Electro-assisted biaxial tensile forming limit test system and method for coated sheet materials

Abstract

This invention relates to an electrically assisted biaxial tensile forming limit testing system and method for coated plates, belonging to the field of bipolar plate testing technology. The testing method includes the following steps: heating the bipolar plate sample with pulsed current and then stretching the bipolar plate sample, while simultaneously acquiring images of temperature, displacement, load, and surface deformation of the bipolar plate sample during stretching, and performing ultrasonic testing. A basic forming limit curve is established based on the plastic instability criterion and dynamically corrected. This invention captures the initiation and expansion of interface delamination in real time through multi-physics field detection, and can synchronously track the dynamic evolution of necking initiation, expansion, and interface failure during stretching. It can more closely meet the actual working conditions and avoid Joule heating exacerbating interface thermal mismatch. Furthermore, multi-physics field detection can compensate for the insufficient sensitivity of general ultrasonic algorithms, redefining the failure criterion as the critical state of coating-substrate interface separation, further reducing the error in predicting the critical failure point of the forming limit curve.

Description

Technical Field

[0001] This invention belongs to the field of bipolar plate testing technology, specifically relating to an electrically assisted biaxial tensile forming limit testing system and method for coated plates. Background Technology

[0002] In the production of ultrathin titanium alloy bipolar plates, an electroplating nickel coating process is required after stamping to improve the corrosion resistance and conductivity of the bipolar plates. However, due to the poor depth and uniformity of electroplating, incomplete plating of the micro-channel structure of the bipolar plate is prone to occur, failing to guarantee the required service life. To effectively solve this problem, the plating process is often advanced by electroplating nickel onto the surface of the ultrathin titanium substrate before stamping, avoiding incomplete plating of the micro-channel structure and allowing the stamped bipolar plates to be used directly. However, to achieve integrated manufacturing of ultrathin nickel-plated titanium alloy bipolar plates, forming limit tests are needed to construct forming limit curves on the pre-coated plates. These curves are then used to evaluate the forming performance of the ultrathin nickel-plated titanium plates, ensuring that no interface delamination defects occur between the substrate and the coating after stamping.

[0003] When conducting forming limit tests on pre-coated sheets, the bipolar plate is typically stretched using mechanical property testing, followed by metallographic analysis or scanning electron microscopy to observe the failure morphology of the substrate and coating, and a forming limit curve is established based on the ductile fracture criterion. However, the aforementioned testing method, which combines mechanical property testing with metallographic analysis or scanning electron microscopy, is based on observing the defect of interfacial delamination between the substrate and the coating. This method not only has limitations such as destructiveness and hysteresis, but also makes it difficult to achieve real-time capture and early warning of failure mechanisms during dynamic loading, and cannot correlate interfacial delamination failure with critical strain in real time. Meanwhile, the forming limit curve established based on the ductile fracture criterion uses tensile fracture of the material as the failure criterion. Its core assumption is the homogeneity and integrity of the material. However, delamination failure at the interface between the coating and the substrate of the bipolar plate often occurs earlier than the fracture of the substrate itself. Once the interface between the coating and the substrate separates, even if the substrate does not fracture, the key functions of the bipolar plate, such as conductivity and corrosion resistance, are completely lost, rendering the bipolar plate unusable. This gives the bipolar plate a unique failure mechanism, causing the established forming limit curve to fail to capture the critical point of interface delamination failure. The output forming limit prediction value is artificially high. Furthermore, the process parameters (such as current density and strain path) obtained from the forming limit curve cannot effectively suppress the expansion of interface delamination. Often, in order to avoid bipolar plate failure, forming process parameters are excessively reduced, sacrificing production efficiency. Therefore, the forming limit curve established based on the ductile fracture criterion is not suitable for evaluating the forming performance of ultrathin nickel-titanium plated sheets.

[0004] Currently, ultrasonic testing is widely used to identify delamination defects within bipolar plates, a technique that avoids damaging the plates. However, ultrasonic testing only provides a static characterization of internal defects and cannot simultaneously track the dynamic evolution of necking initiation, propagation, and interface failure during stretching. It also struggles to quantify the dynamic impact of delamination propagation on forming limits (such as local necking thresholds). Furthermore, during electrically assisted stretching, the current-induced Joule heating significantly alters the local temperature field, accelerating strain localization and interface delamination in the necking region. Existing forming limit testing methods often focus on a single physical field, making it difficult to reproduce failure mechanisms under actual working conditions. Moreover, the significant difference in interfacial bonding strength between the pre-coating and the substrate, coupled with insufficient sensitivity in general ultrasonic algorithms, makes it difficult to predict critical failure points using the obtained forming limit curves. Summary of the Invention

[0005] In view of this, the present invention provides an electrically assisted biaxial tensile forming limit testing system and method for coated sheet materials, in order to overcome the shortcomings of the prior art. The present invention can simultaneously track the dynamic evolution of necking initiation, propagation and interface failure during the stretching process, quantify the dynamic influence of delamination propagation on the forming limit, redefine the failure criterion as the critical state of coating-substrate interface separation, and further reduce the error of the forming limit curve in predicting the critical failure point.

[0006] The technical solution of this invention is: a method for testing the limit of biaxial tensile forming of coated sheets by electric assistance, comprising the following steps:

[0007] The bipolar plate sample is heated to a preset temperature using pulsed current, and then stretched. Simultaneously, images of temperature, displacement, load, and surface deformation of the bipolar plate sample during stretching are acquired, and ultrasonic testing is performed.

[0008] The acquired temperature, displacement, load, surface deformation images and ultrasonic signals are processed to obtain temperature gradient curves, load-displacement curves, reflected wave attenuation rate, strain cloud map, harmonic distortion degree, layer area ratio and ultrasonic signal energy accumulation curve.

