Mathematical and intellectual fusion reliability test and evaluation method, device and computer equipment

Abstract

The present application relates to a reliability test and evaluation method, device and computer equipment that integrates mathematical and rational elements. The performance degradation data of multiple detection components of each subtype under different component types in the product to be tested at different test times are obtained. For multiple detection components of each subtype under the same component type, the first cumulative distribution function corresponding to the detection components of each subtype is determined based on the performance degradation data of the multiple detection components. The first failure life of the product to be tested is determined based on each first cumulative distribution function. The reliability evaluation index of the product to be tested is determined based on the first failure life of the multiple products to be tested. The embodiment of the present application divides the detection components in the product to be tested into detection components of different component types, and uses different algorithms to process the performance degradation data of different component types, making full use of the advantages of multiple algorithms, flexibly processing the performance degradation data of different component types, and improving the accuracy of reliability evaluation.

Description

Technical Field

[0001] The present application relates to the field of product testing technology, and in particular to a reliability testing and evaluation method, device and computer equipment that integrates mathematical and rational fusion. Background Art

[0002] Product reliability evaluation is a crucial component of modern product quality management. It involves assessing a product's ability to perform its specified functions under specified conditions and within a specified timeframe through scientific experimental design and data analysis. Its core objective is to quantify a product's reliability level and identify potential failure modes and causes, thereby providing a critical basis for product design, production, and use.

[0003] Current reliability evaluation mainly uses mathematical statistics methods to analyze product reliability test data. However, this method only infers the overall reliability characteristics based on the statistical laws of sample data, resulting in low accuracy of reliability evaluation. Summary of the Invention

[0004] Based on this, it is necessary to provide a reliability test and evaluation method, device and computer equipment that integrates mathematical and rational elements to improve the accuracy of reliability evaluation in order to address the above technical problems.

[0005] In a first aspect, the present application provides a reliability testing and evaluation method for mathematical and rational fusion, comprising:

[0006] Obtaining performance degradation data of multiple test components of different subtypes under different component types in the product to be tested at different test times; the different component types include mathematical statistics type, failure physics type, and intelligent algorithm type;

[0007] For a plurality of detection components of each subtype under the same component type, determining a first cumulative distribution function corresponding to each detection component of the subtype based on performance degradation data of the plurality of detection components; the first cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the detection component of the subtype;

[0008] determining a first failure life of the product to be tested according to each of the first cumulative distribution functions;

[0009] Determine a reliability evaluation index of the product to be tested based on the first failure life of the plurality of products to be tested.

[0010] In one embodiment, the first cumulative distribution function is a first cumulative distribution function corresponding to the intelligent algorithm class; and determining, based on the performance degradation data of the plurality of detection components, the first cumulative distribution function corresponding to the detection components of each subtype includes:

[0011] performing normalization processing on the performance degradation data of the plurality of detection components and the detection time to obtain a feature normalization sequence and a time normalization sequence;

[0012] Inputting the feature-normalized sequence and the time-normalized sequence into a prediction model to obtain a first prediction sequence corresponding to the feature-normalized sequence and a second prediction sequence corresponding to the time-normalized sequence;

[0013] Performing inverse normalization processing on the first prediction sequence and the second prediction sequence to obtain a first prediction inverse normalized sequence and a second prediction inverse normalized sequence;

[0014] A first cumulative distribution function corresponding to the subtype detection component is determined according to the first predicted inverse normalized sequence and the second predicted inverse normalized sequence.

[0015] In one embodiment, determining the reliability evaluation index of the product to be tested based on the first failure life of the plurality of products to be tested includes:

[0016] Determining a second cumulative distribution function corresponding to the product to be detected based on the first failure life of the plurality of products to be detected; the second cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the product to be detected;

[0017] The average life of the product to be tested is determined according to the second cumulative distribution function; and the reliability evaluation index includes the average life of the product to be tested.

[0018] In one embodiment, the method further comprises:

[0019] Determining a reliability function of the product to be tested based on a target parameter corresponding to the second cumulative distribution function; the reliability function is used to characterize the relationship between the failure life and reliability of the product to be tested;

[0020] The reliability of the product to be tested is determined according to the target failure life of the product to be tested and the reliability function; the reliability evaluation index includes the reliability of the product to be tested.

[0021] In one embodiment, obtaining performance degradation data of multiple test components of different subtypes under different component types in the product to be tested at different test times includes:

[0022] Determine the inspection time corresponding to different component types based on the inspection time intervals corresponding to different component types;

[0023] The detection equipment is controlled to detect the detection time corresponding to different component types, and the performance degradation data of multiple detection components of each subtype under the corresponding component type are collected to obtain the performance degradation data at different detection times.

[0024] In one embodiment, determining the first failure life of the product to be tested according to each of the first cumulative distribution functions includes:

[0025] For each subtype of the same component type, determining a second failure life of the detection component of the subtype according to a first cumulative distribution function and a random failure probability corresponding to the detection component of the subtype;

[0026] The first failure life of the product to be inspected is determined according to the second failure life and the connection relationship of the inspection components of each sub-type under each component type.

