A method for predicting the life of solid dielectrics by equivalent acceleration

Abstract

The present invention discloses a method for equivalent accelerated prediction of solid dielectric life, which belongs to the technical field of solid dielectric life prediction. The present invention obtains E by conducting a conduction current test and a short-time breakdown test of a solid dielectric sample. a and E b , determine the field strength E of each order applied by the step stress test to satisfy E a <E<E b , using the test duration as a common parameter to jointly establish the damage expressions of step stress and constant stress, deriving the influence of step test parameters on n, and finally calculating the n value using the damage characteristic relationship obtained from the step stress test. The present invention can effectively obtain the n value, avoid conducting invalid tests, greatly improve test efficiency, save the number of test products, and provide a basis for parameter selection for other types of accelerated life tests. In addition, it can avoid the equivalence problem of obtaining the n value caused by the type of applied electric field being different from the constant electric field, ensuring that the obtained n and D c The unique correspondence between the two improves the accuracy of solid dielectric life prediction.

Description

Technical Field

[0001] The invention belongs to the technical field of solid dielectric life prediction, and in particular relates to a method for equivalent accelerated prediction of solid dielectric life. Background Art

[0002] The constant stress method is closest to the natural aging of solid dielectrics. The method for obtaining the life index is to linearly fit the (E, t) data points obtained under multiple electric fields, and obtain n and Dc from the slope and intercept respectively: Accelerated life testing is a common method for obtaining the life index. The life index of the inverse power model shown in formula (1) can reflect the remaining life of solid dielectrics:

[0003] Dc=tE n (1);

[0004] Where t represents the remaining electrical life of the dielectric under the field strength E, n is the life index, and Dc represents the cumulative damage threshold of the dielectric to withstand electrical damage. n and Dc of solid dielectric samples of the same specification are constants. Taking the logarithm of both sides of formula (1) can obtain formula (2), which is expressed as a life curve:

[0005]

[0006] However, the (E, t) data of the constant stress test is highly dispersed, which is not conducive to the accuracy of the results. In addition, the selection of the constant stress level is blind, and even at the appropriate stress level, some samples have a problem of too long failure time. Therefore, other forms of stress accelerated aging are currently used to improve efficiency, but there are the following problems: First, accelerating the life of solid dielectrics will make the aging mechanism different from the aging under natural conditions (constant stress), and the obtained life index has no reference value; second, the failure data obtained by the accelerated life test has a form different from the failure data of the constant stress test, and it is difficult to obtain the life index by analyzing the failure data.

[0007] Among the accelerated life methods currently proposed, the step stress method is the most widely used. The cumulative damage threshold caused by step stress can be calculated by formula (3):

[0008]

[0009] Where E 0 is the initial field strength, E i is the field strength at each level, i is the field strength at the i-th level, i=0,1,…,k-1,k; Δt is the boost time step, t k is the final field strength E k action time.

[0010] It can be seen from formula (3) that the cumulative damage increases exponentially with the increase of field strength, so it has higher test efficiency, but there are the following technical problems: First, there is a lack of basis for the selection of test parameters, such as the size of the initial field strength E 0 , boost time step Δt and boost step ΔE, etc., which makes the blindly selected stress level may cause the aging mechanism of the solid dielectric to change, resulting in the accelerated life test process being unable to be equivalent to the aging under the natural state (constant stress). The literature discloses a method for selecting step stress test parameters to obtain the DC voltage tolerance index of XLPE insulation. The method discloses the differences between the damage curves produced by step stress tests with different parameters, and concludes that the test parameters for the same sample should be diversified, but no specific diversity requirements are proposed, so the problem is not fundamentally solved. Second, the failure data obtained by the step stress test does not have the form of constant stress (E, t) data, and the life index n cannot be obtained by linear fitting. The solution provided by the prior art provides a basis for solving this problem to a certain extent. Patent applications CN110007199A and CN112364523A propose methods for obtaining the n value, which lack the selection of the optimal number of test groups. The more tests are conducted, the more accurate the results are. However, too many tests will undoubtedly affect the test efficiency. In addition, the above two methods cannot obtain n and Dc simultaneously. The unique correspondence between the two is weak, which is not conducive to the accuracy of predicting the remaining life. The literature on the evaluation of the XLPE insulation voltage tolerance index based on the mixed stress method discloses that the step stress damage is converted into constant stress damage by combining equations (1) and (3), and then the life index n and the cumulative damage threshold Dc are obtained by fitting according to equation (2). However, this conversion must be based on the equivalence of the two types of test processes, otherwise the results lack reference value.