[0009] Temperature gradient curves, load-displacement curves, and strain contour maps were used to locate the delamination initiation regions on the bipolar plate specimens.

[0010] The crack propagation rate in the layered initiation region is obtained by considering the attenuation rate of reflected waves, harmonic distortion, and energy accumulation curve of ultrasonic signals in the layered initiation region.

[0011] A basic forming limit curve is established based on the plastic instability criterion. The basic forming limit curve is dynamically corrected by crack propagation rate and delamination area ratio to obtain the forming limit curve FLC. The forming limit curve FLC is then used to evaluate the performance of bipolar plate stamping.

[0012] Preferably, the steps for obtaining the reflected wave attenuation rate, harmonic distortion, and layered area ratio include:

[0013] The acquired ultrasonic signals are preprocessed by wavelet denoising and EMD decomposition to remove noise and separate the interface reflected wave signal and transmitted wave signal, thereby obtaining the reflected wave attenuation rate and harmonic distortion.

[0014] Threshold segmentation is performed on the interface reflected wave signal to obtain the boundary of the layered region;

[0015] Calculate the proportion of pixels in the layered region to the total number of pixels to obtain the proportion of the area of ​​the layered region to the total area of ​​the bipolar plate sample.

[0016] Preferably, the location of the delamination initiation region on the bipolar plate specimen is achieved using temperature gradient curves, load-displacement curves, and strain contour maps, including:

[0017] Align the temperature gradient curve with the load-displacement curve. When the abrupt change point on the temperature gradient curve coincides with the abrupt change point on the load-displacement curve, compare the abrupt change point on the temperature gradient curve with the strain localization region on the strain contour map. When the strain localization region coincides with the position of the abrupt change point on the temperature gradient curve, the region is the stratification initiation region.

[0018] Preferably, establishing the basic forming limit curve based on the plastic instability criterion includes:

[0019] Selection of plastic instability criterion:

[0020] The parameters in the plastic instability criterion are calibrated based on historical test data.

[0021] Based on the plastic instability criterion, a staged FLD prediction model is established to obtain models for uniform deformation, local instability, and critical failure stages, thus forming the basic forming limit curve.

[0022] Preferably, the plastic instability criterion is selected based on the difference in interfacial thermal expansion coefficients and bond strength of the bipolar plate specimens:

[0023] FLD on the right: adopts the modified maximum force criterion;

[0024] Left FLD: using Hill'48 yield criterion.

[0025] Preferably, the FLD prediction model is established in stages according to the plastic instability criterion, including:

[0026] Uniform deformation: The diffusion necking initiation point is predicted based on the Swift hardening criterion, and the initial warning is triggered by the slope change of the load-displacement curve;

[0027] Local instability: When the temperature and the intensity of the ultrasonic signal rise synchronously, switch to the MK model, introduce an equivalent initial defect, and simulate the strain localization caused by interface delamination.

[0028] Critical failure: Combining bifurcation theory, the formation of shear bands is determined by the abrupt change in the energy spectrum of ultrasonic signals, and the critical fracture strain is output.

[0029] Preferably, the dynamic correction of the basic forming limit curve using crack propagation rate and delamination area ratio includes:

[0030] Import the crack propagation rate and the percentage of delamination area into the following formula:

[0031] ;

[0032] Among them, FLC base Based on the forming curve, FLC new The corrected forming limit curve is FLC, where k1 is the weighting coefficient for the delamination area ratio, k2 is the weighting coefficient for the crack propagation rate, and A... d V represents the percentage of the area of ​​each layer. c V represents the crack propagation rate, expressed in mm / s. crit The critical crack propagation rate is expressed in mm / s.

[0033] Preferably, the performance evaluation of bipolar plate stamping using the forming limit curve (FLC) includes:

[0034] The forming limit curve FLC is superimposed onto the biaxial tensile strain field;

[0035] A three-dimensional thermal map of the safe-transition-hazardous zone is generated using an interpolation algorithm to identify the probability of critical fracture locations.

[0036] Safe zone: ε1 < 0.8 FLC new ;

[0037] Transition region: 0.8 FLC new <ε1 <FLC new ;

[0038] Danger zone: ε1>FLC new ;

[0039] Where ε1 is the maximum principal strain.

[0040] A biaxial tensile forming limit testing system for coated sheet metal includes a biaxial tensile testing machine. The ends of the bipolar plate specimen are fixedly connected to the clamping ends of the biaxial tensile testing machine to perform biaxial tensile testing on the bipolar plate specimen. The testing system further includes: pulse current elements connected to the ends of the bipolar plate specimen via wires to heat the bipolar plate specimen; multiple thermocouples fixed on the side of the bipolar plate specimen near the clamping ends to collect the temperature of the bipolar plate specimen; multiple ultrasonic detection elements fixed on the side of the bipolar plate specimen near the clamping ends to perform ultrasonic testing on the bipolar plate specimen; multiple displacement sensors fixed on the ends of the bipolar plate specimen to collect the displacement during biaxial tensile testing; load detection elements fixed on the loading shaft connected to the clamping ends of the biaxial tensile testing machine to collect the load applied to the bipolar plate specimen during biaxial tensile testing; and a DIC high-speed camera positioned directly above the center of the bipolar plate specimen to collect surface deformation images of the bipolar plate specimen. Based on the collected temperature, ultrasonic signals, displacement, load, and surface deformation images, the performance of bipolar plate stamping is evaluated using testing methods.

[0041] Preferably, the bipolar plate sample is cross-shaped.

