Method and device for evaluating health state of front-end isolated switching power supply of aging table

Abstract

The embodiment of the invention provides a method and device for evaluating the health state of a front-end isolated switching power supply of an aging table, and relates to the technical field of aging tests.The method comprises the steps that in the aging test process, the front-end isolated switching power supply of the aging testbed can generate monitoring signals such as voltage and current; a directly measured monitoring signal is large in data size and contains a large amount of redundant information, firstly, feature extraction is conducted on directly measured data, and a feature sequence with high information density is obtained; further spatial mapping is carried out on the feature sequence by using a neighbor component analysis algorithm, and automatic segmentation of the health state and the degradation state of the new test sample is realized; and performing quantitative calculation of the health state on the mapped feature sequence by using a multi-distance measurement fusion algorithm with automatic weight optimization, and completing health assessment of the front-end isolated switching power supply of the aging test stand.

Description

Technical Field

[0001] The embodiments of the present specification relate to the technical field of lossless reliability screening of integrated circuits, and more particularly to a method and device for evaluating the health status of an isolated switching power supply at the front end of a burn-in station. Background Art

[0002] The aging test bench simulates various actual use environments and working conditions, so that the product undergoes state changes similar to those after long-term use in a short period of time. Aging tests usually require the test equipment to be powered on for a long time to simulate the aging process of the product in actual use. Among them, the front-end isolated switching power supply can convert the input AC power into stable DC power, providing continuous and stable power supply for many tested devices on the aging test bench. Existing high-temperature aging products can achieve long-term monitoring of the test environment by adding long-term monitoring in the machine and overstress protection mechanisms. When the product fails, it can respond in time and realize post-maintenance based on fault data. However, due to the lack of active protection technology for high-temperature aging products, the existing aging products make it difficult for related products to achieve the integrity of the test process and the consistency of the test environment stress. It is very easy to cause the test process to be interrupted due to product failure, resulting in major property losses such as damage to millions of test devices, or the adverse effects such as the introduction of additional stress during the test due to product performance degradation, resulting in the aging test being identified as a failure test, 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 in integrated circuits, significantly reduce the failure rate of integrated circuits, and avoid large-scale electronic system failures and shutdowns such as new energy vehicles, civil airliners, energy storage and transmission stations due to integrated circuit failures.

[0003] During the aging test, if the front-end isolated switching power supply fails, such as the output voltage suddenly drops or disappears, the aging test will be unexpectedly interrupted. During the use of the aging test bench, problems such as capacitor aging and switch tube performance degradation will gradually occur. Therefore, how to ensure the healthy state 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, an embodiment of this specification provides a method for evaluating the health status of a front-end isolated switching power supply of a burn-in station. One or more embodiments of this specification also relate to a device for evaluating the health status of a front-end isolated switching power supply of a burn-in 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 a first aspect of an embodiment of this specification, a method for evaluating the health status of a front-end isolated switching power supply of a burn-in station is provided, comprising: Acquire the monitoring data of the isolated switching power supply at the front end of the aging test bench at the current moment, and obtain feature data with high information density by extracting features from the monitoring data; By inputting the high-information-density feature data into a trained NCA (Neighborhood Components Analysis) algorithm model, a preliminary health status assessment result of the front-end isolated switching power supply of the aging test bench at the current moment is obtained; If the preliminary health status assessment result is a degraded state, the health status of the high information density feature data is quantitatively calculated using a multi-distance metric fusion method with automatic weight optimization to obtain the health degree of the degraded state at the current moment, and the health degree of the degraded state at the current moment is used as the final health status assessment result of the isolated switching power supply at the front end of the aging test bench.

[0006] Preferably, it also includes: If the evaluation status result is a healthy status, the healthy status is used as the final healthy status evaluation result of the front-end isolated switching power supply of the aging test bench.

[0007] Preferably, it also includes: Acquire first historical monitoring data and second historical monitoring data of the full life cycle degradation cycle of the front-end isolated switching power supply of the aging test bench, and use the first historical monitoring data as a training set and the second historical monitoring data as a test set; By performing feature extraction on the historical monitoring data in the training set, first historical feature data with high information density is obtained; Determine the state label corresponding to each moment of the first high information density historical feature data; wherein the state label includes a healthy state label and a degraded state label; An NCA algorithm model is constructed, and the historical feature data with the first high information density is used as the model input. The state label corresponding to each moment of the historical feature data with the first high information density is used as the model output to train and learn the NCA algorithm model, so as to obtain a trained NCA algorithm model and an optimized mapping matrix.

[0008] Preferably, it also includes: By performing feature extraction on the historical monitoring data in the test set, historical feature data with the second highest information density is obtained; By inputting the historical feature data with the second highest information density into the trained NCA algorithm model, the state label corresponding to each moment of the historical feature data with the second highest information density is obtained; Extracting all healthy state labels from the state labels corresponding to each moment of the historical feature data with the second highest information density, and discarding all degraded state labels; Build health baseline data based on the historical feature data corresponding to all health status tags.

[0009] Preferably, the health status of the high information density feature data is quantitatively calculated using a multi-distance metric fusion method with automatic weight optimization to obtain the health status of the degradation state at the current moment, including: respectively calculating the Mahalanobis distance, cosine similarity and Manhattan distance between the feature data with high information density and the health baseline data; The health of the degradation state at the current moment is obtained by fusing the multi-distance metrics by automatically optimizing the weights of the Mahalanobis distance, the cosine similarity and the Manhattan distance.

[0010] Preferably, calculating the Mahalanobis distance between the high information density feature data and the health baseline data comprises: Based on the optimized mapping matrix, the Mahalanobis distance between the feature data with high information density and the healthy baseline data is calculated.

