Fault Detection and Diagnosis Method and Device for Isolated Switching Power Supply at the Front End of Aging Bench

Abstract

The embodiments of this specification provide a fault detection and diagnosis method and device for an isolated switching power supply at the front end of an aging platform. The fault detection and diagnosis method for the isolated switching power supply at the front end of the aging platform includes: obtaining multiple historical signal values corresponding to analog signals of the isolated switching power supply at the front end of the aging platform in multiple historical time series; determining signal parameters of the isolated switching power supply at the front end of the aging platform based on the multiple historical signal values; performing binning processing on the signal parameters, and respectively constructing feature vectors of the isolated switching power supply at the front end of the aging platform based on at least two bin datasets included in the binning processing results; constructing a training sample set based on the feature vectors and the state labels corresponding to the feature vectors, and training a support vector machine diagnosis model to be trained based on the training sample set. Among them, the trained support vector machine diagnosis model is used for fault detection and diagnosis of the isolated switching power supply at the front end of the aging platform.

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 for fault detection and diagnosis of a front-end isolated switching power supply of an aging station. Background Art

[0002] The integrated circuit high-temperature aging test bench accelerates various physical and chemical reaction processes inside the components by continuously applying a certain electrical stress to the components for a long time, prompting various potential faults inside the components to be exposed early, so as to eliminate early failure products and enable the electronic components to enter a period with low and relatively stable failure rates from the beginning of use. The current integrated circuit high-temperature aging test bench can achieve long-term monitoring of the test environment by means of increasing in-machine long-term monitoring, over-stress protection mechanisms, etc., and can respond in a timely manner when the machine fails, realizing after-sales maintenance based on fault data. However, due to the lack of active guarantee technology for integrated circuit high-temperature aging test benches at present, it is difficult to achieve the integrity of the test process and the consistency of the test environment stress, and it is extremely easy to cause major property losses such as the destruction of millions of test devices due to the forced interruption of the test process caused by machine failures, or the aging test is recognized as a failure test due to adverse effects such as the introduction of additional stress during the test period due to machine 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 is conducive to avoiding large-scale electronic system failures such as new energy vehicles, civil airliners, and energy storage power transmission stations that use the same integrated circuit due to integrated circuit failures.

[0003] The front-end isolated switching power supply provides the power required for the entire aging test bench to ensure that all parts of the test bench can work properly. As the core key equipment of the aging test bench, the front-end isolated switching power supply has a great impact on the overall reliability of the aging test bench. Once there are circuit degradation, failures or sudden faults, at best, the aging test bench stops the test and damages the object under test, and at worst, it causes voltage overload, leading to fires, resulting in major property losses and safety hazards. Therefore, it is very necessary to detect and diagnose the faults of the front-end isolated switching power supply of the aging test bench. And how to detect and diagnose the faults of the front-end isolated switching power supply of the aging test bench has become an urgent problem to be solved. Summary of the Invention

[0004] In view of this, the embodiments of this specification provide a method for fault detection and diagnosis of a front-end isolated switching power supply of an aging station. One or more embodiments of this specification also relate to a device for fault detection and diagnosis of a front-end isolated switching power supply of an aging station, 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 the present specification, a method for detecting and diagnosing faults of a front-end isolated switching power supply of an aging platform is provided, including:

[0006] Obtaining a plurality of historical signal values corresponding to the analog signal of the front-end isolated switching power supply of the aging platform at a plurality of historical time series;

[0007] Based on the plurality of historical signal values, determining signal parameters of the front-end isolated switching power supply of the aging platform;

[0008] Performing binning processing on the signal parameters, and respectively constructing feature vectors of the front-end isolated switching power supply of the aging platform based on at least two binning data sets included in the binning processing result;

[0009] Constructing a training sample set based on the feature vectors and the state labels corresponding to the feature vectors, and training a support vector machine diagnosis model to be trained based on the training sample set, wherein the trained support vector machine diagnosis model is used to detect and diagnose faults of the front-end isolated switching power supply of the aging platform.

[0010] Optionally, the step of respectively constructing the feature vectors of the front-end isolated switching power supply of the aging platform based on at least two binning data sets included in the binning processing result includes:

[0011] Performing feature extraction on the target binning data set based on the target signal parameters included in the target binning data set, and constructing the feature vectors of the front-end isolated switching power supply of the aging platform based on the feature extraction result, wherein the target binning data set is each of the at least two binning data sets.

[0012] Optionally, the step of performing feature extraction on the target binning data set based on the target signal parameters included in the target binning data set includes:

[0013] Performing feature extraction on the target binning data set based on the target signal parameters included in the target binning data set to obtain the average value, standard deviation, and peak value of the corresponding target signal parameters.

[0014] Optionally, the step of performing binning processing on the signal parameters includes:

[0015] Performing binning processing on the signal parameters by using the K-means binning algorithm.

[0016] Optionally, the training sample set includes a training sample subset and a test sample subset, and the state labels include a healthy state label and a fault state label;

[0017] Correspondingly, the step of training the support vector machine diagnosis model to be trained based on the training sample set includes:

[0018] Train the support vector machine diagnosis model to be trained through the training sample subset to generate a trained support vector machine diagnosis model;

[0019] Evaluate the support vector machine diagnosis model through the test sample subset and generate a scoring value corresponding to the support vector machine diagnosis model;

[0020] Iteratively update the support vector machine diagnosis model according to the scoring value until the scoring value of the support vector machine diagnosis model meets the scoring requirements or reaches a preset number of iterations.

