Method and device for evaluating the health state of a secondary power supply of a high-temperature aging test equipment

Abstract

The present invention discloses a method and device for evaluating the health status of the secondary power supply of a high-temperature aging test equipment. The method includes: collecting the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated; extracting features from the current and voltage output monitoring signals, and obtaining a multi-dimensional feature sequence by mixing the degradation curves such as the power supply ripple volatility and the MOSFET case temperature to junction temperature; inputting the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment; in the case that the secondary power supply of the high-temperature aging test equipment is in a degraded state, determining the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated based on the probability space points after mapping of the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model, so as to determine the health status of the secondary power supply of the high-temperature aging test equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of non-destructive reliability screening of integrated circuits, and particularly to a method and device for evaluating the health status of a secondary power supply of a high-temperature aging test equipment. Background Art

[0002] Existing 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 technologies for high-temperature aging products in existing high-temperature aging products, it is difficult to achieve the integrity of the test process and the consistency of the test environment stress for related products, which is extremely likely to cause major property losses such as the destruction of millions of test devices (also known as devices to be detected) due to the forced interruption of the test process caused by product failures, or the aging test is recognized as a failure test due to adverse effects such as additional stress introduced during the test period due to product performance degradation, resulting in ineffective waste of resources. At the same time, ensuring the quality of high-temperature aging tests can avoid over-aging and under-aging failures of integrated circuits, greatly reduce the failure rate of integrated circuits, and avoid the failure shutdown of large-scale electronic systems such as new energy vehicles, civil airliners, and energy storage substations due to integrated circuit failures.

[0003] High-temperature aging products such as aging platforms (also known as aging test equipment) simulate various actual usage environments and working conditions, enabling products to experience 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 and run for a long time to simulate the aging process of products in actual use. Among them, the secondary power supply refers to a power supply device that converts and distributes the primary power supply in a high-temperature aging test bench to meet the specific power requirements of different loads (such as electronic components, circuit boards, etc.) in aging tests.

[0004] During the process of aging tests, if the secondary power supply fails, such as a sudden drop or disappearance of the output voltage, it will cause an unexpected interruption of the aging test. And during the use of aging test equipment, problems such as capacitor aging and performance degradation of switching tubes will gradually occur. By evaluating the health of the secondary power supply of the aging test equipment, potential hidden dangers inside the power supply can be discovered in advance, repaired or replaced in a timely manner, avoiding test interruptions, and ensuring that the aging test can proceed smoothly according to the plan.

[0005] It can be seen that there is an urgent need for those skilled in the art to provide a solution for evaluating the health status of the secondary power supply of high-temperature aging test equipment. Summary of the Invention

[0006] The purpose of the embodiments of the present invention is to provide a method and device for evaluating the health status of a secondary power supply of a high-temperature aging test equipment, and an electronic device, which can solve the above problems existing in the prior art.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] An embodiment of the present invention provides a method for evaluating the health status of the secondary power supply of a high-temperature aging test equipment. Wherein, the method includes:

[0009] Collect the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated, including current signals, voltage output signals, power supply ripple volatility, MOSFET device case temperature to junction temperature, etc.;

[0010] Extract features from the monitoring signals such as current and voltage output, and combine the power supply ripple volatility and the MOSFET case temperature to junction temperature degradation curve to obtain a multi-dimensional feature sequence;

[0011] Input the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment. Wherein, the degradation feature state includes: healthy state and degradation state;

[0012] In the case that the secondary power supply of the high-temperature aging test equipment is in a degradation state, based on the probability space points mapped by the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model, determine the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated. Wherein, the state degradation curve can characterize the health status of the secondary power supply of the high-temperature aging test equipment.

[0013] Optionally, the residual convolutional neural network model includes: a convolutional layer, a batch normalization layer, an activation function, a max pooling layer, a plurality of residual blocks, a global average pooling layer, and a fully connected layer.

[0014] Optionally, the residual convolutional neural network model is trained in the following manner:

[0015] Obtain the training time-series degradation feature samples of the secondary power supply of the high-temperature aging test equipment; wherein, each training time-series degradation feature sample is marked with a degradation feature state;

[0016] Input the training time-series degradation feature samples into a pre-established residual convolutional neural network model to obtain the first probability space points after mapping;

[0017] Discriminate the degradation starting point of the secondary power supply of the high-temperature aging test equipment according to the two-dimensional first probability space points, and generate the health baseline of the secondary power supply of the high-temperature aging test equipment.

[0018] Optionally, the method further includes:

[0019] Obtain the test time-series degradation feature samples of the secondary power supply of the high-temperature aging test equipment;

[0020] Input the sample of the test timing degradation characteristics into the residual convolutional neural network model being trained to obtain the mapped second probability space points;

[0021] Calculate the distance between the second probability space and the healthy baseline based on a preset multi-distance metric algorithm.

[0022] Optionally, the step of calculating the distance between the second probability space and the healthy baseline based on a preset multi-distance metric algorithm includes:

[0023] Calculate the distances between the second probability space and the healthy baseline based on the Mahalanobis distance algorithm, cosine similarity algorithm, and Manhattan distance algorithm respectively to obtain the first distance, second distance, and third distance;

[0024] Perform weighted summation on the first distance, second distance, and third distance to obtain the health degree of the second power supply of the high-temperature aging test equipment.