[0011] Preferably, by fusing the multi-distance metrics with automatic weight optimization of the Mahalanobis distance, the cosine similarity and the Manhattan distance, the health of the degradation state at the current moment is obtained, which includes:

[0012] in, Indicates the health of the degradation state at the current moment, represents the Mahalanobis distance; represents cosine similarity; represents the Manhattan distance; , and denote weights, and + + =1; Among them, by measuring the monotonicity and smoothness of the health curve in the health state of the current degradation state, the monotonicity loss and the smoothness loss are calculated respectively, and the monotonicity loss and the smoothness loss are used as objective functions. The multi-objective genetic algorithm is used to optimize the fusion weights of multiple distance metrics to obtain the best weight combination.

[0013] Preferably, after obtaining the final health curve, the monotonicity and smoothness are evaluated, the monotonicity loss and the smoothness loss are calculated, and the monotonicity loss and the smoothness loss are used as the objective function to perform the multi-objective genetic algorithm. , and Seek excellence.

[0014] According to a second aspect of the embodiments of this specification, a device for evaluating the health status of a front-end isolated switching power supply of a burn-in station is provided, comprising: The feature extraction module is configured to obtain the monitoring data of the isolated switching power supply at the front end of the aging test bench at the current moment, and obtain feature data with high information density by extracting features from the monitoring data; A preliminary evaluation module is configured to obtain a preliminary health status evaluation result of the front-end isolated switching power supply of the aging test bench at the current moment by inputting the high information density feature data into a trained neighbor component analysis (NCA) algorithm model; The final evaluation module is configured to, if the preliminary health status evaluation result is a degraded state, use a multi-distance metric fusion method with automatic weight optimization to quantitatively calculate the health status of the high information density feature data, obtain the health status of the degraded state at the current moment, and use the health status of the degraded state at the current moment as the final health status evaluation result of the front-end isolated switching power supply of the aging test bench.

[0015] According to a third aspect of an embodiment of this specification, a computing device is provided, including: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement any step of the method for evaluating the health status of the front-end isolated switching power supply of the aging station.

[0016] According to a fourth aspect of the embodiments of this 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 evaluating the health status of a front-end isolated switching power supply of a burn-in station are implemented.

[0017] According to a fifth aspect of the embodiments of this specification, a computer program is provided, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station.

[0018] The embodiment of this specification obtains the monitoring data of the isolated switching power supply at the front end of the aging test bench at the current moment, and obtains the feature data with high information density by extracting the features of the monitoring data; the preliminary health status assessment result of the isolated switching power supply at the front end of the aging test bench at the current moment is obtained by inputting the feature data with high information density into the trained nearest component analysis NCA algorithm model; if the preliminary health status assessment result is a degraded state, the health status of the feature data with high information density is quantitatively calculated using the multi-distance metric fusion method with automatic weight optimization to obtain the health degree of the degraded state at the current moment, and the health degree of the degraded state at the current moment is used as the final health status assessment result of the isolated switching power supply at the front end of the aging test bench. By performing a health assessment on the isolated switching power supply at the front end of the aging test bench, the potential hidden dangers inside the power supply can be discovered in advance, and repairs or replacements can be carried out in time to avoid test interruptions, so as to ensure that the aging test can proceed smoothly as planned. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of a method for evaluating the health status of a front-end isolated switching power supply of a burn-in station provided by an embodiment of this specification; Figure 2 It is a schematic diagram of a health status assessment process of a front-end isolated switching power supply of a burn-in station based on multi-distance metric fusion based on nearest component analysis and automatic weight optimization provided by an embodiment of this specification; Figure 3 It is a flow chart of a method for evaluating the health status of a front-end isolated switching power supply of a burn-in station provided by an embodiment of this specification; Figure 4 It is a structural schematic diagram of a health status assessment device for a front-end isolated switching power supply of a burn-in station provided by an embodiment of this specification; Figure 5 It is a structural block diagram of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION

[0020] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0021] 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 of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. 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 associated listed items.

[0022] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this 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 "at the time of" or "when" or "in response to determining".

[0023] In this specification, a method for evaluating the health status of an isolated switching power supply at the front end of a burn-in station is provided. This specification also relates to an apparatus for evaluating the health status of an isolated switching power supply at the front end of a burn-in station, a computing device, a computer-readable storage medium, and a computer program, which are described in detail one by one in the following embodiments.

[0024] Figure 1 A flowchart of a method for evaluating the health status of a front-end isolated switching power supply of a burn-in station according to an embodiment of the present specification is shown, which specifically includes the following steps.

[0025] Step 101: Acquire monitoring data of the isolated switching power supply at the front end of the aging test bench at the current moment, and obtain feature data with high information density by extracting features from the monitoring data; Step 102: obtaining a preliminary health status assessment result of the front-end isolated switching power supply of the aging test bench at the current moment by inputting the high information density feature data into a trained neighbor component analysis (NCA) algorithm model; Step 103: If the preliminary health status assessment result is a degraded state, a multi-distance metric fusion method with automatic weight optimization is used to quantitatively calculate the health status of the high information density feature data to obtain the health degree of the degraded state at the current moment, and the health degree of the degraded state at the current moment is used as the final health status assessment result of the front-end isolated switching power supply of the aging test bench.

[0026] In an optional implementation, the method for evaluating the health status of the front-end isolated switching power supply of the aging station further includes: If the evaluation status result is a healthy status, the healthy status is used as the final healthy status evaluation result of the front-end isolated switching power supply of the aging test bench.

[0027] In an optional implementation, the method for evaluating the health status of the front-end isolated switching power supply of the aging station further includes: Acquire first historical monitoring data and second historical monitoring data of the full life cycle degradation cycle of the front-end isolated switching power supply of the aging test bench, and use the first historical monitoring data as a training set and the second historical monitoring data as a test set; By performing feature extraction on the historical monitoring data in the training set, first historical feature data with high information density is obtained; Determine the state label corresponding to each moment of the first high information density historical feature data; wherein the state label includes a healthy state label and a degraded state label; An NCA algorithm model is constructed, and the historical feature data with the first high information density is used as the model input. The state label corresponding to each moment of the historical feature data with the first high information density is used as the model output to train and learn the NCA algorithm model, so as to obtain a trained NCA algorithm model and an optimized mapping matrix.

