Contact resistance detection method for data line connection stability

Abstract

The invention relates to the technical field of contact resistance detection, in particular to a contact resistance detection method for data line connection stability, and the method comprises the steps: applying high-frequency pulse excitation, obtaining excitation response signals under to-be-detected and standard states, carrying out dynamic impedance modeling decomposition, and extracting an impedance component to be analyzed and a standard impedance component. Constructing an impedance peak position difference change coefficient based on a difference change trend and an average level of time corresponding to an impedance peak position, and calculating a contact interference factor of each component; meanwhile, the fluctuation consistency is evaluated by using the information entropy of the impedance peak amplitude difference, and the impedance amplitude difference consistency is determined. And combining the two to construct a superposition interference characteristic value, screening effective impedance components with small interference, and performing dynamic fusion to obtain a dynamic impedance spectrum after the contact interference is eliminated. And finally, high-precision transient calculation of the contact resistance is realized through a dynamic impedance-resistance conversion model.

Description

Technical Field

[0001] This invention relates to the field of contact resistance testing technology, and in particular to a contact resistance testing method for data cable connection stability. Background Technology

[0002] With the widespread application of smart electronic devices, data cables, as key components for signal transmission and power supply between devices, directly impact system stability and data transmission security due to their connection reliability. Under prolonged plugging and unplugging or environmental factors, data cable interfaces are prone to oxidation, wear, and loosening, leading to increased contact resistance and consequently causing signal attenuation, reduced charging efficiency, and even communication interruptions. Traditional contact resistance detection methods often employ static DC resistance measurement, which struggles to capture the dynamic contact characteristics of data cables under actual operating conditions caused by transient current excitation, particularly failing to effectively reflect the instantaneous response behavior of the contact interface during high-frequency signal transmission. Furthermore, existing technologies are susceptible to the influence of distributed inductance, parasitic capacitance, and external electromagnetic interference during the detection process, resulting in a large amount of noise and non-contact-related components mixed in with the measured impedance signal, affecting the accuracy of contact resistance identification.

[0003] In recent years, dynamic detection technology based on pulse excitation has gradually attracted attention. By applying a high-frequency narrow pulse signal to excite the transient response of the contact end, it can more realistically simulate the electrical conditions of data lines in actual use. However, how to effectively separate the impedance component closely related to the contact state from the complex response signal remains a challenge for current detection technologies. Especially under multi-factor coupling interference, conventional methods struggle to distinguish between impedance changes caused by contact degradation and spurious changes caused by system noise or structural fluctuations, resulting in a high false positive rate.

[0004] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this invention is to provide a contact resistance detection method for data line connection stability, aiming to solve the technical problem that existing data line contact resistance detection technologies are unable to effectively eliminate the influence of interference factors, cannot accurately extract dynamic impedance characteristics that reflect connection stability, and thus have insufficient accuracy in evaluating the contact state.

[0006] To achieve the above objectives, the present invention provides a contact resistance detection method for data cable connection stability, the method comprising: The excitation signal and contact end response signal of the data line under high-frequency pulse excitation are acquired respectively, and recorded as the excitation response signal to be tested and the standard excitation response signal. Dynamic impedance modeling and decomposition were performed on the excitation response signal to be tested and the standard excitation response signal to be standard, and each impedance component to be analyzed and each standard impedance component were extracted. The differences between the impedance component to be analyzed and the standard impedance component with respect to the time corresponding to each impedance peak position are analyzed. The variation coefficient of the impedance peak position difference between the impedance component to be analyzed and the standard impedance component is constructed. Combined with the average level of the difference between the impedance component to be analyzed and the standard impedance component with respect to the time corresponding to the impedance peak position, the contact interference factor of the impedance component to be analyzed is obtained. Based on the fluctuation of the difference between the impedance vector to be analyzed and the standard impedance vector with respect to the peak amplitude of each impedance, the consistency of the impedance amplitude difference of the impedance vector to be analyzed is determined. Based on the contact interference factor and impedance amplitude difference consistency of each impedance vector to be analyzed, the superposition interference characteristic value of each impedance vector to be analyzed is constructed. Based on the superposition interference characteristic value of all impedance components to be analyzed, the effective impedance component vector in the test excitation response signal is extracted. The dynamic impedance spectrum after contact interference elimination is obtained by dynamic fusion. By using a dynamic impedance-resistance conversion model combined with the dynamic impedance spectrum after contact interference elimination, transient calculations are performed on the contact resistance at the data line contact end to obtain the transient value of the data line contact resistance.

[0007] Optionally, the extraction of each impedance component to be analyzed and each standard impedance component specifically includes: Construct the excitation response vector to be analyzed and the standard excitation response vector, and use the impedance components after dynamic impedance modeling decomposition of the excitation response vector to be analyzed as the impedance components to be analyzed. The standard excitation response vector is dynamically modeled and decomposed. The discrete coefficients of each impedance vector in the standard excitation response vector are calculated. The average of the discrete coefficients of all impedance vectors in the standard excitation response vector is used as the judgment threshold. Impedance vectors with discrete coefficients greater than or equal to the judgment threshold are all recorded as standard impedance vectors.

[0008] Optionally, the construction of the stimulus response vector to be analyzed and the standard stimulus response vector further includes: The response signal data of the stimulus response signal to be tested are arranged in ascending order of time to form the stimulus response vector to be analyzed, and the response signal data of the standard stimulus response signal are arranged in ascending order of time to form the standard stimulus response vector.

[0009] Optionally, the formula for calculating the coefficient of variation of the impedance peak position difference between the impedance component to be analyzed and the standard impedance component is as follows: ; In the formula, Let be the coefficient of variation of the impedance peak position difference between the i-th impedance vector to be analyzed and the j-th standard impedance vector, norm() be the exponential normalization function, and M be the number of elements in the impedance peak position difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector. and These are the k-th and (k-1)-th elements in the impedance peak difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector, respectively.

[0010] Optionally, the construction of the impedance peak difference vector further includes: For each impedance vector to be analyzed and each standard impedance vector, the time corresponding to all impedance peak positions in the impedance vector to be analyzed and the standard impedance vector are arranged in the order of peak appearance to form a vector, which is denoted as the impedance peak position time vector of the impedance vector to be analyzed and the impedance peak position time vector of the standard impedance vector. The difference vector between the impedance vector to be analyzed and the standard impedance vector with respect to the impedance peak position time vector is denoted as the impedance peak position difference vector between the impedance vector to be analyzed and the standard impedance vector.

[0011] Optionally, the formula for calculating the contact interference factor of the impedance component to be analyzed is: ; In the formula, Let N be the contact interference factor for the i-th impedance vector to be analyzed, and N be the number of standard impedance vectors. Let be the mean value of the elements in the impedance peak difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector.

[0012] Optionally, the formula for calculating the uniformity of impedance magnitude differences of the impedance vector to be analyzed is: ; In the formula, Let be the impedance magnitude difference uniformity of the i-th impedance vector to be analyzed, and exp() be an exponential function with the natural constant as the base. Let be the information entropy of the impedance amplitude difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector. Here, the amplitudes of all impedance peak positions in the i-th impedance vector to be analyzed and the j-th standard impedance vector are sorted according to their peak positions to form the impedance peak amplitude vector of the i-th impedance vector to be analyzed and the impedance peak amplitude vector of the j-th standard impedance vector. The difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector with respect to the impedance peak amplitude vector is denoted as the impedance amplitude difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector.

[0013] Optionally, the superposition interference characteristic value of each impedance vector to be analyzed is the ratio of the contact interference factor of each impedance vector to the consistency of impedance amplitude difference.

[0014] Optionally, the extraction of the effective impedance component vector further includes: Threshold segmentation is performed on the superimposed interference eigenvalues ​​of all impedance vectors to be analyzed to obtain the segmentation threshold. The impedance vectors to be analyzed whose superimposed interference eigenvalues ​​are lower than the segmentation threshold are recorded as effective impedance component vectors.

[0015] Furthermore, to achieve the above objectives, the present invention also provides a contact resistance detection device for data line connection stability, the device comprising: a memory, a processor, and a contact resistance detection program for data line connection stability stored in the memory and executable on the processor, the contact resistance detection program for data line connection stability being configured to implement the steps of the contact resistance detection method for data line connection stability as described above.

[0016] This invention provides a contact resistance detection method for data cable connection stability. The method introduces a detection mechanism combining high-frequency pulse excitation and dynamic impedance modeling, effectively capturing the transient response characteristics of the data cable contact end under real-world operating conditions, overcoming the low sensitivity and lag issues of traditional static measurement methods. By constructing a standard excitation response signal as a reference benchmark and comprehensively analyzing both impedance peak time variation and amplitude fluctuation, an innovative dual-index collaborative evaluation mechanism of contact interference factor and impedance amplitude difference consistency is proposed. This accurately quantifies the interference level of each impedance component, significantly improving the separation capability of effective signals in complex electromagnetic environments. Furthermore, by superimposing interference characteristic values, adaptive suppression of noise and non-contact-related components is achieved, extracting high-fidelity effective impedance component vectors. Dynamic fusion yields a pure dynamic impedance spectrum, making the transient calculation of contact resistance more accurate and reliable. Ultimately, this method achieves high-precision, anti-interference, and real-time online detection of data cable connection stability, providing strong technical support for early contact degradation warning, and possesses good application prospects and engineering practical value. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an embodiment of the contact resistance detection method for data cable connection stability according to the present invention.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] Reference Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the contact resistance detection method for data line connection stability according to the present invention, which presents an embodiment of the contact resistance detection method for data line connection stability according to the present invention.

[0021] In one embodiment, the contact resistance detection method for data line connection stability includes: Step S100: Obtain the excitation signal and contact end response signal of the data line under high-frequency pulse excitation, and record them as the excitation response signal to be tested and the standard excitation response signal.