[0025] An embodiment of the present invention further provides a device for evaluating the health state of the secondary power supply of a high-temperature aging test equipment, wherein the device includes:

[0026] A collection module for collecting monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated; wherein the monitoring signals include: current signal, voltage output signal, power supply ripple volatility, MOSFET case temperature to junction temperature;

[0027] An extraction module for performing feature extraction on the current signal and the voltage output signal and then mixing the power supply ripple volatility and the MOSFET case temperature to junction temperature degradation curve to obtain a multi-dimensional feature sequence;

[0028] A prediction module for inputting the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment, wherein the degradation feature state includes: healthy state and degradation state;

[0029] A determination module for determining the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated based on the probability space points mapped from the multi-dimensional feature sequence and the healthy baseline in the residual convolutional neural network model in the case that the secondary power supply of the high-temperature aging test equipment is in a degradation state, wherein the state degradation curve can characterize the health state of the secondary power supply of the high-temperature aging test equipment.

[0030] Optionally, the residual convolutional neural network model includes: a convolutional layer, a batch normalization layer, an activation function, a max pooling layer, multiple residual blocks, a global average pooling layer, and a fully connected layer.

[0031] Optionally, the device further includes a training module for training a residual convolutional neural network model, and the training module includes:

[0032] A first sub-module for obtaining training time series degradation feature samples of the secondary power supply of the high-temperature aging test equipment; wherein each training time series degradation feature sample is labeled with a degradation feature state;

[0033] A second sub-module for inputting the training time series degradation feature samples into a pre-established residual convolutional neural network model to obtain a first probability space point after mapping;

[0034] A third sub-module for discriminating the degradation starting point of the secondary power supply of the high-temperature aging test equipment according to the two-dimensional first probability space point and generating a health baseline for the secondary power supply of the high-temperature aging test equipment.

[0035] Optionally, the training module further includes:

[0036] A fourth sub-module for obtaining test time series degradation feature samples of the secondary power supply of the high-temperature aging test equipment;

[0037] A fifth sub-module for inputting the test time series degradation feature samples into the residual convolutional neural network model under training to obtain a second probability space point after mapping;

[0038] A sixth sub-module for calculating the distance between the second probability space and the health baseline based on a preset multi-distance metric algorithm.

[0039] Optionally, the sixth sub-module is specifically configured to:

[0040] Calculate the distances between the second probability space and the health baseline based on the Mahalanobis distance algorithm, the cosine similarity algorithm, and the Manhattan distance algorithm respectively to obtain a first distance, a second distance, and a third distance;

[0041] Perform a weighted sum of the first distance, the second distance, and the third distance to obtain the health degree of the second power supply of the high-temperature aging test equipment.

[0042] An embodiment of the present invention further provides an electronic device, which is characterized by including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete mutual communication through the communication bus; the memory is used for storing a computer program; when the processor executes the program stored in the memory, it implements the process of any one of the above-mentioned health state evaluation methods for the secondary power supply of the high-temperature aging test equipment.

[0043] The secondary power supply health state evaluation scheme for high-temperature aging test equipment disclosed in the embodiments of the present invention collects the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated; obtains a multi-dimensional feature sequence based on the monitoring signals; inputs the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment; in the case that the secondary power supply of the high-temperature aging test equipment is in a degraded state, based on the probability space points after mapping of the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model, determines the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated, so as to accurately determine the health state of the secondary power supply of the high-temperature aging test equipment. The secondary power supply health state evaluation scheme for high-temperature aging test equipment provided by the embodiments of the present invention collects the secondary power supply monitoring signals and extracts the multi-dimensional feature sequence, and predicts the multi-dimensional feature sequence based on the trained model to obtain the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated, so as to accurately determine the health state of the secondary power supply of the high-temperature aging test equipment. Since the health state of the secondary power supply of the high-temperature test equipment can be accurately determined, it is possible to effectively avoid the damage of the test piece due to the failure of the secondary power supply during the aging test of the test piece by the high-temperature test equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flowchart showing the steps of a method for evaluating the health state of the secondary power supply of a high-temperature aging test equipment according to an embodiment of the present application;

[0045] Figure 2 is a flowchart showing the steps of a method for evaluating the health state of the secondary power supply of a high-temperature aging test equipment according to an embodiment of the present application;

[0046] Figure 3 is a flowchart showing the steps of a method for depth fusion of time-frequency features of the secondary power supply of a high-temperature aging test equipment according to an embodiment of the present application;

[0047] Figure 4 is a structural block diagram showing a device for evaluating the health state of the secondary power supply of a high-temperature aging test equipment according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0049] The following will, with reference to the accompanying drawings, through specific embodiments and their application scenarios, describe in detail the method for evaluating the health state of the secondary power supply of a high-temperature aging test equipment provided by the embodiments of the present application.

[0050] As shown in the Figure 1 accompanying drawings, the method for evaluating the health state of the secondary power supply of a high-temperature aging test equipment according to the embodiments of the present application includes the following steps:

[0051] Step 101: Collect the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated.

[0052] The method for evaluating the health status of the secondary power supply of the high-temperature aging test equipment provided by the embodiment of the present application can be applied to an electronic device. An evaluation computer program for the health status of the secondary power supply of the high-temperature aging test equipment is set in the electronic device. When the computer program is executed by a processor, the method for evaluating the health status of the secondary power supply of the high-temperature aging test equipment in the embodiment of the present application is realized.

[0053] The aging test equipment provided by the present application can perform non-destructive aging detection on the test piece. The method for evaluating the health status of the secondary power supply in the embodiment of the present application can evaluate the health status of the secondary power supply of the aging test equipment, and the health evaluation of the secondary power supply of the aging station can ensure the safety of the test piece.

