Mathematical and rational fusion of accelerated testing and life assessment methods, devices and equipment

Abstract

The present application relates to a method, apparatus, and device for accelerated testing and life assessment that integrates mathematical and logical reasoning. The method comprises: placing different components of a product to be tested in different accelerated test chambers, and applying different accelerated stresses to each of the accelerated test chambers; obtaining test data for each component according to the failure type of each component; performing data analysis on each test data according to the data type of each component to obtain a life distribution function for each component; and determining the life assessment test results for the product to be tested based on the failure correlations between the components and the life distribution functions of each component. This method can improve the accuracy of life assessment results.

Description

Technical Field

[0001] The present application relates to the technical field of product life assessment, and in particular to a method, device and equipment for accelerated testing and life assessment that integrates mathematical and rational analysis. Background Art

[0002] As industrial products become more complex and market demands for reliability become increasingly stringent, accelerated testing is often used to assess product lifespan.

[0003] In related technologies, the entire product is usually placed in the same test box, the same stress is applied to all devices in the product, and the same detection method and the same data analysis method are used to achieve accelerated testing and life assessment of the product.

[0004] However, the lifespan assessment method in the related art has a technical problem of low accuracy of the assessment results. Summary of the Invention

[0005] Based on this, it is necessary to provide a mathematical and rational fusion of accelerated testing and life assessment methods, devices and equipment to address the above technical problems, which can improve the accuracy of life assessment results.

[0006] In a first aspect, the present invention provides a method for accelerated testing and lifespan assessment that integrates mathematical and rational analysis, including:

[0007] When different components of the product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers, test data of each component device is obtained according to the failure type of each component device;

[0008] According to the data type of each test data, the test data are analyzed to obtain the life distribution function of each component equipment;

[0009] The life evaluation test results of the product to be tested are determined based on the failure correlation between the component devices and the life distribution function of the component devices.

[0010] In one embodiment, obtaining test data of each component device according to the failure type of each component device includes:

[0011] Obtain the test data acquisition requirements for each component device based on the failure type of each component device;

[0012] According to each test data acquisition requirement, in the mapping relationship between the test data acquisition requirement and the test detection method, obtain the test detection method that matches each test data acquisition requirement;

[0013] Adopt the test and detection method that matches the requirements for obtaining each test data to test each component equipment and obtain the test and detection data of each component equipment.

[0014] In one embodiment, when a component device is a hard failure device, the data type includes failure time data; and based on the data type of each test and detection data, data analysis is performed on each test and detection data to obtain a life distribution function of each component device, including:

[0015] Determine the failure likelihood function of the hard failure device based on the number of failed devices in the hard failure device, the failure time data of each failed device, and the mathematical statistics model of the hard failure device;

[0016] According to the failure likelihood function, the life distribution function of hard failure equipment is determined.

[0017] In one embodiment, when the component device is a first soft failure device, the data type includes first performance degradation data; and based on the data type of each test detection data, data analysis is performed on each test detection data to obtain a life distribution function of each component device, including:

[0018] Acquire, according to the first performance degradation data, a detection time series and a detection value series of a performance parameter of the first soft-failure device;

[0019] According to the detection time series and the detection value series, the preset neural network is trained to obtain a performance detection value prediction model;

[0020] The life distribution function of the first soft failure device is determined according to the performance detection value prediction model.

[0021] In one embodiment, determining the life distribution function of the first soft failure device according to the performance detection value prediction model includes:

[0022] Inputting the candidate time series into the performance test value prediction model to obtain the test value prediction sequence output by the performance test value prediction model;

[0023] Determine the failure time of the first soft-failure device based on the candidate time series and the detection value prediction series;

[0024] According to the failure time, a life distribution function of the first soft failure device is determined.

[0025] In one embodiment, determining the life distribution function of the first soft failure device according to the failure time includes:

[0026] Determining life assessment parameters of the first soft failure device according to the number of devices and failure time of the first soft failure device;

[0027] The life distribution function of the first soft failure device is determined according to the life evaluation parameter.

[0028] In one embodiment, when the component device is a second soft failure device, the data type includes second performance degradation data; the amount of the second performance degradation data is less than the amount of the first performance degradation data; and based on the data type of each test detection data, data analysis is performed on each test detection data to obtain a life distribution function of each component device, including:

[0029] acquiring a detection value sequence of a performance parameter of a second soft-failure device according to the second performance degradation data;

[0030] determining a failure time of the second soft-failure device based on a failure physical model and a detection value sequence of the second soft-failure device;

[0031] According to the failure time, the life distribution function of the second soft failure device is determined.

[0032] In one embodiment, determining the life distribution function of the second soft-failure device based on the failure time includes:

[0033] Determining life assessment parameters of the second soft failure devices according to the number of devices and failure time of the second soft failure devices;

[0034] The life distribution function of the second soft failure device is determined according to the life evaluation parameter.

[0035] In one embodiment, determining the life assessment test result of the product to be tested based on the failure correlation between the component devices and the life distribution function of the component devices includes:

[0036] Determine the life distribution function of the product to be tested based on the failure correlation between each component device and the life distribution function of each component device;

[0037] The life evaluation test result of the product to be tested is determined based on the acceleration factor of the product to be tested under accelerated stress and the life distribution function of the product to be tested.

[0038] In a second aspect, the present application also provides a mathematical and rational fusion accelerated testing and life assessment device, comprising:

[0039] A data acquisition module is used to obtain test data of each component device according to the failure type of each component device when different components of the product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers;

[0040] A data analysis module is used to analyze the test data according to the data type of each test data to obtain the life distribution function of each component equipment;

[0041] The result determination module is used to determine the life evaluation test result of the product to be tested based on the failure correlation between the component devices and the life distribution function of the component devices.

[0042] 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 when the processor executes the computer program, the steps of the method in any one of the embodiments of the first aspect are implemented.

[0043] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method in any one of the embodiments of the first aspect above.

[0044] In a fifth aspect, the present application also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method in any one of the embodiments of the first aspect above.

