A multi-factor aging test platform and test method

Abstract

The present application relates to the technical field of aging of insulation parts of switch cabinets, and discloses a multi-factor aging test platform and a test method, wherein the multi-factor aging test platform includes: a template construction unit, used to obtain multiple potential factors affecting insulation aging; a data set acquisition unit, used to obtain an aging data set under an aging scenario template for any set of aging scenario templates; a weight determination unit, used to parse each aging sample under the aging data set; an instantiation unit, used to instantiate the aging scenario template into one or more aging instances according to a combination of self-weights and cross-weights; a test unit, used to perform an aging test on a target product based on the target aging instance to generate an aging test result for the target product, thereby solving the technical problem of inaccurate aging test conditions in related technologies and improving the accuracy and richness of aging tests.

Description

Technical Field

[0001] The present application relates to the technical field of aging of switch cabinet insulation components, and in particular to a multi-factor aging test platform and test method. Background Art

[0002] The switchgear is one of the key main equipment of the power system, and its operating status has a significant impact on the reliability of the power system. The comprehensive performance of the switchgear and the insulation inside the cabinet is a key factor in determining the service life of the switchgear, safe and stable operation, and the probability of accidents. During long-term operation, the switchgear and the insulation inside the cabinet usually age due to various stresses such as electricity, heat, and mechanical stresses, resulting in irreversible defects or structural damage, which in turn cause power supply failures and shortened service life. At this stage, research on the factors affecting insulation aging mainly focuses on thermal aging and electrical aging. However, in the industry's aging standards for switchgear and insulation inside the cabinet, the test conditions for secondary aging are quite different from the actual operating conditions, and cannot effectively simulate the actual conditions of the product. Summary of the invention

[0003] The present application provides a multi-factor aging test platform and test method, which solves the technical problem of inaccurate aging test conditions in related technologies and achieves the technical effect of improving the accuracy and richness of aging tests.

[0004] In order to achieve the above objectives, the main technical solutions adopted in this application include:

[0005] In a first aspect, an embodiment of the present application provides a multi-factor aging test platform, which includes: a template construction unit, which is used to obtain multiple potential factors affecting insulation aging, and for any target potential factor, construct one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors; a data set acquisition unit, which is used to obtain an aging data set under the aging scenario template for any set of aging scenario templates, wherein the aging data set includes multiple aging samples, and the aging samples are used to characterize the aging parameters of the product to be tested at different time periods; a weight determination unit, which is used to parse each aging sample under the aging data set to determine the self-weight of the target potential factor in the aging scenario template and the cross-weight of each other potential factor; an instantiation unit, which is used to instantiate the aging scenario template into one or more aging instances according to a combination of the self-weight and the cross-weight, and load the aging instances instantiated from each aging scenario template as a preset aging configuration; a test unit, which is used to obtain a target product, and determine a matching target aging instance from the preset aging configuration according to the actual application scenario of the target product, and perform an aging test on the target product based on the target aging instance to generate an aging test result of the target product.

[0006] The multi-factor aging test platform provided by the embodiment of the present application includes a template construction unit, which is used to obtain multiple potential factors affecting insulation aging, and for any target potential factor, construct one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors; a data set acquisition unit, which is used to obtain an aging data set under an aging scenario template for any set of aging scenario templates, wherein the aging data set includes multiple aging samples, and the aging samples are used to characterize the aging parameters of the product to be tested at different time periods; a weight determination unit, which is used to parse each aging sample under the aging data set to determine the self-weight of the target potential factor in the aging scenario template and the cross-weight of each other potential factor; an instantiation unit, which is used to instantiate the aging scenario template into one or more aging instances according to the combination of the self-weight and the cross-weight, and load the aging instances instantiated from each aging scenario template into a preset aging configuration; a test unit, which is used to obtain a target product, and determine a matching target aging instance from the preset aging configuration according to the actual application scenario of the target product, and perform an aging test on the target product based on the target aging instance to generate an aging test result of the target product, thereby solving the technical problem of inaccurate aging test conditions in the related art and improving the accuracy and richness of the aging test.

[0007] Optionally, if the aging scenario template includes only one other potential factor, the weight determination unit is specifically used to identify the target influence ratio of the target potential factor in each aging sample, cluster the target influence ratio, and generate one or more main influence ratios of the target potential factor according to the clustering result, and screen out the largest main influence ratio from the one or more main influence ratios; generate the secondary influence ratio of the other potential factor according to the largest main influence ratio; use the largest main influence ratio as the self-weight of the target potential factor, and use the secondary influence ratio as the cross-weight of the other potential factors.

