Electromagnetic drive Hopkinson multi-field coupling data acquisition and calculation system and method

Abstract

The invention relates to the technical field of material dynamic mechanical property testing, and discloses an electromagnetic drive Hopkinson multi-field coupling data acquisition and calculation system and method, and the system comprises a physical loading subsystem and a multi-field resolving controller. The method comprises the following steps: firstly, performing ultrasonic signal decoupling based on Hilbert envelope and statistical noise features, and identifying a rough damage time window; then accurately positioning the crack initiation moment in the time window by using gray gradient entropy analysis; meanwhile, a conservation model of mechanical deformation work and heat dissipation energy is established in the plastic hardening stage of the test block, and the optimal lag parameter of the infrared temperature data is solved. And finally, aligning multi-field data by using the parameter, eliminating an adiabatic temperature rise effect, and generating a composition spectrum in combination with a crack moment. According to the invention, through acousto-optic heterogeneous cascade and energy conservation calibration, the problem that multi-physical field data time sequence dislocation and weak damage are difficult to identify is solved, and accurate characterization of the dynamic performance of the material is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of material dynamic mechanical property testing, in particular to an electromagnetic driving Hopkinson multi-field coupling data acquisition and calculation system and method. BACKGROUND

[0002] In the study of material dynamic mechanical properties, split Hopkinson pressure bar (SHPB) experiments are often combined with high-speed photography and infrared thermal imaging techniques to synchronously obtain multi-physical field data such as stress wave propagation, macroscopic deformation evolution and temperature field changes of the sample under high-speed impact. The fusion analysis of such multi-dimensional information is crucial for revealing the adiabatic shear, dynamic fracture and thermal softening mechanisms of materials.

[0003] However, in actual multi-field coupling tests, the physical response mechanisms and signal transmission links of different modal sensors differ significantly. In particular, infrared temperature sensors often have inherent photoelectric response hysteresis, and simply relying on traditional hardware TTL level synchronous triggering cannot completely eliminate the time sequence deviation of temperature data relative to stress wave data. This microsecond-level time misalignment can cause errors in subsequent stress correction based on adiabatic temperature rise, making it difficult to accurately distinguish between thermal softening effects and physical damage softening of materials. In addition, in the strong noise environment of high-speed impact, it is difficult to sensitively capture early weak damage signals relying solely on the voltage amplitude mutation of stress wave signals, and directly processing massive high-speed image sequences in the full time domain not only has a huge computational load, but also the conventional gray threshold analysis method is easily disturbed by surface texture changes when facing micro-crack initiation, making it difficult to accurately locate the damage time.

[0004] Therefore, the present application proposes an electromagnetic driving Hopkinson multi-field coupling data acquisition and calculation system and method to solve the deficiencies of the prior art. SUMMARY

[0005] To address the deficiencies of the prior art, the present application provides an electromagnetic driving Hopkinson multi-field coupling data acquisition and calculation system and method, which solves the problems of time sequence lag of infrared temperature data in Hopkinson multi-field coupling experiments and difficulty in accurately positioning early weak damage in a strong noise environment, thereby causing distortion of material dynamic performance characterization.

[0006] To achieve the above purpose, the present application is implemented by the following technical solution: an electromagnetic driving Hopkinson multi-field coupling data acquisition and calculation system, comprising a physical loading subsystem and a multi-field calculation controller; The physical loading subsystem is used to perform impact experiments and provide stress wave signals, ultrasonic signals, image sequences and infrared temperature data; The multi-field solution controller is connected with the physical loading subsystem, and includes a parameter calibration module, a signal decoupling module, a time sequence triggering module, a space-time locking module, an energy calibration module, and a constitutive generation module; The parameter calibration module is used for constructing a system characteristic database containing an ultrasonic background envelope feature and a statistical noise standard deviation. The signal decoupling module is used for performing differential calculation on the measured ultrasonic signal in combination with the system characteristic database to identify a damage rough time window. The space-time locking module is used for locating a crack initiation time through gray gradient entropy analysis within the damage rough time window. The energy calibration module is used for constructing an energy conservation model within a plastic hardening stage interval and solving an optimal time lag parameter of infrared temperature data. The constitutive generation module is used for aligning the infrared temperature data by using the optimal time lag parameter and generating a multi-field coupling constitutive map in combination with the crack initiation time.

[0007] Preferably, the parameter calibration module adopts a Hilbert transform algorithm to analyze a background signal under an unloaded working condition, extracts a modulus value of the analyzed signal as a background envelope signal, selects a non-impact steady-state interval in the background envelope signal for variance analysis, and calculates the statistical noise standard deviation of the background noise; and the background envelope signal and the statistical noise standard deviation are stored in the system characteristic database.

[0008] Preferably, the signal decoupling module calculates a measured envelope signal of the measured ultrasonic signal, calculates an absolute value of a difference between the measured envelope signal and a corresponding background envelope signal in the system characteristic database to obtain a difference signal; the signal decoupling module generates a dynamic judgment threshold value by using the statistical noise standard deviation in combination with a preset sensitivity coefficient, marks a damage initiation time and generates the damage rough time window when the difference signal continuously exceeds the dynamic judgment threshold value within a preset continuous time length.

[0009] Preferably, the time sequence triggering module calculates an elastic longitudinal wave velocity according to a rod material attribute, calculates a theoretical delay time of stress wave propagation in combination with a sensor installation position, subtracts a preset pre-buffer time to obtain a triggering time, and controls a high-speed camera and an infrared temperature sensor in the physical loading subsystem to synchronously collect within an effective collection window.

[0010] Preferably, the spatiotemporal locking module uses the rough damage time window as a time index to intercept the image sequence; for each frame of the intercepted image sequence, the spatiotemporal locking module calculates the probability distribution of the gradient amplitude in the image and calculates the gray gradient entropy; the spatiotemporal locking module performs first-order differential operation on the time sequence curve of the gray gradient entropy, and determines the crack initiation time by searching for the extreme point of the differential curve.

[0011] Preferably, the energy calibration module defines the plastic hardening stage interval by identifying the yield point time and the peak stress point time in the stress-strain curve; in the plastic hardening stage interval, the energy calibration module calculates the cumulative mechanical deformation work density by integrating the dynamic true stress and the plastic strain rate, and calculates the heat dissipation energy density by using the infrared temperature data in combination with the specific pressure heat capacity of the material and the Taylor-Quinney coefficient.