[0045] The embodiment of the present application provides a method, device and equipment for accelerated testing and life assessment that integrates mathematical and rational elements. When different components of the product to be tested are placed in different accelerated test chambers, and different accelerated test chambers are respectively applied with different accelerated stresses, the test detection data of each component device is obtained according to the failure type of each component device, and then the test detection data are analyzed according to the data type of each test detection data to obtain the life distribution function of each component device. Then, the life assessment test result of the product to be tested is determined based on the failure correlation between each component device and the life distribution function of each component device. In this method, when the life assessment test of the product to be tested is performed, the different components of the product to be tested are first placed in different accelerated test chambers and different accelerated stresses are applied to the different accelerated test chambers. On this basis, based on the failure type of each component device, different detection methods are used to detect each component device to obtain the test detection data of each component device. Then, based on the data type of the test detection data, different data analysis methods are used to analyze each test detection data to obtain the life distribution function of each component device, thereby obtaining the life assessment result of the product to be tested based on the life distribution function of each component device. During this process, stress application, equipment testing and data analysis are all performed on different component equipment according to their types, making the life assessment of each component equipment more accurate, thereby improving the accuracy of product life assessment; at the same time, the failure correlation between different component equipment is also considered during the life assessment, further improving the accuracy of the life assessment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] 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 or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 is a diagram of the internal structure of a computer device in one embodiment;

[0048] Figure 2 1. A flow chart of a method for accelerating testing and lifespan assessment by integrating mathematical reasoning in one embodiment;

[0049] Figure 3 is a schematic diagram of stress application in one embodiment;

[0050] Figure 4 A schematic diagram of a process for obtaining test data in one embodiment;

[0051] Figure 5 A schematic diagram of an experimental test in one embodiment;

[0052] Figure 6 1 is a schematic diagram of a process for determining a life distribution function in one embodiment;

[0053] Figure 7 is a schematic diagram of a process for determining a lifetime distribution function in another embodiment;

[0054] Figure 8 is a schematic diagram of a process for determining a lifetime distribution function in another embodiment;

[0055] Figure 9 is a schematic diagram of a process for determining a lifetime distribution function in another embodiment;

[0056] Figure 10 A schematic diagram of a process for determining life assessment test results in one embodiment;

[0057] Figure 11 Schematic diagram of the structure of a mathematical and rational fusion accelerated testing and life assessment device in one embodiment. DETAILED DESCRIPTION

[0058] 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.

[0059] The following first describes the technical background of the embodiments of the present application.

[0060] Accelerated testing is a testing method that artificially applies stresses (such as temperature, voltage, and mechanical loads) exceeding normal operating conditions to stimulate potential product failure mechanisms and shorten their duration. Its core purpose is to simulate, in a relatively short period of time, the performance degradation or failure processes that a product would experience in long-term, real-world use, thereby enabling rapid assessment of product lifespan. Accelerated testing can compress traditional lifespan tests, which typically take years, into months or weeks, making it widely used in the lifespan assessment of high-reliability products.

[0061] Traditional accelerated testing and lifespan assessment methods place the entire product in the same test chamber, apply the same accelerated stress, and employ the same testing method, using a single assessment method to achieve both accelerated testing and lifespan assessment. However, traditional accelerated testing methods lack specificity, resulting in low lifespan assessment accuracy.

[0062] Based on this, the embodiment of the present application provides a method for accelerated testing and life assessment that integrates mathematical reasoning. When conducting a life assessment test on a product to be tested, first, different components of the product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers. On this basis, based on the failure type of each component, different detection methods are used to detect each component to obtain test data of each component. Then, based on the data type of the test data, different data analysis methods are used to analyze each test data to obtain the life distribution function of each component. Thus, based on the life distribution function of each component, the life assessment result of the product to be tested is obtained. In this process, stress application, equipment detection and data analysis are all performed on different components according to their types, making the life assessment of each component more accurate, thereby improving the accuracy of the product's life assessment. At the same time, the failure correlation between different components is also considered during the life assessment, further improving the accuracy of the life assessment result. Of course, the technical solution provided in the embodiment of the present application is not limited to solving the above problems, but also has other technical effects. For details, please refer to the following embodiment.

[0063] It should be noted that the beneficial effects or technical problems solved by the embodiments of the present application are not limited to this one, but may also include other implicit or related problems. For details, please refer to the description of the following embodiments.

[0064] The following mathematical and rational fusion accelerated test and life assessment method provided in the embodiment of the present application can be applied to computer equipment. The computer equipment can be a server, and its internal structure diagram can be as follows: Figure 1As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. Among them, the processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, 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 data. 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 through a network connection. When the computer program is executed by the processor, it realizes a mathematical and rational fusion accelerated test and life assessment method. Those skilled in the art can 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.

[0065] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0066] In an exemplary embodiment, Figure 2 As shown in the figure, a mathematical and rational fusion accelerated test and life assessment method is provided, which is applied to Figure 1 The computer device in the embodiment is used as an example to illustrate the method, including the following steps 201 to 203.

[0067] S201 , placing different components of a product to be tested in different accelerated test chambers, and applying different accelerated stresses to the different accelerated test chambers, and obtaining test data of each component according to the failure type of each component.

[0068] In the embodiments of the present application, the components of the product to be tested are divided into hard failure devices and soft failure devices according to the failure type. Hard failure refers to the sudden and complete loss of the specified function of the equipment at a certain moment, which is usually unpredictable and instantaneous. This failure mode is usually caused by a sudden external shock or internal failure. Soft failure refers to the gradual decline in the function or performance of the equipment over time, eventually reaching a state where it cannot work normally. This failure mode is usually caused by factors such as long-term wear, corrosion, fatigue or aging. At the same time, soft failure devices are further divided into soft failure devices with unknown failure physical models (Class A soft failure devices) and soft failure devices with known failure physical models (Class B soft failure devices).

