A method and system for detecting fatigue damage of precision hardware

Abstract

The application discloses a precision hardware fatigue damage detection method and system, and relates to the field of precision mechanical detection.The method comprises the following steps: coupling a flexible impedance sensing array with a precision hardware to be detected; the array substrate is made of polydimethylsiloxane, and the surface array groove is filled with a conductive hydrophilic composite gel formed by in-situ polymerization of acrylamide monomer, sodium alginate and nanometer carbon fibers, thereby forming a plurality of miniaturized impedance units; a constant pressure is applied through a pressure feedback control system, so that the gel produces elastic deformation to cover and compensate the micron-level roughness of the surface of the detection area; a micro-amplitude sinusoidal voltage excitation is applied, current response data in the swept frequency band is collected, complex impedance information is extracted and fitted to a randall equivalent circuit model, and charge transfer resistance and interface capacitance values are obtained; and the parameter deviation degree is calculated according to the charge transfer resistance of the current detection period and historical reference, so as to determine the fatigue damage condition of the local area.

Description

Technical Field

[0001] This application belongs to the field of precision machinery testing, specifically relating to a method and system for detecting fatigue damage in precision hardware. Background Technology

[0002] With the continuous advancement of industrial manufacturing technology, precision hardware components such as high-performance springs and aerospace microstructures are increasingly widely used in aerospace and precision machinery fields, and their service safety is directly related to the operational reliability of the entire system. Fatigue damage, as the main failure mode of materials under alternating loads, requires accurate non-destructive testing and life assessment, which is crucial for preventing major safety accidents. Establishing a comprehensive material fatigue monitoring system has significant technical value for improving the maintenance efficiency of key components and ensuring the long-term stability of complex engineering systems.

[0003] Fatigue damage detection for precision hardware components with complex, irregular curved surfaces is a crucial research direction in the field of materials reliability assessment. This specific technology aims to establish a correlation model between damage level and physical signals by capturing changes in the microscopic characteristics of the material's surface and subsurface layers, thereby achieving effective monitoring of the damage state of minute components. To ensure the validity of the detection results, the detection system typically needs extremely high physical fit and sensitivity to microstructural evolution to output quantitative assessment data in the early stages of fatigue evolution.

[0004] However, traditional non-destructive testing methods such as ultrasonic or X-ray testing are ill-suited to the complex and irregular structures of precision metal parts. Poor physical fit results in limited inspection coverage and a high likelihood of blind spots. Furthermore, existing surface vision inspection technologies can only identify macroscopically visible geometric cracks, lacking effective quantitative characterization capabilities for lattice distortions and submicron-level damage that appear in the early stages of fatigue. In addition, minute surface roughness fluctuations and geometric variability in precision metal parts create severe background interference on electromagnetic detection signals, making it difficult to balance detection sensitivity and diagnostic accuracy, thus failing to meet the demands of high-precision evaluation. Summary of the Invention

[0005] The purpose of this invention is to provide a precision hardware fatigue damage detection method and system, which can effectively solve the problems in the background art.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: Firstly, a method for detecting fatigue damage in precision hardware includes: A flexible impedance sensing array consisting of a polydimethylsiloxane substrate and multiple array grooves formed on its surface is coupled to a precision hardware component under test. Each array groove is filled with a conductive hydrophilic composite gel formed by in-situ polymerization of acrylamide monomer, sodium alginate and carbon nanofibers to form multiple miniaturized impedance units. A constant pressure is applied to the precision hardware under test through a pressure feedback control system, maintaining a pressure range within a preset pressure range. This causes the conductive hydrophilic composite gel to undergo elastic deformation to cover the test area. The deformation compensates for the micron-level roughness of the test area surface, eliminates air gaps, and reduces contact resistance. A small-amplitude sinusoidal voltage signal is applied between the precision hardware component under test and each of the miniaturized impedance units. The current response data of each miniaturized impedance unit in the frequency sweep band is collected simultaneously. The complex impedance information is extracted and fitted into the Landers equivalent circuit model to obtain the charge transfer resistance and interface capacitance values ​​corresponding to each miniaturized impedance unit. Based on the charge transfer resistance obtained by fitting the current detection cycle of the same miniaturized impedance unit and the historical reference resistance calibrated in the intact state, the parameter deviation is calculated, and based on the comparison result of the parameter deviation and the preset ratio threshold, the fatigue damage status of the precision hardware under test in the corresponding local area is determined.

[0007] Preferably, the step of applying a constant pressure within a preset pressure range to the precision hardware component under test via a pressure feedback control system includes: The deformation of the conductive hydrophilic composite gel is made to reach a preset percentage range of its original height, and the deformation of the conductive hydrophilic composite gel compensates for the micron-level roughness of the surface of the precision hardware part to be tested.

[0008] Preferably, the step of obtaining the charge transfer resistance and interface capacitance values ​​corresponding to each of the miniaturized impedance units includes: Nonlinear least squares fitting is performed on the complex impedance information to extract the characteristic parameters of the charge transfer resistor and the constant phase angle element, and the interface capacitance value is calculated from the characteristic parameters of the constant phase angle element; wherein, the change in the interface capacitance value is positively correlated with the fatigue microcrack propagation depth, and is used to help determine the degree of microcrack propagation.

[0009] Preferred options also include: Nyquist plots are drawn based on the complex impedance information of each miniaturized impedance unit. The diameter of the semicircle in the high-frequency band, the solution resistance value in the high-frequency band, and the phase parameters of the constant phase angle element in the low-frequency band are extracted. By comparing the changes in the diameter of the semicircle, the solution resistance value, and the phase parameters in the current detection cycle with those in the reference state, the fatigue risk determined based on the charge transfer resistance is cross-validated in multiple dimensions.

[0010] Preferably, the array grooves of the flexible impedance sensing array are formed by casting with a nanoscale precision mold prepared by electron beam exposure technology, so that the cross-sectional size of the miniaturized impedance unit and the spacing between adjacent array grooves are adapted to the detection requirements of ultra-miniature hardware parts. The conductive hydrophilic composite gel contains carbon nanofibers that are treated with an ultrasonic exfoliation process to construct a three-dimensional conductive network within a miniaturized array of grooves.