[0054] In the actual implementation process, the directly measured monitoring signal data of the secondary power supply of the high-temperature aging test equipment is large in quantity and contains a large amount of redundant information. Therefore, preliminary feature extraction is performed on the directly measured data to obtain a feature sequence with high information density.

[0055] Step 102: Generate a multi-dimensional feature sequence based on the monitoring signal.

[0056] The monitoring signals include but are not limited to current signals, voltage output signals, power supply ripple volatility, MOSFET case temperature to junction temperature, etc. The specific features extracted can be flexibly set by those skilled in the art according to actual needs, as long as the extracted features can characterize the state of the secondary power supply. Among them, the abbreviation of MOSFET is MOS, which is a metal-oxide-semiconductor field-effect transistor, a voltage-controlled semiconductor device, and refers to the voltage-controlled semiconductor device in the secondary power supply of the high-temperature aging test equipment in this embodiment.

[0057] Specifically, feature extraction is performed on the current signal and the voltage output signal, and then the power supply ripple volatility and the MOSFET device case temperature to junction temperature degradation curve are mixed to obtain a multi-dimensional feature sequence.

[0058] Step 103: Input the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment.

[0059] Among them, the degraded feature states include: healthy state and degraded state. If the degraded feature state of the secondary power supply of the high-temperature aging test equipment is in the healthy state, there is no need to execute step 104 and subsequent steps. If the degraded feature state of the secondary power supply of the high-temperature aging test equipment is in the degraded state, it indicates that the secondary power supply of the high-temperature aging test equipment has degraded. By performing adaptive analysis on the multi-dimensional feature sequence data through the residual convolutional neural network model, the degradation progress of the secondary power supply of the high-temperature aging test equipment can be automatically identified.

[0060] In an optional embodiment, the training process of the residual convolutional neural network model may include the following sub-steps:

[0061] Sub-step 1: Obtain the training time-series degraded feature samples of the secondary power supply of the high-temperature aging test equipment;

[0062] Among them, each training time-series degraded feature sample is labeled with a degraded feature state. The training time-series degraded feature samples of the secondary power supply of the high-temperature aging test equipment can be obtained by monitoring the temperature signal and the degraded feature state during the entire life cycle of the secondary power supply of a high-temperature aging test equipment.

[0063] In the actual implementation process, the training time-series degraded feature samples of the secondary power supply of the high-temperature aging test equipment can be expressed as , the maximum number of iterations max_iter. Mark the corresponding degraded feature state of the trained secondary power supply of the high-temperature aging test equipment , including the healthy state and the degraded state.

[0064] Sub-step 2: Input the training time-series degraded feature samples into the pre-established residual convolutional neural network model to obtain the first probability space point after mapping;

[0065] The residual convolutional neural network model includes: convolutional layer, batch normalization layer, activation function (ReLU), max pooling layer, multiple residual blocks (Residual blocks), global average pooling layer, and fully connected layer.

[0066] In the actual implementation process, with as the input, as the output to train the ResNet model (i.e., the residual convolutional neural network model); obtain the iterative learning to update the ResNet model parameters. When the number of iterations reaches the maximum number of iterations max_iter, stop the iterative learning.

[0067] In the actual implementation process, the maximum number of iterations max_iter can be flexibly set by those skilled in the art, and no specific limitation is made in the embodiments of the present application.

[0068] Sub-step 3: Discriminate the degradation starting point of the secondary power supply of the high-temperature aging test equipment according to the two-dimensional first probability space points, and generate a health baseline for the secondary power supply of the high-temperature aging test equipment.

[0069] In an optional embodiment, the training process of the residual convolutional neural network model may further include the following process:

[0070] Obtain the test time-series degradation feature samples of the secondary power supply of the high-temperature aging test equipment; input the test time-series degradation feature samples into the training residual convolutional neural network model to obtain the mapped second probability space points; based on a preset multi-distance metric algorithm, calculate the distance between the second probability space and the health baseline.

[0071] The test time-series degradation feature samples of the secondary power supply of the high-temperature aging test equipment can be expressed as: . During the training and testing of the residual convolutional neural network model, and are used as the input of the ResNet algorithm model to obtain the mapped probability space points and .

[0072] Automatically discriminate the degradation starting point according to the two-dimensional probability space result, and calculate the distances between the health space and the degradation space of the secondary power supply of the training and testing high-temperature aging test equipment by using a weighted fusion multi-distance metric algorithm . Traverse all failure modes to obtain the health assessment results corresponding to each failure mode. Finally, the trained residual convolutional neural network model outputs and The health assessment curves corresponding to the second power supplies of each high-temperature aging test equipment.

[0073] More specifically, based on the preset multi-distance metric algorithm, the method for calculating the distance between the second probability space and the health baseline can be: calculate the distances between the second probability space and the health baseline based on the Mahalanobis distance algorithm, cosine similarity algorithm, and Manhattan distance algorithm respectively to obtain the first distance, the second distance, and the third distance; perform a weighted sum of the first distance, the second distance, and the third distance to obtain the corresponding health degree.

[0074] In the actual implementation process, it is not limited to using the above three distance algorithms to calculate the distance. It can also use any one of the above three distance algorithms, or use any two of the above three distance algorithms to obtain two distances and then perform a weighted sum of the two distances.

