Failure physical model migration learning method for mos field effect transistor, storage medium and device

Abstract

The application discloses a MOS field effect transistor failure physical model migration learning method, a storage medium and equipment, and belongs to the technical field of MOS field effect transistor bias temperature instability failure physical modeling. In order to solve the problem that the existing migration learning method cannot perform migration learning on the failure mechanism derivation law of the MOS field effect transistor failure physical model, thereby leading to an unsatisfactory model effect after migration, the application combines historical data and a simulation analysis method, constructs different candidate source field correlation models, determines key parameters affecting failure physical model coefficients, then calculates the distance between different candidate source fields and a target field according to the correlation models, and constructs a MOS field effect transistor migration discriminant criterion; then, an electronic component failure physical parameter identification model of the candidate source field is established, and based on the migration domain discriminant criterion, an instance migration learning method is adopted to construct a parameter identification model of the failure physical model coefficients of the target field and the affected failure physical model coefficients.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of MOS field effect tube bias temperature instability failure physical modeling, and relates to a MOS field effect tube failure physical model transfer learning method. BACKGROUND

[0002] The failure physical model is a model for quantitatively describing the internal physical and chemical processes of a basic unit (material, structure, process, etc.) of an electronic component under various failure inducements (electricity, heat, vibration stress, etc.) to cause failure phenomena. MOS field effect tubes are widely used in various types of electronic systems to ensure reliable operation of the electronic systems. However, in actual applications, the MOS field effect tubes are affected by various complex task profiles, which leads to their degradation and further affects the reliability of the electronic systems. Relevant statistics show that the MOS field effect tube degradation accounts for more than 40% of the total electronic system failure. It is of great significance and application value to establish a failure physical model for the typical failure mechanism of the MOS field effect tube and apply it to the reliability analysis of the electronic system.

[0003] At present, for the typical failure mode and failure mechanism of the MOS field effect tube, the performance parameter degradation data obtained based on the accelerated degradation test is combined with the typical task profile of the MOS field effect tube to establish a failure physical model. However, there are still some problems in the current research: the failure physical model established at present is only for a specific failure mechanism of a specific MOS field effect tube, and does not have the ability to generalize to other specifications of the same type of MOS field effect tube. Therefore, how to establish a model with universal applicability is a key problem in the reliability prediction method based on the failure physics. For the MOS field effect tubes with the same failure mechanism, the essence of the change of the failure physical model is caused by the change of the key materials, structure and process of the bottom layer. By analyzing the material, structure and process characteristics, a transfer model representing the relationship between the material, structure, process parameters and the failure physical model coefficients can be established, which can effectively solve the problems of high modeling cost, long cycle and poor applicability of the failure physical model based on the degradation test, and provide protection for reliability prediction. SUMMARY

[0004] The existing transfer learning method cannot perform transfer learning on the failure mechanism derivation law of the MOS field effect tube failure physical model, so that the effect of the transferred model is not ideal.

[0005] A MOS field effect tube failure physical model transfer learning method, comprising:

[0006] Step S1, combining historical data and a simulation analysis method, a different candidate source field correlation model is constructed, that is, a MOS field effect tube failure physical model coefficient set and the key parameter set between the key parameters and the failure physical model coefficients; the key parameters include materials, structures, and process parameters; the complex correlation coefficients are ranked according to the calculation results, and the top several parameters are selected as the key parameters affecting the failure physical model coefficients, denoted as

[0007] Step S2, according to the complex correlation model of step S1, the distances between different candidate source domains and the target domain are calculated, and a MOS field effect tube transferability judgment criterion is constructed:

[0008]

[0009] In the formula, Z is a similarity order number, and rank represents a ranking function; represents the distance mean value of the kth candidate source domain and the target domain, represents the distance variance of the kth candidate source domain and the target domain.

[0010] Step S3, for the candidate source domain, the key parameters in the candidate source domain are taken as inputs, the failure physical model coefficients affected thereby are taken as outputs, and an electronic component failure physical parameter identification model of the candidate source domain is established;

[0011] Step S4, according to the transferable domain judgment criterion of step S2, an instance transfer learning method is used to construct a parameter identification model of the target domain and the failure physical model coefficients affected thereby.

