MOS field effect transistor failure physical modeling and estimation method considering quality consistency

Abstract

The invention discloses an MOS field effect transistor failure physical modeling and estimation method considering quality consistency, and belongs to the technical field of MOS field effect transistor performance and reliability analysis. In order to solve the problems that an existing MOS field effect transistor failure physical modeling scheme does not consider a performance parameter degradation process and a model coefficient in a degradation model cannot reflect the consistency of batch products, the method comprises the following steps: firstly, according to a typical failure mechanism of an MOS field effect transistor, determining a failure physical model general expression form considering quality consistency; the method comprises the following steps: carrying out MOS field effect transistor multi-stress acceleration test design and testing to obtain an acceleration test result, calculating a failure physical model coefficient by adopting a Copula-based multi-parameter correlation method to obtain an electronic component failure physical model, and further determining a failure physical model considering quality consistency. And finally, estimating according to the failure physical model considering the quality consistency.

Description

Technical Field

[0001] The invention belongs to the technical field of MOS field effect tube performance and reliability analysis, and relates to a MOS field effect tube failure physical modeling method. Background Art

[0002] MOS field effect tubes are widely used in various types of power electronic converters. With the continuous improvement of application requirements and the rapid development of process technology, the performance of MOS field effect tubes is getting better and better. However, in practical applications, MOS field effect tubes are affected by electrical, thermal and various complex environmental stresses, and have a high failure rate. Relevant statistics show that among all failures of power electronic converters, failures caused by semiconductor power devices account for more than 1 / 3. It is of great significance and application value to study the mechanisms of various types of failures of MOS field effect tubes and establish failure physical models for the main failure mechanisms, so as to apply them to the reliability analysis of MOS field effect tubes or power electronic converters.

[0003] At present, phased results have been achieved in the failure analysis, accelerated degradation test and failure physical modeling of MOS field effect tubes. Extensive research has been conducted on the main failure mechanisms of MOS field effect tubes, such as hot carrier injection, gate oxide breakdown over time, single particle gate breakdown, bias temperature instability, etc. The occurrence process of various failure mechanisms has been explained through simulation, experiments and other methods, and used for actual evaluation. Based on the performance parameter degradation data obtained from the accelerated degradation test and combined with the failure mechanism, a failure physical model has been established, and the reliability of the device has been further evaluated on this basis.

[0004] However, there are still some problems in the current research: (1) Most of the failure physical models established for MOS field effect tubes are life models, which only focus on the relationship between the device life (failure time) and stress, but not the degradation process of the device performance parameters, ignoring the reliability information implied in the device degradation process; (2) The failure physical modeling of MOS field effect tubes considering quality consistency is still blank. For domestic MOS field effect tubes, how to consider individual differences in the failure physical model and integrate quality consistency information needs to be solved urgently. Therefore, combining the structural characteristics and failure mechanism of MOS field effect tubes themselves, determining the functional form of its degradation model, and based on the manufacturing process data of MOS field effect tubes, the maximum likelihood method is used to estimate the distribution of random influence coefficients in the MOS field effect tube degradation model, so that a degradation model integrating quality consistency information can be constructed, and the foundation for further accurate evaluation of the reliability level of MOS field effect tubes can be laid. Summary of the invention

[0005] In order to solve the problems that the existing MOS field effect tube failure physical modeling scheme has not considered the performance parameter degradation process and the model coefficients in the degradation model cannot reflect the consistency of batch products, the present invention provides a MOS field effect tube failure physical modeling and estimation method taking quality consistency into consideration.