[0011] Preferably, the precision hardware component to be tested is an ultra-miniature hardware component; In the step of coupling the precision hardware component to be tested with the flexible impedance sensing array, the ultra-micro hardware component is pressed into the surface of the flexible impedance sensing array with micron-level stepping accuracy using a precision three-dimensional displacement platform, and the preset pressure range is controlled within a preset safety percentage of the deformation of the conductive hydrophilic composite gel, so that the deformation does not exceed its original height.

[0012] Preferably, the step of fitting the complex impedance information to the Landels equivalent circuit model further includes: In the Landers equivalent circuit model, an edge leakage resistor is added in parallel with the charge transfer resistor to compensate for the additional charge path generated by the concentration of the edge electric field in the ultra-miniature hardware.

[0013] Preferably, the step of calculating the parameter deviation further includes: A spatial weighting algorithm is introduced, wherein the historical reference resistance of the miniaturized impedance unit located in the central region is assigned a higher weighting coefficient than the historical reference resistance of the miniaturized impedance unit located in the edge region, in order to compensate for the difference in reference value caused by the uneven force distribution between the central region and the edge region of the ultra-miniature hardware.

[0014] Preferred options also include: The remaining fatigue life of the precision hardware under test is predicted using a long short-term memory neural network model. The long short-term memory neural network model takes the time series of the charge transfer resistance and the interface capacitance values ​​obtained in each detection cycle as input and outputs the predicted value of the remaining fatigue life.

[0015] Secondly, a precision hardware fatigue damage detection system for performing the above method includes: A flexible impedance sensing array includes a polydimethylsiloxane substrate and multiple array grooves formed on its surface. Each array groove is filled with a conductive hydrophilic composite gel formed by in-situ polymerization of acrylamide monomer, sodium alginate and carbon nanofibers to form multiple miniaturized impedance units. The pressure feedback control system is used to apply a constant pressure within a preset pressure range to the precision hardware part to be tested, so that the conductive hydrophilic composite gel undergoes elastic deformation to cover the area to be tested. The deformation compensates for the micron-level roughness of the surface of the area to be tested, eliminates air gaps and reduces contact resistance. A multi-channel signal acquisition module is used to apply a small-amplitude sinusoidal voltage signal between the precision hardware under test and each of the miniaturized impedance units, and simultaneously acquire the current response data of each of the miniaturized impedance units in the frequency sweep band. The central processing unit is used to extract complex impedance information based on the current response data and fit it into the Landers equivalent circuit model to obtain the charge transfer resistance and interface capacitance values. It also calculates the parameter deviation based on the charge transfer resistance of the same miniaturized impedance unit in the current detection cycle and the historical reference resistance to determine the fatigue damage status.

[0016] In summary, this application includes at least one of the following beneficial technical effects: 1. This application involves in-situ polymerization of a conductive hydrophilic composite gel composed of acrylamide monomer, sodium alginate, and carbon nanofibers in a flexible polydimethylsiloxane substrate array groove. A pressure feedback control system is used to induce elastic deformation of the gel to cover and compensate for the micron-level roughness of the surface of the test area. This eliminates the air gap between the traditional rigid or semi-rigid sensor and the test piece, reduces contact impedance, and ensures high-fidelity coupling and acquisition of weak electrochemical signals.

[0017] 2. This application not only extracts the charge transfer resistance and interface capacitance values ​​by fitting the swept impedance spectrum to a Landers equivalent circuit model containing constant phase angle elements, and uses parameter deviation to quantify the degree of damage; but also independently confirms the decreasing trend of charge transfer resistance through multi-dimensional cross-validation of the high-frequency semicircle diameter, solution resistance value and low-frequency phase parameters of the Nyquist plot.

[0018] 3. This application uses electron beam exposure to prepare nanoscale precision molds, miniaturizes impedance units to a suitable scale, and adds an edge leakage resistance in parallel with the charge transfer resistance in the equivalent circuit model to compensate for the additional charge path generated by the concentration of electric field at the edge of the ultra-micro parts; at the same time, a spatial weighting algorithm is introduced to assign differentiated weights to the historical reference resistance of the center and edge regions. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall technical solution architecture of the precision hardware fatigue damage detection method of the present invention; Figure 2 This is a schematic diagram of the core principle framework of the fatigue damage assessment based on the deviation of electrochemical impedance parameters in this invention. Figure 3This is a logical flowchart of the fabrication of the flexible impedance sensing array and the acquisition of coupled signals in this invention. Figure 4 This is a multi-level interaction diagram of the precision hardware component under test, the multi-channel acquisition module, and the central processing unit in this invention; Figure 5 This is a flowchart illustrating the logic of generating a pseudo-color heat map and outputting damage assessment results based on the spatial distribution of parameter deviation in this invention. Detailed Implementation

[0020] The following will be combined with the appendix Figures 1 to 5 The technical solution of the present invention is clearly and completely described below. Obviously, the following embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. Example 1: This embodiment 1 provides a method for detecting fatigue damage in precision hardware, including the following steps:

[0021] Step S1: Fabricate a flexible impedance sensing array.

[0022] In this step, the flexible impedance sensing array serves as the physical medium for acquiring surface charge transfer information of precision metal parts, and its physical characteristics directly determine the signal fidelity. Specifically, it includes the following sub-steps: Step S101: Prepare a polydimethylsiloxane mixture and degas it.

[0023] Polydimethylsiloxane was selected as the flexible substrate material, and the polydimethylsiloxane monomer and curing agent were mixed evenly at a mass ratio of 10:1.

[0024] The mixture is placed in a vacuum degassing machine and maintained at a vacuum of 0.08 MPa for 30 minutes to remove internal micro-bubbles and prevent pore defects from forming in the substrate after curing.

[0025] Step S102: Prepare the array groove template and cast it into shape.

[0026] A thick layer of photoresist is spin-coated onto a silicon wafer, and ultraviolet exposure and development are performed using a mask to obtain an array structure template composed of photoresist protrusions. The cross-sectional dimensions of the protrusions are 200μm×200μm, the height is 150μm, the spacing between adjacent protrusions is 50μm, and the whole structure is arranged in a 10×10 matrix.

[0027] The polydimethylsiloxane mixture after degassing in step S101 is poured onto the array structure template so that the polydimethylsiloxane mixture completely covers the raised array; as an optional alternative, precision mold forming technology can also be used, that is, the polydimethylsiloxane mixture is directly imprinted using a metal mold that has been pre-processed with a micro-pillar array.

