A method and apparatus for evaluating the aging state of a high-voltage power module package insulation

Abstract

The application discloses a kind of high-voltage power module encapsulation insulation aging state evaluation method and equipment, belongs to high-power power electronic device insulation technical field, to solve the following technical problems: the accuracy of existing evaluation high-voltage power module encapsulation insulation aging state method is not high.The proposed method includes: the relaxation process contained in the frequency domain dielectric spectrum of encapsulation insulation is decomposed;Determine the characteristic variation factor corresponding to the relaxation process respectively decomposed, and determine the relaxation process with the highest correlation degree with the aging state of the encapsulation insulation according to the characteristic variation factor in the relaxation process decomposed;Corresponding relationship between dielectric characteristic parameter and aging time corresponding to the relaxation process with the highest correlation degree is constructed, and the aging state of the encapsulation insulation is determined using the corresponding relationship.The application realizes the accurate evaluation of high-voltage power module encapsulation insulation aging state by the above method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of insulation of high-power power electronic devices, and in particular to a method and device for evaluating the aging state of high-voltage power module packaging insulation. BACKGROUND

[0002] In recent years, high-voltage power modules such as SiC power modules have gradually become an important component of circuit breakers, transformers, and converters due to their high power density, fast switching speed, strong self-protection capability, and accurate and flexible control. Epoxy resin is widely used as the main packaging insulation material for high-voltage power modules due to its excellent electrical and mechanical properties.

[0003] However, the maximum operating temperature of high-voltage power modules can reach 250-280℃. Long-term operation at high temperatures can cause thermal oxidative aging of epoxy resin, which can damage its molecular structure. More seriously, it can also cause partial discharge of epoxy resin, causing irreversible damage to the packaging insulation of high-voltage power modules. As the operating temperature of high-voltage power modules increases, the insulation deterioration problem caused by thermal aging is becoming increasingly serious, so it is particularly important to evaluate the aging state of the packaging insulation of high-voltage power modules.

[0004] The dielectric properties of epoxy resin do not decrease due to external factors, which is a good performance of its insulation properties. The change rule of dielectric characteristic parameters can guide the evaluation of insulation state, so dielectric characteristic parameters are usually used to evaluate the aging state of epoxy resin. However, for the insulation structure of high-voltage power modules in actual use scenarios, the relationship between the dielectric characteristic parameters and the aging state is still unclear, which leads to low accuracy of the aging state evaluation method by directly extracting dielectric characteristic parameters. SUMMARY

[0005] The embodiments of the present application provide a method and device for evaluating the aging state of high-voltage power module packaging insulation, which solves the technical problem of low accuracy of existing methods for evaluating the aging state of high-voltage power module packaging insulation.

[0006] The embodiments of the present application adopt the following technical solutions:

[0007] On the one hand, an embodiment of the present application provides a method for evaluating the aging state of high-voltage power module package insulation, the method comprising: decomposing the relaxation processes contained in the frequency domain dielectric spectrum of the package insulation; determining the characteristic variation factors corresponding to the decomposed relaxation processes, and determining the relaxation process with the highest correlation with the aging state of the package insulation among the decomposed relaxation processes based on the characteristic variation factors; constructing a correspondence between the dielectric characteristic parameters corresponding to the relaxation process with the highest correlation and the aging time, and using the correspondence to determine the aging state of the package insulation.

[0008] In one or more embodiments of the present application, the relaxation process contained in the frequency-domain dielectric spectrum of the package insulation is decomposed, specifically comprising: decomposing the frequency-domain dielectric spectrum into a carrier migration process, a space charge polarization process, and an interface polarization process using a frequency-domain dielectric spectrum decomposition model; the frequency-domain dielectric spectrum decomposition model is as follows:

[0009]

[0010] Among them, ε * is the frequency domain dielectric spectrum, ε ∞ is the optical frequency dielectric constant; Δε1 is the relaxation intensity of the space charge polarization process, τ1 is the relaxation time of the space charge polarization process, α1 is the first shape parameter of the space charge polarization process, β1 is the second shape parameter of the space charge polarization process, and both 11 and β1 are related to τ1; Δε2 is the relaxation intensity of the interface polarization process, τ2 is the relaxation time of the interface polarization process, α2 is the first shape parameter of the interface polarization process, β2 is the second shape parameter of the interface polarization process, and both α2 and β2 are related to τ2; ε0, σ0 are the corresponding parameters of the carrier migration process, σ0 is the DC conductivity, and ε0 is the vacuum dielectric constant.