[0028] In an optional implementation, the method for evaluating the health status of the front-end isolated switching power supply of the aging station further includes: By performing feature extraction on the historical monitoring data in the test set, historical feature data with the second highest information density is obtained; By inputting the historical feature data with the second highest information density into the trained NCA algorithm model, the state label corresponding to each moment of the historical feature data with the second highest information density is obtained; Extracting all healthy state labels from the state labels corresponding to each moment of the historical feature data with the second highest information density, and discarding all degraded state labels; Build health baseline data based on the historical feature data corresponding to all health status tags.

[0029] In an optional implementation, a multi-distance metric fusion method with automatic weight optimization is used to perform quantitative calculation of the health status of the feature data with high information density, and the health status of the degradation state at the current moment is obtained, which includes: respectively calculating the Mahalanobis distance, cosine similarity and Manhattan distance between the feature data with high information density and the health baseline data; The health of the degradation state at the current moment is obtained by fusing the multi-distance metrics by automatically optimizing the weights of the Mahalanobis distance, the cosine similarity and the Manhattan distance.

[0030] In an optional implementation, calculating the Mahalanobis distance between the high information density feature data and the health baseline data includes: Based on the optimized mapping matrix, the Mahalanobis distance between the feature data with high information density and the healthy baseline data is calculated.

[0031] In an optional implementation, by performing multi-distance metric fusion on the Mahalanobis distance, the cosine similarity and the Manhattan distance, the health of the degradation state at the current moment is obtained, which includes:

[0032] in, Indicates the health of the degradation state at the current moment, represents the Mahalanobis distance; represents cosine similarity; represents the Manhattan distance; , and denote weights, and + + =1; Among them, by measuring the monotonicity and smoothness of the health curve in the health state of the current degradation state, the monotonicity loss and the smoothness loss are calculated respectively, and the monotonicity loss and the smoothness loss are used as objective functions. The multi-objective genetic algorithm is used to optimize the fusion weights of multiple distance metrics to obtain the best weight combination.

[0033] The embodiment of this specification obtains the monitoring data of the isolated switching power supply at the front end of the aging test bench at the current moment, and obtains the feature data with high information density by extracting the features of the monitoring data; the preliminary health status assessment result of the isolated switching power supply at the front end of the aging test bench at the current moment is obtained by inputting the feature data with high information density into the trained nearest component analysis NCA algorithm model; if the preliminary health status assessment result is a degraded state, the health status of the feature data with high information density is quantitatively calculated using the multi-distance metric fusion method with automatic weight optimization to obtain the health degree of the degraded state at the current moment, and the health degree of the degraded state at the current moment is used as the final health status assessment result of the isolated switching power supply at the front end of the aging test bench. By performing a health assessment on the isolated switching power supply at the front end of the aging test bench, the potential hidden dangers inside the power supply can be discovered in advance, and repairs or replacements can be carried out in time to avoid test interruptions, so as to ensure that the aging test can proceed smoothly as planned.

[0034] Figure 2A schematic diagram of a health status assessment process of a front-end isolated switching power supply of an aging station based on multi-distance metric fusion based on neighbor component analysis and automatic weight optimization provided by an embodiment of this specification is shown. Figure 2 As shown in the figure, during the aging test, the current, voltage and other signals of the isolated switching power supply at the front end of the test bench are monitored by sensors to obtain large-scale monitoring data for the status evaluation of the switching power supply. Since the amount of data of the directly measured monitoring signal is large and contains a lot of redundant information, it is necessary to first extract features from the direct monitoring data to obtain a feature sequence with high information density.

[0035] Based on the extracted feature sequence, the health assessment process is further realized: the nearest component analysis (NCA) metric learning algorithm is used to realize the adaptive state division and distance measurement of feature data. The main goal of the NCA algorithm is to learn a Mahalanobis distance The NCA algorithm considers the neighbor relationship of feature data sample points. ,in Represents the i-th feature sample. NCA will calculate each feature sample With other characteristic samples The distance between them is calculated through a transformed space, and the calculation formula is as follows:

[0036] The optimization mapping matrix L is a learnable semi-positive metric matrix. The core of the NCA algorithm is to adjust the optimization mapping matrix L through training, and map the feature samples to a new space so that the distance between samples of the same type is as small as possible after mapping, and the distance between samples of different types is as large as possible. This process is completed by maximizing the probability of determining the same type of feature samples as neighbors through the gradient optimization algorithm.

[0037] Compared with the health assessment method based on traditional distance metrics, it is necessary to manually set the health baseline to construct health indicators, which has certain instability. The NCA metric learning algorithm learns to optimize the mapping matrix L through the above-mentioned intra-class clustering and inter-class separation training method, which can effectively separate the feature samples and map them to the new space, so that the healthy feature samples and the degraded feature samples show obvious clustering and separation in the new space, and automatically cut and determine the health baseline in this process.

[0038] The health assessment based on the NCA metric learning algorithm can be trained and learned according to the characteristics of the samples, and can perform adaptive recognition and mapping of features with strong adaptive ability. During the training process, the health and degradation states of the training feature samples are set, and the mapping matrix is ​​obtained through training. Then, it is applied to the test feature samples during the test process to map the test data and separate the states.

[0039] Specifically, during the training phase of the NCA metric learning algorithm model, training data with health status labels are input , h = 0 or 1; is the feature sample, for The corresponding labels, 0 and 1 represent the healthy state and the degraded state respectively, and n is the number of feature samples obtained by sliding window resampling of the entire degradation curve. In this process, the NCA model continuously optimizes the mapping matrix of the input data to the low-dimensional feature space, strengthens the classification ability of the healthy state data and the degraded state data, and enables the NCA model to automatically segment the healthy baseline of the test data; in the test phase of the algorithm model, the test data is input The NCA model can optimize the learned feature mapping matrix according to the training process, map the healthy baseline data and the degraded data to a space with a larger inter-class distance, and easily cut the healthy baseline of the test data through the unsupervised clustering algorithm to obtain the test data set. , represents the test data set of all health labels, Represents a test data set of all degraded labels to avoid the blindness of manually segmenting healthy baselines.

