Method and system for determining limit stress information of reliability enhancement test

Abstract

The invention relates to the technical field of product reliability engineering. The invention provides a method and a system for determining limit stress information of a reliability enhancement test. The method is realized by the following steps: acquiring rated working stress and maximum allowable stress of a product, and calculating an initial stress increment; applying stress step by step by taking the initial stress increment as a stepping value, resetting the initial stress increment by adopting a dichotomy when a fault is excited for the first time, and performing a working limit test again to obtain a working limit upper limit; on the basis that the stress is excited by the first fault, the stress is improved to excite the fault under the stress, and the stress is traced back to the upper limit of the working limit to carry out the failure limit test. The problems that in the prior art, deep mining of test data and a multi-stress coupling effect systematic solution are lacked, and parameter optimization and a real-time feedback mechanism under the multi-stress condition are not involved are solved.

Description

Technical Field

[0001] This invention relates to the field of product reliability engineering technology, and more specifically, to a method and system for determining the ultimate stress information in reliability enhancement tests. Background Technology

[0002] In the field of product reliability engineering, reliability enhancement testing (RET) is a key technology for evaluating the ultimate performance of products under extreme stress conditions. This technology rapidly exposes potential defects in product design and manufacturing by applying progressively increasing environmental or operational stresses (such as temperature, vibration, electrical stress, etc.), thereby determining its ultimate stress threshold and optimizing reliability.

[0003] Currently, patented technologies for determining ultimate stress (such as the HALT / HASS test method) mostly focus on the design of stress application procedures, lacking systematic solutions for in-depth mining of test data and multi-stress coupling effects. For example, although the U.S. Department of Defense standard MIL-STD-883E specifies the basic method of Limittest, it does not cover parameter optimization and real-time feedback mechanisms under multi-stress conditions. Summary of the Invention

[0004] The purpose of this invention is to provide a method for determining the ultimate stress information in reliability enhancement tests, aiming to solve the problems in existing reliability enhancement test ultimate stress determination technologies, such as the lack of in-depth data mining, systematic solutions for multi-stress coupling effects, and the absence of parameter optimization and real-time feedback mechanisms under multi-stress conditions.

[0005] This invention is achieved through the following technical solution: A method for determining the ultimate stress information in a reliability enhancement test includes the following steps: Obtain the rated working stress of the product and maximum allowable stress Calculate the initial stress increment ; From rated working stress Begin with initial stress increment Stress is applied incrementally in steps, and when the fault is first triggered, the initial stress increment is calculated using a bisection method. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress Then, the working limit test was repeated to obtain the upper limit of the working limit. ;in, This indicates the number of backtracking steps in the working limit test. ; stress triggered by the initial fault Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. Conduct destructive limit tests; among them, Indicates the number of destructive limit tests; This represents the stress increment traced back from the last working limit test.

[0006] Optionally, the step of obtaining the rated working stress of the product... and maximum allowable stress Calculate the initial stress increment The specific process is as follows: Determine the rated working stress of the product by referring to historical data of similar products, expert knowledge, or design specifications. and maximum allowable stress ; The initial stress increment is calculated using the following formula (1). :

[0007] in, Indicates the preset stress level number; generates an equal-increment stress sequence. ,and .

[0008] Optionally, the initial stress increment is calculated using the bisection method. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress Then, the working limit test was repeated to obtain the upper limit of the working limit. The specific process is as follows: When the fault is first triggered, take The initial stress increment Reset to Tracing back to the stress triggered by the initial fault The first stage of stress ; If the failure persists after backtracking, continue with... To reduce stress, backtracking is performed each time. Until function is restored, the stress is increased incrementally using the stress increment from the last backtracking until the fault recurs, with the stress at the time of function restoration serving as the upper limit of the working limit. ; If the functionality is restored after rollback, then continue with... Increment the stress until the fault recurs, with each increment... Revert to the previous stress level and verify functional recovery, using the previous stress level at the time of fault reproduction as the upper limit of the operating limit. .