[0069] When conducting an accelerated test on a product to be tested, first place the components of the product to be tested into different accelerated test chambers and apply different sensitive stress types and stress values. Figure 3 As shown, the components of the product under test include hard failure devices, Class A soft failure devices, and Class B soft failure devices. Hard failure devices are primarily sensitive to stresses such as mechanical loads, and the accelerated test stresses are mechanical load stresses. Soft failure devices are primarily sensitive to stresses such as temperature and humidity, and the accelerated test stresses are temperature and humidity stresses. The different components are interconnected, and accelerated testing is performed simultaneously. It should be noted that for multiple products under test, the same type of components for each product are placed in the same accelerated test chamber.

[0070] After placing different components of the product to be tested into different accelerated test chambers and applying different sensitive stress types and stress values to the different accelerated test chambers, the different components are tested to obtain test data for each component.

[0071] In practical applications, based on the failure type of each component device, a detection method matching each failure type can be obtained. Using this detection method, each component device can be tested to obtain test data for each component device. For example, a mapping relationship between failure types and detection methods can be obtained from a database. Then, based on the failure type of each component device, a detection method corresponding to each failure type can be matched from this mapping relationship.

[0072] S202 , performing data analysis on each test data according to the data type of each test data to obtain the life distribution function of each component device.

[0073] After obtaining the test data of each component device, it is necessary to perform data analysis on each test data to obtain the life distribution function of each component device.

[0074] In the embodiments of the present application, when performing data analysis on each test data, a data analysis method that matches each data type is obtained based on the data type of each test data. Using the data analysis method that matches each data type, data analysis is performed on the test data of each component device to obtain a life distribution function for each component device. For example, data analysis methods include mathematical statistics methods, artificial intelligence algorithms, and failure physics models. Different test data corresponds to different data analysis methods. Based on the test data, a data analysis method that is suitable for the test data is selected, and based on the selected data analysis method, the test data is combined to determine the life distribution function of each component device.

[0075] S203 , determining a life evaluation test result of the product to be tested based on the failure correlation between the component devices and the life distribution function of the component devices.

[0076] The failure correlation between the component devices refers to the related competitive failure relationship between the component devices, that is, the failures between the devices are competitive and correlated.

[0077] Based on the lifetime distribution functions of the component devices obtained above, combined with the failure correlations between the component devices, the lifetime of the product under test is evaluated to obtain a lifetime evaluation test result for the product under test. For example, the failure correlations between the component devices are combined with the lifetime distribution functions of the component devices to obtain a lifetime distribution function for the product under test, and the lifetime distribution function of the product under test is analyzed to obtain a lifetime evaluation test result for the product under test.

[0078] In the accelerated test and life assessment method of the embodiment of the present application, when different components of the product to be tested are placed in different accelerated test chambers, and different accelerated test chambers are respectively applied with different accelerated stresses, test detection data of each component device is obtained according to the failure type of each component device, and then data analysis is performed on each test detection data according to the data type of each test detection data to obtain the life distribution function of each component device, and then the life assessment test result of the product to be tested is determined based on the failure correlation between each component device and the life distribution function of each component device. In this method, when the life assessment test of the product to be tested is performed, first, different components of the product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers. On this basis, based on the failure type of each component device, different detection methods are used to detect each component device to obtain test detection data of each component device, and then based on the data type of the test detection data, different data analysis methods are used to analyze each test detection data to obtain the life distribution function of each component device, thereby obtaining the life assessment result of the product to be tested based on the life distribution function of each component device. During this process, stress application, equipment testing and data analysis are all performed on different component equipment according to their types, making the life assessment of each component equipment more accurate, thereby improving the accuracy of product life assessment; at the same time, the failure correlation between different component equipment is also considered during the life assessment, further improving the accuracy of the life assessment results.

[0079] Based on the above embodiment, an embodiment is provided to illustrate the process of obtaining the test detection data.

[0080] In an exemplary embodiment, Figure 4 As shown, according to the failure type of each component equipment, the test data of each component equipment is obtained, including:

[0081] S301 , obtaining a detection data acquisition requirement of each component device according to the failure type of each component device.

[0082] The failure types of each component device include hard failure devices, soft failure devices with unknown failure physical models, and soft failure devices with known failure physical models.

[0083] Different failure types require different test data acquisition requirements. For hard-failure devices, the requirement is to obtain the exact failure time of the device. For soft-failure devices with unknown failure physics models, the requirement is to obtain a large amount of performance degradation data. For soft-failure devices with known failure physics models, the requirement is to obtain a small amount of performance degradation data.

[0084] S302 : According to each test data acquisition requirement, in the mapping relationship between the test data acquisition requirement and the test detection method, a test detection method matching each test data acquisition requirement is acquired.

[0085] The database stores the mapping relationship between different test data acquisition requirements and different test detection methods. When testing each component device, the test data acquisition requirements of each component device can be first based on the mapping relationship to match the test detection method corresponding to each test data acquisition requirement.

[0086] Among them, the test data acquisition requirement is to obtain the accurate failure time of the equipment, and its corresponding test detection method is real-time online detection to obtain the specific failure time. The test data acquisition requirement is to obtain a large amount of performance degradation data, and its corresponding test detection method is offline regular detection with a short time interval. The test data acquisition requirement is a small amount of performance degradation data, and its corresponding test detection method is offline regular detection with a long time interval. The test detection methods corresponding to different components of the equipment are as follows: Figure 5 shown.

[0087] S303: Test each component device using a test detection method that matches the test data acquisition requirements to obtain test detection data of each component device.

[0088] After obtaining the test and detection methods for each component device, each component device is tested using each test and detection method to obtain test and detection data for each component device. Specifically, for hard failure devices, real-time online testing is performed on each hard failure device using real-time online testing equipment; for Class A soft failure devices, offline testing is performed on each Class A soft failure device using offline testing equipment, with a short testing interval; for Class B soft failure devices, offline testing is performed on each Class B soft failure device using offline testing equipment, with a long testing interval.