[0021] Optionally, the binning process for the signal parameters includes:

[0022] Obtain reference information on the value range and reference information on the change law corresponding to the signal parameters;

[0023] Determine an abnormal value interval corresponding to the signal parameter according to the fault information of the front-end isolated switching power supply of the aging platform;

[0024] Determine a binning interval corresponding to the signal parameter based on the value range reference information, the change law reference information, and the abnormal value interval;

[0025] Perform binning on the signal parameters according to the binning interval.

[0026] Optionally, the front-end isolated switching power supply fault detection and diagnosis method of the aging platform further includes:

[0027] Obtain a plurality of real-time simulation monitoring signal values of the analog signal of the front-end isolated switching power supply of the aging platform;

[0028] Input the real-time simulation monitoring signal values into the support vector machine diagnosis model and obtain a target status label generated by the support vector machine diagnosis model through processing the real-time simulation monitoring signal values;

[0029] In the case where the target status label is a fault status label, trigger and execute the target protection strategy of the front-end isolated switching power supply of the aging platform.

[0030] According to the second aspect of the embodiments of the present specification, there is provided a front-end isolated switching power supply fault detection and diagnosis device for an aging platform, including:

[0031] An acquisition module configured to acquire a plurality of historical signal values corresponding to the analog signal of the front-end isolated switching power supply of the aging platform at a plurality of historical time series;

[0032] A determination module, configured to determine signal parameters of the front-end isolated switching power supply of the aging station based on the multiple historical signal values;

[0033] A construction module, configured to perform binning processing on the signal parameters, and respectively construct feature vectors of the front-end isolated switching power supply of the aging station based on at least two binning data sets included in the binning processing results;

[0034] A training module, configured to construct a training sample set based on the feature vectors and the state labels corresponding to the feature vectors, and train a support vector machine diagnosis model to be trained based on the training sample set, wherein the trained support vector machine diagnosis model is used to perform fault detection and diagnosis on the front-end isolated switching power supply of the aging station.

[0035] According to a third aspect of the embodiments of the present specification, a computing device is provided, including:

[0036] A memory and a processor;

[0037] 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 methods for fault detection and diagnosis of the front-end isolated switching power supply of the aging station.

[0038] According to a 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 methods for fault detection and diagnosis of the front-end isolated switching power supply of the aging station are implemented.

[0039] According to a fifth aspect of the embodiments of the present specification, a computer program is provided, wherein when the computer program is executed in a computer, the computer is made to execute the steps of the above-mentioned method for fault detection and diagnosis of the front-end isolated switching power supply of the aging station.

[0040] In the embodiments of this specification, the analog signal values corresponding to the analog quantity signals of the isolated switching power supply at the front end of the aging station in multiple historical time series are obtained; based on the multiple historical signal values, the signal parameters of the isolated switching power supply at the front end of the aging station are determined; the signal parameters are subjected to binning processing, and based on at least two binning data sets included in the binning processing results, the feature vectors of the isolated switching power supply at the front end of the aging station are constructed respectively; a training sample set is constructed based on the feature vectors and the state labels corresponding to the feature vectors, and the support vector machine diagnosis model to be trained is trained based on the training sample set, wherein the trained support vector machine diagnosis model is used to detect and diagnose faults of the isolated switching power supply at the front end of the aging station. In the embodiments of this specification, through binning processing, the signal parameters are grouped according to certain rules, making the grouped signal parameters more discrete, and then a training sample set is constructed through the grouped signal parameters and the support vector machine diagnosis model to be trained is trained, which can help enhance the stability of the trained support vector machine diagnosis model and help avoid overfitting of the trained support vector machine diagnosis model. Description of the Drawings

[0041] Figure 1 is a flowchart of a method for detecting and diagnosing faults of an isolated switching power supply at the front end of an aging station provided by an embodiment of this specification;

[0042] Figure 2 is a schematic structural diagram of a device for detecting and diagnosing faults of an isolated switching power supply at the front end of an aging station provided by an embodiment of this specification;

[0043] Figure 3 is a block diagram of the structure of a computing device provided by an embodiment of this specification. Detailed Embodiments

[0044] Many specific details are set forth in the following description in order 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 generalizations without departing from the connotation of this specification. Therefore, this specification is not limited by the specific embodiments disclosed below.

[0045] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, 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 indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more related listed items.

[0046] 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 information of the same type 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".

[0047] First, the noun terms involved in one or more embodiments of this specification are explained.

[0048] Generalized Sequential Pattern (GSP) Mining: The GSP sequence pattern algorithm is mainly used to process sequence data sets to find frequent sequences that meet the minimum support. The GSP algorithm uses a layer-by-layer mining method, uses a prefix tree structure to organize candidate sequences, and reduces the search space through pruning techniques to improve efficiency.

[0049] Currently, high-temperature aging products can achieve long-term monitoring of the test environment by means of increasing in-machine long-term monitoring, overstress protection mechanisms, etc., and can respond in a timely manner when the product fails, realizing after-sales maintenance based on failure data. However, due to the lack of active guarantee technology for high-temperature aging products, it is difficult for existing aging 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 destruction of millions of test devices due to the forced interruption of the test process caused by product failures, or the aging test being recognized as a failure test due to adverse effects such as additional stress introduced during the test 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 failures of integrated circuits, greatly reduce the failure rate of integrated circuits, and avoid large-scale electronic system failures and shutdowns such as new energy vehicles, civil airliners, and energy storage power transmission stations due to integrated circuit failures.