[0022] The high-frequency pulse excitation can be a voltage or current excitation signal with high frequency and narrow pulse width characteristics, used to excite the transient electrical response of the data line contact. In this embodiment, the high-frequency pulse excitation can be generated by a signal generator producing a pulse sequence of specific frequency and pulse width, and applied to both ends of the data line interface. Furthermore, the high-frequency pulse excitation can simulate the dynamic electrical environment of the data line under actual high-frequency signal transmission or fast charging conditions, exciting the real transient response of the contact interface. For example, the high-frequency pulse excitation can include, but is not limited to, one or more of square wave pulse excitation, Gaussian pulse excitation, and exponentially decaying pulse excitation.

[0023] The test excitation response signal can be a time-domain response signal containing contact state information, acquired from the contact end of the data line under test under high-frequency pulse excitation. In an exemplary embodiment, the test excitation response signal can be synchronously recorded by a high-speed sampling circuit to capture the voltage / current response waveform at the contact end after the excitation signal is applied. Furthermore, the test excitation response signal can serve as raw observation data reflecting the current contact state for subsequent impedance modeling and interference analysis. Exemplarily, the test excitation response signal can include, but is not limited to, one or more of differential response signals, common-mode response signals, and reflection response signals.

[0024] The standard excitation response signal can be a reference response signal obtained from a data line contact in a known good condition (no degradation) under the same high-frequency pulse excitation conditions. In one specific embodiment, the standard excitation response signal can be obtained by exciting a standard sample and recording its response during the device manufacturing or calibration phase, storing it as a benchmark template. Furthermore, the standard excitation response signal can provide a benchmark for comparative analysis, supporting the quantitative assessment of impedance component differences. Exemplarily, the standard excitation response signal can include, but is not limited to, one or more of the following: a new line standard response signal, an aging threshold standard response signal, a temperature control standard response signal, etc.

[0025] The excitation signal and contact end response signal of the data line under high-frequency pulse excitation are acquired and recorded as the test excitation response signal and the standard excitation response signal, respectively. This can be achieved by simultaneously applying high-frequency pulse excitation and acquiring the response signals of the test and standard data lines, which are then marked and stored separately. Furthermore, this operation can be achieved by sequentially exciting the test and standard samples using the same signal source and recording the responses separately, or by using a dual-channel synchronous acquisition system to simultaneously excite and record both response signals. This allows the establishment of a measured and reference signal pair for comparative analysis, providing a data foundation for subsequent modeling.

[0026] Step S200: Perform dynamic impedance modeling decomposition on the test excitation response signal and the standard excitation response signal respectively, and extract each impedance component to be analyzed and each standard impedance component.

[0027] Dynamic impedance modeling decomposition can be a process of converting a time-domain excitation response signal into multiple analytical impedance components in the frequency domain or the time-frequency domain. In this embodiment, dynamic impedance modeling decomposition can decompose the response signal into several physically meaningful impedance components based on methods such as system identification, wavelet transform, or parametric model fitting. Furthermore, dynamic impedance modeling decomposition can decouple complex responses into independently analyzable impedance components, facilitating the identification of effective components related to the contact state. For example, dynamic impedance modeling decomposition can include, but is not limited to, one or more of the following: impedance decomposition based on the ARX model, impedance decomposition based on wavelet packets, and impedance decomposition based on the state-space model.

[0028] The impedance component to be analyzed can be an independent impedance component obtained from the dynamic impedance modeling decomposition of the excitation response signal under test. Furthermore, the impedance component to be analyzed can serve as the basic unit for contact state assessment, used for difference comparison with standard impedance components. For example, the impedance component to be analyzed can be one or more of the following: contact resistance dominant component, distributed inductance dominant component, parasitic capacitance dominant component, etc.

[0029] The standard impedance components can be reference impedance components obtained from the standard excitation response signal through the same dynamic impedance modeling decomposition process. Furthermore, the standard impedance components can serve as a comparison benchmark for the impedance components to be analyzed, used to calculate peak position time shift and amplitude fluctuation. For example, the standard impedance components can be one or more of the following: ideal contact standard components, temperature drift compensation standard components, structural consistency standard components, etc.

[0030] Dynamic impedance modeling decomposition is performed on both the test excitation response signal and the standard excitation response signal to extract each impedance component to be analyzed and each standard impedance component. This can be achieved by applying the same dynamic impedance modeling algorithm to both types of response signals, outputting the corresponding set of impedance components. Furthermore, this operation can be implemented by using wavelet packet decomposition to extract multi-scale impedance components, or by using parametric circuit model fitting to extract equivalent impedance elements. This decouples the original signal into independently evaluable physical components, supporting refined interference analysis.

[0031] Step S300: Analyze the differences between the impedance component to be analyzed and the standard impedance component regarding the time corresponding to each impedance peak position, construct the variation coefficient of the impedance peak position difference between the impedance component to be analyzed and the standard impedance component, and obtain the contact interference factor of the impedance component to be analyzed by combining the average level of the differences between the impedance component to be analyzed and the standard impedance component regarding the time corresponding to the impedance peak position.

[0032] The time corresponding to the impedance peak position can be the moment when the amplitude of a certain impedance component reaches a local maximum in the time domain or time-frequency domain representation. Furthermore, the time corresponding to the impedance peak position can reflect the response delay characteristics of the impedance component under excitation, and can be used to determine the dynamic behavior changes of the contact interface. For example, the time corresponding to the impedance peak position can be, but is not limited to, one or more of the following: the main peak time point, the secondary peak time point, and the resonant peak time point.

[0033] The coefficient of variation of impedance peak position difference can be a statistical measure of the variability of the offset of the impedance component to be analyzed from the standard impedance component at multiple peak time points. Furthermore, the coefficient of variation of impedance peak position difference can characterize the instability of peak position time offset, and is used to assist in constructing the contact interference factor. For example, the coefficient of variation of impedance peak position difference may include, but is not limited to, one or more of the following: peak position offset standard deviation coefficient, peak position offset coefficient of variation, peak position offset entropy value, etc.

[0034] The average level of the time difference corresponding to the impedance peak position can be the mean of the offsets of the impedance component to be analyzed and the standard impedance component at all corresponding peak time points. Furthermore, the average level of the time difference corresponding to the impedance peak position can reflect the overall response delay trend, serving as one of the components of the contact interference factor. For example, the average level of the time difference corresponding to the impedance peak position can be, but is not limited to, one or more of the following: arithmetic mean time offset, weighted average time offset, median time offset, etc.

[0035] The contact interference factor can be an index constructed by combining the variation coefficient of the peak position difference of the comprehensive impedance with the average level, used to quantify the degree of interference of a certain impedance component by non-contact factors. In a specific embodiment, the contact interference factor can be transformed into a single interference metric by weighted fusion or function mapping of two statistics in the time dimension. Furthermore, the contact interference factor can distinguish between response changes caused by contact degradation and pseudo-changes caused by system noise or structural fluctuations. For example, the contact interference factor may include, but is not limited to, one or more of the following: time offset interference factor, dynamic response mismatch factor, phase perturbation interference factor, etc.

[0036] The analysis examines the time-related differences between the impedance component under analysis and the standard impedance component regarding the peak positions of each impedance peak. A coefficient for the variation of the impedance peak position difference between the two components is constructed. This can be achieved by calculating the offset of each corresponding peak time point and statistically analyzing its variability to generate the coefficient. Furthermore, this operation can be implemented by calculating the ratio of the standard deviation to the mean of the peak position offset as the coefficient, or by using information entropy to measure the uncertainty of the peak position offset sequence as the coefficient. This allows for the quantification of response time jitter and reflects the dynamic characteristics of system stability affected by disturbances.

[0037] By combining the average level of the time difference between the impedance component to be analyzed and the standard impedance component regarding the impedance peak position, the contact interference factor of the impedance component to be analyzed can be obtained. This can be achieved by fusing the average level of the time difference with the variation coefficient to generate a single interference metric. Furthermore, this operation can be achieved by using a weighted linear combination: α × average level + β × variation coefficient, or by constructing a nonlinear mapping function to map the two statistics into an interference factor. This allows for a comprehensive reflection of systematic shifts and random disturbances in the time dimension, improving the accuracy of interference identification.

[0038] Step S400: Determine the consistency of impedance amplitude difference between the impedance vector to be analyzed and the standard impedance vector based on the fluctuation of the difference between the peak impedance values ​​of each impedance.

[0039] The impedance vector to be analyzed can be a multi-dimensional vector composed of all impedance components decomposed from the same measured response signal in sequence. Furthermore, the impedance vector to be analyzed can serve as input for amplitude fluctuation analysis, supporting joint evaluation of multiple components. For example, the impedance vector to be analyzed can be, but is not limited to, one or more of the following: a full-band impedance vector, a low-frequency dominant impedance vector, and a high-frequency dominant impedance vector.

[0040] The standard impedance vector can be a reference multidimensional vector composed of all standard impedance components decomposed from the standard response signal in sequence. Furthermore, the standard impedance vector can serve as a benchmark for comparing the amplitude of the impedance vector to be analyzed. For example, the standard impedance vector can be one or more of the following: nominal state impedance vector, temperature-compensated impedance vector, batch-average impedance vector, etc. The peak impedance amplitude can be the magnitude of a certain impedance component at its peak position. Furthermore, the peak impedance amplitude can reflect the energy intensity of the component, and its fluctuation can indicate changes in the conductivity of the contact interface. For example, the peak impedance amplitude can be one or more of the following: main peak amplitude, side lobe amplitude, resonant amplitude, etc.