[0040] In order to further realize the health status assessment of the front-end isolated switching power supply at the current moment, the multi-distance metric fusion method with automatic weight optimization is used to comprehensively consider the Mahalanobis distance, cosine similarity and Manhattan distance between sample data and perform weighted fusion to obtain the health status at the current moment. At the same time, the health degradation curve up to that moment is output, providing reliable data support for further remaining service life prediction research of the front-end isolated switching power supply. Among them, the Mahalanobis distance The calculation formula is as follows:

[0041] For two n-dimensional space vectors and , the calculation formula of its cosine similarity is as follows:

[0042] The calculation formula of Manhattan distance is as follows:

[0043] On this basis, multiple distance metrics fusion with automatic weight optimization is further performed, and the final distance metric result calculation process is as follows:

[0044] in, , and denote weights, and + + =1.

[0045] Furthermore, by measuring the monotonicity and smoothness of the health curve in the health state of the current degradation state, the monotonicity loss and the smoothness loss are calculated respectively, and the monotonicity loss and the smoothness loss are used as objective functions. The multi-objective genetic algorithm is used to optimize the fusion weights of multiple distance metrics to obtain the best weight combination.

[0046] It should be noted that the evaluation indicators include the monotonicity and smoothness of the health curve, and the monotonicity loss and smoothness loss are calculated. The health status of the equipment gradually degrades over time. Assume that the health curve is ,but: Monotonicity loss:

[0047] Smoothness loss:

[0048] Furthermore, the pseudo code of the multi-distance metric fusion health assessment algorithm with NCA and automatic weight optimization is shown in Table 1: Table 1: Pseudo code of the multi-distance metric fusion health assessment algorithm with NCA and automatic weight optimization

[0049] Furthermore, the automatic optimization of weights is reflected in the consistency evaluation of the degradation trend of the health curve, and finally the fusion weights are optimized based on the feedback of the evaluation results.

[0050] Figure 3 1 is a flow chart of a method for evaluating the health status of a front-end isolated switching power supply of a burn-in station provided by an embodiment of the present specification. Figure 3 As shown, including: Step S301: pre-processing of voltage signal of isolated switching power supply at the front end of the aging test bench; The health assessment case of the front-end isolated switching power supply of the aging test bench uses the voltage data during the degradation process of the front-end isolated switching power supply. In this case, two voltage curves that have completed the full life degradation cycle are used for training and testing respectively. First, the training curve is normalized to the maximum and minimum, and the test curve is normalized in the same way using the normalization parameters of the training curve. Then, the two voltage degradation data are resampled by sliding window to achieve sample cutting. The window size of the sliding window is 10, and the sliding window step is 10. Two sets of resampled voltage degradation data sets are obtained, which are used for training the nearest component analysis algorithm and testing the health assessment method respectively. The state of the front-end isolated switching power supply of the aging test bench can be divided into a healthy state and a degraded state. The training data set is marked according to the state of the sample as shown in Table 2.

[0051] Table 2: Training sample labeling

[0052] Step S302: training the nearest component analysis model; The healthy samples and degraded samples of the training data set are used as the input of the NCA metric learning model, and the corresponding calibration state feature labels are used as output to train the metric learning model. The maximum number of iterations is set to 100, and the cross-validation method is used to adjust the parameters, and the random state random_state and learning rate learning_rate of the algorithm model are continuously optimized to achieve the best training effect and accuracy. Table 3 below shows the NCA model parameters.

[0053] Table 3: NCA metric learning model parameters

[0054] Based on the trained NCA metric learning model and the learned feature matrix L, the voltage samples of the front-end isolated switching power supply can be adaptively mapped to the new distribution space, so that the health and degradation features are completely separated, the data aliasing of different states is reduced, and the calculation of subsequent health indicators is facilitated.

[0055] Step S303: Test sample health assessment.

[0056] Input the test sample into the trained NCA model to get whether the test sample is in a healthy state or a degraded state. Perform mean processing on the samples in a healthy state to construct a healthy baseline, and calculate the health index for the samples in a degraded state. Calculate the Mahalanobis distance between the degraded sample and the healthy baseline according to the time sequence. , cosine similarity Distance to Manhattan , where the learnable feature matrix used in the Mahalanobis distance calculation is L obtained in step 302. Considering the normalization of the data, the difference in the degraded features is mainly reflected in the inclination angle. The weight distribution in the final distance measurement result formula is shown in Table 4 below.

[0057] Table 4: Initial weight allocation for multi-distance metric fusion with automatic weight optimization

[0058] An embodiment of the present application also provides a method for extracting health features of a primary (level 1) switching power supply of a burn-in station, which is used to extract features from monitoring data during a health status assessment process of a primary (level 1) switching power supply of a burn-in station.

[0059] Furthermore, a flow chart of a method for extracting health characteristics of a primary switching power supply of a burn-in station according to an embodiment of the present application specifically includes the following steps.

[0060] Step 402: Acquire multiple historical signal values ​​of the analog signal of the primary switching power supply of the aging station.

[0061] Specifically, the aging bench is an integrated circuit high-temperature aging test bench, which accelerates various physical and chemical reaction processes inside the components by continuously applying a certain amount of electrical stress to the components for a long time, prompting various potential faults inside the components to be exposed early, thereby eliminating early failure products and making the electronic components enter a period of low failure rate and relative stability from the beginning of use. The primary switching power supply is used to provide the required power for the entire aging test bench.

[0062] Since the first-level switching power supply of the aging test bench will generate monitoring signals such as current and voltage during operation, the analog signal described in the embodiment of this specification may include voltage signals and current signals, and the current signal may include capacitor current signals and inductor current signals.

[0063] In addition, the multiple historical signal values ​​described in the embodiments of this specification may be historical time series data of analog signals.