[0008] Optionally, if the current aging scene template includes a first other potential factor and a second other potential factor, the weight determination unit is specifically used to obtain a first aging scene template including the target potential factor and the first other potential factor, and obtain a second aging scene template including the target potential factor and the second other potential factor; select a first aging sample of the first aging scene template, a second aging sample of the second aging scene template, and a third aging sample of the current aging scene template; respectively extract a first aging feature of the first aging sample, a second aging feature of the second aging sample, and a third aging feature of the third aging sample; use the first aging feature as an index feature, the second aging feature as a query feature, and the third aging feature as a value feature to calculate a self-attention value of the third aging feature; use the self-attention value of the third aging feature as the self-weight of the target potential factor in the current aging scene template, and set a first cross-weight of the first other potential factor and a second cross-weight of the second other potential factor in the current aging scene template according to the self-weight of the target potential factor.

[0009] Optionally, the weight determination unit is specifically used to generate a self-attention feature of the third aging feature based on the index feature, the query feature and the value feature; input the self-attention feature into a pre-trained weight conversion network to convert the self-attention feature into a corresponding self-attention value through the weight conversion network.

[0010] Optionally, the weight determination unit is specifically used to divide the index feature, the query feature and the value feature into blocks to obtain respective sub-features; for any index sub-feature of the index feature, calculate the similarity between the index sub-feature and each query sub-feature; multiply the calculated multiple similarities with each value sub-feature, and add the multiplication results to obtain a self-attention sub-feature; and concatenate the self-attention sub-features corresponding to each index sub-feature to obtain the self-attention feature of the third aging feature.

[0011] Optionally, the weight determination unit is specifically used to identify the first initial weight of the first other potential factor and the second initial weight of the second other potential factor in the third aging parameter represented by the third aging sample; calculate the sum of the revised weights of the first other potential factor and the second other potential factor according to the self-weight of the target potential factor, and redistribute the sum of the revised weights according to the first initial weight and the second initial weight to obtain the first cross-weight of the first other potential factor and the second cross-weight of the second other potential factor in the current aging scene template.

[0012] In a second aspect, an embodiment of the present application provides a multi-factor aging test method, the method comprising: obtaining multiple potential factors affecting insulation aging, and for any target potential factor, constructing one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors; for any set of aging scenario templates, obtaining an aging data set under the aging scenario template, wherein the aging data set includes multiple aging samples, and the aging samples are used to characterize aging parameters of the product to be tested at different time periods; parsing each aging sample under the aging data set to determine the self-weight of the target potential factor in the aging scenario template and the cross-weight of each of the other potential factors; instantiating the aging scenario template into one or more aging instances according to a combination of the self-weight and the cross-weight, and loading each aging instance instantiated from the aging scenario template as a preset aging configuration; obtaining a target product, and determining a matching target aging instance from the preset aging configuration according to an actual application scenario of the target product, and performing an aging test on the target product based on the target aging instance to generate an aging test result for the target product.

[0013] Optionally, if the aging scenario template includes only one other potential factor, determining the self-weight of the target potential factor in the aging scenario template and the cross-weights of each of the other potential factors includes: identifying the target influence ratio of the target potential factor in each aging sample, clustering the target influence ratio, and generating one or more main influence ratios of the target potential factor based on the clustering results, and screening out the largest main influence ratio from the one or more main influence ratios; generating the secondary influence ratios of the other potential factors based on the largest main influence ratio; using the largest main influence ratio as the self-weight of the target potential factor, and using the secondary influence ratio as the cross-weight of the other potential factors.

[0014] Optionally, if the current aging scene template includes a first other potential factor and a second other potential factor, determining the self-weight of the target potential factor and the cross-weight of each of the other potential factors in the aging scene template includes: obtaining a first aging scene template including the target potential factor and the first other potential factor, and obtaining a second aging scene template including the target potential factor and the second other potential factor; selecting a first aging sample of the first aging scene template, a second aging sample of the second aging scene template, and a third aging sample of the current aging scene template; extracting a first aging feature of the first aging sample, a second aging feature of the second aging sample, and a third aging feature of the third aging sample respectively; using the first aging feature as an index feature, the second aging feature as a query feature, and the third aging feature as a value feature to calculate a self-attention value of the third aging feature; using the self-attention value of the third aging feature as the self-weight of the target potential factor in the current aging scene template, and setting a first cross-weight of the first other potential factor and a second cross-weight of the second other potential factor in the current aging scene template according to the self-weight of the target potential factor.

[0015] Optionally, calculating the self-attention value of the third aging feature includes: generating a self-attention feature of the third aging feature based on the index feature, the query feature and the value feature; inputting the self-attention feature into a pre-trained weight conversion network to convert the self-attention feature into a corresponding self-attention value through the weight conversion network.