[0041] Impedance amplitude difference consistency can be defined as the degree of consistency between the analyzed impedance vector and the standard impedance vector in terms of the fluctuations in the corresponding peak amplitude differences. In an exemplary embodiment, impedance amplitude difference consistency can be quantified by calculating statistics such as variance, correlation coefficient, or entropy of the amplitude difference sequence. Furthermore, impedance amplitude difference consistency can assess whether amplitude changes are caused by systematic contact degradation rather than random noise interference. Exemplarily, impedance amplitude difference consistency may include, but is not limited to, one or more of amplitude difference variance consistency, amplitude difference correlation consistency, and amplitude difference entropy consistency. Determining the impedance amplitude difference consistency of the analyzed impedance vector based on the fluctuations between the analyzed impedance vector and the standard impedance vector regarding the differences in the peak amplitudes can be achieved by calculating a statistical consistency index of the amplitude difference sequence of each component. Further, this operation can be achieved by calculating the reciprocal of the variance of the amplitude difference sequence as the consistency, or by calculating the correlation coefficient between the analyzed and standard vector amplitudes as the consistency, thereby identifying whether amplitude changes exhibit a systematic pattern and excluding random noise interference.

[0042] Step S500: Based on the contact interference factor and impedance amplitude difference consistency of each impedance vector to be analyzed, construct the superimposed interference characteristic value of each impedance vector to be analyzed. Based on the superimposed interference characteristic value of all impedance components to be analyzed, extract the effective impedance component vector in the test excitation response signal and obtain the dynamic impedance spectrum after contact interference elimination through dynamic fusion.

[0043] The superimposed interference feature value can be a composite interference metric constructed from the contact interference factor and the consistency of impedance amplitude differences. In one specific embodiment, the superimposed interference feature value can fuse interference indices of two orthogonal dimensions through weighted summation, product fusion, or machine learning mapping. Furthermore, the superimposed interference feature value can assign interference weights to each impedance component for adaptive suppression of ineffective components. Exemplarily, the superimposed interference feature value can include, but is not limited to, one or more of weighted superimposed interference values, logically fused interference values, and nonlinearly mapped interference values. The effective impedance component vector can be a set of impedance components that primarily reflect the contact state, retained after removing high-interference components from the test excitation response signal. In an exemplary embodiment, the effective impedance component vector can be selected / weighted based on a threshold set for the superimposed interference feature value of each component or by using a soft-weighting method to filter / weight the retained components. Furthermore, the effective impedance component vector can provide high-fidelity contact-related information for constructing a pure dynamic impedance spectrum. Exemplarily, the effective impedance component vector can include, but is not limited to, one or more of hard-threshold effective vector selection, soft-weighted effective vector, and adaptive-threshold effective vector.

[0044] A dynamic impedance spectrum can be a time-frequency or frequency-domain representation of the transient impedance characteristics of a contact end, generated by dynamically fusing effective impedance component vectors. In a specific embodiment, the dynamic impedance spectrum can integrate the effective components into a continuous spectrum through time-frequency reconstruction, spectral synthesis, or model inversion. Furthermore, the dynamic impedance spectrum can serve as a direct input for transient calculation of contact resistance, reflecting the contact electrical characteristics under actual operating conditions. For example, the dynamic impedance spectrum can include, but is not limited to, one or more of the following: time-frequency joint dynamic impedance spectrum, complex domain dynamic impedance spectrum, and parameterized dynamic impedance spectrum. Based on the contact interference factor and impedance amplitude difference consistency of each impedance vector to be analyzed, a superimposed interference feature value of each impedance vector to be analyzed is constructed. This can be an interference index that fuses the time and amplitude dimensions, generating a comprehensive interference score for each component. Furthermore, this operation can be achieved by using a product form: interference factor × (1 - consistency), or by mapping the two indices to interference levels through a pre-trained classifier, thereby achieving a unified representation of multi-dimensional interference information and supporting adaptive component selection.

[0045] Extracting the effective impedance component vector from the measured excitation response signal based on the superimposed interference characteristic values ​​of all impedance components to be analyzed can be achieved by filtering or weighting each component according to the interference characteristic values, retaining low-interference components. Furthermore, this operation can be achieved by setting a threshold to retain only components with interference characteristic values ​​below the threshold, or by inversely weighting all components according to the interference characteristic values ​​to generate a weighted effective vector. This can suppress noise and non-contact related components, improving the effective signal-to-noise ratio.

[0046] The dynamic impedance spectrum obtained after contact interference elimination through dynamic fusion can be generated by synthesizing the effective impedance component vectors through time-frequency synthesis or model reconstruction to produce a continuous impedance spectrum. Furthermore, this operation can be achieved by reconstructing the time-frequency spectrum using inverse wavelet transform, or by fitting the effective components with a parameterized model to generate a smooth impedance curve. This results in a high-fidelity, interference-resistant characterization of the contact state, providing reliable input for resistance calculations.

[0047] Step S600: Using the dynamic impedance-resistance conversion model combined with the dynamic impedance spectrum after contact interference elimination, the contact resistance of the data line contact end is transiently calculated to obtain the transient value of the data line contact resistance.

[0048] The dynamic impedance-resistance conversion model can be a mathematical or data-driven model that maps the dynamic impedance spectrum to transient contact resistance values. In an exemplary embodiment, the dynamic impedance-resistance conversion model can establish the mapping relationship between impedance and resistance based on physical equation derivation or through training with a large amount of experimental data. Furthermore, the dynamic impedance-resistance conversion model can achieve accurate inversion from high-dimensional impedance information to a single contact resistance value. Exemplarily, the dynamic impedance-resistance conversion model can include, but is not limited to, one or more of the following: physical mechanism conversion model, neural network conversion model, lookup table interpolation conversion model, etc.

[0049] The transient value of contact resistance can be the equivalent contact resistance exhibited by the data line contact at a specific moment under high-frequency pulse excitation. Furthermore, the transient value of contact resistance can directly characterize the current connection stability, enabling real-time monitoring and early warning of degradation. For example, the transient value of contact resistance may include, but is not limited to, one or more of the following: peak transient resistance, steady-state transient resistance, and decay-edge transient resistance.

[0050] By employing a dynamic impedance-resistance conversion model combined with the dynamic impedance spectrum after contact interference cancellation, transient calculations are performed on the contact resistance at the data line contact point to obtain the transient value of the data line contact resistance. This can be achieved by inputting the dynamic impedance spectrum into a pre-built conversion model and outputting the contact resistance value at the corresponding moment. Furthermore, this operation can be achieved by using the inverse solution of physical formulas (such as Z=R+jωL) to obtain the real part as the transient resistance, or by using a trained neural network model to map the impedance spectrum into resistance values, thereby achieving a precise mapping from complex impedance information to intuitive contact resistance indicators.

[0051] Taking the health monitoring of a smartphone fast charging interface as an example, the contact resistance detection method for data cable connection stability in this embodiment can be as follows: during the phone charging process, the detection module applies a high-frequency narrow pulse excitation to the USB-C interface, and simultaneously collects the response signals of the cable under test and the built-in standard cable; multiple impedance components are obtained through dynamic impedance modeling decomposition, and the contact interference factor and impedance amplitude difference consistency of each component are calculated and fused to generate superimposed interference characteristic values; based on this, components affected by electromagnetic interference or cable structure fluctuations are eliminated, and effective impedance components related only to contact oxidation or wear are retained; after generating a pure dynamic impedance spectrum through dynamic fusion, it is input into the dynamic impedance-resistance conversion model, and the transient value of contact resistance is output in real time; when the value is continuously higher than the threshold, the system prompts the user to replace the data cable to avoid charging interruption or device damage due to poor contact.

[0052] This embodiment uses high-frequency pulse excitation to stimulate the true transient response of the data line contact end, overcoming the limitation of traditional static measurement in capturing dynamic contact characteristics. Dynamic impedance modeling decomposes the response signal into multiple analyzable impedance components. Using a standard excitation response signal as a reference, it constructs the contact interference factor and impedance amplitude difference consistency from two orthogonal dimensions: impedance peak time offset and peak amplitude fluctuation. This quantifies the degree of influence of non-contact factors on each component. Furthermore, by superimposing interference characteristic values ​​and fusing multi-dimensional interference indices, it adaptively suppresses noise and structural artifacts and accurately extracts effective impedance component vectors related only to the contact state. After dynamic fusion to generate a high-fidelity dynamic impedance spectrum, it combines a specially constructed dynamic impedance-resistance conversion model to complete the high-precision inversion of transient contact resistance. This achieves high sensitivity, strong anti-interference capability, and real-time online detection of early signs of contact degradation under complex electromagnetic environments and multi-physics field coupling interference.

[0053] In one embodiment, the extraction of each impedance component to be analyzed and each standard impedance component specifically includes: Construct the excitation response vector to be analyzed and the standard excitation response vector, and use the impedance components after dynamic impedance modeling decomposition of the excitation response vector to be analyzed as the impedance components to be analyzed. The standard excitation response vector is dynamically modeled and decomposed. The discrete coefficients of each impedance vector in the standard excitation response vector are calculated. The average of the discrete coefficients of all impedance vectors in the standard excitation response vector is used as the judgment threshold. Impedance vectors with discrete coefficients greater than or equal to the judgment threshold are all recorded as standard impedance vectors.

[0054] The excitation response vector to be analyzed can be a multi-dimensional vector formed by organizing multiple response signal samples acquired by the data line under test under high-frequency pulse excitation according to time or channel dimensions. It can be used to provide structured input for dynamic impedance modeling decomposition and support multi-sample joint modeling to improve component extraction stability. In an exemplary embodiment, the excitation response vector to be analyzed can be formed by sequentially arranging the response signals acquired in multiple repeated excitations or parallel channel acquisitions into a vector form. Furthermore, the excitation response vector to be analyzed can include, but is not limited to, one or more of the following: single-sample response vector, multi-cycle averaged response vector, and multi-channel fused response vector.