[0006] A MOS field effect tube failure physical modeling method considering quality consistency includes the following steps:

[0007] Step 1: According to the typical failure mechanism of MOS field effect tube, determine the general expression form of failure physical model considering quality consistency;

[0008] The failure physical model considering quality consistency includes performance degradation failure physical model and functional failure failure physical model; the performance degradation failure physical model coefficient represents all model coefficients in the corresponding model, and the model quality consistency is characterized by the initial value of the characteristic output Y0 and the degradation rate A of the performance degradation failure physical model. D It obeys the two-dimensional normal distribution, using μ A Denotes the degradation rate A D The mean value, σ A Denotes the degradation rate A D The standard deviation of the functional failure type failure physical model coefficient represents all model coefficients in the corresponding model, and the model quality consistency information is represented by the function F(·) rate coefficient A F Obey the normal distribution;

[0009] Step 2: Combined with the general expression of the failure physical model considering quality consistency constructed in step 1, carry out the multi-stress accelerated test design of MOS field effect tube;

[0010] Step 3: According to the accelerated test design of step 2, the accelerated test results are obtained, and the failure physical model coefficients are calculated using the Copula-based multi-parameter correlation method to obtain the failure physical model of the electronic components, and then determine the failure physical model that takes quality consistency into consideration.

[0011] Furthermore, the electronic component characteristics of the performance degradation failure physical model are output as follows:

[0012] Y(t)=Y0+D(S,t) (1)

[0013] Where Y(t) represents the characteristic output of the electronic component at time t, Y0 represents the initial value of the characteristic output, D(·) represents the performance degradation failure physical model, S represents the stress type set, and t represents time;

[0014] The model quality consistency of the performance degradation failure physical model is characterized by the degradation rate A of Y0 and function D(·) DIt obeys the two-dimensional normal distribution, that is, it satisfies formula (2):

[0015]

[0016] In the formula, μ0 represents the mean of Y0, σ0 represents the standard deviation of Y0, and μ A Denotes the degradation rate A D The mean value, σ A Denotes the degradation rate A D The standard deviation of , ρ represents the correlation coefficient;

[0017] The functional failure type failure physical model is constructed as follows:

[0018] τ=F(s) (3)

[0019] Among them, τ represents the life of electronic components, S represents the stress combination, and F(·) represents the failure mechanism as a failure type failure physical model;

[0020] The quality consistency information of the functional failure type failure physical model is represented by the function F(·) rate coefficient A F Obey the normal distribution, that is, satisfy

[0021]

[0022] Among them, μ F ,σ F They represent the functional failure type failure physical model corresponding to A F The mean and standard deviation of .

[0023] Furthermore, the specific process of carrying out the multi-stress accelerated test design of MOS field effect tube includes:

[0024] Select the limit stress and number of groups for the MOS field effect tube accelerated test: the selected maximum stress should not exceed the proportional threshold of the device limit stress, and the selected minimum stress is the device limit rated working condition;

[0025] The intermediate stress level is determined by the stress inverse equal interval formula, as shown in formula (4):

[0026]

[0027] Where Δ is the stress reciprocal interval; l is the number of selected stresses; S1 is the rated stress level; S Max is the ultimate stress level;

[0028] Selecting MOS field effect tube accelerated test samples: Based on the life mean and standard deviation of the samples, select MOS field effect tube accelerated test samples according to the confidence interval;

[0029] Determine the test time interval of MOS field effect tube accelerated test: Determine the test time interval of MOS field effect tube accelerated test based on the quality consistency of MOS field effect tube.

[0030] Furthermore, the threshold value of the ratio of the device limit stress is 90%.

[0031] Furthermore, based on the life mean and standard deviation of the samples, the process of selecting MOS field effect tube accelerated test samples according to the confidence interval includes:

[0032] First, the degradation data is expressed as life data determined by the failure threshold or the life data of the electronic components. Assuming that the mean life is μ and the standard deviation is σ, the true value of the average life of the electronic components can be determined to be μ.

[0033] Secondly, n samples are drawn from all life data, and the n samples show that the mean is μ and the variance is σ 2 / n normal distribution; using the distribution of the above n sample means, the confidence interval of the confidence level P is calculated by formula (5);

[0034]

[0035] Among them, z is the standard score corresponding to the corresponding confidence level;

[0036] Finally, according to the confidence interval of the obtained confidence level P, the error between the upper and lower limits of the confidence interval and the true life span is calculated by formula (6);

[0037]

[0038] Based on the error, the confidence level P under different quality consistency is generated, and the limit relative error of the average life sampling estimation is then selected to select the MOS field effect tube accelerated test samples.

