Secondary power supply fault detection and diagnosis method and device for high-temperature aging test equipment

Abstract

The embodiments of this specification provide a method and device for detecting and diagnosing secondary power supply faults in high-temperature aging test equipment. The method for detecting and diagnosing secondary power supply faults in high-temperature aging test equipment includes: detecting faults in the secondary switching power supply simulation model to obtain the states of multiple virtual measuring points, and initially constructing an original D matrix according to multiple fault types of the secondary switching power supply; using the monitoring parameters corresponding to the data-based fault detection model as virtual measuring points to expand the D matrix, and using the constructed expanded D matrix to match the actual test vector to achieve fault diagnosis of the secondary switching power supply.

Description

Technical Field

[0001] The embodiments of this specification relate to the technical field of non-destructive reliability screening of integrated circuits, and particularly to a method and device for detecting and diagnosing secondary power supply faults in high-temperature aging test equipment. Background Art

[0002] An aging station is an aging test equipment for electronic components and products. By continuously applying a certain electrical stress to the components for a long time, it accelerates various physical and chemical reaction processes inside the components, promotes the early exposure of various potential defects inside the components, and thus evaluates the reliability and durability of the products. The secondary switching power supply of the aging station has a secondary protection function. Its working principle is similar to that of an ordinary switching power supply, but it pays more attention to the protection and monitoring of the circuit. It will continuously monitor parameters such as the voltage and current of the circuit. Once an abnormality is found, it will immediately cut off the power supply or adjust the circuit parameters. Existing high-temperature aging products can achieve long-term monitoring of the test environment by means of increasing in-machine long-term monitoring and over-stress protection mechanisms, and can respond in a timely manner when a product fails, realizing after-sales maintenance based on fault data. However, due to the lack of active guarantee technologies for high-temperature aging products in existing aging products, it is difficult for related products to achieve the integrity of the test process and the consistency of the test environment stress, which is extremely likely to lead to major property losses such as the forced interruption of the test process due to product failures resulting in the damage of millions of test devices, or the aging test being recognized as a failed test due to adverse effects such as additional stress introduced during the test period due to product performance degradation, resulting in ineffective waste of resources. At the same time, ensuring the quality of high-temperature aging tests can avoid over-aging and under-aging faults of integrated circuits, greatly reduce the failure rate of integrated circuits, and avoid large-scale electronic system failures and outages such as new energy vehicles, civil airliners, and energy storage substations caused by integrated circuit failures.

[0003] During the aging test process of components, once a secondary switching power supply fails, such as due to short circuit or overload, it will have a great impact on the aging station equipment, test process, test results, test environment, etc. Therefore, how to detect potential hidden dangers that may exist inside the power supply in advance, perform repairs or replacements in a timely manner, and ensure the safety of equipment and personnel as well as the successful completion of the device aging test is a key technical problem that urgently needs to be solved. Summary of the Invention

[0004] In view of this, the embodiments of this specification provide a method for detecting and diagnosing secondary power supply faults in high-temperature aging test equipment. One or more embodiments of this specification also relate to a device for detecting and diagnosing secondary power supply faults in high-temperature aging test equipment, a computing device, a computer-readable storage medium, and a computer program to solve the technical defects existing in the prior art.

[0005] According to the first aspect of the embodiments of this specification, a method for detecting and diagnosing secondary power supply faults of a high-temperature aging test device is provided, including:

[0006] By performing real-time monitoring on the secondary power supply of the target high-temperature aging test device, real-time monitoring data of the secondary power supply of the target high-temperature aging test device is obtained;

[0007] By setting test thresholds and 0 / 1 logic conversion rules for different fault types, the real-time monitoring data is constructed into a target one-dimensional test vector;

[0008] Based on the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the pre-stored extended D matrix, the fault type of the secondary power supply of the target high-temperature aging test device is determined and output.

[0009] Preferably, it further includes:

[0010] According to the types of electronic components and the connection relationships between electronic components in the secondary power supply of the high-temperature aging test device, a simulation model of the secondary power supply of the high-temperature aging test device is constructed, and by using the simulation model of the secondary power supply of the high-temperature aging test device, fault monitoring data of the secondary power supply of the high-temperature aging test device under different fault types is obtained;

[0011] A data-driven fault diagnosis model is established, and the data-driven fault diagnosis model is trained by using the fault monitoring data of the secondary power supply of the high-temperature aging test device under different fault types to obtain a trained data-driven fault diagnosis model.

[0012] Preferably, it further includes:

[0013] By injecting physical fault data into the simulation model of the secondary power supply of the high-temperature aging test device, fault monitoring data of the secondary power supply of the high-temperature aging test device under different fault types is obtained;

[0014] According to the set test thresholds and the first 0 / 1 logic conversion rules, the fault monitoring data under different fault types is respectively converted into first 0 / 1 logic values;

[0015] Based on the first 0 / 1 logic values, entity one-dimensional test vectors under different fault types are obtained, and an original D matrix is constructed from the entity one-dimensional test vectors under different fault types.

[0016] Preferably, it further includes:

[0017] By respectively inputting the fault monitoring data of the secondary power supply of the high-temperature aging test device under different fault types into the trained data-driven fault diagnosis model, fault diagnosis results under different fault types are obtained;

[0018] According to the set second 0 / 1 logic conversion rule, convert the fault diagnosis results under the different fault types into second 0 / 1 logic values respectively;

[0019] Based on the second 0 / 1 logic values, obtain virtual one-dimensional test vectors under different fault types, and add the virtual one-dimensional test vectors under the different fault types to the original D matrix to obtain an extended D matrix; wherein, the extended D matrix includes different fault types and their corresponding one-dimensional test vectors, and the one-dimensional test vectors are composed of entity one-dimensional test vectors and virtual one-dimensional test vectors.

