Impedance measurement method and system for common mode choke

Abstract

The invention relates to the technical field of impedance measurement, in particular to an impedance measurement method and system for a common-mode inductor, and the method comprises the steps: obtaining voltage and current signals of the common-mode inductor in a test circuit, and obtaining frequency domain response atlas data containing impedance characteristics through time-frequency transformation; carrying out non-linear segmentation division on the frequency axis to form a plurality of frequency segmentation regions; extracting the phase amplitude of each segmented region, carrying out clustering segmentation, and recognizing an effective impedance frequency band; analyzing phase synchronization characteristics in the effective impedance section, and determining a common-mode resonant frequency point through clustering analysis; and fusing multi-section frequency region division, effective impedance section and resonance point information to finish accurate measurement of the impedance characteristic of the common-mode inductor. According to the method, through non-linear scanning and phase information joint analysis, the recognition precision of impedance response key characteristics in a broadband range is improved, the capturing capacity of parasitic parameters and resonance behaviors is enhanced, and efficient and high-precision comprehensive evaluation of the impedance characteristics of the common-mode inductor is achieved.

Description

Technical Field

[0001] This invention relates to the field of impedance measurement technology, and in particular to an impedance measurement method and system for common-mode inductors. Background Technology

[0002] Common-mode inductors, as key components for suppressing electromagnetic interference (EMI), are widely used in switching power supplies, frequency converters, and high-speed digital circuits. Their performance directly depends on the accurate characterization of their impedance characteristics over a wide frequency range, especially the precise measurement of the common-mode impedance amplitude and resonant frequency. Traditional impedance measurement methods often employ linear frequency scanning combined with an LCR bridge or impedance analyzer for point-by-point testing. While these methods offer high accuracy within narrow frequency bands, they struggle to balance measurement efficiency and dynamic response characteristics over a wide frequency range. Furthermore, they cannot effectively capture the nonlinear resonant behavior of common-mode inductors at high frequencies caused by parasitic parameters (such as distributed capacitance and core losses), leading to an underestimation or misjudgment of the actual impedance characteristics.

[0003] Furthermore, existing technologies often neglect the crucial role of phase information in determining the resonant point and effectively suppressing frequency bands, relying solely on amplitude-frequency characteristics for analysis. This makes them susceptible to test noise and circuit interference, leading to errors in common-mode resonant frequency identification. Simultaneously, traditional fixed-step scanning methods suffer from insufficient sampling density in critical frequency regions (such as near resonance) while wasting resources in flat response regions, lacking adaptive adjustment capabilities.

[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 an impedance measurement method and system for common-mode inductors, aiming to solve the technical problems of existing common-mode inductor impedance measurement methods, which are unable to simultaneously achieve efficient scanning and accurate identification of resonance characteristics over a wide frequency range, and lack effective utilization of phase synchronization information, resulting in inaccurate impedance characteristic characterization.

[0006] To achieve the above objectives, the present invention provides an impedance measurement method for common-mode inductors, the method comprising: The voltage and current signal data of the common-mode inductor in the test circuit are acquired, and the voltage and current signal data are processed by time-frequency transformation to obtain frequency domain response spectrum data containing impedance characteristics. The frequency axis in the frequency domain response spectrum data is nonlinearly segmented to obtain multiple frequency segment regions. The phase amplitude of the multi-segment frequency region is obtained, and the phase amplitude of the multi-segment frequency region is clustered and segmented to determine the effective impedance frequency segment; The phase synchronization characteristics of the effective impedance frequency band are obtained, and the phase synchronization characteristics of the effective impedance frequency band are clustered to determine the common-mode resonant frequency point. The common-mode inductance measurement results are determined based on the multi-segment frequency regions, the effective impedance frequency range, and the common-mode resonant frequency point.

[0007] Optionally, the step of performing time-frequency transformation processing on the voltage and current signal data to obtain frequency domain response spectrum data containing impedance characteristics includes: The voltage and current signal data are subjected to Hilbert-Huang transform to obtain time-frequency distribution data; The time-frequency distribution data is phase-normalized to obtain the frequency domain response spectrum data containing impedance characteristics.

[0008] Optionally, the step of clustering and segmenting the phase amplitudes of the multi-segment frequency regions to determine the effective impedance frequency range includes: Based on the phase amplitude of the multi-segment frequency regions, the average phase gradient of each multi-segment frequency region is calculated. Based on the average phase gradient of the multi-segment frequency regions, the phase gradient difference between the starting frequency segment and other frequency segment regions is calculated respectively; wherein, the other frequency segment regions are the remaining frequency segment regions excluding the starting frequency segment. The phase gradient difference between the starting frequency segment and other frequency segment regions is compared with a first threshold. Based on the comparison result, the multiple frequency segment regions are divided to obtain the effective impedance frequency segment.

[0009] Optionally, the step of performing cluster analysis on the phase synchronization characteristics of the effective impedance frequency band to determine the common-mode resonant frequency point includes: The phase synchronization difference between each frequency sub-segment is calculated based on the phase synchronization characteristics of the effective impedance frequency band. The phase synchronization difference between each frequency sub-segment is compared with a second threshold, and the effective impedance frequency segment is divided based on the comparison result to obtain the common-mode resonant frequency point.

[0010] Optionally, determining the common-mode inductance measurement result based on the multi-segment frequency region, the effective impedance frequency segment, and the common-mode resonant frequency point includes: The impedance magnitude is calculated based on the phase value at the cutoff frequency point in the multi-segment frequency region, the phase reference point of the effective impedance frequency segment, and the phase inflection point of the common-mode resonant frequency point. The formula for calculating the impedance magnitude is as follows: in, This represents the impedance magnitude of the common-mode inductor at frequency ω. and These are the effective values ​​of voltage and current, respectively. This indicates the phase inflection point at the common-mode resonant frequency. The phase reference point represents the effective impedance frequency range. This indicates the phase value at the cutoff frequency.

[0011] Optionally, after determining the common-mode inductance measurement result based on the multi-segment frequency region, the effective impedance frequency segment, and the common-mode resonant frequency point, the method further includes: Obtain the impedance measurement value corresponding to each nonlinear frequency scanning cycle in the frequency domain response spectrum data. If the fluctuation rate of the impedance measurement value for consecutive preset cycles is lower than the third threshold, then determine the common mode inductor impedance stability analysis result based on the impedance measurement value corresponding to each cycle.

[0012] Furthermore, to achieve the above objectives, the present invention also provides an impedance measurement system for common-mode inductors, the system comprising: The time-frequency processing module is used to acquire the voltage and current signal data of the common-mode inductor in the test circuit, and to perform time-frequency transformation processing on the voltage and current signal data to obtain frequency domain response spectrum data containing impedance characteristics. The frequency axis segmentation module is used to perform nonlinear segmentation of the frequency axis in the frequency domain response spectrum data to obtain multiple frequency segmentation regions. The phase clustering module is used to obtain the phase amplitude of the multi-segment frequency region, perform clustering and segmentation of the phase amplitude of the multi-segment frequency region, and determine the effective impedance frequency segment; The synchronization and tuning module is used to obtain the phase synchronization characteristics of the effective impedance frequency band, perform cluster analysis on the phase synchronization characteristics of the effective impedance frequency band, and determine the common-mode resonant frequency point. The integrated measurement module is used to determine the common-mode inductance measurement results based on the multi-segment frequency regions, the effective impedance frequency range, and the common-mode resonant frequency point.

[0013] Furthermore, to achieve the above objectives, the present invention also provides an impedance measurement device for common-mode inductors, the device comprising: a memory, a processor, and an impedance measurement program for common-mode inductors stored in the memory and executable on the processor, the impedance measurement program for common-mode inductors being configured to implement the steps of the impedance measurement method for common-mode inductors as described above.