[0039] Further, the process of determining the test time interval of the MOS field effect tube accelerated test based on the MOS field effect tube quality consistency includes the following steps:

[0040] The mean and standard deviation of the performance parameters or life data at the initial moment are counted. The mean and standard deviation of the initial value of the characteristic output are denoted as μ0 and σ0, and the mean and standard deviation of the life are denoted as μ τ , σ τ ;

[0041] If σ0 / μ0≤the characteristic-related coefficient of variation threshold, or σ τ / μ≤threshold value of coefficient of variation related to life span, the test interval gradually increases;

[0042] If σ0 / μ0> the characteristic-related coefficient of variation threshold, or στ / μ>life-related coefficient of variation threshold, the test interval should be gradually reduced.

[0043] Furthermore, the coefficient of variation threshold associated with the characteristic is 0.15.

[0044] Furthermore, the life span-related coefficient of variation threshold is 0.15.

[0045] Furthermore, the process of calculating the failure physical model coefficients includes:

[0046] Assume that the coefficient set of the failure physical model to be identified is θ = (θ1, θ2, …, θ n ), the test data follows a normal distribution; the given likelihood function L(θ) represents the probability of the observed data occurring when the parameter θ is given, and the calculation formula is shown in (7):

[0047]

[0048] In the formula, x i represents the degradation or lifetime data of the sample size, represents the joint probability distribution, L(θ) represents the likelihood function;

[0049] Taking the logarithm of the likelihood function gives the log-likelihood function:

[0050]

[0051] Find the optimal parameters based on the optimization algorithm and calculate the parameters that maximize the log-likelihood function logL(θ)

[0052]

[0053] Among them, arg max represents the variable that makes the posterior logL(θ) reach the maximum value. The value of

[0054] According to the estimated parameters The failure physical model parameters θ=(θ1,θ2,…,θ n ).

[0055] A method for estimating failure of a MOS field effect tube considering quality consistency is disclosed. For a MOS field effect tube, a physical modeling method for failure of a MOS field effect tube considering quality consistency is first used to model the MOS field effect tube, so as to obtain a MOS field effect tube failure model considering quality consistency corresponding to the MOS field effect tube, and then failure estimation of the MOS field effect tube is performed according to the MOS field effect tube failure model considering quality consistency.

[0056] Beneficial effects:

[0057] In the process of failure physical modeling of MOS field effect tubes, the present invention constructs a failure physical model expression that takes quality consistency into consideration, which can more accurately show the differences between individual MOS field effect tubes and more accurately characterize the degradation distribution of MOS field effect performance parameters under specific stress, which is of great significance for the reliability prediction of MOS field effect tubes and power electronic converters. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a flow chart of the MOS field effect tube failure physical modeling method considering quality consistency described in the present invention.

[0059] Figure 2 The upper and lower limits of the 90% confidence interval corresponding to model A.

[0060] Figure 3 The upper and lower limits of the 90% confidence interval corresponding to model B.

[0061] Figure 4 The upper and lower limits of the 90% confidence interval corresponding to model C.

[0062] Figure 5 The upper and lower limits of the 90% confidence interval corresponding to model D.

[0063] Figure 6 V corresponding to model A th Degradation model validation results.

[0064] Figure 7 V corresponding to model B th Degradation model validation results.

[0065] Figure 8 V corresponding to model C th Degradation model validation results.

[0066] Fig. 9 V corresponding to model D th Degradation model validation results. DETAILED DESCRIPTION

[0067] The technical solution of the present invention is further described below in conjunction with the accompanying drawings, but is not limited thereto. Any modification or equivalent replacement of the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention should be included in the protection scope of the present invention.