[0089] In the accelerated test and life assessment method of the mathematical and rational fusion provided in the embodiment of the present application, first, based on the failure type of each component device, the detection data acquisition requirements of each component device are obtained, and then, based on each detection data acquisition requirement, in the mapping relationship between the detection data acquisition requirements and the test detection method, the test detection method that matches the detection data acquisition requirement is obtained, and then the test detection method that matches the detection data acquisition requirement is adopted to test each component device and obtain the test detection data of each component device. In this method, first, based on the failure type of different component devices, the detection data acquisition requirements of each component device are obtained, and different detection data requirements correspond to different test detection methods. Based on this, the test detection method of each component device is obtained from the mapping relationship between the detection data acquisition requirements and the test detection method, so that the corresponding component device is tested based on the test detection method of each component device, and the test detection data of each component device is obtained, thereby improving the accuracy of the detection data and thus improving the accuracy of the life assessment.

[0090] Based on the above embodiment, an embodiment is provided to illustrate the process of determining the lifetime distribution function.

[0091] In an exemplary embodiment, Figure 6 As shown, when the component equipment is a hard failure equipment, the data type includes failure time data; according to the data type of each test detection data, the test detection data is analyzed to obtain the life distribution function of each component equipment, including:

[0092] S401 : determining a failure likelihood function of a hard failure device based on the number of failed devices in the hard failure device, failure time data of each failed device, and a mathematical statistics model of the hard failure device.

[0093] In the embodiment of the present application, the accelerated test data analysis of the hard failure device adopts a mathematical statistics method.

[0094] For n hard failure devices, assuming a total of k ( ) hard failure devices fail, then the failure time of the hard failure devices is sorted from small to large. Assuming that the life distribution function of hard failure equipment is log-normal distribution, the mathematical statistics model of hard failure equipment is as follows (1).

[0095] (1)

[0096] in, is the life distribution function of hard failure equipment; and is the life assessment parameter; n is the number of products to be tested, and is also the number of hard failure devices, Class A soft failure devices, and Class B soft failure devices.

[0097] According to the life distribution function of hard failure equipment, the failure likelihood function of hard failure equipment is obtained as follows (2).

[0098] (2)

[0099] Where S is the failure likelihood function and t0 is the cutoff time of the accelerated test.

[0100] S402: Determine the life distribution function of the hard failure device according to the failure likelihood function.

[0101] By solving the following equation group (3), the life assessment parameters of hard failure equipment are obtained: and , thus obtaining the life distribution function of hard failure equipment.

[0102] (3)

[0103] In the mathematically and statistically integrated accelerated testing and life assessment method provided in the embodiments of the present application, the failure likelihood function of the hard failure device is first determined based on the number of failed devices in the hard failure device, the failure time data of each failed device, and the mathematical statistical model of the hard failure device. Then, based on the failure likelihood function, the life distribution function of the hard failure device is determined. In this method, the data type of the test detection data of the hard failure device is failure time data, and there is no need to predict the failure time. Therefore, a mathematical and statistical method is used to analyze the failure time data of each component device to obtain the life distribution function of the hard failure device. By using a data analysis method suitable for hard failure device detection data to analyze the hard failure device data, the accuracy of the life assessment of the hard failure device is improved.

[0104] Based on the above embodiment, another embodiment is provided to illustrate the process of determining the lifetime distribution function.

[0105] In an exemplary embodiment, Figure 7 As shown, when the component device is a first soft failure device, the data type includes first performance degradation data; according to the data type of each test detection data, data analysis is performed on each test detection data to obtain the life distribution function of each component device, including:

[0106] S501 : Acquire a detection time series and a detection value series of a performance parameter of a first soft-failure device according to first performance degradation data.

[0107] In the embodiment of the present application, the first soft failure device is a Class A soft failure device, and the accelerated test data analysis of the Class A soft failure device uses an artificial intelligence algorithm. The first performance degradation data is a large amount of performance degradation data obtained by testing the Class A soft failure device using an offline detection device.

[0108] For n Class A soft failure devices, assuming a total of times, and the detection times are , the detection values of the performance parameters of n Class A soft failure devices are shown in Table 1 below.

[0109] Table 1

[0110]

[0111] Among them, the detection time series The test value sequence x is shown in the following formula (4).

[0112] (4)

[0113] S502: Training a preset neural network according to the detection time series and the detection value series to obtain a performance detection value prediction model.

[0114] After obtaining the detection time series and detection value series of Class A soft-failure devices, the detection times and detection values are normalized to obtain a normalized detection time series and a normalized detection value series. These normalized detection time series and normalized detection value series are then used as training data for the neural network model training. The normalized detection time series is input into the neural network to obtain the predicted detection values output by the neural network. Loss calculation is performed based on the predicted detection values and the normalized detection value series. The model parameters of the neural network are continuously adjusted based on the calculated loss values until the training is completed, resulting in a performance prediction model.

[0115] Among them, the normalization formula of the detection time and the normalization formula of the detection value of the performance parameter are as follows: Formulas (5) and (6).

[0116] (5)

[0117] in, for Normalized value; is the jth detection time; To obtain The minimum value of To obtain The maximum value of .

[0118] (6)

[0119] in, for Normalized value; is the j-th detection value of the i-th Class A soft failure device; To obtain The minimum value of To obtain The maximum value of .

[0120] Thus, the normalized detection time series is obtained And the normalized detection value sequence .

[0121] (7)

[0122] Artificial intelligence algorithm is used to normalize the detection time series And the normalized detection value sequence Conduct training to build an artificial intelligence model for Class A soft failure devices.

[0123] S503: Determine a life distribution function of the first soft failure device according to the performance detection value prediction model.

[0124] Based on the performance test value prediction model obtained through the above training, the test values at other subsequent times are predicted, and thus the life distribution function of the first soft failure device is determined based on the predicted test values.

[0125] In one embodiment, if Figure 8 As shown, according to the performance detection value prediction model, determining the life distribution function of the first soft failure device includes the following steps:

[0126] S601: Input the candidate time series into the performance prediction model to obtain the detection value prediction sequence output by the performance detection value prediction model.