[0014] Furthermore, to achieve the above objectives, the present invention also provides a medium storing an impedance measurement program for a common-mode inductor, wherein when the impedance measurement program for a common-mode inductor is executed by a processor, the program implements the steps of the impedance measurement method for a common-mode inductor as described above.

[0015] Furthermore, to achieve the above objectives, the present invention also provides a computer program product, including an impedance measurement program for common-mode inductors, wherein when the impedance measurement program for common-mode inductors is executed by a processor, it implements the steps of the impedance measurement method for common-mode inductors as described above.

[0016] This invention provides an impedance measurement method for common-mode inductors. By introducing a strategy combining nonlinear frequency scanning and phase synchronization detection, the method can adaptively and finely segment and focus key frequency regions, effectively improving the identification accuracy near common-mode resonant points. It utilizes phase amplitude clustering segmentation technology to accurately extract effective impedance frequency bands, enhancing the ability to identify nonlinear responses under the influence of parasitic parameters. Through clustering analysis of phase synchronization features, it accurately locates resonant frequency points, overcoming the problem of traditional methods being susceptible to noise interference and misjudgment due to reliance on amplitude-frequency characteristics. The overall method achieves efficient and high-precision measurement of common-mode inductor impedance characteristics over a wide frequency range, significantly improving the accuracy and reliability of impedance modeling and providing more precise technical support for electromagnetic interference filtering design. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an embodiment of the impedance measurement method for common-mode inductors 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 impedance measurement method for common-mode inductors according to the present invention, which presents an embodiment of the impedance measurement method for common-mode inductors according to the present invention.

[0021] In one embodiment, the impedance measurement method for common-mode inductors includes: Step S100: Obtain the voltage and current signal data of the common-mode inductor in the test circuit, perform time-frequency transformation on the voltage and current signal data, and obtain frequency domain response spectrum data containing impedance characteristics.

[0022] A common-mode inductor is a magnetic component used to suppress common-mode electromagnetic interference, consisting of two coils wound in the same direction and with the same number of turns on the same magnetic core. It can be used in switching power supplies, frequency converters, and high-speed digital circuits to filter common-mode noise and improve system electromagnetic compatibility. Exemplarily, the common-mode inductor can be one or more of the following: ferrite core common-mode inductors, amorphous alloy common-mode inductors, and nanocrystalline common-mode inductors. The test circuit can be a measurement loop used to connect the common-mode inductor and apply an excitation signal to obtain its electrical response. It can provide a controllable voltage and current excitation environment, enabling the common-mode inductor to exhibit measurable impedance characteristics. In an exemplary embodiment, the test circuit can consist of a signal source, a sampling resistor, a differential probe, and a ground reference, with the common-mode inductor connected in series in the excitation path.

[0023] Voltage and current signal data can be time-domain waveform sequences of the voltage across the common-mode inductor and the current flowing through it, synchronously acquired in the test circuit. These can be used as raw input data to support subsequent calculations of impedance frequency domain characteristics. Furthermore, voltage and current signal data can be acquired by synchronously recording the output signals of the voltage probe and the current transformer using a high-bandwidth oscilloscope or a dedicated data acquisition card.

[0024] Time-frequency transformation (TF-F) processing can be a mathematical process that converts time-domain voltage and current signals into frequency-domain complex impedance representations. It can be used to extract impedance amplitude and phase information at each frequency point, forming a complete frequency response spectrum. In a specific embodiment, TF-F can employ Fast Fourier Transform (FFT) or Short-Time Fourier Transform (STFT) to perform complex domain operations on the synchronously sampled signal, calculating the complex impedance at each frequency point. The frequency response spectrum data can be a complex frequency response dataset containing a triplet of frequency, impedance amplitude, and phase. It can be used to comprehensively characterize the impedance characteristics of common-mode inductors over a wide frequency range, supporting subsequent nonlinear analysis. Furthermore, the frequency response spectrum data can serve as the input data source for nonlinear segmentation, with its frequency axis structure determining the subsequent segmentation strategy.

[0025] Acquiring voltage and current signal data of the common-mode inductor in the test circuit can be achieved by applying a frequency sweep excitation signal to the test circuit and simultaneously acquiring the time-domain waveforms of the common-mode inductor terminal voltage and loop current. Furthermore, this operation can be achieved by using a differential voltage probe and a high-frequency current probe in conjunction with an oscilloscope for synchronous triggering acquisition, or by using an integrated impedance analysis chip to directly output digital voltage and current samples, thereby obtaining raw measurement data and ensuring the integrity and synchronization of subsequent frequency domain analysis. Time-frequency transformation processing of the voltage and current signal data to obtain frequency domain response spectrum data containing impedance characteristics can be performed by performing Fourier transforms on the synchronously acquired voltage and current time-domain signals respectively, calculating the complex impedance point by point. Furthermore, this operation can be achieved by using a window function-weighted FFT algorithm to reduce spectral leakage, or by using a parametric frequency response estimation method (such as the Prony method) to improve high-frequency resolution, thereby constructing a complete complex impedance frequency response spectrum that retains both amplitude and phase information.

[0026] Step S200: Nonlinear segmentation of the frequency axis in the frequency domain response spectrum data is performed to obtain multiple frequency segmentation regions.

[0027] The frequency axis can be a one-dimensional coordinate sequence representing frequency variables in the frequency domain response spectrum data, used to define the measurement frequency band and the frequency position corresponding to each data point. For example, the frequency axis can be one or more of, including but not limited to, logarithmic frequency axes, linear frequency axes, and adaptive non-uniform frequency axes. Nonlinear segmentation can be an operation of non-uniformly dividing the frequency axis based on the frequency response change rate or a preset sensitivity index. This can be used to increase the sampling density in resonant regions and sparsely segment in flat regions, optimizing the allocation of computational resources. In a specific embodiment, nonlinear segmentation can dynamically generate segment boundaries by setting a segmentation threshold based on impedance amplitude gradient, phase change rate, or a combination of both.

[0028] Multi-segment frequency regions can be sets of continuous and non-overlapping frequency sub-intervals obtained after nonlinear segmentation. These sub-intervals can provide structured analysis units for subsequent phase amplitude clustering, enabling localized fine-tuning. Furthermore, multi-segment frequency regions can include, but are not limited to, one or more of the following: resonant sensitive segments, transition response segments, and flat impedance segments. Nonlinear segmentation of the frequency axis in the frequency domain response spectrum data yields multi-segment frequency regions. This can be achieved by setting a dynamic segmentation threshold based on the gradient of impedance amplitude or phase changes, dividing the frequency axis into non-uniform sub-intervals. Further, this operation can be refined based on regions where the absolute value of the first-order phase derivative exceeds the threshold, or by using an adaptive mesh refinement algorithm (such as h-refinement) to iteratively optimize the segment boundaries. This allows for increased sampling density in critical regions (such as near resonance), improving local analytical capabilities.

[0029] Step S300: Obtain the phase amplitude of the multi-segment frequency region, cluster and segment the phase amplitude of the multi-segment frequency region, and determine the effective impedance frequency segment.

[0030] The phase amplitude can be a two-dimensional feature vector composed of the magnitude and argument of the complex impedance at each frequency point in the frequency domain response spectrum. This vector can be used to jointly characterize the impedance magnitude and phase delay characteristics, identifying the effective operating frequency band. Furthermore, the phase amplitude can serve as an input feature for clustering segmentation, and its distribution pattern determines the boundary of the effective impedance frequency band. Clustering segmentation can be an unsupervised classification process based on phase amplitude similarity across multiple frequency segments, which can be used to automatically distinguish between nonlinear response regions significantly affected by parasitic parameters and stable impedance regions. In an exemplary embodiment, clustering segmentation can employ algorithms such as K-means, DBSCAN, or Gaussian mixture models to partition the phase-amplitude feature space.