[0012] Preferably, the energy calibration module sets a time lag parameter to be solved within a preset physical boundary constraint range, constructs an energy residual error objective function, and the energy residual error objective function represents the aggregation degree of the difference between the cumulative mechanical deformation work density and the heat dissipation energy density after time translation correction; the energy calibration module uses a numerical optimization algorithm to search for the parameter value that makes the energy residual error objective function reach a minimum value, as the optimal time lag parameter.

[0013] Preferably, the constitutive generation module uses the optimal time lag parameter to perform time sequence translation correction on the whole process of the infrared temperature data; the constitutive generation module uses the corrected temperature data and the material thermal softening index to perform reverse compensation on the true stress data to eliminate the adiabatic temperature rise effect, and calculates the isothermal equivalent true stress.

[0014] Preferably, the constitutive generation module takes the crack initiation time as a dividing point and calculates a damage factor after the crack initiation time; the damage factor is obtained by calculating the deviation ratio between the isothermal equivalent true stress and the ideal undamaged extrapolated stress, and the ideal undamaged extrapolated stress is obtained by fitting and predicting based on the plastic hardening data before the crack initiation time.

[0015] The application also provides an electromagnetic driving Hopkinson multi-field coupling data acquisition and calculation method, comprising the following steps: Extracting the ultrasonic background envelope features and the statistical noise standard deviation under the no-load working condition and storing them in the system feature database; Performing difference calculation on the measured ultrasonic signals in combination with the system feature database, setting a dynamic judgment threshold value by using the statistical noise standard deviation, and identifying a rough damage time window; According to the stress wave velocity calculation theory, a delay time is calculated, and the physical loading subsystem is controlled to be synchronized in the effective acquisition window. The image sequence is intercepted by using the damage rough time window, and the crack initiation time is located by calculating the local gray gradient entropy of the image and the change rate of the gray gradient entropy. An energy conservation model of deformation work and heat dissipation energy is established in the plastic hardening stage interval to solve the optimal time lag parameter of the infrared temperature data. The optimal time lag parameter is used to align the infrared temperature data, and the crack initiation time and damage factor are fused to output the multi-field coupling constitutive graph.

[0016] The application provides an electromagnetic drive Hopkinson multi-field coupling data acquisition and calculation system and method. 1、The application uses Hilbert transform to extract signal envelope and combines statistical noise standard deviation to set dynamic threshold, replacing the traditional fixed voltage threshold determination. This method effectively avoids the interference of high-frequency carrier phase jitter and environmental electromagnetic noise, and improves the signal-to-noise ratio and accuracy of identifying early weak damage signals under complex impact conditions.

[0017] 2、The application indexes the image sequence through the ultrasonic rough time window, and calculates the local gray gradient entropy and its change rate. This cascading strategy avoids the algorithm redundancy caused by full-time domain image processing, and solves the problem that the traditional gray threshold method cannot identify early fine cracks by using the sensitivity of entropy value to the complexity of image texture, and improves the time positioning accuracy to the microsecond level.

[0018] 3、The application inverses the optimal time lag parameter of the infrared temperature data based on the energy conservation model of the plastic hardening stage, eliminating the data misplacement caused by sensor response delay. This makes the system accurately remove the thermal softening effect caused by adiabatic temperature rise, so as to construct a material dynamic constitutive graph containing real damage factor and conforming to the law of thermodynamics. DETAILED DESCRIPTION

[0019] Figure 1 The application is an electromagnetic drive Hopkinson multi-field coupling data acquisition and calculation system structure diagram; Figure 2 The application is an electromagnetic drive Hopkinson multi-field coupling data acquisition and calculation method flow chart; Figure 3 The application is a waveform schematic diagram of ultrasonic signal decoupling and damage time window identification; Figure 4 The application is a data processing curve diagram for locking the crack initiation time based on the gray gradient entropy mutation; Figure 5 The application is an energy conservation calibration and time lag parameter optimization comparison diagram in the plastic hardening verification interval; Figure 6 schematic diagram of a physical loading subsystem of the present application; Figure 7 schematic diagram of a physical loading subsystem of the present application; Figure 6 enlarged schematic diagram of A in the middle; Figure 8 schematic diagram of the present application.

[0020] wherein, 1, incident rod; 2, transmission rod; 3, base; 4a, first guide bracket; 4b, second guide bracket; 5a, first strain gauge; 5b, second strain gauge; 6, test block; 7, test block fixed end; 8a, ultrasonic transmitter; 8b, ultrasonic receiver; 9, high-speed camera; 10, infrared temperature sensor; 20, multi-field solution controller; 21, parameter calibration module; 22, signal decoupling module; 23, time sequence triggering module; 24, space-time locking module; 25, energy calibration module; 26, constitutive generation module. DETAILED DESCRIPTION

[0021] The specific embodiments of the present application are further described in detail below with reference to the accompanying drawings and examples.

[0022] Referring to the accompanying Figure 1 , the accompanying Figure 6 , Figure 7 and Figure 8 , the present application provides an electromagnetic drive Hopkinson multi-field coupling data acquisition and calculation system, which comprises a physical loading subsystem and a multi-field solution controller 20.

[0023] The physical loading subsystem is used to perform impact experiments and provide raw physical field data; the physical loading subsystem structurally comprises an incident rod 1, a transmission rod 2, a base 3, a first guide bracket 4a, a second guide bracket 4b, a first strain gauge 5a, a second strain gauge 5b, a test block 6, a test block fixed end 7, an ultrasonic transmitter 8a, an ultrasonic receiver 8b, a high-speed camera 9, and an infrared temperature sensor 10. The incident rod 1 and the transmission rod 2 are coaxially arranged on the base 3 through the first guide bracket 4a and the second guide bracket 4b; the test block 6 is clamped between the incident rod 1 and the transmission rod 2; the first strain gauge 5a and the second strain gauge 5b are respectively pasted on the surfaces of the incident rod 1 and the transmission rod 2; the ultrasonic transmitter 8a and the ultrasonic receiver 8b are respectively embedded in the test block fixed ends 7 on both sides; the high-speed camera 9 and the infrared temperature sensor 10 are arranged on both sides of the test block 6.

[0024] The multi-field solution controller 20 is respectively connected in signal with the first strain gauge 5a, the second strain gauge 5b, the ultrasonic transmitter 8a, the ultrasonic receiver 8b, the high-speed camera 9, and the infrared temperature sensor 10; the multi-field solution controller 20 is used to perform data synchronous acquisition, signal interference elimination, and multi-physical field coupling calculation. For example, Figure 1As shown, the multi-field solution controller 20 logically comprises a parameter calibration module 21, a signal decoupling module 22, a timing trigger module 23, a space-time locking module 24, an energy calibration module 25, and a constitutive generation module 26. The various modules interact with each other through an internal bus.