[0064] Step 404: Encode the multiple historical signal values ​​through the Transformer model encoder in the multi-layer model encoder to obtain self-attention features in the multiple historical signal values, wherein the multi-layer model encoder is a neural network structure that integrates context features and is constructed by the Transformer model encoder and the stacked autoencoder, including an input layer, an intermediate layer, and an output layer, and the intermediate layer includes a sequentially connected Transformer model encoder structure and a stacked autoencoder.

[0065] In an optional embodiment, the multi-layer model encoder further includes a multi-head attention layer, a frequency domain feature extraction layer and a fusion layer, wherein the frequency domain feature extraction layer is used to extract frequency domain features, and the fusion layer is used to fuse the data features extracted by the Transformer model encoder and the frequency domain feature extraction layer; Correspondingly, the method also includes: obtaining the transformation result generated by the multi-head attention layer performing linear transformation on the multiple historical signal values; encoding the transformation result through the Transformer model encoder to generate self-attention features in the multiple historical signal values; performing frequency domain feature extraction on the transformation result through the frequency domain feature extraction layer to obtain the frequency domain features in the multiple historical signal values ​​output by the frequency domain feature extraction layer; fusing the self-attention features and the frequency domain features through the fusion layer to obtain the target fusion features output by the fusion layer.

[0066] Among them, fusing the self-attention feature and the frequency domain feature through the fusion layer to obtain the target fusion feature output by the fusion layer includes: splicing the self-attention feature and the frequency domain feature through the fusion layer to obtain the target fusion feature output by the fusion layer.

[0067] Specifically, the multi-layer model encoder is a multi-layer Transformer model encoder, which includes an input layer, an intermediate layer and an output layer. The intermediate layer includes a sequentially connected Transformer model encoder structure and a stacked autoencoder, wherein the Transformer model encoder includes a multi-head attention sublayer, a fully connected layer and a feedforward neural network; in addition, the multi-layer Transformer model encoder also includes a frequency domain feature extraction layer and a fusion layer. The multi-head attention sublayer is used to extract multi-head attention features in historical time series data; the frequency domain feature extraction layer is used to extract frequency domain features in historical time series data; the fusion layer is used to fuse the multi-head attention features extracted by the multi-head attention sublayer with the frequency domain features extracted by the frequency domain feature extraction layer, so as to achieve the purpose of fully extracting features.

[0068] In an optional embodiment, after obtaining multiple historical signal values ​​of the analog signal of the first-level switching power supply of the aging station, it also includes: performing sliding window cutting on the multiple historical signal values ​​according to the first window width and the first step length to generate a first sliding window cutting result, and determining an initial sample data set based on the first sliding window cutting result; normalizing the initial sample data set to generate a target sample data set; accordingly, encoding the multiple historical signal values ​​through the Transformer model encoder in the multi-layer model encoder includes: encoding the target sample data set through the Transformer model encoder in the multi-layer model encoder.

[0069] The embodiment of this specification provides a feature extraction processing flow, which includes the following steps.

[0070] Step 502: Obtain fault prediction data of the primary switching power supply of the aging station.

[0071] Step 504: Perform comprehensive preprocessing on the fault data.

[0072] Step 506: Perform sliding window cutting on the preprocessing result according to the first window width and the first step length to generate a first sliding window cutting result, and determine an initial sample data set based on the first sliding window cutting result.

[0073] Step 508: normalize the initial sample data set to generate a target sample data set.

[0074] Step 510: construct a training data set and a test data set based on the target sample data set.

[0075] Step 512: Perform multi-head attention feature extraction on the training dataset through the Transformer model encoder.

[0076] Step 514: extract frequency domain features from the training data set.

[0077] Step 516: Perform deep feature fusion on the spliced ​​features through a stacked autoencoder to obtain the health features corresponding to the primary switching power supply of the aging station.

[0078] Step 518: Output health characteristics.

[0079] To obtain the fault prediction data of the first-level switching power supply of the aging station, the voltage and current signals of the first-level switching power supply in the high-temperature dynamic aging detection system of the integrated circuit can be monitored by sensors to obtain the original fault prediction data of the first-level switching power supply of the aging station, that is, the historical time series data of the voltage and current signals.

[0080] For comprehensive preprocessing of fault data, the collected original fault prediction data can be sent to the fault data preprocessing unit for comprehensive processing to obtain a training data set and a test data set. This can be achieved through the following steps: Step 602: Perform sliding window cutting on the historical time series data to construct an initial sample data set.

[0081] Specifically, the historical time series data of any level of switching power supply collected by the sensor is X, , perform sliding window cutting on X to generate the corresponding initial sample data set. When the first window width is W and the first step length is S, the number of samples generated by cutting is:

[0082] The corresponding initial sample data set is generated as ,right Each sample in Take the length The data is used as training data, and W- The data of length is used as the prediction data corresponding to this section of training data.

[0083] Step 604: Perform maximum and minimum value normalization processing on the training data set.

[0084] In order to improve the data expression ability and speed up the convergence of subsequent model training, it is necessary to normalize the training data set, mainly by scaling the amplitude of the original parameters through the maximum and minimum normalization method to complete the linear transformation of the data. , the normalization process is realized by the following formula to obtain the normalized sample data set .

[0085]

[0086] Step 608: Construct a training data set and a test data set.

[0087] The first r% of the data are selected from all the data as the training data set, and the remaining data are used as the test data set to verify the model prediction performance. Generally speaking, r is generally 60-80, and r in the embodiment of this specification is 70.

[0088] For multi-head attention feature extraction of processed fault data based on the Transformer model encoder, the training data set obtained after processing by the comprehensive data preprocessing module can be sent to the Transformer model encoder and the frequency domain feature extraction layer respectively to obtain self-attention features and frequency domain features.

[0089] Another feature extraction process provided in the embodiments of this specification specifically includes the following steps.

[0090] Step 702: Obtain a training data set.

[0091] Step 704: Construct a Transformer model encoder.