[0011] Therefore, the electric field interval and parameter selection basis for electrothermal aging have not been solved. This is because the aging process of solid dielectrics is affected by many factors, which makes its aging characteristics have obvious nonlinear characteristics. In addition, the two test methods (constant stress method and step stress method) have not been effectively unified in obtaining n. This is because the optimal number of tests is difficult to determine, and it is very difficult to synchronously obtain n and Dc to ensure their unique corresponding relationship. Therefore, it is very necessary to solve the existing technical problems, but it is difficult. Summary of the invention

[0012] In view of the above-mentioned shortcomings of the prior art, the present invention proposes a method for equivalent accelerated prediction of the life of solid dielectrics.

[0013] A method for equivalent accelerated prediction of solid dielectric life, comprising the following steps:

[0014] (1) Prepare three or more groups of solid dielectric samples of the same specification, with no less than 10 samples in each group, and conduct current tests in the temperature range of 30 to 90°C. The relationship between the conduction current density J and the field strength E is as shown in formula (4):

[0015] J=αE x (4);

[0016] x is an index that characterizes the power-law relationship between E and J. The (E, J) data points obtained are linearly fitted in a double logarithmic coordinate system to obtain the EJ characteristic curve. When the conduction current transitions to the non-ohmic region, the boundary between the two regions causes the EJ curve to deflect. The field intensity corresponding to the deflection is the threshold field intensity E a .

[0017] (2) Prepare no less than 15 solid dielectric samples of the same specification and conduct short-time breakdown tests. Use the Weibull distribution shown in formula (5) to calculate the average breakdown field strength E b Statistical analysis was performed and the field strength E of F(E) = 63.2% was taken as the average breakdown field strength E of the sample. b :

[0018] F(E)=1-exp[-(E / a) β ] (5);

[0019] Where F(E) is the failure rate when the field strength is E, α and β are the scale parameter and shape parameter respectively.

[0020] (3) The field strength applied in the step stress test is a step-up test, and each step of the field strength E satisfies E a <E<E b .

[0021] (4) Clarify the influence of step stress test parameters on n:

[0022] S4-1: Assume that the breakdown of the solid dielectric occurs at the final field strength E k The action time of is complete Δt, then

[0023]

[0024] ΔE is the step length of the voltage increase, Dc is the threshold of the dielectric to withstand electrical damage, and E i is the field strength at each level, i is the field strength at the i-th level, i=0,1,…,k-1,k; Δt is the boost time step, t k is the final field strength E k action time.

[0025] S4-2: Simplify equation (6) to equation (8):

[0026]

[0027] The ratio of ΔE / Δt is defined as h, t c is the total time of the experiment.

[0028] (5) Based on step (3) and step (4), the damage eigenvalue method is used to calculate the n value of the inverse power model and the Dc corresponding to n. The life curve is fitted according to the obtained n and Dc, and the life prediction is performed based on the life curve.

[0029] As a preferred embodiment of the present invention, in the steps (1) and (2), a flat plate vulcanizing machine is used to hot-press the solid dielectric sample raw material into sheet-shaped solid dielectric samples of the same size.

[0030] As a preferred embodiment of the present invention, solid dielectric samples of the same specification refer to solid dielectric samples of the same size.

[0031] As a preferred embodiment of the present invention, the step stress test parameters include the initial field strength E 0 , boost step ΔE, boost time step Δt.

