the lower mold assembly 30 and is distributed on the non-working side of the lower mold assembly 30. The ultrasonic guided wave transceiver array 50 consists of a number of transmitting transceivers 51 and a number of receiving transceivers 52.

[0044] Several transmitting transducers 51 and several receiving transducers 52 are arranged in a ring array along the outer periphery of the lower mold assembly 30, or in a multi-layer matrix arrangement along the height of the sidewall of the lower mold assembly 30. The transmitting transducers 51 and receiving transducers 52 are arranged alternately in spatial position. A high-temperature resistant acoustic wedge block 53 is provided between each transmitting transducer 51 and each receiving transducer 52 and the outer sidewall surface of the lower mold assembly 30. The wedge angle of the high-temperature resistant acoustic wedge block 53 is determined according to the geometric dimensions of the lower mold assembly 30 and the depth of the target detection area. The high-temperature resistant acoustic wedge block 53 is used to adjust the incident angle of the ultrasonic guided wave so that the main beam path of the ultrasonic guided wave can cover the cavity surface at the top of the lower mold assembly 30.

[0045] Each transmitting transducer 51 and each receiving transducer 52 includes a piezoelectric ceramic wafer, an acoustic matching layer, a backing block, and a metal shielding shell. The piezoelectric ceramic wafer is made of modified lead zirconate titanate material or lithium niobate single crystal material with a Curie temperature higher than 400 degrees Celsius. The transmitting transducer 51, the receiving transducer 52, and the high-temperature resistant acoustic wedge block 53 are bonded and fixed together, as are the high-temperature resistant acoustic wedge block 53 and the metal sidewall surface of the lower mold assembly 30, using high-temperature resistant inorganic adhesives or mechanical clamping. The mechanical clamping method is used in conjunction with gold foil coupling gaskets to ensure the coupling stability of the acoustic energy transmission interface under high-temperature forging conditions.

[0046] The transmitting transducer 51 is a wideband transducer. The bandwidth of the transmitting transducer 51 covers the frequency range required for impedance spectrum analysis by the acoustic-mechanical coupling control unit 60. The transmitting transducer 51 is used to convert the electrical excitation pulse output by the acoustic-mechanical coupling control unit 60 into ultrasonic longitudinal waves or ultrasonic transverse waves, and inject them into the solid material of the lower mold assembly 30 through the high-temperature resistant acoustic wedge block 53. The receiving transducer 52 has frequency response characteristics that match those of the transmitting transducer 51. The receiving transducer 52 is used to pick up the elastic wave signal after propagation inside the lower mold assembly 30 and reflection at the interface, and convert the elastic wave signal into a voltage signal.

[0047] The construction and parameter definition process of the acoustic-elastic coupling model includes: the acoustic-mechanical coupling control unit 60 establishes a physical field coupling model of the lower mold assembly 30; the acoustic-mechanical coupling control unit 60 defines several dynamic actuation units 41 in the multi-point array active compliant mold base 40 as the boundary input source of the model; and sets the total number of dynamic actuation units 41 to be... The acoustic-mechanical coupling control unit 60 uses the principle of linear superposition to describe any spatial position inside the lower mold assembly 30. Total stress tensor distribution at point .

[0048] Total stress tensor distribution The passive deformation stress field generated during the forging process is superimposed with the active prestress field applied by all dynamic actuation units 41. The acoustic-mechanical coupling control unit 60 calculates the total stress tensor distribution using the following formula. :

[0049] in, =(x,y,z) represents the Cartesian three-dimensional coordinate vector of the lower mold assembly 30 relative to its geometric center. This represents the passive load stress field caused by the deformation resistance of the aluminum alloy billet. The passive load stress field is estimated by the acoustic-mechanical coupling control unit 60 based on the finite element simulation database or feedback data from the real-time load detection sensor 43. Indicates the first The scalar of the vertical axial force applied by the dynamic actuation unit 41 Indicates the first The influence function matrix of the dynamic actuation unit 41 on the internal stress distribution of the lower mold assembly 30, influence function matrix During the system initialization phase, the calibration is performed using finite element simulation and stored in the memory of the acoustic-mechanical coupling control unit 60.