[0068] The present invention first constructs the basic form of the failure physical model between threshold voltage and stress according to the typical failure mechanism of MOS field effect tube. Then, the influence of key material, structure and process vector fluctuations on the model coefficients is analyzed, and the quality consistency information is introduced into the failure physical model to construct a failure physical model considering quality consistency. Secondly, the multi-stress accelerated test design of MOS field effect tube is carried out. Finally, the maximum likelihood method is used to identify the failure physical model parameters, and the model coefficients with distribution characteristics are obtained to complete the failure physical modeling of MOS field effect tube considering quality consistency. It mainly includes four steps, step one: construct a general form of MOS field effect tube failure physical model according to the failure mechanism; step two: accelerated test design; step three: failure physical model parameter identification and verification (not considering quality consistency comparison + only comparing the mean). The present invention is further described below in conjunction with specific embodiments.

[0069] This embodiment is a MOS field effect tube failure physical modeling method and a corresponding failure estimation method considering quality consistency. Figure 1 A method for physical modeling of MOS field effect tube failure considering quality consistency is described, which includes the following steps:

[0070] Step 1: According to the typical failure mechanism of MOS field effect tube, determine the general expression form of failure physical model considering quality consistency;

[0071] According to the typical failure mechanism of MOS field effect tube, the failure physical model considering quality consistency is specifically divided into performance degradation type failure physical model and functional failure type failure physical model. The specific model is determined according to the actual failure mechanism of electronic components.

[0072] A. The electronic component characteristics output of the performance degradation failure physical model are as follows:

[0073] Y(t)=Y0+D(S,t) (1)

[0074] Among them, Y(t) represents the characteristic output of the electronic component at time t, Y0 represents the initial value of the characteristic output, D(·) represents the performance degradation failure physical model, S represents the stress type set (sensitive stress type is selected according to the specific failure mechanism of the electronic component), and t represents time. The performance degradation failure physical model coefficient represents all model coefficients in D(·). The quality consistency of the model is characterized by the degradation rate A of Y0 and function D(·) D It obeys the two-dimensional normal distribution, that is, it satisfies formula (2):

[0075]

[0076] In the formula, μ0 represents the mean of Y0, σ0 represents the standard deviation of Y0, and μ ADenotes the degradation rate A D The mean value, σ A Denotes the degradation rate A D The standard deviation of the correlation coefficient is 0<|ρ|<1. When ρ>0, it means that the quality consistency of the product at the initial moment is positively correlated with the degradation rate. When ρ<0, it means that the quality consistency of the product at the initial moment is negatively correlated with the degradation rate. When ρ|>0.5, it means that the correlation between the two is strong. When |ρ|<0.5, it means that the correlation between the two is weak.

[0077] B. The functional failure type failure physical model is constructed as shown in formula (3):

[0078] τ=F(s) (3)

[0079] Where τ represents the life of electronic components, S represents the stress combination, F(·) represents the failure mechanism is the failure type failure physical model, the functional failure type failure physical model coefficient represents all model coefficients in F(·), and the quality consistency information of the model is represented by the function F(s) rate coefficient A F It obeys the normal distribution, that is, μ F ,σ F They represent the corresponding rate coefficient A of the functional failure type failure physical model F The mean and standard deviation of the function F(s) rate coefficient refers to the coefficient corresponding to the rate of change of the failure physical model;

[0080] Step 2: Combined with the general expression of the failure physical model considering quality consistency constructed in step 1, carry out multi-stress accelerated test design for MOS field effect tubes. The scope of application of the test design in this step covers performance degradation failure mechanism and functional failure mechanism:

[0081] Select the ultimate stress and number of groups for the accelerated test of MOS field effect tubes: the selected maximum stress should not exceed 90% of the device's ultimate stress, and the selected minimum stress is the device's ultimate rated operating condition.

[0082] The rated stress of MOS field effect tube is S0, and the limit stress that keeps the failure mechanism unchanged is S Max The intermediate stress level is determined by the stress reciprocal equal interval formula, as shown in formula (4):

[0083]

[0084] Where Δ is the stress reciprocal interval; l is the number of selected stresses; S1 is the rated stress level; S Max is the ultimate stress level.