[0027] In a second aspect, the present application also provides a reliability testing and evaluation device for mathematical and rational fusion, comprising:

[0028] An acquisition module is used to obtain performance degradation data of multiple test components of different subtypes under different component types in the product to be tested at different test times; the different component types include mathematical statistics, failure physics, and intelligent algorithm types;

[0029] a first determining module configured to determine, for a plurality of detection components of each subtype under the same component type, a first cumulative distribution function corresponding to each detection component of the subtype based on performance degradation data of the plurality of detection components; wherein the first cumulative distribution function is configured to represent a relationship between a failure life and a failure probability of the detection component of the subtype;

[0030] A second determining module is configured to determine a first failure life of the product to be tested according to each of the first cumulative distribution functions;

[0031] The third determining module is used to determine the reliability evaluation index of the product to be detected according to the first failure life of the plurality of products to be detected.

[0032] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the method steps provided in the first aspect when executing the computer program.

[0033] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the method steps provided in the first aspect when the computer program is executed by a processor.

[0034] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which implements the method steps provided in the first aspect when executed by a processor.

[0035] The above-mentioned reliability test and evaluation method, device and computer equipment of mathematical and rational fusion obtains the performance degradation data of multiple detection components of each subtype under different component types in the product to be tested at different detection times. For multiple detection components of each subtype under the same component type, according to the performance degradation data of multiple detection components, the first cumulative distribution function corresponding to the detection components of each subtype is determined, and the first failure life of the product to be tested is determined according to each first cumulative distribution function. According to the first failure life of multiple products to be tested, the reliability evaluation index of the product to be tested is determined; different component types include mathematical statistics, failure physics and intelligent algorithm; the first cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the detection component under the subtype. The embodiment of the present application divides the detection components in the product to be tested into mathematical statistics, failure physics and intelligent algorithm types of detection components. Different algorithms are used to process the performance degradation data of different component types. The quantitative analysis ability of mathematical statistics, the mechanism description ability of failure physics models and the complex data processing ability of intelligent algorithms can be combined to fully utilize the advantages of multiple algorithms to flexibly process the performance degradation data of different component types, and obtain the first cumulative distribution function of the sub-type detection components under each component type. Therefore, the first failure life of the product to be detected can be determined based on the first cumulative distribution function of the detection components of each sub-type. The first failure life of multiple products to be detected is used to comprehensively evaluate the reliability of the product to be detected, comprehensively reflect the reliability evaluation index of the product to be detected, and improve the accuracy of the reliability evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 2. It is an application environment diagram of a reliability test and evaluation method for mathematical and rational fusion in one embodiment;

[0038] Figure 2 1. It is a flowchart of a reliability test and evaluation method of mathematical and rational fusion in one embodiment;

[0039] Figure 3 1 is a flow chart of a method for determining a first cumulative distribution function in one embodiment;

[0040] Figure 4 1 is a flow chart of a method for determining a reliability evaluation index in one embodiment;

[0041] Figure 5 Schematic diagram of a flow chart of a method for determining a reliability evaluation index in another embodiment;

[0042] Figure 6 Schematic diagram of a method for collecting performance degradation data in one embodiment;

[0043] Figure 7 1 is a flow chart of a first failure life determination method in one embodiment;

[0044] Figure 8 1 is a structural block diagram of a reliability testing and evaluation device for mathematical and rational fusion in one embodiment. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0046] The reliability test and evaluation method of mathematical and rational fusion provided in the embodiment of the present application can be applied to Figure 1 The application environment shown in FIG. The application environment includes a computer device, which may be a server, and its internal structure diagram may be as shown in FIG. Figure 1 As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, memory and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store relevant data for reliability testing and evaluation of mathematical and rational fusion. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a reliability testing and evaluation method for mathematical and rational fusion is implemented. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0047] Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0048] In an exemplary embodiment, Figure 2 As shown in the figure, a reliability test and evaluation method of mathematical and rational fusion is provided. Figure 1 The computer device in the example is used to illustrate, including the following S201 to S204.

[0049] S201, obtaining performance degradation data of multiple test components of different subtypes under different component types in the product to be tested at different test times; different component types include mathematical statistics type, failure physics type and intelligent algorithm type.

[0050] Optionally, the performance degradation data may be data obtained based on a reliability test. The reliability test is to test the product to be tested by simulating the actual use environment or accelerated test conditions of the product to be tested, and obtain performance degradation data of each test component in the product to be tested.

[0051] In the embodiment of the present application, first, the functional structure of the product to be inspected is divided, and the various inspection components in the entire product to be inspected are divided into mathematical statistics type inspection components, failure physics type inspection components, and intelligent algorithm type inspection components. Among them, the mathematical statistics type inspection components mainly refer to components whose performance parameters have a significant degradation trend and whose failure physics model is unknown. The failure physics type inspection components mainly refer to components whose performance parameters have a significant degradation trend and whose failure physics model is known. The intelligent algorithm type inspection components mainly refer to components whose performance parameters have no significant degradation trend and whose test data volume is large.