[0055] A standard excitation response vector can be a reference multidimensional vector composed of multiple response signal samples acquired by a standard state data line under the same high-frequency pulse excitation conditions. It can be used as the basis for constructing a highly stable standard impedance component for subsequent discrete coefficient calculation and threshold determination. For example, the standard excitation response vector can be generated by exciting multiple standard samples or the same standard sample multiple times during the calibration phase, acquiring and integrating the responses to form a vector. In a specific embodiment, the standard excitation response vector may include, but is not limited to, batch average standard response vector, temperature-stable standard response vector, and aging zero-point standard response vector.

[0056] Constructing the excitation response vector to be analyzed and the standard excitation response vector can be achieved by organizing the excitation response signals acquired through multiple acquisitions or parallel channels into multi-dimensional vector forms. Furthermore, the construction of the excitation response vector to be analyzed and the standard excitation response vector can be achieved by continuously exciting a single data line N times and stacking the N responses column-wise into a vector, or by using a multi-channel synchronous acquisition system to combine the responses of different pins or differential pairs into a vector. This provides structured input to support subsequent modeling and statistical analysis, enhancing the robustness of impedance component extraction.

[0057] The coefficient of variation (COP) is a dimensionless statistical indicator that measures the degree of numerical fluctuation of an impedance vector across multiple measurements or samples. It is typically defined as the ratio of the standard deviation to the mean. COP can be used to quantify the stability of the impedance vector under the influence of non-contact factors (such as manufacturing tolerances or environmental disturbances) and to screen reliable standard components. In this embodiment, the COP can be calculated by taking the standard deviation and mean of the same type of impedance vector in different standard response samples, and then obtaining the ratio. For example, the COP may include, but is not limited to, amplitude COP, phase COP, and time delay COP.

[0058] The coefficients of variation of each impedance vector in the standard excitation response vector can be calculated by decomposing the standard excitation response vector through dynamic impedance modeling and calculating the coefficients of variation for each type of impedance vector in multiple samples or multiple measurements. Furthermore, the coefficients of variation of each impedance vector in the standard excitation response vector can be calculated by taking the ratio of the standard deviation to the mean of the amplitude of each impedance component in M ​​standard samples, or by calculating the coefficients of variation of the impedance components extracted from the same standard sample in K repeated measurements. This allows for the quantification of the stability of each standard impedance component, providing a basis for selecting reliable reference benchmarks.

[0059] The judgment threshold can be a filtering threshold set based on the average of the discrete coefficients of all standard impedance vectors. It can be used as an adaptive benchmark to distinguish between high-stability and low-stability standard impedance vectors. In this embodiment, the judgment threshold can be calculated as the arithmetic mean of the discrete coefficients of each impedance vector obtained after dynamic impedance modeling of the standard excitation response vector. For example, the judgment threshold may include, but is not limited to, a global mean threshold, a weighted mean threshold, and a sliding window mean threshold.

[0060] Using the average of the discrete coefficients of all impedance vectors in the standard excitation response vector as the judgment threshold can be achieved by summing the discrete coefficients of all impedance vectors and taking their average as a unified screening threshold. Furthermore, using the average of the discrete coefficients of all impedance vectors in the standard excitation response vector as the judgment threshold can be achieved by calculating the arithmetic mean of all discrete coefficients, or by calculating the mean of the remaining discrete coefficients after removing outliers. This allows for adaptive threshold setting, avoiding biases caused by manual presets and improving the objectivity of standard component selection.

[0061] Impedance vectors with discrete coefficients greater than or equal to the judgment threshold are all recorded as standard impedance vectors. Alternatively, impedance vectors with discrete coefficients not lower than the threshold can be retained as standard impedance vectors for subsequent comparisons. Furthermore, recording all impedance vectors with discrete coefficients greater than or equal to the judgment threshold as standard impedance vectors can be achieved through hard threshold screening, retaining only components with discrete coefficients ≥ the threshold, or by using a soft labeling mechanism to assign confidence weights to all components based on their discrete coefficients, but only formally recording those ≥ the threshold as standard impedance vectors. This ensures that the standard impedance components used have sufficient stability, eliminates unstable components susceptible to interference, and improves the reliability of the evaluation.

[0062] For example, in the scenario of factory quality verification of USB-C data cables, the contact resistance detection method for data cable connection stability in this embodiment can be as follows: at the production line testing station, a high-frequency pulse excitation is applied to each batch of standard good cables, and 10 repeated responses are collected to construct a standard excitation response vector; 20 impedance components are obtained through dynamic impedance modeling decomposition, and the dispersion coefficient of each component in 10 measurements is calculated; the average value of the 20 dispersion coefficients is taken as the judgment threshold; only the components with dispersion coefficients ≥ the threshold (such as low-frequency components reflecting contact resistance) are recorded as standard impedance vectors; after the response of the subsequent cable under test is decomposed in the same way, only the peak time and amplitude are compared with these high-stability standard components, thereby avoiding misjudgment caused by unstable factors such as parasitic inductance fluctuations and significantly improving the consistency of production line testing.

[0063] This embodiment achieves the technical effect of effectively eliminating unstable impedance components that are susceptible to manufacturing tolerances, environmental disturbances, or parasitic parameters by providing structured multi-sample input for dynamic impedance modeling, introducing discrete coefficients to quantify the stability of standard impedance components, and using the average discrete coefficients to adaptively generate a judgment threshold to screen high-stability standard impedance vectors. This ensures that the subsequent calculation of the consistency between the contact interference factor and the impedance amplitude difference is based on high-quality standard data, thereby improving the accuracy of interference separation, the robustness of effective component extraction, and the reliability and anti-interference capability of the final transient assessment of contact resistance.

[0064] In one embodiment, the construction of the stimulus response vector to be analyzed and the standard stimulus response vector further includes: The response signal data of the stimulus response signal to be tested are arranged in ascending order of time to form the stimulus response vector to be analyzed, and the response signal data of the standard stimulus response signal are arranged in ascending order of time to form the standard stimulus response vector.

[0065] The response signal data can be a sequence of raw voltage or current sampling points acquired from the data line contact under high-frequency pulse excitation. This sequence can be used to construct the basic unit of the test or standard excitation response signal to support subsequent vectorization and modeling. In this embodiment, the response signal data can be organized based on its accompanying timestamp or fixed sampling interval. For example, the response signal data can include, but is not limited to, one or more of the following: rising edge sampling data, steady-state sampling data, and decaying edge sampling data.

[0066] Ascending time sorting is a process of reorganizing response signal data according to their acquisition time in ascending order. It can be used to ensure that each element in the vector strictly follows the physical time causality relationship and maintain the temporal integrity of the transient response. In an exemplary embodiment, ascending time sorting can be implemented based on hardware clock sorting, trigger synchronization sorting, or interpolation alignment sorting.

[0067] The stimulus response vector to be analyzed is formed by arranging all the response signal data in the stimulus response signal in ascending time order. This can be achieved by extracting all sampling points in the stimulus response signal and arranging them in ascending time order to form a one-dimensional vector. Furthermore, this operation can be implemented by directly arranging the data according to the ADC sampling order (assuming strict timing), or, in the case of multi-channel asynchronous sampling, by first aligning the data by timestamps and then merging and sorting them. This allows for the construction of an input vector with a clear temporal causal structure, avoiding modeling distortion caused by out-of-order sampling.

[0068] Arranging all response signal data in the standard excitation response signal in ascending time order to form a standard excitation response vector can be achieved by performing the same timing sorting operation on the standard excitation response signal as on the signal under test, generating a reference vector with a consistent structure. In a specific embodiment, this operation can be achieved by directly sorting the data after acquisition using the same sampling rate and triggering mechanism as the signal under test, or by resampling historically stored standard waveforms according to a unified time grid and then sorting them, so that the signal under test and the standard vector are aligned in the same time coordinate system, supporting subsequent accurate comparison of peak positions and amplitudes.

[0069] For example, in the scenario of online health monitoring of the in-vehicle Type-C interface, the contact resistance detection method for data cable connection stability in this embodiment can be as follows: the vehicle infotainment system automatically applies a high-frequency pulse excitation and collects the contact end response each time a mobile phone is connected for charging; due to the strong noise of the in-vehicle power supply, the original response signal contains multiple discontinuous sampling segments; the system arranges all sampling points in ascending order according to high-precision timestamps to construct the excitation response vector to be analyzed; at the same time, it calls the standard excitation response vector stored at the factory and processed in the same ascending time order; the two are dynamically impedance modeled and decomposed on a completely aligned time axis to ensure that the subsequent peak time difference calculation is not affected by sampling offset, thereby accurately identifying the microsecond-level response delay caused by interface oxidation and realizing a high-anti-interference contact degradation early warning.

[0070] This embodiment constructs a vector representation with consistent structure and time alignment by uniformly arranging the response signal data in the test and standard response signals in ascending time order. This achieves the following technical effects: ensuring that the input data maintains the causality and phase consistency of the physical transient process; avoiding human distortion introduced by sampling disorder or asynchronous acquisition; making the test and standard signals comparable in a unified time coordinate system; and providing a reliable foundation for subsequent accurate extraction of impedance peak positions, calculation of peak time differences and amplitude fluctuations, etc. This improves the stability of dynamic impedance modeling, the reliability of the standard reference, and the effectiveness of the entire interference separation mechanism, ultimately supporting high-precision, strong anti-interference transient detection capability of contact resistance.

[0071] In one embodiment, the formula for calculating the coefficient of variation of the impedance peak position difference between the impedance component to be analyzed and the standard impedance component is as follows: ; In the formula, Let be the coefficient of variation of the impedance peak position difference between the i-th impedance vector to be analyzed and the j-th standard impedance vector, norm() be the exponential normalization function, and M be the number of elements in the impedance peak position difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector. and These are the k-th and (k-1)-th elements in the impedance peak difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector, respectively.