[0028] Step S103: Control the curing process to obtain the target flexible substrate.

[0029] An array structure template filled with polydimethylsiloxane is placed in an oven and heated for curing. The Young's modulus of the substrate material is precisely controlled by changing the curing temperature and curing time. For example, curing at 80°C for 2 hours yields a substrate with a Young's modulus of approximately 1.5 MPa. If the curing temperature is lowered to 60°C and the curing time is extended to 4 hours, the Young's modulus can be reduced to approximately 0.5 MPa. Thus, the Young's modulus is controlled within a preset range of 0.5 MPa to 2.0 MPa, while the mass fraction of the crosslinked network of polydimethylsiloxane after curing is between 90% and 95%.

[0030] After curing, the polydimethylsiloxane substrate is peeled off from the array structure template, and a regular array of grooves is formed on the surface of the polydimethylsiloxane. Each array groove has a cross-sectional size of 200μm×200μm and a depth of 150μm. The spacing between adjacent array grooves is 50μm, forming a 10×10 impedance unit matrix, and the coverage area achieves millimeter-level accuracy.

[0031] Through steps S101 to S103, a sensing substrate with preset flexibility and surface microstructure is obtained, which can be closely attached to complex irregular curved surfaces with cylindrical surfaces, spherical vertices and steep transition edges, such as high-performance springs and aerospace micro-structural components.

[0032] Step S104: Prepare a high conductivity hydrophilic composite gel precursor solution.

[0033] Acrylamide monomer, sodium alginate and carbon nanofibers were dissolved in deionized water at a mass ratio of 8:1.5:0.5. Crosslinking agent N,N'-methylenebisacrylamide and photoinitiator 2-hydroxy-2-methylphenylacetone were added to the solution and stirred until all components were uniformly dispersed to obtain a gel precursor solution.

[0034] Acrylamide will polymerize to form a three-dimensional polymer network framework of polyacrylamide, sodium alginate will enhance the hydrophilicity and interfacial contact ability of the system, and carbon nanofibers will be used to construct nanoscale three-dimensional conductive pathways. By adjusting the amount of carbon nanofibers added, the conductivity of the final gel can be stabilized within a preset range of 5S / m to 15S / m.

[0035] Step S105: Fill the array grooves using vacuum-assisted suction and gel in situ.

[0036] The polydimethylsiloxane substrate with arrayed grooves obtained in step S103 is placed in a vacuum chamber, and a sufficient amount of gel precursor liquid from step S104 is dropped onto the surface of the arrayed grooves. Vacuum suction is started to extract the air from the arrayed grooves, and the gel precursor liquid completely penetrates into each micro-groove under the pressure difference until the liquid surface is level with the surface of the polydimethylsiloxane substrate.

[0037] Subsequently, ultraviolet light source irradiation is turned on to initiate in-situ polymerization and cross-linking reaction of the gel precursor liquid, forming a resilient conductive hydrophilic composite gel. After polymerization, the surface of the hydrophilic composite gel is flush with the surface of the polydimethylsiloxane substrate, resulting in multiple miniaturized impedance units that are independent of each other and have consistent electrical conductivity.

[0038] Steps S104 and S105 construct highly conductive and well-coupled sensitive sites through in-situ gelation, providing a reliable physical contact interface for subsequent extraction of charge transfer resistance and interface capacitance signals.

[0039] In summary, step S1 completes all the fabrication processes of the flexible impedance sensing array. The flexible impedance sensing array, through its controllable low-modulus flexibility, can fit closely to the complex curved surface of the precision hardware being tested, eliminating air gaps. Through the arrayed micro gel impedance units, it achieves the capture of electrochemical information at multiple points with high spatial resolution in the area to be tested.

[0040] Based on this, proceed to step S2, where the hardware component under test is coupled with the flexible impedance sensing array to collect data and perform frequency sweep measurement.

[0041] Step S2: Achieve coupling and data acquisition between the precision hardware component under test and the sensing unit.

[0042] In this step, after completing step S1 to prepare the flexible impedance sensing array, the precision metal part to be tested is stably physically coupled and electrically connected to the flexible impedance sensing array to construct a signal path for electrochemical impedance measurement. Specifically, this includes the following sub-steps: Step S201: Place the precision hardware to be tested and apply constant pressure.

[0043] The precision hardware component to be tested is placed above the flexible impedance sensing array, so that the test area of ​​the precision hardware component is aligned with the 10×10 miniaturized impedance unit matrix in the flexible impedance sensing array; in response to the micron-level roughness on the surface of the precision hardware component to be tested, a constant pressure is applied to the precision hardware component to be tested through a pressure feedback control system, so that the hydrophilic composite gel undergoes elastic deformation.

[0044] The pressure feedback control system maintains a constant pressure within a preset pressure range of 0.2MPa to 0.5MPa. Under this pressure range, the deformation of the hydrophilic composite gel reaches 15% to 25% of its original height, thereby allowing the hydrophilic composite gel to cover the test area of ​​the precision hardware under test and eliminating the air gap between the flexible impedance sensing array and the precision hardware under test.

[0045] Step S202: Reduce contact impedance and ensure signal coupling.

[0046] Through the deformation compensation mechanism in step S201, a physical contact interface is formed between the hydrophilic composite gel and the surface of the precision hardware under test. The elimination of the air gap reduces the contact impedance, allowing the subsequently applied AC excitation signal to couple into the double layer of the surface of the precision hardware under test.

[0047] Thus, mechanical and electrical coupling between the precision hardware component under test and the flexible impedance sensing array is completed through steps S201 and S202.

[0048] Step S203: Configure the multi-channel signal acquisition module.

[0049] The system is equipped with a multi-channel signal acquisition module, which has 100 synchronous acquisition channels, each corresponding to one of the 100 miniaturized impedance units of the flexible impedance sensing array. Each acquisition channel has a built-in high-precision current sensor with a current detection limit of 10nA, which is a preset detection threshold. The multi-channel signal acquisition module uses a 16-bit high-resolution analog-to-digital converter to capture the weak current fluctuations caused by changes in the microstructure of the precision hardware material under test.

[0050] Therefore, step S203 constructs a signal acquisition hardware link that matches the scale of the flexible impedance sensing array, ensuring that the current response of each miniaturized impedance unit can be recorded independently and synchronously.

