Electromagnetic compatibility intelligent test and diagnosis system and method

Abstract

The invention relates to the field of electromagnetic compatibility testing, and particularly discloses an electromagnetic compatibility intelligent test diagnosis system and method, and the method comprises the steps: firstly obtaining an electromagnetic emission spectrogram of a tested object, and collecting current values and voltage values of the tested object at a plurality of preset time points through a sensor, and then carrying out the deep learning technology, the three components are subjected to feature extraction and correlation analysis, and finally an electromagnetic interference grade classification label is obtained through a classifier, so that site limitation is reduced, the test flexibility is improved, valuable insight is provided for product design and improvement, and design optimization and performance improvement are promoted.

Description

Technical Field

[0001] The present application relates to the field of electromagnetic compatibility testing, and more specifically, to an electromagnetic compatibility intelligent testing and diagnosis system and method. Background Art

[0002] Electromagnetic compatibility refers to the coexistence of electrical and electronic equipment in a common electromagnetic environment, which requires that the above-mentioned various devices in the same electromagnetic environment can work normally and do not interfere with each other, achieving a "compatible" state. When a device, equipment or system is working, it is inevitable to generate external electromagnetic emissions. When this emission exceeds a certain tolerance, it may cause the performance of other devices, equipment or systems in the environment to deteriorate, that is, electromagnetic interference occurs.

[0003] In order to check whether the electromagnetic emission of the test product exceeds the standard, electromagnetic interference testing is required. At present, the most commonly used laboratory method for electromagnetic interference testing is to use a microwave darkroom. However, the construction cost of a microwave darkroom is high and the requirements for the test site and test conditions are strict.

[0004] Therefore, an electromagnetic compatibility intelligent testing and diagnosis system and method are desired. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides an electromagnetic compatibility intelligent test and diagnosis system and method, which first obtains the electromagnetic emission spectrum of the test product, the current value and voltage value of the test product at multiple predetermined time points collected by the sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally obtains the electromagnetic interference level classification label through a classifier, thereby reducing site restrictions and increasing the flexibility of testing, thereby providing valuable insights into product design and improvement, and promoting design optimization and performance improvement.

[0006] According to one aspect of the present application, an electromagnetic compatibility intelligent test and diagnosis system is provided, which includes:

[0007] The electromagnetic compatibility test data acquisition module is used to obtain the electromagnetic emission spectrum of the test product, and the current value and voltage value of the test product at multiple predetermined time points collected by the sensor;

[0008] An electromagnetic compatibility test data processing module, used to extract an electromagnetic emission spectrum enhanced attention feature vector and a test product current-voltage time domain waveform feature vector from the electromagnetic emission spectrum diagram of the test product and the current values ​​and voltage values ​​of the test product at multiple predetermined time points collected by the sensor;

[0009] The electromagnetic interference level label analysis module is used to obtain an electromagnetic interference level classification label based on the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector.

[0010] According to another aspect of the present application, there is provided an electromagnetic compatibility intelligent test and diagnosis method, which includes:

[0011] Obtaining the electromagnetic emission spectrum of the test object, and the current values ​​and voltage values ​​of the test object collected by the sensor at multiple predetermined time points;

[0012] Extracting an electromagnetic emission spectrum enhanced attention feature vector and a test product current-voltage time domain waveform feature vector from the electromagnetic emission spectrum diagram of the test product and the current values ​​and voltage values ​​of the test product at multiple predetermined time points collected by the sensor;

[0013] Based on the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector, an electromagnetic interference level classification label is obtained.

[0014] Compared with the prior art, the present application provides an electromagnetic compatibility intelligent testing and diagnosis system and method, which first obtains the electromagnetic emission spectrum diagram of the test product, the current value and voltage value of the test product at multiple predetermined time points collected by the sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally uses a classifier to obtain an electromagnetic interference level classification label, thereby reducing site restrictions and increasing testing flexibility, thereby providing valuable insights into product design and improvement, and promoting design optimization and performance improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 4 is a block diagram of an electromagnetic compatibility intelligent test and diagnosis system according to an embodiment of the present application.

[0017] Figure 2 4 is a block diagram of an electromagnetic compatibility test data processing module in an electromagnetic compatibility intelligent test and diagnosis system according to an embodiment of the present application.