[0050] Therefore, the development of an intelligent guarantee system for test quality not only has important economic value, but also helps China's high-end test equipment to move to the forefront of the world.

[0051] Currently, it is difficult for active guarantee technology to adapt to high-temperature aging products. The key difficulties mainly include: (1) It is difficult to determine the health benchmark due to inconsistent test environments; (2) It is difficult to calculate the performance degradation trends of different levels due to the multi-structural levels of products; (3) It is difficult to construct a fault self-healing strategy due to the complex composition of fault sources.

[0052] In addition, during the aging test process, the front-end isolated switching power supply of the integrated circuit high-temperature aging test bench generates monitoring signals such as voltage and current. The directly collected monitoring signal data is large and contains a lot of redundant information.

[0053] In response to the above problems, facing the urgent need for autonomous guarantee of the performance of high-end test equipment in key fields, by accurately monitoring the characteristic parameters of high-temperature aging products, constructing a characteristic vector of the front-end isolated switching power supply of the aging bench based on the monitored characteristic parameters, training a model based on the characteristic vector and its corresponding labels, and using the trained model to detect and diagnose faults in the front-end isolated switching power supply of the aging bench to ensure the quality of the long-term test operation environment and meet the mass aging requirements of large-scale integrated circuits. The relevant technical system can be effectively extended to the same type of aging system to enhance the key scientific and technological strength of the integrated circuit testing industry.

[0054] In the embodiments of this specification, the performance parameters of the ripple voltage are first extracted from these monitoring data, and then the binned discretization method is used to process the extracted multi-performance parameters into a characteristic matrix with high information density. After binning, the values of the characteristics are more stable, and the model's tolerance for outliers is enhanced. Binning groups the data according to certain rules to make the data more discrete, enhancing the stability of the model and avoiding overfitting.

[0055] Based on the support vector machine diagnosis model, learn and classify the characteristic matrix, complete the fault detection and diagnosis of the primary switching power supply of the aging test bench, and combine the overvoltage protection and overcurrent protection mechanisms to achieve real-time response and ensure the normal progress of the aging test.

[0056] In this specification, a method for fault detection and diagnosis of the front-end isolated switching power supply of the aging bench is provided. This specification also relates to a device for fault detection and diagnosis of the front-end isolated switching power supply of the aging bench, 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.

[0057] Figure 1 The flowchart of a method for fault detection and diagnosis of the front-end isolated switching power supply of the aging bench provided according to an embodiment of this specification is shown, which specifically includes the following steps.

[0058] Step 102: Obtain multiple historical signal values corresponding to the analog signals of the front-end isolated switching power supply of the aging bench at multiple historical time series.

[0059] Specifically, the aging bench, i.e., the integrated circuit high-temperature aging test bench, accelerates various physical and chemical reaction processes inside the components by continuously applying a certain electrical stress to the components for a long time, prompting various potential faults inside the components to be exposed early, so as to eliminate early failure products and enable the electronic components to enter a period with low failure rate and relatively stable from the beginning of use. The front-end isolated switching power supply is used to provide the required power for the entire aging test bench.

[0060] Since the front-end isolated switching power supply of the aging test bench generates monitoring signals such as current and voltage during operation, therefore, the analog signals described in the embodiments of this specification can include output voltage signals and current accuracy signals.

[0061] In addition, the multiple historical signal values described in the embodiments of this specification can be the historical time-series data of the analog signals. Each historical time series can be composed of time points in the time period corresponding to each complete use process of the front-end isolated switching power supply.

[0062] In practical applications, the historical signal values corresponding to the analog signals of a switching power supply can be obtained through simulation. In the simulation environment, by setting the input parameters (mains input voltage, load conditions, etc.) and operating characteristics of the switching power supply, the signal values of its corresponding voltage and current signals can be generated and output. The simulation system can simulate a variety of actual operating conditions (different load changes, component aging states, etc.), so as to generate monitoring signal data covering normal operation and fault states.

[0063] The signal data generated by simulation has high controllability and diversity, and can provide reliable data support for performance parameter extraction and fault diagnosis.

[0064] Step 104: Based on the multiple historical signal values, determine the signal parameters of the front-end isolated switching power supply of the aging bench.

[0065] Specifically, on the basis that the aforementioned analog signal is a voltage signal, the signal parameter can be the ripple voltage.

[0066] The embodiments of this specification can analyze the voltage signals generated by simulation, extract the key performance parameter, i.e., the ripple voltage. The ripple voltage is used to evaluate the AC component in the DC output voltage and is an important indicator of the output quality. The AC part is retained by subtracting the DC component (average value).

[0067] Step 106: Perform binning processing on the signal parameters, and respectively construct the feature vectors of the front-end isolated switching power supply of the aging bench based on at least two binning data sets included in the binning processing results.

[0068] Specifically, the sequential pattern mining algorithm (GSP) can be used to identify regular temporal changes. By introducing temporal patterns, it is possible to identify fault patterns that gradually change over a long period of time, rather than relying solely on instantaneous data, thus improving the accuracy of fault detection. Based on the acquired historical signal values, the variation patterns of voltage and current are analyzed to identify the characteristics of the load. By analyzing indicators such as voltage and current fluctuations and instantaneous changes, the type of load, such as resistive load, inductive load, or capacitive load, is determined.