[0092] Step 706: Train and construct the Transformer model encoder.

[0093] Step 708: Perform self-attention feature extraction on the training data set based on the trained Transformer model encoder.

[0094] Step 710: Perform sliding window cutting on the normalized target sample data set.

[0095] Step 712: Extract at least two initial frequency domain features based on the sliding window cutting result.

[0096] Step 714: normalize at least two initial frequency domain features.

[0097] Step 716: perform feature concatenation on the self-attention features and the frequency domain features.

[0098] When performing multi-head attention feature extraction on processed fault data based on the Transformer model encoder, the Transformer model encoder can be first constructed using the training data set. Specifically, the training data set can be used to pre-train the model. The Transformer model usually requires that the input data is three-dimensional, so a training data set needs to be constructed. The purpose of constructing a training data set is to meet the input requirements of the model. The method is to convert the data format of the training data set into ,in, is the sample size, is the data length of each sample, and 3 is the number of features, which represents the three features of output voltage signal, capacitor current signal and inductor current signal at each time point. After the training data set is constructed, the constructed training sample data set can be input into the Transformer model encoder.

[0099] The Transformer model is a model built on the Seq-to-Seq framework. Compared with the classic deep learning model, the most prominent advantage of the Transformer model is the use of a multi-head attention mechanism. The purpose of the multi-head attention sublayer is to give different importance to words / tokens in the sequence from multiple aspects. The Transformer model is mainly composed of input, encoder, decoder and output.

[0100] The position encoding layer in the Transformer model is to determine the position information of the sequence. Since there is no recursive layer and convolution layer in RNN and CNN, the input order information cannot be obtained by relying solely on the self-attention mechanism. The order information of the sequence needs to be actively passed to the model. The Transformer model uses a combination of sine and cosine functions to encode the position of the sequence. The calculation method is as follows:

[0101]

[0102] Among them, pos is the position of the current sequence; i is the dimension; is the dimension of the input features.

[0103] The multi-head attention mechanism in the Transformer model uses multiple attention mechanisms to perform operations in parallel, and then concatenates the operation results through linear transformation. The core technology in the Transformer model is the multi-head attention mechanism, which is used to extract the dependency features between data, capture the correlation between data and establish a context prediction model. The calculation method is as follows:

[0104]

[0105]

[0106]

[0107]

[0108]

[0109] Among them, Q is the query matrix; K is the key matrix; V is the value matrix; , , is a trainable parameter matrix; X is the processed input; is the dimension of the key matrix; , , , is the learnable parameter matrix.

[0110] In the Transformer model, the encoding and decoding parts also include feedforward networks, summation, and normalization. The calculation formula of the feedforward neural network is as follows:

[0111] Where x is the input; , , , are parameters that can be obtained through training.

[0112] The calculation formula for summation and normalization is as follows:

[0113] Where x is the input; This is the result after module processing.

[0114] The model processing flow is as follows: the input data is sent to the encoding part after position encoding, the output of the encoding part is flattened and sent to the decoder to map the feature parameters in the high-dimensional hidden layer to the original input data, thereby training the model's feature extraction ability.

[0115] Secondly, select the appropriate number of iterations and loss function, input the constructed data set into the feature extraction model to repeatedly perform forward propagation and back propagation iterative calculation processes; in this process, the model parameters of the embedding dimension, number of attention heads, encoder and decoder layers are continuously adjusted to complete the pre-training of the model.

[0116] Again, take out the encoding layer and decoding layer of the pre-trained model and retain their weight parameters to construct them as the trained Transformer model encoder.

[0117] Finally, based on the pre-trained Transformer model encoder, the training dataset Perform self-attention feature extraction to obtain the self-attention feature set .

[0118] Step 406: extract the frequency domain features of the multiple historical signal values, and fuse the self-attention features with the frequency domain features to generate target fusion features.

[0119] In an optional implementation, extracting frequency domain features of the multiple historical signal values ​​includes: The target sample data set is subjected to sliding window cutting according to a second window width and a second step length to generate a second sliding window cutting result, and frequency domain features of the plurality of historical signal values ​​are extracted based on the second sliding window cutting result.

[0120] Furthermore, the extracting of frequency domain features of the multiple historical signal values ​​based on the second sliding window cutting result includes: performing Fourier transform on each sample data set to be processed contained in the second sliding window cutting result to generate a time series energy spectrum corresponding to each sample data set to be processed; respectively extracting eigenvalues ​​corresponding to at least two initial frequency domain features of the time series energy spectrum; and normalizing the eigenvalues ​​using a maximum-minimum normalization algorithm to generate frequency domain features of the multiple historical signal values.

[0121] The at least two initial frequency domain features include: peak value, bandwidth, center frequency, shape factor, and power spectral density.

[0122] Specifically, the frequency domain features of the multiple historical signal values ​​are extracted, and the training data set generated by sliding window cutting can be specifically Perform frequency domain feature extraction based on expert knowledge.

[0123] Furthermore, the target sample data set generated by the sliding window cutting mentioned above can be subjected to sliding window cutting according to the second window width and the second step size to generate a second sliding window cutting result, and frequency domain features of multiple historical signal values ​​can be extracted based on the second sliding window cutting result.

[0124] Among them, when the target sample data set is cut by sliding window, if the second window width is , the second step length is 1, for the sample Can be cut out samples, each sample length is , that is, .

[0125] After the sliding window cutting is completed, Fourier transform can be performed on each sample data set to be processed contained in the second sliding window cutting result to convert it into the frequency domain to obtain the corresponding frequency domain representation F′ (time series energy spectrum).

[0126] For each frequency domain representation F′ of the sample data set to be processed, five initial frequency domain features are extracted, namely, the peak value, bandwidth, center frequency, shape factor, and power spectrum density of the energy spectrum. Each initial frequency domain feature includes a frequency domain feature and the eigenvalue corresponding to the frequency domain feature. The initial frequency domain features extracted are , so for the sample The initial frequency domain features extracted are After the extraction is completed, the eigenvalues ​​of each initial frequency domain feature can be normalized using the maximum and minimum normalization algorithm to generate frequency domain features of multiple historical signal values.