[0012] Further, the process of constructing complex correlation models of different candidate source domains includes the following steps:

[0013] Step S11, the MOS field effect tube manufacturing process is analyzed to determine

[0014] Step S12, according to the key parameters and the corresponding failure physical model coefficients obtained in step S11, a complex correlation model of the key parameters and the failure physical model coefficients is constructed, as shown in formulas (1) and (2):

[0015]

[0016] In the formula, is a complex correlation model coefficient set, P represents an actual value of a model coefficient, and P corresponds to the coefficient in for the source domain, represents a mean value of the model coefficient; R represents a complex correlation coefficient, and the value range is [0, 1]; the greater the complex correlation coefficient, the closer the correlation degree between the key parameters and the failure physical model coefficients.

[0017] Step S13: Sort the complex correlation coefficients according to the calculation results and select the top s (k)n The parameters are key parameters that affect the failure physical model coefficients. The key parameters selected according to the calculation results are recorded as

[0018] Furthermore, the key parameter set in step S11 Select [T ox ,W,L,N ch ,N sub ], where T ox is the gate oxide thickness, W is the channel width, L is the channel length, N ch is the channel doping concentration, N sub is the substrate doping concentration.

[0019] Furthermore, the actual value of the model coefficient in step S12 is represented

[0020]

[0021] Furthermore, the key parameters in step S13 are For [T ox ,W,L].

[0022] Furthermore, the specific process of constructing the criterion for determining the mobility of the MOS field effect transistor in step S2 includes the following steps:

[0023] Step S21: Assume that the target domain is Datasets in Each candidate source field has a labeled dataset, and the labeled dataset of the kth candidate source field is represented as Then the MMD distance of the key parameters of the candidate source domain and the target domain is shown in formula (3):

[0024]

[0025] Where, Characterizes the maximum mean difference between the candidate source domain and the target domain, X T Indicates D T All instance features in Indicates D s(k) All instance features in ; Indicates D s(k) The jth and j'th instance features in the candidate source domain are the key parameters that affect the failure physical model coefficients; i 、x i' Indicates D T The i-th and i'th instance features in are the key parameters that affect the coefficients of the failure physical model in the target domain; is the number of instances of the candidate source domain dataset; n T is the number of instances of the target domain dataset; k(·,·) represents a kernel function in the RKHS space;

[0026] Step S22, calculate the variance of different candidate source domains and the target domain, to obtain the variance of the candidate source domain and the target domain, as shown in formula (5):

[0027]

[0028] In the formula, k(·,·) represents a kernel function in the RKHS space;

[0029] Step S23, use the ordinal function to sort the mean and variance of the distance of different candidate source domains and the target domain, to form a transferability discrimination criterion

[0030] Further, the process of establishing the electronic component failure physical parameter identification model of the candidate source domain in step S3 includes the following steps:

[0031] Step S31, use a neural network to construct a parameter identification model reflecting the key parameters and failure physical model coefficients ;

[0032] Step S32, use an optimization algorithm to construct a parameter identification model of the adaptive search parameter unit and the best combination of epoch.

[0033] Further, the specific process of step S4 includes the following steps:

[0034] Step S41, select the candidate source domain dataset T closest to the target domain D to construct a new dataset represents data set fusion;

[0035] Step S42, initialize the sample weight ω Ti of the source domain and the target domain Si , and construct a target domain failure physical parameter identification model based on the TrAdaBoost algorithm; wherein the instance feature of the new dataset is the key parameter, the instance space label is the failure physical model coefficient, and the mapping model established is the target domain failure physical parameter identification model;

[0036] The failure physical parameter identification model is established as shown in formula (6):

[0037] p D = F(x D) (6)

[0038] In the formula, F represents a parameter identification model constructed based on AdaBoost, P D represents an instance space label of the reconstructed data set D, x D represents an instance feature of the reconstructed data set;

[0039] Step S43, calculate the model coefficient prediction result and the actual model coefficient error, and the error function is defined as:

[0040]

[0041] wherein e iD is the relative error of sample i D , is the label space true value, characterizes error normalization;

[0042] Step S44, according to the model error of step S43, update the source domain data set D s(k) weight until the error of formula (6) meets the requirements, and the weight update formula is as follows:

[0043] ω T(N+1) = ω TN · β (7)

[0044] ω S(N+1) = ω SN · (1-β) (8)

[0045] In the formula, ω TN , ω T(i+1) represent the current target domain weight and the adjusted target domain weight; β represents an adjustment coefficient; ω SN , ω S(i+1) represent the current source domain weight and the adjusted source domain weight, and N represents the number of iterations.