[0025] The parameter calibration module 21 is used to establish a correlation model of electromagnetic drive parameters and system response baseline before the experiment, and to construct a system feature database containing ultrasonic background envelope features and statistical noise standard deviation; the timing trigger module 23 is used to monitor the strain gauge signal, calculate the waveform arrival time according to the wave velocity, and control each sensor to start acquisition within the effective time window; the signal decoupling module 22 is used to call the system feature database to perform envelope processing and difference calculation on the measured ultrasonic signal, identify ultrasonic transmission abnormalities and mark the rough time window of the damage; the space-time locking module 24 is used to call the high-speed image sequence within the rough time window, and determine the crack initiation time through gray gradient entropy analysis; the energy calibration module 25 is used to construct an energy conservation model and solve the optimal time lag parameter of the infrared temperature data within the plastic hardening stage interval of the test block; the constitutive generation module 26 is used to align the infrared data using the time lag parameter, and combine the crack initiation time to fuse the damage factor, and generate a multi-field coupled constitutive map.

[0026] Referring to the accompanying drawings Figure 2 The present application provides a kind of electromagnetic drive Hopkinson multi-field coupling data acquisition calculation method, comprising the following steps: S100, construct a system feature database with statistical characteristics. The parameter calibration module 21 establishes a correlation model of electromagnetic drive parameters and system response baseline, extracts ultrasonic background envelope features and statistical noise standard deviation under no-load operating conditions, and stores the correlation data into the system feature database; S200, signal decoupling based on Hilbert envelope and statistical criteria. The signal decoupling module 22 acquires the measured ultrasonic signal and calculates the measured envelope, performs difference calculation in combination with the background envelope in the system feature database, sets a dynamic threshold using the statistical noise standard deviation, and identifies the rough time window of the damage; S300, execute zero-redundancy timing control based on wave velocity calculation. The timing trigger module 23 monitors the stress wave signals of the incident rod 1 and the transmission rod 2, calculates the delay time according to the sound velocity, and controls the high-speed camera 9 and the infrared temperature sensor 10 to synchronously acquire within the effective window; S400, perform acoustic-optical heterogeneous cascade space-time locking. The space-time locking module 24 intercepts the high-speed image sequence using the rough time window of the damage, calculates the local gray gradient entropy of the image and its change rate, and locates the crack initiation time; S500, perform bounded energy anchoring based on plastic hardening segment. The energy calibration module 25 identifies the plastic hardening interval of the test block 6, establishes an energy conservation model of deformation work and heat dissipation energy in the interval, and solves the optimal time lag parameter of the infrared data within the physical constraint range; S600, generate a full-process multi-field coupling constitutive relationship. The constitutive generation module 26 uses the optimal time lag parameter to perform time sequence alignment on the full-process infrared data, uses the crack initiation time as a demarcation point to fuse the ultrasonic damage factor, and outputs a dynamic constitutive graph containing true stress, strain and damage state.

[0027] Referring to the drawings Figure 1 The electromagnetic drive Hopkinson multi-field coupling data acquisition and calculation system provided by the embodiment comprises a physical loading subsystem, which is a basic hardware platform for obtaining original data of dynamic mechanical properties of materials.

[0028] The main structure of the physical loading subsystem comprises an incident rod 1 and a transmission rod 2, both of which are coaxially arranged along the horizontal direction. The incident rod 1 is installed on the base 3 through the first guide bracket 4a, the first guide bracket 4a limits the radial displacement of the incident rod 1 and allows the axial sliding of the incident rod 1. The transmission rod 2 is installed on the base 3 through the second guide bracket 4b, the second guide bracket 4b also limits the radial displacement of the transmission rod 2 and allows the axial sliding of the transmission rod 2. The first guide bracket 4a and the second guide bracket 4b are the same in structure, which ensures the coaxiality of the incident rod 1 and the transmission rod 2 during the impact process. The test block 6 is clamped between the two end faces of the incident rod 1 and the transmission rod 2 close to each other, and is kept in a fixed position by the end face friction force.

[0029] In order to realize effective integration of the ultrasonic sensor and the rod system and prevent impact damage, the end of the incident rod 1 close to the transmission rod 2 is connected with the test block fixing end head 7 through threads; the end of the test block fixing end head 7 close to the test block 6 is designed as a solid flat structure, which can ensure the smooth transmission of stress waves when passing through the contact interface and reduce the reflection interference caused by wave impedance mismatch; the test block fixing end head 7 is made of high-strength metal material to withstand the compression load generated by the bullet impact.

[0030] In terms of damage monitoring, the ultrasonic transmitter 8a is installed in the test block fixing end head 7 connected to the incident rod 1, and the ultrasonic receiver 8b is installed in the test block fixing end head 7 connected to the transmission rod 2. The ultrasonic transmitter 8a and the ultrasonic receiver 8b both adopt an embedded installation structure, the sensing surface of which is located inside the solid flat structure of the test block fixing end head 7, and does not expose the end face of the test block fixing end head 7 in contact with the test block 6. This non-contact embedded layout avoids the mechanical impact of the ultrasonic probe directly bearing the test block 6 breaking, and also prevents the wear caused by the direct contact of the probe and the test block 6, and uses the metal medium of the test block fixing end head 7 as an acoustic coupling layer to realize the transmission and reception of ultrasonic waves.

[0031] In the aspect of stress wave monitoring, the middle surface of the incident rod 1 is pasted with a first strain gauge 5a for collecting incident wave and reflected wave signals; the middle surface of the transmission rod 2 is pasted with a second strain gauge 5b for collecting transmission wave signals; the first strain gauge 5a and the second strain gauge 5b are connected to the input end of the controller through shielding wires, so as to provide original voltage data for subsequent time sequence control and constitutive calculation.

[0032] In the aspect of non-contact field data acquisition, a high-speed camera 9 and an infrared temperature sensor 10 are arranged on the two sides of the test block 6 respectively; the high-speed camera 9 is fixed on the base 3 through an independent support, and the lens optical axis thereof is perpendicular to the side surface of the test block 6, for recording the surface deformation image sequence of the test block 6 in the impact process; the infrared temperature sensor 10 is also fixed through a support, and is located on the other side of the test block 6, with the field center thereof being aligned with the center of the side surface of the test block 6, for collecting infrared radiation temperature data of the surface of the test block 6; the spatial positions of the high-speed camera 9 and the infrared temperature sensor 10 do not interfere with each other, and both are electrically connected with the controller to accept unified time sequence triggering instructions of the controller.