[0032] As a preferred embodiment of the present invention, in the short-time breakdown test, the voltage is increased uniformly at a rate of v=1 kV / s until the sample breaks down.

[0033] As a preferred embodiment of the present invention, in the constant stress test, the remaining life of the solid dielectric is:

[0034] Dc=tE n (1).

[0035] As a preferred embodiment of the present invention, in the step stress test, the cumulative damage caused by the step stress is:

[0036]

[0037] As a preferred embodiment of the present invention, in step S4-2, equation (6) is simplified by Bernoulli formula to obtain equation (8):

[0038]

[0039] Where C i p+1 are polynomial coefficients, and B is the Bernoulli number.

[0040] As a preferred embodiment of the present invention, the step (5), the damage characteristic value method comprises the following steps:

[0041] S5-1: Based on formula (8), m h values ​​are selected for each step stress test, and m groups of ΔE and Δt values ​​are selected at the same h value to perform step stress tests, where m ≥ 3;

[0042] S5-2: Substitute the data of each level generated by the m×m step stress test in S5-1 into the cumulative damage threshold formula caused by step stress. Then the data obtained from the m×m tests are arranged in sequence to form an m×m square matrix D(n), and the rank of D(n) is 1, where n and Dc are unknown quantities;

[0043] Iterate the value of n starting from 0 and substitute n into formula (10) equal to 0 to calculate the corresponding eigenvalue, so that the eigenvalue of the D(n) matrix satisfies λ max =mDc: The m eigenvalues ​​corresponding to the D(n) matrix are λ 1 ,λ 2 ,λ 3 …,λ m , set the accuracy requirement α, set I ＝ S / λ max , when I≤α, when I≤α, and the n value when I is the minimum value is the true n value of the solid dielectric, corresponding to Dc=λ max / m; where S = λ 1 +λ 2 +λ 3+ …+λ m-1 ,λ max＝ λ m ; If I≤α is not satisfied, then n=n+δn and the n value iteration in S5-3 is continued until I≤α is satisfied and the true n value and the corresponding Dc are calculated, where δn is the iteration step length;

[0044] The formula (10) is

[0045]

[0046] As a preferred embodiment of the present invention, in the step (5), the life curve is a life curve in which the logarithm of both sides of the formula (1) is taken, and the vertical coordinate is lnE and the horizontal coordinate is lnt;

[0047] Dc=tE n (1);

[0048] Where t represents the remaining electrical life of the dielectric under the field strength E, n is the life index, and Dc represents the cumulative damage threshold of the dielectric to withstand electrical damage.

[0049] As a preferred embodiment of the present invention, n and Dc of solid dielectric samples with the same specifications are constant.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] (1) The present invention proposes a method for equivalent accelerated prediction of solid dielectric life. By selecting parameters of the solid dielectric life index of the step stress test, it can effectively avoid conducting invalid tests, greatly improve the test efficiency, save the number of test samples, and improve the credibility of the results. It also provides a basis for parameter selection for other types of accelerated life tests, such as constant stress tests.

[0052] (2) The method for equivalent accelerated prediction of the life of solid dielectrics proposed in the present invention first selects multiple h values ​​in the step stress test based on the effectiveness of the test process and the equivalence of the test results. The same h value arbitrarily selects ΔE and Δt, and the constant h assignment principle is implemented. Finally, the inverse power model parameters are obtained by solving the eigenvalues ​​of the cumulative damage matrix. The functional relationship between n and Dc is obtained through experiments and an m-order matrix is ​​constructed. The matrix is ​​iterated to obtain the n value that meets the conditions, and then the life of the solid dielectric is predicted. This method is applicable to the method of obtaining n by step stress and constant stress, and can avoid the result equivalence problem caused by different methods of obtaining n, ensure the unique corresponding relationship between n and Dc, and improve the accuracy of the prediction of the remaining life of the solid dielectric. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A curve diagram of the life index obtained under constant stress according to the present invention.

[0054] Figure 2 This is a diagram showing the influence of the external applied field strength on n in the present invention.