[0046] A computer storage medium, the storage medium stores at least one instruction, the at least one instruction is loaded and executed by the processor to realize the MOS field effect tube failure physical model transfer learning method.

[0047] A MOS field effect tube failure physical model transfer learning device, characterized in that the device comprises a processor and a memory, and the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to realize the MOS field effect tube failure physical model transfer learning method.

[0048] Advantages:

[0049] The present invention provides a method for transferring and learning the failure physical model of electronic components. The method reveals the evolution law of MOS field effect tube materials, structures, processes and their internal failure mechanisms, analyzes the interaction law between key materials, structures, processes and their performance parameters and degradation processes, and quantitatively establishes the law of key parameters and failure physical model coefficients to form a migration model for the failure physical models established for several specifications of products through an inductive summary method. Therefore, the present invention can effectively improve the effect of the model after migration. Furthermore, through the migration model, the failure physical model coefficients of any new specification product can be quickly extrapolated and calculated to achieve the rapid establishment of the failure physical model, effectively saving the resources and time required for MOS field effect tube failure physical modeling, and laying the foundation for accurately establishing a MOS field effect tube reliability prediction model. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a flow chart of a MOS field effect transistor failure physical model transfer learning method;

[0051] Figure 2 The result diagram of the target domain failure physical model coefficient A based on the TradaBoost parameter identification method;

[0052] Figure 3 The result diagram of the target domain failure physical model coefficient B based on the TradaBoost parameter identification method;

[0053] Figure 4 The result diagram of the target domain failure physical model coefficient Ea based on the TradaBoost parameter identification method;

[0054] Figure 5 The result diagram of the target domain failure physical model coefficient C based on the TradaBoost parameter identification method;

[0055] Figure 6 This shows the comparison between the validation specification 1 based on the TradaBoost fitting results and the actual degradation data.

[0056] Figure 7 This shows the comparison between the TradaBoost fitted degradation data and the actual degradation data for validation specification 2.

[0057] Figure 8 This shows the comparison between the validation specification 3 based on the TradaBoost fitting results and the actual degradation data.

[0058] Figure 9 This figure shows the comparison between the validation specification 4 based on the TradaBoost fitting results and the actual degradation data. DETAILED DESCRIPTION

[0059] To address the aforementioned issues in the existing technology, the present invention provides a method for transferring learning the failure physical model of MOS field-effect transistors. This method first combines historical data and simulation analysis methods to construct a complex correlation coefficient model between key parameters and failure physical models. Then, an improved maximum mean difference method is used to calculate the distance between key parameters of MOS field-effect transistors made of different materials, thereby constructing a transferable domain model for MOS field-effect transistors made of different materials. Subsequently, a functional model is constructed between the parameter vectors of the same material and the failure physical model coefficients affected by them. Finally, an example transfer learning method is used to construct a parameter identification and transfer model between the key material, structure, and process vectors of MOS field-effect transistors made of different materials and the failure physical model coefficients affected by them.

[0060] This method reveals the evolutionary laws of MOS field effect tube materials, structures, processes and their internal failure mechanisms, analyzes the interaction between key materials, structures, processes and their performance parameters and degradation processes, and establishes a migration model for different key materials, structures, processes and failure physical model coefficients, effectively saving the resources and time required for MOS field effect tube failure physical modeling. The present invention is further described below in conjunction with specific embodiments. Before proceeding with the specific description, the parameters and definitions are first explained:

[0061] (1) Key parameters: electronic component materials, structures, and process parameters that affect the failure physical model coefficients.

[0062] (2) Candidate source fields: represents the candidate source domain, that is, there are a large number of labeled samples available for learning, Denotes the candidate source domain feature space. s(k) represents the kth candidate source domain dataset, Represents the instance characteristics of the candidate source domain, that is, the key parameters affecting the failure physical model, specifically expressed as Represents the source domain instance space annotation, that is, the failure physical model coefficient (the mathematical model coefficient of the MOS field effect tube paranoid temperature instability failure mechanism), specifically expressed as is the number of instances in the candidate source domain dataset.