[0075] During the training process of the residual convolutional neural network model, introduce the time-sequence degradation feature samples of the secondary power supply test of the high-temperature aging test equipment. The accuracy of the prediction of the residual convolutional neural network model can be tested through the prediction results. In the actual implementation process, when the prediction accuracy of the residual convolutional neural network model reaches the accuracy threshold, it can be determined that the training of this residual convolutional neural network model is completed. The subsequent residual convolutional neural network model after training can be used to predict the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated.

[0076] Step 104: In the case where the secondary power supply of the high-temperature aging test equipment is in a degraded state, based on the probability space points after mapping of the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model, determine the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated.

[0077] Among them, the state degradation curve can characterize the health state of the secondary power supply of the high-temperature aging test equipment.

[0078] By analyzing the multi-dimensional features of the secondary power supply of the high-temperature aging test equipment to be evaluated through the residual convolutional neural network model, the health degree of the secondary power supply at the current moment can be obtained, and at the same time, the state degradation curve (also known as the health degree degradation curve) up to this moment is output, providing reliable data support for the further prediction research of the remaining service life of the secondary power supply of the high-temperature aging test equipment.

[0079] The method for evaluating the health state of the secondary power supply of the high-temperature aging test equipment disclosed in the embodiments of the present invention collects the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated, and generates a multi-dimensional feature sequence based on the monitoring signals; inputs the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment; in the case where the secondary power supply of the high-temperature aging test equipment is in a degraded state, based on the probability space points after mapping of the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model, determine the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated, so as to accurately determine the health state of the secondary power supply of the high-temperature aging test equipment. The solution for evaluating the health state of the secondary power supply of the high-temperature aging test equipment provided by the embodiments of the present invention collects the secondary power supply monitoring signals and extracts the multi-dimensional feature sequence, and based on the trained model, predicts the multi-dimensional feature sequence to obtain the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated, so as to accurately determine the health state of the secondary power supply of the high-temperature aging test equipment.

[0080] The following combines Figure 2 Take a specific example to illustrate the solution for evaluating the health state of the secondary power supply of the high-temperature aging test equipment provided by this application.

[0081] This specific example provides a method for evaluating the health status of the secondary power supply of a high-temperature aging test equipment based on the fusion of a residual convolutional neural network and multiple distance metrics. The core is as follows: During the aging test, monitoring signals such as voltage, current, power supply ripple volatility, and MOSFET case temperature to junction temperature will be generated by the secondary power supply of the high-temperature aging test equipment. The directly measured monitoring signals have a large amount of data and contain a lot of redundant information. First, feature extraction is performed on the directly measured data to obtain a feature sequence with high information density. The ResNet (Residual Neural Network) model is used to perform adaptive analysis on the feature sequence data to automatically identify the degradation starting point of the secondary power supply of the high-temperature aging test equipment.

[0082] The distance metric algorithm is used to quantitatively calculate the health status of the feature sequence in the degradation stage, and the health assessment of the secondary power supply of the high-temperature aging test equipment is completed.

[0083] The process of the method for evaluating the health status of the secondary power supply of the high-temperature aging test equipment based on the fusion of a residual convolutional neural network and multiple distance metrics can specifically include the following processes:

[0084] S1: Monitoring the signals of the secondary power supply of the high-temperature aging test equipment:

[0085] The high-temperature aging test equipment is Figure 2 the high-temperature aging platform shown in. During the aging test, the voltage and current signals of the secondary power supply of the high-temperature aging test equipment are monitored through sensors to obtain a large amount of monitoring data for the state assessment of the switching power supply.

[0086] S2: Preliminary feature extraction.

[0087] Since the directly measured monitoring signals have a large amount of data and contain a lot of redundant information, first, feature extraction needs to be performed on the directly monitored data to obtain a feature sequence with high information density, that is, the multi-dimensional degradation features of the secondary power supply of the high-temperature aging test equipment.

[0088] S3: Input the multi-dimensional degradation features of the secondary power supply of the high-temperature aging test equipment into the residual convolutional neural network model, perform automatic identification of the degradation starting point, and obtain the health baseline and degradation features.

[0089] On the basis of the extracted feature sequence, the health assessment process is further realized: First, a residual convolutional neural network model is used to judge the degradation starting point of the secondary power supply temperature data and its feature data of the high-temperature aging test equipment, eliminating the instability brought by artificially setting the health baseline for constructing health indicators. The ResNet model is based on the residual structure as the core. By constructing "shortcut connections", it cleverly adds the input of each layer directly to the output after operations such as convolution, ensuring the stability of data feature extraction and learning. The ResNet model uses convolutional layers to extract features. Its convolutional layers act on the input time series data of the secondary power supply of the high-temperature aging test equipment in the form of a sliding window. During the process of the convolutional kernel sliding along the time axis, it focuses on different local time intervals, accurately captures the feature patterns therein, and excavates the internal laws hidden in the time series data that can reflect the changes in the equipment operation state. The ResNet model stacks multiple modules containing convolutional layers, activation functions, and residual connections reasonably. After the time series data of the secondary power supply of the high-temperature aging test equipment is processed layer by layer, rich and valuable feature information can be extracted. Then the fully connected layer integrates the numerous features extracted previously, maps the feature vectors to a two-dimensional probability space, and realizes the accurate discrimination of the degradation starting point of the secondary power supply of the high-temperature aging test equipment, providing a strong basis for knowing in advance the changes in the health status of the equipment and reasonably arranging maintenance plans, etc.