[0042] Compared with existing technologies, the present invention provides an electrically assisted biaxial tensile forming limit testing system and method for coated plates. This system utilizes pulsed current to heat the bipolar plate sample before subjecting it to biaxial tensile testing. Simultaneously, it acquires images of temperature, displacement, load, and surface deformation, and performs ultrasonic testing. This yields information such as temperature gradient curves, load-displacement curves, reflected wave attenuation rate, strain contour maps, harmonic distortion, and delamination area ratio. The system can locate delamination initiation regions and obtain crack propagation rates within these regions. Through multiphysics detection, it captures the initiation and propagation of interfacial delamination in real time, and can synchronously track the dynamic evolution of necking initiation, propagation, and interfacial failure during the tensile process. The dynamic impact of quantified delamination on forming limits enables dynamic monitoring of interface failure under thermo-electric-mechanical coupling conditions. This approach more closely reflects actual working conditions, avoiding Joule heating that exacerbates interface thermal mismatch. Furthermore, multi-physics detection can compensate for the insufficient sensitivity of general ultrasonic algorithms. A basic forming limit curve is established based on the plastic instability criterion and dynamically corrected. By dynamically capturing interface delamination failure characteristics, a quantitative relationship between the curve and the bipolar plate forming limit is established. This integrates classical plastic instability theory with multi-physics dynamic detection technology, redefining the failure criterion as the critical state of coating-substrate interface separation. This further reduces the error in predicting critical failure points using the forming limit curve. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the testing system of the present invention;

[0044] Figure 2 This is a flowchart of the forming limit prediction method of the present invention.

[0045] Explanation of reference numerals in the attached figures:

[0046] 1. Clamping end; 5. Ultrasonic testing element; 6. Load testing element; 8. Temperature sensing system; 9. Bipolar plate sample; 10. Pulse current element; 11. Ultrasonic testing system; 12. Upper support; 13. Lower support; 14. Left support; 15. Right support; 16. Insulating gasket; 17. Thermocouple; 18. Insulating sleeve; 19. Displacement sensor; 20. DIC high-speed camera; 21. DIC analysis system. Detailed Implementation

[0047] This invention provides an electrically assisted biaxial tensile forming limit testing system and method for coated sheets, which is described below in conjunction with... Figures 1 to 2 The present invention is illustrated by the structural diagram shown below.

[0048] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the technical solution of this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0049] A method for testing the biaxial tensile forming limit of coated sheet metal using electrical assistance includes the following steps:

[0050] The bipolar plate sample is heated to a preset temperature using pulsed current, and then stretched. Simultaneously, images of temperature, displacement, load, and surface deformation of the bipolar plate sample during stretching are acquired, and ultrasonic testing is performed.

[0051] The acquired temperature, displacement, load, surface deformation images and ultrasonic signals are processed to obtain temperature gradient curves, load-displacement curves, reflected wave attenuation rate, strain cloud map, harmonic distortion degree, layer area ratio and ultrasonic signal energy accumulation curve.

[0052] Temperature gradient curves, load-displacement curves, and strain contour maps were used to locate the delamination initiation regions on the bipolar plate specimens.

[0053] The crack propagation rate in the layered initiation region is obtained by considering the attenuation rate of reflected waves, harmonic distortion, and energy accumulation curve of ultrasonic signals in the layered initiation region.

[0054] A basic forming limit curve is established based on the plastic instability criterion. The basic forming limit curve is dynamically corrected by crack propagation rate and delamination area ratio to obtain the forming limit curve FLC. The forming limit curve FLC is then used to evaluate the performance of bipolar plate stamping.

[0055] In this embodiment, the electrically assisted biaxial tensile forming limit test method for coated plates first heats the bipolar plate sample using pulsed current. The temperature of the bipolar plate sample is monitored to ensure it reaches a preset temperature and current density, thus meeting the tensile requirements. Then, the bipolar plate sample is subjected to bidirectional tensile testing. Images of temperature, displacement, load, and surface deformation during the tensile process are collected, along with ultrasonic testing. This allows for dynamic monitoring of multi-physics interface failure under thermo-electric-mechanical coupling conditions. Furthermore, multi-physics detection compensates for the insufficient sensitivity of general ultrasonic algorithms, enabling real-time capture of interface delamination initiation and expansion, and synchronous tracking of the tensile process. The dynamic evolution of necking initiation, propagation, and interface failure is analyzed. The dynamic impact of delamination propagation on forming limit is quantified. The above information is then processed to obtain information such as temperature gradient curve, load-displacement curve, reflected wave attenuation rate, strain cloud map, harmonic distortion degree, and delamination area ratio. This allows for the location of delamination initiation areas and the acquisition of crack propagation rates in these areas. A basic forming limit curve is then established based on the plastic instability criterion and dynamically corrected. By dynamically capturing interface delamination failure characteristics, a quantitative relationship between the failure and bipolar plate forming limit is established. The classical plastic instability theory is integrated with multiphysics dynamic detection technology, and the failure criterion is redefined as the critical state of coating-substrate interface separation. This can further reduce the error in predicting the critical failure point of the forming limit curve.

[0056] The electro-assisted biaxial tensile forming limit test method for coated plates in the above embodiments is based on the pulsed current field assisted Nakajima forming limit test method. It can obtain the dynamic damage evolution law of interface integrity under the action of electroplastic coupling size effect, thereby establishing the electro-assisted stamping forming limit curve based on substrate / coating interface delamination failure, and thus accurately evaluating the stamping forming performance of ultrathin nickel-titanium plate.

[0057] Traditional detection methods are destructive and have delays, meaning that the coating has already failed by the time interface defects are observed. However, the forming limit curve obtained by the detection method in this embodiment can be used to predict interface failure before it occurs.

[0058] The electrically assisted biaxial tensile forming limit test method for coated materials in this embodiment has the following advantages:

[0059] More precise failure criteria: The forming limit assessment is changed from "material fracture" to "loss of interface function", which is more in line with the actual failure mode of pre-coated sheets;

[0060] Resource efficiency improvement: Avoid production efficiency losses caused by overly conservative process parameters and increase production capacity;

[0061] Industrial application value: It provides theoretical support for optimizing the forming process of pre-coated parts such as bipolar plates for new energy vehicles and composite laminates for aerospace.