[0052] In one possible implementation, a computer device can determine different inspection time intervals based on the test data volume requirements of algorithms corresponding to different component types, thereby determining the inspection times corresponding to different component types based on the inspection time intervals. For each component type, the corresponding inspection time is used to control the inspection equipment to collect degradation data for all subtypes of the inspection components of that component type. The computer device then obtains performance degradation data for all subtypes of the inspection components of different component types at different inspection times from each inspection equipment.

[0053] For example, for the detection components of mathematical statistics, the mathematical statistics category includes Detection components for different seed types , that is, the detection components with different subtypes in A all belong to the mathematical statistics category. For example, the detection component There are m in total, assuming the number of detection times is , the detection time is , the performance degradation data is , collect the detection components of mathematical statistics The performance degradation data of the reliability test is shown in Table 1.

[0054] Table 1

[0055]

[0056] For the detection components of failure physics, the failure physics category includes Detection components for different seed types , that is, existence Detection components with different seed types all belong to the failure physics class. For example, the detection component Including m, assuming the number of detection times is , the detection time is , the performance degradation data is , collect the detection components of the failed physical class The performance degradation data of the reliability test is shown in Table 2.

[0057] Table 2

[0058]

[0059] For intelligent algorithm detection components, the intelligent algorithm category also includes Detection components , that is, existence Detection components with different seed types all belong to the intelligent algorithm category. For example, the detection component Including m, assuming the number of detection times is , the detection time is , the performance degradation data is , collect the detection components of intelligent algorithm The performance degradation data of the reliability test is shown in Table 3.

[0060] Table 3

[0061]

[0062] In another possible implementation, the computer device can set the same time detection interval for different component types, and the total detection duration for different component types is also the same. That is, the detection time for different component types is the same, and the amount of performance degradation data obtained is also the same.

[0063] In another possible implementation, the same time interval can be set for different component types, and the total testing duration can be set differently for each component type based on the test data volume required by the corresponding algorithm. For each component type, the corresponding testing time is used to control the testing equipment to collect performance degradation data for all subtypes of the component type.

[0064] S202 , for multiple detection components of each subtype under the same component type, determine a first cumulative distribution function corresponding to each subtype of the detection components based on performance degradation data of the multiple detection components; the first cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the detection components under the subtype.

[0065] For mathematical statistics, the subtypes are The performance degradation data is fitted using a mathematical and statistical model for the detection components. The mathematical and statistical model for performance degradation is as follows:

[0066] (1)

[0067] In formula (1): Detection components for mathematical statistics The corresponding first cumulative distribution function; is the critical failure value of performance degradation, which is determined by the detection component Determine the usage specifications; is the first unknown parameter, is the second unknown parameter ( and are two unknown parameters of the mathematical statistics model); The failure life.

[0068] Among them, the use of Mathematical statistics detection components The performance degradation data of the first unknown parameter is calculated using formulas (2) and (3). and the second unknown parameter To solve, For the Mathematical statistics detection components The initial performance degradation value of is the number of detection times, then Detection components The first parameter value and the second parameter value They are

[0069] (2)

[0070] (3)

[0071] Based on formulas (4) and (5), we can get The first parameter mean of the first parameter values and The second parameter mean of the second parameter values .

[0072] (4)

[0073] (5)

[0074] Thus, we can solve the detection components of mathematical statistics The first cumulative distribution function The same method can be used to obtain the first cumulative distribution function of other detection components of the mathematical statistics class, thereby obtaining the first cumulative distribution function of all detection components of the mathematical statistics class. .

[0075] For the failure physics category, the subtypes are For the detection components, the failure physical model is used to fit the performance degradation data. The failure physical model of different sub-type detection components is different. For example, the failure physical model of rolling bearing is as follows:

[0076] (6)

[0077] in: is the failure life of the rolling bearing; It is the design life under rated torque, which is determined by the design manual; is the rated speed, determined by the design manual; is the rated torque, determined by the design manual; is the average speed of the test, obtained from the test; is the average load torque of the test, obtained from the reliability test; It is the characteristic parameter of life span and is determined by the design manual.

[0078] The average speed and average load torque (i.e., performance degradation data) of the rolling bearing are obtained through reliability test. Substituting the average speed and average load torque into formula (6), we can get Failure life of rolling bearings Assume that the first cumulative distribution function of the rolling bearing is Weibull distribution, as shown in formula (7):

[0079] (7)

[0080] Where: is the first cumulative distribution function of rolling element bearings; is the third unknown parameter, is the fourth unknown parameter ( and are two unknown parameters of the failure physical model); the third and fourth unknown parameters are solved by the failure life of the rolling bearing and formula (8).

[0081] (8)

[0082] Thus, the first cumulative distribution function of the rolling bearing is obtained , using the same method, the first cumulative distribution functions of all failed physical detection components are obtained as follows: .