[0033] Referring to the accompanying drawings Figure 2 The parameter calibration module 21 performs step S100 before the compression test starts, to construct a system characteristic database with statistical characteristics; this process aims to eliminate inherent errors of the measurement system itself, and to establish reference data for subsequent damage judgment.

[0034] The electromagnetic drive estimation model pre-stored in the parameter calibration module 21 is used to describe the nonlinear mapping relationship between the charging voltage of the electromagnetic launching system and the bullet impact speed; this model is constructed by using a polynomial fitting method. Specifically, through a plurality of groups of no-load launching experiments under different charging voltages performed in advance, discrete data points of the voltage and the measured speed are obtained, and a quadratic function relationship formula about the voltage is obtained by using the least square method fitting: , wherein is a system inherent coefficient determined by experimental data fitting. The system indexes the corresponding ultrasonic reference data in the system characteristic database according to the estimated bullet impact speed, so as to realize adaptive matching under different loading rates.

[0035] In the process of constructing the system characteristic database, the system needs to run under the no-load working condition without clamping the test block. At this time, the ultrasonic transmitter 8a transmits an ultrasonic pulse, the pulse passes through the test block fixing end head 7 of the incident rod 1, is directly coupled to the test block fixing end head 7 of the transmission rod 2, and is received by the ultrasonic receiver 8b, to form a background signal Since the propagation characteristics of ultrasonic waves in metal rods are extremely sensitive to minor environmental changes, directly using the time-domain original waveform as a reference can produce false differential signals due to phase jitter. Therefore, the parameter calibration module 21 uses the Hilbert transform algorithm to analyze the background signal and extract its energy envelope features.

[0036] Specifically, the parameter calibration module 21 performs Hilbert transform on the collected background signal to construct the imaginary part of the analytic signal thereof; the Hilbert transform is achieved by shifting the phase of all frequency components of the original signal by 90 degrees, and its calculation expression is: ; In the formula, represents the Hilbert transform conjugate signal of the background signal , represents the time variable, represents the integral variable; and represents the Cauchy principal value integral.

[0037] Based on the conjugate signal, the parameter calibration module 21 further calculates the modulus of the analytic signal to obtain the background envelope signal , and the calculation formula is: ; In the formula, represents the extracted background envelope signal. The background envelope signal characterizes the instantaneous distribution of ultrasonic energy in the system structure and removes the phase information of the high-frequency carrier. Using the background envelope signal instead of the original radio frequency waveform as the fingerprint feature can effectively eliminate the phase jitter interference caused by temperature drift and slight changes in contact pressure, ensuring the robustness of subsequent damage determination. The system stores the background envelope signals under different voltage levels and their corresponding driving parameters into a database to form a system feature database.

[0038] With further reference to the accompanying Figure 2 , after the parameter calibration module 21 extracts the background envelope signal, the parameter calibration module 21 continues to perform statistical quantization processing of the system background noise features; since the Hopkinson pressure bar experiment site usually has a complex electromagnetic environment, and the capacitor discharge process can introduce random line thermal noise and mechanical micro-vibration, it is difficult to simultaneously consider the sensitivity and anti-interference ability of the detection using a fixed voltage threshold for signal determination; therefore, the parameter calibration module 21 uses a statistical analysis method to extract the dispersion characteristics of the background noise, providing data support for subsequent adaptive threshold generation.

[0039] The parameter calibration module 21 first calculates the mean value of the background envelope signal A steady-state interval within the non-impact period is extracted as a statistical sample. This steady-state interval is selected as the signal stability segment after the ultrasonic emission pulse excitation and before the echo signal arrives, to eliminate the interference of structural reflection waves on noise statistics. The parameter calibration module 21 performs variance analysis on the discrete sampling points within this steady-state interval to calculate the standard deviation of noise statistics. The calculation formula is as follows: ; In the formula, The statistical standard deviation of background noise. This represents the total number of sampling points within the selected steady-state interval. Indicates the first term within the steady-state interval Background envelope amplitude at each sampling time, This represents the arithmetic mean of the background envelope signal amplitude within the steady-state interval.

[0040] Calculated noise statistical standard deviation It does not directly participate in waveform reconstruction, but is stored in the system feature database as a statistical indicator to measure the signal-to-noise ratio level of the current test system; the physical meaning of this indicator lies in quantifying the background fluctuation range of the system under the current operating conditions; in the subsequent damage identification steps, the system will utilize this... Values ​​combined with statistical criteria (such as...) (Criteria) Dynamically generate judgment thresholds. Compared with traditional fixed absolute value thresholds, this calibration method based on statistical standard deviation can automatically adjust the trigger sensitivity according to changes in environmental noise, ensuring that the system does not trigger falsely in high-noise environments, while being able to identify weak crack signals submerged in the background in low-noise environments, thereby improving the robustness and accuracy of damage monitoring.

[0041] See attached document Figure 2 During the compression test, the signal decoupling module 22 executes step S200, which involves decoupling the signal based on the Hilbert envelope and statistical criteria. The core of this step is to perform time-domain alignment and differential analysis between the real-time acquired ultrasonic signal containing damage information and the non-destructive reference signal in the system feature database, thereby separating the signal distortion component caused solely by damage inside the test block.

[0042] The signal decoupling module 22 receives the measured signal output by the ultrasonic receiver 8b in real time. Meanwhile, the signal decoupling module 22 retrieves the corresponding background envelope signal from the system feature database based on the electromagnetic emission voltage set in the current experiment and the estimated impact velocity. To eliminate interference caused by high-frequency carrier phase asynchrony in the measured signal, the signal decoupling module 22 uses the same Hilbert transform algorithm as the parameter calibration module 21 to perform the measurement on the measured signal. The measured envelope signal is obtained through processing and calculation. On this basis, the signal decoupling module 22 performs baseline difference operation, calculates the absolute value of the difference between the measured envelope and the background envelope, and obtains the difference signal representing the damage degree , and the calculation formula is: ; In the formula, represents the difference signal, and the amplitude directly reflects the scattering and attenuation degree of the ultrasonic wave when penetrating the test block 6, represents the energy envelope of the measured signal, represents the corresponding background reference envelope.

[0043] In view of the difference in acoustic response characteristics of different brittle or ductile materials in the damage evolution process and the fluctuation of the noise level in the experimental environment, the signal decoupling module 22 discards the fixed voltage threshold determination method, and instead uses a self-adaptive criterion based on the statistical noise standard deviation to generate a dynamic determination threshold. This method uses the statistical noise standard deviation prestored by the parameter calibration module 21, and combines the sensitivity coefficient to calculate the dynamic determination threshold : ; In the formula, represents the dynamic amplitude threshold for determining the damage initiation, represents the sensitivity coefficient. The value range of the sensitivity coefficient is usually set to 3 to 5, and the specific value is pre-set according to the acoustic attenuation characteristics of the test block material: for metal materials with good acoustic impedance matching and internal uniformity, a smaller value is selected to improve the sensitivity to micro-cracks; for non-homogeneous materials such as concrete or rock, a larger value is selected to reduce the misjudgment rate caused by background scattering.