[0055] Figure 3 Flowchart for calculating n value using damage eigenvalue method.

[0056] Figure 4 This is the damage curve of the step stress test of Group A test in Example 1 of the present invention.

[0057] Figure 5 This is the damage curve of the step stress test of Group B test in Example 1 of the present invention.

[0058] Figure 6 This is the damage curve of the step stress test of Group C test in Example 1 of the present invention.

[0059] Figure 7 This is the damage curve of the step stress test of group D test in Example 1 of the present invention.

[0060] Figure 8 This is a graph of n and Dc obtained from the experiment of Group A in Example 1 of the present invention.

[0061] Fig. 9 This is a graph of n and Dc obtained from the experiment of Group B in Example 1 of the present invention.

[0062] Fig.10 This is a graph of n and Dc obtained from the experiment of Group C in Example 1 of the present invention.

[0063] Fig.11 This is a graph of n and Dc obtained in the D group experiment in Example 1 of the present invention. DETAILED DESCRIPTION

[0064] In order to better illustrate the purpose, technical solutions and advantages of the present invention, the present invention will be further described below in conjunction with specific embodiments.

[0065] Example 1

[0066] A method for equivalent accelerated prediction of solid dielectric life, comprising the following steps:

[0067] (1) Prepare multiple groups of solid dielectric samples with the same specifications, with 15 samples in each group, and conduct current tests in the temperature range of 30 to 90°C. The relationship between the conduction current density J and the applied electric field E is as shown in formula (4):

[0068] J=αE x (4);

[0069] The obtained (E, J) data points are linearly fitted in the double logarithmic coordinate system to obtain the EJ characteristic curve; when E>E a When the conduction current transitions to the non-ohmic region, the boundary between the two regions causes the EJ curve to deflect. The field strength corresponding to the deflection is the threshold field strength E a , since the conduction current test experiment is carried out at room temperature, E a =30.73kV / mm.

[0070] (2) Fifteen solid dielectric samples of the same specification were prepared and subjected to short-time breakdown tests. The breakdown field strength was statistically analyzed using the Weibull distribution shown in equation (5). The field strength E at F(E) = 63.2% was taken as the average breakdown field strength E of the sample. b :

[0071] F(E)=1-exp[-(E / a) β ] (5);

[0072] Where F(E) is the failure rate when the field strength is E, α and β are the scale parameter and shape parameter respectively;

[0073] By fitting the curve of the above equation, the horizontal axis is E, the vertical axis is F(E), and the short-time breakdown field strength E corresponding to F(E) = 63.2% b =354kV / mm. b 、v=1kV / s、n 1=10Substitute into formula (11) and get Dc 1 =9.95×10 15 ;

[0074]

[0075] (3) The field strength applied in the step stress test is a step-up test. The field strength E of each step may exist: ①E <E a , ②E a <E<E b ③E>E b In the three cases, the natural aging process is the electrothermal aging process, so the accelerated aging should also ensure this aging mechanism. a The accumulation of space charge can increase the local field strength, thus accelerating aging. Therefore, the starting field strength of the stepping field strength is E 0 Should be greater than E a . And E>E b When the breakdown is in the form of short-time thermal breakdown rather than electrothermal breakdown, the final field strength E of the stepping field strength is k Should be less than E b Therefore, the initial parameters of the step electric field should meet condition ② to ensure the equivalence of the test process.

[0076] (4) Clarify the influence of step stress test parameters on n to ensure the validity of the test results:

[0077] S4-1: Assume that the breakdown of the solid dielectric occurs at the final field strength E k The action time is the complete Δt, so the formula (3) of the cumulative damage caused by the step stress in the step stress test is transformed into formula (6):

[0078]

[0079]

[0080] ΔE is the stress rise gradient.

[0081] S4-2: Formula (6) is the form of sum of natural number powers, which can be simplified by the Bernoulli formula shown in formula (7), and the result is shown in formula (8):

[0082]

[0083] Where C i p+1 are polynomial coefficients, and B is the Bernoulli number.