[0083] For each subtype of detection components under the intelligent algorithm class, in a possible implementation method, an artificial intelligence algorithm is used to fit the performance degradation data of the detection components of the intelligent algorithm class. In the embodiment of the present application, the performance degradation data and the detection time are normalized by performing data feature standardization, and the feature standardization sequence and time standardization sequence obtained by the data feature standardization are input into the artificial intelligence algorithm to obtain a first prediction sequence and a second prediction sequence. The first prediction sequence and the second prediction sequence are subjected to inverse standardization processing to obtain a first prediction inverse standardization sequence and a second prediction inverse standardization sequence. According to the first prediction inverse standardization sequence and the second prediction inverse standardization sequence, the first cumulative distribution function corresponding to the detection component of each subtype is determined. Optionally, the artificial intelligence algorithm may include support vector machine, naive Bayes, logistic regression, decision tree, K nearest neighbor algorithm, random forest, etc.

[0084] For each subtype of detection components under the intelligent algorithm category, in another possible implementation method, the performance degradation data and detection time can be directly input into the artificial intelligence algorithm to obtain a time feature sequence and a data feature sequence. Based on the time feature sequence and the data feature sequence, the first cumulative distribution function corresponding to each subtype of detection component is determined.

[0085] S203: Determine a first failure life of the product to be tested according to each first cumulative distribution function.

[0086] In an embodiment of the present application, for each subtype under the same component type, the second failure life of each detection component is determined based on the first cumulative distribution function and the random failure probability, and the first failure life of the product to be detected is determined based on the second failure life and connection relationship of the detection components of the subtype under each component type.

[0087] In one possible implementation, for each subtype under the same component type, the second failure life of the detection component of each subtype is determined based on the first cumulative distribution function and the random failure probability, and the smallest second failure life among the detection components of all subtypes is used as the first failure life of the product to be detected.

[0088] S204 , determining a reliability evaluation index of the product to be tested based on the first failure life of the plurality of products to be tested.

[0089] Among them, reliability evaluation is to conduct statistical analysis, model fitting and comprehensive evaluation of the performance degradation data of multiple products to be tested, calculate reliability evaluation indicators (such as reliability, mean time before failure, etc.), and evaluate whether the reliability level of the products to be tested meets the design requirements.

[0090] In an embodiment of the present application, a first number of products to be tested that fail within a preset time can be obtained from a first failure life of multiple products to be tested, and the number of multiple products to be tested is a second number. The failure rate of the products to be tested is calculated according to the formula failure rate = (first number ÷ second number) × 100%. The larger the failure rate, the lower the product reliability.

[0091] In the above-mentioned reliability test and evaluation method of mathematical and rational fusion, by obtaining the performance degradation data of multiple detection components of each subtype under different component types in the product to be tested at different detection times, for multiple detection components of each subtype under the same component type, according to the performance degradation data of the multiple detection components, the first cumulative distribution function corresponding to the detection components of each subtype is determined, and the first failure life of the product to be tested is determined according to each first cumulative distribution function. According to the first failure life of the multiple products to be tested, the reliability evaluation index of the product to be tested is determined; different component types include mathematical statistics, failure physics and intelligent algorithm; the first cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the detection component under the subtype. The embodiment of the present application divides the detection components in the product to be tested into mathematical statistics, failure physics and intelligent algorithm types of detection components. Different algorithms are used to process the performance degradation data of different component types. The quantitative analysis ability of mathematical statistics, the mechanism description ability of failure physics models and the complex data processing ability of intelligent algorithms can be combined to fully utilize the advantages of multiple algorithms to flexibly process the performance degradation data of different component types, and obtain the first cumulative distribution function of the sub-type detection components under each component type. Therefore, the first failure life of the product to be detected can be determined based on the first cumulative distribution function of the detection components of each sub-type. The first failure life of multiple products to be detected is used to comprehensively evaluate the reliability of the product to be detected, comprehensively reflect the reliability evaluation index of the product to be detected, and improve the accuracy of the reliability evaluation.

[0092] Figure 3 FIG. 1 is a flow chart of a method for determining a first cumulative distribution function in one embodiment. Figure 3 As shown, the embodiment of the present application relates to a possible implementation method of determining a first cumulative distribution function corresponding to each subtype of detection component based on performance degradation data of multiple detection components, including the following steps:

[0093] S301 , normalizing the performance degradation data and detection time of multiple detection components to obtain a feature normalization sequence and a time normalization sequence.

[0094] In the embodiment of the present application, the subtype of the intelligent algorithm is The detection components are described using the support vector machine as an example. Intelligent algorithm detection components No. The detection time is , the performance degradation data is , for each detection time and detection component Each performance degradation data in is standardized using the following formulas (9) and (10) to obtain the time standard value and feature standard value.

[0095] (9)

[0096] (10)

[0097] Where: and They are and The time standard value and characteristic standard value of .

[0098] Thus, we can obtain the detection components of intelligent algorithm Time-standardized series and feature-standardized series, time-standardized series and feature normalization sequence As shown below:

[0099]

[0100]

[0101] Through normalization, the original performance degradation data and detection time can be scaled to a distribution with a mean of 0 and a variance of 1, avoiding the influence of large numerical differences on the weight of the support vector machine.

[0102] The support vector machine in the embodiment of the present application is a pre-built and trained model, and the parameters of the support vector machine are set. For example, a radial basis function is used, a regularization strength of 100, a kernel width of 0.1, an insensitivity loss of 0.01, etc. The support vector machine is trained using a sample time normalization sequence and a sample feature normalization sequence.