[0085] Select MOS field effect tube accelerated test samples:

[0086] First, when there is enough life data of electronic components, the life data represents the life data of the electronic components, or represents degradation data to determine the life data according to the failure threshold. Assuming that the mean life value is μ and the standard deviation is σ, then it can be determined that the true value of the average life of the electronic component is μ.

[0087] Secondly, we extract n samples from all the life data. According to the central limit theorem, the average values ​​of these n samples will be around the average value μ of the overall sample, and will obey the mean μ and variance σ. 2 / n is a normal distribution.

[0088] Then, using the distribution of the above n sample means, the confidence interval of the confidence level P is calculated by formula (5).

[0089]

[0090] Among them, z is the standard score corresponding to the corresponding confidence level, which can be obtained by looking up the table. When the sample size n is less than 30, z is obtained by looking up the t distribution table; when the sample size n is not less than 30, it is obtained by looking up the standard normal distribution table.

[0091] Finally, according to the confidence interval of the obtained confidence level P, the error between the upper and lower limits of the confidence interval and the true life can be calculated by formula (6), thereby generating the relative error of the average life sampling estimate limit of the confidence level P under different quality consistency.

[0092]

[0093] Where μ is the actual mean life span, σ 2 / n is the actual standard deviation, z is the standard score corresponding to the corresponding confidence level, and n is the number of samples used for verification. If the error value is very small, it means that the boundary of the interval is very close to the true value of the average life, which means that when the sampling sample size is n, its average value is probability P very close to the true value of the average life. Therefore, when the error value is very small, the mean distribution of n samples can be used to characterize the true average life distribution of electronic components. It can be seen from formula (6) that the error value is related to the standard score z, the sampling sample size n, and the overall sample dispersion (i.e., quality consistency) σ / μ.

[0094] Determine the test time interval of the MOS field effect tube accelerated test: Statistical initial performance parameters or life data mean and standard deviation, characteristic output initial value mean and standard deviation are recorded as μ0, σ0, life mean and standard deviation are recorded as μ τ , σ τ .

[0095] If the quality consistency of MOS field effect tube is good (σ0 / μ0≤0.15 or στ / μ τ ≤0.15), the test interval gradually increases: Δt={Δt1<Δt2<Δt3…<Δt n-2 <Δt n-1};

[0096] If the quality consistency of MOS field effect tube is poor (σ0 / μ0>0.15 or σ τ / μ τ >0.15), the test interval should be gradually reduced: Δt={Δt1>Δt2>Δt3…>Δt n-2 >Δt n-1};

[0097] Step 3: According to the accelerated test design of step 2, the accelerated test results are obtained, and the failure physical model coefficient is calculated using the Copula-based multi-parameter correlation method (the model coefficient represents the performance degradation failure physical model coefficient or the functional failure type failure physical model coefficient) to obtain the electronic component failure physical model.

[0098] The process of calculating the failure physics model coefficients includes:

[0099] Assume that the coefficient set of the failure physical model to be identified is θ = (θ1, θ2, …, θ n ), the test data follows a normal distribution. Given the likelihood function L(θ), it represents the probability of the observed data occurring when the parameter θ is given. The calculation formula is shown in (7):

[0100]

[0101] In the formula, x i represents the degradation or lifetime data of the sample size, represents the joint probability distribution, and L(θ) represents the likelihood function.

[0102] In order to simplify the calculation, the logarithm of the likelihood function is usually taken to obtain the logarithmic likelihood function:

[0103]

[0104] Find the optimal parameters based on the optimization algorithm and calculate the parameters that maximize the log-likelihood function logL(θ)

[0105]

[0106] Among them, arg max represents the variable that makes the posterior logL(θ) reach the maximum value. The value of .

[0107] According to the estimated parameters The failure physical model parameters θ=(θ1,θ2,…,θ n ).