[0072] The impedance peak position difference variation coefficient can be a normalized index used to quantify the drastic change in adjacent offsets between the analyzed impedance vector and the standard impedance vector in the offset sequence at each peak time point. It can reflect the dynamic instability of peak position time offset, improve sensitivity to timing distortions caused by contact degradation, and suppress spurious changes caused by minor structural fluctuations or measurement drift. In this embodiment, the impedance peak position difference variation coefficient can be obtained by constructing a rate of change sequence based on the difference between adjacent elements in the impedance peak position difference vector, processing it with an exponential normalization function, and normalizing it according to the number of elements. Furthermore, the impedance peak position difference variation coefficient can include, but is not limited to, one or more of the following: first-order difference variation coefficient, exponential compression variation coefficient, and sequence entropy variation coefficient.

[0073] The exponential normalization function can be a nonlinear normalization mapping function used to compress the dynamic range of the input vector while maintaining its relative trend. It can be used to unify the scale and nonlinearly compress the rate of change of peak position differences, reducing the dominant influence of large outliers on the overall assessment and enhancing the ability to identify weak but consistent change patterns. In an exemplary embodiment, the exponential normalization function maps the original difference sequence to the [0, 1] interval. Exemplary examples include, but are not limited to, Softmax normalization, exponential decay normalization, and logarithmic-exponential composite normalization. The impedance peak position difference vector can be an ordered sequence composed of the time offsets of the i-th impedance vector to be analyzed and the j-th standard impedance vector at their corresponding impedance peak positions. It can be used as the basic data structure for constructing the change coefficients, carrying system response deviation information in the time dimension. In a specific embodiment, the impedance peak position difference vector can be formed by extracting the time coordinates of each corresponding peak in the two vectors, calculating the difference to obtain the offset, and arranging them in peak order. Furthermore, the impedance peak position difference vector may include, but is not limited to, the main peak offset vector, the multi-resonant peak offset vector, and the cross-frequency band peak position difference vector.

[0074] M (the number of elements in the impedance peak position difference vector) can be the number of peak time offsets contained in the impedance peak position difference vector. It can be used to normalize the length of the sum of the rates of change, eliminating evaluation bias caused by different component numbers. For example, M can be one or more of the following: single-peak counting, multi-mode peak counting, effective peak screening counting, etc. The k-th element can be the peak time offset corresponding to the k-th position in the impedance peak position difference vector. It can be used to participate in the calculation of adjacent differences, forming the basic unit of the rate of change sequence. Further, the k-th element can be, but is not limited to, the first peak offset element, the intermediate peak offset element, the last peak offset element, etc. The (k-1)-th element can be the peak time offset located one position before the k-th element in the impedance peak position difference vector. It can be used together with the k-th element to form a local rate of change, used to capture the temporal gradient of the peak position offset. In an exemplary embodiment, the (k-1)-th element can be, but is not limited to, the preceding peak offset element, the hysteresis reference element, the adjacent offset element, etc.

[0075] The coefficient of variation of the impedance peak position difference between the i-th impedance vector to be analyzed and the j-th standard impedance vector can be calculated by generating a single coefficient of variation value based on the impedance peak position difference vector between the two vectors through three steps: difference of adjacent elements, exponential normalization, and length normalization. Furthermore, this operation can be achieved by first calculating the first-order difference sequence, then summing its absolute values ​​and normalizing, or by applying a sliding window variance to the difference sequence and then normalizing. This allows for highly sensitive quantification of the dynamic characteristics of peak position time shift, enhancing the ability to identify early signs of contact degradation.

[0076] Applying an exponential normalization function to the impedance peak position difference vector can be achieved by inputting the impedance peak position difference vector or its difference result into a predefined exponential nonlinear mapping function for scaling. Furthermore, this operation can be implemented using Softmax normalization or saturated exponential compression, thereby suppressing the excessive influence of extreme offset values ​​on the overall evaluation and preserving weak but consistent variation patterns. Constructing a rate of change sequence based on the difference between adjacent elements (k-th and (k-1)-th) in the impedance peak position difference vector can be achieved by traversing the impedance peak position difference vector and calculating the differences sequentially to form a first-order difference sequence. In a specific embodiment, this operation can be achieved by calculating the center difference or using weighted differences to emphasize high-frequency variation components, thereby converting static offsets into dynamic rates of change and highlighting discontinuities or distortions in the response time series. Normalization based on the number of elements M in the impedance peak position difference vector can be achieved by dividing the cumulative amount (e.g., the sum of absolute values) of the rate of change sequence by M, achieving length-independent normalization. For example, this operation can be achieved by dividing by M for arithmetic mean normalization, or by dividing by √M for energy normalization, thereby ensuring that impedance components with different numbers of peak values ​​are comparable in terms of variation coefficients and avoiding evaluation deviations due to differences in component complexity.

[0077] Taking early oxidation detection of USB-C interface as an example, the contact resistance detection method for data cable connection stability in this embodiment can be as follows: During the detection process, the system acquires the response signals of the USB-C cable under test and the standard cable under high-frequency pulse excitation, and obtains multiple impedance components after dynamic impedance modeling decomposition. For a certain high-frequency dominant component, its five peak time points are extracted, and the offsets are calculated with the corresponding time points of the standard component to form a 5-dimensional impedance peak position difference vector. Then, the first-order difference between adjacent elements of this vector is calculated to obtain four rates of change; the dynamic range is compressed by applying an exponential normalization function to this difference sequence, and normalization is completed by dividing by M=5, finally outputting the change coefficient of impedance peak position difference. When this coefficient is significantly higher than that of normal samples, it indicates that there is transient response timing jitter caused by oxidation at the contact interface. Even if the absolute offset is small, it can be effectively identified, thereby triggering an early warning.

[0078] This embodiment provides a contact resistance detection method for data line connection stability. By calculating the variation coefficient of the impedance peak position difference between the i-th impedance vector to be analyzed and the j-th standard impedance vector, and constructing a change rate sequence based on the difference between adjacent elements in the impedance peak position difference vector, applying an exponential normalization function to the change rate sequence, and normalizing it according to the number of elements M of the impedance peak position difference vector, this method can upgrade the traditional static evaluation that only focuses on absolute time offset to a sensitive capture of the dynamic change trend of offset. The exponential normalization function effectively suppresses spurious change interference caused by small fluctuations in system structure or measurement drift, and the combination of the number of elements M ensures the evaluation consistency between impedance components of different complexities. This significantly improves the ability to distinguish real contact degradation (such as oxidation and loosening), and enhances the technical effect of the entire detection method in the robustness of complex electromagnetic environments and the sensitivity of early degradation identification.

[0079] In one embodiment, the construction of the impedance peak difference vector further includes: For each impedance vector to be analyzed and each standard impedance vector, the time corresponding to all impedance peak positions in the impedance vector to be analyzed and the standard impedance vector are arranged in the order of peak appearance to form a vector, which is denoted as the impedance peak position time vector of the impedance vector to be analyzed and the impedance peak position time vector of the standard impedance vector. The difference vector between the impedance vector to be analyzed and the standard impedance vector with respect to the impedance peak position time vector is denoted as the impedance peak position difference vector between the impedance vector to be analyzed and the standard impedance vector.

[0080] The impedance peak position time vector can be an ordered vector composed of the time values ​​corresponding to all impedance peak positions in a given impedance vector, arranged in chronological order of their appearance in the time-domain response. It can be used to structurally characterize the multi-peak response timing features of the impedance component and preserve the dynamic evolution information of the transient response. In this embodiment, the impedance peak position time vector can identify local maxima points from the impedance components after dynamic impedance modeling decomposition, extract their corresponding time coordinates, and arrange them into a vector according to their chronological order. For example, the impedance peak position time vector can include, but is not limited to, one or more of the following: the main resonance peak time vector, the multimodal reflection peak time vector, and the damped oscillation peak time vector.

[0081] The impedance peak time vector of the impedance vector to be analyzed can be a peak time sequence extracted and sorted from the impedance vector to be analyzed corresponding to the data line under test. It can be used to reflect the response delay characteristics of each impedance component under the current contact state. In an exemplary embodiment, the impedance peak time vector of the impedance vector to be analyzed can include, but is not limited to, the peak time vector of the degraded sample, the peak time vector of online monitoring, and the peak time vector of environmental disturbance. The impedance peak time vector of the standard impedance vector can be a reference peak time sequence extracted and sorted from the standard impedance vector corresponding to the standard (healthy) data line. It can be used as a benchmark template to quantify the time series deviation of the sample under test. Furthermore, the impedance peak time vector of the standard impedance vector can include, but is not limited to, the factory calibration peak time vector, the temperature control calibration peak time vector, and the batch average peak time vector.

[0082] The difference vector of impedance peak time vectors can be a new vector formed by the element-wise differences between the impedance peak time vector of the impedance vector to be analyzed and the impedance peak time vector of the standard impedance vector at corresponding positions. This can be used to eliminate common-mode system delays and highlight structural timing distortions caused by contact interface degradation. In a specific embodiment, the difference vector of impedance peak time vectors can include, but is not limited to, absolute offset difference vectors, relative timing difference vectors, and normalized time difference vectors. The times corresponding to all impedance peak positions in the impedance vector to be analyzed are arranged in the order of peak appearance to form a vector, denoted as the impedance peak time vector of the impedance vector to be analyzed. This can be achieved by traversing the time domain or time-frequency representation of the impedance vector to be analyzed, detecting local maxima, recording their time coordinates, and arranging them in ascending time order to form a vector. Furthermore, this operation can be implemented by extracting peak times through sliding window local maximum detection, or by combining wavelet transform modulus maxima tracing to determine the peak appearance order, thereby transforming unstructured peak information into an ordered sequence and preserving the dynamic evolution logic of multi-peak responses.

[0083] The time corresponding to all impedance peak positions in the standard impedance vector is arranged in the order of peak appearance to form a vector, denoted as the impedance peak time vector of the standard impedance vector. This can be achieved by performing the same peak detection and sorting process on the standard impedance vector as on the vector to be analyzed. Furthermore, this operation can directly construct the vector using a pre-stored list of calibrated peak times, or by synchronously acquiring standard samples online and generating the peak time vector in real time, thereby establishing a reference benchmark aligned with the structure of the sample to be tested.