[0020] Preferably, the 0 / 1 logic conversion rule includes the first 0 / 1 logic conversion rule and the second 0 / 1 logic conversion rule, wherein, constructing the real-time monitoring data into a target one-dimensional test vector by setting test thresholds and 0 / 1 logic conversion rules for different fault types includes:

[0021] By comparing the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment with the test thresholds set for different fault types, a comparison result is obtained, and based on the comparison result and the first 0 / 1 logic conversion rule, an entity one-dimensional test vector is obtained;

[0022] By inputting the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment into the data-driven fault diagnosis model, the fault diagnosis result of the secondary power supply of the target high-temperature aging test equipment is obtained, and based on the fault diagnosis result and the second 0 / 1 logic conversion rule, a virtual one-dimensional test vector is obtained;

[0023] Based on the entity one-dimensional test vector and the virtual one-dimensional test vector, the target one-dimensional test vector of the secondary power supply of the target high-temperature aging test equipment is obtained.

[0024] Preferably, determining and outputting the fault type of the secondary power supply of the target high-temperature aging test equipment based on the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the pre-stored extended D matrix includes:

[0025] Calculate the similarity between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix;

[0026] Determine the maximum value among the similarities between the target one-dimensional test vector and the one-dimensional test vectors corresponding to the multiple fault types, and determine the fault type corresponding to the maximum value as the fault type of the secondary power supply of the target high-temperature aging test equipment and output it.

[0027] Preferably, calculating the similarity between the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type in the extended D matrix includes:

[0028] Calculating the Mahalanobis distance, cosine similarity, and Manhattan distance between the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type in the extended D matrix respectively;

[0029] By performing multi-distance metric fusion on the Mahalanobis distance, the cosine similarity, and the Manhattan distance, the similarity between the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type in the extended D matrix is obtained, which includes:

[0030]

[0031] Wherein, represents the similarity between the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type in the extended D matrix, represents the Mahalanobis distance; represents the cosine similarity; represents the Manhattan distance; , and represent weights respectively, and + + = 1.

[0032] According to the second aspect of the embodiments of the present specification, a secondary power supply fault detection and diagnosis device for high-temperature aging test equipment is provided, including:

[0033] An acquisition module, configured to obtain real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment by performing real-time monitoring on the secondary power supply of the target high-temperature aging test equipment;

[0034] A construction module, configured to construct the real-time monitoring data into a target one-dimensional test vector by setting test thresholds and 0 / 1 logic conversion rules for different fault types;

[0035] A fault detection and diagnosis module, configured to determine and output the fault type of the secondary power supply of the target high-temperature aging test equipment based on the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type stored in the pre-stored extended D matrix.

[0036] According to the third aspect of the embodiments of the present specification, a computing device is provided, including:

[0037] A memory and a processor;

[0038] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the steps of any one of the secondary power supply fault detection and diagnosis methods for the high-temperature aging test equipment.

[0039] According to the fourth aspect of the embodiments of the present specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of any one of the secondary power supply fault detection and diagnosis methods for the high-temperature aging test equipment are implemented.

[0040] According to the fifth aspect of the embodiments of the present specification, a computer program is provided, wherein when the computer program is executed on a computer, the computer is made to execute the steps of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment.

[0041] In the aging test process of the embodiments of the present specification, monitoring signals such as voltage and current will be generated by the secondary switching power supply of the aging test bench. First, fault detection is performed on the secondary switching power supply simulation model to obtain the states of multiple virtual measurement points, and an original D matrix is initially constructed according to multiple fault types of the secondary switching power supply; secondly, since the D matrix may have undetectable faults, fuzzy groups, or fault types not covered, the monitoring parameters corresponding to the data-based fault detection model can be used as virtual measurement points to expand the D matrix and solve the problems and deficiencies existing in the above-mentioned original D matrix. Finally, the constructed extended D matrix is used to match the actual test vector to achieve fault diagnosis of the secondary switching power supply. Description of the Drawings

[0042] Figure 1 is a flowchart of a secondary power supply fault detection and diagnosis method for a high-temperature aging test equipment provided by an embodiment of the present specification;

[0043] Figure 2 is a schematic diagram of another secondary power supply fault detection and diagnosis process for a high-temperature aging test equipment provided by an embodiment of the present specification;

[0044] Figure 3 is a schematic diagram of the process from the original D matrix to the extended D matrix provided by an embodiment of the present specification;

[0045] Figure 4 is a schematic diagram of the structure of a secondary power supply fault detection and diagnosis device for a high-temperature aging test equipment provided by an embodiment of the present specification;

[0046] Figure 5 is a block diagram of the structure of a computing device provided by an embodiment of the present specification. Detailed Embodiments

[0047] Numerous specific details are set forth in the following description to provide a thorough understanding of this specification. However, this specification can be implemented in many other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this specification. Therefore, this specification is not limited by the specific implementations disclosed below.

[0048] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms "a", "the", and "said" used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0049] It should be understood that although the terms first, second, etc. may be used in one or more embodiments of this specification to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining".

[0050] In this specification, a method for detecting and diagnosing secondary power supply faults of a high-temperature aging test equipment is provided. This specification also relates to a device for detecting and diagnosing secondary power supply faults of a high-temperature aging test equipment, a computing device, a computer-readable storage medium, and a computer program, which will be described in detail one by one in the following embodiments.

[0051] Figure 1 A flowchart of a method for detecting and diagnosing secondary power supply faults of a high-temperature aging test equipment according to an embodiment of this specification is shown, which specifically includes the following steps.

[0052] Step S101: Obtain real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment by performing real-time monitoring on the secondary power supply of the target high-temperature aging test equipment;

[0053] Step S102: Construct a target one-dimensional test vector from the real-time monitoring data by setting test thresholds and 0 / 1 logic conversion rules for different fault types;

[0054] Step S103: Based on the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the pre-stored extended D matrix, determine the fault type of the secondary power supply of the target high-temperature aging test equipment and output it.