[0016] In a third aspect, an embodiment of the present application provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the above-mentioned multi-factor aging test method by executing the computer instructions.

[0017] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the above-mentioned multi-factor aging test method.

[0018] In a fifth aspect, an embodiment of the present application provides a computer program product, including computer instructions, which are used to enable a computer to execute the above-mentioned multi-factor aging test method. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A schematic diagram of a multi-factor aging test platform provided in an embodiment of the present application;

[0021] Figure 2 A flowchart of a multi-factor aging test method provided in an embodiment of the present application;

[0022] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0024] The switchgear is one of the key main equipment of the power system, and its operating status has a significant impact on the reliability of the power system. The comprehensive performance of the switchgear and the insulation inside the cabinet is a key factor in determining the service life of the switchgear, safe and stable operation, and the probability of accidents. During long-term operation, the switchgear and the insulation inside the cabinet usually age due to various stresses such as electricity, heat, and mechanical stresses, resulting in irreversible defects or structural damage, which in turn cause power supply failures and shortened service life. At this stage, research on the factors affecting insulation aging mainly focuses on thermal aging and electrical aging. However, in the industry's aging standards for switchgear and insulation inside the cabinet, the test conditions for secondary aging are quite different from the actual operating conditions, and cannot effectively simulate the actual conditions of the product.

[0025] The embodiment of the present application provides a multi-factor aging test platform, which constructs different aging scenario templates through multiple potential factors affecting aging, sets self-weights for target potential factors in the aging scenario templates, and sets cross-weights for each other potential factor, so as to accurately characterize different aging scenarios. Subsequently, according to actual scenario requirements, suitable aging instances can be selected for aging tests, thereby improving the accuracy and richness of aging tests.

[0026] See also Figure 1 , Figure 1 Schematic diagram of the multi-factor aging test platform provided in the embodiment of the present application. Figure 1 As shown, an embodiment of the present application provides a multi-factor aging test platform, including: a template construction unit, used to obtain multiple potential factors affecting insulation aging, and for any target potential factor, construct one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors; a data set acquisition unit, used to obtain an aging data set under an aging scenario template for any set of aging scenario templates, wherein the aging data set includes multiple aging samples, and the aging samples are used to characterize the aging parameters of the product to be tested at different time periods; a weight determination unit, used to parse each aging sample under the aging data set to determine the self-weight of the target potential factor in the aging scenario template and the cross-weight of each other potential factor; an instantiation unit, used to instantiate the aging scenario template into one or more aging instances according to a combination of the self-weight and the cross-weight, and load the aging instances instantiated from each aging scenario template into a preset aging configuration; a test unit, used to obtain a target product, and determine a matching target aging instance from the preset aging configuration according to the actual application scenario of the target product, and perform an aging test on the target product based on the target aging instance to generate an aging test result of the target product.

[0027] Among them, the latent factors include target latent factors and other latent factors. The latent factors can be electrical aging factors, thermal aging factors, wet aging factors, etc. The aging scenario template includes a target latent factor and one or more other latent factors. The target latent factor can be understood as the main factor affecting aging, and other latent factors can be understood as the secondary factors affecting aging. For example, the latent factors include three factors A, B, and C. Assuming that A is the target latent factor, B and C are other latent factors, four aging scenario templates A, AB, AC, and ABC can be obtained. In these four aging scenario templates, A is the most important influencing factor. The latent factors in the aging scenario template will involve weight configuration, and the same latent factor may have different weights in different aging scenario templates.

[0028] Specifically, for the same aging scenario template, the latent factors may have different weight effects under different aging samples. The aging effects of different latent factors on the product to be tested can be expressed by aging parameters. In fact, this aging parameter can be understood as the actual value of the self-weight of the target latent factor and the cross-weight of other latent factors. For example, the aging scenario template is AB, that is, the current aging result is the aging test result under the combined influence of A and B. Due to the difference in aging samples, A and B have different effects on the product. For the aging sample A (0.7) B (0.3), A is the target latent factor, A has a greater influence, and the weight is set to 0.7. B is the other latent factor, B has a smaller influence, and the weight is set to 0.3. For the aging sample B (0.7) A (0.3), B is the target latent factor, B has a greater influence, and the weight is set to 0.7. A is the other latent factor, and the weight is set to 0.3. The different effects of different latent factors will directly lead to different aging samples, that is, the aging parameters of the product to be tested are different in different time periods.