[0011] In one or more embodiments of the present application, after decomposing the relaxation process contained in the frequency-domain dielectric spectrum of the package insulation, the method further includes: constructing a multi-parameter optimization model, and solving the parameters of the frequency-domain dielectric spectrum decomposition model using the multi-parameter optimization model; the multi-parameter optimization model is as follows:

[0012]

[0013] Among them, g is the matching degree between the fitting curve corresponding to the frequency domain dielectric spectrum decomposition model and the measured curve corresponding to the frequency domain dielectric spectrum; g1 is the difference between the real part of the frequency domain dielectric spectrum decomposition model and the real part of the frequency domain dielectric spectrum, g2 is the difference between the imaginary part of the frequency domain dielectric spectrum decomposition model and the imaginary part of the frequency domain dielectric spectrum; d is an adjustment parameter used to adjust the proportion of g1 and g2.

[0014] In one or more embodiments of the present application, determining the characteristic variation factor corresponding to each of the relaxation processes includes: selecting a preset number of test points on a curve corresponding to the relaxation process; calculating a variance and a mean value of the dielectric constant of the test points; determining a ratio between the variance and the mean value, and determining the ratio as the characteristic variation factor corresponding to the relaxation process.

[0015] In one or more embodiments of the present application, the characteristic variation factor is used to represent the dispersion of the curve corresponding to the relaxation process under different aging states, and the characteristic variation factor is positively correlated with the dispersion, and the dispersion is used to represent the degree of influence of the aging state on the relaxation process; and determining, according to the characteristic variation factor, the relaxation process with the highest correlation degree to the aging state of the packaging insulation from the relaxation processes, specifically including: determining the maximum characteristic variation factor from the characteristic variation factors corresponding to each of the relaxation processes; and determining the relaxation process corresponding to the maximum characteristic variation factor as the relaxation process with the highest correlation degree to the aging state of the packaging insulation.

[0016] In one or more embodiments of the present application, after determining the relaxation process with the highest correlation degree to the aging state of the packaging insulation, the method further includes: extracting a dielectric characteristic parameter included in the relaxation process with the highest correlation degree; and the dielectric characteristic parameter at least includes a relaxation time τ of the relaxation process with the highest correlation degree, a relaxation intensity Δε of the relaxation process with the highest correlation degree, a first shape parameter α of the relaxation process with the highest correlation degree, and a second shape parameter β of the relaxation process with the highest correlation degree.

[0017] In one or more embodiments of the present application, constructing the corresponding relationship between the dielectric characteristic parameter and the aging time includes: selecting, on a curve corresponding to the relaxation process with the highest correlation degree, dielectric constants corresponding to the packaging insulation at 0 days, 5 days, 15 days and 45 days as a fitting data set; obtaining, through the fitting data set, dielectric characteristic values of the dielectric characteristic parameter under different aging times in the relaxation process; and fitting the corresponding relationship between the dielectric characteristic values and the aging time to obtain the corresponding relationship between the dielectric characteristic parameter and the aging time.

[0018] In one or more embodiments of the present application, the corresponding relationship between the dielectric characteristic parameter and the aging time includes: a corresponding relationship between the first shape parameter α and the aging time t:

[0019]

[0020] The second shape parameter β and the corresponding relationship of aging time t:

[0021]

[0022] The relaxation time τ and the corresponding relationship of aging time t:

[0023]

[0024] The relaxation strength Δε and the corresponding relationship of aging time t:

[0025]

[0026] In one or more embodiments of the present application, after obtaining the corresponding relationship between the dielectric characteristic parameter and the aging time, the method further comprises: selecting the dielectric constant corresponding to the 25th day of the encapsulation insulation on the relaxation process corresponding curve with the highest correlation degree as a verification data set; obtaining the dielectric characteristic value corresponding to the 25th day in the relaxation process through the verification data set; using the corresponding relationship between the dielectric characteristic parameter and the aging time to obtain the aging time corresponding to the dielectric characteristic value; calculating the evaluation error corresponding to the aging time; and using the corresponding relationship with the minimum evaluation error to determine the aging state of the encapsulation insulation.

[0027] In another aspect, the embodiments of the present application also provide a high-voltage power module encapsulation insulation aging state evaluation device, the device comprising: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to execute a high-voltage power module encapsulation insulation aging state evaluation method as described above.

[0028] The high-voltage power module encapsulation insulation aging state evaluation method and device provided by the embodiments of the present application have the following beneficial effects:

[0029] 1) A frequency domain dielectric spectrum deconvolution analysis model is constructed, and based on a multi-parameter optimization model, each relaxation process contained in the frequency domain dielectric spectrum is decomposed, providing theoretical guidance for high-voltage power module insulation degradation mechanism interpretation.