[0090] Specifically, in the training stage of the ResNet model, training data with health status labels (i.e., training time series degradation feature samples and the degradation feature state labels marked for each sample) are input , where is the feature sample, is the corresponding label, 0 and 1 represent the healthy state and the degraded state respectively, and n is the number of feature samples obtained by resampling the entire degradation curve with a sliding window. In this process, the ResNet model continuously optimizes the mapping parameters of the input data to the low-dimensional feature space, strengthens the classification ability of healthy state data and degraded state data, so that the ResNet model obtains the ability to automatically segment the health baseline of the test data.

[0091] In the testing stage of the ResNet model, test data (i.e., the secondary power supply test time series degradation feature samples) are input . The ResNet model can map the health baseline data and the degraded data to a two-dimensional probability space according to the model parameters optimized and learned during the training process, complete the identification of the degradation starting point of the test data, and obtain , avoiding the blindness of manually segmenting the health baseline.

[0092] S4: Adopt the multi-distance metric fusion method to calculate the distance between the healthy baseline and the degradation features, and determine the state degradation curve and the current health degree of the secondary power supply of the high-temperature aging test equipment.

[0093] To further realize the health state assessment of the secondary power supply of the high-temperature aging test equipment at the current moment, use the multi-distance metric fusion method, comprehensively consider the Mahalanobis distance, cosine similarity, and Manhattan distance between sample data and perform weighted fusion to obtain the health degree at the current moment. At the same time, output the health degree degradation curve up to this moment, providing reliable data support for the further remaining service life prediction research of the secondary power supply of the high-temperature aging test equipment.

[0094] The embodiment of the present application also provides a time-frequency feature deep fusion method for the secondary power supply of the high-temperature aging test equipment, which is used to extract features from the secondary power supply monitoring signals during the health state assessment of the secondary power supply of the high-temperature aging test equipment.

[0095] As shown in the appendix Figure 3 The time-frequency feature deep fusion method for the secondary power supply of the high-temperature aging test equipment in the embodiment of the present application includes the following steps:

[0096] Step 301: Collect the fault data set of the components of the high-temperature aging test equipment.

[0097] The high-temperature aging test equipment provided by the present application can perform non-destructive aging detection on the test pieces. The time-frequency domain feature processing method in the embodiment of the present application can extract features from the state data, health data, etc. of the components of the high-temperature aging test equipment. The extracted features can be used for the health assessment, fault prediction, etc. of the components of the aging test equipment, thus ensuring the safety of the test pieces. The aging test equipment is the high-temperature aging test equipment, also known as the high-temperature aging table or simply the aging table for short. The secondary power supply is the secondary power supply described above.

[0098] Among them, the fault data includes multiple key parameter time series data, and the key parameters are different for different components. For example: when the component is a secondary switching power supply, the key parameters can be the power supply voltage, current signal, and the thermal resistance characteristic from the MOSFET case temperature to the junction temperature synchronously monitored by an infrared thermal imager and an embedded temperature sensor. When the component is a test chamber, the key parameter can be the temperature monitoring signal, etc. These monitoring signals can be used to evaluate the health state of the components of the aging test equipment. In the actual implementation process, the directly measured key parameter data is large in quantity and contains a large amount of redundant information. Therefore, it is necessary to use the feature extraction model trained by the time-frequency feature deep fusion method for the secondary power supply of the high-temperature aging test equipment in the embodiment of the present application to perform preliminary feature extraction on the directly measured data to obtain a feature sequence with high information density.

[0099] In the embodiments of the present application, the fault data set collected from the old test equipment devices can be the fault data set during the entire process from the start of degradation to the stop of operation (i.e., the end of the remaining life cycle) in the device's full life cycle.

[0100] When collecting the fault data set of the aging test equipment devices, it is necessary to synchronously monitor the load change characteristics of the secondary switching power supply, including the output voltage and current transient response under dynamic load switching. For the load fluctuation when the device under test is connected, record the power supply ripple volatility caused by it. The calculation formula is: Ripple volatility = ( / ) × 100%

[0101] where is the peak-to-peak value of the output voltage, and is the average value of the output voltage. Capture the ripple waveform through a high-precision oscilloscope and analyze its spectral characteristics to identify abnormal harmonic components. The devices can include, but are not limited to: the switching power supply of the aging platform, the high-temperature test chamber of the aging platform, the high-temperature and high-humidity test chamber of the aging platform, etc. The switching power supply of the aging platform can be further divided into: the primary switching power supply, the secondary switching power supply, and the MOSFET switch tube of the secondary switching power supply. Moreover, the devices of the aging platform can also include: the drive control detection board.

[0102] The specific devices included in the aging test equipment can be flexibly adjusted according to the type of the aging test equipment. In the embodiments of the present application, the specific type of the aging platform is not limited. For example: the aging test equipment can be a high-temperature and high-humidity aging platform, or a constant-temperature and constant-pressure aging platform, etc.

[0103] Step 302: Preprocess the fault data set to obtain a training data set and a test data set.

[0104] An optional way to preprocess the fault data set to obtain a training data set and a test data set can include the following sub-steps:

[0105] Sub-step 1: For each key parameter in the fault data set, perform sliding window cutting on the key parameter time series data to construct a sample data set;

[0106] When constructing the sample data set by sliding window cutting, it is necessary to adaptively segment the data during the load mutation period. For the interval where the load change rate exceeds the threshold (such as ±10% / ms), use a smaller window width and step size to capture the detailed features of the transient response. At the same time, perform dynamic baseline calibration on the current and voltage output data to eliminate the baseline drift caused by the connection of the device under test.

[0107] Sub-step 2: Normalize each data sample in the sample data set to obtain a normalized sample data set;

[0108] Sub-step 3: Select the sample data of the first preset percentage from the normalized sample dataset as the training dataset, and use the remaining sample data as the test dataset.