[0029] Assume that the aging scene template AB has four aging samples, which are recorded as A(0.7)B(0.3), A(0.65)B(0.35), A(0.4)B(0.6), and A(0.35)B(0.65). The corresponding four aging parameters can be expressed as (0.7, 0.3) (0.65, 0.35) (0.4, 0.6) (0.35, 0.65). Then, these four aging parameters are clustered. Obviously, (0.7, 0.3) and (0.65, 0.35) are in one class, and (0.4, 0.6) and (0.35, 0.65) are in another class. According to the clustering results of the aging parameters, the self-weight of A can be 0.68 (the average of 0.7 and 0.65) or 0.38 (the average of 0.4 and 0.35), and the corresponding cross-weight of B can be 0.32 or 0.62. Different aging samples under the same aging scene template may have different self-weight-cross-weight correspondences. In the above case, there are two sets of self-weight-cross-weight correspondences, namely (0.68, 0.32) and (0.38, 0.62). For aging scene template AB, if A is the target latent factor, the combination with the largest self-weight of A, that is, A (0.68) B (0.32), can be selected as the weight value of the aging parameter representation.

[0030] According to the weight value represented by the above aging parameters, i.e. A(0.68)B(0.32), instantiation is performed to generate one or more aging instances of the aging scene template. For each aging scene template, the above instantiation process is performed to obtain corresponding rich aging instances, wherein the target latent factors and other latent factors in each aging instance have their own weights. For example, according to the above A(0.68)B(0.32), instantiation can obtain aging instances such as A(0.71)B(0.29), A(0.7)B(0.3), A(0.68)B(0.37), and A(0.65)B(0.35).

[0031] Through the above instantiation process, the aging scenario template is enriched. In subsequent actual application scenarios, instance matching can be performed according to the actual scenario requirements. The actual application scenario will limit the potential factors that actually affect aging, and limit the degree of influence (weight) of each potential factor on the aging result, so that the corresponding aging instance can be matched. After the aging instance is determined, the aging simulation environment can be configured according to the weight, that is, the matching aging instance is loaded as the preset aging configuration, so as to perform aging tests on the product and obtain the aging test results.

[0032] The multi-factor aging test platform provided by the embodiment of the present application includes a template construction unit, which is used to obtain multiple potential factors affecting insulation aging, and for any target potential factor, construct one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors; a data set acquisition unit, which is used to obtain an aging data set under an aging scenario template for any set of aging scenario templates, wherein the aging data set includes multiple aging samples, and the aging samples are used to characterize the aging parameters of the product to be tested at different time periods; a weight determination unit, which is used to parse each aging sample under the aging data set to determine the self-weight of the target potential factor in the aging scenario template and the cross-weight of each other potential factor; an instantiation unit, which is used to instantiate the aging scenario template into one or more aging instances according to the combination of the self-weight and the cross-weight, and load the aging instances instantiated from each aging scenario template into a preset aging configuration; a test unit, which is used to obtain a target product, and determine a matching target aging instance from the preset aging configuration according to the actual application scenario of the target product, and perform an aging test on the target product based on the target aging instance to generate an aging test result of the target product, thereby solving the technical problem of inaccurate aging test conditions in the related art and improving the accuracy and richness of the aging test.

[0033] In some embodiments, if the aging scenario template includes only one other potential factor, the weight determination unit is specifically used to identify the target influence ratio of the target potential factor in each aging sample, cluster the target influence ratio, and generate one or more main influence ratios of the target potential factor according to the clustering results, and screen out the largest main influence ratio from the one or more main influence ratios; generate the secondary influence ratios of other potential factors according to the largest main influence ratio; use the largest main influence ratio as the self-weight of the target potential factor, and use the secondary influence ratio as the cross-weight of other potential factors.

[0034] The sum of the main influence ratio and the secondary influence ratio is 1. For example, for the aging scene template AB, A is the target latent factor and B is the other latent factor. The aging samples of the aging scene template AB are A (0.7) B (0.3), A (0.65) B (0.35), A (0.4) B (0.6), and A (0.35) B (0.65). The target influence ratios of the target latent factors include 0.7, 0.65, 0.4, and 0.35. Clustering the target influence ratios, it is obvious that 0.7 and 0.65 are in one category, and 0.4 and 0.35 are in another category. According to the clustering results, the main influence ratios of A include 0.68 (the mean of 0.7 and 0.65) and 0.38 (the mean of 0.4 and 0.35). Based on the main influence ratio of A, the secondary influence ratio of B is generated, and the corresponding secondary influence ratios, i.e., 0.32 and 0.62, are obtained by subtracting the main influence ratio from 1. The largest main effect ratio was used as the self-weight of the target latent factor, and the secondary effect ratio was used as the cross-weight of other latent factors, that is, 0.68 was used as the self-weight of A and 0.32 was used as the cross-weight of B.