[0051] Step S204: Complete the electrical connection using shielded wires.

[0052] The multi-channel signal acquisition module is electrically connected to each miniaturized impedance unit through shielded wires. The shielded wires adopt a coaxial double-layer shielding structure to suppress the interference of power frequency noise and spatial stray electromagnetic fields on weak current signals in the industrial environment.

[0053] Thus, steps S203 and S204 together complete the construction of a high-fidelity signal transmission link from the sensing unit to the acquisition circuit, so that the current response generated by the miniaturized impedance unit is not polluted by the external electromagnetic environment during transmission.

[0054] In summary, step S2 achieves physical bonding and signal path construction between the precision hardware under test and the flexible impedance sensing array through two steps: mechanical pressure coupling and electrical shielding connection. This eliminates the air gap at the interface and reduces the contact impedance. At the same time, the multi-channel synchronous acquisition architecture and shielded transmission design ensure the accuracy of capturing weak current signals.

[0055] After completing step S2, proceed to step S3, where the central processing unit controls the frequency generator to output a micro-amplitude sinusoidal voltage signal to perform frequency sweep measurement and current response recording.

[0056] Step S3: Perform frequency sweep measurement and current response recording.

[0057] In this step, after completing the coupling acquisition in step S2, the electrochemical impedance spectroscopy (EIS) frequency sweep measurement stage begins. By applying a wide-frequency-domain micro-amplitude excitation and simultaneously acquiring the current response, the EIS, reflecting the surface condition of the precision metal part under test, is obtained. Specifically, this includes the following sub-steps: Step S301: Configure sweep frequency excitation parameters.

[0058] The central processing unit controls the frequency generator to output a micro-sinusoidal voltage signal. The amplitude of the micro-sinusoidal voltage signal is set to 5mV, and the amplitude does not exceed a preset voltage threshold of 10mV. The operating frequency range of the frequency generator is set to 10Hz to 100kHz. A logarithmic step frequency scanning mode is adopted, and the scanning step size is set to 10 sampling points per octave, with a total number of sampling points of no less than 50, thereby obtaining a complete coverage of the electrochemical impedance spectroscopy characteristic curve in the 10Hz to 100kHz range.

[0059] Step S302: Apply a small-amplitude sinusoidal voltage signal between the precision hardware component under test and the flexible impedance sensing array.

[0060] The small-amplitude sinusoidal voltage signal output by the frequency generator is applied between the precision hardware under test and each miniaturized impedance unit. Since the excitation voltage amplitude is limited to below 10 mV, the entire measurement process is in the electrochemical linear response region, avoiding nonlinear electrochemical polarization reaction on the surface of the precision hardware under test caused by excessive electrical excitation, thereby protecting the surface integrity of the precision hardware under test.

[0061] Therefore, steps S301 and S302 jointly define the waveform parameters and application method of the frequency sweep excitation. By performing a low-amplitude logarithmic frequency sweep, a foundation is provided for obtaining high signal-to-noise ratio and lossless impedance spectrum data.

[0062] Step S303: Synchronously acquire the current response of each miniaturized impedance unit.

[0063] The multi-channel signal acquisition module has 100 synchronous acquisition channels to simultaneously acquire the current response data generated by each of the 100 miniaturized impedance units in the flexible impedance sensing array. The current sensor built into each acquisition channel has a current detection limit of 10nA. With the help of a 16-bit high-resolution analog-to-digital converter, it can distinguish the slight current differences caused by changes in the microstructure of the precision hardware under test.

[0064] At each frequency sweep point, the multi-channel signal acquisition module completes one current signal acquisition for all miniaturized impedance units and sends the acquired data to the central processing unit for processing.

[0065] Step S304: Extract the real and imaginary parts of the complex impedance and calculate the magnitude and phase.

[0066] The central processing unit calls the quadrature vector lock-in amplifier and uses the sinusoidal voltage signal output by the frequency generator as a reference signal to perform quadrature demodulation on the current signal acquired in step S303. The quadrature vector lock-in amplifier separates the component in phase with the excitation voltage and the quadrature component from the current signal, which are respectively used as the real part and the imaginary part of the complex impedance.

[0067] The central processing unit uses the real and imaginary parts to calculate the complex impedance modulus and phase of each miniaturized impedance unit at the frequency point, and obtains local electrochemical impedance spectral data.

[0068] Therefore, steps S303 and S304 realize the signal conversion from the original current waveform to the quantitative impedance parameter. Multi-point synchronous acquisition combined with lock-in amplification and demodulation ensures the accuracy of impedance data.

[0069] Step S305: Monitor signal stability and execute a retest mechanism.

[0070] At each frequency point, the central processing unit performs excitation-acquisition-demodulation three times consecutively on the same miniaturized impedance unit to obtain three sets of complex impedance magnitude values. The central processing unit calculates the volatility of the three consecutive complex impedance magnitude values. If the volatility exceeds 1%, it determines that the data at the frequency point has not reached a stable state. It automatically triggers a remeasurement of the frequency point and repeats steps S303 and S304 until the volatility of the complex impedance magnitude value drops to less than 1%.

[0071] Therefore, through the stability monitoring in step S305, occasional data fluctuations caused by external electromagnetic interference or unstable contact are eliminated, ensuring that the final recorded electrochemical impedance spectroscopy data is stable and reliable.

[0072] In summary, step S3 obtains the electrochemical impedance spectral characteristics of each miniaturized impedance unit in the range of 10Hz to 100kHz through parameterized wideband micro-amplitude excitation, 100-channel synchronous acquisition, and lock-in amplification and demodulation. The robustness of the data is then ensured by online stability monitoring and retesting mechanisms.

[0073] After completing step S3, the recorded complex impedance magnitude, phase, and real and imaginary part data will be sent to step S4 for fitting the Landers equivalent circuit model and quantifying the parameter deviation.

[0074] Step S4: Perform equivalent circuit model fitting and parameter quantization.

[0075] After obtaining the electrochemical impedance spectroscopy data of each miniaturized impedance unit in step S3, this step uses a Landers equivalent circuit model to fit the impedance spectrum, converting it into electrical parameters with clear physical meaning, and then uses the parameter deviation to achieve a quantitative assessment of fatigue damage. Specifically, this includes the following sub-steps: Step S401: Import electrochemical impedance spectroscopy data and call the Landers model.