[0063] (3) Target areas: represents the target domain, which usually has fewer labeled samples to learn from. represents the target domain feature space, P T (x) represents the marginal probability distribution of the target domain data. T represents the target domain dataset, x i is the target domain instance feature, and the target domain instance feature is expressed as Represents instance space annotation, and the target domain instance space annotation is represented as n T is the number of instances in the target domain dataset.

[0064] Specific implementation method 1: Combination Figure 1 To explain this embodiment,

[0065] This embodiment is a method for transferring and learning a failure physical model of a MOS field effect transistor, including the following steps:

[0066] Step S1: Combine historical data and simulation analysis methods to construct complex correlation models for different candidate source fields, that is, to reflect the set of coefficients of the MOS field effect tube failure physical model. With the key parameter set The multiple correlation model between

[0067] The specific steps are as follows:

[0068] Step S11: Analyze the MOS field effect tube manufacturing process to determine Its specific form depends on the actual situation. In this embodiment, the materials, structures, and process parameters related to the failure physical model are selected [T ox ,W,L,N ch ,N sub ], where T ox is the gate oxide thickness, W is the channel width, L is the channel length, N ch is the channel doping concentration, N sub is the substrate doping concentration.

[0069] Step S12: The key parameters obtained in step S11 [T ox ,W,L,N ch ,N sub ] and the corresponding failure physical model coefficients Construct a complex correlation model between key parameters and failure physical model coefficients, as shown in formulas (1) and (2):

[0070]

[0071] Where, It is a set of complex correlation model coefficients, P represents the actual value of the model coefficient, and for the source domain P corresponds to The coefficients in the target domain P correspond to The coefficients in ; represents the average value of the model coefficient; R represents the multiple correlation coefficient, and its value range is [0,1]. The larger the multiple correlation coefficient, the closer the correlation between the key parameters and the failure physical model coefficient.

[0072] Step S13: Sort the complex correlation coefficients according to the calculation results, select some parameters as key parameters affecting the failure physical model coefficients (select according to the complex correlation coefficient R result), and record the key parameters selected according to the calculation results as (Specifically, in this embodiment, [T ox ,W,L]).

[0073] The same method as the candidate source field is used to establish a complex correlation model for the target field. The materials, structures, and process parameters related to the failure physical model are selected. After verification, the influence of the failure physical model coefficients of the target field is basically consistent. The number of some parameters for the target field is selected to be the same as the number of some parameters in the source field.

[0074] Step S2: Based on the complex correlation model of step S1 and the improved maximum mean difference method, the distances between different candidate source domains and target domains are calculated to construct a criterion for determining the transferability of MOS field effect transistors. The specific steps are as follows:

[0075] Step S21: Assume that the target domain is Datasets in There are a large number of labeled datasets in each candidate source field. The labeled dataset of the kth candidate source field is represented as Then the MMD distance of the key parameters of the candidate source domain and the target domain is shown in formula (3):

[0076]

[0077] Where, Characterizes the maximum mean difference between the candidate source domain and the target domain, X T Indicates D T All instance features in Indicates D s(k) All instance features in ; Indicates D s(k) The jth and j'th instance features in the candidate source domain are the key parameters that affect the failure physical model coefficients; i 、x i' Indicates D T The i-th and i'th instance features in are the key parameters that affect the coefficients of the failure physical model in the target domain; is the number of instances in the candidate source domain dataset; n T is the number of instances in the target domain dataset; k(·,·) represents the kernel function in the RKHS space.

[0078] Step S22: Calculate the variance between different candidate source domains and target domains according to the statistical knowledge variance calculation formula. The calculation process is as shown in formula (5):

[0079] Variance formula:

[0080]

[0081] Where E(x) represents the expectation and x represents the data set to be calculated.

[0082] Substituting formula (3) into formula (4), we can obtain the variance of the candidate source domain and the target domain, as shown in formula (5):

[0083]

[0084] Where k(·,·) represents the kernel function in the RKHS space;

[0085] Step S23: Use an ordinal function to sort the distance means and variances between different candidate source domains and target domains to form a transferability criterion. The specific formula is shown in (5):

[0086]

[0087] Wherein, the similarity ordinal number Z ranges from [1, M] and is an integer. represents the mean distance between the kth candidate source domain and the target domain, represents the distance variance between the kth candidate source domain and the target domain; rank represents the ranking function.