[0030] 2) The influence mechanism of each relaxation process on the degree of aging is analyzed one by one, and finally the dielectric characteristic parameter corresponding to the space charge polarization process is selected for insulation state evaluation, greatly improving the evaluation accuracy of the aging state.

[0031] 3) Analyze the specific relationship between the four dielectric characteristic parameters and the aging time in the space charge polarization process, obtain the empirical formula representing the aging state of the packaging insulation, and optimize the dielectric characteristic parameter with the smallest evaluation error as the evaluation parameter, thereby further improving the accuracy of the evaluation result of the aging state of the packaging insulation. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art without creative labor. In the drawings:

[0033] Figure 1 A flow chart of an aging state evaluation method of a high-voltage power module packaging insulation provided by an embodiment of the present application;

[0034] Figure 2 A frequency domain dielectric spectrum curve diagram under different aging states in an application scenario provided by an embodiment of the present application;

[0035] Figure 3 A comparison diagram of a fitting curve and a measured curve corresponding to the real part of the frequency domain dielectric spectrum in an application scenario provided by an embodiment of the present application;

[0036] Figure 4 A curve diagram corresponding to the imaginary part of the carrier migration process in an application scenario provided by an embodiment of the present application;

[0037] Figure 5 A curve diagram corresponding to the imaginary part of the space charge polarization process in an application scenario provided by an embodiment of the present application;

[0038] Figure 6 A curve diagram corresponding to the imaginary part of the interface polarization process in an application scenario provided by an embodiment of the present application;

[0039] Figure 7 A curve diagram of the corresponding relationship between the first shape parameter a and the aging time t in an application scenario provided by an embodiment of the present application;

[0040] Figure 8 A curve diagram of the corresponding relationship between the second shape parameter β and the aging time t in an application scenario provided by an embodiment of the present application;

[0041] Figure 9 A curve diagram of the corresponding relationship between the relaxation time τ and the aging time t in an application scenario provided by an embodiment of the present application;

[0042] Figure 10A corresponding relationship curve diagram of the relaxation strength Δε and the aging time t is provided in an application scenario of the embodiment of the present application.

[0043] Figure 11 A structural schematic diagram of an aging state evaluation device of a high-voltage power module packaging insulation is provided in the embodiment of the present application. DETAILED DESCRIPTION

[0044] In order for those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0045] The embodiment of the present application provides an aging state evaluation method of a high-voltage power module packaging insulation. By using a multi-parameter optimization model, a frequency domain dielectric spectrum spectral decomposition model for decomposing multiple relaxation processes contained in the frequency domain dielectric spectrum is constructed, and the relaxation process most affected by the aging state is determined. Then, the dielectric characteristic parameters corresponding to the relaxation process are optimally selected to accurately evaluate the aging state of the packaging insulation.

[0046] The method in the embodiment of the present application will be described in detail below with reference to the drawings.

[0047] Figure 1 A flowchart of the aging state evaluation method of the high-voltage power module packaging insulation provided in the embodiment of the present application is shown in FIG. 1, and the method in the embodiment of the present application at least includes the following execution steps: Figure 1

[0048] Step 101, decompose the relaxation processes contained in the frequency domain dielectric spectrum of the packaging insulation.

[0049] The aging state evaluation method of the high-voltage power module packaging insulation provided in the embodiment of the present application first uses the multi-relaxation process decomposition idea to construct a frequency domain dielectric spectrum spectral decomposition model based on a multi-parameter optimization model, decomposes the frequency domain dielectric spectrum of the packaging insulation, and in an example of the present application, decomposes the frequency domain dielectric spectrum curve of the packaging insulation into three independent spectral lines of the carrier migration process, the space charge polarization process and the interface polarization process.

[0050] Specifically, the frequency domain dielectric spectrum spectral decomposition model of the packaging insulation needs to be constructed first.

[0051] ​The classic Debye model can be used to explain the dielectric response, which is verified in single ferroelectric and dilute solution medium. However, there is a non-Debye relaxation phenomenon in the dielectric response of the actual composite medium, which can be explained by introducing the H-N model, which describes the function of the complex dielectric constant with frequency, that is, the frequency domain dielectric spectrum, as shown in the following formula:

[0052]

[0053] In the formula, ω is the angular frequency of the applied electric field, τ HN is the relaxation time, ε ∞ is the optical frequency dielectric constant, Δε is the relaxation strength, α, β are shape parameters related to the relaxation time distribution, 0≤α≤1, 0≤β≤1.