[0062] The electrically assisted biaxial tensile forming limit test method for coated plates in the above embodiments, targeting the electrically assisted tensile scenario of bipolar plates, utilizes dynamic capture of interface delamination failure characteristics to establish a quantitative relationship between them and the forming limit of the plate (such as critical fracture strain and local necking threshold). It integrates classical plastic instability theory with multi-physics dynamic detection technology and proposes a "interface delamination damage-forming limit dynamic mapping" model to obtain the forming limit curve FLC. The forming limit curve FLC is used to evaluate the performance of bipolar plate stamping, which can provide an important basis for optimizing coating processes and forming parameters.

[0063] In the above embodiments, the attenuation rate of reflected waves, harmonic distortion degree, and energy accumulation curve of ultrasonic signals in the layered initiation region are input into the 1D-CNN + LSTM model to output the crack propagation rate in mm / s, so as to predict the crack propagation rate. At the same time, the direction of crack propagation can also be obtained. By utilizing the dynamic change law of characteristic parameters, combined with physical models or data-driven algorithms, quantitative assessment and early warning of crack propagation can be achieved.

[0064] For example, based on the energy accumulation curve of ultrasonic signals, when the energy value increases from 100 mV²*s to 500 mV²*s within 1 second, the crack propagation rate V is determined. c =0.1mm / s.

[0065] Crack propagation rate threshold determination: If the reflected wave attenuation rate exceeds the preset threshold (e.g., ≥20 dB / μs), delamination is determined to have started; if the harmonic distortion continues to rise (e.g., THD increases from 3% to 8%), it indicates that the crack is propagating at an accelerated rate.

[0066] The reflected wave attenuation rate in the above embodiments can reflect the energy loss of ultrasound at the delamination interface and is directly related to the geometric dimensions of the delamination (such as crack length and depth). The harmonic distortion degree (THD) can characterize the degree of nonlinear distortion of the signal and is related to the stress field change at the dynamically expanding crack tip.

[0067] As a further optimization, the steps for obtaining the reflected wave attenuation rate, harmonic distortion, and layered area ratio in this embodiment include:

[0068] The acquired ultrasonic signals are preprocessed by wavelet denoising and EMD decomposition to remove noise and separate the interface reflected wave signal and transmitted wave signal, thereby obtaining the reflected wave attenuation rate and harmonic distortion.

[0069] Threshold segmentation is performed on the interface reflected wave signal to obtain the boundary of the layered region;

[0070] Calculate the proportion of pixels in the layered region to the total number of pixels to obtain the proportion of the area of ​​the layered region to the total area of ​​the bipolar plate sample.

[0071] In this embodiment, wavelet denoising and EMD decomposition are used to preprocess the acquired ultrasonic signal, which can separate physical quantities directly related to the layering (such as attenuation rate and THD) from the complex acoustic signal, thus solving the signal-to-noise ratio problem in dynamic environments.

[0072] The specific implementation of obtaining the percentage of area in each layer is as follows:

[0073] Ultrasound image segmentation: Threshold segmentation of reflected wave signals to identify the boundaries of layered regions;

[0074] Pixel statistics: Calculate the proportion of pixels in the layered regions to the total number of pixels;

[0075] Real-time output: The multi-channel signal acquisition system dynamically updates the value of the layered area ratio. That is, the ultrasonic signal processing algorithm calculates the proportion of the layered area to the total detection area in real time, and when the attenuation rate of the reflected wave exceeds 20 dB / μs, the area is determined to be a layered damage area.

[0076] As a further optimization, this embodiment utilizes temperature gradient curves, load-displacement curves, and strain contour maps to locate the delamination initiation region on the bipolar plate sample, including:

[0077] Align the temperature gradient curve with the load-displacement curve. When the abrupt change point on the temperature gradient curve coincides with the abrupt change point on the load-displacement curve, compare the abrupt change point on the temperature gradient curve with the strain localization region on the strain contour map. When the strain localization region coincides with the position of the abrupt change point on the temperature gradient curve, the region is the stratification initiation region.

[0078] During electrically assisted stretching, the initiation of interfacial delamination can lead to abrupt changes in local thermodynamic behavior, specifically manifested as: frictional heating, where friction occurs in the delamination area due to interfacial misalignment or microcrack propagation, resulting in a local temperature increase; strain localization, where delamination is accompanied by strain concentration in the necking region, exacerbating the Joule heating effect and creating abrupt temperature gradient changes; and impeded heat conduction, where delamination interrupts the material's heat conduction path, causing heat to accumulate near the delamination interface and forming temperature anomalies.

[0079] Therefore, this embodiment utilizes multiphysics to detect spatiotemporal synchronization:

[0080] Time synchronization: Align the timestamps of abrupt temperature gradient changes with the inflection points of the load-displacement curve to eliminate temperature fluctuations caused by non-stratification (such as environmental interference).

[0081] Spatial mapping: Real-time strain field is acquired through the DIC high-speed camera system. If the temperature abrupt change point coincides with the strain localization region, it is confirmed as a stratification initiation region.

[0082] Among them, the strain localization region is the region in the strain cloud map where the strain is greater than or equal to the threshold.

[0083] Inflection point identification of load-displacement curve characteristics: When delamination occurs, the load-displacement curve shows a decrease in slope or fluctuation, indicating a decrease in material stiffness. Temperature data should be analyzed simultaneously at this time point.

[0084] Energy dissipation: The stratification process is accompanied by energy release. The energy abrupt change point is calculated by load integration and matched with the temperature gradient change time.