[0035] In some embodiments, if the current aging scene template includes a first other potential factor and a second other potential factor, the weight determination unit is specifically used to obtain a first aging scene template including a target potential factor and the first other potential factor, and obtain a second aging scene template including the target potential factor and the second other potential factor; select a first aging sample of the first aging scene template, a second aging sample of the second aging scene template, and a third aging sample of the current aging scene template; extract a first aging feature of the first aging sample, a second aging feature of the second aging sample, and a third aging feature of the third aging sample respectively; use the first aging feature as an index feature, the second aging feature as a query feature, and the third aging feature as a value feature to calculate a self-attention value of the third aging feature; use the self-attention value of the third aging feature as the self-weight of the target potential factor in the current aging scene template, and set a first cross-weight of the first other potential factor and a second cross-weight of the second other potential factor in the current aging scene template according to the self-weight of the target potential factor.

[0036] If there is only one other potential factor, then the aging parameter corresponding to the aging sample can accurately represent the weight value. However, if there are more other potential factors, the weight value represented by the aging parameter may not be accurate, so it needs to be corrected to get the accurate weight value. The self-weight of A is corrected by calculating its self-attention value, and then the cross-weights of other potential factors are calculated based on the corrected self-weight.

[0037] For example, for the current aging scene template ABC, A is the target latent factor, and B and C are other latent factors. The first aging scene template includes the target latent factor and the first other latent factor, set to AB; the second aging scene template includes the target latent factor and the second other latent factor, set to BC. Select the first aging sample of the first aging scene template, and extract the first aging feature of the first aging sample as the index feature. Select the second aging sample of the second aging scene template, and extract the second aging feature of the second aging sample as the query feature. Select the third aging sample of the current aging scene template, and extract the third aging feature of the third aging sample as the value feature. Among them, the selected first aging sample, second aging sample, and third aging sample need to meet the following conditions: among the potential factors causing sample aging, the target latent factor is the main factor, that is, A needs to be the target latent factor. For example, the first aging scene template AB has four aging samples, which are recorded as A(0.7)B(0.3), A(0.65)B(0.35), A(0.4)B(0.6), and A(0.35)B(0.65). When selecting the first aging sample, it is necessary to select the sample with A as the main influencing factor, that is, one of A(0.7)B(0.3) or A(0.65)B(0.35) can be selected. For samples A(0.4)B(0.6) and A(0.35)B(0.65), their main influencing factor is B, so these two samples cannot be used as the first aging samples.

[0038] The self-attention value is calculated according to the self-attention mechanism, and the self-attention value is used as the self-weight of the target potential factor in the current aging scene template. Among them, the index feature is k, the query feature is q, and the value feature is v. With these three features, the self-attention feature can be calculated, and then the self-attention feature can be converted into a self-attention value. According to the weight value of each potential factor in the aging sample ABC, the weight ratio relationship between B and C can be obtained. According to 1 minus the self-attention value of A, the sum of the weights of B and C is obtained, and then according to the weight ratio relationship between B and C, the first cross weight of B and the second cross weight of C can be obtained.

[0039] In the aging scene template, if there are multiple other potential factors at the same time, these other potential factors will affect the target potential factor. This complex cross-effect cannot be accurately represented by the aging parameters of the aging sample. The embodiment of the present application calculates the self-attention value of the target potential factor in the aging sample to obtain the actual impact of the target potential factor on aging in the aging sample, thereby improving the accuracy of the aging instance.

[0040] Specifically, the weight determination unit is used to generate a self-attention feature of the third aging feature according to the index feature, the query feature and the value feature; the self-attention feature is input into a pre-trained weight conversion network to convert the self-attention feature into a corresponding self-attention value through the weight conversion network.

[0041] Among them, during the training process of the weight conversion network, the input is the sample of the self-attention feature, and the label is the weight value. The weight value of the label can be a standard value, such as the 9 standard values ​​​​from 0.1 to 0.9. In this way, the training process of the network becomes a classification process, depending on which interval the final result falls into. For example, if it falls into the interval of 0.1 to 0.2, the output weight value is 0.2, which is the self-attention value.

[0042] In some embodiments, the weight determination unit is specifically used to divide the index feature, query feature and value feature into blocks respectively to obtain respective sub-features; for any index sub-feature of the index feature, calculate the similarity between the index sub-feature and each query sub-feature; multiply the calculated multiple similarities with each value sub-feature respectively, and add the multiplication results to obtain a self-attention sub-feature; splice the self-attention sub-features corresponding to each index sub-feature to obtain the self-attention feature of the third aging feature.