[0127] The candidate time series Input the performance test value prediction model to obtain the test value prediction sequence .

[0128] (8)

[0129] S602: Determine the failure time of the first soft-failure device according to the candidate time series and the detection value prediction series.

[0130] After obtaining the detection value prediction sequence, the candidate time series and the detection value prediction sequence are respectively subjected to inverse normalization processing to obtain the inverse normalized candidate time series and the inverse normalized detection value prediction sequence. Then, based on the inverse normalized candidate time series and the inverse normalized detection value prediction sequence, the failure time of the first soft failure device is predicted.

[0131] The inverse normalization formula of the candidate time series is: .in, for The inverse normalized value. The inverse normalization formula for the detection value prediction sequence is: .in, for The inverse normalized value.

[0132] Thus, the candidate time series after inverse normalization is obtained And the detection value prediction sequence after inverse normalization As shown in formula (9).

[0133] (9)

[0134] Based on the candidate time series obtained after inverse normalization And the detection value prediction sequence after inverse normalization , combined with the failure threshold of Class A soft failure devices, the failure time is evaluated, and the failure times of n Class A soft failure devices are obtained as follows: .

[0135] Exemplarily, each predicted detection value in the detection value prediction sequence is compared with the failure threshold of the Class A soft failure device, a predicted detection value greater than the failure threshold is obtained, and the candidate time corresponding to the predicted detection value greater than the failure threshold is determined as the failure time of the first soft failure device.

[0136] S603: Determine a life distribution function of the first soft failure device according to the failure time.

[0137] Based on the estimated failure time and in combination with the life assessment parameters of the first soft failure device, a life distribution function of the first soft failure device is determined.

[0138] Exemplarily, based on the number of devices and failure time of the first soft failure devices, life assessment parameters of the first soft failure devices are determined; based on the life assessment parameters, a life distribution function of the first soft failure devices is determined.

[0139] Assuming that the life distribution function of Class A soft failure equipment is exponential distribution, its life distribution function is as shown in the following formula (10).

[0140] (10)

[0141] in, is the life distribution function of Class A soft failure equipment, It is the life assessment parameter of Class A soft failure equipment.

[0142] By solving the following equation (11), the life assessment parameters of Class A soft failure devices are evaluated: .

[0143] (11)

[0144] Thus, the life assessment parameters to be solved are Substituting into the above formula (10), the life distribution function of the first soft failure device is obtained: .

[0145] In the accelerated test and life assessment method of mathematical and rational fusion provided in the embodiment of the present application, first, based on the first performance degradation data, the detection time series and detection value series of the performance parameters of the first soft failure device are obtained, and then the preset neural network is trained based on the detection time series and detection value series to obtain a performance detection value prediction model, and then the life distribution function of the first soft failure device is determined based on the performance detection value prediction model. In this method, the data type of the test detection data of the first soft failure device is a large amount of performance degradation data, and the failure physical model of the first soft failure device is unknown, and an artificial intelligence algorithm is required to predict the failure time. Therefore, after obtaining the detection time series and detection value series of the first soft failure device, they are used as training data for model training, so as to determine the life distribution function of the first soft failure device based on the trained model. By adopting a data analysis method suitable for the detection data of the first soft failure device, the detection data of the first soft failure device is analyzed, thereby improving the accuracy of the life assessment of the first soft failure device.

[0146] Based on the above embodiment, another embodiment is provided to illustrate the process of determining the lifetime distribution function.

[0147] In an exemplary embodiment, Figure 9 As shown, when the component device is a second soft failure device, the data type includes second performance degradation data; the amount of the second performance degradation data is less than the amount of the first performance degradation data; based on the data type of each test detection data, each test detection data is analyzed to obtain the life distribution function of each component device, including:

[0148] S701: Acquire a detection value sequence of a performance parameter of a second soft-failure device according to second performance degradation data.

[0149] In the embodiment of the present application, the second soft failure device is a Class B soft failure device, and the accelerated test data analysis of the Class B soft failure device uses a failure physics model. The second performance degradation data is a small amount of performance degradation data obtained by testing the Class B soft failure device using offline testing equipment.

[0150] For n Class B soft failure devices, assuming a total of times, and the detection times are ,get The test values of the performance parameters of Class B soft failure devices are shown in Table 2 below.

[0151] Table 2

[0152]

[0153] S702 : Determine the failure time of the second soft-failure device according to the failure physical model and the detection value sequence of the second soft-failure device.

[0154] The failure physical models of different Class B soft failure devices are different. For example, the main failure mechanism of a certain board is the metal electromigration failure of the microwave power device, and its failure physical model is , where t is the failure time; J is the current density of the microwave power device of the plate, obtained through detection; is the threshold value of current density, obtained from the device manual; E is the activation energy, obtained from the device manual; is the model parameter obtained from engineering experience; b is the Boltzmann constant; W is the temperature stress value of the accelerated test.

[0155] The current density (test value) obtained from the test is brought into the above failure physical model to evaluate the failure time of the Class B soft failure device. The failure time of Class B soft failure devices is .

[0156] S703: Determine a life distribution function of the second soft failure device according to the failure time.

[0157] Based on the estimated failure time and in combination with the life assessment parameters of the second soft failure device, a life distribution function of the second soft failure device is determined.

[0158] Exemplarily, based on the number of devices and failure time of the second soft failure devices, life assessment parameters of the second soft failure devices are determined; based on the life assessment parameters, a life distribution function of the second soft failure devices is determined.

[0159] Similarly, assuming that the life distribution function of Class B soft failure equipment is exponential distribution, its life distribution function is as shown in the following formula (12).

[0160] (12)

[0161] in, is the life distribution function of Class B soft failure equipment; Life assessment parameters for Class B soft failure devices.

[0162] By solving the following equation (13), the life assessment parameters of Class B soft failure devices are evaluated: .

[0163] (13)

[0164] Thus, the life assessment parameters to be solved are Substituting into the above formula (12), the life distribution function of the second soft failure device is obtained: .