[0085] Location of abrupt change points in the temperature gradient curve: The slope of the temperature gradient (ST=∆T / ∆x) is monitored in real time by a thermocouple array. When ST suddenly increases in a certain region (e.g., from 2℃ / mm to 8℃ / mm), it is marked as a potential stratification initiation point.

[0086] Alternatively, thermal imaging can be used as an aid, such as an infrared thermal imager, to supplement thermocouple data and capture the spatial non-uniformity of the temperature field.

[0087] As a further optimization, the basic forming limit curve based on the plastic instability criterion is established in this embodiment as follows:

[0088] Selection of plastic instability criterion:

[0089] The parameters in the plastic instability criterion are calibrated based on historical test data.

[0090] Based on the plastic instability criterion, a staged FLD prediction model is established to obtain models for uniform deformation, local instability, and critical failure stages, thus forming the basic forming limit curve.

[0091] As a further optimization, this embodiment selects the plastic instability criterion based on the difference in the interfacial thermal expansion coefficients and the bond strength of the bipolar plate samples:

[0092] FLD on the right: adopts the modified maximum force criterion;

[0093] Left FLD: using Hill'48 yield criterion.

[0094] In this embodiment, the Modified Maximum Force Criterion measures the influence of current-induced thermal stress on the necking threshold by extracting harmonic distortion from ultrasonic signals. It uses the Hill'48 yield criterion combined with the temperature gradient abrupt change point to locate the interface layered initiation region and correct the local strain path.

[0095] In this embodiment, calibrator parameters are established by using historical test data, such as the strain path sensitivity coefficient in the maximum force criterion and the anisotropy coefficient in the Hill'48 yield criterion, thereby establishing a mapping relationship with ultrasonic characteristics (reflection wave attenuation rate, delamination area ratio).

[0096] As a further optimization, this embodiment establishes an FLD prediction model in stages based on the plastic instability criterion, including:

[0097] Uniform deformation: The diffusion necking initiation point is predicted based on the Swift hardening criterion, and the initial warning is triggered by the slope change of the load-displacement curve (dF / dε=0);

[0098] Local instability: When the temperature and the intensity of the ultrasonic signal rise synchronously, switch to the MK model and introduce an equivalent initial defect (defect factor f0=0.99) to simulate the strain localization caused by interface delamination;

[0099] Critical failure: Combining bifurcation theory (Loss of Ellipticity criterion), the formation of shear bands is determined by the abrupt change in the energy spectrum of ultrasonic signals, and the critical fracture strain is output.

[0100] In this embodiment, a FLD prediction model is established in stages according to the plastic instability criterion. The proportion of stratified area and the slope of temperature gradient can be input into the Random Forest model to output the probability of stratification, necking and microcracks, thereby realizing static damage classification.

[0101] In this embodiment, the temperature gradient slope is calculated by real-time acquisition of temperature data and outputting the temperature distribution at various locations on the bipolar plate sample. For example, thermocouples with a spacing of 2 mm can be placed in the necking region of the bipolar plate sample. When the temperature difference between adjacent thermocouples reaches 5℃, the gradient slope ST = 2.5℃ / mm is calculated. This abrupt change point is also the initiation point of interface delamination.

[0102] As a further optimization, this embodiment utilizes crack propagation rate and delamination area ratio to dynamically correct the basic forming limit curve, including:

[0103] Import the crack propagation rate and the percentage of delamination area into the following formula:

[0104] ;

[0105] Among them, FLC base The basic forming curve, i.e., the theoretical curve without considering interfacial delamination damage, is the FLC. new The corrected forming limit curve FLC is dynamically adjusted by introducing interface delamination damage parameters, which better reflect actual working conditions. k1 is the weighting coefficient for the delamination area ratio, k2 is the weighting coefficient for the crack propagation rate, and A d V represents the percentage of the area of ​​each layer. c V represents the crack propagation rate, expressed in mm / s. crit Critical crack propagation rate, representing the crack propagation rate at which interface delamination reaches the failure threshold, is calibrated through experiments or historical data, and is measured in mm / s.

[0106] In this embodiment, the basic forming limit curve is dynamically corrected, and the crack propagation rate V is adjusted. c , Area ratio of each layer A d And the second derivative of the load curve is imported into the XGBoost model to obtain FLC. new A safety threshold is established to achieve accurate threshold regression.

[0107] Additionally, k1 (weight of delamination area ratio): This value is determined by fitting historical experimental data, such as comparing the deviation between measured FLD and theoretical FLD under different degrees of interfacial delamination, and using regression analysis. k1 can reflect the influence of static damage (such as delamination area) on the forming limit. A larger weight indicates that the delamination area of ​​the interface has a more significant weakening effect on FLD.

[0108] k2 (crack propagation rate weight): This is determined through fatigue tests or dynamic tensile tests, based on the ratio of crack propagation rate to the critical value. For example, when monitoring the sensitivity of crack propagation to FLD (Fluid Degradation), the physical meaning of k2 reflects the influence of dynamic damage (such as crack propagation rate). A higher weight indicates that rapid crack propagation will significantly reduce the critical strain threshold.

[0109] In this embodiment, damage weights k1 and k2 are trained using historical test data of the pre-coated coating to adapt to the interface thermal-mechanical coupling characteristics.

[0110] As a further optimization, this embodiment uses the forming limit curve (FLC) to evaluate the performance of bipolar plate stamping, including:

[0111] The forming limit curve FLC is superimposed onto the biaxial tensile strain field (plane).

[0112] A three-dimensional thermal map of the safe-transition-hazardous zone is generated using an interpolation algorithm to identify the probability of critical fracture locations.

[0113] Safe zone: ε1 < 0.8 FLCnew ;

[0114] Transition region: 0.8 FLC new <ε1 <FLC new ;

[0115] Danger zone: ε1>FLC new ;

[0116] Where ε1 is the maximum principal strain.