[0127] After the frequency domain features and self-attention features (high-dimensional hidden layer features) are extracted, the extracted frequency domain features and self-attention features can be concatenated.

[0128] For the training data set Each sample Perform self-attention feature extraction and frequency domain feature extraction and feature fusion. Let the dimension of the fusion feature be , the training data set can be reorganized into The two-dimensional fusion feature matrix of .

[0129] Step 408: Encode the target fusion features through the stacked autoencoder in the multi-layer model encoder to generate health features corresponding to the primary switching power supply of the aging station.

[0130] Specifically, an autoencoder is an unsupervised learning method that aims to learn effective representations or encodings of data and achieve self-learning by reconstructing input data. Stacked autoencoders (SAEs) are an extension of this method. Through multiple layers of nonlinear transformations, the original data is mapped to an increasingly abstract and compact feature space, thereby achieving deeper feature learning and representation learning, which is of great significance for dimensionality reduction, feature extraction, and classification tasks of complex data.

[0131] Corresponding to the above method embodiment, this specification also provides an embodiment of a health status assessment device for an isolated switching power supply at the front end of a burn-in station, Figure 4 FIG. 1 is a schematic diagram showing a structure of a device for evaluating the health status of an isolated switching power supply at the front end of a burn-in station provided by an embodiment of the present specification. Figure 4 As shown, the device comprises: The feature extraction module 401 is configured to obtain the monitoring data of the isolated switching power supply at the front end of the aging test bench at the current moment, and obtain feature data with high information density by performing feature extraction on the monitoring data; The preliminary evaluation module 402 is configured to obtain a preliminary health status evaluation result of the front-end isolated switching power supply of the aging test bench at the current moment by inputting the high information density feature data into a trained neighbor component analysis (NCA) algorithm model; The final evaluation module 403 is configured to, if the preliminary health status evaluation result is a degraded state, use a multi-distance metric fusion method with automatic weight optimization to perform a quantitative calculation of the health status of the feature data with high information density, obtain the health status of the degraded state at the current moment, and use the health status of the degraded state at the current moment as the final health status evaluation result of the front-end isolated switching power supply of the aging test bench.

[0132] Furthermore, the final evaluation module 403 is also configured to use the healthy state as the final healthy state evaluation result of the front-end isolated switching power supply of the aging test bench if the evaluation state result is a healthy state.

[0133] Optionally, a training module 404 is further included, configured to: Acquire first historical monitoring data and second historical monitoring data of the full life cycle degradation cycle of the front-end isolated switching power supply of the aging test bench, and use the first historical monitoring data as a training set and the second historical monitoring data as a test set; By performing feature extraction on the historical monitoring data in the training set, first historical feature data with high information density is obtained; Determine the state label corresponding to each moment of the first high information density historical feature data; wherein the state label includes a healthy state label and a degraded state label; An NCA algorithm model is constructed, and the historical feature data with the first high information density is used as the model input. The state label corresponding to each moment of the historical feature data with the first high information density is used as the model output to train and learn the NCA algorithm model, so as to obtain a trained NCA algorithm model and an optimized mapping matrix.

[0134] Optionally, the training module 404 is further configured to: By performing feature extraction on the historical monitoring data in the test set, historical feature data with the second highest information density is obtained; By inputting the historical feature data with the second highest information density into the trained NCA algorithm model, the state label corresponding to each moment of the historical feature data with the second highest information density is obtained; Extracting all healthy state labels from the state labels corresponding to each moment of the historical feature data with the second highest information density, and discarding all degraded state labels; Build health baseline data based on the historical feature data corresponding to all health status tags.

[0135] Optionally, the final evaluation module 403 is specifically configured to respectively calculate the Mahalanobis distance, cosine similarity and Manhattan distance between the feature data with high information density and the health baseline data; By performing multi-distance metric fusion with weighted automatic optimization on the Mahalanobis distance, the cosine similarity and the Manhattan distance, the health of the degradation state at the current moment is obtained, and the calculation formula is:

[0136] in, Indicates the health of the degradation state at the current moment, represents the Mahalanobis distance; represents cosine similarity; represents Manhattan distance; , and denote weights, and + + =1.

[0137] Furthermore, by measuring the monotonicity and smoothness of the health curve in the health state of the current degradation state, the monotonicity loss and the smoothness loss are calculated respectively, and the monotonicity loss and the smoothness loss are used as objective functions. The multi-objective genetic algorithm is used to optimize the fusion weights of multiple distance metrics to obtain the best weight combination.

[0138] The evaluation indicators include the monotonicity and smoothness of the health curve, and the monotonicity loss and smoothness loss are calculated. The health status of the equipment gradually degrades over time. Assume that the health curve is ,but: Monotonicity loss:

[0139] Smoothness loss:

[0140] The above is a schematic scheme of a device for evaluating the health status of an isolated switching power supply at the front end of an aging station in this embodiment. It should be noted that the technical scheme of the device for evaluating the health status of an isolated switching power supply at the front end of an aging station and the technical scheme of the method for evaluating the health status of an isolated switching power supply at the front end of an aging station belong to the same concept. For details not described in detail in the technical scheme of the device for evaluating the health status of an isolated switching power supply at the front end of an aging station, please refer to the description of the technical scheme of the method for evaluating the health status of an isolated switching power supply at the front end of an aging station.

[0141] Figure 5 The structure block diagram of a computing device 500 provided according to an embodiment of the present specification is shown. 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 the database 550 is used to store data.

[0142] The computing device 500 also includes an access device 540 that enables the computing device 500 to communicate via one or more networks 560. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a 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 network interface (e.g., a network interface card (NIC)) that is wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a World Wide 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 the like.

[0143] In one embodiment of the present specification, the above components of the computing device 500 and Figure 5 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 5 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0144] Computing device 500 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. Computing device 500 may also be a mobile or stationary server.