[0076] The central processing unit imports the electrochemical impedance spectroscopy data of each miniaturized impedance unit recorded in step S3 into a Landers equivalent circuit model for fitting. The Landers equivalent circuit model is based on the solution resistance. Charge transfer resistance It is composed of a constant phase angle element CPE connected in series.

[0077] Among them, solution resistance Characterizing the ion conduction properties and charge transfer resistance at the interface between the hydrophilic composite gel and the surface of the precision hardware component under test. The constant phase angle element (CPE) is used to characterize the non-ideal capacitance characteristics of the surface of a precision metal part under test, representing the ease or difficulty of charge transfer in the electric double layer.

[0078] Compared to an ideal capacitor, a constant phase angle element (CPE) can more accurately reflect the uneven surface charge distribution caused by lattice distortion and dislocation accumulation in the early stages of fatigue.

[0079] Step S402: Perform nonlinear least squares fitting to obtain component parameters.

[0080] The central processing unit (CPU) performs nonlinear least-squares fitting on the electrochemical impedance spectroscopy data of each miniaturized impedance unit, with frequency as the independent variable and the real and imaginary parts of the complex impedance as the dependent variables. During the fitting process, the CPU iteratively approximates the impedance expression of the Landels equivalent circuit model with the measured impedance spectrum until the fitting residual converges to below a preset threshold. After the fitting is completed, the CPU outputs the solution resistance corresponding to each miniaturized impedance unit. Charge transfer resistance And the two characteristic parameters of the constant phase angle element CPE, CPE-T and CPE-P.

[0081] CPE-T characterizes the magnitude of the interface capacitance, while CPE-P is a dimensionless exponent with a value ranging from 0.5 to 1, used to measure the degree of non-ideality of the electrode surface.

[0082] Thus, steps S401 and S402 complete the extraction process from impedance spectrum data to equivalent circuit element parameters, enabling the surface electrochemical state of the precision hardware under test in each miniaturized impedance unit to be quantitatively expressed using physical element parameters.

[0083] Step S403: Analyze the correlation between changes in charge transfer resistance and interface capacitance and fatigue damage.

[0084] The central processing unit extracts the charge transfer resistance from the fitting results. The interface capacitance value is calculated from the CPE-T and CPE-P parameters of the constant phase angle element CPE.

[0085] Charge transfer resistance The change in charge transfer resistance is negatively correlated with the degree of fatigue damage. When microcracks or lattice defects appear on the surface of the precision metal part under test, the energy barrier for charge transfer decreases, and the charge transfer resistance increases. The change in interfacial capacitance is used to help determine the propagation depth of microcracks. The generation of microcracks increases the effective contact area between the surface of the precision hardware part under test and the hydrophilic composite gel, resulting in an increase in the interfacial capacitance.

[0086] In this embodiment, a correlation model between capacitance change rate and crack evolution stage is pre-established in the central processing unit. The correlation model is obtained by fitting calibration experimental data of different fatigue levels and is used to quantitatively evaluate microscopic damage.

[0087] Step S404: Calculate the parameter deviation.

[0088] For each miniaturized impedance unit, the central processing unit extracts the charge transfer resistance fitted during the current detection cycle. And read the historical reference resistance calibrated by the miniaturized impedance unit under the condition of intact materials. .

[0089] Central processing unit calculation parameter deviation The calculation formula is as follows: ; in, This refers to the parameter deviation, expressed as a percentage. The charge transfer resistance is fitted to the current detection cycle, in Ω. This is the historical reference resistance of the miniaturized impedance unit for the precision hardware component under test, in its intact state, expressed in Ω. Parameter deviation. The larger the value, the more severe the material damage in the local area corresponding to the miniaturized impedance unit.

[0090] Step S405: Determine fatigue risk.

[0091] The central processing unit calculates the parameter deviation of each miniaturized impedance unit. Compare with the preset ratio threshold.

[0092] The preset ratio threshold is set to 20%, when the charge transfer resistance The value decreases by more than a preset percentage threshold, i.e., parameter deviation. When the value exceeds 20%, the central processing unit determines that there is a risk of fatigue in the local area corresponding to the miniaturized impedance unit.

[0093] Therefore, steps S403 to S405 are performed through the charge transfer resistor. The physical changes in interfacial capacitance explain the electrochemical mechanism of fatigue evolution, and the normalized parameter deviation is used to explain this mechanism. As a quantitative criterion, it enables the automated identification of fatigue risks.

[0094] In summary, step S4 uses the Landers equivalent circuit model to analyze the electrochemical impedance spectroscopy into component parameters with clear physical meanings, such as solution resistance, charge transfer resistance, and constant phase angle elements. The decrease in charge transfer resistance and the increase in interfacial capacitance are used to characterize the initiation and propagation of microcracks on the surface of the precision metal part under test, and the parameter deviation is used to further analyze these parameters. The quantitative calculation enables point-by-point determination of fatigue risk for each miniaturized impedance unit.

[0095] After completing step S4, the parameter deviation of each miniaturized impedance unit is... The fatigue risk assessment result is sent to step S5 to generate the fatigue damage assessment result of the precision hardware part to be tested.

[0096] Based on the detection process constructed in steps S1 to S4, this embodiment also integrates supplementary processing steps such as signal quality assurance, remaining lifetime prediction, damage grading early warning, and cross-verification of complex plane impedance characteristics to improve the reliability and engineering applicability of the detection results.

[0097] T1: In terms of signal preprocessing, a data preprocessing unit is configured. After the current response data is acquired in step S3, the data preprocessing unit sequentially performs 50Hz power frequency notch filtering and Butterworth bandpass filtering on the data. The 50Hz notch filter is used to filter out power frequency interference introduced by the power system, and the passband frequency range of the Butterworth bandpass filter is set according to the frequency sweep excitation from 10Hz to 100kHz to eliminate stray electromagnetic signal interference caused by surface roughness fluctuations in precision hardware parts.

[0098] T2: For remaining fatigue life prediction, a database storage module is configured to record the impedance characteristic evolution curves of precision hardware parts made of different materials such as stainless steel, titanium alloy, and spring steel during service. The central processing unit calls a Long Short-Term Memory (LSTM) neural network model, which includes one input layer, at least one hidden layer with a forget gate, input gate, and output gate structure, and one regression output layer.