[0084] Calculating the difference vector between the impedance vector to be analyzed and the standard impedance vector with respect to the impedance peak time vector can be achieved by subtracting the two impedance peak time vectors of the same dimension element by element. Furthermore, this operation can be performed by using Dynamic Time Warping (DTW) alignment before subtracting to handle inconsistencies in the number of peaks, or by directly subtracting point by point under a strict one-to-one correspondence. This allows for the elimination of common-mode interference such as overall system delay, focusing on local timing distortions caused by contact degradation.

[0085] The difference vector is denoted as the impedance peak difference vector between the impedance vector to be analyzed and the standard impedance vector. This can be achieved by assigning a semantic label to the calculated difference vector. Furthermore, this operation can be performed by attaching metadata (such as component indexes and frequency ranges) and storing it as a structured object, or by directly passing it to the next processing module for calculating the variation coefficients, thus completing the transformation from the raw response to structured interference characteristics.

[0086] Taking Type-C interface contact degradation monitoring as an example, the contact resistance detection method for data line connection stability in this embodiment can be implemented in a high-frequency plugging and unplugging scenario in industrial equipment. The system applies a high-frequency pulse excitation to a Type-C data line in service, acquires its response signal, and decomposes it into multiple impedance components. For an impedance component containing four obvious peaks, the system extracts its peak time [12.3ns, 28.7ns, 45.1ns, 62.0ns] to form the impedance peak time vector to be analyzed. At the same time, the standard vector calibrated by the factory [12.0ns, 28.0ns, 44.5ns, 61.2ns] is called. The peak difference vector is obtained by subtracting the peaks point by point [0.3ns, 0.7ns, 0.6ns, 0.8ns]. This vector shows that the peak offset in the later part increases, indicating that there is an accumulation of high-frequency signal propagation delay caused by fretting wear at the contact interface. This difference vector is then used to calculate the change coefficient, which is significantly higher than the threshold, triggering an early warning, while the traditional static resistance measurement still shows normal.

[0087] This embodiment achieves the following technical effects by using structured characterization of multi-peak response timing features, establishing aligned reference benchmarks, eliminating common-mode system delays, and completing interference feature transformation: effectively extracting multi-peak response timing distortion patterns caused by contact degradation, preserving transient response dynamic evolution features, suppressing non-contact related interference, and providing a physically meaningful and high signal-to-noise ratio input basis for high-precision contact state assessment. This significantly improves the sensitivity to early and weak contact degradation signs and enhances anti-interference capabilities.

[0088] In one embodiment, the formula for calculating the contact interference factor of the impedance component to be analyzed is: ; In the formula, Let N be the contact interference factor for the i-th impedance vector to be analyzed, and N be the number of standard impedance vectors. Let be the mean value of the elements in the impedance peak difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector.

[0089] The contact interference factor of the i-th impedance vector to be analyzed can be a statistical average index calculated based on multiple standard impedance vectors, characterizing the overall peak position shift of the i-th impedance component to be analyzed in the time dimension. It can be used to quantify the degree of interference of this impedance component by non-contact factors, reflecting its systematic temporal shift level away from a healthy state, and is used for subsequent interference suppression and effective component screening. In this embodiment, the contact interference factor of the i-th impedance vector to be analyzed can be calculated by taking the arithmetic mean of the element-wise averages of the impedance peak position difference vectors constructed by the i-th impedance vector to be analyzed and all N standard impedance vectors. For example, the contact interference factor of the i-th impedance vector to be analyzed can be one or more of the following, including but not limited to multi-reference mean interference factor, temporal shift robust interference factor, and cross-sample consistency interference factor.

[0090] The number N of standard impedance vectors can be the total number of standard impedance vectors involved in the contact interference factor calculation. This can be used to determine the breadth of multi-reference fusion and improve the robustness of the evaluation results to anomalies in individual standard samples. In an exemplary embodiment, the number N of standard impedance vectors can include, but is not limited to, the number of single-batch calibration samples, the number of multi-environment calibration samples, and the number of cross-temperature standard samples. The mean value of the elements within the impedance peak position difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector can be the arithmetic mean of all elements in the impedance peak position difference vector formed by the i-th impedance vector to be analyzed and the j-th standard impedance vector. This can be used to characterize the overall response delay trend of the measured component relative to the healthy state under the j-th standard reference, serving as the basic unit of multi-reference fusion. Furthermore, the mean value of the elements within the impedance peak position difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector can be obtained by summing the impedance peak position difference vector and dividing it by the number of its elements M to obtain the average level of the overall time offset. In a specific embodiment, this mean value can include, but is not limited to, the absolute offset mean, the mean after common-mode suppression, and the structural alignment offset mean.

[0091] The mean of the impedance peak position difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector is calculated. This can be achieved by taking the arithmetic mean of all elements of the constructed impedance peak position difference vector. Furthermore, this operation can be performed by directly calculating the arithmetic mean, or by using a truncated mean (averaging after removing the maximum and minimum values) to enhance the ability to resist outliers. This allows for the extraction of the overall time offset trend under this comparison, filtering out local fluctuation noise, and retaining systematic degradation characteristics.

[0092] The contact interference factor for the i-th impedance vector to be analyzed is obtained by averaging the mean values ​​of the impedance peak position difference vectors corresponding to all j (j=1 to N). Alternatively, the mean values ​​of peak position offsets obtained from N standard references can be averaged again to form the final interference factor. Furthermore, this operation can be achieved by using an equal-weighted arithmetic mean or a weighted average based on the confidence level of the standard samples (such as weighting based on the distance of calibration time). This allows the fusion of multiple standard references to weaken the bias introduced by a single standard sample due to environmental drift or measurement error, thereby improving the robustness and generalization ability of the interference factor.

[0093] Taking online monitoring of quick-change interfaces for industrial robots as an example, the contact resistance detection method for data cable connection stability in this embodiment can be applied to situations where frequent plugging and unplugging of a robot's end effector in an automated production line causes micro-motion wear on the Type-C interface. The detection system collects its high-frequency pulse response and decomposes it into a key impedance component. The system calls N=5 standard impedance vectors calibrated under different temperature and humidity conditions, and constructs impedance peak difference vectors with the component under test, respectively. The mean value of each difference vector is calculated to obtain [0.42ns, 0.38ns, 0.45ns, 0.40ns, 0.43ns], and then the average is calculated to obtain a contact interference factor of 0.416ns. This value is significantly higher than the new line threshold (0.15ns), indicating the existence of a systematic contact delay. Although not completely failed, it has entered the early deterioration stage. Based on this, the system arranges preventive replacement to avoid sudden communication interruptions.

[0094] This embodiment provides a contact resistance detection method for data line connection stability. By defining the contact interference factor as the statistical average of the impedance peak position difference under multiple standard impedance vectors, and by using N standard samples for multi-reference fusion to suppress the evaluation bias caused by environmental fluctuations or measurement noise of a single calibration sample, and by reflecting the systematic electromagnetic propagation delay caused by contact degradation rather than random disturbances through the element mean of the impedance peak position difference vector, this method can achieve the technical effects of robustly quantifying the degree of contact state deviation from the healthy baseline, enhancing the ability to identify real degradation signals, weakening the influence of spurious changes caused by distributed parameter fluctuations or external interference, and providing highly reliable input for subsequent construction of superimposed interference feature values ​​and extraction of effective impedance components.

[0095] In one embodiment, the formula for calculating the uniformity of impedance magnitude differences of the impedance vector to be analyzed is: ; In the formula, Let be the impedance magnitude difference uniformity of the i-th impedance vector to be analyzed, and exp() be an exponential function with the natural constant as the base. Let be the information entropy of the impedance amplitude difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector. Here, the amplitudes of all impedance peak positions in the i-th impedance vector to be analyzed and the j-th standard impedance vector are sorted according to their peak positions to form the impedance peak amplitude vector of the i-th impedance vector to be analyzed and the impedance peak amplitude vector of the j-th standard impedance vector. The difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector with respect to the impedance peak amplitude vector is denoted as the impedance amplitude difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector.

[0096] The impedance peak amplitude vector can be an ordered vector formed by arranging the amplitudes corresponding to all impedance peak positions in a given impedance vector according to the chronological order of peak occurrence. It can be used to achieve structural alignment between the signal to be analyzed and the standard signal in the amplitude dimension, providing a basis for constructing the difference vector. In this embodiment, the impedance peak amplitude vector can be formed by identifying local maxima points and their amplitudes in the impedance components and sorting them according to timestamps. Furthermore, the impedance peak amplitude vector can be used in conjunction with the impedance peak positions to ensure that the amplitudes are arranged in the correct temporal order; its output serves as the input for constructing the impedance amplitude difference vector. For example, the impedance peak amplitude vector can include, but is not limited to, one or more of the following: time-aligned peak amplitude vector, frequency-band filtered peak amplitude vector, and multi-scale peak amplitude vector.

[0097] An impedance amplitude difference vector can be a vector composed of the amplitude difference between the impedance vector to be analyzed and the standard impedance vector at the corresponding impedance peak positions. It can be used to characterize the degree of systematic or random deviation between two signals in the amplitude dimension, serving as input for information entropy calculation. In an exemplary embodiment, the impedance amplitude difference vector can be formed by subtracting the amplitudes of the two vectors sequentially according to the same peak position, thus creating a difference sequence. Exemplarily, the impedance amplitude difference vector can include, but is not limited to, one or more of the following: a dominant peak difference vector, a full-peak aligned difference vector, and a weighted amplitude difference vector.