[0055] In an alternative implementation, it further includes:

[0056] Construct a simulation model of the secondary power supply of the high-temperature aging test equipment according to the types of electronic components and the connection relationships between the electronic components in the secondary power supply of the high-temperature aging test equipment, and use the simulation model of the secondary power supply of the high-temperature aging test equipment to obtain the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types;

[0057] Establish a data-driven fault diagnosis model, and use the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types to train the data-driven fault diagnosis model to obtain a trained data-driven fault diagnosis model.

[0058] In an alternative implementation, it further includes:

[0059] Inject physical fault data into the simulation model of the secondary power supply of the high-temperature aging test equipment to obtain the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types;

[0060] According to the set test threshold and the first 0 / 1 logic conversion rule, convert the fault monitoring data under different fault types into first 0 / 1 logic values respectively;

[0061] Based on the first 0 / 1 logic values, obtain the entity one-dimensional test vectors under different fault types, and construct the original D matrix from the entity one-dimensional test vectors under different fault types.

[0062] In an alternative implementation, it further includes:

[0063] Input the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types into the trained data-driven fault diagnosis model respectively to obtain the fault diagnosis results under different fault types;

[0064] According to the set second 0 / 1 logic conversion rule, convert the fault diagnosis results under different fault types into second 0 / 1 logic values respectively;

[0065] Based on the second 0 / 1 logical value, virtual one-dimensional test vectors under different fault types are obtained, and the virtual one-dimensional test vectors under different fault types are added to the original D matrix to obtain an extended D matrix; wherein, the extended D matrix includes different fault types and their corresponding one-dimensional test vectors, and the one-dimensional test vectors are composed of entity one-dimensional test vectors and virtual one-dimensional test vectors.

[0066] In an alternative embodiment, the 0 / 1 logical conversion rule includes the first 0 / 1 logical conversion rule and the second 0 / 1 logical conversion rule, wherein constructing the real-time monitoring data into a target one-dimensional test vector by setting test thresholds and 0 / 1 logical conversion rules for different fault types includes:

[0067] By comparing the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment with the test thresholds set for different fault types, a comparison result is obtained, and based on the comparison result and the first 0 / 1 logical conversion rule, an entity one-dimensional test vector is obtained;

[0068] By inputting the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment into the data-driven fault diagnosis model, a fault diagnosis result of the secondary power supply of the target high-temperature aging test equipment is obtained, and based on the fault diagnosis result and the second 0 / 1 logical conversion rule, a virtual one-dimensional test vector is obtained;

[0069] Based on the entity one-dimensional test vector and the virtual one-dimensional test vector, a target one-dimensional test vector of the secondary power supply of the target high-temperature aging test equipment is obtained.

[0070] In an alternative embodiment, determining and outputting the fault type of the secondary power supply of the target high-temperature aging test equipment based on the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type in the pre-stored extended D matrix includes:

[0071] Calculating the similarity between the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type in the extended D matrix;

[0072] Determining the maximum value among the similarities between the target one-dimensional test vector and the one-dimensional test vectors corresponding to the multiple fault types, and determining and outputting the fault type corresponding to the maximum value as the fault type of the secondary power supply of the target high-temperature aging test equipment.

[0073] In an alternative embodiment, calculating the similarity between the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type in the extended D matrix includes:

[0074] Calculate the Mahalanobis distance, cosine similarity, and Manhattan distance between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix respectively;

[0075] By performing multi-distance metric fusion on the Mahalanobis distance, the cosine similarity, and the Manhattan distance, obtain the similarity between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix, which includes:

[0076]

[0077] Among them, represents the similarity between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix, represents the Mahalanobis distance; represents the cosine similarity; represents the Manhattan distance; , and represent weights respectively, and + + = 1.

[0078] In summary, in the aging test process of the embodiments of this specification, the secondary switching power supply of the aging test bench will generate monitoring signals such as voltage and current. First, through fault detection of the secondary switching power supply simulation model, obtain the states of multiple virtual measurement points, and initially construct the original D matrix according to multiple fault types of the secondary switching power supply; secondly, since the D matrix may have undetectable faults, fuzzy groups, or fault types not covered, the monitoring parameters corresponding to the data-based fault detection model can be used as virtual measurement points to expand the D matrix to solve the problems and deficiencies existing in the above-mentioned original D matrix. Finally, use the constructed extended D matrix to match the actual test vector to achieve fault diagnosis of the secondary switching power supply.

[0079] Figure 2 is a schematic diagram of another process for fault detection and diagnosis of the secondary power supply of the high-temperature aging test equipment provided by an embodiment of this specification. As Figure 2 shown, it includes:

[0080] Step S1: Fault simulation model of the secondary switching power supply of the aging test bench (obtain fault data);

[0081] According to the types of electronic components and the connection relationships among the electronic components in the secondary switching power supply, an analog simulation model of the secondary switching power supply is constructed; current and voltage data of the secondary switching power supply of the aging platform in all states (including fault states and normal states) are obtained through the method of fault simulation injection or physical fault injection.

[0082] Step S2: Use the monitoring data of the data-driven fault detection model as virtual measuring points;

[0083] Based on the current and voltage data of the secondary switching power supply fault obtained in Step S1, a data-driven fault diagnosis model is established to detect faults in the secondary switching power supply, and the states of the virtual measuring points are obtained. Among them, the virtual measuring points are the detection results of the data-driven secondary switching power supply, including fault modes and normal states. After obtaining the states of the virtual measuring points, they are expanded into the D matrix to obtain the extended D matrix. The advantage of using the fault characteristics of the monitoring parameters corresponding to the data-driven fault diagnosis model as virtual measuring points is that the diagnosis model has self-adaptability, robustness, and contains richer fault information. Among them, the diagnosis model is: the secondary switching power supply fault detection model. Among them, the main function of the secondary switching power supply fault detection model is to extract the corresponding fault characteristics from the detected current and voltage signals and complete fault diagnosis, and monitor whether there are faults and the fault modes that occur.