[0109] Among them, the training dataset is used to train the preset model, and the test dataset is used to verify the prediction accuracy of the trained feature extraction model. The preset percentage can be flexibly set by those skilled in the art, and this application embodiment does not make specific restrictions thereon. For example, it is set to 60%-80%, preferably 70%.

[0110] Step 303: Perform short-time Fourier transform on the time series data of each key parameter in the training dataset to obtain a time-frequency diagram.

[0111] Most traditional time-domain and frequency-domain analysis methods are for stationary signals and can only obtain information in one aspect of the time domain or the frequency domain. In the application of actual devices such as secondary switching power supplies, the collected signals are mostly non-stationary, and the analysis in one aspect can no longer meet the needs. Therefore, it is necessary to determine the relationship between the signal frequency and time. The joint time-frequency analysis method is a very effective tool in the current process of processing non-stationary signals. Therefore, short-time Fourier transform is used in this application to process variable time-series and non-stationary signals. The basic principle of STFT (i.e., short-time Fourier transform) is: use a window function h(t) with a finite duration to intercept the vibration signal, perform Fourier transform on the obtained signal to obtain the local spectrum in a small range of this time period, and gradually analyze the signal band by moving the window function h(t) on the time axis to obtain a set of local "spectrums" of the signal. The essence of STFT is the transformation of the basis function.

[0112] An optional way to perform short-time Fourier transform on the time series data in the training dataset to obtain a time-frequency diagram may include the following sub-steps:

[0113] Sub-step 1: For each time series data in the training dataset, use a window function with a finite duration to intercept the vibration signal in the time series data;

[0114] Sub-step 2: Perform Fourier transform on the intercepted vibration signal to obtain the local spectrum in a small range of the corresponding time period;

[0115] The transformation formula for the signal collected by the aging test bench device such as a secondary power supply can be:

[0116]

[0117] In the formula: is the source signal, is the analysis window function; is the time-frequency spectrum at time t.

[0118] Sub-step 3: By moving the window function on the time axis, gradually analyze each vibration signal band to obtain a set of local spectra of the vibration signal, and generate a time-frequency diagram.

[0119] Preferably, the window function is selected as the Hanning window function. The Hanning window function has good smoothness and is suitable for reducing spectral leakage. The expression of the Hanning window function is as follows:

[0120]

[0121] Step 304: Randomly mask each time-frequency diagram and input it into a preset model for training to generate a feature extraction model.

[0122] Through Steps 301 to 304, time-frequency diagrams corresponding to different time segments of multiple key parameters can be obtained. For each time-frequency diagram, the time-frequency diagram can be masked in a random masking manner. Different from time series features and frequency domain features, time series features and frequency domain features are one-dimensional data, while the time-frequency diagram is a two-dimensional picture. The abscissa represents time, the ordinate represents frequency, and the different depths of color represent different frequency amplitudes.

[0123] Send the randomly masked time-frequency diagrams into a preset model such as the Transformer autoencoder for training respectively. After the training is completed, a feature extraction model is generated.

[0124] Transformer is a model established based on the Seq-to-Seq framework. Compared with classical deep learning models, the most prominent advantage of Transformer is the use of the multi-head attention mechanism. The purpose of the multi-head attention layer is to assign different importance to words / tokens in the sequence from multiple aspects. The Transformer model mainly consists of parts such as input, encoder, decoder, and output.

[0125] Position encoding: The position encoding layer is to determine the position information of the sequence. Since there are no recurrent layers and convolutional layers in RNN and CNN, and only relying on the self-attention mechanism cannot obtain the order information of the input, it is necessary to actively transmit the order information of the sequence to the model. The Transformer model uses a combination of sine and cosine functions to perform position encoding on the sequence, and the calculation method is as follows:

[0126]

[0127]

[0128] In the formula: is the position of the current sequence; is the dimension; is the dimension of the input feature.

[0129] Multi-Head Attention Mechanism: The multi-head attention mechanism performs operations in parallel using multiple attention mechanisms, and then stitches together the operation results through a linear transformation. The core technology in Transformer is the multi-head attention mechanism, which is used to extract the dependency relationship features between data, capture the correlation between data, and establish a context prediction model. The calculation method is as follows:

[0130]

[0131]

[0132]

[0133]

[0134]

[0135]

[0136] In the formula: is the query matrix; is the key matrix; is the value matrix; , , is the trainable parameter matrix; is the processed input; is the dimension of the key matrix; , , , is the learnable parameter matrix.

[0137] Feed-Forward Network and Summation and Normalization: In the Transformer model, the encoding part and the decoding part also include a feed-forward network and summation and normalization. The calculation formula of the feed-forward neural network is as follows:

[0138]

[0139] In the formula: is the input; , , , are the parameters that can be obtained through training.

[0140] The calculation formula of summation and normalization is as follows:

[0141]

[0142] In the formula: is the input; is the result after being processed by the module.

[0143] In an alternative embodiment, the method of randomly masking each time-frequency diagram and inputting it into a preset model for training to generate a feature extraction model may include the following sub-steps:

[0144] Sub-step 1: Input the randomly masked time-frequency diagram into the preset model, send it to the encoding module after position encoding, and send the output of the encoding module into the decoder. The decoder restores the feature parameters in the high-dimensional hidden layer into the time-frequency diagram before masking;

[0145] Sub-step 2: Use the restored time-frequency diagram to iteratively train the preset model multiple times to adjust the model parameters;

[0146] Among them, the model parameters include: embedding dimension, number of attention heads, number of encoder layers, and number of decoder layers.