[0088] When Z takes the minimum value of 1, it means that the similarity between the candidate source domain and the target domain is the smallest; when Z takes the maximum value of M, it means that the similarity between the candidate source domain and the target domain is the largest.

[0089] Step S3: For the candidate source field, the key parameters in the candidate source field are As input, the failure physics model coefficients affected by it As an output, a physical parameter identification model of electronic component failure in the candidate source area is established; the specific steps are as follows:

[0090] Step S31: Use a neural network to construct a system that reflects the key parameters and the failure physics model coefficients Parameter identification model.

[0091] Step S32: Using the bat optimization algorithm, a parameter identification model for adaptively searching for the best combination of parameters unit and epoch is constructed.

[0092] Step S4: Based on the transferable domain discrimination criteria in step S2, the target domain is constructed using the instance transfer learning method. and the failure physical model coefficients affected by it The parameter identification model; the specific steps are as follows:

[0093] Step S41, selecting the candidate source domain dataset closest to the target domain D T Construct a new dataset Indicates that the dataset is fused.

[0094] Step S42, initialize the sample weight ω of the source domain and the target domain Ti = ω Si , based on the TrAdaBoost algorithm to construct the target domain failure physical parameter identification model, that is, although the data of the candidate source domain and the target domain are different, there will be a part of the candidate source domain similar to the target domain data, find out those instance features suitable for the target domain from the candidate source domain data, and reconstruct the data set with these instance features and target domain features Take the new data set instance feature x D as input, the new data set instance space label p D as output, and establish a mapping model based on TrAdaBoost. Wherein, the instance feature of the new data set is the key parameter, the instance space label is the failure physical model coefficient, and the mapping model established is the target domain failure physical parameter identification model.

[0095] The failure physical parameter identification model is established as shown in formula (6):

[0096] p D = F(x D ) (6)

[0097] In the formula, F represents the parameter identification model constructed based on AdaBoost, P D represents the instance space label of the reconstructed data set D, and x D represents the instance feature of the reconstructed data set.

[0098] Step S43, calculate the model coefficient prediction result and the actual model coefficient error, and the error function is defined as:

[0099]

[0100] Wherein, e iD is the relative error of sample i D , is the label space true value, characterizes the error normalization;

[0101] Step S44, according to the model error of step S43, update the source domain dataset D s(k) ​Weight, until the error of formula (6) meets the requirements, the weight update formula is as follows;

[0102] ω T(N+1) =ω TN β (7)

[0103] ω S(N+1) =ω SN ·(1-β) (8)

[0104] Where ω TN 、ω T(i+1) represents the current target domain weight and the adjusted target domain weight; β represents the adjustment coefficient; ω SN 、ω S(i+1) Represents the current source domain weight and the adjusted source domain weight, and N represents the number of iterations.

[0105] Example:

[0106] Step S1: Determine the MOS field effect tube failure physical model

[0107]

[0108] In the formula, V th (t) is the threshold voltage at time t, V th (0) is the initial value of the threshold voltage, 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.

[0109] Combining historical data and simulation analysis methods, a physical model coefficient P=[P1, P2, ... P n ] and the materials, structures, and processes of MOS field effect tubes X=[x1,x2,...,x n ] to determine the key parameters that affect the failure physical model coefficients:

[0110] First, combining historical data and simulation methods, the key parameters of MOS field effect tube are determined as follows: M=[T ox ,W,L,N ch ,N sub ]; where T ox is the gate oxide thickness, W is the channel width, L is the channel length, N ch is the channel doping concentration, N sub is the substrate doping concentration, and the failure physical model coefficient is [A, B, E a ,C].

[0111] Then, according to formula (1) and formula (2), the gate oxide thickness, channel width, channel length, channel doping concentration, N sub is the complex correlation model between the substrate doping concentration and the failure physical model coefficient, and the complex correlation coefficient is calculated. The specific results are shown in Table 1:

[0112] Table 1 Multiple correlation coefficients of different parameter combinations

[0113]

[0114] Finally, according to the results in Table 1, T ox The complex correlation coefficient under the combination of W and L parameters is the largest, so the key parameter combination affecting the MOS failure physical model is determined to be T ox ,W,L, that is, the first three parameters are selected as the key parameters affecting the failure physical model coefficient.