[0084] The fault characteristics of the monitoring parameters at least include: the ripple voltage of the current and voltage, the average output voltage, the peak-to-peak value of the inductor current, the peak value of the inductor current, the peak-to-peak value of the capacitor current, and the peak value of the capacitor current. Among them, the essential characteristics of the change in the ripple voltage: the root cause of the sudden increase in high-frequency ripple is: the increase in the ESR of the output capacitor (drying out of the electrolytic capacitor), abnormal switching timing of the synchronous rectifier MOSFET, and failure of the input capacitor; the root cause of the distortion of the low-frequency ripple: the failure of the feedback loop compensation (phase margin < 30°), and the voltage fluctuation of the PWM controller power supply. The essential characteristics of the deviation of the average output voltage: voltage increase: the PWM duty cycle is locked at the maximum value, and the optocoupler isolation fails (CTR value drops > 50%); voltage decrease: the output impedance increases significantly, and the parasitic resistance of the power loop increases (PCB copper foil temperature rise > 40°C); voltage oscillation; the loop gain exceeds the critical value, and the capacitance value of the compensation network capacitor drifts (such as Ccomp deviation > 20%). The essential characteristics of the abnormal inductor current: the increase in the peak-to-peak value: inductor saturation (L value drops > 30%) or input voltage overshoot; the decrease in the peak-to-peak value: the switching frequency increases abnormally. The essential characteristics of the abnormal capacitor current: the change law of the peak-to-peak value: spike burrs: resonance caused by the increase in ESL, flat-top distortion: attenuation of the capacitor capacitance value.

[0085] The impact of the DUT (Device Under Test) on the secondary power supply is a complex and multi-dimensional issue, which can mainly be attributed to four dimensions: load characteristics, parasitic coupling, thermal interaction, and feedback interference. Load characteristics are one of the key factors directly affecting the secondary power supply by the DUT. As the load of the secondary power supply, the characteristics of the DUT, such as its impedance, power demand, and dynamic load changes, will directly affect the output voltage, current, and stability of the secondary power supply. Parasitic coupling is also an influencing factor that cannot be ignored. In a complex electronic system, there may be a coupling effect between the DUT and the secondary power supply through parasitic elements such as capacitors and inductors, resulting in problems such as power supply noise and ripple. Thermal interaction is another important dimension. During the operation of an electronic device, both the DUT and the secondary power supply will generate a certain amount of heat. If the thermal design is improper, this heat may accumulate inside the device, leading to an increase in temperature. A high-temperature environment will not only affect the performance and lifespan of the secondary power supply but may also have an adverse impact on the reliability and stability of the DUT. Feedback interference is also an important aspect of the impact of the DUT on the secondary power supply. This feedback interference may cause problems such as unstable output voltage and increased ripple of the secondary power supply. In severe cases, it may even damage the power supply or cause the system to crash.

[0086] Step S3: Construct the original D matrix based on the monitoring parameters of the entity object;

[0087] The current and voltage data of all states (including fault states and normal states) of the secondary switching power supply of the aging platform obtained through the physical fault injection method. According to the detection thresholds of the current and voltage data and the first 0 / 1 logic conversion rule, the measured values in different states are converted into 0 / 1 logic values (set to 1 if exceeding the monitoring threshold, and set to 0 if not exceeding the monitoring threshold). Traverse all states to construct the original D matrix represented by 0 / 1 logic.

[0088] Furthermore, if the monitored data of the current and voltage exceeds the test thresholds set for different fault types, the set data of the monitored data of the current and voltage is set to 1 according to the first 0 / 1 logic conversion rule; and if the real-time monitored data does not exceed the test thresholds set for different fault types, the set data of the monitored data of the current and voltage is set to 0 according to the first 0 / 1 logic conversion rule.

[0089] Among them, the monitoring parameters of the current and voltage are: ripple voltage, average output voltage, peak-to-peak inductor current, peak inductor current, peak-to-peak capacitor current, and peak capacitor current. According to whether the monitored parameter values exceed the monitoring threshold, if they exceed the threshold, they are set to 1, representing a fault, and if they do not exceed the threshold, they are set to 0, representing a normal state.

[0090] Step S4: Supplement the virtual measurement points into the original D matrix to obtain the extended D matrix;

[0091] From step S3, the original D matrix in the physical fault injection mode can be obtained. Since the original D matrix may have undetectable faults, fuzzy groups, or fault modes that are not covered, the fault characteristics of the monitoring parameters represented by the data-driven fault detection model are used as virtual measurement points to supplement the original D matrix, expand the D matrix, and construct a complete extended D matrix model to solve the above problems. The monitoring parameters can be parameters such as the input, output, and state variables of the system.

[0092] In this specification, to solve the problems of undetectable faults, fuzzy groups, or fault types not covered in the fault diagnosis of the original D matrix, a data-driven detection model is proposed as a virtual measurement point to expand the D matrix. The advantages are that the diagnostic model has self-adaptability, robustness, and contains richer fault information.

[0093] Specifically, the result of extracting fault-related feature information from the data and completing the fault diagnosis by the secondary switching power supply fault detection model is used as the newly added virtual measurement point after expansion. According to the correlation relationship between the monitoring parameters represented by the fault detection model and different fault modes, specifically, whether the fault detection model can detect the fault mode, and according to the second 0 / 1 logic conversion rule (set to 1 if the fault mode can be detected, set to 0 if it cannot be detected) to expand the original D matrix. A specific example of the expansion of the D matrix is shown in the following figure. The expanded D matrix mainly includes physical measurement points: measurement point A, measurement point B, measurement point C…, and virtual measurement points: the secondary switching power supply fault detection model. Among them, the result of the fault diagnosis completed by one or more fault-related feature information extracted by the secondary switching power supply fault detection model according to the monitoring parameters is used as the virtual measurement point. According to the correlation relationship between the monitoring parameters represented by these fault detection models and different fault modes, the original D matrix can be expanded, as Figure 3 shown.