[0054] In one or more possible implementation manners of the embodiments of the present application, the epoxy material used for the encapsulation insulation of the high-voltage power module belongs to a composite material, and the frequency domain dielectric spectrum of the composite material is often composed of a conduction process and multiple dielectric relaxation processes. In order to more finely analyze the dielectric properties of the material, according to the basic principle of dielectric relaxation, the above function of the complex dielectric constant with frequency is transformed into three independent processes including a carrier migration process, a space charge polarization process and an interface polarization process, as follows:

[0055]

[0056] The formula is the constructed frequency domain dielectric spectrum deconvolution model. In the formula, ε * is the frequency domain dielectric spectrum, ε ∞ is the optical frequency dielectric constant; Δε1 is the relaxation strength of the space charge polarization process, τ1 is the relaxation time of the space charge polarization process, α1 is the first shape parameter of the space charge polarization process, β1

[0057] is the second shape parameter of the space charge polarization process, and both α1 and β1 are related to τ1; Δε2 is the relaxation strength of the interface polarization process, τ2 is the relaxation time of the interface polarization process, α2 is the first shape parameter of the interface polarization process, β2 is the second shape parameter of the interface polarization process, and both α2 and β2 are related to τ2; ε0, σ0 are corresponding parameters of the carrier migration process, and σ0 is the direct current conductivity and ε0 is the vacuum dielectric constant.

[0058] Further, after the relaxation processes contained in the frequency domain dielectric spectrum are decomposed by the above frequency domain dielectric spectrum deconvolution model, the parameters in the relaxation processes need to be solved.

[0059] In one or more possible implementation manners of the embodiments of the present application, the number of parameters to be solved in the frequency domain dielectric spectrum deconvolution process reaches 10, and the traditional numerical method cannot quickly solve it. In order to accurately and quickly determine the above-mentioned parameters, the multi-parameter optimization model is constructed in the embodiments of the present application, and the essential idea is to determine the parameter value with the highest matching degree with the original data based on the minimum principle of the penalty function.

[0060] Therefore, the optimization variable of the multi-parameter optimization model is:

[0061]

[0062] The optimization goal of the multi-parameter optimization model is to obtain the matching degree of the real part and the imaginary part of the frequency domain dielectric spectrum fitting curve with the measured curve.

[0063] In an example of the present application, the measured curve of the above-mentioned frequency domain dielectric spectrum can be obtained by the following way:

[0064] After the complex dielectric constant of the above-mentioned H-N model is decomposed into the real part ε' and the imaginary part ε'', the following can be obtained:

[0065] ε'(ω) = ε ∞ + Δε cos(βθ)t -β / 2

[0066]

[0067] wherein,

[0068]

[0069] t = 1 + (ωτ) 2α + 2(ωτ) α cos(απ / 2)

[0070] wherein, σ0 is the direct current conductivity, and ε0 is the vacuum dielectric constant.

[0071] According to the above-mentioned real part and imaginary part, the measured curve of the frequency domain dielectric spectrum is drawn as shown in Figure 2 . Figure 2 In an application scenario provided by the embodiments of the present application, the frequency domain dielectric spectrum curve graphs in different aging states are shown in Figure 2 , wherein the curve of the frequency domain dielectric spectrum is divided into the real part curve and the imaginary part curve, and the horizontal coordinates of the real part curve and the imaginary part curve are both frequency, and the vertical coordinates are both dielectric constant, and no matter the real part curve or the imaginary part curve, the corresponding curve of different aging states is contained therein.

[0072] After that, g1 represents the matching degree of the real part of the fitting curve and the real part of the measured curve, which can be represented by the difference between the real part of the frequency domain dielectric spectrum and the real part of the frequency domain dielectric spectrum, and g2 represents the matching degree of the imaginary part of the fitting curve and the imaginary part of the measured curve, which can be represented by the difference between the imaginary part of the frequency domain dielectric spectrum and the imaginary part of the frequency domain dielectric spectrum.

[0073] And the expressions of g1 and g2 are:

[0074]

[0075]

[0076] In summary, the control equation of the multi-parameter optimization model can be obtained as:

[0077]

[0078] Wherein, d is an adjustment parameter, used to adjust the proportion of g1 and g2.