[0044] The signal decoupling module 22 compares the difference signal with the dynamic determination threshold point by point. In order to prevent false triggering caused by transient electromagnetic peak pulses, the signal decoupling module 22 sets a continuity determination logic; when the difference signal is continuously higher than the dynamic determination threshold for a certain length of time, the system determines that irreversible damage has occurred inside the test block, and marks the time as the damage initiation time . The determination logic is described as: ; In the formula, represents preset minimum continuous determination time, used to filter out sporadic spike noise. Once the above conditions are met, the signal decoupling module 22 will lock the time period containing , generating a damage rough time window . The damage rough time window covers the entire interval from backwards in time to the signal recovery smoothness, and the window information is then transmitted to the time-space locking module 24 as a time index for subsequent high-speed image retrieval.

[0045] Referring to the accompanying Figure 3 , the figure directly shows the internal signal processing process when the signal decoupling module 22 of the present application performs the above steps. The horizontal axis of the figure represents time , and the vertical axis represents signal amplitude . As shown in the figure, the background envelope Figure 3 is the reference signal obtained by the system calling the system feature database, which is expressed as a low-amplitude smooth noise baseline; and the measured envelope is the ultrasonic signal collected in the experiment and subjected to Hilbert transform.

[0046] In the early stage of the experiment (about to 40), the measured envelope substantially coincides with the background envelope, indicating that no damage has occurred inside the test block. When the time reaches the moment marked in the figure , the measured envelope mutates due to acoustic scattering caused by internal damage of the test block, and its amplitude exceeds the dynamic threshold . The system captures the intersection point, not only locking the damage starting time , but also generating a damage rough time window containing the mutation point before and after it. The result proves that the present application can effectively extract the weak damage feature signal in a strong background noise environment by using difference and statistical threshold logic.

[0047] Referring to the accompanying Figure 2 , the time sequence triggering module 23 performs step S300, i.e. zero-redundancy time sequence control based on wave velocity calculation, during the experiment. This step aims to solve the storage limitation problem caused by the huge data throughput of the high-speed camera 9 and the infrared temperature sensor 10, and realizes accurate locking of the effective observation window by accurately calculating the propagation delay of the stress wave in the physical rod.

[0048] The time sequence triggering module 23 monitors the voltage signal output by the first strain gauge 5a in real time. When the amplitude of the voltage signal is detected to exceed the preset trigger level, the time sequence triggering module 23 records the moment ​At this point, the stress wave has not yet reached test block 6. The timing trigger module 23 uses the material properties of the Hopkinson bar to calculate the propagation velocity of the one-dimensional elastic longitudinal wave in the bar. For the uniformly materialed incident bar 1, its elastic longitudinal wave velocity depends on the elastic modulus and density of the bar material, and the calculation formula is: ; In the formula, This indicates the elastic longitudinal wave velocity of the incident rod. This represents the elastic modulus of the incident rod material. This indicates the density of the incident rod material.

[0049] After obtaining the wave velocity, the timing trigger module 23 determines the physical distance between the installation position of the first strain gauge 5a and the contact end face of the test block 6 based on the geometric dimension data of the incident rod 1. Based on this distance and the calculated wave velocity, the timing triggering module 23 calculates the theoretical delay time required for the stress wave to propagate from the monitoring point to the test block 6. : ; In the formula, Indicates the theoretical propagation delay time. This represents the axial distance from the first strain gauge to the front end face of the specimen.

[0050] To ensure that the high-speed camera 9 and the infrared temperature sensor 10 can completely record the initial state of the test block 6 before the impact and the dynamic changes at the moment of impact, and to avoid invalid data occupying storage space due to premature start of acquisition, the timing trigger module 23 sets a microsecond-level pre-buffer time. This pre-buffer time is typically set to tens of microseconds to cover the device's hardware startup latency and establish a steady-state baseline. The timing trigger module 23 generates a synchronization trigger command based on this, and the timing of the command's transmission... Set as: ; In the formula, This indicates the absolute moment when the acquisition command is sent to the high-speed camera and infrared temperature sensor. This indicates the moment when the first strain gauge detected a signal. This indicates the preset pre-buffer time.

[0051] In addition to determining the trigger time, the timing trigger module 23 also needs to determine the duration of the acquisition window. This achieves "zero redundancy" acquisition. The duration depends on the length of the electromagnetically driven bullet (impact rod), as the pulse width of the stress wave is proportional to the bullet length. The timing trigger module 23 adjusts the timing based on the bullet length. Set the effective data collection window duration: ; wherein, represents the collection opening duration of the sensor, represents the axial length of the bullet, represents the residual time reserved for recording the unloading process after impact. The time sequence triggering module 23 will calculate the and parameters and issue them to each sensor controller, ensuring that the multi-physical field data collection is strictly limited to the effective period of dynamic deformation of the test block, achieving an optimal balance between data storage efficiency and capture accuracy.

[0052] With reference to the accompanying drawings, Figure 2 the space-time locking module 24 performs step S400, i.e., acoustic-optical heterogeneous cascade space-time locking, in the data processing stage of the compression test. This step uses the previously obtained rough time window of ultrasonic damage as the index of the time dimension to perform dimension reduction retrieval on the massive image data collected by the high-speed camera 9. Since the high-speed camera 9 usually operates at a rate of hundreds of thousands of frames per second, the amount of data generated is extremely large, and direct full-time domain processing is extremely inefficient. The space-time locking module 24 only retrieves the image frame sequence corresponding to the time window as the region of interest (ROI) sequence, thereby excluding invalid image frames before impact and after fragmentation, and achieving a cascade jump from acoustic rough positioning to optical fine analysis.

[0053] After obtaining the region of interest sequence, the space-time locking module 24 further uses an algorithm based on gray level gradient entropy to accurately locate the crack initiation time. Traditional edge detection methods based on pixel gray level threshold are easily disturbed by uneven lighting and surface texture of the test block in the Hopkinson bar experiment. In contrast, image information entropy can reflect the complexity and disorder of image texture. When microcracks appear on the surface of the test block, the texture complexity of the local image will change abruptly, causing an increase in information entropy. In order to improve the sensitivity to the edges of microcracks, the space-time locking module 24 calculates the gray level gradient entropy of the image instead of simply the gray level entropy.