[0103] S302: Input the feature-normalized sequence and the time-normalized sequence into a prediction model to obtain a first prediction sequence corresponding to the feature-normalized sequence and a second prediction sequence corresponding to the time-normalized sequence.

[0104] In the embodiment of the present application, the feature normalization sequence and the time normalization sequence are input into the support vector machine to generate the first prediction sequence and the second prediction sequence , the first prediction sequence and the second prediction sequence As shown below:

[0105]

[0106]

[0107] That is, the number of columns (number of detections) of the first prediction sequence and the second prediction sequence is larger than the number of columns (number of detections) of the original performance degradation data and detection time, that is, , the data after the maximum detection time can be predicted.

[0108] S303 : Perform inverse normalization processing on the first prediction sequence and the second prediction sequence to obtain a first prediction inverse normalized sequence and a second prediction inverse normalized sequence.

[0109] In the embodiment of the present application, for the first prediction sequence and the second prediction sequence , use formula (11) and (12) to perform inverse normalization and obtain the first predicted inverse normalized sequence and the second predicted denormalized sequence , formulas (11) and (12) are as follows:

[0110] (11)

[0111] (12)

[0112] First forecast denormalized sequence The specific expression is: .

[0113] The second predicted denormalized sequence The specific expression is: .

[0114] S304: Determine a first cumulative distribution function corresponding to the subtype detection component according to the first predicted inverse normalized sequence and the second predicted inverse normalized sequence.

[0115] In the embodiment of the present application, according to the first predicted inverse normalization sequence, the second predicted inverse normalization sequence and the detection component The performance degradation threshold of the intelligent algorithm is determined to determine the detection components of the intelligent algorithm The failure life is obtained Intelligent algorithm detection components The failure life is For example, for testing components For the first detection component in the prediction, the detection time corresponding to the time when the detection time is less than the performance degradation critical value is determined according to the characteristic value corresponding to each detection time in the second prediction inverse normalized sequence and each detection time in the first prediction inverse normalized sequence, and the detection time is used as the failure life of the first detection component.

[0116] Similarly, assuming that the detection component of the intelligent algorithm The first cumulative distribution function is the Weibull distribution, as shown in formula (13):

[0117] (13)

[0118] in, and are the fifth and sixth unknown parameters. The fifth unknown parameter is calculated by formula (14): and the sixth unknown parameter Solving it, formula (14) is as follows:

[0119] (14)

[0120] Thus, the detection components of the intelligent algorithm class are obtained The first cumulative distribution function , using the same method, the first cumulative distribution functions of the detection components of all subtypes of intelligent algorithm classes are obtained as follows: .

[0121] In an embodiment of the present application, the performance degradation data and detection time of multiple detection components are standardized to obtain a feature standardized sequence and a time standardized sequence, the feature standardized sequence and the time standardized sequence are input into a prediction model to obtain a first prediction sequence corresponding to the feature standardized sequence and a second prediction sequence corresponding to the time standardized sequence, the first prediction sequence and the second prediction sequence are inversely standardized to obtain a first prediction inverse standardized sequence and a second prediction inverse standardized sequence, and the first cumulative distribution function corresponding to the subtype detection component is determined based on the first prediction inverse standardized sequence and the second prediction inverse standardized sequence. In an embodiment of the present application, data standardization is performed on the performance degradation data and the detection time, and the first prediction sequence and the second prediction sequence output by the prediction model are inversely standardized, which improves the convergence speed and evaluation efficiency of the artificial intelligence algorithm, thereby improving the efficiency of determining the first cumulative distribution function.

[0122] Figure 4 FIG. 1 is a flow chart of a method for determining a reliability evaluation index in one embodiment. Figure 4 As shown, the embodiment of the present application relates to a possible implementation method of determining the reliability evaluation index of a product to be tested based on the first failure life of multiple products to be tested, including the following steps:

[0123] S401 , determining a second cumulative distribution function corresponding to a plurality of products to be tested based on first failure lives of the products to be tested; the second cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the products to be tested.

[0124] In the embodiment of the present application, random sampling is performed For each product to be tested, repeat the steps of obtaining the first failure life of the product to be tested, and obtain The first failure life of the product to be tested .

[0125] Similarly, it is assumed that the second cumulative distribution function of the product to be tested is Weibull distribution, as shown in formula (15):

[0126] (15)

[0127] Where: is the second cumulative distribution function; is the seventh unknown parameter and The eighth unknown parameter. The first failure life of the product to be tested The seventh unknown parameter and the eighth unknown parameter are solved by using formula (16) to obtain the first cumulative distribution function of the product to be tested. Formula (16) is as follows:

[0128] (16)

[0129] S402 , determining an average lifespan of the product to be tested according to the second cumulative distribution function; the reliability evaluation index includes the average lifespan of the product to be tested.

[0130] In the embodiment of the present application, formula (17) can be solved to obtain the average life of the product to be tested. Formula (17) is as follows:

[0131] (17)

[0132] In an embodiment of the present application, the second cumulative distribution function corresponding to the products to be tested is determined based on the first failure life of multiple products to be tested, and the average life of the products to be tested is determined based on the second cumulative distribution function. In an embodiment of the present application, the second cumulative distribution function corresponding to the products to be tested is determined based on the first failure life of multiple products to be tested, so that the average life of the products to be tested is obtained based on the second cumulative distribution function, thereby improving the accuracy of determining the average life.