[0094] Furthermore, by inputting the monitoring parameters of the current and voltage of the secondary power supply of the high-temperature aging test equipment into the data-driven fault diagnosis model, a fault diagnosis result is obtained. If the fault diagnosis result is that a fault type is detected, the setting data of the data-driven fault diagnosis model is set to 1 according to the second 0 / 1 logic conversion rule; and if the fault diagnosis result is that no fault type is detected, the setting data of the data-driven fault diagnosis model is set to 0 according to the second 0 / 1 logic conversion rule.

[0095] Step S5: Construct a physical monitoring vector and obtain the final diagnostic inference result.

[0096] Based on the extended D-matrix model, the model is analyzed to achieve model simplification and optimal selection of test points, and finally a diagnostic strategy is obtained. First, real-time monitoring data is collected from multiple measurement points (physical measurement points and virtual measurement points) of the secondary switching power supply, and the real-time test results of each test point are obtained. Through the setting of test thresholds and 0 / 1 logic conversion rules, a one-dimensional test vector (the test results are combined into a vector) is constructed. Then, the test vector is compared with each row of the D-matrix to find the row that is closest to the test vector, which is usually achieved by calculating the distance (such as Manhattan distance, cosine similarity, etc.) between the test vector and each row of the D-matrix. Finally, the fault state corresponding to the row closest to the test vector is the diagnosed fault state.

[0097] An embodiment of the present application also provides a time-frequency domain feature processing method based on an alternating mask reconstruction mechanism, which is used to obtain real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment during the fault detection and diagnosis process of the high-temperature aging test equipment. Among them, the real-time monitoring data includes target time-frequency domain feature data.

[0098] Furthermore, the time-frequency domain feature processing method based on the alternating mask reconstruction mechanism in the embodiment of the present application includes the following steps:

[0099] Step 201: Collect the fault data set of the aging bench device.

[0100] The aging bench provided by the present application can perform non-destructive aging detection on the test piece. The time-frequency domain feature processing method in the embodiment of the present application can extract features from the state data, health data, etc. of the aging bench device, and the extracted features can be used for health assessment, fault prediction and diagnosis, etc. of the aging bench device, so as to ensure the safety of the test piece.

[0101] Among them, the fault data includes multiple key parameter time series data, and the key parameters are different for different devices. For example: in the case of a two-level switching power supply device, the key parameters can be power supply voltage, current signal, and the thermal resistance characteristic from the MOSFET case temperature to the junction temperature synchronously monitored by an infrared thermal imager and an embedded temperature sensor. In the case of a test chamber device, the key parameter can be a temperature monitoring signal, etc. These monitoring signals can be used to evaluate the health status of the aging bench device. In the actual implementation process, the directly measured key parameter data is large in quantity and contains a large amount of redundant information. Therefore, it is necessary to use the feature extraction model trained by the time-frequency domain feature processing method based on the alternating mask reconstruction mechanism in the embodiment of the present application to perform preliminary feature extraction on the directly measured data to obtain a feature sequence with high information density.

[0102] In the embodiments of the present application, the fault data set of the aging bench devices can be the fault data set during the entire process from the start of degradation to the stop of operation (i.e., the end of the remaining life cycle) in the whole life cycle of the devices.

[0103] When collecting the fault data set of the aging bench devices, it is necessary to synchronously monitor the load change characteristics of the secondary switching power supply, including the output voltage and current transient response under dynamic load switching. For the load fluctuation when the device under test is connected, record the power supply ripple volatility caused by it. The calculation formula is: Ripple volatility = ( / ) × 100%

[0104] Where, is the peak-to-peak value of the output voltage, is the average value of the output voltage. Capture the ripple waveform through a high-precision oscilloscope and analyze its spectral characteristics to identify abnormal harmonic components. The devices can include but are not limited to: aging bench switching power supply, aging bench high-temperature test chamber, aging bench high-temperature and high-humidity test chamber, etc. The aging bench switching power supply can be further divided into: primary switching power supply, secondary switching power supply, and secondary switching power supply MOSFET switch tube. Moreover, the aging bench devices can also include: drive control detection board.

[0105] The specific devices included in the aging bench can be flexibly adjusted according to the type of the aging bench. In the embodiments of the present application, the specific type of the aging bench is not limited. For example: the aging bench can be a high-temperature and high-humidity aging bench, can be a constant temperature and constant pressure aging bench, or can also be a high-temperature aging test equipment, etc.

[0106] Step 202: Preprocess the fault data set to obtain a training data set and a test data set.

[0107] An optional way to preprocess the fault data set to obtain a training data set and a test data set can include the following sub-steps:

[0108] Sub-step 2021: For each key parameter in the fault data set, perform sliding window cutting on the key parameter time series data to construct a sample data set;

[0109] When constructing the sample data set by sliding window cutting, it is necessary to perform adaptive segmentation on the data during the load mutation period. For the interval where the load change rate exceeds the threshold (such as ±10% / ms), use a smaller window width and step size to capture the detailed features of the transient response. At the same time, perform dynamic baseline calibration on the current and voltage output data to eliminate the baseline drift caused by the connection of the device under test.

[0110] Sub-step 2022: Normalize each data sample in the sample data set to obtain a normalized sample data set;

[0111] Sub-step 2023: Select the first preset percentage of the sample data from the normalized sample dataset as the training dataset, and use the remaining sample data as the test dataset.