[0031] The effective impedance frequency band can be a frequency range determined to have actual filtering function after clustering and segmentation. It can be used to eliminate spurious responses caused by distributed capacitance or core loss, focusing on the true effective suppression frequency band. For example, the effective impedance frequency band can include, but is not limited to, one or more of the following: the effective segment of the primary resonance, the effective segment of the secondary resonance, and the effective segment of broadband suppression. Obtaining the phase amplitude of multi-segment frequency regions can be achieved by extracting the impedance amplitude and phase values ​​of all frequency points within each segment from the frequency domain response spectrum data. Furthermore, this operation can be achieved by directly reading the magnitude and argument of the complex impedance, or by calculating the normalized amplitude and detrended phase to enhance feature discriminative power, thereby providing structured feature input for clustering and segmentation and supporting region-level pattern recognition.

[0032] Clustering the phase amplitudes of multi-segment frequency regions to determine the effective impedance frequency range can be achieved by treating the phase amplitudes as two-dimensional feature vectors and applying unsupervised clustering algorithms to divide the effective and ineffective response regions. Furthermore, this operation can be implemented by using the DBSCAN algorithm to separate the effective and noise segments based on density, or by using a Gaussian mixture model to fit the multimodal phase amplitude distribution and selecting the frequency bands corresponding to the principal components. This allows for the automatic identification of nonlinear frequency bands affected by parasitic effects, while retaining impedance ranges with practical engineering value.

[0033] Step S400: Obtain the phase synchronization characteristics of the effective impedance frequency band, perform cluster analysis on the phase synchronization characteristics of the effective impedance frequency band, and determine the common-mode resonant frequency point.

[0034] Phase synchronization characteristics can be a pattern where the phase difference between voltage and current signals tends to stabilize or exhibits a specific regularity within the effective impedance frequency range. This pattern can be used as a criterion for resonant points, avoiding misjudgments caused by relying solely on peak amplitude. In a specific embodiment, phase synchronization characteristics can be extracted by calculating the rate of change of phase difference between adjacent frequency points or the extreme points of the phase derivative. Cluster analysis can be a statistical analysis process of pattern recognition and grouping of phase synchronization characteristics. It can be used to accurately locate common-mode resonant frequency points and improve the robustness of resonance identification. For example, cluster analysis can be implemented by clustering based on the feature point density near the zero point of the phase derivative, or by using phase synchronization stability indicators (such as phase variance) for hierarchical clustering. The common-mode resonant frequency point can be the frequency position corresponding to the parallel resonance formed by the common-mode inductor's own inductance and distributed capacitance. It can be used to determine the filter's cutoff characteristics and maximum impedance point, and is a key parameter in EMI design. Furthermore, the common-mode resonant frequency point can include, but is not limited to, one or more of the following: fundamental mode resonant point, higher-order parasitic resonant point, and core loss-dominated resonant point.

[0035] Obtaining phase synchronization characteristics within the effective impedance frequency range can be achieved by calculating the stability index or synchronization measure of phase changes within this range. Further, this can be accomplished by calculating the phase standard deviation within a sliding window (low variance regions are considered synchronization zones) or by solving for the phase derivative with respect to frequency (extracting the region near zero as a synchronization feature set). This allows for the extraction of key phase characteristics for resonant point identification, reducing the impact of amplitude noise interference. Clustering the phase synchronization characteristics within the effective impedance frequency range to determine the common-mode resonant frequency can be performed by spatially clustering the phase synchronization feature point set and identifying the densest or most representative frequency point as the resonant point. Further, this can be achieved by using the MeanShift algorithm to find the density peak of the zero point of the phase derivative, or by using hierarchical clustering to merge neighboring synchronization feature points and taking the cluster center as the resonant frequency. This allows for precise location of the resonant frequency, avoiding misjudgments under noise conditions using the traditional amplitude peak method.

[0036] Step S500: Determine the common-mode inductance measurement results based on the multi-segment frequency region, the effective impedance frequency segment, and the common-mode resonant frequency point.

[0037] The common-mode inductor impedance measurement results can be a complete impedance characteristic description generated by integrating multiple frequency segmentation regions, effective impedance frequency bands, and common-mode resonant frequency points. This can be used to provide high-fidelity modeling basis for electromagnetic interference filter design and improve the accuracy of system EMC performance prediction. Determining the common-mode inductor impedance measurement results based on multiple frequency segmentation regions, effective impedance frequency bands, and common-mode resonant frequency points can integrate the segmentation structure, effective frequency band boundaries, and resonant point locations to reconstruct a high-fidelity impedance curve. Furthermore, this operation can be achieved by retaining the original data in the effective segment, performing smooth interpolation or confidence marking in the invalid segment, or constructing a local high-resolution sub-model centered on the resonant point and using low-complexity fitting for the remaining regions. This allows for the output of a full-band impedance characterization that balances efficiency and accuracy, supporting high-reliability modeling.

[0038] Taking the design of an EMI filter for a switching power supply as an example, the impedance measurement method for common-mode inductors in this embodiment can be to measure the impedance of a ferrite common-mode inductor in the 500kHz–30MHz frequency band. The system first acquires its voltage and current time-domain signals and obtains a complex frequency response spectrum through FFT. Then, the frequency axis is nonlinearly divided according to the phase change rate, generating dense sub-segments in the resonant region around 2.1MHz. Next, DBSCAN clustering is performed on the phase amplitude of each sub-segment to identify the 1.8–2.4MHz as the main effective impedance segment. Feature points with phase derivatives close to zero are extracted in this segment, and 2.15MHz is determined as the common-mode resonant frequency point through MeanShift clustering. Finally, the segmented structure and effective segment information are fused to output an impedance curve containing the accurate resonant point and effective suppression bandwidth, which is used to guide the matching of filter LC parameters.

[0039] This embodiment provides an impedance measurement method for common-mode inductors. It acquires voltage and current signal data of the common-mode inductor in a test circuit and performs time-frequency transformation to obtain frequency domain response spectrum data containing impedance characteristics. The frequency axis in the frequency domain response spectrum data is nonlinearly segmented to obtain multiple frequency segment regions. The effective impedance frequency range is determined by acquiring the phase amplitude of each frequency segment region and performing clustering analysis on the phase amplitude. The common-mode resonant frequency point is determined by acquiring the phase synchronization characteristics of the effective impedance frequency range and performing clustering analysis on the phase synchronization characteristics. By determining the common-mode inductor impedance measurement result based on the multiple frequency segment regions, the effective impedance frequency range, and the common-mode resonant frequency point, this method achieves efficient, robust, and high-precision characterization of the common-mode inductor impedance characteristics (including amplitude, phase, and resonant point) over a wide frequency range, significantly improving its reliability and practicality in EMI filter design and electromagnetic compatibility modeling.

[0040] In one embodiment, the voltage and current signal data undergo time-frequency transformation processing to obtain frequency domain response spectrum data containing impedance characteristics, including: The voltage and current signal data are subjected to Hilbert-Huang transform to obtain time-frequency distribution data; The time-frequency distribution data is phase-normalized to obtain frequency domain response spectrum data containing impedance characteristics.