[0147] Sub-step 3: Take out the encoder layer and decoder layer of the model after multiple iterative adjustments, and retain the weight parameters of each layer.

[0148] Sub-step 4: Generate a feature extraction model based on each encoder layer, each decoder layer, and the weight parameters of each layer.

[0149] Step 305: Test the feature extraction model according to the test data set.

[0150] After training (also known as pre-training) the preset model based on the training data set to generate a feature extraction model, the feature extraction model can be tested based on the test data in the test data set to determine the accuracy of the prediction result of the feature extraction model.

[0151] Based on the pre-trained Transformer auto-encoding model for the training data set perform self-attention feature extraction to obtain a self-attention feature set .

[0152] Step 306: After passing the test, input the key parameter time series data of the device to be predicted into the feature extraction model to predict the target time-frequency domain features.

[0153] The method for deeply fusing the time-frequency characteristics of the secondary power supply of the high-temperature aging test equipment provided by the embodiment of the present application collects the fault data set of the devices of the aging test equipment; preprocesses the fault data set to obtain a training data set and a test data set; performs short-time Fourier transform on the time series data of each key parameter in the training data set to obtain a time-frequency diagram; performs random masking on each time-frequency diagram and inputs it into a preset model for training to generate a feature extraction model; tests the feature extraction model based on the test data set; after passing the test, inputs the time series data of the key parameters of the device to be predicted into the feature extraction model to predict the target time-frequency domain features. The method provided by the embodiment of the present invention fully coordinates and refines a large amount of fault data of the devices of the aging test equipment, performs random masking on the time-frequency diagrams obtained by performing short-time Fourier transform on the segmented time series data, and uses the randomly masked time-frequency diagrams to train the preset model, which can fully refine the local semantic information of each part of the aging table device, overcome the restrictions on the training model such as data scarcity and rough annotation, and the time-frequency features predicted by the trained feature extraction model are more accurate and reliable. Further, the health state of the devices of the aging test equipment analyzed based on the predicted time-frequency features is more reliable.

[0154] Figure 4 The structural block diagram of a device for evaluating the health state of the secondary power supply of a high-temperature aging test equipment according to an embodiment of the present application.

[0155] The device for evaluating the health state of the test chamber of the high-temperature aging test equipment provided by the embodiment of the present application includes the following functional modules:

[0156] The acquisition module 401 is used to acquire the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated, where the monitoring signals include: current signal, voltage output signal, power supply ripple volatility, MOSFET device case temperature to junction temperature, etc.;

[0157] The extraction module 402 is used to extract features from the current signal and voltage output signal, and mix the degradation curves such as power supply ripple volatility and MOSFET case temperature to junction temperature to obtain a multi-dimensional feature sequence;

[0158] The prediction module 403 is used to input the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment, where the degradation feature state includes: healthy state and degradation state;

[0159] A determination module 404, configured to determine a state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated based on the probability space points after mapping of the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model when the secondary power supply of the high-temperature aging test equipment is in a degraded state, where the state degradation curve can characterize the health state of the secondary power supply of the high-temperature aging test equipment.

[0160] Optionally, the residual convolutional neural network model includes: a convolutional layer, a batch normalization layer, an activation function, a max pooling layer, a plurality of residual blocks, a global average pooling layer, and a fully connected layer.

[0161] Optionally, the device further includes a training module, configured to train a residual convolutional neural network model, and the training module includes:

[0162] A first sub-module, configured to obtain training time-series degradation feature samples of the secondary power supply of the high-temperature aging test equipment; wherein each training time-series degradation feature sample is labeled with a degradation feature state;

[0163] A second sub-module, configured to input the training time-series degradation feature samples into a pre-established residual convolutional neural network model to obtain first probability space points after mapping;

[0164] A third sub-module, configured to discriminate a degradation starting point of the secondary power supply of the high-temperature aging test equipment according to the two-dimensional first probability space points and generate a health baseline of the secondary power supply of the high-temperature aging test equipment.

[0165] Optionally, the training module further includes:

[0166] A fourth sub-module, configured to obtain test time-series degradation feature samples of the secondary power supply of the high-temperature aging test equipment;

[0167] A fifth sub-module, configured to input the test time-series degradation feature samples into the residual convolutional neural network model under training to obtain second probability space points after mapping;

[0168] A sixth sub-module, configured to calculate a distance between the second probability space and the health baseline based on a preset multi-distance metric algorithm.

[0169] Optionally, the sixth sub-module is specifically configured to:

[0170] Calculate the distances between the second probability space and the health baseline based on the Mahalanobis distance algorithm, the cosine similarity algorithm, and the Manhattan distance algorithm respectively to obtain a first distance, a second distance, and a third distance;

[0171] Perform a weighted sum of the first distance, the second distance, and the third distance to obtain the health degree of the second power supply of the high-temperature aging test equipment.

[0172] The secondary power supply health status evaluation device for high-temperature aging test equipment provided by the embodiments of the present application Figure 4 shown can implement Figure 1 each process implemented by the method embodiments. To avoid repetition, it will not be elaborated here.