[0050] The acoustic-mechanical coupling control unit 60 constructs a mapping relationship between sound velocity and stress field based on the acoustoelastic theory. The propagation velocity of the ultrasonic guided wave in the medium of the lower mold assembly 30 changes with the local stress state. The acoustic-mechanical coupling control unit 60 uses the acoustoelastic equation to calculate the local longitudinal wave velocity after being affected by stress. :

[0051] in, This indicates the nominal longitudinal wave velocity of the material in the lower mold component 30 under stress-free conditions. and These are the first-order and second-order acoustoelastic constants of the material of the lower mold component 30, respectively. Represents the trace of the stress tensor. Characterizing the isotropic effect of volumetric stress components on sound velocity. The unit vector representing the direction of ultrasonic guided wave propagation. The term represents the effect of the normal stress component along the wave propagation direction on the anisotropy of the sound velocity.

[0052] The acoustic-mechanical coupling control unit 60 utilizes the calculated local longitudinal wave velocity The spatial distribution of the ultrasonic guided wave is calculated based on the equation of process or Fermat's principle, and the local longitudinal wave velocity is determined. The uneven distribution of sound velocity within the lower mold assembly 30 forms a sound velocity distribution field with gradient refractive index characteristics. The acoustic-mechanical coupling control unit 60 calculates the required output force for each dynamic actuation unit 41 by solving the inverse problem of the sound ray trajectory between the target detection area and the emission source. The combination causes the sound wave beam to be continuously refracted under the action of the sound speed gradient field and finally converge on the target detection area.

[0053] The acoustic-mechanical coupling control unit 60 defines the optimization objective function. Optimize the objective function The acoustic-mechanical coupling control unit 60 calculates the Euclidean distance deviation between the actual landing point of the ultrasonic guided wave main beam and the center of the target detection area, and then converts the output force vectors of each dynamic actuation unit 41 in the multi-point array active compliant mode base 40 into the Euclidean distance. Let these be the variables to be optimized, and let the objective function be optimized. The mathematical expression is as follows:

[0054] in, This represents the three-dimensional coordinate vector of the center of the target detection area relative to the geometric center of the lower mold assembly 30. This indicates the current output force vector. Under the influence of the generated sound velocity gradient field, the coordinate vector of the intersection point between the main ultrasonic wave beam originating from the activated transmitting transducer 51 and the surface of the inner cavity of the lower mold assembly 30 is... The regularization parameter is used to constrain the output force vector. The modulus is set to prevent the solution result from exceeding the upper limit of the saturation output force of the dynamic actuation unit 41.

[0055] The acoustic-mechanical coupling control unit 60 sets a set of constraints. This set includes the maximum and minimum output force limits for each dynamic actuation unit 41, as well as the allowable yield strength threshold of the material in the lower mold assembly 30. The acoustic-mechanical coupling control unit 60 limits the output force vector through this set of constraints. The feasible region ensures that the calculated output force vector It will not cause permanent plastic deformation of the lower mold assembly 30 and ensures the output force vector. It is located within the linear operating range of the magnetorheological damper 411 and the piezoelectric stack actuator 412.

[0056] The acoustic-mechanical coupling control unit 60 uses the gradient descent method or the Gauss-Newton iteration method to solve the optimization objective function. The minimum value is calculated by the acoustic-mechanical coupling control unit 60 based on the current iterative output force vector during each iteration. Update the stress field distribution data and corresponding sound velocity field distribution data inside the mold component 30. The acoustic-mechanical coupling control unit 60 uses the Runge-Kutta numerical integration method to solve the equation of motion and obtain the propagation path and actual arrival point of the ultrasonic guided wave in the non-uniform medium. .

[0057] The acoustic-mechanical coupling control unit 60 calculates and optimizes the objective function. Regarding the output force vector The Jacobian matrix represents the sensitivity of the output force change of each dynamic actuation unit 41 to the coordinates of the sound wave landing point. The acoustic-mechanical coupling control unit 60 updates the output force vector based on the Jacobian matrix and the current position deviation. The acoustic-mechanical coupling control unit 60 cyclically performs forward ray tracing calculations and reverse gradient update calculations until the position deviation is less than the preset focusing accuracy threshold.