[0133] Figure 5 FIG. 1 is a flow chart of a method for determining a reliability evaluation index in another embodiment. Figure 5 As shown, the embodiment of the present application relates to another possible implementation method of determining the reliability evaluation index of a product to be tested based on the first failure life of multiple products to be tested, including the following steps:

[0134] S501 , determining a reliability function of the product to be tested based on a target parameter corresponding to the second cumulative distribution function; the reliability function is used to characterize the relationship between the failure life and reliability of the product to be tested.

[0135] In the embodiment of the present application, the target parameters are the seventh unknown parameter and the eighth unknown parameter in the above formula (17). The reliability function can be obtained by solving formula (17). The reliability function is shown in formula (18):

[0136] (18)

[0137] S502 , determining the reliability of the product to be tested based on the target failure life and reliability function of the product to be tested; the reliability evaluation index includes the reliability of the product to be tested.

[0138] In the embodiment of the present application, the target failure life is substituted into formula (18) to obtain the reliability of the product to be tested under the target failure life.

[0139] In the embodiment of the present application, the reliability function of the product to be tested is determined according to the target parameter corresponding to the second cumulative distribution function, and the reliability of the product to be tested is determined according to the target failure life of the product to be tested and the reliability function. Figure 4 The illustrated embodiment proposes a comprehensive product reliability evaluation method that takes failure competition into consideration, thereby improving the accuracy of reliability evaluation of the product to be tested.

[0140] In one embodiment, obtaining performance degradation data of multiple inspection components of each subtype under different component types in a product to be inspected at different inspection times includes: determining the inspection times corresponding to the different component types based on the inspection time intervals corresponding to the different component types; and controlling the inspection equipment to collect performance degradation data of multiple inspection components of each subtype under the corresponding component type at the inspection times corresponding to the different component types, so as to obtain the performance degradation data at different inspection times.

[0141] During the reliability test process, the traditional method is to conduct the same test on the entire product to be tested, that is, the test time intervals of the reliability test are set the same. Different test methods are not implemented according to the characteristics of the test components of different types of components in the product to be tested, which often leads to high reliability test costs.

[0142] In the embodiments of this application, Figure 6As shown, the mathematical statistics detection components include detection components of subtype A, the failure physics detection components include detection components of subtype B, and the intelligent algorithm detection components include detection components of subtype C. For example, the number of detection components included in each subtype is 1. For mathematical statistics detection components, the required test data volume is medium, so the detection interval can be medium. For failure physics detection components, the required test data volume is small, so the detection interval can be long. For intelligent algorithm detection components, the required test data volume is large, so the detection interval can be short. For mathematical statistics detection components, detection devices 11, 12, ..., 1A are used to obtain performance degradation data for detection components of subtype A. For failure physics detection components, detection devices 21, 22, ..., 2B are used to obtain performance degradation data for detection components of subtype B. For intelligent algorithm detection components, detection devices 31, 32, ..., 3C are used to obtain performance degradation data for detection components of subtype C.

[0143] In the embodiment of the present application, the detection time corresponding to different component types is determined based on the detection time intervals corresponding to different component types; the detection equipment is controlled to detect the detection time corresponding to different component types, and the performance degradation data of the detection components of the corresponding component types is collected to obtain the performance degradation data of the detection components under different component types at different detection times. In the embodiment of the present application, in the reliability test, different detection methods are implemented for different component types in the product to be tested to obtain corresponding performance degradation data. This method is more targeted and can reduce the cost of reliability testing, thereby improving the accuracy of reliability evaluation and expanding the scope of application of reliability testing and reliability evaluation.

[0144] Figure 7 FIG. 1 is a flow chart of a first failure life determination method in one embodiment. Figure 7 As shown, the embodiment of the present application relates to a possible implementation method of how to determine the first failure life of the product to be tested according to each first cumulative distribution function, including the following steps:

[0145] S701 , for each subtype of the same component type, determine a second failure life of the detection component of the subtype according to a first cumulative distribution function and a random failure probability corresponding to the detection component of the subtype.

[0146] According to the data analysis of the above reliability test, the first cumulative distribution functions of all the detection components of mathematical statistics can be obtained, which are , the first cumulative distribution functions of all failed physical class detection components are , the first cumulative distribution functions of all intelligent algorithm detection components are Since there is a failure competition relationship between the mathematical statistics detection components, the failure physics detection components, and the intelligent algorithm detection components in the product to be detected, the embodiment of the present application adopts a comprehensive reliability evaluation method based on failure competition. The steps of this method are as follows:

[0147] Randomly generate a number between 0 and 1 As the random failure probability, all first cumulative distribution functions are equal to ,like , , , solve it and you can get the second failure life of all sub-types of detection components. Assume that there are 10 sub-types of products to be tested, of which 3 sub-types belong to mathematical statistics, 3 sub-types belong to failure physics, and 4 sub-types belong to intelligent algorithms. The first failure life of all sub-types of detection components obtained by one sampling is .