[0098] Information entropy can be a statistic used to measure the uncertainty of the distribution of elements in an impedance amplitude difference vector. It can be used to quantify the degree of disorder in the amplitude differences: low entropy indicates systematic shifts (possibly due to contact degradation), while high entropy indicates random fluctuations (possibly due to noise). In one specific embodiment, information entropy can be calculated as the Shannon entropy of the difference vector based on a probability distribution model (such as a histogram or kernel density estimation). Exemplarily, information entropy can include, but is not limited to, one or more of Shannon information entropy, approximate entropy, and sample entropy.

[0099] The amplitudes of all impedance peak positions in the i-th impedance vector to be analyzed are sorted according to their peak positions to form the impedance peak amplitude vector of the i-th impedance vector to be analyzed. This can be achieved by identifying the peak positions of each component in the impedance vector to be analyzed, extracting the corresponding amplitudes, and arranging them into a vector in chronological order. Furthermore, this operation can be automatically extracted and sorted using a local maximum detection algorithm, or by pre-recording the peak timestamps of each component during the dynamic impedance modeling stage and directly extracting the amplitudes by time index. This allows for the establishment of a structured, time-consistent amplitude representation, supporting alignment and comparison with standard signals.

[0100] The amplitudes of all impedance peak positions in the j-th standard impedance vector are sorted according to their peak positions to form the impedance peak amplitude vector of the j-th standard impedance vector. This can be achieved by performing the same peak extraction and sorting process on the standard impedance vector as on the vector to be analyzed. Furthermore, this operation can be implemented by directly constructing the vector using the peak time-amplitude mapping table stored during calibration, or by performing peak detection and sorting online synchronously to adapt to environmental changes such as temperature drift. This allows the generation of a reference amplitude sequence that matches the structure of the signal to be analyzed, ensuring the physical comparability of the difference vectors.

[0101] The difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector with respect to the impedance peak amplitude vector is calculated and denoted as the impedance amplitude difference vector. This can be achieved by subtracting the two peak amplitude vectors of the same dimension element by element to obtain the difference sequence. Furthermore, this operation can be implemented by directly generating the difference vector through point-by-point subtraction, or by introducing a normalization factor (such as dividing by the standard amplitude) to generate a relative difference vector. This allows for an explicit expression of the amplitude deviation pattern of the two signals at each peak, providing a data foundation for subsequent uncertainty analysis.

[0102] Calculating the information entropy based on the impedance amplitude difference vector can be achieved by modeling the probability distribution of the difference vector and applying the Shannon entropy formula to calculate its uncertainty measure. Furthermore, this operation can be implemented by dividing the difference vector into several intervals, constructing a discrete probability distribution based on statistical frequencies, and then calculating the entropy; or by using kernel density estimation to construct a continuous probability density function and then integrating to calculate the differential entropy. This allows the distribution characteristics of the amplitude difference to be transformed into a single numerical index, distinguishing between systematic changes and random disturbances.

[0103] By taking the negative exponent of the information entropy using an exponential function, the impedance magnitude difference consistency of the i-th impedance vector to be analyzed can be obtained. This can be achieved by applying information entropy... Substitute exp(- The function outputs the uniformity value in the interval [0, 1]. Furthermore, this operation can be performed by directly using the natural exponent function exp(- Mapping can be performed using exp(-α), or a scaling parameter α can be introduced, using exp(-α) By adjusting the sensitivity, high uncertainty (high entropy) can be mapped to low consistency, and low uncertainty (low entropy) can be mapped to high consistency, thereby enhancing the effective contact variation and suppressing noise.

[0104] Taking early oxidation detection of the Type-C interface as an example, the contact resistance detection method for data line connection stability in this embodiment can be achieved by applying a high-frequency pulse excitation and acquiring the response signal during device charging. For the decomposed impedance vector to be analyzed, its peak amplitudes are extracted and sorted by time to form an impedance peak amplitude vector; simultaneously, the standard peak amplitude vector stored at the factory is called. The two are subtracted point by point to obtain the impedance amplitude difference vector. If the overall amplitude systematically decreases due to contact oxidation, the difference vector shows a concentrated negative bias and low information entropy; after exp(- After mapping, a high consistency value is obtained. This consistency value, together with the contact interference factor, participates in the calculation of superimposed interference characteristic value, so that the effective components affected by oxidation are retained, and finally accurately reflects the upward trend of contact resistance in the dynamic impedance spectrum, triggering an early warning.

[0105] This embodiment provides a contact resistance detection method for data line connection stability. It sorts the amplitudes of all impedance peak positions in the impedance vector to be analyzed and the standard impedance vector according to their peak positions to form their respective impedance peak amplitude vectors. The difference vector between the two is calculated as the impedance amplitude difference vector. Based on this difference vector, information entropy is calculated, and the impedance amplitude difference consistency is obtained by taking the negative exponent of the information entropy using an exponential function. By establishing a time-aligned amplitude structure to support physical comparability, constructing a difference vector to explicitly characterize amplitude deviation patterns, using information entropy to quantify the uncertainty of the difference distribution, and transforming uncertainty into a consistency metric through negative exponential mapping, it is possible to quantitatively evaluate the uncertainty of amplitude fluctuation patterns. This effectively distinguishes between systematic amplitude shifts caused by contact degradation and random fluctuations caused by random noise. This, combined with the contact interference factor based on peak position time shift, forms a positive interactive complement, jointly supporting the accurate construction of superimposed interference characteristic values, improving the accuracy of effective impedance component extraction, and ensuring the purity of the dynamic impedance spectrum and the reliability of transient contact resistance calculation.

[0106] In one embodiment, the superposition interference characteristic value of each impedance vector to be analyzed is the ratio of the contact interference factor of each impedance vector to the consistency of impedance amplitude difference.

[0107] The ratio can be the result of a numerical division operation between the contact interference factor and the impedance amplitude difference consistency, used to construct the superimposed interference characteristic value. In this embodiment, the ratio can be calculated by dividing the calculated contact interference factor as the numerator and the impedance amplitude difference consistency as the denominator. Furthermore, the ratio can be one or more of the following: a linear ratio, a reciprocal weighted ratio, a logarithmic domain ratio, etc.

[0108] The superimposed interference characteristic value of each impedance vector to be analyzed is calculated as the ratio of the contact interference factor to the impedance amplitude difference consistency. This can be achieved by dividing the corresponding contact interference factor of each impedance vector by its impedance amplitude difference consistency. For example, calculating the superimposed interference characteristic value of each impedance vector to be analyzed as the ratio of the contact interference factor to the impedance amplitude difference consistency can be achieved by directly performing a division operation; that is, the superimposed interference characteristic value equals the contact interference factor divided by the impedance amplitude difference consistency. In a specific embodiment, a smoothing term can also be introduced when the denominator is close to zero to prevent numerical instability; that is, the superimposed interference characteristic value equals the contact interference factor divided by the impedance amplitude difference consistency and a small constant, thereby enabling nonlinear fusion of multi-dimensional interference indicators, enhancing sensitivity to non-contact related components, and improving the discriminative power for effective signal screening.

[0109] Taking the USB-C data cable aging online monitoring system as an example, the contact resistance detection method for data cable connection stability in this embodiment can be as follows: during fast charging, the system collects the response of the cable under test in real time and decomposes it into multiple impedance components; for each component, its contact interference factor (based on peak position time offset) and impedance amplitude difference consistency (based on amplitude fluctuation stability) are calculated, and the ratio of the two is taken to obtain the superimposed interference characteristic value; if a component causes a severe amplitude fluctuation due to external electromagnetic interference (low consistency), even if the time offset is not large, its ratio is still significantly increased and is judged as a high interference component and suppressed; while the systematic peak position delay caused by contact oxidation is usually accompanied by stable amplitude change (high consistency), the ratio rises moderately but is still below the threshold, and is retained as an effective component; finally, the low ratio components are fused to generate a dynamic impedance spectrum, which is used to accurately invert transient contact resistance and realize reliable identification of real degradation.

[0110] This embodiment provides a contact resistance detection method for data line connection stability. The superimposed interference characteristic value of each impedance vector to be analyzed is the ratio of the contact interference factor to the impedance amplitude difference consistency of each impedance vector. By normalizing and fusing the contact interference factor and the impedance amplitude difference consistency, a composite interference index with adaptive discrimination capability is formed. Based on this ratio structure, higher characteristic values ​​are assigned to high interference components to achieve adaptive suppression. This method can enhance the ability to distinguish between contact state-related signals and environmental / structural artifacts, improve the purity of the dynamic impedance spectrum, ensure the accuracy and robustness of transient contact resistance calculation, and support high-precision anti-interference online connection stability assessment.

[0111] In one embodiment, the extraction of the effective impedance component vector further includes: Threshold segmentation is performed on the superimposed interference eigenvalues ​​of all impedance vectors to be analyzed to obtain the segmentation threshold. The impedance vectors to be analyzed whose superimposed interference eigenvalues ​​are lower than the segmentation threshold are recorded as effective impedance component vectors.

[0112] The superimposed interference characteristic values ​​of all impedance vectors to be analyzed can be a set of composite interference metrics generated by fusing the contact interference factor and the consistency of impedance amplitude difference corresponding to each impedance vector. This set can be used as input data for threshold segmentation to distinguish between high-interference and low-interference impedance components. In this embodiment, the superimposed interference characteristic values ​​of all impedance vectors to be analyzed can be one or more of the following: statistically distributed interference characteristic value set, clustered structure interference characteristic value set, and empirical rule-based interference characteristic value set.

[0113] The segmentation threshold can be a critical value used to binarize superimposed interference feature values. For example, the segmentation threshold can be adaptively determined based on the overall distribution characteristics of the superimposed interference feature values, such as using the Otsu method, cluster boundaries, or percentile criteria. Furthermore, the segmentation threshold can include, but is not limited to, the Otsu optimal segmentation threshold, the K-means cluster boundary threshold, and the empirical percentile segmentation threshold.