[0165] In the accelerated test and life assessment method of mathematical and rational fusion provided in the embodiment of the present application, first, based on the second performance degradation data, the detection value sequence of the performance parameters of the second soft failure device is obtained, and then the failure time of the second soft failure device is determined based on the failure physical model and the detection value sequence of the second soft failure device. Then, based on the failure time, the life distribution function of the second soft failure device is determined. In this method, the data type of the test detection data of the second soft failure device is a small amount of performance degradation data, and the failure physical model of the second soft failure device is known, so the failure physical model can be directly used to predict the failure time. Therefore, after obtaining the detection value sequence of the second soft failure device, the failure time is estimated in combination with the failure physical model of the second soft failure device, so as to determine the life distribution function of the first soft failure device based on the estimated failure time. By adopting a data analysis method suitable for the detection data of the second soft failure device, the detection data of the first soft failure device is analyzed, thereby improving the life assessment accuracy of the second soft failure device.

[0166] Based on the above embodiment, an embodiment is provided to illustrate the process of determining the life evaluation test result.

[0167] In an exemplary embodiment, Figure 10 As shown, based on the failure correlation between each component device and the life distribution function of each component device, the life assessment test results of the product to be tested are determined, including:

[0168] S801 , determining a life distribution function of a product to be tested based on the failure correlation between component devices and the life distribution function of each component device.

[0169] In the embodiment of the present application, the correlated competitive failure relationship between hard failure devices, Class A failure devices, and Class B failure devices is considered, that is, the failures between devices are competitive and correlated, thereby constructing a correlated competitive failure model of the product to be tested, that is, the life distribution function of the product to be tested, as shown in the following formula (14).

[0170] (14)

[0171] in, is the life distribution function of the product to be tested; for The correlation function of are the unknown parameters of the correlation function, which are obtained by fitting the data.

[0172] Furthermore, the life distribution function of the product to be tested is for

[0173] .

[0174] S802 , determining a life evaluation test result of the product to be tested according to an acceleration factor of the product to be tested under accelerated stress and a life distribution function of the product to be tested.

[0175] Average life of the product under test under accelerated stress for

[0176] .

[0177] Assuming that the acceleration factor of the product under test is A, the average life T of the product under normal working stress is

[0178] .

[0179] The average life of the product under normal working stress obtained above is the life evaluation test result of the product under test.

[0180] In the accelerated test and life assessment method of the mathematical and rational fusion provided in the embodiment of the present application, the life distribution function of the product to be tested is first determined based on the failure correlation between the component devices and the life distribution function of the component devices, and then the life assessment test result of the product to be tested is determined based on the acceleration factor of the product to be tested under accelerated stress and the life distribution function of the product to be tested. In this method, when performing life assessment on the product to be tested, the relevant competitive failure relationship between different component devices is considered to construct the life distribution function of the product to be tested, and then the life assessment test result of the product to be tested is obtained in combination with the acceleration factor of the product to be tested under accelerated stress. By additionally considering the competitive correlation between different component devices in the product, the accuracy of the product life assessment is improved.

[0181] In addition, in an exemplary embodiment, the accelerated test and life evaluation process in the embodiment of the present application are described.

[0182] S1. Place the components of the product into different accelerated test chambers and apply different sensitive stress types and stress values.

[0183] Hard failure devices are often sensitive to stresses such as mechanical loads, and accelerated testing stresses are often mechanical loads. Soft failure devices are often sensitive to stresses such as temperature and humidity, and accelerated testing stresses are often temperature and humidity. The different components of the devices are interconnected, and accelerated testing is performed simultaneously.

[0184] S2. Different testing methods are implemented for different equipment due to different testing data requirements.

[0185] For hard-failure devices, the test data requirement is to obtain the device's exact failure time, so real-time online testing is necessary to obtain the specific failure time. For soft-failure devices (including Class A and Class B soft-failure devices), the test data requirement is to obtain the device's performance degradation data for failure prediction, without obtaining the exact failure time. To save testing costs, offline periodic testing is sufficient. Further subdivided, for Class A soft-failure devices, the test data requirement is to obtain a large amount of performance degradation data and use intelligent algorithms to predict failure times, so shorter offline periodic testing intervals are required. For Class B soft-failure devices, the test data requirement is to obtain a small amount of performance degradation data and use failure physics models to predict failure times. To save testing costs, longer offline periodic testing intervals are sufficient.

[0186] S3. For different component equipment, different data analysis methods are used to analyze the accelerated test data to obtain the life distribution function of each component equipment.

[0187] Mathematical statistics are used to analyze the accelerated test data of hard failure devices. Artificial intelligence algorithms are used to analyze the accelerated test data of Class A soft failure devices. The physical failure model is used to analyze the accelerated test data of Class B soft failure devices.

[0188] S4. Consider the relevant competitive failure relationships among hard failure devices, Class A soft failure devices, and Class B soft failure devices, construct the product life distribution function, and combine the acceleration factor under the accelerated stress of the product to determine the average life of the product under normal working stress.

[0189] In this embodiment, a mathematical and rational fusion accelerated test and life assessment method with higher test efficiency, lower test cost and higher assessment accuracy is provided, which can quickly and accurately assess the life of the product. The component equipment of the product is divided into hard failure equipment and soft failure equipment (including Class A soft failure equipment and Class B soft failure equipment) according to the failure type. Different equipment types are placed in different accelerated test chambers and different acceleration stresses are applied to achieve the maximum acceleration effect of the product accelerated test and improve the efficiency of the accelerated test. Different detection methods are implemented to reduce the cost of accelerated testing in view of the different test data requirements of different equipment. Different evaluation methods are adopted according to the characteristics of the test data of different equipment, and then a comprehensive evaluation is carried out to make the test data analysis more targeted and improve the accuracy of life assessment. Considering the competitiveness and correlation between the failures of hard failure equipment, Class A soft failure equipment and Class B soft failure equipment, a related competitive failure model of the product is constructed, and then the average life of the product is evaluated to improve the accuracy of life assessment.