[0054] Specifically, the space-time locking module 24 calculates the gradient of each frame of image in the sequence using the Sobel operator or the Prewitt operator to calculate the gradient and the gradient of the image in the horizontal direction and the vertical direction, respectively, and obtains the gradient amplitude of each pixel point in the image according to the formula . Subsequently, the probability distribution of different gradient amplitudes appearing in the entire image is calculated. Assuming that the gradient amplitude is quantized into levels, the probability of the gradient amplitude of the level appearing in the image is the gray level gradient entropy of the image at the moment, is defined as: ; wherein, represents the gray level gradient entropy of the image at the moment, represents the occurrence probability of the gradient amplitude of the level in the image at the moment.

[0055] As the loading process proceeds, the spatiotemporal locking module 24 generates a time series curve of the gray level gradient entropy varying with time. Since the crack initiation is a structural mutation process caused by instantaneous energy release, the process corresponds to the moment when the rate of change of the entropy value curve is the largest. Therefore, the spatiotemporal locking module 24 performs first-order differentiation on the entropy value curve and searches for the extreme point of the differentiated curve to lock the crack initiation moment : ; wherein, represents the calculated accurate crack initiation moment, represents the first-order derivative of the gray level gradient entropy with respect to time. The moment has a time resolution of microseconds, providing an accurate time reference point for subsequent multi-field data fusion.

[0056] Referring to the accompanying Figure 4 , the figure shows the specific calculation results of the entropy analysis of the high-speed image sequence by the spatiotemporal locking module 24 within the rough time window of damage.

[0057] As Figure 4 shown, the gray level gradient entropy presents a nonlinear change with the progress of impact loading. The curve shows that the entropy value remains relatively flat at the beginning, and presents a sharp rising trend of S type near the moment marked in the figure , which corresponds to the instantaneous initiation and expansion of micro-cracks on the surface of the test block, leading to a sudden mutation of optical texture complexity. The entropy change rate dH / dt (marked as a dashed line in the legend) as the first-order derivative of the entropy curve presents a sharp peak at the most severe mutation of the entropy value. The present application accurately locks the crack initiation moment by searching for the global maximum point of the derivative curve, thereby improving the positioning accuracy of the damage moment from the rough range of ultrasound to the microsecond accuracy of the image frame level.

[0058] Referring to the accompanying Figure 2When executing step S500, the energy calibration module 25 first defines the plastic hardening verification interval. Due to the physical characteristics of the photoelectric conversion and integration circuit of the infrared detector, the acquired temperature data has an inherent time lag relative to the stress wave data. To eliminate this systematic error, the energy calibration module 25 needs to establish a time alignment reference based on the law of physical conservation. The energy calibration module 25 first identifies feature points on the stress-strain curves obtained by the first strain gauge 5a and the second strain gauge 5b to accurately locate the yield point of the material. and the time of peak stress point .

[0059] Energy calibration module 25 will be from arrive The time period is defined as the plastic hardening verification interval. Within this interval, the specimen has not yet experienced macroscopic through-cracks, and the deformation satisfies the uniformity assumption. Selecting this specific interval as the energy calibration window has significant physical implications: in the elastic stage before the yield point, the deformation work of the material is mainly converted into elastic potential energy, with extremely low heat dissipation, failing to provide a sufficient signal-to-noise ratio for thermo-mechanical alignment; however, after the peak stress point, specimen 6 often experiences macroscopic failure, accompanied by the penetration of microcracks and structural disintegration, even resulting in the splashing of specimen fragments. These splashed fragments may obstruct the field of view of the infrared temperature sensor 10, or cause the sensor to capture thermal radiation from the background environment, severely undermining the physical premise of adiabatic temperature rise. In contrast, in the plastic hardening verification interval... Inside, specimen 6 is in the stage of uniform plastic deformation. The heat generated by dislocation motion is significant and stable, and the specimen structure remains intact, satisfying the energy conservation condition of mechanical work converted into heat energy under adiabatic conditions, thus ensuring the effectiveness and accuracy of physical quantities in the calibration model.

[0060] After determining the plastic hardening verification range Subsequently, the energy calibration module 25, based on the first law of thermodynamics, calculates the mechanical deformation work density and theoretical heat dissipation energy density within the interval. For any time within the interval... The energy calibration module 25 calculates the accumulated mechanical deformation work density using dynamic real stress and plastic strain rate. Its calculation expression is: ; In the formula, Indicates the accumulation from the yield moment to Mechanical deformation work density at any given moment express The actual stress at any moment, express Plastic strain rate at time t, It is the integral variable.

[0061] Meanwhile, the energy calibration module 25 calculates the corresponding heat dissipation energy density based on the temperature rise data collected by the infrared temperature sensor 10 and the thermophysical parameters of the material. Considering that the proportion of plastic work converted into heat energy during high-speed deformation of metallic materials is not 100%, but is controlled by the Taylor-Quinney coefficient, the formula for calculating heat dissipation energy density is: ; In the formula, This represents the equivalent energy density derived from temperature data. This indicates the density of the test block material. This indicates the specific isobaric heat capacity of the test block material. Indicates that the infrared sensor is in The temperature rise relative to the initial temperature is recorded at all times. This represents the Taylor-Quinney coefficient, which can be set as a constant (e.g., 0.9) or as a function of strain rate. This is achieved by establishing... and In the plastic hardening test range Based on the correspondence within the system, the system can provide a quantitative physical constraint model for subsequent solutions to the optimal time lag parameters.

[0062] See attached document Figure 2 After establishing the correspondence between deformation work and heat dissipation energy, the energy calibration module 25 further executes a time delay parameter optimization process under bounded constraints. The core of this process lies in finding an optimal time shift amount through a mathematical optimization algorithm, ensuring that within the plastic hardening verification interval... Within the interior, the mechanical deformation work curve and the heat dissipation energy curve have the highest degree of overlap.

[0063] Energy calibration module 25 first defines the time lag parameter to be solved. Considering the photoelectric response delay of the infrared temperature sensor 10 and the physical limits of the integration time of the signal processing circuit, this time lag parameter is not an arbitrary value, but is constrained by the physical characteristics of the hardware device. Therefore, the energy calibration module 25 sets boundary constraints for the solution domain. The lower bound is set to 0, indicating no delay; the upper bound is set to... This value is set according to the maximum response time (e.g., 100 microseconds) specified in the technical manual of the infrared temperature sensor 10. That is, the solution process is strictly limited to... The calculation is performed within a closed interval to prevent the mathematical solution from getting stuck in local extrema that do not conform to physical facts.