[0043] In some embodiments, the weight determination unit is specifically used to identify a first initial weight of a first other potential factor and a second initial weight of a second other potential factor in a third aging parameter represented by a third aging sample; calculate the sum of revised weights of the first other potential factor and the second other potential factor according to the self-weight of the target potential factor, and redistribute the sum of revised weights according to the first initial weight and the second initial weight to obtain a first cross-weight of the first other potential factor and a second cross-weight of the second other potential factor in the current aging scenario template.

[0044] For example, in the third aging sample A (0.7) B (0.2) C (0.1), the ratios of the first other latent factor and the second other latent factor are 0.2 and 0.1 respectively (the initial ratio of the target latent factor is 0.7). Since the weight value of the target latent factor needs to be corrected, for example, corrected to 0.65, then 1-0.65=0.35 is the sum of the corrected weights of the two other latent factors, and then 0.35 is allocated according to the ratio of 0.2:0.1, and the first cross-weight of the first latent factor and the second cross-weight of the second latent factor after correction are obtained.

[0045] The embodiment of the present application also provides an embodiment of a multi-factor aging test method. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0046] See also Figure 2 , Figure 2 This is a flow chart of the multi-factor aging test method provided in the embodiment of the present application. Figure 2 As shown, the process includes the following steps:

[0047] Step S1, obtaining multiple potential factors affecting insulation aging, and for any target potential factor, constructing one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors.

[0048] Among them, the potential factors include target potential factors and other potential factors. The potential factors can be electrical aging factors, thermal aging factors, and wet aging factors. The aging scene template includes a target potential factor and one or more other potential factors. The target potential factor can be understood as the main factor affecting aging, and the other potential factors can be understood as the secondary factors affecting aging.

[0049] Step S3, for any set of aging scene templates, an aging data set under the aging scene template is obtained, where the aging data set includes a plurality of aging samples, and the aging samples are used to characterize aging parameters of the product to be tested at different time periods.

[0050] For example, the main influencing factors of aging of the product to be tested at different time periods may change. By selecting different aging samples, it can be used to characterize the aging parameters at different time periods to meet actual application requirements.

[0051] Step S5, analyzing each aging sample in the aging data set to determine the self-weight of the target latent factor and the cross-weight of each other latent factor in the aging scene template.

[0052] The sum of the main influence ratio and the secondary influence ratio is 1. For the same aging scenario template, the latent factors may have different weight effects under different aging samples. According to the clustering results of the aging parameters, the combination with the largest self-weight of the target latent factors is selected as the weight value of the aging parameter representation.

[0053] Step S7: instantiate the aging scene template into one or more aging instances according to the combination of the self-weight and the cross-weight, and load each aging instance instantiated from the aging scene template into a preset aging configuration.

[0054] Step S9, obtaining a target product, and determining a matching target aging instance from a preset aging configuration according to an actual application scenario of the target product, and performing an aging test on the target product based on the target aging instance to generate an aging test result of the target product.

[0055] Through the above instantiation process, the aging scenario template is enriched. In subsequent actual application scenarios, instance matching can be performed according to the actual scenario requirements. The actual application scenario will limit the potential factors that actually affect aging, and limit the degree of influence (weight) of each potential factor on the aging result, so that the corresponding aging instance can be matched. Once the aging instance is determined, the aging simulation environment can be configured according to the weight, that is, the matching aging instance is loaded as the preset aging configuration, so as to perform aging tests on the target product and obtain the aging test results.

[0056] The multi-factor aging test method provided in this embodiment includes: obtaining multiple potential factors affecting insulation aging, and for any target potential factor, constructing one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors; for any set of aging scenario templates, obtaining an aging data set under the aging scenario template, wherein the aging data set includes multiple aging samples, and the aging samples are used to characterize the aging parameters of the product to be tested at different time periods; parsing each aging sample under the aging data set to determine the self-weight of the target potential factor in the aging scenario template and the cross-weight of each other potential factor; instantiating the aging scenario template into one or more aging instances according to the combination of the self-weight and the cross-weight, and loading the aging instances instantiated from each aging scenario template as a preset aging configuration; obtaining a target product, and according to the actual application scenario of the target product, determining a matching target aging instance from the preset aging configuration, and performing an aging test on the target product based on the target aging instance to generate an aging test result of the target product, thereby solving the technical problem of inaccurate aging test conditions in the related art and improving the accuracy and richness of the aging test.