[0108] At this point, the MOS field effect tube failure physical modeling considering quality consistency is completed to obtain the failure physical model considering quality consistency. The model can be used to estimate the failure of the MOS field effect, that is: during the actual estimation, for the MOS field effect tube, first use the MOS field effect tube failure physical modeling method considering quality consistency to model it, and obtain the MOS field effect tube failure model considering quality consistency corresponding to the MOS field effect tube, and then perform failure estimation of the MOS field effect tube according to the MOS field effect tube failure model considering quality consistency.

[0109] Example:

[0110] This embodiment uses the typical failure mechanism of MOS field effect transistor, bias temperature instability (BTI), as an example to verify the effectiveness of the proposed method, and bias temperature instability is a performance degradation type failure physical model. The modeling ideas and processes of the functional failure type failure physical model and the performance degradation type failure physical model are the same.

[0111] Construct its threshold voltage V th With the impact of V th The basic form of the failure physical model of sensitive stress voltage V and temperature T is:

[0112]

[0113] In the formula, V th (t) is the threshold voltage at time t, corresponding to Y(t) in formula (1); V th (0) is the initial value of the threshold voltage, corresponding to Y0 in formula (1); A BTI is the BTI effect factor, B is the electric field acceleration factor, E a is the failure activation energy, k is the Boltzmann constant, and C is the time power exponent. BTI It follows a normal distribution.

[0114] Sample selection for MOS field effect tube accelerated test: For the number of samples used in the test, the quality consistency information of the initial value of the threshold voltage of the MOS field effect tube is determined to be 0.1 (σ0 / μ0). The confidence interval of the MOS field effect tube at a 90% confidence level is calculated according to formula (5). Based on the confidence interval at a 90% confidence level obtained, the error between the upper and lower limits of the confidence interval and the true value can be calculated by formula (6), thereby generating a quick lookup table of the relative error of the average life sampling estimation limit at a 90% confidence level. Some results are shown in Table 1.

[0115] Table 1 Quick reference table of relative error limit of mean life sampling estimation at 90% confidence level

[0116]

[0117] According to the results shown in Table 1, in order to ensure the representativeness of the test data, at least 22 samples are required under the verification stress.

[0118] Determine the accelerated test stress of MOS field effect transistor: On this basis, consider the level and combination of applied stress. Under the NBTI effect, the key influencing stresses of PMOSFET are the gate-source drive voltage and ambient temperature. Combined with the relevant standards, the number of existing test samples and the data requirements of subsequent modeling work, it can be determined that each stress should be divided into three levels. According to the degradation of the device under different stresses in the baseline test, the degradation rate of the device can be estimated, and then the range of applied stress can be determined. Taking into account the above information, the driving voltage is divided into three levels: -36V, -38V, and -40V, and the ambient temperature is divided into three levels: 348K, 378K, and 408K, which are cross-combined into 5 groups.

[0119] Table 2 Statistics of the number of samples under various stresses in the accelerated test (“ / ” in the specific data indicates no sample)

[0120]

[0121] Determine the test time interval of the MOS field effect tube accelerated test: Since the quality consistency information of the initial value of the MOS field effect tube threshold voltage is 0.1 (σ0 / μ0), the test interval of this accelerated test is: the initial test interval is once a day, the duration is 14 days, and then the test is once every two or three days until the threshold voltage reaches the failure threshold.

[0122] According to formula (10) and the data obtained from the accelerated test, μ A , σ A , B, E a The maximum likelihood estimation of C, where the threshold voltage degradation model of a certain type of MOS field effect tube is:

[0123]

[0124] 1000, 5000 and 10000 degradation trajectories were randomly generated under the validation stress, and the upper and lower limits of the 90% confidence interval were plotted. The results are shown in Figure 2-Figure 5 As shown, Figure 2-Figure 5 are the upper and lower limits of the 90% confidence interval corresponding to model AD, respectively. Figure 2-Figure 5 It can be seen from the figure that when the sample population is 1000, the degradation trajectory can represent the degradation distribution of the population. The comparison results are as follows Figure 6-Figure 9 As shown, Figure 6-Figure 9 They are V corresponding to models AD respectively. thDegradation model validation results show that most of the actual degradation data are within the confidence interval.