[0117] In this embodiment, the process parameters can be fed back in real time through the above-mentioned three-dimensional thermal map:

[0118] When the proportion of dangerous areas exceeds the threshold, a tiered warning is triggered:

[0119] Level 1 warning: Reduce the electrical pulse intensity to a preset lower limit (e.g., current density J=50 A / mm²). 2 ), to slow down the accumulation of Joule heat;

[0120] Level 2 warning: Switch the strain path to plane strain mode (β=0) to suppress interface delamination propagation;

[0121] The optimized process parameters are automatically stored in the database for use as initial parameter settings in subsequent experiments.

[0122] Note: Triggering conditions and thresholds:

[0123] Level 1 Warning:

[0124] Trigger flag: Triggered when the detected dangerous area accounts for 15% of the threshold;

[0125] Response measures: Reduce the intensity of the pulse current to a preset safe value (e.g., reduce the current density from 100 A / mm² to 50 A / mm²) to delay damage propagation by reducing Joule heat accumulation;

[0126] Intervention objective: To slow down damage accumulation and buy time for process adjustments.

[0127] Level 2 warning:

[0128] Triggering criteria: If the proportion of dangerous areas does not decrease or continues to rise to 25% of the threshold after a Level 1 warning, or if abnormal dynamic parameters are detected (such as crack propagation rate exceeding the critical value Vcrit, or temperature gradient abrupt change rate ≥10℃ / s), it will be triggered.

[0129] Response measures: Switch the strain path to plane strain mode to suppress strain localization and interface delamination.

[0130] Intervention objective: To forcibly terminate the high-risk state and prevent failure.

[0131] This embodiment enables closed-loop feedback of process parameters, automatically adjusting parameters such as current intensity and strain path based on the real-time detected interface damage threshold, thereby improving the forming performance of the sheet metal.

[0132] The electrically assisted biaxial tensile forming limit test method for coated substrates in this invention breaks through the traditional fracture failure assumption of FLC and proposes a "interface delamination failure-forming limit" correlation model. The failure criterion is redefined as the critical state of coating-substrate interface separation, including:

[0133] 1. Failure Criterion Reconstruction:

[0134] Real-time capture of interface layering initiation and expansion using multi-physics field detection (ultrasonic reflection wave attenuation rate, abrupt change points on temperature gradient curves);

[0135] Establish interface layer area ratio (A) d ≥15%) and crack propagation rate (V c ≥V crit The dual threshold failure criterion.

[0136] 2. Dynamic FLC Correction:

[0137] FLC curves are dynamically corrected based on interface damage parameters.

[0138] 3. Specialization optimization:

[0139] The selected plastic instability criterion is based on the interfacial characteristics of the pre-coating layer and substrate (such as the difference in thermal expansion coefficient and bonding strength), and optimizes the ultrasonic detection frequency (such as 20MHz) and the feature selection of machine learning algorithms to improve sensitivity.

[0140] Reference Figure 1 As shown, Figure 1This is a schematic diagram of the testing system in this embodiment. An electrically assisted biaxial tensile forming limit testing system for coated sheet metal includes a biaxial tensile testing machine. The ends of the bipolar plate specimen 9 are fixedly connected to the clamping ends 1 of the biaxial tensile testing machine in a one-to-one correspondence to perform biaxial tensile testing on the bipolar plate specimen 9. The testing system further includes: pulse current elements 10 connected to the ends of the bipolar plate specimen 9 via wires to heat the bipolar plate specimen 9; multiple thermocouples 17 are fixedly mounted on the side of the bipolar plate specimen 9 near the clamping end 1 to collect the temperature of the bipolar plate specimen 9; and multiple ultrasonic detection elements 5 are fixedly mounted on the side of the bipolar plate specimen 9 near the clamping end 1. On one side of the holding end 1, ultrasonic testing is performed on the bipolar plate sample 9. Multiple displacement sensors 19 are respectively fixed at the ends of the bipolar plate sample 9 to collect the displacement during biaxial tension. Load detection elements 6 are respectively fixed on the loading shaft connected to the clamping end 1 of the biaxial tensile testing machine to collect the load applied to the bipolar plate sample 9 during biaxial tension. The DIC high-speed camera 20 is set directly above the middle of the bipolar plate sample 9 to collect the surface deformation image of the bipolar plate sample 9. Based on the collected temperature, ultrasonic signal, displacement, load and surface deformation image, the performance of bipolar plate stamping is evaluated using the test method.

[0141] Reference Figure 2 As shown, Figure 2 This is a flowchart of the forming limit prediction in this embodiment. The specific operation process of the electrically assisted biaxial tensile forming limit testing system for coated sheet in this embodiment is as follows:

[0142] Step 1: Install the upper support 12, lower support 13, left support 14, and right support 15 onto the bitensile testing machine;

[0143] Step 2: In order to prevent the pulse current from damaging equipment such as the bitensile testing machine, insulating pads 16 are installed on the clamping ends respectively. Then, based on the fastening screws and insulating sleeves 18, the clamping ends 1 are fixedly connected to the upper support 12, lower support 13, left support 14 and right support 15.

[0144] Step 3: Prepare bipolar plate sample 9 and use a digital image correlation speckle strain measurement system to perform real-time strain measurement under multi-energy field assisted bi-tension orthogonal loading.

[0145] Step 4: Install ultrasonic testing element 5, displacement sensor 19 and thermocouple 17 at the predetermined position of bipolar plate sample 9. Ultrasonic testing element 5 is connected to ultrasonic testing system 11. Displacement sensor 19 and load detection element are connected to load-displacement signal detection system. Thermocouple 17 is connected to temperature sensing system 8.