[0057] In some embodiments, if the aging scenario template includes only one other potential factor, determining the self-weight of the target potential factor in the aging scenario template and the cross-weights of each other potential factor includes: identifying the target influence ratio of the target potential factor in each aging sample, clustering the target influence ratio, and generating one or more main influence ratios of the target potential factor based on the clustering results, and screening out the largest main influence ratio from the one or more main influence ratios; generating secondary influence ratios of other potential factors based on the largest main influence ratio; using the largest main influence ratio as the self-weight of the target potential factor, and using the secondary influence ratio as the cross-weight of other potential factors.

[0058] In some embodiments, if the current aging scene template includes a first other potential factor and a second other potential factor, determining the self-weight of the target potential factor and the cross-weight of each other potential factor in the aging scene template includes: obtaining a first aging scene template including the target potential factor and the first other potential factor, and obtaining a second aging scene template including the target potential factor and the second other potential factor; selecting a first aging sample of the first aging scene template, a second aging sample of the second aging scene template, and a third aging sample of the current aging scene template; extracting a first aging feature of the first aging sample, a second aging feature of the second aging sample, and a third aging feature of the third aging sample respectively; using the first aging feature as an index feature, the second aging feature as a query feature, and the third aging feature as a value feature, and calculating a self-attention value of the third aging feature; using the self-attention value of the third aging feature as the self-weight of the target potential factor in the current aging scene template, and setting a first cross-weight of the first other potential factor and a second cross-weight of the second other potential factor in the current aging scene template according to the self-weight of the target potential factor.

[0059] In some embodiments, calculating the self-attention value of the third aging feature includes: generating a self-attention feature of the third aging feature based on an index feature, a query feature, and a value feature; inputting the self-attention feature into a pre-trained weight conversion network to convert the self-attention feature into a corresponding self-attention value through the weight conversion network.

[0060] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0061] The multi-factor aging test platform in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0062] See also Figure 3 , Figure 3 is a schematic diagram of the structure of a computer device provided in an embodiment of the present application, such as Figure 3As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 3 A processor 10 is taken as an example.

[0063] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0064] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0065] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0066] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory. The computer device also includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0067] The embodiment of the present application also provides a computer-readable storage medium. The above method according to the embodiment of the present application can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0068] The embodiment of the present application provides a computer program product, which includes computer instructions, which are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method of any embodiment of the present application.

[0069] Although the embodiments of the present application are described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

[0070] The test platform described in the above embodiments can be implemented by a computer chip or entity, or by a product with a certain function. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0071] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0072] Those skilled in the art will appreciate that the embodiments of the present application may be provided as a test platform, a test method or a computer program product. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may 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.) that contain computer-usable program code.

[0073] The present application is described with reference to the flowcharts and / or block diagrams of the test methods, test platforms, and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0074] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0075] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0076] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity or device including the element.

[0077] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0078] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

[0079] Although the embodiments of the present application have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A multi-factor aging test platform, characterized in that: The multi-factor aging test platform includes: A template construction unit, used for obtaining multiple potential factors affecting insulation aging, and for any target potential factor, constructing one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors; A data set acquisition unit, used for acquiring an aging data set under any set of aging scene templates, wherein the aging data set includes a plurality of aging samples, and the aging samples are used for characterizing aging parameters of the product to be tested at different time periods; A weight determination unit is used to parse each aging sample under the aging data set to determine the self-weight of the target potential factor in the aging scene template and the cross-weight of each other potential factor; wherein, if the aging scene template includes only one other potential factor, the weight determination unit is specifically used to identify the target influence ratio of the target potential factor in each aging sample, cluster the target influence ratio, and generate one or more main influence ratios of the target potential factor according to the clustering result, and screen out the largest main influence ratio from one or more main influence ratios; generate the secondary influence ratio of other potential factors according to the largest main influence ratio; use the largest main influence ratio as the self-weight of the target potential factor, and use the secondary influence ratio as the cross-weight of other potential factors; if the current aging scene template includes a first other potential factor and a second other potential factor, the weight determination unit is specifically used to obtain a target influence ratio including the target potential factor and the first other potential factor. A first aging scene template of a latent factor, and a second aging scene template including the target latent factor and the second other latent factor are obtained; a first aging sample of the first aging scene template, a second aging sample of the second aging scene template, and a third aging sample of the current aging scene template are selected; a first aging feature of the first aging sample, a second aging feature of the second aging sample, and a third aging feature of the third aging sample are extracted respectively; the first aging feature is used as an index feature, the second aging feature is used as a query feature, and the third aging feature is used as a value feature to calculate a self-attention value of the third aging feature; the self-attention value of the third aging feature is used as the self-weight of the target latent factor in the current aging scene template, and a first cross-weight of the first other latent factor and a second cross-weight of the second other latent factor in the current aging scene template are set according to the self-weight of the target latent factor; An instantiation unit, configured to instantiate the aging scene template into one or more aging instances according to a combination of the self-weight and the cross-weight, and load each aging instance instantiated from the aging scene template as a preset aging configuration; The test unit is used to obtain a target product, and determine a matching target aging instance from the preset aging configuration according to an actual application scenario of the target product, and perform an aging test on the target product based on the target aging instance to generate an aging test result of the target product.