[0079] Figure 3 For an application scenario provided by the embodiment of the present application, a comparison chart of the fitting curve and the measured curve corresponding to the real part of the frequency domain dielectric spectrum is provided. As shown in Figure 3 The fitting curve of the real part obtained by fitting the frequency domain dielectric spectrum curve with the parameters solved by the above multi-parameter optimization model has good matching degree with the measured curve of the real part of the original frequency domain dielectric spectrum, which proves that the relaxation process contained in the frequency domain dielectric spectrum can be well decomposed in the embodiment of the present application, and the accuracy of the parameter value obtained after decomposition is also high, which provides guarantee for the accuracy of subsequent aging state evaluation of the packaging insulation.

[0080] At this point, the decomposition process and the parameter solving process of the relaxation process contained in the frequency domain dielectric spectrum of the packaging insulation are completed.

[0081] Step 102, determining the relaxation process with the highest correlation degree with the aging state of the packaging insulation according to the characteristic variation factor.

[0082] In this step, a discrete degree evaluation method of dielectric relaxation process under different aging states based on the characteristic variation factor is proposed, the influence mechanism of the aging degree on each relaxation process is analyzed, and the relaxation process most affected by the aging degree is finally determined.

[0083] Specifically, after the frequency domain dielectric spectrum curve of the packaging insulation is decomposed into three independent spectra of carrier migration, space charge polarization and interface polarization by using the frequency domain dielectric spectrum decomposition model, the discrete degrees of the three independent spectra under different aging states are calculated based on the characteristic variation factor. The independent spectra of the above three processes are respectively as Figure 4 、 Figure 5 andFigure 6 The characteristic variation factor is shown.

[0084] It should be noted that the aforementioned characteristic variation factor can represent the dispersion of the curve corresponding to each relaxation process under different aging states, and the characteristic variation factor is positively correlated with the dispersion, and the dispersion can represent the degree of influence of the relaxation process by the aging state, that is, the greater the value of the characteristic variation factor, the higher the dispersion of the relaxation process curve, which represents that the relaxation process is more affected by the aging state.

[0085] In one or more possible implementation manners of the embodiments of the present application, the calculation method of the characteristic variation factor is defined as follows:

[0086]

[0087] Wherein, p represents the value of the characteristic variation factor, x i represents a test point on the relaxation process curve, and N represents the number of test points.

[0088] Further, after calculating the characteristic variation factors corresponding to the three relaxation processes respectively, according to the correlation between the aforementioned characteristic variation factor and the dispersion of the relaxation curve, the relaxation process with the largest value of the characteristic variation factor is determined as the relaxation process most affected by the aging state.

[0089] In actual application scenarios, the characteristic variation factors of the three independent processes are calculated as follows: carrier migration process: p = 0.000259; space charge polarization process: p = 0.0014; interface polarization process: p = 0.0000116.

[0090] Therefore, it can be determined that the relaxation process most affected by the aging state in the frequency domain dielectric spectrum is the space charge polarization process, that is, the space charge polarization process is the relaxation process with the highest correlation degree with the aging state of the junction package.

[0091] Step 103, evaluating the aging state of the package insulation by constructing the corresponding relationship between the dielectric characteristic parameter and the aging time.

[0092] Taking the space charge polarization process as an example, the aging state evaluation method in the embodiments of the present application is further described.

[0093] After determining the relaxation process with the highest correlation degree with the aging state of the package insulation, the corresponding relationship between the dielectric characteristic parameter and the aging time is established, which can be represented by a nonlinear correlation model, and the accuracy of each model is verified, and the model with the highest accuracy is selected as the final aging state evaluation model to evaluate the aging state of the package insulation.

[0094] Specifically, first, the dielectric characteristic parameter in the space charge polarization process is extracted.

[0095] In one or more possible implementation manners of the embodiments of the present application, the extracted dielectric characteristic parameters at least include: a relaxation time τ, a relaxation strength Δε, a first shape parameter α, and a second shape parameter β.

[0096] It should be noted that the first shape parameter α and the second shape parameter β only contain mathematical meaning and are used to explain the “non-Debye semicircle” phenomenon in dielectric.

[0097] Secondly, the correspondence between the aforementioned dielectric characteristic parameters and the aging time is fitted.