[0125] Assume that the time required for the threshold voltage to degrade by 5% is t 5％ , respectively calculate the expected t under the verification stress 5％ And the actual 5％ The average value of the average value is calculated, and the relative error of the average value is calculated. The results are shown in Table 1.

[0126] Table 1 Comparison between expected and actual values

[0127]

[0128] According to the results in the table, when the mean is used as the model accuracy evaluation index, the model error is less than 10%.

[0129] The above calculation examples of the present invention are only used to explain the calculation model and calculation process of the present invention in detail, and are not intended to limit the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A MOS field effect tube failure physical modeling method considering quality consistency, characterized in that: The following steps are involved: Step 1: According to the typical failure mechanism of MOS field effect tube, determine the general expression form of failure physical model considering quality consistency; The failure physical model considering quality consistency includes performance degradation failure physical model and functional failure failure physical model; the performance degradation failure physical model coefficient represents all model coefficients in the corresponding model, and the model quality consistency is characterized by the initial value of the characteristic output Y0 and the degradation rate A of the performance degradation failure physical model. D It obeys the two-dimensional normal distribution, using μ A Denotes the degradation rate A D The mean value, σ A Denotes the degradation rate A D The standard deviation of the functional failure type failure physical model coefficient represents all model coefficients in the corresponding model, and the model quality consistency information is represented by the function F(·) rate coefficient A F Obey the normal distribution; Step 2: Combined with the general expression of the failure physical model considering quality consistency constructed in step 1, carry out the multi-stress accelerated test design of MOS field effect tube; Step 3: According to the accelerated test design of step 2, the accelerated test results are obtained, and the failure physical model coefficients are calculated using the Copula-based multi-parameter correlation method to obtain the failure physical model of the electronic components, and then determine the failure physical model that takes quality consistency into consideration.

2. A MOS field effect tube failure physical modeling method considering quality consistency according to claim 1, characterized in that: The electronic component characteristics output of the performance degradation failure physical model are as follows: Y(t)=Y0+D(S,t) (1) Where Y(t) represents the characteristic output of the electronic component at time t, Y0 represents the initial value of the characteristic output, D(·) represents the performance degradation failure physical model, S represents the stress type set, and t represents time; The model quality consistency of the performance degradation failure physical model is characterized by the degradation rate A of Y0 and function D(·) D It obeys the two-dimensional normal distribution, that is, it satisfies formula (2): In the formula, μ0 represents the mean of Y0, σ0 represents the standard deviation of Y0, and μ A Denotes the degradation rate A D The mean value, σ A Denotes the degradation rate A D The standard deviation of , ρ represents the correlation coefficient; The functional failure type failure physical model is constructed as follows: τ=F(s) (3) Among them, τ represents the life of electronic components, S represents the stress combination, and F(·) represents the failure mechanism as a failure type failure physical model; The quality consistency information of the functional failure type failure physical model is represented by the function F(·) rate coefficient A F Obey the normal distribution, that is, satisfy Among them, μ F ,σ F They represent the functional failure type failure physical model corresponding to A F The mean and standard deviation of .

3. A MOS field effect tube failure physical modeling method considering quality consistency according to claim 2, characterized in that: The specific process of carrying out the multi-stress accelerated test design of MOS field effect tubes includes: Select the ultimate stress and number of groups for the accelerated test of MOS field effect tubes: the selected maximum stress should not exceed the proportional threshold of the device ultimate stress, and the selected minimum stress is the device's ultimate rated working condition; The intermediate stress level is determined by the stress inverse equal interval formula, as shown in formula (4): Where Δ is the stress reciprocal interval; l is the number of selected stresses; S1 is the rated stress level; S Max is the ultimate stress level; Selecting MOS field effect tube accelerated test samples: Based on the life mean and standard deviation of the samples, select MOS field effect tube accelerated test samples according to the confidence interval; Determine the test time interval of MOS field effect tube accelerated test: Determine the test time interval of MOS field effect tube accelerated test based on the quality consistency of MOS field effect tube.