[0009] Optionally, the stress during the initial fault triggering... Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. The specific process of conducting a destructive limit test is as follows: initialization Increase the stress to And conduct experiments; If the product is operating normally, the stress is increased to trigger a failure again, and the stress is then pushed back to the upper limit of the operating limit. Continue to conduct destructive limit tests; If the product still fails, reduce the stress level to the minimum stress value. If it still fails, determine the upper limit stress of failure. , recorded as .

[0010] Optionally, the working limit test employs a multi-stress coupling analysis engine, specifically including: Real-time acquisition of multi-dimensional sensor data on temperature, vibration, and electrical stress is used to construct a time-varying stress matrix. Identify stress coupling feature vectors using convolutional neural networks; When a fault occurs, the gradient backpropagation algorithm is used to calculate the contribution weight of each stress component to the fault. ; Based on contribution weight Adjust the subsequent stress application strategy: If the weight difference between any two stress components Then, a bisection backtracking method is used for the high-weight stress components, while the low-weight components maintain a linear increasing trend; where... and These represent the weights corresponding to any two stress components in the contribution weights; This indicates the preset threshold.

[0011] Optionally, the destructive limit test phase integrates failure prediction and root cause analysis, specifically including: Under stress When a fault is triggered, internal status data of the product is collected simultaneously. The product's internal status data; Including voltage waveform and temperature field distribution; Product internal status data Input a pre-trained fault knowledge graph, output fault mode confidence. and a set of potential design flaws; like This activates the dynamic stress backtracking mechanism: Immediately reduce stress to ; Reconstruct the stress application sequence based on the set of potential design defects, and strengthen the defect-related stress components; in, Indicates the attenuation coefficient; This indicates a pre-set confidence threshold.

[0012] Optionally, the specific process of reconstructing the stress application sequence based on the set of potential design defects and strengthening the defect-related stress components is as follows: Analyze the set of potential design flaws and identify the types of key design flaws; Based on a predefined defect-stress mapping relationship, target stress components associated with key design defect types are determined; wherein, the defect-stress mapping relationship includes: thermally related defects mapped to temperature stress, mechanically related defects mapped to vibration stress, and electrically related defects mapped to electrical stress. Calculate the stress increment after strengthening for the target stress components. ;in, Indicates the strengthening coefficient. ; This represents the stress increment retrospectively from the last working limit test; For non-target stress components, the original stress increment remains unchanged; Based on the stress increment after strengthening Reconstruct the multi-stress application sequence using the original stress increment; Apply the reconstructed stress application sequence, starting from the stress level set by the dynamic stress backtracking mechanism. Begin, continue with the destructive limit test. Based on the same inventive concept, the present invention also provides a reliability enhancement test limit stress information determination system, used to implement the aforementioned reliability enhancement test limit stress information determination method, comprising: The stress loading control module is used to obtain the rated working stress of the product. and maximum allowable stress Calculate the initial stress increment Generate an equal-increment stress sequence and start from the rated working stress. Start with initial stress increment Apply stress step by step to the increment value; The working limit test module, connected to the stress loading control module, is used to calculate the initial stress increment using a binary method upon first triggering of a fault. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress And re-conduct the working limit test, by adjusting Value until the upper limit of the working limit is determined. ; The destructive limit test module, connected to the working limit test module, is used to test the initial failure excitation stress. Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. Conduct destructive limit tests, by adjusting The value determines the upper limit stress of the failure limit. ; The multi-stress coupling analysis module, connected to the destructive limit test module, is used to collect multi-dimensional sensor data of temperature, vibration, and electrical stress in real time during the working limit test, construct a time-varying stress matrix, identify stress coupling feature vectors through a convolutional neural network, calculate the contribution weight of each stress component to the fault when a fault occurs using a gradient backpropagation algorithm, and adjust the subsequent stress application strategy based on the contribution weight. The fault prediction and root cause analysis module, connected to the multi-stress coupling analysis module, is used to perform fault prediction and root cause analysis during the failure limit test phase, under stress... Synchronously collect internal status data of the product when a fault is triggered. Product internal status data Input a pre-trained fault knowledge graph, output fault mode confidence. and a set of potential design flaws, if If this is not the case, the dynamic stress backtracking mechanism is activated, and the stress application sequence is reconstructed based on the set of potential design defects, and the failure limit test is continued.