[0058] The acoustic-mechanical coupling control unit 60 will eventually converge the output force vector. Parsing the commands into specific hardware drive instructions, the acoustic-mechanical coupling control unit 60 sends high excitation current commands to the magnetorheological dampers 411 in each dynamic actuation unit 41, causing the magnetorheological dampers 411 to enter a high-stiffness locked state to provide reverse support force. Simultaneously, the acoustic-mechanical coupling control unit 60, based on the piezoelectric constant and stiffness characteristics of the piezoelectric stack actuator 412, adjusts the output force vector... The static DC voltage command is converted into a corresponding static DC voltage command. The acoustic-mechanical coupling control unit 60 sends the static DC voltage command to each piezoelectric stack actuator 412 through a multi-channel digital-to-analog converter. Each piezoelectric stack actuator 412 generates axial elongation displacement under the rigid support of the magnetorheological damper 411, and constructs the required sound velocity gradient field inside the lower mold assembly 30 to achieve precise focusing of the ultrasonic guided wave on the target detection area.

[0059] The broadband impedance spectrum inversion algorithm is executed by the signal preprocessing module 601, the spectrum analysis module 602, and the impedance inversion module 603 integrated within the acoustic-mechanical coupling control unit 60. The broadband impedance spectrum inversion algorithm includes: The signal preprocessing module 601 receives the time-domain echo signal sequence from the receiving transducer 52. The signal preprocessing module 601 performs a windowing truncation operation on the time-domain echo signal sequence. The signal preprocessing module 601 uses a Hanning window function or a rectangular window function to extract an effective time-domain segment containing the reflected wave from the contact interface between the lower mold assembly 30 and the aluminum alloy billet from the time-domain echo signal sequence. The signal preprocessing module 601 aligns the effective time-domain segment with a pre-stored reference echo signal. The reference echo signal is a calibration signal pre-collected and stored under the free boundary condition of the lower mold assembly 30 in contact with air, i.e., when the lower mold assembly 30 is not holding the aluminum alloy billet.

[0060] The spectrum analysis module 602 performs Fast Fourier Transform on the effective time-domain segment and the reference echo signal after time-domain alignment processing. The spectrum analysis module 602 converts the time-domain signal into a complex frequency-domain sequence, obtaining the measured signal spectrum. and reference signal spectrum , This represents the angular frequency variable. The spectrum analysis module 602 calculates the complex reflection coefficient spectrum of the contact interface. Complex reflection coefficient spectrum The calculation formula is as follows:

[0061] in, The reference reflection coefficient constant is the known constant of the reference interface. Since the reference echo signal is taken from the steel-air interface, here... The value is -1; The phase compensation factor is used to compensate for the linear phase deviation introduced by the difference in the signal truncation window position. The spectrum analysis module 602 performs Wiener filtering on the calculated spectrum ratio result. Wiener filtering is used to suppress the linear phase deviation introduced by the signal truncation window position difference. High-frequency numerical instability noise caused by spectral zeros.

[0062] Impedance inversion module 603 constructs a contact interface acoustic impedance calculation model based on one-dimensional acoustic transmission line theory. Impedance inversion module 603 utilizes the acoustic impedance characteristics of the material in the lower mold assembly 30. and the calculated complex reflection coefficient spectrum Inversion calculation of the complex acoustic impedance spectrum of the contact interface Complex acoustic impedance spectrum The calculation formula is as follows:

[0063] in, It is a constant. It is equal to the product of the material density of the lower mold component 30 and the longitudinal wave velocity of the material. It is a complex value that varies with frequency.

[0064] The rheological state feature decoupling logic is executed by the feature extraction module 604 and the state determination module 605 integrated within the acoustic-mechanical coupling control unit 60. The rheological state feature decoupling logic includes: Feature extraction module 604 receives the real part acoustic impedance component data output by impedance inversion module 603. and imaginary part acoustic impedance component data The feature extraction module 604 synchronously acquires the real-time slider displacement data fed back by the main body 10 of the precision forging press. The feature extraction module 604 combines the real acoustic impedance component data, the imaginary acoustic impedance component data, and the slider displacement data to calculate two key feature indicators characterizing the interface contact state: the interface contact density coefficient. Spectral flatness index .