4. A MOS field effect tube failure physical modeling method considering quality consistency according to claim 3, characterized in that: The threshold value of the ratio of the device limit stress is 90%.

5. The method for physical modeling of MOS field effect tube failure considering quality consistency according to claim 3, characterized in that: Based on the life mean and standard deviation of the samples, the process of selecting MOS field effect tube accelerated test samples according to the confidence interval includes: First, the degradation data is expressed as life data determined by the failure threshold or the life data of the electronic components. Assuming that the mean life is μ and the standard deviation is σ, the true value of the average life of the electronic components can be determined to be μ. Secondly, n samples are drawn from all life data, and the n samples show that the mean is μ and the variance is σ 2 / n normal distribution; using the distribution of the above n sample means, the confidence interval of the confidence level P is calculated by formula (5); Among them, z is the standard score corresponding to the corresponding confidence level; Finally, according to the confidence interval of the obtained confidence level P, the error between the upper and lower limits of the confidence interval and the true life span is calculated by formula (6); Based on the error, the confidence level P under different quality consistency is generated, and the limit relative error of the average life sampling estimation is then selected to select the MOS field effect tube accelerated test samples.

6. The method for physical modeling of MOS field effect tube failure considering quality consistency according to claim 3, characterized in that: The process of determining the test time interval of the MOS field effect tube accelerated test based on the MOS field effect tube quality consistency includes the following steps: The mean and standard deviation of the performance parameters or life data at the initial moment are counted. The mean and standard deviation of the initial value of the characteristic output are denoted as μ0 and σ0, and the mean and standard deviation of the life are denoted as μ τ , σ τ ; If σ0 / μ0≤the characteristic-related coefficient of variation threshold, or σ τ / μ≤threshold value of coefficient of variation related to life span, the test interval gradually increases; If σ0 / μ0> the characteristic-related coefficient of variation threshold, or σ τ / μ>life-related coefficient of variation threshold, the test interval should be gradually reduced.

7. A MOS field effect tube failure physical modeling method considering quality consistency according to claim 6, characterized in that: The coefficient of variation threshold associated with the characteristics was taken as 0.

15.

8. The method for physical modeling of MOS field effect tube failure considering quality consistency according to claim 6, characterized in that: The threshold value of the coefficient of variation related to life span is 0.

15.

9. A MOS field effect transistor failure physical modeling method considering quality consistency according to any one of claims 2 to 8, characterized in that: The process of calculating the failure physics model coefficients includes: Assume that the coefficient set of the failure physical model to be identified is θ = (θ1, θ2, …, θ n ), the test data follows a normal distribution; the given likelihood function L(θ) represents the probability of the observed data occurring when the parameter θ is given, and the calculation formula is shown in (7): In the formula, x i represents the degradation or lifetime data of the sample size, represents the joint probability distribution, L(θ) represents the likelihood function; Taking the logarithm of the likelihood function gives the log-likelihood function: Find the optimal parameters based on the optimization algorithm and calculate the parameters that maximize the log-likelihood function logL(θ) Among them, arg max represents the variable that makes the posterior logL(θ) reach the maximum value. The value of According to the estimated parameters The failure physical model parameters θ=(θ1,θ2,…,θ n ).

10. A MOS field effect tube failure estimation method considering quality consistency, characterized in that: For MOS field effect transistors, firstly, a MOS field effect transistor failure physical modeling method considering quality consistency as described in any one of claims 1 to 9 is used for modeling to obtain a MOS field effect transistor failure model considering quality consistency corresponding to the MOS field effect transistor, and then failure estimation of the MOS field effect transistor is performed according to the MOS field effect transistor failure model considering quality consistency.

Images