[0146] Step 5: Before the test begins, use the pulse current element 10 to apply a pulse current so that the bipolar plate sample 9 reaches the preset temperature and current density.

[0147] Step 6: Perform an electric-assisted biaxial tensile test. The upper support 12, lower support 13, left support 14, and right support 15 are controlled to move simultaneously by the CNC system on the biaxial tensile testing machine to perform biaxial tensile testing at a set strain rate, simulating the electro-thermal-mechanical coupling failure scenario under actual working conditions.

[0148] Step 7: The load-displacement signal detection system and temperature sensing system 8 output load-displacement curves and temperature gradient curves, and combine the abrupt change points on the temperature gradient curves to locate the stratification initiation region.

[0149] Step 8: The ultrasonic signal obtained from the ultrasonic testing system 11 is subjected to wavelet denoising and EMD decomposition to extract the feature parameters related to delamination (such as reflected wave attenuation rate and harmonic distortion degree) in the ultrasonic signal, and the interface crack propagation is predicted based on the real-time detected acoustic signal.

[0150] Step 9: Train machine learning models based on historical experimental data: Random Forest, 1D-CNN + LSTM and XGBoost. Input parameters include layer area ratio, crack propagation rate, temperature gradient curve slope, etc.

[0151] Step 10: Using the trained machine learning model, output static damage classification, crack propagation rate and FLCnew safety threshold for factors such as the percentage of layered area, slope of temperature gradient curve, attenuation rate of reflected wave, harmonic distortion degree and second derivative of load curve, thereby obtaining the forming limit curve FLC under the current loading conditions. This curve is then superimposed on the biaxial tensile strain field to generate a thermogram of the "safe-dangerous" region and mark the critical fracture region.

[0152] Step 11: When the proportion of the dangerous area exceeds the threshold, automatically reduce the electrical pulse intensity or switch the strain path to delay failure and record the optimized forming limit parameters.

[0153] Step 12: After the test, remove the four clamping ends 1 and disassemble the bipolar plate sample 9 for surface observation and microstructure observation.

[0154] The digital image correlation speckle strain measurement system in this embodiment includes a DIC high-speed camera 20 and a DIC analysis system 21 for real-time strain measurement. The DIC high-speed camera 20 is positioned directly above the center of the bipolar plate sample 9, with a frame rate ≥ 500 fps and a resolution ≥ 2K. A black and white speckle pattern (spot size 0.1-0.3 mm) is sprayed onto the surface of the bipolar plate sample 9.

[0155] Specifically, a high-contrast speckle pattern is sprayed onto the surface of the bipolar plate sample 9, using a matte black base coat and white spots (0.2 mm in diameter) to ensure speckle coverage > 60%. During the biaxial tensile test, the DIC high-speed camera 20 acquires surface deformation images at 500 fps and transmits them to the DIC analysis system 21 in real time.

[0156] The DIC Analysis System 21 incorporates a digital image correlation algorithm to calculate the full-field strain distribution (such as the Green-Lagrange strain tensor) in real time and output strain contour maps and coordinates of high-risk areas.

[0157] In this embodiment, the DIC analysis system 21 is used to achieve sub-pixel-level analysis of the strain field on the sample surface (accuracy ±0.05%). Combined with thermocouple array and ultrasonic detection, the interface delamination positioning error is reduced from ±1 mm in the traditional method to ±0.2 mm.

[0158] In this embodiment, the displacement sensor 19 can be a laser displacement meter or an extensometer, and the ultrasonic detection element 5 uses a solid coupling agent to achieve a stable connection with the bipolar plate sample 9. The ultrasonic detection element 5, the load detection element 6, the displacement sensor 19, and the thermocouple 17 acquire dynamic data at a sampling rate of 1 kHz.

[0159] As a further optimization, the bipolar plate sample 9 in this embodiment is cross-shaped.

[0160] In this embodiment, the bipolar plate specimen 9 is cross-shaped and uses a different wire cutting method, which can realize multiaxial stress state simulation. The four-arm structure of the cross-shaped specimen allows for simultaneous application of biaxial tensile loads on the X / Y axes, reproducing the complex stress state in actual stamping (such as the combined loading of plane strain and biaxial tension during bipolar plate stamping). Controllable induction of interface delamination failure: The central region of the cross-shaped specimen is designed as a necking sensitive area. Through geometric transition and Joule heating effect coupling of pulse current, interface delamination is artificially induced to initiate in a predetermined area, which is convenient for the accurate monitoring of ultrasonic detection elements and thermocouples. In contrast, the stress distribution of traditional specimens (such as rectangular specimens) is dispersed, making it difficult to locate the initiation point of interface failure.

[0161] Specifically, when the bipolar plate sample 9 is cross-shaped, multiple thermocouples 17 are arranged in the necking sensitive areas of the four extension sections of the bipolar plate sample 9. The multiple thermocouples 17 in each necking sensitive area are arranged at equal intervals along the length of the extension section. The thermocouples 17 can be fixed to the bipolar plate sample 9 by electric welding.