[0098] After the dielectric characteristic parameters are determined, the dielectric constants of the packaging insulation corresponding to 0 days, 5 days, 15 days, and 45 days are extracted as a fitting data set by the curve corresponding to the aforementioned space charge polarization process, and the dielectric constant of the packaging insulation corresponding to the 25th day is extracted as a verification data set. Then, the dielectric characteristic values (i.e., the values of the dielectric characteristic parameters) corresponding to each aging time are calculated in the formula corresponding to the relaxation process by the fitting data set, so as to fit the relationship between the aforementioned extracted dielectric characteristic parameters and the aging time, and obtain the following relationship model:

[0099] The correspondence between the first shape parameter α and the aging time t is as follows:

[0100]

[0101] The correspondence between the second shape parameter β and the aging time t is as follows:

[0102]

[0103] The correspondence between the relaxation time τ and the aging time t is as follows:

[0104]

[0105] The correspondence between the relaxation strength Δε and the aging time t is as follows:

[0106]

[0107] The relationship curves corresponding to the aforementioned relationship model are as follows: Figure 7 The correspondence between the first shape parameter α and the aging time t in an application scenario provided by the embodiments of the present application is shown in the following graph: Figure 8 The correspondence between the second shape parameter β and the aging time t in an application scenario provided by the embodiments of the present application is shown in the following graph: Figure 9 The correspondence between the relaxation time τ and the aging time t in an application scenario provided by the embodiments of the present application is shown in the following graph: Figure 10A corresponding relationship curve between the relaxation strength Δε and the aging time t is provided in an application scenario of the embodiment.

[0108] In Figures 7-10 , the abscissa represents the aging time, i.e., the number of days, and the ordinate represents the value of the dielectric characteristic parameter. The blocks in the figure represent the values actually calculated in the formula of the space charge polarization process, and the curve represents the fitting curve. Figures 7-10 It can also be seen in that the first shape parameter α and the relaxation time τ both decrease exponentially with the increase of the aging time, while the second shape parameter β and the relaxation strength Δε both increase with the increase of the aging time, and the slope of the increasing trend first increases and then decreases.

[0109] It should be noted that the gradual decrease of the relaxation time τ indicates that new polar substances are generated in the aging process of the encapsulation insulation, and the corresponding characteristic frequency increases; the increase of the relaxation strength Δε is due to the generation of small-molecule polar substances in the aging process of the encapsulation insulation, and a larger number of polar molecules participate in the polarization process after the excitation voltage is applied, so the relaxation strength increases obviously.

[0110] After that, the established corresponding relationship is verified to determine the corresponding relationship with the highest accuracy / smallest error.

[0111] After obtaining the above four corresponding relationships, the dielectric characteristic value corresponding to the 25th day is calculated by substituting the above verification data set into the corresponding formula of the space charge polarization process, and then the dielectric characteristic value is substituted into the above four corresponding relationships to calculate the aging time, i.e., the number of days. The error is calculated by comparing the 25th day with the calculated number of days, and the corresponding relationship with the smallest error in the above four corresponding relationships is found.

[0112] In actual use scenarios, it can be found that the corresponding relationship composed of the first shape parameter α has the smallest error of 0.0046.

[0113] Finally, the aging state of the encapsulation insulation is evaluated through the above-mentioned corresponding relationship with the highest accuracy / smallest error.

[0114] According to the above error calculation, the corresponding relationship of the first shape parameter α is selected for the evaluation of the aging state of the encapsulation insulation.

[0115] In order to more detailedly explain the aging state evaluation method in the embodiment, the following supplementary description is made in the embodiment.

[0116] Firstly, a frequency domain dielectric spectroscopy (FDS) model of encapsulation insulation is constructed based on a multi-parameter optimization algorithm. In an example of the present application, the material of the encapsulation insulation is epoxy resin.

[0117] To verify the accuracy of the FDS model, pure epoxy resin samples with different aging times are prepared. In order to obtain samples with different aging degrees within a limited time, an accelerated thermal aging experiment is performed on the samples to be tested. When the temperature is increased, the aging rate of the sample will be accelerated. Therefore, based on the normal operating temperature of 90°C of the high-voltage power module, epoxy resin samples aged for 0, 5, 15, 25 and 45 days are obtained at 150°C.

[0118] The real part of the dielectric constant of the pure epoxy resin with different aging times is measured, and the measured value is compared with the fitted value. As shown in FIG. 1, the measured value and the fitted value are highly consistent, verifying the accuracy of the FDS model. Figure 3

[0119] Then, the influence mechanism of different aging times on the numerical dielectric properties of epoxy is analyzed, as shown in FIG. 2. Figures 4-6

[0120] 1) The aging of epoxy resin is specifically manifested as the breaking of the main chain and the shedding of the branched chain groups of macromolecules, but unlike the aging of oil-paper insulation, no impurity ions are generated, and therefore the change in the carrier migration process is not obvious.

[0121] 2) Aging causes the molecular chain of epoxy resin to break, generating small molecular polar substances such as acids and aldehydes. At the same time, more defects are generated, causing the increase of the trap density and the total amount of space charges in the epoxy resin. As a result, the space charge polarization loss in the frequency domain will be significantly increased.