[0013] Based on the same inventive concept, the present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the above-described method for determining the ultimate stress information of the reliability enhancement test.

[0014] Based on the same inventive concept, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-described method for determining the ultimate stress information of reliability enhancement tests.

[0015] The technical solution of the present invention has at least the following advantages and beneficial effects: Initially, step loading is used to quickly approximate the first failure point. Then, a binary backtracking mechanism is introduced, which can significantly narrow the search range. Compared with a single incremental or fixed step size method, this combined strategy can locate the upper limit of the working limit more accurately with fewer tests and more refined stress step size, avoiding misjudgments caused by overstress or understress, and greatly improving testing efficiency and data accuracy.

[0016] By clearly distinguishing and sequentially performing working limit tests and destructive limit tests, especially destructive limit tests which begin by applying higher stresses from the precisely determined upper limit of the working limit and ultimately back to the upper limit of the working limit for verification or exploration of proximity points, the structured hierarchical testing process systematically reveals the complete failure boundary of a product from functional failure to physical damage, providing more comprehensive data support for design margin assessment.

[0017] By using the fine stress increment obtained from the last working limit test as the reference step size for the destructive limit test, the stress increment is intelligently and dynamically adjusted based on the results of previous tests. This effectively adapts to the changes in the product's response characteristics under different stress levels. It is particularly beneficial for handling complex multi-stress coupling scenarios, ensuring the accuracy and controllability of stress application in high-risk areas, and avoiding excessive stress jumps that could lead to the omission of critical failure modes or instantaneous excessive damage to the product.

[0018] The operational limit test phase involves refined binary backtesting near the initial failure point, significantly enhancing the ability to expose potential design flaws and process weaknesses in the critical stress region; the destructive limit test starts from the upper limit of the operational limit, effectively exposing deeper defects that may cause permanent damage; the entire process provides a high-confidence data foundation for quantifying the actual operational and destructive margins of the product.

[0019] Through its unique backtracking mechanism, hierarchical test design, and dynamic stress stepping, it effectively compensates for the lack of in-depth data mining and multi-stress parameter optimization guidance in extreme tests, as exemplified by standards such as MIL-STD-883E. The inherent feedback loop (adjusting subsequent step sizes and starting points based on previous results) provides a real-time feedback mechanism, making the test process more adaptable and intelligent, and increasing its practical engineering value. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the method for determining the ultimate stress information in a reliability enhancement test according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the system for determining the ultimate stress information in reliability enhancement testing according to an embodiment of the present invention. Figure 3 This is a schematic diagram of a typical existing reliability enhancement test profile (the stress increment remains consistent throughout the entire reliability enhancement test). Figure 4 This is a schematic cross-sectional view of the reliability enhancement test corresponding to the failure after backtracking in the reliability enhancement test limit stress information determination method of the present invention. Figure 5 This is a schematic cross-sectional view of the reliability enhancement test corresponding to the function recovery after backtracking in the reliability enhancement test limit stress information determination method of the present invention embodiment; Figure 6 This is a cross-sectional schematic diagram of the reliability enhancement test, which determines the upper limit of the working limit corresponding to the failure after backtracking, in the reliability enhancement test limit stress information determination method of this embodiment of the invention. Figure 7 This is a cross-sectional schematic diagram of the reliability enhancement test in the reliability enhancement test limit stress information determination method of the present invention, which determines the upper limit of the working limit corresponding to the function recovery after backtracking. Figure 8 This is a schematic diagram of the pre-trained fault knowledge graph structure constructed in the reliability enhancement test limit stress information determination method of this invention. Detailed Implementation

[0021] The following is a detailed description of the embodiments, in conjunction with the accompanying drawings.

[0022] Reference Figure 1 A method for determining the ultimate stress information in a reliability enhancement test includes the following steps: Step 1: Obtain the product's rated working stress and maximum allowable stress Calculate the initial stress increment .