[0115] Step S2: Based on the complex correlation model of step S1, an improved maximum mean difference method is proposed to calculate the distance between the target domain MOS field effect transistor and the candidate source domain MOS field effect transistor. The detailed construction parameters of the candidate source domain dataset experiment are shown in Table 2, where D s(1) -D s(6) Indicates candidate source areas, and 1-6 represent codes for different materials.

[0116] Table 2 Experimental settings of the bi-moon artificial dataset

[0117]

[0118] The effectiveness of MMD variance divergence in measuring the distribution differences between domains: The experimental results are shown in Table 3

[0119] Table 3 MMD distance Distance variance σ 2 , MMD similarity order Z and transfer accuracy Acc

[0120]

[0121] Table 3 lists the target areas and candidate source fields MMD distance between Distance variance σ 2 , MMD similarity ordinal Z, gives the transfer learning algorithm TrAdaBoost, which is used to transfer each candidate source domain to the target domain, and the transfer accuracy Acc. As the k value increases, the candidate source domain and target areas The larger the actual distribution difference is, the greater the difference is. The following conclusions can be drawn from the analysis of the experimental results: As the distribution difference between the candidate source domain and the target domain gradually increases, the MMD distance between the target domain and the candidate source domain obtained by calculation Distance variance σ 2 , the MMD similarity ordinal number Z gradually decreases.

[0122] Step S3: Construct the same material parameter vector M = [T ox ,W,L] and the failure physical model coefficient P=[A,B,E a ,C] function mapping model; the specific steps are as follows:

[0123] Based on different stress combinations, failure physical models are established, and the failure physical model coefficients under different stress combinations are obtained. Research on transfer learning of failure physical models based on KNN is then conducted. The source domain data is converted into multivariate feature sample data using Polynomial Features processing. After data normalization, high-dimensional spatial data samples are obtained. The high-dimensional Euclidean distance between the key parameters of the MOSFETs of the test set specifications and the key parameters of the training specifications is calculated and sorted in ascending order of distance. K points with the smallest distance to the key parameters of the MOSFETs of the specifications to be predicted are then selected. Finally, the average value of the values ​​of these K points is calculated as the mapping model output.

[0124] Step S4: Based on the transferable domain model in step S2, a physical migration model of MOS field effect transistor failure in the target domain is constructed using the instance transfer learning method:

[0125] Select the candidate source domain dataset closest to the target domain of the key parameter and construct a new dataset Taking the key manufacturing parameters of the target field as input and the model coefficients as output, the TradaBoost method is used to establish the physical migration model of electronic component failure of new materials. Figure 2 The target domain failure physical model coefficient A is based on the result of the TradaBoost parameter identification method. Figure 3 The target domain failure physical model coefficient B is based on the result of the TradaBoost parameter identification method. Figure 4 The target domain failure physical model coefficient Ea is based on the result of the TradaBoost parameter identification method. Figure 5 The coefficient C of the target domain failure physical model is based on the result of the TradaBoost parameter identification method. Figures 2-5 It shows that the PoF model coefficients based on TradaBoost parameter identification are not much different from the actual ones, which is consistent with the actual change trend.

[0126] In order to further demonstrate the effectiveness of this method, the model coefficients calculated based on TradaBoost are brought into the failure physical model to obtain the average degradation trajectory of the verification specification, as shown in Figures 6-9 As shown, Figure 6 Indicates that the validation specification 1 is based on the comparison between the TradaBoost fitting results and the actual degradation data. Figure 7 Indicates that the validation specification 2 is based on the comparison of TradaBoost fitted degradation data and actual degradation data. Figure 8 Indicates that the validation specification 3 is based on the comparison between the TradaBoost fitting results and the actual degradation data. Figure 9 This figure shows a comparison between the TradaBoost-fitted results and actual degradation data for validation specification 4. The degradation curves of the fitted and actual data were analyzed. Table 5 shows the error between the actual average lifetime of MOSFET stress and the failure physical migration method based on TradaBoost when the degradation time reaches 500 hours. This comparison reveals that the maximum average lifetime error of the proposed method is only 0.0281, demonstrating its effectiveness.

[0127] Table 4 Comparison of fitting RMSE of the two methods

[0128] Specific implementation method two:

[0130] This embodiment is a computer storage medium, in which at least one instruction is stored. The at least one instruction is loaded and executed by a processor to implement the failure physical model transfer learning method of a MOS field effect transistor.