2. The multi-factor aging test platform according to claim 1, characterized in that: The weight determination unit is specifically used to generate a self-attention feature of the third aging feature based on the index feature, the query feature and the value feature; input the self-attention feature into a pre-trained weight conversion network to convert the self-attention feature into a corresponding self-attention value through the weight conversion network.

3. The multi-factor aging test platform according to claim 2, characterized in that: The weight determination unit is specifically used to divide the index feature, the query feature and the value feature into blocks to obtain respective sub-features; for any index sub-feature of the index feature, calculate the similarity between the index sub-feature and each query sub-feature; multiply the calculated multiple similarities with each value sub-feature, and add the multiplication results to obtain a self-attention sub-feature; and splice the self-attention sub-features corresponding to each index sub-feature to obtain the self-attention feature of the third aging feature.

4. The multi-factor aging test platform according to claim 1, characterized in that: The weight determination unit is specifically used to identify the first initial weight of the first other potential factor and the second initial weight of the second other potential factor in the third aging parameter represented by the third aging sample; calculate the sum of the revised weights of the first other potential factor and the second other potential factor according to the self-weight of the target potential factor, and redistribute the sum of the revised weights according to the first initial weight and the second initial weight to obtain the first cross-weight of the first other potential factor and the second cross-weight of the second other potential factor in the current aging scene template.

5. A multi-factor aging test method, applied to the multi-factor aging test platform as described in any one of claims 1 to 4, characterized in that: The method comprises: Acquire multiple potential factors that affect insulation aging, and for any target potential factor, construct one or more sets of aging scenario templates between the target potential factor and other potential factors in the multiple potential factors; For any set of aging scene templates, an aging data set under the aging scene template is obtained, wherein the aging data set includes a plurality of aging samples, and the aging samples are used to characterize aging parameters of the product to be tested at different time periods; Parsing each aging sample under the aging data set to determine the self-weight of the target latent factor and the cross-weight of each of the other latent factors in the aging scene template; According to a combination of the self-weight and the cross-weight, the aging scene template is instantiated into one or more aging instances, and each aging instance instantiated from the aging scene template is loaded as a preset aging configuration; A target product is obtained, and according to an actual application scenario of the target product, a matching target aging instance is determined from the preset aging configuration, and an aging test is performed on the target product based on the target aging instance to generate an aging test result of the target product.

6. The method according to claim 5, characterized in that If the aging scene template includes only one other potential factor, determining the self-weight of the target potential factor in the aging scene template and the cross-weights of each of the other potential factors includes: Identifying the target influence ratio of the target latent factor in each aging sample, clustering the target influence ratio, generating one or more main influence ratios of the target latent factor according to the clustering result, and screening out the largest main influence ratio from the one or more main influence ratios; The secondary influence ratios of the other potential factors are generated according to the maximum primary influence ratio; the maximum primary influence ratio is used as the self-weight of the target potential factor, and the secondary influence ratios are used as the cross-weights of the other potential factors.

7. The method according to claim 5 or 6, characterized in that: If the current aging scene template includes a first other latent factor and a second other latent factor, determining the self-weight of the target latent factor in the aging scene template and the cross-weights of each of the other latent factors includes: Acquire a first aging scenario template including the target latent factor and the first other latent factor, and acquire a second aging scenario template including the target latent factor and the second other latent factor; Selecting a first aging sample of the first aging scene template, a second aging sample of the second aging scene template, and a third aging sample of the current aging scene template; respectively extracting a first aging feature of the first aging sample, a second aging feature of the second aging sample, and a third aging feature of the third aging sample; Taking the first aging feature as an index feature, taking the second aging feature as a query feature, taking the third aging feature as a value feature, and calculating a self-attention value of the third aging feature; The self-attention value of the third aging feature is used as the self-weight of the target latent factor in the current aging scene template, and according to the self-weight of the target latent factor, a first cross-weight of the first other latent factor and a second cross-weight of the second other latent factor in the current aging scene template are set.

8. The method according to claim 7, characterized in that Calculating the self-attention value of the third aging feature includes: generating a self-attention feature of the third aging feature according to the index feature, the query feature, and the value feature; The self-attention feature is input into a pre-trained weight conversion network to convert the self-attention feature into a corresponding self-attention value through the weight conversion network.

Images