[0190] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence 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 executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0191] Based on the same inventive concept, embodiments of the present application also provide a device for implementing the aforementioned method for accelerating testing and life assessment using mathematical and rational 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 accelerating testing and life assessment using mathematical and rational fusion can be found in the aforementioned definition of the accelerated testing and life assessment method, and will not be further elaborated here.

[0192] In an exemplary embodiment, Figure 11 As shown, a mathematical and rational fusion accelerated testing and life assessment device 1 is provided, comprising: a data acquisition module 10, a data analysis module 20 and a result determination module 30, wherein:

[0193] The data acquisition module 10 is used to obtain test data of each component device according to the failure type of each component device when different components of the product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers;

[0194] The data analysis module 20 is used to analyze the test data according to the data type of each test data to obtain the life distribution function of each component device;

[0195] The result determination module 30 is used to determine the life evaluation test result of the product to be tested based on the failure correlation between the component devices and the life distribution function of the component devices.

[0196] In one embodiment, the data acquisition module 10 is further configured to:

[0197] According to the failure type of each component equipment, the test data acquisition requirements of each component equipment are obtained; according to each test data acquisition requirement, in the mapping relationship between the test data acquisition requirements and the test detection method, the test detection method that matches the each test data acquisition requirement is obtained; using the test detection method that matches the each test data acquisition requirement, each component equipment is tested to obtain the test detection data of each component equipment.

[0198] In one embodiment, the data analysis module 20 is further configured to:

[0199] According to the number of failed devices in the hard failure devices, the failure time data of each failed device, and the mathematical statistics model of the hard failure devices, the failure likelihood function of the hard failure devices is determined; according to the failure likelihood function, the life distribution function of the hard failure devices is determined.

[0200] In one embodiment, the data analysis module 20 is further configured to:

[0201] Based on the first performance degradation data, a detection time series and a detection value series of the performance parameters of the first soft failure device are obtained; based on the detection time series and the detection value series, a preset neural network is trained to obtain a performance detection value prediction model; based on the performance detection value prediction model, a life distribution function of the first soft failure device is determined.

[0202] In one embodiment, the data analysis module 20 is further configured to:

[0203] Determining the life distribution function of the first soft failure device based on the performance test value prediction model includes:

[0204] The candidate time series is input into the performance test value prediction model to obtain the test value prediction sequence output by the performance test value prediction model; the failure time of the first soft failure device is determined based on the candidate time series and the test value prediction sequence; and the life distribution function of the first soft failure device is determined based on the failure time.

[0205] In one embodiment, the data analysis module 20 is further configured to:

[0206] Determine the life evaluation parameters of the first soft failure devices according to the device quantity and failure time of the first soft failure devices; and determine the life distribution function of the first soft failure devices according to the life evaluation parameters.

[0207] In one embodiment, the data analysis module 20 is further configured to:

[0208] A detection value sequence of a performance parameter of a second soft failure device is obtained based on the second performance degradation data; a failure time of the second soft failure device is determined based on a failure physical model and the detection value sequence of the second soft failure device; and a life distribution function of the second soft failure device is determined based on the failure time.

[0209] In one embodiment, the data analysis module 20 is further configured to:

[0210] According to the number of devices and failure time of the second soft failure devices, life assessment parameters of the second soft failure devices are determined; and according to the life assessment parameters, a life distribution function of the second soft failure devices is determined.

[0211] In one embodiment, the result determination module 30 is further configured to:

[0212] The life distribution function of the product to be tested is determined based on the failure correlation between the component devices and the life distribution function of each component device; the life assessment test results of the product to be tested are determined based on the acceleration factor of the product to be tested under accelerated stress and the life distribution function of the product to be tested.

[0213] Each module in the aforementioned mathematically-intelligible fusion accelerated testing and life assessment 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 within a computer device in hardware form, or stored in a computer device memory in software form, allowing the processor to call and execute the corresponding operations of each module.

[0214] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0215] When different components of the product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers, test data of each component device is obtained according to the failure type of each component device;

[0216] According to the data type of each test data, the test data are analyzed to obtain the life distribution function of each component equipment;

[0217] The life evaluation test results of the product to be tested are determined based on the failure correlation between the component devices and the life distribution function of the component devices.

[0218] The implementation principles and technical effects of each step implemented by the processor in the embodiment of the present application are similar to the principles of the above-mentioned accelerated test and life assessment method of mathematical and rational fusion, and will not be repeated here.

[0219] 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 following steps are implemented:

[0220] When different components of the product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers, test data of each component device is obtained according to the failure type of each component device;

[0221] According to the data type of each test data, the test data are analyzed to obtain the life distribution function of each component equipment;

[0222] The life evaluation test results of the product to be tested are determined based on the failure correlation between the component devices and the life distribution function of the component devices.

[0223] The implementation principles and technical effects of the various steps implemented when the computer program in the embodiment of the present application is executed by the processor are similar to the principles of the above-mentioned accelerated test and life assessment method of mathematical and rational fusion, and will not be repeated here.

[0224] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0225] When different components of the product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers, test data of each component device is obtained according to the failure type of each component device;

[0226] According to the data type of each test data, the test data are analyzed to obtain the life distribution function of each component equipment;

[0227] The life evaluation test results of the product to be tested are determined based on the failure correlation between the component devices and the life distribution function of the component devices.

[0228] The implementation principles and technical effects of the various steps implemented when the computer program in the embodiment of the present application is executed by the processor are similar to the principles of the above-mentioned accelerated test and life assessment method of mathematical and rational fusion, and will not be repeated here.

[0229] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0230] Those skilled in the art will appreciate 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 above-mentioned embodiments. 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 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, distributed databases based on blockchains. 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), data processing logic devices based on quantum computing, and the like.

[0231] The various technical types in the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the various technical types in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical types, they should be considered to be within the scope of this specification.