[0065] Feature extraction module 604 extracts features through the effective bandwidth. The interface contact density coefficient is calculated by normalizing the integral of the real acoustic impedance component. , interface contact density coefficient The interface contact density coefficient is used to quantify the macroscopic fit between the lower mold component 30 and the aluminum alloy blank. The calculation formula is as follows:

[0066] in, To effectively analyze bandwidth, and These represent the upper and lower frequency limits of the effective analysis bandwidth, and the interface contact density coefficient, respectively. The value range is from 0 to 1, where 1 indicates that the interface is in complete rigid contact and 0 indicates that the interface is in a completely separated state.

[0067] Feature extraction module 604 calculates the spectral flatness index Spectral flatness index The spectral flatness index is used to characterize the degree of fluctuation in the acoustic response of a contact interface as a function of frequency. The feature extraction module 604 characterizes the uniformity of the dielectric layer at the contact interface by calculating the ratio of the geometric mean to the arithmetic mean of the real acoustic impedance component data to obtain the spectral flatness index. Spectral flatness index The calculation formula is as follows:

[0068] in, This represents the total number of frequency domain sampling points. For the first Discrete frequency points. If microscopic lubrication pits or uneven oxide scale exist at the interface, sound waves will scatter and resonate, leading to an increase in the spectral flatness index. The value decreases; if metal atomic-level bonding occurs at the interface, the acoustic wave transmission characteristics tend to be uniform across a wide frequency band, and the spectral flatness index decreases. The value approaches 1.

[0069] The state determination module 605 receives the contact density coefficient at the interface. Spectral flatness index The system performs logical judgments based on a preset set of state thresholds, and the state determination module 605 outputs the current rheological contact state flag of the target detection area.

[0070] If the interface contact density coefficient Greater than the preset adhesion threshold And the spectral flatness index Greater than the preset flatness threshold The state determination module 605 determines that the target detection area is in an interface bonding state. In the interface bonding state, the surface of the aluminum alloy billet and the surface of the lower mold assembly 30 undergo atomic diffusion connection, the friction coefficient increases, and the metal flow is hindered.

[0071] If the interface contact density coefficient Within the preset sliding range The interface contact density coefficient calculated by the feature extraction module 604 is within the range. For slider displacement first derivative Less than the preset stability threshold The state determination module 605 determines that the target detection area is in a normal sliding state. In the normal sliding state, a stable lubricating film exists at the interface, and the metal flows smoothly.

[0072] If the interface contact density coefficient Continuously greater than the upper limit of the sliding range Meanwhile, the feature extraction module 604 calculates the average value of the imaginary acoustic impedance component data. For slider displacement rate of change Greater than the preset hardening rate threshold The state determination module 605 determines that the target detection area is in a rheological locked state. In the rheological locked state, the aluminum alloy billet undergoes volumetric incompressible accumulation in a local area, and the local hydrostatic pressure increases sharply, resulting in passive hardening of the interface stiffness.

[0073] The high-frequency micro-motion debonding control mode is executed by the vibration generation module 606 integrated inside the acoustic-mechanical coupling control unit 60. The high-frequency micro-motion debonding control mode includes: If the state determination module 605 determines that the target detection area is in an interface bonding state, the acoustic-mechanical coupling control unit 60 activates the vibration generation module 606. The rigid reaction force support point prevents the high-frequency vibration energy from dissipating towards the rigid base plate 401, ensuring that the vibration energy is transmitted unidirectionally to the lower mold assembly 30.

[0074] After the magnetorheological damper 411 is locked, the vibration generation module 606 sends a high-frequency drive signal to the piezoelectric stack actuator 412. The vibration generation module 606 obtains the complex acoustic impedance spectrum data at the current moment from the spectrum analysis module 602, and extracts the frequency corresponding to the minimum value of the real part acoustic impedance component in the complex acoustic impedance spectrum as the resonance driving frequency. High-frequency drive signal The mathematical expression for the superposition of DC bias voltage and AC sinusoidal voltage is as follows:

[0075] in, This is a DC bias voltage, used to prevent depolarization damage to the piezoelectric stack actuator 412 during the negative half-cycle. For AC drive amplitude, This is the resonant driving frequency.