[0069] In an alternative embodiment, the binning process for the signal parameters includes:

[0070] The signal parameters are binned using the K-means binning algorithm.

[0071] Specifically, the embodiments of this specification adopt the K-means binning algorithm (K-means Binning) and bin the signal parameters according to the data distribution pattern of the signal parameters. Additionally, the signal parameters can also be binned with reference to the classification requirements of the support vector machine diagnosis model to ensure that the discrete data after binning can effectively reflect the differences between different operating states of the isolated switching power supply at the front end of the aging bench.

[0072] In practical applications, when binning the signal parameters using the K-means binning algorithm, specifically, the K-means model can be initialized first, and then the number of bins is selected as the number of clusters for K-means binning. Then, the sample data is used to train the K-means model, and the signal parameters are binned using the trained K-means model and the number of clusters, generating at least two corresponding binning intervals. The signal parameters included in one binning interval form a binned data set.

[0073] In another alternative embodiment, the binning process for the signal parameters includes:

[0074] Obtain the reference information on the value range and the reference information on the variation pattern corresponding to the signal parameters;

[0075] Based on the fault information of the isolated switching power supply at the front end of the aging bench, determine the abnormal value interval corresponding to the signal parameters;

[0076] Based on the value range reference information, the variation pattern reference information, and the abnormal value interval, determine the binning interval corresponding to the signal parameters;

[0077] Bin the signal parameters according to the binning interval.

[0078] Specifically, the basis for binning the signal parameters in the embodiments of this specification mainly refers to the actual value range (i.e., the value range reference information) and its variation law (i.e., the variation law reference information) of the signal parameters in the aging bench simulation data, and also combines the following considerations:

[0079] 1) Determine the normal interval of the signal parameters based on the parameter distribution range of the front-end isolated switching power supply in the normal operating state (healthy state) in the aging bench simulation data;

[0080] 2) Combine the influence of possible fault scenarios in the simulation process on the signal parameters to delimit the abnormal interval of the signal parameters to facilitate the distinction of fault types;

[0081] 3) Consider the distribution characteristics of the signal parameters and reasonably set the binning interval to reduce information loss.

[0082] Through the above binning discretization process, continuous performance parameters can be converted into discrete performance parameters, improving the comparability of data and reducing the complexity in the model training process. The interval setting and discretization results of binning provide efficient and accurate input features for the subsequent support vector machine diagnosis model to be trained.

[0083] In an optional implementation manner, constructing the feature vector of the front-end isolated switching power supply of the aging bench based on at least two binning data sets included in the binning processing results includes:

[0084] Based on the target signal parameters included in the target binning data set, perform feature extraction on the target binning data set, and construct the feature vector of the front-end isolated switching power supply of the aging bench based on the feature extraction results, where the target binning data set is each of the at least two binning data sets.

[0085] Further, the performing feature extraction on the target binning data set based on the target signal parameters included in the target binning data set includes:

[0086] Based on the target signal parameters included in the target binning data set, perform feature extraction on the target binning data set to obtain the average value, standard deviation, and peak value of the corresponding target signal parameters.

[0087] Specifically, the target binning data set is each of the at least two binning data sets, and the signal parameters included in the target binning data set are the target signal parameters.

[0088] Before performing feature extraction, median filtering can be applied to the ripple voltage to reduce the influence of noise on feature extraction.

[0089] In practical applications, based on the target signal parameters included in the target binned dataset, feature extraction is performed on the target binned dataset. Specifically, three features can be extracted from the filtered ripple voltage: the average value of the ripple voltage, the standard deviation of the ripple voltage, and the peak value of the ripple voltage. After extracting these feature parameters, a feature vector of the front-end isolated switching power supply of the aging platform can be constructed based on these feature parameters to form a feature sequence with high information density, providing a basis for subsequent analysis.

[0090] Step 108: Construct a training sample set based on the feature vector and the status label corresponding to the feature vector, and train the support vector machine diagnostic model to be trained based on the training sample set. After training, the support vector machine diagnostic model is used to detect and diagnose faults in the front-end isolated switching power supply of the aging platform.

[0091] Specifically, after the feature vector is constructed, a training sample set can be constructed based on the feature vector and the status label corresponding to the feature vector, and the support vector machine diagnostic model to be trained is trained based on the training sample set.

[0092] In the embodiments of this specification, the support vector machine diagnostic model to be trained can be a support vector machine diagnostic model optimized by particle swarm optimization based on GSP. Therefore, particle swarm optimization (PSO) can be used to optimize the hyperparameters of the support vector machine diagnostic model to be trained, enabling it to globally search for the optimal hyperparameter combination in a high-dimensional space and avoiding the problem of local optimal solutions.

[0093] In an optional implementation manner, the training sample set includes a training sample subset and a test sample subset, and the status label includes a healthy status label and a fault status label;

[0094] Correspondingly, the training of the support vector machine diagnostic model to be trained based on the training sample set includes:

[0095] Training the support vector machine diagnostic model to be trained through the training sample subset to generate a trained support vector machine diagnostic model;

[0096] Evaluating the support vector machine diagnostic model through the test sample subset and generating a score value corresponding to the support vector machine diagnostic model;

[0097] Iteratively updating the support vector machine diagnostic model according to the score value until the score value of the support vector machine diagnostic model meets the scoring requirements or reaches the preset number of iterations.