[0232] 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 mathematical and rational fusion accelerated testing and life assessment method, characterized in that: The method comprises: When different components of a product to be tested are placed in different accelerated test chambers, and different accelerated stresses are applied to the different accelerated test chambers, test detection data of each component device is obtained according to the failure type of each component device; the detection method of each test detection data is matched with the corresponding failure type; the failure types include hard failure devices, first soft failure devices, and second soft failure devices; the failure physical model of the first soft failure device is unknown; and the failure physical model of the second soft failure device is known; According to the data type of each of the test detection data, data analysis is performed on each of the test detection data to obtain the life distribution function of each of the component devices; when the component device is a hard failure device, the data type includes failure time data; the accelerated test data analysis of the hard failure device adopts a mathematical statistics method; when the component device is a first soft failure device, the data type includes first performance degradation data; the accelerated test data analysis of the first soft failure device adopts an artificial intelligence algorithm; when the component device is a second soft failure device, the data type includes second performance degradation data; the amount of the second performance degradation data is less than the amount of the first performance degradation data; the accelerated test data analysis of the second soft failure device adopts a failure physics model; The life evaluation test result of the product to be tested is determined based on the failure correlation between the component devices and the life distribution function of the component devices.

2. The method according to claim 1, characterized in that The step of obtaining test data of each component device according to the failure type of each component device includes: Obtaining detection data acquisition requirements for each of the component devices according to the failure type of each of the component devices; According to each of the test data acquisition requirements, in the mapping relationship between the test data acquisition requirements and the test detection methods, obtaining a test detection method that matches each of the test data acquisition requirements; Each of the component devices is tested using a test detection method that matches the test data acquisition requirements to obtain the test detection data of each of the component devices.

3. The method according to claim 1 or 2, characterized in that When the component device is a hard failure device, the data type includes failure time data; and performing data analysis on each of the test and detection data according to the data type of each of the test and detection data to obtain the life distribution function of each of the component devices includes: Determining a failure likelihood function of the hard failure device according to the number of failed devices in the hard failure device, failure time data of each of the failed devices, and a mathematical statistical model of the hard failure device; A life distribution function of the hard failure device is determined according to the failure likelihood function.

4. The method according to claim 1 or 2, characterized in that When the component device is a first soft failure device, the data type includes first performance degradation data; and performing data analysis on each of the test and detection data according to the data type of each of the test and detection data to obtain the life distribution function of each of the component devices includes: acquiring, according to the first performance degradation data, a detection time series and a detection value series of a performance parameter of the first soft-failure device; Training a preset neural network according to the detection time series and the detection value series to obtain a performance detection value prediction model; Determine the life distribution function of the first soft failure device according to the performance detection value prediction model.

5. The method according to claim 4, characterized in that Determining the life distribution function of the first soft failure device according to the performance detection value prediction model includes: Inputting the candidate time series into the performance detection value prediction model to obtain the detection value prediction sequence output by the performance detection value prediction model; determining a failure time of the first soft-failure device according to the candidate time sequence and the detection value prediction sequence; A life distribution function of the first soft-failure device is determined according to the failure time.

6. The method according to claim 5, characterized in that Determining the life distribution function of the first soft-failure device according to the failure time includes: Determining a life assessment parameter of the first soft-failure device according to the number of devices of the first soft-failure device and the failure time; A life distribution function of the first soft-failure device is determined according to the life assessment parameter.

7. The method according to claim 1 or 2, characterized in that When the component device is a second soft failure device, the data type includes second performance degradation data; the amount of the second performance degradation data is less than the amount of the first performance degradation data; and performing data analysis on each of the test detection data according to the data type of each of the test detection data to obtain the life distribution function of each of the component devices includes: acquiring, according to the second performance degradation data, a sequence of detection values of a performance parameter of the second soft-failure device; determining a failure time of the second soft-failure device according to a failure physical model of the second soft-failure device and the detection value sequence; A life distribution function of the second soft-failure device is determined according to the failure time.

8. The method according to claim 7, characterized in that Determining the life distribution function of the second soft-failure device according to the failure time includes: determining a life assessment parameter of the second soft-failure device according to the number of devices of the second soft-failure device and the failure time; A life distribution function of the second soft-failure device is determined according to the life assessment parameter.

9. The method according to claim 1 or 2, characterized in that Determining the life evaluation test result of the product to be tested based on the failure correlation between the component devices and the life distribution function of the component devices includes: Determining the life distribution function of the product to be tested based on the failure correlation between the component devices and the life distribution function of the component devices; A life evaluation test result of the product to be tested is determined according to the acceleration factor of the product to be tested under accelerated stress and the life distribution function of the product to be tested.

10. A mathematical and rational fusion accelerated testing and life assessment device, characterized in that: The device comprises: A data acquisition module is configured to acquire test detection data for each component device according to a failure type of each component device, when different components of a product to be tested are placed in different accelerated test chambers, each of which is subjected to different accelerated stresses. The detection method of each test detection data is matched to the corresponding failure type. The failure types include hard failure devices, first soft failure devices, and second soft failure devices. The failure physical model of the first soft failure device is unknown, while the failure physical model of the second soft failure device is known. a data analysis module, configured to perform data analysis on each of the test and inspection data according to a data type of each of the test and inspection data to obtain a life distribution function of each of the component devices; when the component device is a hard failure device, the data type includes failure time data; the accelerated test data analysis of the hard failure device adopts a mathematical statistics method; when the component device is a first soft failure device, the data type includes first performance degradation data; the accelerated test data analysis of the first soft failure device adopts an artificial intelligence algorithm; when the component device is a second soft failure device, the data type includes second performance degradation data; the amount of the second performance degradation data is less than the amount of the first performance degradation data; the accelerated test data analysis of the second soft failure device adopts a failure physics model; The result determination module is used to determine the life evaluation test result of the product to be tested according to the failure correlation between the component devices and the life distribution function of the component devices.

11. 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 9 are implemented.

12. 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 9 are implemented.

13. 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 9 are implemented.

Images