[0076] The piezoelectric stack actuator 412 converts the electrical signal into high-frequency mechanical vibration in the vertical direction. The high-frequency mechanical vibration is transmitted to the bottom of the lower mold assembly 30 through the ball joint connection assembly 414, and induces alternating normal stress at the contact interface between the lower mold assembly 30 and the aluminum alloy billet. The alternating normal stress periodically reduces the effective normal pressure at the contact interface, thereby periodically reducing the sliding friction resistance at the contact interface. At the same time, the high-frequency mechanical vibration excites dislocation motion inside the aluminum alloy billet, generating a sonoplastic effect. The sonoplastic effect reduces the yield strength and deformation resistance of the aluminum alloy material.

[0077] While outputting a high-frequency drive signal, the vibration generation module 606 continuously monitors the interface contact density coefficient output by the feature extraction module 604. If the interface contact density coefficient Descend to the preset sliding range Inside, the vibration generation module 606 determines that the interface unbinding operation is complete, the vibration generation module 606 stops outputting high-frequency drive signals, and the vibration generation module 606 sends a damping release command to the magnetorheological damper 411 to reduce the excitation current to zero or maintain it at a low damping level, so that the dynamic actuation unit 41 returns to the passive compliant support state.

[0078] The magnetorheological compliance compensation control mode is executed by the compliance adjustment module 607 integrated within the acoustic-mechanical coupling control unit 60. The magnetorheological compliance compensation control mode includes: If the state determination module 605 determines that the target detection area is in a rheological locked state, the acoustic-mechanical coupling control unit 60 activates the compliance adjustment module 607.

[0079] Compliance adjustment module 607 reads the average value of the imaginary acoustic impedance component data output by feature extraction module 604. And by combining the slider displacement data, the real-time hardening rate is calculated. Real-time hardening rate Defined as the derivative of the imaginary acoustic impedance component data with respect to the slider displacement. The compliance adjustment module 607 calculates the real-time hardening rate. With the preset ideal flow hardening rate Deviation between .

[0080] The compliance adjustment module 607 uses an incremental proportional-integral control algorithm to calculate the target yield force correction required for the magnetorheological damper 411. Target yield force correction amount The calculation formula is as follows:

[0081] in, This is the proportional gain coefficient. This is the integral gain coefficient. This is due to the hardening rate deviation. Under rheological lockout conditions, Usually much larger ,lead to A negative value indicates that the support force threshold of the magnetorheological damper 411 needs to be reduced to increase the compliance of the system.

[0082] The compliance adjustment module 607 adjusts the pressure support force according to the initial preset value. and the calculated target yield force correction amount Determine the target yield force at the current moment. The compliance adjustment module 607 utilizes the static inverse mechanical model of a magnetorheological damper to adjust the target yield force. Converted to excitation control current Excitation control current The calculation formula is as follows:

[0083] in, and The force-current characteristic parameters of the magnetorheological damper 411 are determined by the material properties of the magnetorheological fluid and the magnetic circuit geometry of the damper, and are pre-determined during the system calibration phase. This model ignores the contribution of viscous damping force at low speeds and mainly sets the action threshold of the damper by controlling the shear yield strength of the magnetorheological fluid.

[0084] The compliance adjustment module 607 controls the excitation current through a current drive circuit. The excitation coil of the magnetorheological damper 411 is applied. When the excitation control current... When the load decreases, the yield threshold of the magnetorheological damper 411 decreases. When the forming load transmitted from the aluminum alloy billet to the lower die assembly 30 exceeds the reduced yield threshold, the magnetorheological damper 411 is compressed, and the lower die assembly 30 undergoes a slight downward displacement. This slight displacement increases the metal deformation space in the locked region, leading to a decrease in local hydrostatic pressure. Consequently, the imaginary acoustic impedance component data falls back. The compliance adjustment module 607 continues to execute the above closed-loop adjustment process until the real-time hardening rate... Restore to ideal flow hardening rate Within the allowable error range, it signifies that the rheological blocking state has been released.