[0112] Among them, the training dataset is used to train a preset model, and the test dataset is used to verify the prediction accuracy of the trained feature extraction model. The preset percentage can be flexibly set by those skilled in the art, and no specific limitation is made in the embodiments of the present application. For example, it can be set to 60%-80%, preferably 70%.

[0113] Step 203: Perform short-time Fourier transform on the time-series data of each key parameter in the training dataset to obtain a time-frequency diagram.

[0114] Most traditional time-domain and frequency-domain analysis methods are for stationary signals and can only obtain information in one aspect of the time domain or frequency domain. In the application of actual devices such as secondary switching power supplies, the collected signals are mostly non-stationary, and the analysis in one aspect can no longer meet the needs. Therefore, it is necessary to determine the relationship between the signal frequency and time. The joint time-frequency analysis method is a very effective tool in the current process of processing non-stationary signals. Therefore, short-time Fourier transform is used in this application to process variable time-series and non-stationary signals. The basic principle of STFT (i.e., short-time Fourier transform) is: use a window function h(t) with a finite duration to intercept the vibration signal, perform Fourier transform on the obtained signal to obtain the local spectrum in a small range of this time period, and gradually analyze the signal band by moving the window function h(t) on the time axis to obtain a set of local "spectrums" of the signal. STFT is essentially a transformation of the basis function.

[0115] An optional way to perform short-time Fourier transform on the time-series data in the training dataset to obtain a time-frequency diagram may include the following sub-steps:

[0116] Sub-step 2031: For each time-series data in the training dataset, use a window function with a finite duration to intercept the vibration signal in the time-series data;

[0117] Sub-step 2032: Perform Fourier transform on the intercepted vibration signal to obtain the local spectrum in a small range of the corresponding time period;

[0118] The transformation formula for the signal collected by the aging bench device such as a secondary power supply can be:

[0119]

[0120] In the formula: is the source signal, is the analysis window function; is the time-frequency spectrum at time t.

[0121] Sub-step 2033: By moving the window function on the time axis, gradually analyze each vibration signal band to obtain a set of local spectra of the vibration signal, and generate a time-frequency diagram.

[0122] Preferably, the window function is selected as the Hanning window function, which has good smoothness and is suitable for reducing spectral leakage. The expression of the Hanning window function is as follows:

[0123]

[0124] Step 204: Randomly mask each time-frequency diagram and input it into a preset model for training to generate a feature extraction model.

[0125] Through steps 201 to 204, time-frequency diagrams corresponding to different time segments of multiple key parameters can be obtained. For each time-frequency diagram, the time-frequency diagram can be masked in a random masking manner. The time-frequency diagram is different from the time series feature and the frequency domain feature. The time series feature and the frequency domain feature are one-dimensional data, and the time-frequency diagram is a two-dimensional picture. The abscissa represents time, the ordinate represents frequency, and the different depths of color represent different frequency amplitudes.

[0126] Send the randomly masked time-frequency diagrams into a preset model such as the Transformer autoencoder for training respectively. After the training is completed, a feature extraction model is generated.

[0127] Transformer is a model established based on the Seq-to-Seq framework. Compared with classical deep learning models, the most prominent advantage of Transformer is the use of the multi-head attention mechanism. The purpose of the multi-head attention layer is to assign different importance to words / tokens in the sequence from multiple aspects. The Transformer model mainly consists of parts such as input, encoder, decoder, and output.

[0128] Position encoding: The position encoding layer is to determine the position information of the sequence. Since there are no recursive layers and convolutional layers in RNN and CNN, and only relying on the self-attention mechanism cannot obtain the order information of the input, it is necessary to actively transmit the order information of the sequence to the model. The Transformer model uses a combination of sine and cosine functions to perform position encoding on the sequence, and the calculation method is as follows:

[0129]

[0130]

[0131] In the formula: is the position of the current sequence; is the dimension; is the dimension of the input feature.

[0132] Multi-Head Attention Mechanism: The multi-head attention mechanism performs operations in parallel using multiple attention mechanisms and then concatenates the operation results through a linear transformation. The core technology in Transformer is the multi-head attention mechanism, which is used to extract the dependency relationship features between data, capture the correlation between data, and establish a context prediction model. The calculation method is as follows:

[0133]

[0134]

[0135]

[0136]

[0137]

[0138]

[0139] Where: is the query matrix; is the key matrix; is the value matrix; , , is the trainable parameter matrix; is the processed input; is the dimension of the key matrix; , , , is the learnable parameter matrix.

[0140] Feed-Forward Network and Summation and Normalization: In the Transformer model, the encoding part and the decoding part also include a feed-forward network and summation and normalization. The calculation formula of the feed-forward neural network is as follows:

[0141]

[0142] Where: is the input; , , , are the parameters that can be obtained through training.

[0143] The calculation formula of summation and normalization is as follows:

[0144]

[0145] Where: is the input; is the result after processing by the module.

[0146] In an alternative embodiment, the method of randomly masking each time-frequency diagram and inputting it into a preset model for training to generate a feature extraction model may include the following sub-steps:

[0147] Sub-step 2041: Input the randomly masked time-frequency diagram into the preset model, send it to the encoding module after position encoding, and send the output of the encoding module to the decoder. The decoder restores the feature parameters in the high-dimensional hidden layer into the time-frequency diagram before masking;

[0148] Sub-step 2042: Use the restored time-frequency diagram to iteratively train the preset model multiple times to adjust the model parameters;

[0149] Among them, the model parameters include: embedding dimension, number of attention heads, number of encoder layers, and number of decoder layers.

[0150] Sub-step 2043: Take out the encoder layer and decoder layer of the model after multiple iterative adjustments, and retain the weight parameters of each layer.

[0151] Sub-step 2044: Generate a feature extraction model based on each encoder layer, each decoder layer, and the weight parameters of each layer.