[0041] The Hilbert-Huang transform (HWH) is a time-frequency analysis method applicable to nonlinear and non-stationary signals. It consists of Empirical Mode Decomposition (EMD) and the Hilbert transform, and can be used to extract instantaneous frequency and amplitude information from voltage and current signals, effectively characterizing the nonlinear dynamic response of a common-mode inductor under the influence of parasitic parameters over a wide bandwidth. In this embodiment, the HWH first adaptively decomposes the original signal into several intrinsic mode functions (IMFs) through EMD, and then performs a Hilbert transform on each IMF to obtain an analytic signal and calculate the instantaneous frequency and amplitude. For example, the HWH can include, but is not limited to, one or more of Empirical Mode Decomposition (EMD), Ensemble Empirical Mode Decomposition (EEMD), and Complementary Ensemble Empirical Mode Decomposition (CEEMDAN). The time-frequency distribution data can be a nonlinear time-frequency representation containing four-dimensional information of time, frequency, amplitude, and instantaneous phase obtained after the HWH, which can be used to accurately characterize the local nonlinear behavior of the common-mode inductor impedance as frequency changes, especially suitable for the resonant transition region and regions with significant parasitic effects. Furthermore, the time-frequency distribution data serves as input data for phase normalization processing, and its phase structure directly affects the consistency of the subsequent frequency domain response spectrum.

[0042] The voltage and current signal data are subjected to Hilbert-Huang transform to obtain time-frequency distribution data. This can be achieved by performing Empirical Mode Decomposition (EMD) on the synchronously acquired voltage and current time-domain signals separately, and then performing Hilbert transform on each intrinsic mode function (IMF) to construct their respective instantaneous frequency-amplitude-phase relationships. In an exemplary embodiment, this operation can be achieved by using EEMD to suppress mode aliasing and enhance the physical meaning of IMF components, or by jointly performing multivariate EMD (MEMD) on the voltage and current signals to ensure the consistency of the decomposition scale before performing the Hilbert transform. This overcomes the limitation of the traditional Fourier transform on the stationarity assumption and effectively captures the transient impedance changes of the common-mode inductor near the resonant point caused by core nonlinearity or distributed capacitance coupling.

[0043] Phase normalization can be used to systematically correct the phase components in time-frequency distribution data, eliminating fixed delays or phase shifts introduced by the acquisition link. This ensures the physical comparability of phases at different frequencies and improves the accuracy of phase synchronization feature extraction. In one specific embodiment, phase normalization can compensate for the original phase through reference channel calibration or based on a known system group delay model, aligning the phase zeros with the excitation signal. For example, phase normalization can include, but is not limited to, group delay compensation normalization, reference signal alignment normalization, and minimum phase reconstruction normalization. Phase normalization of time-frequency distribution data yields frequency domain response spectrum data containing impedance characteristics. This can be achieved by extracting the voltage and current phases corresponding to each frequency point from the time-frequency distribution data, calculating the phase difference, and then applying system-level phase correction to form a normalized complex impedance representation. Furthermore, this operation can be achieved by establishing a phase error model using open / short circuit calibration data to perform reverse compensation on the measured phase, or by reconstructing the theoretical phase from the amplitude response based on the minimum phase system assumption and normalizing it based on this. This eliminates the inherent phase distortion of the test system, allows the phase synchronization characteristics to truly reflect the physical characteristics of the device, and provides high-fidelity input for subsequent cluster analysis.

[0044] Taking the verification of a common-mode filter in a high-frequency inverter as an example, the impedance measurement method for common-mode inductors in this embodiment can be to test a nanocrystalline common-mode inductor in the 1–100MHz frequency band. The traditional FFT method only shows a wide peak at 35MHz, which cannot distinguish whether it is a true resonance. This scheme performs Hilbert-Huang transform on the acquired voltage and current signals and finds an IMF component with instantaneous frequency clustering and abrupt phase change near 34.2MHz. Then, group delay compensation phase normalization is performed on the time-frequency distribution data to obtain a phase-continuous impedance spectrum. The spectrum clearly shows that the phase derivative crosses zero and the amplitude rises sharply at 34.3MHz, which is accurately identified as the main resonance point by subsequent cluster analysis, avoiding misjudgment caused by phase distortion in traditional methods.

[0045] This embodiment provides an impedance measurement method for common-mode inductors. It obtains time-frequency distribution data by performing a Hilbert-Huang transform on voltage and current signal data, and then performs phase normalization on this data to obtain frequency domain response spectrum data containing impedance characteristics. By replacing the traditional linear time-frequency transform with the Hilbert-Huang transform, the nonlinear and non-stationary impedance response of the common-mode inductor caused by parasitic parameters over a wide frequency range is effectively analyzed. Phase normalization eliminates system phase distortion, constructing a phase-consistent, high-fidelity frequency domain response spectrum. This method retains amplitude information while accurately presenting phase synchronization characteristics, providing high-quality foundational data for nonlinear frequency segmentation, effective impedance band clustering, and resonant point location. This significantly improves the modeling accuracy and noise robustness of the full-band impedance characteristics of common-mode inductors, supporting more reliable EMI filter designs.

[0046] In one embodiment, the phase amplitude of the multi-segment frequency region is clustered and segmented to determine the effective impedance frequency segment, including: Based on the phase amplitude of the multi-segment frequency regions, the average phase gradient of each multi-segment frequency region is calculated. The average phase gradient can be the average rate of phase change with frequency within a single frequency segment region, and can be used to characterize the dynamic intensity of the phase response within that frequency band. In this embodiment, the average phase gradient can be calculated by performing first-order difference or linear fitting on the phase values ​​of each frequency point within the frequency segment region and calculating the mean of its slope. Furthermore, the average phase gradient can serve as a criterion feature to distinguish between effective and ineffective impedance frequency bands, reflecting the nonlinear phase behavior caused by parasitic parameters. Based on the phase amplitude of multiple frequency segment regions, the average phase gradient of each frequency segment region is calculated. This can be done by calculating the average rate of phase change with frequency using the phase values ​​of each frequency point within each frequency segment region. For example, this operation can be achieved by fitting the phase-frequency curve using least-squares linear regression and taking the slope as the average phase gradient, or by taking the mean of the absolute values ​​after phase difference between adjacent frequency points as an approximation of the gradient. This allows for the extraction of dynamic phase characteristics of each frequency band, providing a structured indicator for subsequent difference comparisons.

[0047] Based on the average phase gradient of the multi-segment frequency regions, the phase gradient difference between the starting frequency segment and other frequency segment regions is calculated respectively; where other frequency segment regions are the remaining frequency segment regions excluding the starting frequency segment. The starting frequency segment can be the first sub-interval at the lowest frequency end of a multi-segment frequency region, serving as a reference segment for phase gradient comparison. Its phase change is relatively gentle and exhibits typical inductive characteristics. In an exemplary embodiment, the starting frequency segment can be the low-frequency inductive operating region of a common-mode inductor. Furthermore, the average phase gradient of the starting frequency segment can be used to calculate the phase gradient difference with other frequency segment regions, forming the basis for the segmentation criterion. Other frequency segment regions can be the set of all remaining frequency sub-intervals in a multi-segment frequency region excluding the starting frequency segment. These can be used as comparison objects to evaluate their dynamic phase deviation relative to the reference segment. In a specific embodiment, other frequency segment regions can include, but are not limited to, one or more of the following: a resonant transition region, a parasitic capacitance-dominated region, and a high-frequency loss attenuation region.