[0148] S702 : Determine the first expiration life of the product to be inspected according to the second expiration life and connection relationship of the inspection components of each subtype under each component type.

[0149] In the embodiments of the present application, the connection relationships of the detection components of each subtype are analyzed based on the functional structure of the product to be detected. The connection relationships include non-redundant connections, redundant connections, and mixed connections. Non-redundant connections refer to connections in which the failure of any of the components to be detected will cause the product to fail. Redundant connections refer to connections in which the product to be detected will fail only when all components to be detected fail. Mixed connections refer to connections that include both non-redundant connections and redundant connections.

[0150] Assume that all the k subtypes of detection components are connected without redundancy, and the first failure life of the product to be detected is:

[0151] (19)

[0152] Where: is the second failure life of the detection component of each subtype.

[0153] Assume that all the k subtypes of detection components have redundant connections, and the first failure life of the product to be detected is:

[0154] (20)

[0155] For mixed connections, the first failure life of the product to be tested is analyzed based on the specific connection situation. For example, if the product to be tested has four sub-type detection components, the detection components of sub-type 1 and sub-type 2 have redundant connections, and the module composed of the detection components of sub-type 1 and sub-type 2 has non-redundant connections with the detection components of sub-type 3 and sub-type 4, then the first failure life of the product to be tested is:

[0156] (twenty one)

[0157] In an embodiment of the present application, for each subtype under the same component type, the second failure life of the detection component of the subtype is determined based on the first cumulative distribution function and random failure probability corresponding to the detection component of the subtype, and the first failure life of the product to be detected is determined based on the second failure life and connection relationship of the detection components of each subtype under each component type, thereby improving the accuracy of determining the first failure life of the product to be detected.

[0158] It should be understood that, although the steps in the flowcharts of the above embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily to be performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0159] Based on the same inventive concept, embodiments of the present application also provide a device for implementing the aforementioned method for reliability testing and evaluation of mathematical and logical fusion. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the device for reliability testing and evaluation of mathematical and logical fusion provided below can be found in the above-mentioned limitations of the reliability testing and evaluation method for mathematical and logical fusion, and will not be repeated here.

[0160] In an exemplary embodiment, Figure 8 As shown, a reliability test and evaluation device for integrating mathematical and rational information is provided, including: an acquisition module 81, a first determination module 82, a second determination module 83 and a third determination module 84, wherein:

[0161] An acquisition module 81 is used to acquire performance degradation data of multiple test components of different sub-types in the product to be tested at different test times; the different component types include mathematical statistics, failure physics, and intelligent algorithm types;

[0162] A first determining module 82 is configured to determine, for each of the plurality of detection components of each subtype under the same component type, a first cumulative distribution function corresponding to each of the detection components of each subtype based on the performance degradation data of the plurality of detection components; the first cumulative distribution function is configured to characterize the relationship between the failure life and failure probability of the detection components of each subtype;

[0163] A second determination module 83 is configured to determine a first failure life of the product to be tested according to each first cumulative distribution function;

[0164] The third determining module 84 is configured to determine a reliability evaluation index of the product to be inspected according to the first failure life of the plurality of products to be inspected.

[0165] In one embodiment, the first determination module 82 is specifically used to standardize the performance degradation data and detection time of multiple detection components to obtain a feature standardized sequence and a time standardized sequence; input the feature standardized sequence and the time standardized sequence into the prediction model to obtain a first prediction sequence corresponding to the feature standardized sequence and a second prediction sequence corresponding to the time standardized sequence; perform inverse normalization on the first prediction sequence and the second prediction sequence to obtain a first predicted inverse normalized sequence and a second predicted inverse normalized sequence; and according to the first predicted inverse normalized sequence and the second predicted inverse normalized sequence, obtain a first cumulative distribution function corresponding to the subtype of the detection component.

[0166] In one embodiment, the third determination module 84 is specifically used to determine the second cumulative distribution function corresponding to the product to be tested based on the first failure life of multiple products to be tested; the second cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the product to be tested; the average life of the product to be tested is determined based on the second cumulative distribution function; and the reliability evaluation index includes the average life of the product to be tested.

[0167] In one embodiment, the third determination module 84 is specifically used to determine the reliability function of the product to be tested based on the target parameters corresponding to the second cumulative distribution function; the reliability function is used to characterize the relationship between the failure life and reliability of the product to be tested; the reliability of the product to be tested is determined based on the target failure life and the reliability function of the product to be tested; the reliability evaluation index includes the reliability of the product to be tested.

[0168] In one embodiment, the acquisition module 81 is specifically used to determine the detection time corresponding to different component types based on the detection time intervals corresponding to different component types; control the detection equipment at the detection time corresponding to different component types, and collect performance degradation data of multiple detection components of each subtype under the corresponding component type to obtain performance degradation data at different detection times.

[0169] In one embodiment, the second determination module 83 is specifically configured to determine, for each subtype under the same component type, the second failure life of the detection component of the subtype according to the first cumulative distribution function and random failure probability corresponding to the detection component of the subtype; and determine the first failure life of the product to be detected according to the second failure life and connection relationship of the detection components of each subtype under each component type.