[0122] 3) The breaking of the molecular chain of epoxy resin also causes the increase of the internal micro area, causing the increase of the interfacial polarization loss, but compared with the inherent micro interface formed in the crosslinking process, the increase of the interface caused by the chain breaking is limited, and therefore the contribution to the interfacial polarization is relatively small.

[0123] Moreover, through calculation, it is determined that the characteristic variation factor of the space charge polarization in the above three relaxation processes is the largest p=0.0014. Therefore, it is determined that the influence of aging on the space charge polarization process is the largest.

[0124] Finally, the nonlinear relationship models of the four dielectric characteristic parameters in the space charge polarization and the aging time are fitted respectively, as shown in FIG. 3, and the errors of the models are evaluated in sequence. The model with the smallest error is selected as the aging state evaluation model, which can realize the accurate evaluation of the aging state of the encapsulation insulation of the high-voltage power module. Figures 7-10

[0125] ​​​Based on the same inventive concept, the application also provides an aging state evaluation device for high-voltage power module packaging insulation, which has a structure as shown in Figure 11

[0126] Figure 11 FIG. 1 shows a structural schematic diagram of an aging state evaluation device for high-voltage power module packaging insulation provided by the application. As shown in Figure 11 the device 1100 in the application specifically includes at least one processor 1101, and a memory 1103 connected with the at least one processor 1101 (connected through a bus 1102), wherein the memory 1103 stores instructions executable by the at least one processor 1101, so that the at least one processor 1101 can perform the aging state evaluation method for high-voltage power module packaging insulation as described in the above embodiments.

[0127] In one or more possible implementation manners of the application, the processor is configured to perform decomposition on relaxation processes contained in the frequency domain dielectric spectrum of the packaging insulation, determine characteristic variation factors corresponding to the relaxation processes respectively, and determine a relaxation process with the highest correlation degree with the aging state of the packaging insulation according to the characteristic variation factors; construct a corresponding relationship between a dielectric characteristic parameter corresponding to the relaxation process with the highest correlation degree and an aging time, and determine the aging state of the packaging insulation by using the corresponding relationship.

[0128] Those skilled in the art should understand that the embodiments of the present specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of the present specification can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present specification can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0129] The present specification is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present specification. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions described in the flowcharts and / or block diagrams. Figure 1 one flow or multiple flows and / or blocks Figure 1 ​means for performing the function specified in the block or blocks.

[0130] It should also be noted that the terms "comprising," "including," and any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0131] The description can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of steps and methods described in this specification is not the only sequence or method that can be implemented. For example, acts described herein can be performed in an order other than the order described. Alternatively, acts can be performed at the same time. Alternatively, acts can be performed in a different order or at different times. Further, various steps can be combined or omitted.

[0132] Various embodiments of the present specification are described in the progressive manner, and the same or similar parts between various embodiments can be referred to each other, and each embodiment focuses on the difference from other embodiments. Especially, for the device, apparatus, and non-transitory computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant part can be referred to the part of the method embodiment.

[0133] The above described embodiments of the present specification have been described. Other embodiments are within the scope of the following claims. In some cases, the acts or steps recited in the claims can be performed in a different order and still accomplish desirable results. Also, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order to achieve desirable results. In certain implementations, multitasking and parallel processing can be advantageous.

[0134] The above descriptions only serve the one or more embodiments of the present specification, and are not intended to limit the present specification. For those skilled in the art, the one or more embodiments of the present specification can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the one or more embodiments of the present specification shall be included in the scope of claims of the present specification.