[0098] Specifically, when constructing a training sample set based on feature vectors and the corresponding state labels of the feature vectors in the embodiments of this specification, the training sample set can be divided into two parts. One part is the training sample subset, which is used to train the support vector machine diagnosis model to be trained; the other part is the test sample subset, which is used to test the trained support vector machine diagnosis model.

[0099] Based on this, the support vector machine diagnosis model to be trained is trained based on the training sample set. Specifically, the support vector machine diagnosis model to be trained is trained through the feature vectors and the corresponding state labels included in the training sample subset. The state labels can include two types, namely the healthy state label and the fault state label.

[0100] After training is completed, the feature vectors and the corresponding state labels included in the test sample subset can be input into the trained support vector machine diagnosis model, so as to process the input data through the support vector machine diagnosis model and generate corresponding output results. The output results can include the prediction results (healthy state or fault state) generated by the model for fault prediction of the front-end isolated switching power supply based on the input feature vectors, and can also include the score value corresponding to the output results.

[0101] Furthermore, after obtaining the score value, it can be determined whether the training iteration times of the support vector machine diagnosis model to be trained reach the preset iteration times, and it can be determined whether the score value is less than the preset score threshold; if the training iteration times of the support vector machine diagnosis model to be trained do not reach the preset iteration times, and the score value is less than the preset score threshold, then the support vector machine diagnosis model needs to be iteratively updated according to the score value until the score value of the support vector machine diagnosis model is greater than or equal to the score threshold, or the training iteration times reach the preset iteration times, then the training process is stopped to obtain the trained support vector machine diagnosis model.

[0102] In practical applications, a support vector machine (SVM) diagnosis model is trained using multiple groups of data generated by simulation (including monitoring signals under normal operation and fault states). The training data includes a feature matrix and state labels. Among them, the feature matrix is the multi-performance signal parameter data after binning and discretization; the state labels include the normal state (healthy state) marked as "0" and the fault state marked as "1". The SVM optimizes the classification boundary, learns the pattern relationship between different state features, and constructs an accurate fault diagnosis model.

[0103] During the training process, the model performance can be evaluated through cross-validation, and the kernel function parameters (RBF kernel and regularization parameter C) of the model can be optimized using grid search, so that the training results of the model are more accurate.

[0104] In an alternative embodiment, the fault detection and diagnosis method for the front-end isolated switching power supply of the aging platform further includes:

[0105] Obtain multiple real-time simulation monitoring signal values of the analog signal of the front-end isolated switching power supply of the aging platform;

[0106] Input the real-time simulation monitoring signal values into the support vector machine diagnosis model, and obtain the target status label generated by the support vector machine diagnosis model through processing the real-time simulation monitoring signal values;

[0107] When the target status label is a fault status label, trigger and execute the target protection strategy for the front-end isolated switching power supply of the aging platform.

[0108] Specifically, after the model training is completed, multiple real-time simulation monitoring signal values of the analog signal of the front-end isolated switching power supply of the aging platform can be obtained, and these multiple real-time simulation monitoring signal values are input into the support vector machine diagnosis model to process the multiple real-time simulation monitoring signal values through the support vector machine diagnosis model to generate corresponding target status labels. Among them, the target status label may be a healthy status label or a fault status label; and when the target status label is a fault status label, the target protection strategy for the front-end isolated switching power supply of the aging platform can be triggered and executed.

[0109] In practical applications, the obtained multiple real-time simulation monitoring signal values are input into the trained SVM diagnosis model for classification, and the model outputs the operation status category of the front-end isolated switching power supply and its confidence level. Among them, the normal state (value of 0) indicates that all performance parameters of the front-end isolated switching power supply are within the normal range; the fault state (value of 1) indicates that abnormal voltage parameters of the front-end isolated switching power supply are detected. When the operation status category output by the model is the fault state, the protection mechanism needs to be triggered according to the output result, and when the ripple voltage exceeds the standard, protection measures are started.

[0110] In addition, when it is determined that the front-end isolated switching power supply of the aging platform is in a degraded state according to the fault status label, the multi-distance metric fusion method with automatically optimized weights is used to quantitatively calculate the health status of the real-time simulation monitoring signal values, obtain the health degree of the degraded state at the current moment, and use the health degree of the degraded state at the current moment as the final fault detection result of the front-end isolated switching power supply of the aging platform.

[0111] In an alternative embodiment, using the multi-distance metric fusion method with automatically optimized weights to quantitatively calculate the health status of the real-time simulation monitoring signal values and obtain the health degree of the degraded state at the current moment includes:

[0112] Calculate the Mahalanobis distance, cosine similarity, and Manhattan distance between the real-time simulation monitoring signal values and the previously obtained healthy baseline data respectively;

[0113] Through multi-distance metric fusion with automatic weight optimization for the Mahalanobis distance, the cosine similarity, and the Manhattan distance, obtain the health degree of the degradation state at the current moment.

[0114] Among them, the healthy baseline data is constructed based on the historical signal values corresponding to all the fault state labels.