[0048] The phase gradient difference can be the absolute difference between the average phase gradient of other frequency segment regions and the average phase gradient of the starting frequency segment. It can be used to quantify the degree of difference in phase response characteristics of different frequency bands and serve as a numerical basis for clustering and segmentation. In this embodiment, the phase gradient difference can be obtained by subtracting the average phase gradients segment by segment, reflecting the degree of local phase nonlinearity deviation from the reference. Based on the average phase gradients of multiple frequency segment regions, the phase gradient difference between the starting frequency segment and other frequency segment regions is calculated respectively. This can be achieved by using the average phase gradient of the starting frequency segment as the reference and successively calculating the absolute difference with the average phase gradients of each other segment. Furthermore, this operation can be achieved by directly calculating the absolute value of the difference between the two gradient values, or by calculating the relative difference after normalization to eliminate the influence of dimensions, thereby establishing a unified comparison scale and quantifying the phase behavior offset of each frequency band relative to the low-frequency reference.

[0049] The phase gradient difference between the starting frequency segment and other frequency segment regions is compared with a first threshold. Based on the comparison result, multiple frequency segment regions are divided to obtain the effective impedance frequency segment.

[0050] The first threshold can be a preset phase gradient difference judgment threshold, which can be used to distinguish between significant changes in phase behavior and normal fluctuations. In an exemplary embodiment, the first threshold can be set based on historical measurement data statistics, or dynamically adjusted according to the overall phase gradient distribution through an adaptive algorithm. Furthermore, the first threshold can control the identification sensitivity of the effective impedance frequency band, avoiding missegmentation caused by noise or small fluctuations. Comparing the phase gradient difference between the starting frequency band and other frequency segment regions with the first threshold can be used to determine whether each phase gradient difference exceeds the preset first threshold. For example, this operation can use a fixed empirical threshold for hard decision, or combine a sliding window with local statistical dynamic adjustment of the threshold for soft decision, thereby achieving automatic discrimination: a difference less than the threshold is considered as similar phase behavior and belongs to the effective impedance frequency band; otherwise, it is considered as an abnormal or invalid segment.

[0051] Based on the comparison results, multiple frequency segments are divided to obtain the effective impedance frequency segment. This can be achieved by retaining all frequency segments whose phase gradient differences do not exceed a first threshold and merging them into an effective impedance frequency segment. Furthermore, this operation can merge segments that continuously meet the conditions into a single effective segment, or it can perform neighborhood expansion or confidence filtering on isolated short segments that meet the conditions. This allows for the precise selection of frequency bands whose phase behavior is consistent with the low-frequency inductive region, eliminating spurious response regions affected by parasitic resonances or losses.

[0052] Taking the verification of EMI suppression components for high-speed digital interfaces as an example, the impedance measurement method for common-mode inductors in this embodiment can be used to measure the impedance of a common-mode inductor used in a USB 3.2 interface within the range of 1MHz–500MHz. The system first performs nonlinear segmentation to obtain 12 frequency sub-segments. The average phase gradient of each segment is calculated, with the gradient of the initial frequency segment (1–5MHz) being -0.8° / MHz. In the remaining segments, the gradient of the 5–45MHz segment is -1.1° / MHz, with a difference of 0.3° / MHz; while the 60–120MHz segment experiences a steep phase change due to distributed capacitance resonance, resulting in a gradient of +2.5° / MHz, with a difference of 3.3° / MHz. Setting a first threshold of 1.0° / MHz, the former is included in the effective impedance frequency range, while the latter is excluded. The final effective range covers 1–45MHz, accurately reflecting the actual suppression bandwidth of the device and avoiding misjudging parasitic resonance peaks as effective filtering regions.

[0053] This embodiment provides an impedance measurement method for common-mode inductors. It calculates the average phase gradient of each segment based on the phase amplitude of multiple frequency segments, and calculates the phase gradient difference between the starting frequency segment and other frequency segments. This difference is then compared with a first threshold to segment the frequency region, achieving accurate identification of effective impedance frequency segments with significant impedance effects. This method utilizes the local variation characteristics of phase information for fine-grained discrimination, effectively capturing nonlinear responses caused by parasitic parameters. It significantly improves sensitivity and robustness to frequencies near resonance and those dominated by parasitic effects, avoiding misjudgments of effective frequency segments due to noise interference or sampling redundancy in flat response regions. This provides a reliable data foundation for high-precision resonant point location and impedance modeling, ultimately supporting wideband, high-efficiency, and noise-resistant common-mode inductor impedance measurement.

[0054] In one embodiment, cluster analysis is performed on the phase synchronization characteristics of the effective impedance frequency range to determine the common-mode resonant frequency point, including: Calculate the phase synchronization difference between each frequency sub-segment based on the phase synchronization characteristics of the effective impedance frequency band; The phase synchronization difference between each frequency segment is compared with the second threshold. Based on the comparison result, the effective impedance frequency segment is divided to obtain the common mode resonant frequency point.

[0055] The frequency segments can be fine-grained continuous frequency intervals further subdivided within the effective impedance frequency range, used for local phase synchronization analysis. Frequency segments can provide high-resolution analysis units, supporting fine detection of phase abrupt change regions near the resonance point. In this embodiment, frequency segments can include, but are not limited to, one or more of the following: pre-resonance transition segment, resonance center segment, and post-resonance attenuation segment. The phase synchronization difference can be a measure of the difference in phase synchronization characteristics between adjacent or specified frequency segments, reflecting the degree of change in local phase consistency. The phase synchronization difference can be used as a key criterion for identifying phase abrupt change regions, used to locate nonlinear inflection points caused by common-mode resonance. Furthermore, the phase synchronization difference can be obtained by calculating the absolute difference of phase synchronization indices (such as phase variance, mean phase derivative, or cross-correlation coefficient) within two frequency segments. The second threshold can be a preset phase synchronization difference judgment threshold, used to distinguish between normal phase transitions and phase abrupt changes caused by resonance. The second threshold can be used to control the sensitivity and noise immunity of resonance point identification, avoiding misjudging small fluctuations as resonance. In one exemplary embodiment, the second threshold can be set based on the statistical characteristics of the phase synchronization distribution within the effective impedance frequency band, or dynamically adjusted through an adaptive algorithm.

[0056] Calculating the phase synchronization difference between frequency sub-segments based on the phase synchronization characteristics of the effective impedance frequency band can be achieved by dividing the effective impedance frequency band into multiple frequency sub-segments, extracting their phase synchronization characteristics, and calculating the synchronization difference between any two sub-segments (usually adjacent sub-segments). Furthermore, this operation can be achieved by using a sliding window to divide the effective segment into overlapping sub-segments of equal width and calculating the difference in the mean phase derivatives of adjacent windows, or by dividing non-uniform sub-segments based on the curvature change of the phase-frequency curve and then calculating the difference in phase synchronization indices (such as phase standard deviation) between sub-segments. This allows for the quantification of local changes in phase consistency in the frequency domain, highlighting the nonlinear abrupt changes near the resonant point.

[0057] Comparing the phase synchronization difference between each frequency segment with a second threshold can be done by determining whether the phase synchronization difference of each pair of frequency segments exceeds the second threshold. For example, this operation can be achieved by using a fixed engineering experience threshold for hard threshold determination, or by dynamically setting the second threshold by combining the median and interquartile range of the local difference distribution. This allows for the automatic identification of boundaries where significant abrupt changes in phase synchronization occur, thus locking in potential resonant regions.

[0058] Based on the comparison results, the effective impedance frequency band is segmented to obtain the common-mode resonant frequency. This can be achieved by considering the boundary of adjacent sub-segments where the phase synchronization difference exceeds a second threshold as a phase abrupt change point, and taking the frequency corresponding to the midpoint or maximum difference of this boundary as the common-mode resonant frequency. In a specific embodiment, this operation can be achieved by selecting the frequency corresponding to the extreme point of the second derivative of the phase in the sub-segment pair with the difference exceeding the threshold as the resonant point, or by clustering multiple threshold boundaries and taking the cluster center frequency as the final common-mode resonant frequency. This allows for precise location of the phase nonlinearity transition frequency caused by parasitic parameter resonance, avoiding amplitude noise interference.