[0131] It should be understood that the instructions include computer program products, software, or computerized methods corresponding to any method described in the present invention; the instructions can be used to program a computer system or other electronic device. Computer storage media may include readable media on which instructions are stored, and may include but are not limited to magnetic storage media, optical storage media; magneto-optical storage media include read-only memory ROM, random access memory RAM, erasable programmable memory (e.g., EPROM and EEPROM) and flash memory layers, or other types of media suitable for storing electronic instructions. Specific implementation method three:

[0133] This embodiment is a device for transferring learning of failure physical models of MOS field-effect transistors. The device includes a processor and a memory. It should be understood that the device includes any device including a processor and a memory described in the present invention. The device may also include other units or modules that perform display, interaction, processing, control, and other functions through signals or instructions.

[0134] The memory stores at least one instruction, which is loaded by the processor and executed to implement the failure physical model migration learning method of the MOS field effect tube.

[0135] Those skilled in the art should understand that the stored at least one instruction is a computer program product corresponding to the method or system. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes. The solutions in the embodiments of the present application can be implemented in various computer languages, such as object-oriented programming languages Java and interpreted scripting language JavaScript.

[0136] The present application is described with reference to flowcharts and / or block diagrams of the methods, systems and computer program products according to the embodiments of the present application, and can also be used for corresponding devices. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in the flowcharts and / or block diagrams. Figure One one flow or multiple flows and / or blocks Figure One an apparatus that performs the functions specified in one or more blocks or multiple blocks.

[0137] These computer program instructions can also be stored in a computer readable memory capable of guiding the computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer readable memory produce a product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure One one flow or multiple flows and / or blocks Figure One an apparatus that performs the functions specified in one or more blocks or multiple blocks.

[0138] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer implemented process, so that the instructions executed on the computer or other programmable device provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure One one flow or multiple flows and / or blocks Figure One an apparatus that performs the functions specified in one or more blocks or multiple blocks.

[0139] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments by those skilled in the art once they learn of the basic inventive concepts. Such additional variations and modifications, including, but not limited to, those pertaining to the preferred embodiments, to the particular designs of the application, to the described compositions and methods, to the particular materials used, and to the claims, can be practiced within the scope of the application.

[0140] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

[0141] The above examples of the application only explain the calculation model and calculation process of the application, and are not intended to limit the embodiments of the application. Based on the above description, other different forms of changes or modifications can be made by those skilled in the art, and it is impossible to enumerate all the embodiments here. Any obvious changes or modifications derived from the technical solutions of the application are still within the protection scope of the application.

Claims

1. A method for transferring learning of failure physical models of MOS field effect transistors, characterized in that: include: Step S1: Combine historical data and simulation analysis methods to construct complex correlation models for different candidate source fields, that is, to reflect the set of coefficients of the MOS field effect tube failure physical model. With key parameter set The complex correlation model between them; the key parameters include material, structure, and process parameters; the complex correlation coefficients are sorted according to the calculation results, and the first multiple parameters are selected as the key parameters affecting the failure physical model coefficients and recorded as The process of building complex correlation models for different candidate source domains includes the following steps: Step S11: Analyze the MOS field effect tube manufacturing process to determine Key parameter set Select [T ox ,W,L,N ch ,N sub ], where T ox is the gate oxide thickness, W is the channel width, L is the channel length, N ch is the channel doping concentration, N sub is the substrate doping concentration; Step S12: Key parameters and corresponding failure physical model coefficients obtained in step S11 Construct a complex correlation model between key parameters and failure physical model coefficients, as shown in formulas (1) and (2): Where, is the set of complex correlation model coefficients, P represents the actual value of the model coefficient, Corresponding to the source domain P The coefficients in , the actual values ​​of the model coefficients; Represents the average value of the model coefficient; R represents the multiple correlation coefficient, which ranges from [0,1]. The larger the multiple correlation coefficient, the closer the correlation between the key parameters and the failure physical model coefficient. Step S13: Sort the complex correlation coefficients according to the calculation results and select the top s (k)n The parameters are key parameters that affect the failure physical model coefficients. The key parameters selected according to the calculation results are recorded as Key parameters For [T ox ,W,L]; Step S2: Based on the complex correlation model of step S1, the distances between different candidate source domains and target domains are calculated to construct a criterion for determining the transferability of MOS field effect transistors: In the formula, Z is the similarity ordinal number, and rank represents the ranking function; represents the mean distance between the kth candidate source domain and the target domain, represents the distance variance between the kth candidate source domain and the target domain; Step S3: For the candidate source field, the key parameters in the candidate source field are As input, the failure physics model coefficients affected by it As output, a physical parameter identification model of electronic component failure in the candidate source field is established; Step S4: Based on the transferable domain discrimination criteria in step S2, the target domain is constructed using the instance transfer learning method. and the failure physical model coefficients affected by it Parameter identification model.