[0152] Step 205: Test the feature extraction model according to the test data set.

[0153] After training (also known as pre-training) the preset model based on the training data set to generate a feature extraction model, the feature extraction model can be tested based on the test data in the test data set to determine the accuracy of the prediction result of the feature extraction model.

[0154] Based on the pre-trained Transformer auto-encoding model for the training data set perform self-attention feature extraction to obtain a self-attention feature set .

[0155] Step 206: After passing the test, input the key parameter time series data of the device to be predicted into the feature extraction model to predict the target time-frequency domain features.

[0156] The time-frequency domain feature processing method based on the alternating mask reconstruction mechanism provided by the embodiments of the present application collects the fault data set of the aging bench devices; preprocesses the fault data set to obtain the training data set and the test data set; performs short-time Fourier transform on the time series data of each key parameter in the training data set to obtain the time-frequency diagram; randomly masks each time-frequency diagram and inputs it into a preset model for training to generate a feature extraction model; tests the feature extraction model based on the test data set; after passing the test, inputs the time series data of the key parameters of the device to be predicted into the feature extraction model to predict the target time-frequency domain features. The method provided by the embodiments of the present invention fully coordinates and refines a large amount of fault data of the aging bench devices, randomly masks the time-frequency diagrams obtained by performing short-time Fourier transform on the segmented time series data, and uses the randomly masked time-frequency diagrams to train the preset model, which can fully refine the local semantic information of each part of the aging bench devices, overcome the restrictions on training models such as data scarcity and rough annotation, and the time-frequency features predicted by the trained feature extraction model are more accurate and reliable. Further, the diagnosis result of the aging bench devices analyzed based on the predicted time-frequency features is more reliable.

[0157] Corresponding to the embodiment of the secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment described above, this specification also provides an embodiment of a secondary power supply fault detection and diagnosis device for the high-temperature aging test equipment. Figure 4 The structural schematic diagram of a secondary power supply fault detection and diagnosis device for a high-temperature aging test equipment provided by an embodiment of this specification is shown. As Figure 4 shown, the device includes:

[0158] An acquisition module 401, configured to obtain the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment by performing real-time monitoring on the secondary power supply of the target high-temperature aging test equipment;

[0159] A construction module 402, configured to construct the real-time monitoring data into a target one-dimensional test vector by setting test thresholds and 0 / 1 logic conversion rules for different fault types;

[0160] A fault detection and diagnosis module 403, configured to determine and output the fault type of the secondary power supply of the target high-temperature aging test equipment based on the target one-dimensional test vector and the one-dimensional test vector corresponding to each fault type in the pre-stored extended D matrix.

[0161] The above is a schematic solution of a secondary power supply fault detection and diagnosis device for a high-temperature aging test equipment according to this embodiment. It should be noted that the technical solution of this secondary power supply fault detection and diagnosis device for a high-temperature aging test equipment belongs to the same concept as the above technical solution of the secondary power supply fault detection and diagnosis method for a high-temperature aging test equipment. For the details not described in the technical solution of the secondary power supply fault detection and diagnosis device for a high-temperature aging test equipment, reference can be made to the description of the technical solution of the secondary power supply fault detection and diagnosis method for a high-temperature aging test equipment above.

[0162] Figure 5 FIG. shows a block diagram of a computing device 500 according to an embodiment of the present specification. The components of the computing device 500 include, but are not limited to, a memory 510 and a processor 520. The processor 520 is connected to the memory 510 via a bus 530, and a database 550 is used to store data.

[0163] The computing device 500 further includes an access device 540, and the access device 540 enables the computing device 500 to communicate via one or more networks 560. Examples of these networks include the Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or a combination of communication networks such as the Internet. The access device 540 may include one or more of any type of wired or wireless network interfaces (e.g., Network Interface Card (NIC)), such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, Worldwide Interoperability for Microwave Access (Wi-MAX) interface, Ethernet interface, Universal Serial Bus (USB) interface, cellular network interface, Bluetooth interface, Near Field Communication (NFC) interface, and so on.

[0164] In an embodiment of the present specification, the above components of the computing device 500 and Figure 5 other components not shown in Figure 5 may also be connected to each other, for example, via a bus. It should be understood that

[0165] the block diagram of the computing device shown is only for illustrative purposes and is not a limitation on the scope of the present specification. Those skilled in the art can add or replace other components as needed.

[0166] Among them, the processor 520 is configured to execute the following computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment are implemented.

[0167] The above is a schematic solution of a computing device according to this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment belong to the same concept. For the details not described in detail in the technical solution of the computing device, reference can be made to the description of the technical solution of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment.

[0168] An embodiment of this specification also provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment are implemented.

[0169] The above is a schematic solution of a computer-readable storage medium according to this embodiment. It should be noted that the technical solution of this storage medium and the technical solution of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment belong to the same concept. For the details not described in detail in the technical solution of the storage medium, reference can be made to the description of the technical solution of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment.

[0170] An embodiment of this specification also provides a computer program. Among them, when the computer program is executed on a computer, the computer is made to execute the steps of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment.

[0171] The above is a schematic solution of a computer program according to this embodiment. It should be noted that the technical solution of this computer program and the technical solution of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment belong to the same concept. For the details not described in detail in the technical solution of the computer program, reference can be made to the description of the technical solution of the above-mentioned secondary power supply fault detection and diagnosis method for the high-temperature aging test equipment.

[0172] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain implementations, multitasking and parallel processing are also possible or may be advantageous.

[0173] The computer instructions include computer program code, which may be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, removable hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0174] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the embodiments of this specification are not limited by the described action sequence, because according to the embodiments of this specification, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments of this specification.

[0175] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0176] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not elaborate on all the details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the embodiments of this specification, so that those skilled in the art can well understand and utilize this specification. This specification is only limited by the claims and their full scope and equivalents.