[0023] In some embodiments, the method of obtaining the rated working stress of the product and maximum allowable stress Calculate the initial stress increment The specific process is as follows: Determine the rated working stress of the product by referring to historical data of similar products, expert knowledge, or design specifications. and maximum allowable stress ; The initial stress increment is calculated using the following formula (1). :

[0024] in, Indicates the preset stress level number; generates an equal-increment stress sequence. ,and .

[0025] Step 2: From the rated working stress Begin with initial stress increment Stress is applied incrementally in steps, and when the fault is first triggered, the initial stress increment is calculated using a bisection method. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress Then, the working limit test was repeated to obtain the upper limit of the working limit. ;in, This indicates the number of backtracking steps in the working limit test. .

[0026] In some embodiments, the initial stress increment is calculated using the bisection method. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress Then, the working limit test was repeated to obtain the upper limit of the working limit. The specific process is as follows: When the fault is first triggered, take The initial stress increment Reset to Tracing back to the stress triggered by the initial fault The first stage of stress ; Reference Figure 4 , Figure 6 If the failure persists after backtracking, then continue with... To reduce stress, backtracking is performed each time. Until function is restored, the stress is increased incrementally using the stress increment from the last backtracking until the fault recurs, with the stress at the time of function restoration serving as the upper limit of the working limit. ; Reference Figure 5 , Figure 7 If the function is restored after rollback, then continue with Increment the stress until the fault recurs, with each increment... Revert to the previous stress level and verify functional recovery, using the previous stress level at the time of fault reproduction as the upper limit of the operating limit. .

[0027] Step 3: Stress triggered by the initial fault Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. Conduct destructive limit tests; among them, Indicates the number of destructive limit tests; This represents the stress increment traced back from the last working limit test.

[0028] In some embodiments, the initial fault-induced stress Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. The specific process of conducting a destructive limit test is as follows: initialization Increase the stress to And conduct experiments; If the product is operating normally, the stress is increased to trigger a failure again, and the stress is then pushed back to the upper limit of the operating limit. Continue to conduct destructive limit tests; If the product still fails, reduce the stress level to the minimum stress value. If it still fails, determine the upper limit stress of failure. , recorded as .

[0029] In some embodiments, the working limit test employs a multi-stress coupling analysis engine, specifically including: Real-time acquisition of multi-dimensional sensor data on temperature, vibration, and electrical stress is used to construct a time-varying stress matrix. Temperature sensors (e.g., thermocouples), vibration sensors (accelerometers), and electrical stress sensors (voltage / current probes) are deployed at key locations on the test sample to synchronously acquire raw data at a sampling rate of ≥1kHz. The time-series data is sliced ​​according to fixed time windows (e.g., 10ms) and organized into a three-dimensional matrix. ,in, , and These are stress components measured simultaneously for temperature, vibration, and electrical stress, respectively.

[0030] Stress coupling feature vectors are identified using a convolutional neural network; the time-varying stress matrix is ​​normalized and then input into the CNN model, with the following structure: First convolutional layer: 32 1×5 convolutional kernels, extracting temporal features of single stress components (ReLU activation); The second convolutional layer consists of 64 3×3 convolutional kernels, which extract cross-stress coupling features (ReLU activation). Fully connected layer: outputs a 128-dimensional feature vector to characterize the multi-stress interaction effect.

[0031] Training data: The CNN was trained using historical test failure data, and the stress coupling patterns at the time of failure were labeled (such as "high temperature + high frequency vibration").

[0032] When a fault occurs, the gradient backpropagation algorithm is used to calculate the contribution weight of each stress component to the fault. When a fault trigger signal (such as functional abnormality) is detected, the CNN weights are frozen; a contribution weight calculation layer (linear layer + Softmax) is added to the end of the CNN; the fault signal gradient is backpropagated, and the gradient sensitivity of each stress component is calculated, as shown in the following equation (2):

[0033] in, This represents the fault determination loss function; Represents the input stress components The gradient; Output normalized weights ,satisfy ; where , and represent the weights corresponding to the stress components measured simultaneously for temperature, vibration, and electrical stress, respectively.