[0064] Under the aforementioned physical boundary constraints, the energy calibration module 25 constructs parameters related to the time lag. Energy residual objective function The function aims to quantify the accumulated deviation of mechanical energy density and thermal energy density over the entire calibration interval under a given time-lag parameter. The energy residual objective function is constructed in the form of squared integral of the difference between the two, with the specific formula as follows: ; wherein, represents the energy residual objective function value, represents the starting time (yielding time) of the plastic hardening calibration interval, represents the ending time (peak stress time) of the plastic hardening calibration interval, represents the accumulated mechanical deformation work density at time , represents the specimen material density, represents the specimen material specific heat capacity, represents the temperature rise data at time after time translation correction, represents the Taylor-Quinney coefficient.

[0065] After the construction of the objective function, the energy calibration module 25 uses a one-dimensional numerical optimization algorithm (such as the golden section search method or the grid approximation method) to search for a point within the preset boundary that makes the reach a minimum value. The corresponding to this minimum value is determined as the optimal time-lag parameter : ; wherein, represents the finally calculated optimal time-lag parameter.

[0066] Once the is obtained, the energy calibration module 25 considers that the system has captured the true physical field delay characteristics. This parameter is then transmitted to the constitutive generation module 26 for global time correction of all infrared temperature data collected during the entire experimental process (including the elastic segment, plastic segment, and failure segment), i.e., making , so as to ensure that each stress-strain data point in the subsequently generated constitutive relationship map accurately corresponds to its true temperature state at the same time.

[0067] Referring to the attached Figure 5 , the figure shows the physical basis and effect of the energy calibration module 25 for aligning the timing of infrared data based on the plastic hardening calibration interval. The gray shaded area in the figure indicates the plastic hardening calibration interval , which is defined between the yielding time and the peak stress time Between; within this range, the specimen structure remains intact and undergoes uniform plastic deformation, satisfying the physical prerequisite for adiabatic temperature rise.

[0068] like Figure 5 As shown, mechanical deformation work This is a baseline energy curve calculated based on stress-strain data. Thermal energy before calibration. This is an energy curve calculated based on raw infrared temperature data. Due to the physical response lag of the infrared sensor, this curve lags behind the mechanical deformation work curve on the time axis (manifested as...). (It only started to rise after a period of time), resulting in a huge discrepancy between the two. And the calibrated thermal energy... The system uses the above objective function to solve for the optimal time lag parameter. The result is obtained after shifting and correcting the infrared data. It can be seen that the corrected thermal energy curve... The curve within the interval highly coincides with the mechanical deformation work curve.

[0069] See attached document Figure 2 In the final stage of multiphysics data acquisition, constitutive generation module 26 executes step S600, which involves constructing a multidimensional constitutive map that integrates thermo-mechanical decoupling and damage fusion. This step aims to transform the discrete, heterogeneous, and spatiotemporally biased raw experimental data into a physical model that can truly reflect the intrinsic mechanical properties of materials under dynamic impact, while eliminating the softening artifact caused by adiabatic temperature rise and clearly defining the critical nodes of damage evolution.

[0070] The constitutive generation module 26 first calls the optimal time lag parameter calculated by the energy calibration module 25. A unified phase correction is performed on the raw temperature data collected by the infrared temperature sensor 10 across the entire time domain. Since the response delay of the infrared detector is determined by the hardware characteristics of its photoelectric conversion element and subsequent integration circuit, this delay is considered a time-invariant system parameter in a single experiment. Therefore, the constitutive generation module 26 shifts the time of all raw temperature sampling points to the left. Generate a corrected temperature sequence that is strictly synchronized with the stress wave signal. This process ensures that each stress-strain data point in subsequent analyses accurately corresponds to the material's true temperature state at that moment, eliminating the "pseudo-hysteresis" or "pseudo-leadership" phenomena caused by time misalignment.

[0071] After the time alignment of data is completed, the constitutive generation module 26 performs a thermal-mechanical decoupling operation. Under high strain rate impact loading, the plastic work generated inside the test block 6 will be converted into heat in a very short time, causing the material temperature to rise. This adiabatic temperature rise effect will cause the lattice thermal vibration inside the material to intensify, which macroscopically manifests as a decrease in flow stress, i.e., thermal softening effect. In order to obtain the intrinsic mechanical behavior of the material under isothermal conditions, or to separate the stress drop purely caused by microstructure damage, the constitutive generation module 26 adopts the temperature term in the Johnson-Cook constitutive model to reverse compensate the experimentally measured true stress, to calculate the isothermal equivalent true stress: ; wherein, represents the isothermal equivalent true stress after removing the thermal softening effect, represents the true stress directly measured by the Hopkinson bar experiment, which contains the thermal softening effect, represents the instantaneous temperature after time synchronization correction, represents the reference temperature of the experimental environment (usually room temperature), represents the melting point of the test block material, represents the thermal softening index of the material. Through this formula, the system can restore the theoretical hardening curve of the material under the influence of the adiabatic temperature rise, so as to separate the stress drop component caused by the temperature rise in the experimental curve.

[0072] On this basis, the constitutive generation module 26 introduces the crack initiation time output by the space-time locking module 24 as a time demarcation point for distinguishing the deformation mechanism of the material, and constructs a constitutive map containing a damage dimension. The constitutive generation module 26 divides the whole process data into a “plastic dominant domain” and a “damage dominant domain” with as the boundary. In the interval of , the mechanical response of the material is mainly controlled by dislocation multiplication and motion, and the system outputs the plastic hardening parameters after thermal decoupling; in the interval of , the decline of the carrying capacity of the material is mainly due to the nucleation and expansion of internal microcracks.

[0073] In order to quantify the influence of damage on mechanical properties, the constitutive generation module 26 calculates the damage factor in the stage. This factor is defined as the deviation ratio between the isothermal equivalent true stress and the ideal undamaged extrapolated stress, and its calculation logic is described as follows: ; wherein, represents the instantaneous damage factor, whose value is between 0 and 1, represents the isothermal equivalent true stress, The ideal undamaged stress value extrapolated from the plastic hardening trend before the time; in the specific calculation, the constitutive generation module 26 selects the stress value at the time when the effective plastic strain is 0.1 to The isothermal stress data in the interval as a sample, the least square method is used to fit the Ludwik hardening model (wherein is the effective plastic strain), the hardening parameters are obtained, and the stress prediction is carried out for the strain range of .