[0059] Taking the EMC verification of an on-board charger (OBC) as an example, the impedance measurement method for common-mode inductors in this embodiment can be used to measure the common-mode inductor used in the OBC input of an electric vehicle within the range of 150kHz–30MHz. After preprocessing, the effective impedance frequency range has been determined to be 1.2–8.5MHz. The system divides this range into 0.5MHz wide frequency sub-segments and calculates the average first-order derivative of phase with respect to frequency within each sub-segment as a synchronization characteristic. It was found that the difference in the average phase derivative between the 6.8–7.3MHz and 7.3–7.8MHz sub-segments reached 4.2° / MHz², far exceeding the set second threshold (1.5° / MHz²), indicating a phase abrupt change. 7.3MHz was chosen as the common-mode resonant frequency point. This result highly matches the measured zero point of the S-parameters by the network analyzer, while the traditional amplitude peak method misjudged it as 7.6MHz due to the presence of multiple parasitic peaks nearby.

[0060] This embodiment provides an impedance measurement method for common-mode inductors. It calculates the phase synchronization difference between frequency segments based on the phase synchronization characteristics of the effective impedance frequency band. This phase synchronization difference is compared with a second threshold. Based on the comparison result, the effective impedance frequency band is segmented to obtain the common-mode resonant frequency point. By further subdividing the selected effective impedance frequency band into frequency segments and calculating the phase synchronization difference between segments using phase synchronization characteristics, the method quantifies the phase consistency changes of voltage and current signals at different frequency points. Near the common-mode resonant point, the parasitic capacitance and inductance resonate, causing the system to exhibit strong nonlinearity and phase relationship... A sudden change occurs, causing a significant decrease in the synchronization of adjacent frequency segments, forming a clear inflection point of difference. By introducing a preset second threshold to discriminate this difference, the boundary region where the phase synchronization changes drastically can be adaptively identified, thereby accurately segmenting and locating the common-mode resonant frequency point. This achieves the technical effect of abandoning the traditional approach of relying solely on the peak value of impedance amplitude to determine the resonant point, instead using the structural abrupt change in phase synchronization as a criterion, effectively avoiding false peak misjudgments caused by noise, interference, or flat regions in the amplitude-frequency response, significantly improving the robustness and accuracy of resonant frequency identification, and providing key parameter support for high-fidelity impedance modeling and EMI filter optimization design.

[0061] In one embodiment, the common-mode inductance measurement result is determined based on a multi-segment frequency region, an effective impedance frequency segment, and a common-mode resonant frequency point, including: The impedance magnitude is calculated based on the phase value at the cutoff frequency point in the multi-segment frequency region, the phase reference point of the effective impedance frequency segment, and the phase inflection point of the common-mode resonant frequency. The formula for calculating the impedance magnitude is as follows: in, This represents the impedance magnitude of the common-mode inductor at frequency ω. and These are the effective values ​​of voltage and current, respectively. This indicates the phase inflection point at the common-mode resonant frequency. The phase reference point represents the effective impedance frequency range. This indicates the phase value at the cutoff frequency.

[0062] The phase value at the cutoff frequency point can be the impedance phase value corresponding to the boundary of each sub-segment (i.e., the cutoff frequency point) in a multi-segment frequency region. It can be used to characterize the transition characteristics of the impedance phase at the frequency segment boundaries and to assist in correcting boundary effects in impedance magnitude calculation. In an exemplary embodiment, the phase value at the cutoff frequency point can be extracted from the frequency response spectrum data, specifically the phase values ​​at the endpoints of each frequency sub-segment obtained through nonlinear segmentation. Furthermore, the phase value at the cutoff frequency point can be used in conjunction with the phase reference point and phase inflection point to participate in the correction calculation of the impedance magnitude. The phase reference point can be a representative stable phase reference value selected within the effective impedance frequency band. It can be used as a benchmark for phase characteristics within the main suppression frequency band to calibrate the phase shift in impedance magnitude calculation. For example, the phase reference point can be determined by cluster centers, mean, or median within the effective impedance frequency band. In a specific embodiment, the phase reference point and the phase value at the cutoff frequency point can jointly reflect the overall trend of the impedance phase within the effective operating frequency band.

[0063] A phase inflection point can be the phase value corresponding to a significant inflection point (such as a derivative extremum or abrupt curvature change) in the impedance phase curve near the common-mode resonant frequency. It can be used to accurately identify the phase state at the time of resonance, improving the interpretability of the physical meaning of the resonant point and the accuracy of modeling. In this embodiment, the phase inflection point can be located by analyzing the first or second derivative of the phase with respect to frequency within the effective impedance frequency range to pinpoint the point of maximum phase change rate or curvature change. Furthermore, the phase inflection point can be compared with a phase reference point to identify phase anomalies caused by resonance. The effective voltage and current values ​​can be the root mean square values ​​of the common-mode inductor terminal voltage and the flowing current measured in the test circuit. They can be used as the basic amplitude input for impedance magnitude calculation, reflecting the energy level of excitation and response. In an exemplary embodiment, the effective voltage and current values ​​are obtained by RMS calculation from synchronously acquired voltage and current signal data. Exemplarily, the effective voltage and current values ​​constitute the original amplitude basis for impedance magnitude calculation and participate in the correction calculation together with the three phase feature points.

[0064] The impedance magnitude can be the amplitude of the complex impedance presented by a common-mode inductor at a specific frequency ω, and can be used to characterize its ability to suppress common-mode current. In this embodiment, the impedance magnitude is calculated based on the voltage-to-current RMS ratio and modified by incorporating three key phase characteristic points. Furthermore, the impedance magnitude, as a core parameter in EMI filter design, directly affects the filter's insertion loss and noise suppression performance. The frequency ω can be an angular frequency variable used to represent the specific operating frequency point in impedance measurement, and can be used as the independent variable in calculating the impedance magnitude, relating the time-domain excitation and frequency-domain response. For example, the frequency ω can include, but is not limited to, one or more of the fundamental angular frequency, resonant angular frequency, and parasitic resonant angular frequency.

[0065] The impedance magnitude can be calculated based on the phase value at the cutoff frequency in a multi-segment frequency region, the phase reference point of the effective impedance frequency band, and the phase inflection point of the common-mode resonant frequency. This can be achieved by introducing three physically meaningful phase characteristic points as correction factors based on known effective voltage and current values, and substituting them into a specific formula to calculate the impedance magnitude at frequency ω. Furthermore, this operation can be implemented by using the three phase characteristic points as weighting coefficients in a nonlinear regression model of the impedance magnitude, or by constructing a lightweight correction network with phase characteristics as input and impedance magnitude as output for real-time correction. This improves the accuracy of impedance amplitude estimation and compensates for the nonlinear phase distortion caused by parasitic parameters.

[0066] For example, in the scenario of verifying a common-mode filter for a high-frequency switching power supply, the impedance measurement method for common-mode inductors in this embodiment can be implemented when performing wideband impedance characterization on a common-mode inductor used in a 1MHz switching power supply. The system has already completed nonlinear segmentation, effective frequency band clustering, and resonant point location. In the final impedance magnitude calculation stage, the algorithm extracts the phase value of the cutoff frequency at 2.4MHz (upper limit of the effective band) as +78°, determines the phase reference point as +85° within the 1.9–2.3MHz effective band, and identifies the phase inflection point as +90° at the 2.15MHz resonant point. Combining the effective values ​​of voltage and current at this frequency (V = 1.2V, I = 15mA), the three phase characteristics are substituted into the correction formula to calculate an impedance magnitude of 82Ω. This more accurately reflects the actual resonant peak value than the traditional result of only using V / I = 80Ω, effectively compensating for the phase lag effect caused by core losses.