[0034] Based on contribution weight Adjust the subsequent stress application strategy; if the weight difference between any two stress components Then, a bisection backtracking method is used for the high-weight stress components, while the low-weight components maintain a linear increasing trend; where... and These represent the weights corresponding to any two stress components in the contribution weights; Indicates the preset threshold (e.g.) Example: If temperature weight Vibration weight , Then, the temperature stress is backtracked using the dichotomy method, and the vibration stress is incremented step by step according to the original increment.

[0035] In some embodiments, the destructive limit test phase integrates failure prediction and root cause analysis, specifically including: Under stress When a fault is triggered, internal status data of the product is collected simultaneously. The product's internal status data; This includes voltage waveforms and temperature field distribution; it can capture waveforms of key circuit nodes using a high-precision oscilloscope (sampling rate ≥10MHz) to identify abnormal fluctuations (such as voltage drops / glitches); it can use an infrared thermal imager or an embedded temperature sensor array (resolution ±0.5℃) to generate a two-dimensional temperature cloud map and locate hotspot areas.

[0036] Product internal status data Input a pre-trained fault knowledge graph, such as Figure 8 As shown, the output fault mode confidence level And a set of potential design flaws. The pre-trained fault knowledge graph can be built based on historical fault databases (such as nuclear power valves and aerospace relays) to construct a network linking "defect modes - stress characteristics - internal states"; fault mode confidence. The range is This reflects the degree of matching between the current fault and the map node (e.g., If the result is negative, it is considered a high confidence level; output the defect type and related parameters (e.g., "Heat dissipation defect: thermal resistance exceeds the standard by 37%").

[0037] like (Default settings) If the value is 0.8, then the dynamic stress backtracking mechanism is activated: Immediately reduce stress to The stress is about to be As a safety stress; The stress application sequence is reconstructed based on a set of potential design defects, and the defect-related stress components are strengthened; among them... Indicates the attenuation coefficient (default setting) ); This indicates a pre-set confidence threshold.

[0038] In some embodiments, the specific process of reconstructing the stress application sequence based on the set of potential design defects and strengthening the defect-related stress components is as follows: Analyze the set of potential design flaws and identify the types of key design flaws; Based on a predefined defect-stress mapping relationship, target stress components associated with key design defect types are determined; wherein, the defect-stress mapping relationship includes: thermally related defects mapped to temperature stress, mechanically related defects mapped to vibration stress, and electrically related defects mapped to electrical stress. Calculate the stress increment after strengthening for the target stress components. ;in, Indicates the strengthening coefficient. ; This represents the stress increment retrospectively from the last working limit test; For non-target stress components, the original stress increment remains unchanged; Based on the stress increment after strengthening Reconstruct the multi-stress application sequence using the original stress increment; Apply the reconstructed stress application sequence, starting from the stress level set by the dynamic stress backtracking mechanism. Begin, continue with the destructive limit test.

[0039] Based on the same inventive concept, corresponding to any of the above embodiments, the present invention also provides a reliability enhancement test limit stress information determination system, used to implement the reliability enhancement test limit stress information determination method, including: The stress loading control module is used to obtain the rated working stress of the product. and maximum allowable stress Calculate the initial stress increment Generate an equal-increment stress sequence and start from the rated working stress. Start with initial stress increment Apply stress step by step to the increment value; The working limit test module, connected to the stress loading control module, is used to calculate the initial stress increment using a binary method upon first triggering of a fault. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress And re-conduct the working limit test, by adjusting Value until the upper limit of the working limit is determined. ; The destructive limit test module, connected to the working limit test module, is used to test the initial failure excitation stress. Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. Conduct destructive limit tests, by adjusting The value determines the upper limit stress of the failure limit. ; The multi-stress coupling analysis module, connected to the destructive limit test module, is used to collect multi-dimensional sensor data of temperature, vibration, and electrical stress in real time during the working limit test, construct a time-varying stress matrix, identify stress coupling feature vectors through a convolutional neural network, calculate the contribution weight of each stress component to the fault when a fault occurs using a gradient backpropagation algorithm, and adjust the subsequent stress application strategy based on the contribution weight. The fault prediction and root cause analysis module, connected to the multi-stress coupling analysis module, is used to perform fault prediction and root cause analysis during the failure limit test phase, under stress... Synchronously collect internal status data of the product when a fault is triggered. Product internal status data Input a pre-trained fault knowledge graph, output fault mode confidence. and a set of potential design flaws, if If this is not the case, the dynamic stress backtracking mechanism is activated, and the stress application sequence is reconstructed based on the set of potential design defects, and the failure limit test is continued.