[0162] The above-disclosed embodiments are merely preferred embodiments of the present invention. However, the embodiments of the present invention are not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for testing the biaxial tensile forming limit of coated sheet metal by electric assistance, characterized in that, Includes the following steps: The bipolar plate sample is heated to a preset temperature using pulsed current, and then stretched. Simultaneously, images of temperature, displacement, load, and surface deformation of the bipolar plate sample during stretching are acquired, and ultrasonic testing is performed. The acquired temperature, displacement, load, surface deformation images and ultrasonic signals are processed to obtain temperature gradient curves, load-displacement curves, reflected wave attenuation rate, strain cloud map, harmonic distortion degree, layer area ratio and ultrasonic signal energy accumulation curve. Temperature gradient curves, load-displacement curves, and strain contour maps were used to locate the delamination initiation regions on the bipolar plate specimens. The crack propagation rate in the layered initiation region is obtained by considering the attenuation rate of reflected waves, harmonic distortion, and energy accumulation curve of ultrasonic signals in the layered initiation region. A basic forming limit curve is established based on the plastic instability criterion. The basic forming limit curve is dynamically corrected by crack propagation rate and delamination area ratio to obtain the forming limit curve FLC. The forming limit curve FLC is then used to evaluate the performance of bipolar plate stamping. The method of locating the delamination initiation region on the bipolar plate sample using temperature gradient curves, load-displacement curves, and strain contour maps includes: Align the temperature gradient curve with the load-displacement curve. When the abrupt change point on the temperature gradient curve coincides with the abrupt change point on the load-displacement curve, compare the abrupt change point on the temperature gradient curve with the strain localization region on the strain contour map. When the strain localization region coincides with the position of the abrupt change point on the temperature gradient curve, the region is the delamination initiation region. The establishment of the basic forming limit curve based on the plastic instability criterion includes: Selection of plastic instability criteria: The parameters in the plastic instability criterion are calibrated based on historical test data. Based on the plastic instability criterion, a staged FLD prediction model is established to obtain models of uniform deformation, local instability and critical failure stages, so as to form the basic forming limit curve. The dynamic correction of the basic forming limit curve using crack propagation rate and delamination area ratio includes: Import the crack propagation rate and the percentage of delamination area into the following formula: Among them, FLC base Based on the forming curve, FLC new The corrected forming limit curve is FLC, where k1 is the weighting coefficient for the delamination area ratio, k2 is the weighting coefficient for the crack propagation rate, and A... d V represents the percentage of the area of ​​each layer. c V represents the crack propagation rate, expressed in mm / s. crit The critical crack propagation rate is expressed in mm / s.

2. The method for testing the limit of electrically assisted biaxial tensile forming of coated sheets according to claim 1, characterized in that, The steps to obtain the reflected wave attenuation rate, harmonic distortion, and layered area ratio include: The acquired ultrasonic signals are preprocessed by wavelet denoising and EMD decomposition to remove noise and separate the interface reflected wave signal and transmitted wave signal, thereby obtaining the reflected wave attenuation rate and harmonic distortion. Threshold segmentation is performed on the interface reflected wave signal to obtain the boundary of the layered region; Calculate the proportion of pixels in the layered region to the total number of pixels to obtain the proportion of the area of ​​the layered region to the total area of ​​the bipolar plate sample.

3. The method for testing the limit of electrically assisted biaxial tensile forming of coated sheets according to claim 1, characterized in that, The plastic instability criterion is selected based on the difference in interfacial thermal expansion coefficients and bond strength of the bipolar plate specimens: Right-side FLD: Adopts modified maximum force criterion; Left FLD: using Hill'48 yield criterion.

4. The method for testing the limit of electrically assisted biaxial tensile forming of coated sheets according to claim 1, characterized in that, The step-by-step establishment of the FLD prediction model based on the plastic instability criterion includes: Uniform deformation: The diffusion necking initiation point is predicted based on the Swift hardening criterion, and the initial warning is triggered by the slope change of the load-displacement curve; Local instability: When the temperature and the intensity of the ultrasonic signal rise synchronously, switch to the MK model, introduce an equivalent initial defect, and simulate the strain localization caused by interface delamination. Critical failure: Combining bifurcation theory, the formation of shear bands is determined by the abrupt change in the energy spectrum of ultrasonic signals, and the critical fracture strain is output.

5. The method for testing the limit of electrically assisted biaxial tensile forming of coated sheets according to claim 1, characterized in that, The evaluation of the bipolar plate stamping performance using the forming limit curve FLC includes: The forming limit curve FLC is superimposed onto the biaxial tensile strain field; A three-dimensional thermal map of the safe-transition-hazardous zone is generated using an interpolation algorithm to identify the probability of critical fracture locations. Safe zone: ε1 < 0.8 FLC new ; Transition region: 0.8 FLC new <ε1 <FLC new ; Danger zone: ε1 > FLC new ; Where ε1 is the maximum principal strain.

6. An electrically assisted biaxial tensile forming limit testing system for coated sheet materials, comprising: A bipolar tensile testing machine, wherein the ends of the bipolar plate specimens (9) are fixedly connected one-to-one with the clamping ends (1) of the bipolar tensile testing machine to perform bidirectional tensile testing on the bipolar plate specimens (9), characterized in that the testing system further includes: A pulse current element (10) is connected to the ends of the bipolar plate sample (9) via wires to heat the bipolar plate sample (9); Multiple thermocouples (17) are respectively fixed on one side of the bipolar plate sample (9) near the clamping end (1) to collect the temperature of the bipolar plate sample (9); Multiple ultrasonic testing elements (5) are respectively fixed on one side of the bipolar plate sample (9) near the clamping end (1) to perform ultrasonic testing on the bipolar plate sample (9); Multiple displacement sensors (19) are respectively fixed at the ends of the bipolar plate sample (9) to collect displacement during biaxial tension. Load detection elements (6) are respectively fixed on the loading shaft connecting the bipolar test machine and the clamping end (1) to collect the load applied to the bipolar plate specimen (9) during bidirectional tensile testing; A DIC high-speed camera (20) is positioned directly above the center of the bipolar plate sample (9) to acquire surface deformation images of the bipolar plate sample (9). The performance of bipolar plate stamping is evaluated using the test method described in any one of claims 1-5, based on the collected temperature, ultrasonic signals, displacement, load, and surface deformation images.

7. The electrically assisted biaxial tensile forming limit testing system for coated sheets according to claim 6, characterized in that, The bipolar plate sample (9) is cross-shaped.

Images