[0099] The LSTM neural network is trained using historical fatigue test data stored in the database module. During training, the time series segments of the impedance feature evolution curve are used as input, and the corresponding remaining fatigue life is used as the label. The network weights are updated through the backpropagation algorithm.

[0100] After training is complete, the charge transfer resistance obtained in the current detection cycle will be... The time series of interface capacitance values ​​are input into the trained LSTM neural network, and the network outputs the predicted value of the remaining fatigue life of the precision hardware part under test, with a prediction accuracy of greater than or equal to 85%.

[0101] T3: Regarding damage level classification and early warning, the central processing unit calculates the parameter deviation based on step S404. The local area corresponding to each miniaturized impedance unit is automatically divided into 4 damage levels: A value between 0% and 10% is classified as Level 1, indicating no damage. A level of 2 is defined as the fatigue induction period, with a range of 10% to 25%. A score between 25% and 45% is classified as Level 3, indicating the microcrack initiation stage. A value greater than 45% is classified as Level 4, indicating a structural failure risk zone.

[0102] This embodiment also includes an alarm module, which detects when the parameters of any miniaturized impedance unit deviate from the specified value. When the damage exceeds the 30% safety threshold, the alarm module sends a warning signal to the monitoring terminal via a wireless transmission protocol so that operators can be informed of the damage status in a timely manner.

[0103] T4: Regarding data cross-validation, the central processing unit introduces impedance characteristic point analysis logic in the complex plane. It plots a Nyquist plot based on the real and imaginary parts of each miniaturized impedance unit, extracts the diameter of the high-frequency semicircle in the plot, and independently verifies the charge transfer resistance by comparing the change in the semicircle diameter between the current detection cycle and the baseline state. The downward trend.

[0104] In the high-frequency range above 50kHz, the solution resistance is analyzed in detail. Changes, when When the deviation from the reference value exceeds the preset fluctuation range, it is determined that the coupling stability between the hydrophilic composite gel and the surface of the precision hardware under test is abnormal. In the low frequency band below 100Hz, the CPE-P parameter of the constant phase angle element CPE is extracted, the phase deviation between the current detection cycle and the reference state is calculated, and the stress concentration under the surface of the precision hardware under test is inferred accordingly.

[0105] Through cross-analysis of three dimensions—Nyquist plot, high-frequency solution resistance, and low-frequency phase deviation—independent parameters of deviation are provided. This serves as an auxiliary criterion for judgment, enhancing the understanding of charge transfer resistance. The reliability of the downward trend assessment.

[0106] Example 2: This embodiment 2, based on the precision hardware fatigue damage detection method disclosed in embodiment 1, further provides a high-resolution detection scheme for ultra-miniature hardware parts, such as miniature clamps or interventional medical stents in medical devices. The following describes the distinguishing technical features of embodiment 2 compared to embodiment 1 step by step.

[0107] Miniaturization iteration of step S1: Fabrication of a high-density flexible impedance sensing array.

[0108] To adapt to the size characteristics of ultra-miniature hardware components, this embodiment iterates the miniaturization of the fabrication process of the flexible impedance sensing array in step S1, specifically including the following sub-steps: Step S1.1: Fabricate nanoscale precision molds using electron beam lithography.

[0109] Electron beam exposure directly writes an array pattern onto a silicon wafer, forming an array structure template with a protrusion cross-sectional size of 50μm×50μm, a height of 50μm, and a spacing of 10μm between adjacent protrusions. The array structure template is used for subsequent polydimethylsiloxane casting to replace the ultraviolet lithography or precision mold forming technology in Example 1.

[0110] Step S1.2: Adjust the content of carbon nanofibers in the hydrophilic composite gel precursor solution.

[0111] The mass fraction of carbon nanofibers in the gel precursor solution was increased by 2 times compared to Example 1. During the preparation process, the precursor solution containing carbon nanofibers was subjected to an ultrasonic exfoliation process with an ultrasonic power of 200W and a processing time of 30min, so that the length of the carbon nanofibers was distributed in the range of 2μm to 5μm.

[0112] This treatment allows for the construction of a well-connected three-dimensional conductive network within a 50μm wide array groove, maintaining the conductivity within a preset range of 5S / m to 15S / m.

[0113] Step S1.3: Complete the curing and in-situ gel polymerization of polydimethylsiloxane.

[0114] Following the processes in steps S103 and S105 of Example 1, the in-situ polymerization of polydimethylsiloxane curing and gel precursor liquid was completed, and a miniaturized impedance unit matrix with an element size of 50 μm and a spacing of 10 μm was finally obtained.

[0115] Compared to the 200μm array element in Example 1, the spatial resolution of this example is improved by about 4 times.

[0116] Thus, through steps S1.1 to S1.3, this embodiment completes the fabrication of a high-density flexible impedance sensing array, which can adapt to the detection requirements of ultra-miniature hardware parts in terms of physical size and spatial resolution.

[0117] Step S2 refinement: high-precision coupled acquisition.

[0118] To achieve a stable reduction in contact impedance without damaging the structure of the ultra-miniature hardware components, this embodiment refines the coupling acquisition step in step S2, specifically including the following sub-steps: Step S2.1: Introduce a precision three-dimensional displacement platform and a vacuum suction head.

[0119] The micro-sized hardware component under test is fixed to a precision three-dimensional displacement platform by a vacuum suction head. The precision three-dimensional displacement platform presses the micro-sized hardware component under test into the surface of the flexible impedance sensing array along the vertical direction with a step accuracy of 0.1μm.

[0120] Step S2.2: Configure a micro-pressure sensor.

[0121] The pressure feedback control system uses a micro-level pressure sensor with a range of 0N to 10N to control the constant pressure within a preset pressure range of 0.01MPa to 0.05MPa, with a pressure control accuracy of ±0.001MPa. The pressure range allows the hydrophilic composite gel to cover the test area of ​​the micro-sized hardware part under test without deforming by more than 10% of its original height, thus avoiding plastic deformation or structural damage to the micro-sized parts due to excessive pressure.

[0122] Step S2.3: Optimize the current detection limit of the multi-channel signal acquisition module.