[0115] In an optional implementation manner, obtaining the health degree of the degradation state at the current moment through multi-distance metric fusion for the Mahalanobis distance, the cosine similarity, and the Manhattan distance includes:

[0116]

[0117] Among them, represents the health degree of the degradation state at the current moment, represents the Mahalanobis distance; represents the cosine similarity; represents the Manhattan distance; , and respectively represent weights, and + + = 1.

[0118] Alternatively, in the case where it is determined that the front-end isolated switching power supply of the aging platform is in a degraded state according to the fault prediction result, based on the probability space points after mapping of the multi-dimensional feature sequence and the healthy baseline in the residual convolutional neural network model, determine the state degradation curve of the front-end isolated switching power supply of the aging platform, where the state degradation curve can characterize the health state of the front-end isolated switching power supply of the aging platform; the multi-dimensional feature sequence is generated by extracting features from the historical signal values.

[0119] In an optional implementation manner, the residual convolutional neural network model is trained through the following method:

[0120] Obtain the training time-series degradation feature samples of the front-end isolated switching power supply of the aging platform; among them, each training time-series degradation feature sample is labeled with a degradation feature state;

[0121] Input the training time-series degradation feature samples into the previously established residual convolutional neural network model to obtain the first probability space points after mapping;

[0122] Discriminate the degradation starting point of the front-end isolated switching power supply of the aging platform according to the two-dimensional first probability space points, and generate the healthy baseline of the front-end isolated switching power supply of the aging platform.

[0123] In the actual implementation process, the training time series degradation feature samples of the isolated switching power supply at the front end of the aging table can be expressed as , the maximum number of iterations max_iter. Mark the degradation feature state corresponding to the training isolated switching power supply at the front end of the aging table , including the healthy state and the degradation state.

[0124] The embodiments of this specification provide an active guarantee technology for the integrated circuit high-temperature aging test bench to achieve the integrity of the test process and the consistency of the test environmental stress, and can reduce the significant property losses caused by the destruction of millions of test devices due to forced interruption during the test process caused by machine failures, or can reduce the adverse effects such as the introduction of additional stress during the test due to the degradation of machine performance, so as to avoid the aging test being recognized as a failure test, which is conducive to avoiding waste of resources. At the same time, ensuring the quality of the high-temperature aging test can avoid over-aging and under-aging failures of integrated circuits, greatly reduce the failure rate of integrated circuits, and is conducive to avoiding the failure shutdown of large-scale electronic systems such as new energy vehicles, civil airliners, and energy storage power transmission stations using the same integrated circuits due to integrated circuit failures.

[0125] The embodiments of this specification obtain multiple historical signal values corresponding to the analog signals of the isolated switching power supply at the front end of the aging table in multiple historical time series; based on the multiple historical signal values, determine the signal parameters of the isolated switching power supply at the front end of the aging table; perform binning processing on the signal parameters, and respectively construct feature vectors of the isolated switching power supply at the front end of the aging table based on at least two bin datasets included in the binning processing results; construct a training sample set based on the feature vectors and the state labels corresponding to the feature vectors, and train the support vector machine diagnosis model to be trained based on the training sample set, where the trained support vector machine diagnosis model is used to detect and diagnose faults of the isolated switching power supply at the front end of the aging table. The embodiments of this specification group the signal parameters according to certain rules through binning processing, making the grouped signal parameters more discrete, and then construct a training sample set through the grouped signal parameters and train the support vector machine diagnosis model to be trained, which can help enhance the stability of the trained support vector machine diagnosis model and avoid overfitting of the trained support vector machine diagnosis model.

[0126] Corresponding to the above method embodiments, this specification also provides embodiments of an apparatus for detecting and diagnosing faults of an isolated switching power supply at the front end of an aging table, Figure 2 showing a schematic structural diagram of an apparatus for detecting and diagnosing faults of an isolated switching power supply at the front end of an aging table provided by an embodiment of this specification. As Figure 2 shown, the apparatus includes:

[0127] An acquisition module 202, configured to acquire a plurality of historical signal values corresponding to analog signals of a front-end isolated switching power supply of an aging station at a plurality of historical time series;

[0128] A determination module 204, configured to determine signal parameters of the front-end isolated switching power supply of the aging station based on the plurality of historical signal values;

[0129] A construction module 206, configured to perform binning processing on the signal parameters, and respectively construct a feature vector of the front-end isolated switching power supply of the aging station based on at least two binning data sets included in the binning processing result;

[0130] A training module 208, configured to construct a training sample set based on the feature vector and a status label corresponding to the feature vector, and train a support vector machine diagnosis model to be trained based on the training sample set, wherein the trained support vector machine diagnosis model is used to perform fault detection and diagnosis on the front-end isolated switching power supply of the aging station.

[0131] Optionally, the construction module 206 is further configured to:

[0132] Extract features from the target binning data set based on target signal parameters included in the target binning data set, and construct a feature vector of the front-end isolated switching power supply of the aging station based on the feature extraction result, wherein the target binning data set is each of the at least two binning data sets.

[0133] Optionally, the construction module 206 is further configured to:

[0134] Extract features from the target binning data set based on target signal parameters included in the target binning data set, and obtain an average value, a standard deviation, and a peak value of the corresponding target signal parameters.

[0135] Optionally, the construction module 206 is further configured to:

[0136] Perform binning processing on the signal parameters by using a K-means binning algorithm.