[0067] This embodiment provides an impedance measurement method for common-mode inductors. It calculates the impedance magnitude based on the phase value at the cutoff frequency point in a multi-segment frequency region, the phase reference point of the effective impedance frequency band, and the phase inflection point of the common-mode resonant frequency. By introducing the phase value at the cutoff frequency point to correct for frequency band boundary effects, using the phase reference point to calibrate the phase shift within the main suppression frequency band, and combining the phase inflection point to accurately identify the phase state at the resonance moment, and by integrating the effective values ​​of voltage and current as the amplitude basis, it can achieve the technical effects of improving the accuracy of impedance amplitude estimation, compensating for nonlinear phase distortion caused by parasitic parameters, and overcoming the problems of traditional amplitude-frequency response being susceptible to noise interference and unable to distinguish between true resonance and spurious peaks. Thus, it achieves a complete and robust characterization of the electromagnetic characteristics of common-mode inductors over a wide frequency range, providing more reliable parameter basis for the optimized design of EMI filters.

[0068] In one embodiment, after determining the common-mode inductance impedance measurement result based on the multi-segment frequency region, the effective impedance frequency segment, and the common-mode resonant frequency point, the method further includes: Obtain the impedance measurement value corresponding to each nonlinear frequency scanning cycle in the frequency domain response spectrum data. If the fluctuation rate of the impedance measurement value for consecutive preset cycles is lower than the third threshold, then determine the common mode inductor impedance stability analysis result based on the impedance measurement value corresponding to each cycle.

[0069] The nonlinear frequency scan period can be a measurement time unit corresponding to a complete nonlinear frequency scan process, including one round of excitation and response acquisition from the start frequency to the end frequency according to the nonlinear step size. It can be used as a time reference unit for impedance measurement, supporting multi-cycle repeated observations to evaluate stability. In this embodiment, the nonlinear frequency scan period can be one or more of the following: initial convergence scan period, steady-state verification scan period, and abnormal retry scan period. The impedance measurement value can be the complex impedance value calculated for the common-mode inductor at a specific frequency point or effective frequency band within a single nonlinear frequency scan period. It can be used to form the basic data unit of a multi-cycle observation sequence for fluctuation analysis and stability judgment. Furthermore, the impedance measurement value can be the basis for fluctuation rate calculation due to its time-series changes, and the impedance measurement values ​​of multiple cycles jointly support the stability analysis results.

[0070] A continuous preset period can be a number of nonlinear frequency scanning cycles that are executed continuously in time and whose number is determined by preset parameters. It can be used to provide a minimum observation window for stability criteria, ensuring that the judgment is statistically significant. In an exemplary embodiment, the continuous preset period can be triggered continuously by the control system at fixed or adaptive intervals for the nonlinear frequency scanning process until the preset number of cycles is reached. Volatility can be a quantitative indicator of the relative change in impedance measurements within the continuous preset period, and can be used as a core criterion for determining whether the measurement has entered a stable state. For example, volatility can be obtained by calculating the normalized standard deviation, coefficient of variation, or maximum deviation rate of the impedance amplitude or complex impedance during the continuous period.

[0071] The third threshold can be a preset upper limit value used to determine whether the volatility meets the stability condition. It can be used to set the stability acceptance boundary and control the measurement termination or retry logic. In a specific embodiment, the third threshold may include, but is not limited to, a high-precision mode threshold, a fast test mode threshold, and a temperature drift compensation mode threshold. The common-mode inductor impedance stability analysis result can be an evaluation output on the time-domain consistency of the common-mode inductor impedance generated based on the fluctuation characteristics of multi-cycle impedance measurements. It can be used to reflect the performance repeatability of the common-mode inductor under repeated excitation, providing a basis for engineering reliability. Furthermore, the common-mode inductor impedance stability analysis result may include, but is not limited to, one or more of the following: short-term thermal stability results, long-term aging trend indication, and transient interference robustness rating.

[0072] Obtaining the impedance measurement value corresponding to each nonlinear frequency scan cycle in the frequency domain response spectrum data can be achieved by extracting the impedance value of a specified frequency point or effective frequency band from the frequency domain response spectrum data generated after each nonlinear frequency scan. Furthermore, this operation can be achieved by extracting the impedance amplitude at the common-mode resonant frequency point as a representative value, or by generating a periodic comprehensive measurement value by weighted averaging of the impedance at all frequency points within the effective impedance frequency band. This allows the construction of a cross-cycle impedance observation sequence, providing a time-series data basis for stability analysis. Determining whether the volatility of impedance measurements over a consecutive preset period is lower than a third threshold can be achieved by calculating the volatility of impedance measurements within the most recent consecutive preset periods and comparing it with the third threshold to determine whether stability conditions are met. Further, this operation can be achieved by using a sliding window to update the volatility in real time and trigger threshold judgment, or by calculating the volatility separately for different frequency bands and only determining overall stability when key frequency bands (such as resonant points) meet the threshold. This allows for dynamic identification of whether the system has entered a reliable steady-state measurement stage, avoiding misleading conclusions from single abnormal data.

[0073] The stability analysis results of the common-mode inductor impedance are determined based on the impedance measurements corresponding to each cycle. This can be achieved by integrating the impedance measurements from each cycle to generate a stability assessment report after the volatility meets the threshold condition. Furthermore, this operation can be implemented by outputting the standard deviation and confidence interval of the impedance for each cycle as a quantitative indicator of stability, or by generating stability level labels (such as "highly stable", "requires monitoring", "unstable") for subsequent system decision-making. This allows for the output of multi-dimensional stability information that includes repeatability, consistency, and potential drift trends.

[0074] For example, in the scenario of common-mode inductor reliability assessment during high-temperature aging testing, the impedance measurement method for common-mode inductors in this embodiment can be to continuously monitor the impedance of the common-mode inductor in an 85°C environmental chamber. The system performs a nonlinear frequency scan every 5 minutes for a total of 10 cycles. After each scan, the impedance amplitude at the 2.15MHz resonant point is extracted as the impedance measurement value. The coefficient of variation of the impedance amplitude over the most recent 5 cycles is calculated as the volatility. When the volatility is continuously below 0.8% (the third threshold), it is determined that the inductor has entered a thermal steady state, and the common-mode inductor impedance stability analysis result of "good short-term thermal stability" is output. This result is used to screen high-reliability magnetic components suitable for automotive power supplies.

[0075] This embodiment provides an impedance measurement method for common-mode inductors. By acquiring the impedance measurement value corresponding to each nonlinear frequency scan cycle in the frequency domain response spectrum data, if the fluctuation rate of the impedance measurement value for consecutive preset cycles is lower than a third threshold, the stability analysis result of the common-mode inductor impedance is determined based on the impedance measurement value corresponding to each cycle. By constructing a multi-cycle impedance observation sequence, using the fluctuation rate being lower than the third threshold as a convergence criterion to dynamically determine the steady state, and integrating the cycle data under stable conditions to generate a multi-dimensional stability assessment, a time-dimensional repeated scanning and stability assessment mechanism can be introduced on the basis of completing the extraction of common-mode inductor impedance characteristics. This effectively avoids single measurement deviations caused by transient interference, temperature drift, or test circuit instability. It not only verifies the repeatability and robustness of the measurement results, but also provides a quantitative basis for the long-term reliability of common-mode inductors in actual working environments, upgrading impedance characterization from a "single snapshot" to a "dynamic reliable assessment," significantly enhancing the ability to grasp the performance boundaries of components in EMI filter design, thereby improving the robustness and engineering practicality of the overall electromagnetic compatibility system.