[0040] Based on the same inventive concept, corresponding to any of the above embodiments, the present invention provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform a reliability enhancement test limit stress information determination method of the embodiment.

[0041] Alternatively, the aforementioned electronic device may be a server.

[0042] In addition, this embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method for determining the ultimate stress information of the reliability enhancement test of the embodiment.

[0043] It is understood that the processor in the embodiments of the present invention can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor.

[0044] The method steps in the embodiments of the present invention can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an ASIC.

[0045] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a storage medium or transmitted through a storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

Claims

1. A method for determining the ultimate stress information in a reliability enhancement test, characterized in that, Includes the following steps: Obtain the rated working stress of the product and maximum allowable stress Calculate the initial stress increment ; From rated working stress Begin with initial stress increment Stress is applied incrementally in steps, and when the fault is first triggered, the initial stress increment is calculated using a bisection method. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress Then, the working limit test was repeated to obtain the upper limit of the working limit. ;in, This indicates the number of backtracking steps in the working limit test. ; stress triggered by the initial fault Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. Conduct destructive limit tests; among them, Indicates the number of destructive limit tests; This represents the stress increment traced back from the last working limit test.

2. The method for determining the ultimate stress information in reliability enhancement tests as described in claim 1, characterized in that, The rated working stress of the product is obtained. and maximum allowable stress Calculate the initial stress increment The specific process is as follows: Determine the rated working stress of the product by referring to historical data of similar products, expert knowledge, or design specifications. and maximum allowable stress ; The initial stress increment is calculated using the following formula (1). : in, Indicates the preset stress level number; generates an equal-increment stress sequence. ,and .

3. The method for determining the ultimate stress information in reliability enhancement tests as described in claim 1, characterized in that, The initial stress increment is calculated using a bisection method. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress Then, the working limit test was repeated to obtain the upper limit of the working limit. The specific process is as follows: When the fault is first triggered, take The initial stress increment Reset to Tracing back to the stress triggered by the initial fault The first stage of stress ; If the failure persists after backtracking, continue with... To reduce stress, backtracking is performed each time. Until function is restored, the stress is increased incrementally using the stress increment from the last backtracking until the fault recurs, with the stress at the time of function restoration serving as the upper limit of the working limit. ; If the functionality is restored after rollback, then continue with... Increment the stress until the fault recurs, with each increment... Revert to the previous stress level and verify functional recovery, using the previous stress level at the time of fault reproduction as the upper limit of the operating limit. .

4. The method for determining the ultimate stress information in reliability enhancement tests as described in claim 1, characterized in that, The stress triggered by the initial fault Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. The specific process of conducting a destructive limit test is as follows: initialization Increase the stress to And conduct experiments; If the product is operating normally, the stress is increased to trigger a failure again, and the stress is then pushed back to the upper limit of the operating limit. Continue to conduct destructive limit tests; If the product still fails, reduce the stress level to the minimum stress value. If it still fails, determine the upper limit stress of failure. , recorded as .

5. The method for determining the ultimate stress information in reliability enhancement tests as described in claim 1, characterized in that, The working limit test employs a multi-stress coupling analysis engine, specifically including: Real-time acquisition of multi-dimensional sensor data on temperature, vibration, and electrical stress is used to construct a time-varying stress matrix. Identify stress coupling feature vectors using convolutional neural networks; When a fault occurs, the gradient backpropagation algorithm is used to calculate the contribution weight of each stress component to the fault. Adjusting subsequent stress application strategy based on contribution weight: If the weight difference between any two stress components Then, a bisection backtracking method is used for the high-weight stress components, while the low-weight components maintain a linear increasing trend; where... and These represent the weights corresponding to any two stress components in the contribution weights; This indicates the preset threshold.