[0085] The focus follow-up mechanism is executed by the dynamic tracking module 608 integrated within the acoustic-mechanical coupling control unit 60. The focus follow-up mechanism includes: The dynamic tracking module 608 periodically collects the stress distribution data of the lower die assembly 30 fed back by the real-time load detection sensor 43 at preset time intervals. During the precision forging of aluminum alloy, due to the plastic flow and hardening of the aluminum alloy billet, the stress field distribution inside the lower die assembly 30 is in a dynamic state of change. The dynamic tracking module 608 calls the pre-stored acoustoelastic constants of the lower die assembly 30 material. Based on the acoustoelastic effect theory, the dynamic tracking module 608 updates the current time using the real-time collected stress distribution data and acoustoelastic constants. Longitudinal wave sound velocity distribution field inside the lower mold assembly 30 .

[0086] The dynamic tracking module 608 is based on the updated longitudinal wave velocity distribution field. and the output force vector currently applied to each dynamic actuation unit 41 The dynamic tracking module 608 performs fast ray tracing simulation calculations. It calculates the actual coordinates of the landing point between the main ultrasonic waveguide beam and the inner surface of the lower mold assembly 30 under the current non-uniform medium sound velocity environment. .

[0087] The dynamic tracking module 608 calculates the actual landing point coordinates. With respect to the preset target detection area center coordinates Focus drift vector between Focus drift vector The calculation formula is as follows:

[0088] The dynamic tracking module 608 calculates the output force increment vector used to compensate for focus drift. To ensure the real-time response speed of the control system, the dynamic tracking module 608 reuses the Jacobian matrix obtained in the initial acoustic lens inverse problem solving stage. Alternatively, a local linearization approximation can be performed using the Jacobian matrix from the previous control cycle. Output force increment vector The calculation formula is as follows:

[0089] in, The sensitivity Jacobian matrix, It represents the linear mapping relationship between small changes in the output force vector and changes in the landing point coordinates. The damping factor is used to prevent numerical divergence during matrix inversion. It is an identity matrix.

[0090] The dynamic tracking module 608 calculates the output force increment vector. Superimposed on the current output force vector The updated target output force vector is generated. The dynamic tracking module 608 converts the target output force vector into a corresponding voltage control command and sends it to the multi-point array active compliant mold base 40.

[0091] The piezoelectric stacked actuator 412 in the multi-point array active compliant mold base 40 responds to voltage control commands to fine-tune the distribution of extrusion stress applied to the bottom of the lower mold assembly 30. The fine-tuning of the extrusion stress distribution changes the refractive index distribution inside the lower mold assembly 30, counteracting the acoustic path deflection caused by the time-varying forging load, and causing the main ultrasonic guided wave beam to refocus on the center of the target detection area.

[0092] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An adaptive optimization system for precision forging process parameters of aluminum alloys, characterized in that, include: The precision forging press body, upper mold assembly, lower mold assembly, multi-point array active compliant mold base, ultrasonic wave guided transceiver array, and acoustic-mechanical coupling control unit; The upper mold assembly is connected to the slider component of the precision forging press body, and the lower mold assembly is correspondingly disposed below the upper mold assembly; The multi-point array type active compliant die holder is installed between the base worktable of the precision forging press body and the lower die assembly, serving as a support base for the lower die assembly. The multi-point array type active compliant die holder integrates several independently controlled dynamic actuation units, and the distribution of the several dynamic actuation units covers the various stress areas at the bottom of the lower die assembly. The ultrasonic guided wave transceiver array is disposed on the outer wall surface or the bottom non-working surface of the lower mold assembly, and is used to excite ultrasonic guided waves inside the structure of the lower mold assembly and collect the reflected sound wave signals. The acoustic-mechanical coupling control unit is electrically connected to the main body of the precision forging press, each dynamic actuation unit in the multi-point array active compliant mold base, and the ultrasonic guided wave transceiver array. The acoustic-mechanical coupling control unit is used to control each of the dynamic actuation units to generate a static prestress field to guide the ultrasonic guided wave to focus on the target detection area, and to adjust the boundary support state of the lower mold assembly according to the collected acoustic wave signal.