[0084]

[0085] The ratio of ΔE / Δt is defined as h, t c is the total time of the experiment.

[0086] S4-3: Due to the same specification of solid insulation style n and D c is a constant, so equation (8) shows the relationship between h and t c This means that when only the values ​​of ΔE and Δt are changed without changing their ratio h, the t corresponding to each step stress test c From the perspective of the inverse power model, both the step stress and constant stress methods are intended to cause cumulative damage to the electrical stress. Therefore, by combining equations (1) and (3), we can obtain equation (9): Combining the step stress test and the constant stress test, we can obtain equation (9);

[0087]

[0088] According to formula (9), the test time is t c The step stress can be expressed as c This approach establishes a correlation between the two methods based on the equivalence of the test process. Therefore, when the step stress test with different parameter combinations has similar t c When the test result is similar to the repeated step stress test with the same parameter combination. Figure 1 (Life curve diagram represented by formula (2)), if the constant electric fields are very close, such as Figure 2 , then each (E, t) data point can be represented by multiple life curves, and the numerical deviations of n and Dc are large and cannot be used for life prediction.

[0089] Therefore, when setting the step stress parameters ΔE and Δt in this step, the constant h assignment principle should be satisfied, that is, the ratio h of each group is different, and the values ​​of ΔE and Δt in the same h value can be selected arbitrarily.

[0090] (5) The process and calculation formula of this step are as follows: Figure 3 Based on the validity of the test process in step (3) and the validity of the test results in step (4), the damage eigenvalue method is used to calculate the n value of the inverse power model and the Dc corresponding to n. The life curve is fitted according to the obtained n and Dc, and the life prediction is performed according to the life curve:

[0091] S5-1: Four groups of step stress tests were conducted, namely Group A, Group B, Group C, and Group D. Three h values ​​were selected for each group of tests, namely h 1 =1 / 600,h 2 =1 / 800,h 3=1 / 1200, arbitrarily select the values ​​of ΔE and Δt under the same h value (as shown in Table 1-4), and conduct three step stress tests. According to formula (8), the cumulative damage curves of the four groups of tests are drawn ( Figure 4-7 ), as shown in the figure, in each group of tests, the tc of the tests with the same h is slightly different, while the tc of the tests with different h is quite different, which proves that the constant h assignment principle can make the dispersion of failure data meet the accuracy requirements.

[0092] S5-2: Substitute the data of each level generated by the 9 step stress tests conducted in each group into expression (3). Then the data obtained from the 9 tests can be arranged in sequence to form a 3×3 square matrix D(n), where n and Dc are unknown quantities.

[0093] S5-3: Since samples of the same specification have the same Dc, the rank of the D(n) matrix should be 1. At this time, the 3rd-order matrix has 2 eigenvalues ​​of 0, and the largest eigenvalue remaining that is not 0 is λ max =3Dc, and its corresponding three eigenvalues ​​are λ 1 =λ max =3Dc,λ 2 =0,λ 3 = 0. When formula (10) is equal to 0, λ is the eigenvalue of the matrix. Therefore, the solution of n value is also transformed into finding a suitable n value so that the eigenvalue of the matrix D(n) satisfies λ max =3Dc. In practice, we iterate the n value from 0 and calculate the corresponding eigenvalue. Then, when the sum of the two smaller eigenvalues ​​S and the largest eigenvalue λ max Ratio I min When it is the smallest, the n value at this time is considered to be the true n value of the solid dielectric, corresponding to Dc = λ max / 3.

[0094] I≤α=5% is selected as the constraint condition to meet the accuracy of engineering error. min The smaller the value, the better the uniqueness of n. When I>α, n=n+δn is used to continue the n value iteration and eigenvalue calculation in S5-3 until I≤α is satisfied, and the true n value and the corresponding Dc are calculated. δn is the iteration step length. The n value and Dc value obtained from each group of experiments are as follows: Figure 8-11 As shown in the figure, the change curve of the inverse of the I value reflects the change of the I value. The horizontal coordinate corresponding to the highest point of the curve is the n value that meets the conditions. At this time, λ max =3Dc. The n values ​​and Dc values ​​of the four groups of step stress tests A to D are all obtained at the peak value, and the results are shown in Table 5.