[0137] Optionally, the training sample set includes a training sample subset and a test sample subset, and the status label includes a healthy status label and a fault status label;

[0138] Correspondingly, the training module 208 is further configured to:

[0139] Train the support vector machine diagnosis model to be trained by using the training sample subset to generate a trained support vector machine diagnosis model;

[0140] Evaluate the support vector machine diagnosis model using the subset of the test samples, and generate a scoring value corresponding to the support vector machine diagnosis model;

[0141] Iteratively update the support vector machine diagnosis model according to the scoring value until the scoring value of the support vector machine diagnosis model meets the scoring requirements or reaches the preset number of iterations.

[0142] Optionally, the construction module 206 is further configured to:

[0143] Obtain the reference information on the value range and the reference information on the variation law corresponding to the signal parameter;

[0144] Determine the abnormal value interval corresponding to the signal parameter according to the fault information of the isolated switching power supply at the front end of the aging platform;

[0145] Based on the reference information on the value range, the reference information on the variation law, and the abnormal value interval, determine the binning interval corresponding to the signal parameter;

[0146] Perform binning processing on the signal parameter according to the binning interval.

[0147] Optionally, the fault detection and diagnosis device for the isolated switching power supply at the front end of the aging platform further includes a processing module 210, which is configured to:

[0148] Obtain a plurality of real-time simulation monitoring signal values of the analog signal of the isolated switching power supply at the front end of the aging platform;

[0149] Input the real-time simulation monitoring signal values into the support vector machine diagnosis model, and obtain the target status label generated by the support vector machine diagnosis model through processing the real-time simulation monitoring signal values;

[0150] When the target status label is a fault status label, trigger and execute the target protection strategy for the isolated switching power supply at the front end of the aging platform.

[0151] The above is a schematic solution of a fault detection and diagnosis device for an isolated switching power supply at the front end of an aging platform in this embodiment. It should be noted that the technical solution of this fault detection and diagnosis device for an isolated switching power supply at the front end of an aging platform belongs to the same concept as the technical solution of the above-mentioned fault detection and diagnosis method for an isolated switching power supply at the front end of an aging platform. For the details not described in the technical solution of the fault detection and diagnosis device for an isolated switching power supply at the front end of an aging platform, reference can be made to the description of the technical solution of the above-mentioned fault detection and diagnosis method for an isolated switching power supply at the front end of an aging platform.

[0152] Figure 3FIG. 0 shows a structural block diagram of a computing device 300 provided according to an embodiment of this specification. The components of the computing device 300 include, but are not limited to, a memory 310 and a processor 320. The processor 320 is connected to the memory 310 via a bus 330, and a database 350 is used to store data.

[0153] The computing device 300 further includes an access device 340, which enables the computing device 300 to communicate via one or more networks 360. 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 340 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, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC) interface, and so on.

[0154] In an embodiment of this specification, the above components of the computing device 300 and Figure 3 other components not shown may also be connected to each other, for example, via a bus. It should be understood that Figure 3 the shown structural block diagram of the computing device is only for illustrative purposes and is not a limitation on the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0155] The computing device 300 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet computer, personal digital assistant, laptop computer, notebook computer, netbook, etc.), a mobile phone (e.g., smartphone), a wearable computing device (e.g., smartwatch, smart glasses, etc.) or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. The computing device 300 can also be a mobile or stationary server.

[0156] Wherein, the processor 320 is used to execute the following computer-executable instructions, and when the computer-executable instructions are executed by the processor, the steps of the above-mentioned aging bench front-end isolated switching power supply fault detection and diagnosis method are implemented.

[0157] The above is a schematic solution of a computing device in this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the above-mentioned aging bench front-end isolated switching power supply fault detection and diagnosis method belong to the same concept. For the details not described in the technical solution of the computing device, reference can be made to the description of the technical solution of the above-mentioned aging bench front-end isolated switching power supply fault detection and diagnosis method.

[0158] One embodiment of this specification also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the above-mentioned fault detection and diagnosis method for the isolated switching power supply at the front end of the aging station.

[0159] The above is a schematic solution of a computer-readable storage medium of this embodiment. It should be noted that the technical solution of this storage medium and the technical solution of the above-mentioned fault detection and diagnosis method for the isolated switching power supply at the front end of the aging station 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 fault detection and diagnosis method for the isolated switching power supply at the front end of the aging station.

[0160] One embodiment of this specification also provides a computer program, which, when executed in a computer, causes the computer to execute the steps of the above-mentioned fault detection and diagnosis method for the isolated switching power supply at the front end of the aging station.

[0161] The above is a schematic solution of a computer program of this embodiment. It should be noted that the technical solution of this computer program and the technical solution of the above-mentioned fault detection and diagnosis method for the isolated switching power supply at the front end of the aging station 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 fault detection and diagnosis method for the isolated switching power supply at the front end of the aging station.

[0162] 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.

[0163] 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, mobile hard disk, magnetic disk, optical disk, 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.

[0164] It should be noted that for the foregoing method embodiments, for the sake of simplicity of 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.

[0165] In the above embodiments, the descriptions of the respective 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.