[0170] Each module in the aforementioned mathematical and intellectual fusion reliability testing and evaluation device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a computer device memory in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0171] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of any of the above method embodiments when executing the computer program.

[0172] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above method embodiments are implemented.

[0173] In one embodiment, a computer program product is provided, comprising a computer program, which implements the steps of any of the above method embodiments when executed by a processor.

[0174] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0175] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0176] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A reliability test and evaluation method based on mathematical and rational fusion, characterized in that: The method comprises: Obtaining performance degradation data of multiple test components of different subtypes under different component types in the product to be tested at different test times; the different component types include mathematical statistics type, failure physics type, and intelligent algorithm type; For a plurality of detection components of each subtype under the same component type, determining a first cumulative distribution function corresponding to each detection component of the subtype based on performance degradation data of the plurality of detection components; the first cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the detection component of the subtype; determining a first failure life of the product to be tested according to each of the first cumulative distribution functions; Determining a reliability evaluation index of the product to be tested based on a plurality of first failure lives of the product to be tested; The first cumulative distribution function is a first cumulative distribution function corresponding to the intelligent algorithm class; and determining, based on the performance degradation data of the plurality of detection components, the first cumulative distribution function corresponding to the detection components of each subtype includes: The performance degradation data of the plurality of detection components and the detection time are standardized to obtain a feature-standardized sequence and a time-standardized sequence; the feature-standardized sequence and the time-standardized sequence are input into a prediction model to obtain a first prediction sequence corresponding to the feature-standardized sequence and a second prediction sequence corresponding to the time-standardized sequence; the first prediction sequence and the second prediction sequence are inverse-standardized to obtain a first prediction inverse-standardized sequence and a second prediction inverse-standardized sequence; and a first cumulative distribution function corresponding to the detection component of the subtype is determined based on the first prediction inverse-standardized sequence and the second prediction inverse-standardized sequence.

2. The method according to claim 1, characterized in that Determining the reliability evaluation index of the product to be tested based on the first failure life of the plurality of products to be tested includes: Determining a second cumulative distribution function corresponding to the product to be detected based on the first failure life of the plurality of products to be detected; the second cumulative distribution function is used to characterize the relationship between the failure life and failure probability of the product to be detected; The average life of the product to be tested is determined according to the second cumulative distribution function; and the reliability evaluation index includes the average life of the product to be tested.

3. The method according to claim 2, characterized in that The method further comprises: Determining a reliability function of the product to be tested based on a target parameter corresponding to the second cumulative distribution function; the reliability function is used to characterize the relationship between the failure life and reliability of the product to be tested; The reliability of the product to be tested is determined according to the target failure life of the product to be tested and the reliability function; the reliability evaluation index includes the reliability of the product to be tested.

4. The method according to claim 1, wherein The obtaining of performance degradation data of multiple test components of each subtype under different component types in the product to be tested at different test times includes: Determine the inspection time corresponding to different component types based on the inspection time intervals corresponding to different component types; The detection equipment is controlled to detect the detection time corresponding to different component types, and the performance degradation data of multiple detection components of each subtype under the corresponding component type are collected to obtain the performance degradation data at different detection times.

5. The method according to claim 1, characterized in that Determining the first failure life of the product to be tested according to each of the first cumulative distribution functions includes: For each subtype of the same component type, determining a second failure life of the detection component of the subtype according to a first cumulative distribution function and a random failure probability corresponding to the detection component of the subtype; The first failure life of the product to be inspected is determined according to the second failure life and the connection relationship of the inspection components of each sub-type under each component type.

6. A reliability test and evaluation device integrating mathematical and rational fusion, characterized in that: The device comprises: An acquisition module is used to obtain performance degradation data of multiple test components of different subtypes under different component types in the product to be tested at different test times; the different component types include mathematical statistics, failure physics, and intelligent algorithm types; a first determining module configured to determine, for a plurality of detection components of each subtype under the same component type, a first cumulative distribution function corresponding to each detection component of the subtype based on performance degradation data of the plurality of detection components; wherein the first cumulative distribution function is configured to represent a relationship between a failure life and a failure probability of the detection component of the subtype; A second determining module is configured to determine a first failure life of the product to be tested according to each of the first cumulative distribution functions; A third determining module is configured to determine a reliability evaluation index of the product to be detected based on a plurality of first failure lives of the product to be detected; The first cumulative distribution function is the first cumulative distribution function corresponding to the intelligent algorithm class; the first determination module is specifically used to standardize the performance degradation data and the detection time of multiple detection components to obtain a feature standardized sequence and a time standardized sequence; input the feature standardized sequence and the time standardized sequence into the prediction model to obtain a first prediction sequence corresponding to the feature standardized sequence and a second prediction sequence corresponding to the time standardized sequence; perform inverse normalization on the first prediction sequence and the second prediction sequence to obtain a first predicted inverse normalized sequence and a second predicted inverse normalized sequence; determine the first cumulative distribution function corresponding to the detection component of the subtype based on the first predicted inverse normalized sequence and the second predicted inverse normalized sequence.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images