[0123] The current detection limit of the multi-channel signal acquisition module is optimized to 1nA to cope with the reduced current response amplitude of the miniaturized impedance unit due to the reduction in area. The multi-channel signal acquisition module still maintains a 16-bit high-resolution analog-to-digital converter, the same as in Example 1.

[0124] Thus, through steps S2.1 to S2.3, this embodiment achieves precise coupling and weak signal acquisition of ultra-miniature hardware parts, and completes stable control of contact impedance while ensuring the integrity of the parts.

[0125] Step S3 parameter adjustment: high frequency sweep measurement.

[0126] The electrochemical response time constant of ultra-miniature hardware parts is smaller than that of conventionally sized parts. Therefore, the frequency sweep parameters in step S3 are adjusted, specifically including the following sub-steps: Step S3.1: Increase the upper limit of the sweep frequency.

[0127] The upper limit of the operating frequency of the frequency generator is increased from 100kHz in Example 1 to 500kHz, the frequency range is set to 10Hz to 500kHz, the scanning step size is still 10 sampling points per octave, and the total number of sampling points is not less than 60.

[0128] Step S3.2: Reduce the amplitude of the excitation voltage.

[0129] The amplitude of the micro-sinusoidal voltage signal is set to 3mV, which is no more than the preset voltage threshold of 10mV and is further reduced compared to 5mV in Example 1. The reduction of the excitation voltage amplitude is to reduce the influence of Joule heating generated by microcurrent on the properties of hydrophilic composite gel.

[0130] The remaining frequency sweep measurement and current response recording processes are the same as steps S302 to S305 in Example 1, and will not be repeated here. By adjusting the parameters in steps S3.1 and S3.2, this embodiment adapts the frequency sweep measurement bandwidth and excitation intensity to the electrochemical response characteristics of the ultra-miniature hardware component.

[0131] Model improvement in step S4: distributed element fitting and spatial weighting algorithm.

[0132] To address the significant edge effect characteristic of ultra-miniature hardware components, this embodiment improves the equivalent circuit model and parameter deviation calculation method in step S4, specifically including the following sub-steps: Step S4.1: Add an edge leakage current compensation branch.

[0133] Based on the Landers equivalent circuit model, an edge leakage current compensation branch is added. This edge leakage current compensation branch is a circuit connected to the charge transfer resistor. parallel leakage resistance It is used to compensate for the extra charge paths generated by the concentration of electric field at the edge of ultra-micro parts.

[0134] The impedance expression for the improved equivalent circuit model is: ; in, The total complex impedance is expressed in Ω. The resistance of the solution is expressed in Ω. This is the charge transfer resistance, measured in Ω. Ω represents the edge leakage resistance; j is the imaginary unit. Angular frequency, in rad / s; This is an interface capacitance parameter, in F. This is the CPE-P parameter, with a value ranging from 0.5 to 1.

[0135] Step S4.2: Perform nonlinear least squares fitting.

[0136] Central processing unit at frequency Using the independent variable and the real and imaginary parts of the measured complex impedance as the dependent variables, the above impedance expression containing leakage resistance is iteratively fitted using the nonlinear least squares method. The fitting convergence condition is that the fitting residual is less than... After fitting, the output of each miniaturized impedance unit is given. , , , and .

[0137] Step S4.3: Introduce a spatial weighted algorithm to calculate the parameter deviation.

[0138] When calculating the parameter deviation, a spatial weighting algorithm is introduced to compensate for the difference in reference values ​​caused by the uneven force distribution between the central and edge regions of the tested micro-sized hardware during the coupling pressurization process.

[0139] The spatial weighting coefficients are allocated according to the following rules: miniaturized impedance units located in the central region. Weighting coefficients The value is 1.0; miniaturized impedance units located in the edge region. Weighting coefficients Based on its linearly decreasing distance from the center, the minimum weight coefficient It is 0.7.

[0140] Weighted parameter deviation The calculation formula is: ; in, This refers to the parameter deviation, expressed as a percentage. The charge transfer resistance is fitted to the current detection cycle, in Ω. The historical reference resistance of the micro-sized hardware component under test in its intact state is expressed in Ω. This is the spatial weighting coefficient, with a value ranging from 0.7 to 1.0.

[0141] Parameter deviation The larger the value, the more severe the material damage in the corresponding local area of ​​the miniaturized impedance unit.

[0142] Step S4.4: Introduce the radius of curvature analysis of the complex plane impedance arc as an auxiliary criterion.

[0143] The central processing unit calculates the radius of curvature based on the shape of the semicircular arc in the Nyquist plot.

[0144] When a specific miniaturized impedance unit If the value shows a monotonically decreasing trend after three consecutive measurements, and the absolute value of the decreasing slope exceeds the preset slope threshold of 0.5Ω / measurement, then even if the current parameter deviation is... If the preset threshold of 20% is not reached, the central processing unit will still automatically increase the risk weight of the miniaturized impedance unit by 1 level, which will be reflected in the subsequent damage level classification.

[0145] Therefore, through steps S4.1 to S4.4, this embodiment employs distributed component equivalent circuits considering edge leakage current and spatially weighted deviation calculations to improve the accuracy of fatigue damage assessment for ultra-miniature hardware components; step S4.4 also introduces curvature radius trend analysis as an independent method. The threshold-based auxiliary criterion enables early micro-damage to be identified and marked when the deviation is not yet obvious.

[0146] In addition, this embodiment adds a wavelet transform-based denoising algorithm to the data preprocessing unit in the data processing stage. The data preprocessing unit performs multi-scale wavelet decomposition on the current response signal, identifies the components related to surface scratches and dirt of the micro-sized hardware parts under test as high-frequency noise and removes them, and retains the mid-to-low frequency characteristic signals related to internal fatigue damage of the material. After wavelet denoising, the diagnostic accuracy of rough micro-sized hardware parts that have not been finely polished is guaranteed.

[0147] In summary, this embodiment extends the detection method of Embodiment 1 from conventional precision hardware parts to ultra-micro hardware parts through improvements in array miniaturization, coupling precision, frequency sweeping frequency, model distribution, and algorithm spatial weighting. This enables high-resolution fatigue damage detection and early warning for complex micro-structured parts such as medical device micro-clamps and interventional medical stents.

[0148] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects.