[0173] The secondary power supply health status evaluation device for high-temperature aging test equipment provided by the embodiments of the present application collects the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated; generates a multi-dimensional feature sequence based on the monitoring signals; inputs the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment; in the case that the secondary power supply of the high-temperature aging test equipment is in a degraded state, based on the probability space points after mapping of the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model, determines the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated, so as to accurately determine the health status of the secondary power supply of the high-temperature aging test equipment. The secondary power supply health status evaluation device for high-temperature aging test equipment provided by the embodiments of the present invention collects the secondary power supply monitoring signals and extracts the multi-dimensional feature sequence, and predicts the multi-dimensional feature sequence based on the trained model to obtain the state degradation curve of the secondary power supply of the high-temperature aging test equipment to be evaluated, so as to accurately determine the health status of the secondary power supply of the high-temperature aging test equipment.

[0174] The embodiments of the present invention also provide an electronic device, including a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus.

[0175] The memory is used to store a computer program;

[0176] The processor, when executing the program stored on the memory, implements the secondary power supply health status evaluation method for high-temperature aging test equipment shown in the above method embodiments.

[0177] The communication bus mentioned in the above terminal may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0178] The communication interface is used for communication between the above terminal and other devices.

[0179] The memory may include a Random Access Memory (RAM), or may also include a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0180] In another embodiment provided by the present invention, there is also provided a computer-readable storage medium storing instructions, which, when running on an electronic device, enable the electronic device to implement the method for evaluating the health status of the secondary power supply of the high-temperature aging test equipment described in any one of the above embodiments.

[0181] In another embodiment provided by the present invention, there is also provided a computer program product containing instructions, which, when running on an electronic device, enable the electronic device to implement the method for evaluating the health status of the secondary power supply of the high-temperature aging test equipment described in any one of the above embodiments.

[0182] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the presence of additional identical elements in the process, method, article or device including such element.

[0183] The above are the preferred embodiments of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for evaluating the health state of a secondary power supply of a high-temperature aging test equipment, characterized in that, The method includes: Collecting the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated, and extracting the features of the monitoring signals to obtain a multi-dimensional feature sequence; Inputting the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment; When the secondary power supply of the high-temperature aging test equipment is in a degraded state, based on the probability space points after mapping of the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model, determining a state degradation curve that can characterize the health state of the secondary power supply of the high-temperature aging test equipment; Among them, the residual convolutional neural network model is trained in the following manner: Obtaining the training time-series degradation feature samples of the secondary power supply of the high-temperature aging test equipment; inputting the training time-series degradation feature samples into a pre-established residual convolutional neural network model to obtain the first probability space points; discriminating the degradation starting point of the secondary power supply of the high-temperature aging test equipment according to the first probability space points, and generating a health baseline for the secondary power supply of the high-temperature aging test equipment; Obtaining the test time-series degradation feature samples of the secondary power supply of the high-temperature aging test equipment, and inputting the test time-series degradation feature samples into the residual convolutional neural network model under training to obtain the second probability space points after mapping; Calculating the distances between the second probability space points and the health baseline respectively based on the Mahalanobis distance algorithm, the cosine similarity algorithm, and the Manhattan distance algorithm to obtain the first distance, the second distance, and the third distance, and performing weighted summation on the first distance, the second distance, and the third distance to obtain the health degree of the secondary power supply of the high-temperature aging test equipment.

2. The method according to claim 1, wherein The residual convolutional neural network model includes: a convolutional layer, a batch normalization layer, an activation function, a max pooling layer, a plurality of residual blocks, a global average pooling layer, and a fully connected layer.

3. A secondary power supply health state evaluation device for high-temperature aging test equipment, characterized in that, The device includes: A collection module for collecting the monitoring signals of the secondary power supply of the high-temperature aging test equipment to be evaluated; an extraction module for extracting the features of the monitoring signals to obtain a multi-dimensional feature sequence; A prediction module for inputting the multi-dimensional feature sequence into a pre-trained residual convolutional neural network model to obtain the degradation feature state of the secondary power supply of the high-temperature aging test equipment; A determination module for, when the secondary power supply of the high-temperature aging test equipment is in a degraded state, determining a state degradation curve that can characterize the health state of the secondary power supply of the high-temperature aging test equipment based on the probability space points after mapping of the multi-dimensional feature sequence and the health baseline in the residual convolutional neural network model; A training module, configured to obtain training time-series degradation feature samples of the secondary power supply of a high-temperature aging test device; input the training time-series degradation feature samples into a pre-established residual convolutional neural network model to obtain a first probability space point; discriminate the degradation starting point of the secondary power supply of the high-temperature aging test device according to the first probability space point, and generate a health baseline of the secondary power supply of the high-temperature aging test device; obtain test time-series degradation feature samples of the secondary power supply of the high-temperature aging test device, input the test time-series degradation feature samples into the residual convolutional neural network model being trained to obtain a mapped second probability space point; calculate the distances between the second probability space point and the health baseline respectively based on the Mahalanobis distance algorithm, the cosine similarity algorithm and the Manhattan distance algorithm to obtain a first distance, a second distance and a third distance, and perform weighted summation on the first distance, the second distance and the third distance to obtain the health degree of the secondary power supply of the high-temperature aging test device.

4. The device according to claim 3, characterized in that, The residual convolutional neural network model includes: a convolutional layer, a batch normalization layer, an activation function, a max pooling layer, a plurality of residual blocks, a global average pooling layer and a fully connected layer.

5. An electronic device, characterized in that, It includes a processor, a communication interface, a memory and a communication bus. Among them, the processor, the communication interface and the memory complete mutual communication through the communication bus; The memory is used to store computer programs; When the processor is used to execute the programs stored on the memory, it implements the method for evaluating the health state of the secondary power supply of the high-temperature aging test device according to any one of claims 1-2.

Images