2. The adaptive optimization system for aluminum alloy precision forging process parameters according to claim 1, characterized in that, Each of the aforementioned dynamic actuation units is composed of a magnetorheological damper and a piezoelectric stacked actuator connected in series along the axial direction; The fixed end of the magnetorheological damper is fixedly connected to the base plate of the multi-point array active compliant mold base, the piston rod output end of the magnetorheological damper is fixedly connected to the bottom surface of the piezoelectric stack actuator, and the top surface output end of the piezoelectric stack actuator is fixedly connected to the bottom surface of the lower mold assembly through a ball joint connection assembly. The magnetorheological damper is used to adjust the axial support stiffness and damping coefficient by changing the excitation current, and the piezoelectric stacked actuator is used to generate static preload displacement to construct a prestress field or generate mechanical vibration.

3. The adaptive optimization system for aluminum alloy precision forging process parameters according to claim 1, characterized in that, The ultrasonic guided wave transceiver array includes several transmitting transducers and several receiving transducers. The transmitting transducers and the receiving transducers are arranged in a ring array along the outer periphery of the lower mold assembly or in a multi-layer matrix arrangement along the height of the sidewall. A high-temperature resistant acoustic wedge is provided between each of the transmitting transducers and each of the receiving transducers and the side wall surface of the lower mold assembly. The high-temperature resistant acoustic wedge is used to adjust the incident angle of the ultrasonic guided wave so that the main beam path of the ultrasonic guided wave covers the cavity surface at the top of the lower mold assembly.

4. The adaptive optimization system for aluminum alloy precision forging process parameters according to claim 2, characterized in that, The acoustic-mechanical coupling control unit is configured to perform the following acoustic elastic stress field focusing detection operation: The acoustic-mechanical coupling control unit determines the target detection area based on the inner contour geometric data of the lower mold assembly and the position data of the metal flow front in finite element simulation, and calculates the required stress field distribution parameters based on the acoustic elastic effect. The acoustic-mechanical coupling control unit sends control commands to each of the dynamic actuation units, driving the piezoelectric stack actuator to generate independent static elongation displacement, constructing a non-uniform static prestress field inside the lower mold assembly, and using the static prestress field to change the sound wave propagation speed distribution inside the lower mold assembly, causing the propagation path of the ultrasonic guided wave to be refracted and deflected toward the target detection area.

5. The adaptive optimization system for aluminum alloy precision forging process parameters according to claim 4, characterized in that, The acoustic-mechanical coupling control unit integrates a dynamic tracking module, which is used to perform the following focus follow-up update operation: The dynamic tracking module periodically collects the force distribution data of the lower mold assembly, updates the longitudinal wave sound velocity distribution field by combining the acoustoelastic constant of the lower mold assembly material, and calculates the actual landing point coordinates of the ultrasonic guided wave main beam through fast ray tracing simulation calculation. The dynamic tracking module calculates the focus drift vector between the actual landing point coordinates and the center coordinates of the target detection area, calculates the output force increment vector to compensate for the focus drift, and adjusts the voltage control command applied to each of the piezoelectric stack actuators so that the focus point of the ultrasonic guided wave moves with the flow front of the aluminum alloy billet.

6. The adaptive optimization system for aluminum alloy precision forging process parameters according to claim 3, characterized in that, The acoustic-mechanical coupling control unit integrates a signal preprocessing module, a spectrum analysis module, and an impedance inversion module, used to perform the following broadband impedance spectrum inversion calculation: The signal preprocessing module receives the echo signal collected by the receiving transducer and extracts the effective time domain segment; the spectrum analysis module performs a fast Fourier transform on the effective time domain segment and calculates the complex reflection coefficient spectrum of the contact interface. The impedance inversion module is based on the one-dimensional acoustic transmission line theory and uses the complex reflection coefficient spectrum to invert and calculate the complex acoustic impedance spectrum data of the contact interface. The complex acoustic impedance spectrum data includes real part acoustic impedance component data and imaginary part acoustic impedance component data.