[0149] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment includes only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for detecting fatigue damage in precision hardware, characterized in that, include: A flexible impedance sensing array consisting of a polydimethylsiloxane substrate and multiple array grooves formed on its surface is coupled to a precision hardware component under test. Each array groove is filled with a conductive hydrophilic composite gel formed by in-situ polymerization of acrylamide monomer, sodium alginate and carbon nanofibers to form multiple miniaturized impedance units. A constant pressure is applied to the precision hardware under test through a pressure feedback control system, maintaining a pressure range within a preset pressure range. This causes the conductive hydrophilic composite gel to undergo elastic deformation to cover the test area. The deformation compensates for the micron-level roughness of the test area surface, eliminates air gaps, and reduces contact resistance. A small-amplitude sinusoidal voltage signal is applied between the precision hardware component under test and each of the miniaturized impedance units. The current response data of each miniaturized impedance unit in the frequency sweep band is collected simultaneously. The complex impedance information is extracted and fitted into the Landers equivalent circuit model to obtain the charge transfer resistance and interface capacitance values ​​corresponding to each miniaturized impedance unit. Based on the charge transfer resistance obtained by fitting the current detection cycle of the same miniaturized impedance unit and the historical reference resistance calibrated in the intact state, the parameter deviation is calculated, and based on the comparison result of the parameter deviation and the preset ratio threshold, the fatigue damage status of the precision hardware under test in the corresponding local area is determined.

2. The method for detecting fatigue damage in precision hardware according to claim 1, characterized in that, The step of applying a constant pressure within a preset pressure range to the precision hardware component under test through a pressure feedback control system includes: The deformation of the conductive hydrophilic composite gel is made to reach a preset percentage range of its original height, and the deformation of the conductive hydrophilic composite gel compensates for the micron-level roughness of the surface of the precision hardware part to be tested.

3. The method for detecting fatigue damage in precision hardware according to claim 1, characterized in that, The step of obtaining the charge transfer resistance and interface capacitance values ​​corresponding to each miniaturized impedance unit includes: Nonlinear least squares fitting is performed on the complex impedance information to extract the characteristic parameters of the charge transfer resistor and the constant phase angle element, and the interface capacitance value is calculated from the characteristic parameters of the constant phase angle element; wherein, the change in the interface capacitance value is positively correlated with the fatigue microcrack propagation depth, and is used to help determine the degree of microcrack propagation.

4. The method for detecting fatigue damage in precision hardware according to claim 3, characterized in that, Also includes: Nyquist plots are drawn based on the complex impedance information of each miniaturized impedance unit. The diameter of the semicircle in the high-frequency band, the solution resistance value in the high-frequency band, and the phase parameters of the constant phase angle element in the low-frequency band are extracted. By comparing the changes in the diameter of the semicircle, the solution resistance value, and the phase parameters in the current detection cycle with those in the reference state, the fatigue risk determined based on the charge transfer resistance is cross-validated in multiple dimensions.

5. The method for detecting fatigue damage in precision hardware according to claim 1, characterized in that, The array grooves of the flexible impedance sensing array are formed by casting with nanoscale precision molds prepared by electron beam exposure technology, so that the cross-sectional size of the miniaturized impedance unit and the spacing between adjacent array grooves are adapted to the detection requirements of ultra-miniature hardware parts. The conductive hydrophilic composite gel contains carbon nanofibers that are treated with an ultrasonic exfoliation process to construct a three-dimensional conductive network within a miniaturized array of grooves.

6. The method for detecting fatigue damage in precision hardware according to claim 5, characterized in that, The precision hardware component to be tested is an ultra-miniature hardware component; In the step of coupling the precision hardware component to be tested with the flexible impedance sensing array, the ultra-micro hardware component is pressed into the surface of the flexible impedance sensing array with micron-level stepping accuracy using a precision three-dimensional displacement platform, and the preset pressure range is controlled within a preset safety percentage of the deformation of the conductive hydrophilic composite gel, so that the deformation does not exceed its original height.

7. The method for detecting fatigue damage in precision hardware according to claim 5, characterized in that, The step of fitting the complex impedance information to the Landers equivalent circuit model further includes: In the Landers equivalent circuit model, an edge leakage resistor is added in parallel with the charge transfer resistor to compensate for the additional charge path generated by the concentration of the edge electric field in the ultra-miniature hardware.

8. The method for detecting fatigue damage in precision hardware according to claim 7, characterized in that, The step of calculating the parameter deviation also includes: A spatial weighting algorithm is introduced, wherein the historical reference resistance of the miniaturized impedance unit located in the central region is assigned a higher weighting coefficient than the historical reference resistance of the miniaturized impedance unit located in the edge region, in order to compensate for the difference in reference value caused by the uneven force distribution between the central region and the edge region of the ultra-miniature hardware.

9. The method for detecting fatigue damage in precision hardware according to claim 1, characterized in that, Also includes: The remaining fatigue life of the precision hardware under test is predicted using a long short-term memory neural network model. The long short-term memory neural network model takes the time series of the charge transfer resistance and the interface capacitance values ​​obtained in each detection cycle as input and outputs the predicted value of the remaining fatigue life.

10. A precision hardware fatigue damage detection system, used to perform the precision hardware fatigue damage detection method according to any one of claims 1 to 9, characterized in that, include: A flexible impedance sensing array includes a polydimethylsiloxane substrate and multiple array grooves formed on its surface. Each array groove is filled with a conductive hydrophilic composite gel formed by in-situ polymerization of acrylamide monomer, sodium alginate and carbon nanofibers to form multiple miniaturized impedance units. The pressure feedback control system is used to apply a constant pressure within a preset pressure range to the precision hardware part to be tested, so that the conductive hydrophilic composite gel undergoes elastic deformation to cover the area to be tested. The deformation compensates for the micron-level roughness of the surface of the area to be tested, eliminates air gaps and reduces contact resistance. A multi-channel signal acquisition module is used to apply a small-amplitude sinusoidal voltage signal between the precision hardware under test and each of the miniaturized impedance units, and simultaneously acquire the current response data of each of the miniaturized impedance units in the frequency sweep band. The central processing unit is used to extract complex impedance information based on the current response data and fit it into the Landers equivalent circuit model to obtain the charge transfer resistance and interface capacitance values. It also calculates the parameter deviation based on the charge transfer resistance of the same miniaturized impedance unit in the current detection cycle and the historical reference resistance to determine the fatigue damage status.

Images