[0095]

[0096] D(n) cmm Represents the Dc value of the mth element in the mth row.

[0097] As shown in Table 5, the life index n obtained from the four groups of experiments is within the existing reference range of 9 to 20, indicating that the proposed method can make the test meet the validity. min The value does not exceed 5%, satisfying the max / 3 as the constraint condition of Dc value. In addition, the present invention adopts the proposed damage characteristic value method to obtain n and Dc values ​​at the same time, and the two have a unique corresponding relationship. Among the four groups of n and Dc obtained, the closer the n values ​​obtained by the step stress method and the constant stress method are, the closer the corresponding Dc values ​​are, indicating that the damage characteristic value method can effectively analyze the failure data obtained by the proposed parameter selection method. The setting of the test parameters of group D in Table 5 is most in line with the proposed equivalence principle (result equivalence and process equivalence). The obtained n=13.04 is closest to the n=12.83 obtained by the constant stress experiment. The two have the best equivalence and better accuracy, and I min It is the minimum value among the four groups and has better uniqueness, so n=12.83 is used as the voltage withstand life index of the insulation sample.

[0098] S5-4: Fitting a life curve based on n and Dc obtained in step S5-3, and performing life prediction based on the life curve:

[0099] The life curve is obtained by taking the logarithm of both sides of equation (1), fitting the life curve with the ordinate as lnE and the abscissa as lnt;

[0100] Dc=tE n (1);

[0101] Where t represents the remaining electrical life of the dielectric under the field strength E, n is the life index, and Dc represents the cumulative damage threshold of the dielectric to withstand electrical damage.

[0102] Table 1 Test parameters of group A

[0103]

[0104] Table 2 Test parameters of group B

[0105]

[0106] Table 3 Experimental parameters of group C

[0107]

[0108]

[0109] Table 4 Test parameters of group D

[0110]

[0111] Table 5 Comparison of four groups of step stress test results and constant stress test results

[0112] Group No. n <![CDATA[D c ]]> <![CDATA[I min ]]> A 15.83 <![CDATA[1.83×10 26 ]]> 1.62％ B 15.63 <![CDATA[9.24×10 25 ]]> 1.46％ C 13.16 <![CDATA[1.13×10 22 ]]> 2.17％ D 13.04 <![CDATA[7.44×10 21 ]]> 0.38％ Constant stress 12.83 <![CDATA[3.42×10 21 ]]> --

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

Claims

1. A method for equivalent accelerated prediction of solid dielectric life, characterized in that: The steps include: (1) Prepare multiple groups of solid dielectric samples of the same specifications, with no less than 10 samples in each group, and conduct current tests in the temperature range of 30 to 90°C. The relationship between the conduction current density J and the field strength E is as shown in formula (4): J=dE x (4); d and x are parameters that characterize the relationship between the conduction current density J and the electric field strength E. x is an exponent that characterizes the power-law relationship between E and J. The obtained (E, J) data points are linearly fitted in a double logarithmic coordinate system to obtain the EJ characteristic curve. When the conduction current transitions to the non-ohmic region, the boundary between the two regions causes the EJ curve to deflect. The electric field strength corresponding to the deflection is the threshold electric field strength E. a ; (2) Prepare no less than 15 solid dielectric samples of the same specification and conduct short-time breakdown tests. Use the Weibull distribution shown in formula (5) to calculate the average breakdown field strength E b Statistical analysis was performed and the field strength E of F(E) = 63.2% was taken as the average breakdown field strength E of solid dielectric samples of the same specification. b : F(E)=1-exp[-(She) β ] (5); Where F(E) is the failure rate when the field strength is E, a and β are the scale parameter and shape parameter respectively; (3) The field strength applied in the step stress test is a step-up test, and each step of the field strength E satisfies E a <E<E b ; (4) Clarify the influence of step stress test parameters on n: S4-1: Assume that the breakdown of the solid dielectric occurs at the final field strength E k The action time of is complete Δt, then ΔE is the step length of the voltage increase, Dc is the threshold of the dielectric to withstand electrical damage, and E i is the field strength at each level, i is the field strength at the i-th level, i=0,1,…,k-1,k; Δt is the boost time step, t k is the final field strength E k The duration of action; S4-2: According to the Bernoulli formula, equation (6) is simplified to equation (8): The ratio of ΔE / Δt is defined as h, t c is the total time of the experiment; (5) Based on step (3) and step (4), the damage eigenvalue method is used to calculate the n value of the inverse power model and the Dc corresponding to n. The life curve is fitted according to the obtained n and Dc, and the life prediction is performed based on the life curve.