[0166] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The alternative embodiments do not elaborate on all details and do not 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. These embodiments are selected and specifically described in this specification 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 fault detection and diagnosis method for the front-end isolated switching power supply of an aging platform, including: Obtain multiple historical signal values of the front-end isolated switching power supply of the aging platform, and based on the multiple historical signal values, determine the signal parameters of the front-end isolated switching power supply of the aging platform; wherein, the signal parameter is the ripple voltage; Perform binning processing on the signal parameters, and based on at least two binning data sets included in the binning processing result, respectively construct the feature vectors of the front-end isolated switching power supply of the aging platform; Construct a training sample set based on the feature vectors and their corresponding state labels, and train the support vector machine diagnosis model to be trained based on the training sample set; Obtain multiple real-time simulation monitoring signal values of the front-end isolated switching power supply of the aging platform, and input the real-time simulation monitoring signal values into the support vector machine diagnosis model to obtain the target state label; When the target state label is the degradation state, calculate the Mahalanobis distance, cosine similarity, and Manhattan distance between the real-time simulation monitoring signal values and the pre-obtained healthy baseline data respectively; through the multi-distance metric fusion with automatic weight optimization of the Mahalanobis distance, the cosine similarity, and the Manhattan distance, obtain the final fault detection result of the front-end isolated switching power supply of the aging platform, which includes: ； Among them, represents the final fault detection result of the isolated switching power supply at the front end of the aging table; represents the Mahalanobis distance; represents the cosine similarity; represents the Manhattan distance; , and respectively represent weights, and + + = 1.

2. The fault detection and diagnosis method for the front-end isolated switching power supply of the aging platform according to claim 1, wherein the constructing the feature vectors of the front-end isolated switching power supply of the aging platform respectively based on at least two binning data sets included in the binning processing result includes: Based on the target signal parameters included in the target binning data set, perform feature extraction on the target binning data set, and construct the feature vectors of the front-end isolated switching power supply of the aging platform based on the feature extraction result, wherein the target binning data set is each of the at least two binning data sets.

3. The fault detection and diagnosis method for the front-end isolated switching power supply of the aging platform according to claim 2, wherein the performing feature extraction on the target binning data set based on the target signal parameters included in the target binning data set includes: Based on the target signal parameters included in the target binning data set, perform feature extraction on the target binning data set to obtain the average value, standard deviation, and peak value of the corresponding target signal parameters.

4. The fault detection and diagnosis method for the front-end isolated switching power supply of the aging platform according to claim 1, wherein the performing binning processing on the signal parameters includes: Perform binning processing on the signal parameters through the K-means binning algorithm.

5. The fault detection and diagnosis method for the front-end isolated switching power supply of the aging platform according to claim 1, wherein the training sample set includes a training sample subset and a test sample subset, and the state labels include a healthy state label and a fault state label; Correspondingly, the training the support vector machine diagnosis model to be trained based on the training sample set includes: Train the support vector machine diagnosis model to be trained through the training sample subset to generate the trained support vector machine diagnosis model; Evaluate the support vector machine diagnosis model using the subset of test samples, and generate a score value corresponding to the support vector machine diagnosis model; Iteratively update the support vector machine diagnosis model according to the score value until the score value of the support vector machine diagnosis model meets the score requirement or reaches the preset number of iterations.

6. The method for detecting and diagnosing faults of the front-end isolated switching power supply of the aging table according to claim 1, wherein the binning process of the signal parameters includes: Obtain the reference information on the value range and the reference information on the variation law corresponding to the signal parameters; Determine the abnormal value interval corresponding to the signal parameters according to the fault information of the front-end isolated switching power supply of the aging table; Based on the reference information on the value range, the reference information on the variation law, and the abnormal value interval, determine the binning interval corresponding to the signal parameters; Perform binning processing on the signal parameters according to the binning interval.

7. The method for detecting and diagnosing faults of the front-end isolated switching power supply of the aging table according to claim 1, further comprising: When the target state label is a fault state label, trigger and execute the target protection strategy of the front-end isolated switching power supply of the aging table.

8. A device for detecting and diagnosing faults of the front-end isolated switching power supply of the aging table, comprising: An acquisition module configured to acquire a plurality of historical signal values of the front-end isolated switching power supply of the aging table; A determination module configured to determine the signal parameters of the front-end isolated switching power supply of the aging table based on the plurality of historical signal values; The signal parameter is the ripple voltage; A construction module configured to perform binning processing on the signal parameters, and respectively construct feature vectors of the front-end isolated switching power supply of the aging table based on at least two binning data sets included in the binning processing results; A training module configured to construct a training sample set based on the feature vectors and their corresponding state labels, and train a support vector machine diagnosis model to be trained based on the training sample set; The determination module is further configured to acquire a plurality of real-time simulation monitoring signal values of the front-end isolated switching power supply of the aging table, and input the real-time simulation monitoring signal values into the support vector machine diagnosis model to obtain a target state label; When the target state label is a degradation state, calculate the Mahalanobis distance, cosine similarity, and Manhattan distance between the real-time simulation monitoring signal values and the pre-acquired healthy baseline data respectively; through multi-distance metric fusion with automatic weight optimization of the Mahalanobis distance, the cosine similarity, and the Manhattan distance, obtain the final fault detection result of the front-end isolated switching power supply of the aging table, which includes: ； Among them, represents the final fault detection result of the isolated switching power supply at the front end of the aging table, represents the Mahalanobis distance; represents the cosine similarity; represents the Manhattan distance; , and respectively represent weights, and + + = 1.

9. 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 method for detecting and diagnosing faults of the front-end isolated switching power supply of the aging table according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps of the method for detecting and diagnosing faults of the front-end isolated switching power supply of the aging station according to any one of claims 1 to 7 are implemented.

Images