[0145] The processor 520 is used to execute the following computer executable instructions, which, when executed by the processor, implement the steps of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station.

[0146] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station belong to the same concept. For details not described in detail in the technical scheme of the computing device, please refer to the description of the technical scheme of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station.

[0147] An embodiment of the present 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 method for evaluating the health status of the isolated switching power supply at the front end of the aging station.

[0148] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station belong to the same concept. For details not described in detail in the technical scheme of the storage medium, please refer to the description of the technical scheme of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station.

[0149] An embodiment of the present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer is instructed to execute the steps of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station.

[0150] The above is a schematic scheme of a computer program of this embodiment. It should be noted that the technical scheme of the computer program and the technical scheme of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station belong to the same concept. For details not described in detail in the technical scheme of the computer program, please refer to the description of the technical scheme of the above-mentioned method for evaluating the health status of the isolated switching power supply at the front end of the aging station.

[0151] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0152] The computer instructions include computer program codes, which may be in source code form, object code form, 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), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0153] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0154] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0155] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods 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 in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is only limited by the claims and their full scope and equivalents.

Claims

1. A method for evaluating the health status of a front-end isolated switching power supply of a burn-in station, characterized in that: include: Acquire the monitoring data of the isolated switching power supply at the front end of the aging test bench at the current moment, and obtain feature data with high information density by extracting features from the monitoring data; By inputting the high information density feature data into the trained nearest neighbor component analysis (NCA) algorithm model, a preliminary health status assessment result of the front-end isolated switching power supply of the aging test bench at the current moment is obtained; If the preliminary health status assessment result is a degraded state, the health status of the high information density feature data is quantitatively calculated using a multi-distance metric fusion method with automatic weight optimization to obtain the health degree of the degraded state at the current moment, and the health degree of the degraded state at the current moment is used as the final health status assessment result of the isolated switching power supply at the front end of the aging test bench.

2. The method according to claim 1, characterized in that Also includes: If the evaluation status result is a healthy status, the healthy status is used as the final healthy status evaluation result of the front-end isolated switching power supply of the aging test bench.

3. The method according to claim 1, characterized in that Also includes: Acquire first historical monitoring data and second historical monitoring data of the full life cycle degradation cycle of the front-end isolated switching power supply of the aging test bench, and use the first historical monitoring data as a training set and the second historical monitoring data as a test set; By performing feature extraction on the historical monitoring data in the training set, first historical feature data with high information density is obtained; Determine the state label corresponding to each moment of the first high information density historical feature data; wherein the state label includes a healthy state label and a degraded state label; An NCA algorithm model is constructed, and the historical feature data with the first high information density is used as the model input. The state label corresponding to each moment of the historical feature data with the first high information density is used as the model output to train and learn the NCA algorithm model, so as to obtain a trained NCA algorithm model and an optimized mapping matrix.

4. The method according to claim 3, characterized in that Also includes: By performing feature extraction on the historical monitoring data in the test set, historical feature data with the second highest information density is obtained; By inputting the historical feature data with the second highest information density into the trained NCA algorithm model, the state label corresponding to each moment of the historical feature data with the second highest information density is obtained; Extracting all healthy state labels from the state labels corresponding to each moment of the historical feature data with the second highest information density, and discarding all degraded state labels; Build health baseline data based on the historical feature data corresponding to all health status tags.

5. The method according to claim 4, characterized in that The health status of the high information density feature data is quantitatively calculated using a multi-distance metric fusion method with automatic weight optimization to obtain the health status of the degradation state at the current moment, including: respectively calculating the Mahalanobis distance, cosine similarity and Manhattan distance between the feature data with high information density and the health baseline data; The health of the degradation state at the current moment is obtained by fusing the multi-distance metrics by automatically optimizing the weights of the Mahalanobis distance, the cosine similarity and the Manhattan distance.

6. The method according to claim 5, characterized in that Calculating the Mahalanobis distance between the high information density feature data and the health baseline data includes: Based on the optimized mapping matrix, the Mahalanobis distance between the feature data with high information density and the healthy baseline data is calculated.

7. The method according to claim 5, characterized in that By fusing the Mahalanobis distance, the cosine similarity and the Manhattan distance with multi-distance metrics with automatic weight optimization, the health of the degradation state at the current moment is obtained, including: ； in, Indicates the health of the degradation state at the current moment, represents the Mahalanobis distance; represents cosine similarity; represents Manhattan distance; , and denote weights, and + + =1; Among them, by measuring the monotonicity and smoothness of the health curve in the health state of the current degradation state, the monotonicity loss and the smoothness loss are calculated respectively, and the monotonicity loss and the smoothness loss are used as objective functions. The multi-objective genetic algorithm is used to optimize the fusion weights of multiple distance metrics to obtain the best weight combination.

8. A health status assessment device for the front-end isolated switching power supply of a burn-in station, characterized in that: include: The feature extraction module is configured to obtain the monitoring data of the isolated switching power supply at the front end of the aging test bench at the current moment, and obtain feature data with high information density by extracting features from the monitoring data; A preliminary evaluation module is configured to obtain a preliminary health status evaluation result of the front-end isolated switching power supply of the aging test bench at the current moment by inputting the high information density feature data into a trained neighbor component analysis (NCA) algorithm model; The final evaluation module is configured to, if the preliminary health status evaluation result is a degraded state, use a multi-distance metric fusion method with automatic weight optimization to quantitatively calculate the health status of the high information density feature data, obtain the health status of the degraded state at the current moment, and use the health status of the degraded state at the current moment as the final health status evaluation result of the front-end isolated switching power supply of the aging test bench.

9. A computing device, characterized in that include: Memory and 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 health status assessment method of the front-end isolated switching power supply of the aging station described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: It stores computer executable instructions, which, when executed by a processor, implement the steps of the method for evaluating the health status of the front-end isolated switching power supply of the aging station as described in any one of claims 1 to 7.

Images