[0114] Impedance vectors whose superimposed interference eigenvalues ​​are below the segmentation threshold can be subsets of impedance vectors that, after threshold segmentation, are determined to be less affected by non-contact factors and more likely to reflect the true contact state. These subsets can be used as a direct source of effective impedance component vectors for subsequent dynamic impedance spectrum construction. In a specific embodiment, impedance vectors whose superimposed interference eigenvalues ​​are below the segmentation threshold may include, but are not limited to, low-time-disturbance impedance vectors, high-amplitude uniform impedance vectors, and two-dimensional low-interference impedance vectors.

[0115] Thresholding is performed on the superimposed interference characteristic values ​​of all impedance vectors to be analyzed to obtain the segmentation threshold. This can be achieved by analyzing the statistical distribution or structural characteristics of the superimposed interference characteristic values ​​and automatically calculating an optimal segmentation point as the threshold. Furthermore, this operation can be implemented by using the Otsu algorithm to maximize the inter-class variance to determine the segmentation threshold, or by performing K-means clustering (K=2) on the interference characteristic values ​​and taking the midpoint between the centers of the two classes as the segmentation threshold. This allows for adaptive quantization of the interference level boundary setting, avoiding subjective bias caused by manual setting.

[0116] Impedance vectors whose superimposed interference eigenvalues ​​are below a segmentation threshold are designated as valid impedance component vectors. This can be achieved by comparing the superimposed interference eigenvalue of each impedance vector with the segmentation threshold, retaining those below the threshold as valid components. In an exemplary embodiment, this operation can be achieved through hard truncation, retaining only vectors with eigenvalues ​​strictly less than the threshold, or by using soft labeling to mark vectors below the threshold as valid and assigning them unit weights for subsequent fusion. This allows for the automatic selection of high-reliability contact-related impedance information, improving the fidelity of subsequent impedance spectra.

[0117] For example, in the scenario of predicting the lifespan of data cables in industrial automated production lines, the contact resistance detection method for data cable connection stability in this embodiment can be as follows: in a high-speed automated testing station, the system applies high-frequency pulse excitation to each data cable and collects the response; after dynamic impedance modeling, the superposition interference characteristic value of each impedance vector is calculated; the segmentation threshold is automatically determined through Otsu threshold segmentation to filter out the effective impedance component vector with low interference; this process can eliminate false degradation signals caused by micro-vibration of the test fixture or environmental EMI without manual intervention, ensuring that only impedance changes caused by real contact degradation (such as terminal oxidation) are included in the evaluation, thereby accurately determining whether the product meets the connection stability standard.

[0118] This embodiment provides a contact resistance detection method for data line connection stability. It obtains a segmentation threshold by thresholding the superimposed interference characteristic values ​​of all impedance vectors to be analyzed, and records the impedance vectors whose superimposed interference characteristic values ​​are lower than the segmentation threshold as effective impedance component vectors. An adaptive threshold segmentation mechanism based on the distribution characteristics of superimposed interference characteristic values ​​automatically determines the segmentation threshold and filters effective components. This achieves objective and robust extraction of effective signals in multi-source coupled noise environments, avoiding the limitations of fixed thresholds or manual interpretation. The system can dynamically adjust the screening criteria according to the actual interference level, more reliably distinguishing impedance changes caused by contact degradation from non-contact artifacts. This further improves the purity of the dynamic impedance spectrum and the accuracy of transient inversion of contact resistance, enhancing the overall adaptability and reliability of the detection method in engineering applications.

[0119] Furthermore, to achieve the above objectives, the present invention also provides a contact resistance detection device for data line connection stability, the device comprising: a memory, a processor, and a contact resistance detection program for data line connection stability stored in the memory and executable on the processor, the contact resistance detection program for data line connection stability being configured to implement the steps of the contact resistance detection method for data line connection stability as described above.

[0120] Other embodiments or specific implementations of the contact resistance detection device for data line connection stability described in this invention can be found in the above-described method embodiments, and will not be repeated here.

[0121] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for detecting contact resistance for data cable connection stability, characterized in that, The method includes: The excitation signal and contact end response signal of the data line under high-frequency pulse excitation are acquired respectively, and recorded as the excitation response signal to be tested and the standard excitation response signal. Dynamic impedance modeling and decomposition were performed on the excitation response signal to be tested and the standard excitation response signal to be standard, and each impedance component to be analyzed and each standard impedance component were extracted. The differences between the impedance component to be analyzed and the standard impedance component with respect to the time corresponding to each impedance peak position are analyzed. The variation coefficient of the impedance peak position difference between the impedance component to be analyzed and the standard impedance component is constructed. Combined with the average level of the difference between the impedance component to be analyzed and the standard impedance component with respect to the time corresponding to the impedance peak position, the contact interference factor of the impedance component to be analyzed is obtained. Based on the fluctuation of the difference between the impedance vector to be analyzed and the standard impedance vector with respect to the peak amplitude of each impedance, the consistency of the impedance amplitude difference of the impedance vector to be analyzed is determined. Based on the contact interference factor and impedance amplitude difference consistency of each impedance vector to be analyzed, the superposition interference characteristic value of each impedance vector to be analyzed is constructed. Based on the superposition interference characteristic value of all impedance components to be analyzed, the effective impedance component vector in the test excitation response signal is extracted. The dynamic impedance spectrum after contact interference elimination is obtained by dynamic fusion. By using a dynamic impedance-resistance conversion model combined with the dynamic impedance spectrum after contact interference elimination, transient calculations are performed on the contact resistance at the data line contact end to obtain the transient value of the data line contact resistance.

2. The contact resistance detection method for data cable connection stability as described in claim 1, characterized in that, The extraction of each impedance component to be analyzed and each standard impedance component specifically includes: Construct the excitation response vector to be analyzed and the standard excitation response vector, and use the impedance components after dynamic impedance modeling decomposition of the excitation response vector to be analyzed as the impedance components to be analyzed. The standard excitation response vector is dynamically modeled and decomposed. The discrete coefficients of each impedance vector in the standard excitation response vector are calculated. The average of the discrete coefficients of all impedance vectors in the standard excitation response vector is used as the judgment threshold. Impedance vectors with discrete coefficients greater than or equal to the judgment threshold are all recorded as standard impedance vectors.

3. The contact resistance detection method for data cable connection stability as described in claim 2, characterized in that, The construction of the stimulus response vector to be analyzed and the standard stimulus response vector further includes: The response signal data of the stimulus response signal to be tested are arranged in ascending order of time to form the stimulus response vector to be analyzed, and the response signal data of the standard stimulus response signal are arranged in ascending order of time to form the standard stimulus response vector.

4. The contact resistance detection method for data cable connection stability as described in claim 1, characterized in that, The formula for calculating the coefficient of variation of the impedance peak position difference between the impedance component to be analyzed and the standard impedance component is as follows: ； In the formula, Let be the coefficient of variation of the impedance peak position difference between the i-th impedance vector to be analyzed and the j-th standard impedance vector, norm() be the exponential normalization function, and M be the number of elements in the impedance peak position difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector. and These are the k-th and (k-1)-th elements in the impedance peak difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector, respectively.

5. The contact resistance detection method for data cable connection stability as described in claim 4, characterized in that, The construction of the impedance peak difference vector further includes: For each impedance vector to be analyzed and each standard impedance vector, the time corresponding to all impedance peak positions in the impedance vector to be analyzed and the standard impedance vector are arranged in the order of peak appearance to form a vector, which is denoted as the impedance peak position time vector of the impedance vector to be analyzed and the impedance peak position time vector of the standard impedance vector. The difference vector between the impedance vector to be analyzed and the standard impedance vector with respect to the impedance peak position time vector is denoted as the impedance peak position difference vector between the impedance vector to be analyzed and the standard impedance vector.

6. The contact resistance detection method for data cable connection stability as described in claim 5, characterized in that, The formula for calculating the contact interference factor of the impedance component to be analyzed is as follows: ； In the formula, Let N be the contact interference factor for the i-th impedance vector to be analyzed, and N be the number of standard impedance vectors. Let be the mean value of the elements in the impedance peak difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector.

7. The contact resistance detection method for data cable connection stability as described in claim 1, characterized in that, The formula for calculating the consistency of impedance magnitude difference of the impedance vector to be analyzed is: ； In the formula, Let be the impedance magnitude difference uniformity of the i-th impedance vector to be analyzed, and exp() be an exponential function with the natural constant as the base. Let be the information entropy of the impedance amplitude difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector. Here, the amplitudes of all impedance peak positions in the i-th impedance vector to be analyzed and the j-th standard impedance vector are sorted according to their peak positions to form the impedance peak amplitude vector of the i-th impedance vector to be analyzed and the impedance peak amplitude vector of the j-th standard impedance vector. The difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector with respect to the impedance peak amplitude vector is denoted as the impedance amplitude difference vector between the i-th impedance vector to be analyzed and the j-th standard impedance vector.

8. The contact resistance detection method for data cable connection stability as described in claim 1, characterized in that, The superposition interference characteristic value of each impedance vector to be analyzed is the ratio of the contact interference factor of each impedance vector to the consistency of the impedance amplitude difference.

9. The contact resistance detection method for data cable connection stability as described in claim 1, characterized in that, The extraction of the effective impedance component vector further includes: Threshold segmentation is performed on the superimposed interference eigenvalues ​​of all impedance vectors to be analyzed to obtain the segmentation threshold. The impedance vectors to be analyzed whose superimposed interference eigenvalues ​​are lower than the segmentation threshold are recorded as effective impedance component vectors.

10. A contact resistance testing device for data cable connection stability, characterized in that, The device includes: a memory, a processor, and a contact resistance detection program for data line connection stability stored in the memory and executable on the processor, the contact resistance detection program for data line connection stability being configured to implement the steps of the contact resistance detection method for data line connection stability as described in any one of claims 1 to 9.

Images