[0076] Furthermore, to achieve the above objectives, the present invention also provides an impedance measurement system for common-mode inductors, the system comprising: The time-frequency processing module is used to acquire the voltage and current signal data of the common-mode inductor in the test circuit, and to perform time-frequency transformation processing on the voltage and current signal data to obtain frequency domain response spectrum data containing impedance characteristics. The frequency axis segmentation module is used to perform nonlinear segmentation of the frequency axis in the frequency domain response spectrum data to obtain multiple frequency segmentation regions. The phase clustering module is used to obtain the phase amplitude of the multi-segment frequency region, perform clustering and segmentation of the phase amplitude of the multi-segment frequency region, and determine the effective impedance frequency segment; The synchronization and tuning module is used to obtain the phase synchronization characteristics of the effective impedance frequency band, perform cluster analysis on the phase synchronization characteristics of the effective impedance frequency band, and determine the common-mode resonant frequency point. The integrated measurement module is used to determine the common-mode inductance measurement results based on the multi-segment frequency regions, the effective impedance frequency range, and the common-mode resonant frequency point.

[0077] Other embodiments or specific implementations of the impedance measurement system for common-mode inductors described in this invention can be found in the above-described method embodiments, and will not be repeated here.

[0078] Furthermore, to achieve the above objectives, the present invention also provides an impedance measurement device for common-mode inductors, the device comprising: a memory, a processor, and an impedance measurement program for common-mode inductors stored in the memory and executable on the processor, the impedance measurement program for common-mode inductors being configured to implement the steps of the impedance measurement method for common-mode inductors as described above.

[0079] Furthermore, to achieve the above objectives, the present invention also provides a medium storing an impedance measurement program for a common-mode inductor, wherein when the impedance measurement program for a common-mode inductor is executed by a processor, the program implements the steps of the impedance measurement method for a common-mode inductor as described above.

[0080] Furthermore, to achieve the above objectives, the present invention also provides a computer program product, including an impedance measurement program for common-mode inductors, wherein when the impedance measurement program for common-mode inductors is executed by a processor, it implements the steps of the impedance measurement method for common-mode inductors as described above.

[0081] 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. An impedance measurement method for common-mode inductors, characterized in that, The method includes: The voltage and current signal data of the common-mode inductor in the test circuit are acquired, and the voltage and current signal data are processed by time-frequency transformation to obtain frequency domain response spectrum data containing impedance characteristics. The frequency axis in the frequency domain response spectrum data is nonlinearly segmented to obtain multiple frequency segment regions. The phase amplitude of the multi-segment frequency region is obtained, and the phase amplitude of the multi-segment frequency region is clustered and segmented to determine the effective impedance frequency segment; The phase synchronization characteristics of the effective impedance frequency band are obtained, and the phase synchronization characteristics of the effective impedance frequency band are clustered to determine the common-mode resonant frequency point. The common-mode inductance measurement results are determined based on the multi-segment frequency regions, the effective impedance frequency range, and the common-mode resonant frequency point.

2. The impedance measurement method for common-mode inductors as described in claim 1, characterized in that, The step of performing time-frequency transformation processing on the voltage and current signal data to obtain frequency domain response spectrum data containing impedance characteristics includes: The voltage and current signal data are subjected to Hilbert-Huang transform to obtain time-frequency distribution data; The time-frequency distribution data is phase-normalized to obtain the frequency domain response spectrum data containing impedance characteristics.

3. The impedance measurement method for common-mode inductors as described in claim 1, characterized in that, The step of clustering and segmenting the phase amplitudes of the multi-segment frequency regions to determine the effective impedance frequency segments includes: Based on the phase amplitude of the multi-segment frequency regions, the average phase gradient of each multi-segment frequency region is calculated. Based on the average phase gradient of the multi-segment frequency regions, the phase gradient difference between the starting frequency segment and other frequency segment regions is calculated respectively; wherein, the other frequency segment regions are the remaining frequency segment regions excluding the starting frequency segment. The phase gradient difference between the starting frequency segment and other frequency segment regions is compared with a first threshold. Based on the comparison result, the multiple frequency segment regions are divided to obtain the effective impedance frequency segment.

4. The impedance measurement method for common-mode inductors as described in claim 3, characterized in that, The step of clustering the phase synchronization characteristics of the effective impedance frequency band to determine the common-mode resonant frequency includes: The phase synchronization difference between each frequency sub-segment is calculated based on the phase synchronization characteristics of the effective impedance frequency band. The phase synchronization difference between each frequency sub-segment is compared with a second threshold, and the effective impedance frequency segment is divided based on the comparison result to obtain the common-mode resonant frequency point.

5. The impedance measurement method for common-mode inductors as described in claim 1, characterized in that, The determination of the common-mode inductance measurement result based on the multi-segment frequency region, the effective impedance frequency segment, and the common-mode resonant frequency point includes: The impedance magnitude is calculated based on the phase value at the cutoff frequency point in the multi-segment frequency region, the phase reference point of the effective impedance frequency segment, and the phase inflection point of the common-mode resonant frequency point. The formula for calculating the impedance magnitude is as follows: in, This represents the impedance magnitude of the common-mode inductor at frequency ω. and These are the effective values ​​of voltage and current, respectively. This indicates the phase inflection point at the common-mode resonant frequency. The phase reference point represents the effective impedance frequency range. This indicates the phase value at the cutoff frequency.

6. The impedance measurement method for common-mode inductors as described in claim 1, characterized in that, After determining the common-mode inductance measurement result based on the multi-segment frequency region, the effective impedance frequency segment, and the common-mode resonant frequency point, the method further includes: Obtain the impedance measurement value corresponding to each nonlinear frequency scanning cycle in the frequency domain response spectrum data. If the fluctuation rate of the impedance measurement value for consecutive preset cycles is lower than the third threshold, then determine the common mode inductor impedance stability analysis result based on the impedance measurement value corresponding to each cycle.

7. An impedance measurement system for common-mode inductors, characterized in that, The system includes: The time-frequency processing module is used to acquire the voltage and current signal data of the common-mode inductor in the test circuit, and to perform time-frequency transformation processing on the voltage and current signal data to obtain frequency domain response spectrum data containing impedance characteristics. The frequency axis segmentation module is used to perform nonlinear segmentation of the frequency axis in the frequency domain response spectrum data to obtain multiple frequency segmentation regions. The phase clustering module is used to obtain the phase amplitude of the multi-segment frequency region, perform clustering and segmentation of the phase amplitude of the multi-segment frequency region, and determine the effective impedance frequency segment; The synchronization and tuning module is used to obtain the phase synchronization characteristics of the effective impedance frequency band, perform cluster analysis on the phase synchronization characteristics of the effective impedance frequency band, and determine the common-mode resonant frequency point. The integrated measurement module is used to determine the common-mode inductance measurement results based on the multi-segment frequency regions, the effective impedance frequency range, and the common-mode resonant frequency point.

8. An impedance measurement device for common-mode inductors, characterized in that, The device includes: a memory, a processor, and an impedance measurement program for common-mode inductors stored in the memory and executable on the processor, the impedance measurement program for common-mode inductors being configured to implement the steps of the impedance measurement method for common-mode inductors as claimed in any one of claims 1 to 6.

9. A medium, characterized in that, The medium stores an impedance measurement program for common-mode inductors, which, when executed by a processor, implements the steps of the impedance measurement method for common-mode inductors as described in any one of claims 1 to 6.

10. A computer program product comprising an impedance measurement program for common-mode inductors, characterized in that, When the impedance measurement program for common-mode inductors is executed by the processor, it implements the steps of the impedance measurement method for common-mode inductors as described in any one of claims 1 to 6.

Images