2. The method for transferring the failure physical model of a MOS field effect transistor according to claim 1, wherein: The specific process of constructing the criterion for determining the mobility of the MOS field effect transistor in step S2 includes the following steps: Step S21: Assume that the target domain is Datasets in Represents the target domain instance space annotation; each candidate source domain has an annotated dataset, and the kth candidate source domain annotated dataset is represented as represents the spatial annotation of the source domain instance, i.e., the failure physical model coefficient; then the MMD distance between the key parameters of the candidate source domain and the target domain is as shown in formula (3): Where, Characterizes the maximum mean difference between the candidate source domain and the target domain, X T Indicates D T All instance features in Indicates D s(k) All instance features in ; Indicates D s(k) The jth and j'th instance features in the candidate source domain are the key parameters that affect the failure physical model coefficients; i 、x i' Indicates D T The i-th and i'th instance features in are the key parameters that affect the coefficients of the failure physical model in the target domain; is the number of instances in the candidate source domain dataset; n T is the number of instances in the target domain dataset; k(·,·) represents the kernel function in the RKHS space; Step S22: Calculate the variances between different candidate source domains and target domains to obtain the variances between the same candidate source domain and target domain, as shown in formula (5): Where k(·,·) represents the kernel function in the RKHS space; Step S23: Use the ordinal function to sort the distance means and variances between different candidate source domains and target domains to form the transferability judgment criterion.

3. The method for transferring the failure physical model of a MOS field effect transistor according to claim 2, wherein: The process of establishing the electronic component failure physical parameter identification model in the candidate source field in step S3 includes the following steps: Step S31: Use a neural network to construct a system that reflects the key parameters and the failure physics model coefficients Parameter identification model of Step S32: Using an optimization algorithm, construct a parameter identification model for adaptively searching for the best combination of parameters unit and epoch.

4. The method for transferring the failure physical model of a MOS field effect transistor according to claim 2, wherein: The specific process of step S4 includes the following steps: Step S41: Select the target area D T The closest candidate source domain dataset Building a new dataset Indicates dataset fusion; Step S42: Initialize the sample weights ω of the source domain and the target domain Ti =ω Si , a target domain failure physical parameter identification model is constructed based on the TrAdaBoost algorithm; wherein, the instance features of the new dataset are key parameters, the instance space annotations are failure physical model coefficients, and the established mapping model is the target domain failure physical parameter identification model; The failure physical parameter identification model is established as shown in formula (6): p D =F(x D ) (6) Where F represents the parameter identification model based on AdaBoost, P D represents the instance space annotation of the reconstructed dataset D, x D Represents the instance features of the reconstructed dataset; Step S43: Calculate the error between the predicted model coefficient and the actual model coefficient. The error function is defined as: Among them, e iD is sample i D The relative error, is the ground truth value in the annotation space, Representation error normalization; Step S44: Update the source domain dataset D according to the model error of step S43. s(k) Weight, until the error of formula (6) meets the requirements, the weight update formula is as follows; oh T(N+1) =ω TN ·b (7) oh S(N+1) =ω SN ·(1-b) (8) Where ω TN 、ω T(N+1) represents the current target domain weight and the adjusted target domain weight; β represents the adjustment coefficient; ω SN 、ω S(N+1) Represents the current source domain weight and the adjusted source domain weight, and N represents the number of iterations.

5. A computer storage medium, characterized in that The storage medium stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the failure physical model transfer learning method of a MOS field effect transistor according to any one of claims 1 to 4.

6. A MOS field effect transistor failure physical model transfer learning device, characterized in that: The device includes a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the failure physical model transfer learning method of a MOS field effect transistor as described in any one of claims 1 to 4.

Images