7. The adaptive optimization system for aluminum alloy precision forging process parameters according to claim 6, characterized in that, The acoustic-mechanical coupling control unit integrates a feature extraction module and a state determination module, which are used to perform the following rheological state feature decoupling operation: The feature extraction module calculates the interface contact density coefficient and spectral flatness index based on the real part acoustic impedance component data, and calculates the real-time hardening rate of the imaginary part acoustic impedance component data in combination with the slider displacement data. The state determination module determines the contact mechanical state of the target detection area based on the interface contact density coefficient, the spectral flatness index, and the real-time hardening rate. The contact mechanical state includes normal sliding state, interface bonding state, and rheological locking state.

8. The adaptive optimization system for aluminum alloy precision forging process parameters according to claim 7, characterized in that, The acoustic-mechanical coupling control unit integrates a vibration generation module for performing the following high-frequency micro-motion debonding control: When the state determination module determines that the interface is in the bonding state, the acoustic-mechanical coupling control unit sends a high excitation current command to the magnetorheological damper in the corresponding area to make the magnetorheological damper enter a high stiffness locking state, and the vibration generation module drives the piezoelectric stacked actuator in the corresponding area to generate high frequency axial mechanical vibration. The vibration generation module extracts the frequency corresponding to the minimum real part acoustic impedance component in the complex acoustic impedance spectrum data as the resonance driving frequency, and generates a high-frequency driving signal containing DC bias voltage and AC sinusoidal voltage.

9. The adaptive optimization system for aluminum alloy precision forging process parameters according to claim 7, characterized in that, The acoustic-mechanical coupling control unit integrates a compliance adjustment module for performing the following magnetorheological compliance compensation control: When the state determination module determines that the rheological lock-in state is reached, the compliance adjustment module calculates the deviation between the real-time hardening rate and the preset ideal flow hardening rate. The compliance adjustment module calculates the target yield force correction based on the deviation value, reduces the excitation current of the magnetorheological damper in the corresponding area to reduce the axial support stiffness and damping coefficient, and drives the piezoelectric stack actuator to generate a slight retraction displacement to implement local compliance yielding on the lower mold assembly.

10. An adaptive optimization method for aluminum alloy precision forging process parameters, executed by an acoustic-mechanical coupling control unit in the aluminum alloy precision forging process parameter adaptive optimization system as described in any one of claims 1 to 9, characterized in that, Includes the following steps: Step S100: In the initial forming stage, the stress field distribution parameters are calculated in reverse according to the target detection area of ​​the lower mold assembly, and the first control command is sent to each dynamic actuation unit in the multi-point array active compliant mold base to drive the piezoelectric stack actuator to generate static elongation displacement to construct a static prestress field, so that the ultrasonic guided wave excited by the ultrasonic wave transceiver array is focused on the target detection area. Step S200: Under the state of maintaining the static prestress field, the echo signal reflected by the contact interface between the lower mold assembly and the aluminum alloy billet is collected, and the complex acoustic impedance spectrum data of the contact interface is calculated. The complex acoustic impedance spectrum data includes real acoustic impedance component data and imaginary acoustic impedance component data. Step S300: Based on the complex acoustic impedance spectrum data calculated in step S200, determine the current contact mechanical state of the target detection area. The contact mechanical state includes normal sliding state, interface bonding state, and rheological locking state. Step S400: Adjust the dynamic actuation unit according to the contact mechanical state determined in step S300: when it is determined to be the interface bonding state, drive the piezoelectric stack actuator to superimpose high-frequency axial mechanical vibration while maintaining static elongation displacement; when it is determined to be the rheological lock-up state, reduce the excitation current of the magnetorheological damper and drive the piezoelectric stack actuator to generate a slight retraction displacement. Step S500: During the downward movement of the slider component, the position coordinates of the target detection area are updated according to the real-time acquired acoustic impedance change position. Based on the updated position coordinates, the stress field distribution parameters are recalculated and the static elongation displacement of each piezoelectric stack actuator is adjusted. Steps S200 to S400 are then executed cyclically based on the updated static prestress field.

Images