6. The method for determining the ultimate stress information in reliability enhancement tests as described in claim 4, characterized in that, The destructive limit test phase integrates failure prediction and root cause analysis, specifically including: Under stress When a fault is triggered, internal status data of the product is collected simultaneously. The product's internal status data; Including voltage waveform and temperature field distribution; Product internal status data Input a pre-trained fault knowledge graph, output fault mode confidence. and a set of potential design flaws; like This activates the dynamic stress backtracking mechanism: Immediately reduce stress to ; Reconstruct the stress application sequence based on the set of potential design defects, and strengthen the defect-related stress components; in, Indicates the attenuation coefficient; This indicates a pre-set confidence threshold.

7. The method for determining the ultimate stress information in reliability enhancement tests as described in claim 6, characterized in that, The specific process of reconstructing the stress application sequence based on the set of potential design defects and strengthening the defect-related stress components is as follows: Analyze the set of potential design flaws and identify the types of key design flaws; Based on a predefined defect-stress mapping relationship, target stress components associated with key design defect types are determined; wherein, the defect-stress mapping relationship includes: thermally related defects mapped to temperature stress, mechanically related defects mapped to vibration stress, and electrically related defects mapped to electrical stress. Calculate the stress increment after strengthening for the target stress components. ;in, Indicates the strengthening coefficient. ; This represents the stress increment retrospectively from the last working limit test; For non-target stress components, the original stress increment remains unchanged; Based on the stress increment after strengthening Reconstruct the multi-stress application sequence using the original stress increment; Apply the reconstructed stress application sequence, starting from the stress level set by the dynamic stress backtracking mechanism. Begin, continue with the destructive limit test.

8. A system for determining the ultimate stress information in a reliability enhancement test, used to implement the method for determining the ultimate stress information in a reliability enhancement test as described in any one of claims 1-7, characterized in that, include: The stress loading control module is used to obtain the rated working stress of the product. and maximum allowable stress Calculate the initial stress increment Generate an equal-increment stress sequence and start from the rated working stress. Start with initial stress increment Apply stress step by step to the increment value; The working limit test module, connected to the stress loading control module, is used to calculate the initial stress increment using a binary method upon first triggering of a fault. Reset to Tracing back to the stress triggered by the initial fault The first stage of stress And re-conduct the working limit test, by adjusting Value until the upper limit of the working limit is determined. ; The destructive limit test module, connected to the working limit test module, is used to test the initial failure excitation stress. Based on this, the stress will be increased to Under stress After a fault is triggered, the stress is pushed back to the upper limit of the working limit. Conduct destructive limit tests, by adjusting The value determines the upper limit stress of the failure limit. ; The multi-stress coupling analysis module, connected to the destructive limit test module, is used to collect multi-dimensional sensor data of temperature, vibration, and electrical stress in real time during the working limit test, construct a time-varying stress matrix, identify stress coupling feature vectors through a convolutional neural network, calculate the contribution weight of each stress component to the fault when a fault occurs using a gradient backpropagation algorithm, and adjust the subsequent stress application strategy based on the contribution weight. The fault prediction and root cause analysis module, connected to the multi-stress coupling analysis module, is used to perform fault prediction and root cause analysis during the failure limit test phase, under stress... Synchronously collect internal status data of the product when a fault is triggered. Product internal status data Input a pre-trained fault knowledge graph, output fault mode confidence. and a set of potential design flaws, if If this is not the case, the dynamic stress backtracking mechanism is activated, and the stress application sequence is reconstructed based on the set of potential design defects, and the failure limit test is continued.

9. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to cause the electronic device to perform the reliability enhancement test limit stress information determination method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for determining the ultimate stress information of a reliability enhancement test as described in any one of claims 1-7.

Images