2. The method for equivalent accelerated prediction of solid dielectric life according to claim 1, characterized in that: In the steps (1) and (2), a flat plate vulcanizing machine is used to hot-press the solid dielectric sample raw material into sheet-shaped solid dielectric samples of the same size.

3. The method for equivalent accelerated prediction of solid dielectric life according to claim 1, characterized in that: In the short-time breakdown test, the voltage is increased uniformly at a rate of v=1 kV / s until the sample breaks down.

4. The parameter selection method for equivalent acceleration to obtain solid dielectric life index according to claim 1 is characterized in that: In the step (4), the step stress test parameters include the initial field intensity E0, the boost step length ΔE, and the boost time step length Δt.

5. The method for equivalent accelerated prediction of solid dielectric life according to claim 1, characterized in that: In step S4-2, the Bernoulli formula is: In the formula are polynomial coefficients, and B is the Bernoulli number.

6. The method for equivalent accelerated prediction of solid dielectric life according to claim 1, characterized in that: The step (5), damage characteristic value method comprises the following steps: S5-1: Based on formula (8), m h values ​​are selected for each step stress test, and m groups of ΔE and Δt values ​​are selected at the same h value to perform step stress tests, where m ≥ 3; S5-2: Substitute the data of each level generated by the m×m step stress test in S5-1 into the cumulative damage threshold formula caused by step stress. Then the data obtained from the m×m tests are arranged in sequence to form an m×m square matrix D(n), and the rank of D(n) is 1, where n and Dc are unknown quantities; S5-3: Iterate the value of n starting from 0 and substitute n into formula (10) equal to 0 to calculate the corresponding eigenvalue, so that the eigenvalue of the D(n) matrix satisfies λ max =mDc: The m eigenvalues ​​corresponding to the D(n) matrix are λ1,λ2,λ3…,λ m , set the accuracy requirement α, set I = S / λ max , when I≤α, and the n value when I is the minimum value is the true n value of the solid dielectric, corresponding to Dc=λ max / m; where S = λ1+λ2+λ 3+ …+λ m-1 ,λ max =λ m ; If I≤α is not satisfied, then n=n+δn and the n value iteration in S5-3 is continued until I≤α is satisfied and the real n value and the corresponding Dc are calculated; The formula (10) is 7. The method for equivalent accelerated prediction of solid dielectric life according to claim 6, characterized in that: The cumulative damage threshold formula caused by the step stress is: E0 is the initial field strength, E i is the field strength at each level, i is the field strength at the i-th level, i=0,1,…,k-1,k; Δt is the boost time step, t k is the final field strength E k action time.

8. The method for equivalent accelerated prediction of solid dielectric life according to claim 1, characterized in that: In the step (5), the life curve is obtained by taking the logarithm of both sides of the equation (1) and fitting a life curve with the ordinate being lnE and the abscissa being lnt; Dc=tEn(1); Where t represents the remaining electrical life of the dielectric under the field strength E, n is the life index, and Dc represents the cumulative damage threshold of the dielectric to withstand electrical damage.

9. The method for equivalent accelerated prediction of solid dielectric life according to claim 6, characterized in that: The n and Dc of solid dielectric samples with the same specifications are constants.

Images