[0074] Finally, the constitutive generation module 26 carries out data fusion on the temperature field synchronized, the isothermal stress decoupled, and the damage factor, and outputs a four-dimensional constitutive data atlas containing the strain rate, temperature, stress and damage state. The atlas can not only be directly used for calibrating the material parameters in numerical simulation, but also can be combined with the crack morphology captured by high-speed photography to provide a full-quantitative physical basis for studying the dynamic failure mechanism of materials under extreme working conditions.

Claims

1. An electromagnetically driven Hopkinson multi-field coupled data acquisition and computing system, characterized in that, Includes a physical loading subsystem and a multi-field solution controller; The physical loading subsystem is used to perform impact experiments and provide stress wave signals, ultrasonic signals, image sequences, and infrared temperature data; The multi-field solution controller is connected to the physical loading subsystem and includes a parameter calibration module, a signal decoupling module, a timing triggering module, a spacetime locking module, an energy calibration module, and a constitutive generation module. The parameter calibration module is used to construct a system feature database containing ultrasonic background envelope features and statistical noise standard deviation; The signal decoupling module is used to perform differential calculations on the measured ultrasonic signals in conjunction with the system feature database to identify a rough time window of damage. The spatiotemporal locking module is used to locate the crack initiation time within the rough damage time window by analyzing gray-scale gradient entropy. The energy calibration module is used to construct an energy conservation model within the plastic hardening stage range and solve for the optimal time lag parameters of the infrared temperature data. The constitutive generation module is used to align the infrared temperature data using the optimal time lag parameter and generate a multi-field coupled constitutive spectrum in conjunction with the crack initiation time.

2. The electromagnetically driven Hopkinson multi-field coupled data acquisition and calculation system according to claim 1, characterized in that, The parameter calibration module uses the Hilbert transform algorithm to analyze the background signal under no-load conditions and extracts the magnitude of the analyzed signal as the background envelope signal. The parameter calibration module selects the non-impact steady-state interval in the background envelope signal for variance analysis and calculates the statistical noise standard deviation of the background noise. The background envelope signal and the statistical noise standard deviation are stored in the system feature database.

3. The electromagnetically driven Hopkinson multi-field coupled data acquisition and calculation system according to claim 1, characterized in that, The signal decoupling module calculates the measured envelope signal of the measured ultrasound signal, and calculates the absolute value of the difference between the measured envelope signal and the background envelope signal corresponding to the system feature database to obtain the differential signal. The signal decoupling module uses the statistical noise standard deviation combined with a preset sensitivity coefficient to generate a dynamic judgment threshold. When the differential signal is continuously higher than the dynamic judgment threshold within a preset continuous time length, the damage initiation time is marked and the damage coarse time window is generated.

4. The electromagnetically driven Hopkinson multi-field coupled data acquisition and calculation system according to claim 1, characterized in that, The timing triggering module calculates the elastic longitudinal wave velocity based on the material properties of the rod, calculates the theoretical delay time of stress wave propagation based on the sensor installation position, and subtracts the preset pre-buffer time to obtain the triggering time, thereby controlling the high-speed camera and infrared temperature sensor in the physical loading subsystem to synchronously acquire data within the effective acquisition window.

5. The electromagnetically driven Hopkinson multi-field coupled data acquisition and calculation system according to claim 1, characterized in that, The spatiotemporal locking module uses the damage coarse time window as a time index to extract the image sequence; for each frame in the extracted image sequence, the spatiotemporal locking module calculates the probability distribution of gradient magnitude in the image and calculates the gray-level gradient entropy; the spatiotemporal locking module performs a first-order differential operation on the time series curve of the gray-level gradient entropy, and determines the crack initiation time by searching for the extreme point of the differential curve.

6. The electromagnetically driven Hopkinson multi-field coupled data acquisition and calculation system according to claim 1, characterized in that, The energy calibration module defines the plastic hardening stage interval by identifying the yield point and peak stress point in the stress-strain curve. Within the plastic hardening stage interval, the energy calibration module calculates the cumulative mechanical deformation work density by integrating the dynamic real stress and plastic strain rate, and calculates the heat dissipation energy density by combining the infrared temperature data with the material's specific isobaric heat capacity and Taylor-Quini coefficient.

7. The electromagnetically driven Hopkinson multi-field coupled data acquisition and calculation system according to claim 6, characterized in that, The energy calibration module sets the time lag parameter to be solved within the preset physical boundary constraints and constructs the energy residual objective function. The energy residual objective function characterizes the degree of convergence of the difference between the cumulative mechanical deformation work density and the heat dissipation energy density after time translation correction. The energy calibration module uses a numerical optimization algorithm to search for the parameter value that minimizes the energy residual objective function, and uses it as the optimal time lag parameter.

8. The electromagnetically driven Hopkinson multi-field coupled data acquisition and calculation system according to claim 1, characterized in that, The constitutive generation module uses the optimal time lag parameter to perform time-series translation correction on the infrared temperature data throughout the process; the constitutive generation module uses the corrected temperature data and the material thermal softening index to perform reverse compensation on the real stress data to eliminate the adiabatic temperature rise effect, and calculates the isothermal equivalent real stress.

9. The electromagnetically driven Hopkinson multi-field coupled data acquisition and calculation system according to claim 8, characterized in that, The constitutive generation module uses the crack initiation time as the dividing point and calculates the damage factor after the crack initiation time. The damage factor is obtained by calculating the deviation ratio between the isothermal equivalent true stress and the ideal non-destructive extrapolated stress. The ideal non-destructive extrapolated stress is obtained by fitting and predicting based on the plastic hardening data before the crack initiation time.

10. An electromagnetically driven Hopkinson multi-field coupling data acquisition and calculation method, applied to the system described in any one of claims 1-9, characterized in that, Includes the following steps: Under no-load conditions, the ultrasonic background envelope features and the statistical noise standard deviation are extracted and stored in the system feature database. The measured ultrasonic signal is differentially calculated by combining the system feature database, and a dynamic judgment threshold is set by using the statistical noise standard deviation to identify the rough time window of the damage. Based on the theoretical delay time calculated according to the stress wave velocity, the physical loading subsystem is controlled to synchronously acquire data within the effective acquisition window. The image sequence is extracted using the aforementioned coarse damage time window, and the moment of crack initiation is located by calculating the local gray-level gradient entropy and the rate of change of the gray-level gradient entropy. An energy conservation model for deformation work and heat dissipation energy is established within the plastic hardening stage interval, and the optimal time lag parameter of the infrared temperature data is solved. The infrared temperature data is aligned using the optimal time lag parameter, and the crack initiation time and damage factor are fused to output the multi-field coupled constitutive spectrum.

Images