Claims

1. A method for detecting and diagnosing secondary power supply faults in a high-temperature aging test equipment, characterized in that, Including: By performing real-time monitoring on the secondary power supply of the target high-temperature aging test equipment, obtaining the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment; By setting test thresholds for different fault types and 0 / 1 logic conversion rules, constructing the real-time monitoring data into a target one-dimensional test vector; Based on the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the pre-stored extended D matrix, determining and outputting the fault type of the secondary power supply of the target high-temperature aging test equipment; Wherein, the pre-stored extended D matrix includes: obtaining the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types, and respectively inputting the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types into a trained data-driven fault diagnosis model to obtain the fault diagnosis results under different fault types; according to the set second 0 / 1 logic conversion rules, respectively converting the fault diagnosis results under different fault types into second 0 / 1 logic values; based on the second 0 / 1 logic values, obtaining virtual one-dimensional test vectors under different fault types, and adding the virtual one-dimensional test vectors under different fault types to the original D matrix to obtain an extended D matrix; wherein, the extended D matrix includes different fault types and their corresponding one-dimensional test vectors, and the one-dimensional test vector is composed of an entity one-dimensional test vector and a virtual one-dimensional test vector.

2. The method according to claim 1, characterized in that, The trained data-driven fault diagnosis model includes: According to the types of electronic components in the secondary power supply of the high-temperature aging test equipment and the connection relationship between the electronic components, constructing a simulation model of the secondary power supply of the high-temperature aging test equipment, and using the simulation model of the secondary power supply of the high-temperature aging test equipment to obtain the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types; Establishing a data-driven fault diagnosis model, and training the data-driven fault diagnosis model by using the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types to obtain a trained data-driven fault diagnosis model.

3. The method according to claim 2, wherein The original D matrix includes: By injecting physical fault data into the simulation model of the secondary power supply of the high-temperature aging test equipment, obtaining the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types; According to the set test thresholds and the first 0 / 1 logic conversion rules, respectively converting the fault monitoring data under different fault types into first 0 / 1 logic values; Based on the first 0 / 1 logic values, obtaining entity one-dimensional test vectors under different fault types, and constructing the original D matrix from the entity one-dimensional test vectors under different fault types.

4. The method according to claim 3, wherein The 0 / 1 logic conversion rules include the first 0 / 1 logic conversion rule and the second 0 / 1 logic conversion rule, wherein, by setting test thresholds for different fault types and 0 / 1 logic conversion rules, constructing the real-time monitoring data into a target one-dimensional test vector includes: By comparing the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment with the test thresholds set for different fault types, a comparison result is obtained, and based on the comparison result and the first 0 / 1 logic conversion rule, a physical one-dimensional test vector is obtained; By inputting the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment into the data-driven fault diagnosis model, a fault diagnosis result of the secondary power supply of the target high-temperature aging test equipment is obtained, and based on the fault diagnosis result and the second 0 / 1 logic conversion rule, a virtual one-dimensional test vector is obtained; Based on the physical one-dimensional test vector and the virtual one-dimensional test vector, a target one-dimensional test vector of the secondary power supply of the target high-temperature aging test equipment is obtained.

5. The method according to claim 4, characterized in that Based on the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the pre-stored extended D matrix, the fault type of the secondary power supply of the target high-temperature aging test equipment is determined and output, including: Calculating the similarity between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix; Determining the maximum value among the similarities between the target one-dimensional test vector and the one-dimensional test vectors corresponding to multiple fault types, and determining the fault type corresponding to the maximum value as the fault type of the secondary power supply of the target high-temperature aging test equipment and outputting it.

6. The method according to claim 5, characterized in that, Calculating the similarity between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix includes: Respectively calculating the Mahalanobis distance, cosine similarity and Manhattan distance between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix; By performing multi-distance metric fusion on the Mahalanobis distance, the cosine similarity and the Manhattan distance, the similarity between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix is obtained, which includes: ； Among them, represents the similarity between the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the extended D matrix, represents the Mahalanobis distance; represents the cosine similarity; represents the Manhattan distance; , and respectively represent weights, and + + = 1.

7. A fault detection and diagnosis device for the secondary power supply of a high-temperature aging test equipment, comprising: An acquisition module configured to obtain the real-time monitoring data of the secondary power supply of the target high-temperature aging test equipment by performing real-time monitoring on the secondary power supply of the target high-temperature aging test equipment; A construction module configured to construct the real-time monitoring data into a target one-dimensional test vector by setting test thresholds and 0 / 1 logic conversion rules for different fault types; A fault detection and diagnosis module configured to determine and output the fault type of the secondary power supply of the target high-temperature aging test equipment based on the target one-dimensional test vector and the one-dimensional test vectors corresponding to each fault type in the pre-stored extended D matrix; Among them, the pre-stored extended D matrix includes: obtaining the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types, and respectively inputting the fault monitoring data of the secondary power supply of the high-temperature aging test equipment under different fault types into a trained data-driven fault diagnosis model to obtain the fault diagnosis results under different fault types; according to the set second 0 / 1 logic conversion rule, respectively converting the fault diagnosis results under different fault types into second 0 / 1 logic values; based on the second 0 / 1 logic values, obtaining virtual one-dimensional test vectors under different fault types, and adding the virtual one-dimensional test vectors under different fault types to the original D matrix to obtain an extended D matrix; wherein, the extended D matrix contains different fault types and their corresponding one-dimensional test vectors, and the one-dimensional test vectors are composed of entity one-dimensional test vectors and virtual one-dimensional test vectors.

8. A computing device, comprising: a memory and a processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the secondary power supply fault detection and diagnosis method of the high-temperature aging test equipment according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, the steps of the secondary power supply fault detection and diagnosis method of the high-temperature aging test equipment according to any one of claims 1 to 6 are implemented.

Images