Claims

1. A method of evaluating an aging state of insulation of a high-voltage power module package, characterized by, The method comprises: The relaxation process contained in the frequency-domain dielectric spectrum of the package insulation is decomposed, including: decomposing the frequency-domain dielectric spectrum into a carrier migration process, a space charge polarization process, and an interface polarization process through a frequency-domain dielectric spectrum decomposition model. The frequency-domain dielectric spectrum decomposition model is as follows: wherein is the frequency domain dielectric spectrum, is the optical frequency dielectric constant; is the relaxation strength of the space charge polarization process, is the relaxation time of the space charge polarization process, is a first shape parameter of the space charge polarization process, is a second shape parameter of the space charge polarization process, and and are related to ; is the relaxation strength of the interface polarization process, is the relaxation time of the interface polarization process, is a first shape parameter of the interface polarization process, is a second shape parameter of the interface polarization process, and and are related to ; , is the corresponding parameter of the carrier transport process, and is the direct current conductivity, is the vacuum permittivity; A multi-parameter optimization model is constructed, and the parameters of the frequency domain dielectric spectrum decomposition model are solved by the multi-parameter optimization model. The multi-parameter optimization model is as follows: wherein, is a matching degree between a fitting curve corresponding to the frequency domain dielectric spectrum deconvolution model and a measured curve corresponding to the frequency domain dielectric spectrum; is a difference between a real part of the frequency domain dielectric spectrum deconvolution model and a real part of the frequency domain dielectric spectrum, is a difference between an imaginary part of the frequency domain dielectric spectrum deconvolution model and an imaginary part of the frequency domain dielectric spectrum; d is an adjustment parameter, used for adjusting and a proportion of Determining characteristic variation factors corresponding to the decomposed relaxation processes respectively, and determining, according to the characteristic variation factors, a relaxation process among the decomposed relaxation processes that has the highest correlation with the aging state of the package insulation; A correspondence between the dielectric characteristic parameter corresponding to the relaxation process with the highest correlation and the aging time is constructed, and the aging state of the package insulation is determined using the correspondence.

2. The method of claim 1, wherein the method is characterized by: Determining the characteristic variation factors corresponding to the decomposed relaxation processes, specifically including: Selecting a preset number of test points on the curve corresponding to the relaxation process; Calculate the variance and mean of the dielectric constant corresponding to the test point; A ratio between the variance and the mean is determined, and the ratio is determined as a characteristic variation factor corresponding to the relaxation process.

3. The method of claim 2, wherein the method further comprises: The characteristic variation factor is used to characterize the dispersion of the corresponding curve of the relaxation process under different aging conditions, and the characteristic variation factor is positively correlated with the dispersion, and the dispersion is used to characterize the degree to which the relaxation process is affected by the aging condition; Determining, according to the characteristic variation factor, a relaxation process having the highest correlation with the aging state of the package insulation among the decomposed relaxation processes specifically includes: Determining the maximum characteristic variation factor among the characteristic variation factors corresponding to the decomposed relaxation processes; The relaxation process corresponding to the maximum characteristic variation factor is determined as the relaxation process with the highest correlation with the aging state of the package insulation.

4. The method of claim 1, wherein the method further comprises: After determining the relaxation process most correlated with the aging state of the package insulation, the method further comprises: extracting dielectric characteristic parameters contained in the relaxation process with the highest correlation; the dielectric characteristic quantity comprises at least a relaxation time of the relaxation process having the highest correlation a relaxation strength of the relaxation process having the highest correlation a first shape parameter of the relaxation process having the highest correlation a second shape parameter of the relaxation process having the highest correlation .

5. The method of claim 4, wherein the method further comprises: Establishing the corresponding relationship between the dielectric characteristic parameter and the aging time specifically includes: On the relaxation process corresponding curve with the highest correlation, the dielectric constants of the package insulation at 0 days, 5 days, 15 days, and 45 days are selected as fitting data sets; Obtaining dielectric characteristic values ​​of the dielectric characteristic parameters at different aging times during the relaxation process using the fitted data set; The corresponding relationship between the dielectric characteristic value and the aging time is fitted to obtain the corresponding relationship between the dielectric characteristic parameter and the aging time.

6. The method of claim 5, wherein the method further comprises: The corresponding relationship between the dielectric characteristic parameter and the aging time includes: the first shape parameter correspondence with the aging time t: the second shape parameter correspondence with the aging time t: the relaxation time correspondence with the aging time t: the relaxation strength correspondence with the aging time t: 。 7. The method of claim 6, wherein the method further comprises: After obtaining the corresponding relationship between the dielectric characteristic parameter and the aging time, the method further includes: On the relaxation process corresponding curve with the highest correlation, the dielectric constant corresponding to the package insulation on the 25th day is selected as the verification data set; Obtaining, by using the validation data set, a dielectric characteristic value corresponding to the dielectric characteristic parameter on the 25th day during the relaxation process; The aging time corresponding to the dielectric characteristic value is obtained by using the correspondence between the dielectric characteristic value and the aging time; An evaluation error corresponding to the aging time is calculated; The aging state of the encapsulation insulation is determined by using the correspondence with the minimum evaluation error.

8. An apparatus for evaluating an aging state of insulation of a high-voltage power module package, characterized by, The device comprises: at least one processor; and a memory connected to the at least one processor in communication; wherein The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the aging state evaluation method of the high-voltage power module encapsulation insulation according to any one of claims 1-7.

Images