[0018] Figure 3 4 is a block diagram of an electromagnetic emission spectrum feature extraction unit in an electromagnetic compatibility intelligent test and diagnosis system according to an embodiment of the present application.

[0019] Figure 4 4 is a block diagram of an electromagnetic interference level label analysis module in an electromagnetic compatibility intelligent test and diagnosis system according to an embodiment of the present application.

[0020] Figure 5 Flow chart of an electromagnetic compatibility intelligent testing and diagnosis method according to an embodiment of the present application.

[0021] Figure 6 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0023] Figure 1 FIG. 1 is a block diagram of an electromagnetic compatibility intelligent test and diagnosis system according to an embodiment of the present application. Figure 1 As shown, according to the electromagnetic compatibility intelligent test and diagnosis system 100 of the embodiment of the present application, it includes: an electromagnetic compatibility test data acquisition module 110, which is used to obtain the electromagnetic emission spectrum diagram of the test product, the current value and the voltage value of the test product at multiple predetermined time points collected by the sensor; an electromagnetic compatibility test data processing module 120, which is used to extract the electromagnetic emission spectrum enhanced attention feature vector and the test product current-voltage time domain waveform feature vector from the electromagnetic emission spectrum diagram of the test product and the current value and voltage value of the test product at multiple predetermined time points collected by the sensor; an electromagnetic interference level label analysis module 130, which is used to obtain an electromagnetic interference level classification label based on the electromagnetic emission spectrum enhanced attention feature vector and the test product current-voltage time domain waveform feature vector.

[0024] In the above-mentioned electromagnetic compatibility intelligent test and diagnosis system 100, the electromagnetic compatibility test data acquisition module 110 is used to obtain the electromagnetic emission spectrum of the test product, the current value and voltage value of the test product at multiple predetermined time points collected by the sensor. It should be understood that electromagnetic compatibility refers to the ability of electrical and electronic equipment to operate normally in the same electromagnetic environment without interfering with each other, ensuring that these devices can achieve a "compatible" state in a common environment. During the operation of the equipment, electromagnetic emissions will inevitably be generated. If these emissions exceed a certain threshold, the performance of other equipment may be degraded. This phenomenon is called electromagnetic interference. In order to detect whether the equipment exceeds the standard of electromagnetic emission, an electromagnetic interference test is usually required. At present, the most commonly used test method is to use a microwave darkroom, but the construction cost of this darkroom is high, and there are strict requirements on the experimental site and conditions. Therefore, in the technical solution of the present application, by obtaining the electromagnetic emission spectrum of the test product, the current value and voltage value of the test product at multiple predetermined time points collected by the sensor, and combining deep learning technology, the electromagnetic interference level classification label is obtained, thereby reducing the restrictions on the test site and improving the flexibility of the test. Provide valuable insights for product design and improvement, driving design optimization and performance improvements.

[0025] Specifically, the electromagnetic emission spectrum diagram records the electromagnetic emission intensity of the device at different frequencies, which can help engineers identify and analyze potential sources of electromagnetic interference. By analyzing these spectrum diagrams, it is possible to detect whether the device exceeds the electromagnetic emission standard or whether it will cause interference to other devices. The current and voltage values ​​provide the electrical characteristics of the device when it is working. By collecting this data at multiple predetermined time points, the current and voltage fluctuations of the device under different working conditions can be monitored. These fluctuations may be related to the intensity of electromagnetic emissions, especially when the device load changes or the operating conditions are different. This data is very important for understanding how electrical characteristics affect electromagnetic emissions, and helps to identify the performance and potential problems of the device under different working conditions. Combining the electromagnetic emission spectrum diagram with current and voltage data for analysis can provide more comprehensive electromagnetic compatibility test results. This comprehensive approach not only helps to identify and locate electromagnetic interference problems, but also guides design optimization to improve the electromagnetic compatibility of equipment.

[0026] In the above-mentioned electromagnetic compatibility intelligent test and diagnosis system 100, the electromagnetic compatibility test data processing module 120 is used to extract the electromagnetic emission spectrum enhanced attention feature vector and the test product current-voltage time domain waveform feature vector from the electromagnetic emission spectrum diagram of the test product, the current value and voltage value of the test product at multiple predetermined time points collected by the sensor. It should be understood that by combining these two feature vectors, a more comprehensive electromagnetic compatibility analysis framework can be established. This method can not only provide an in-depth understanding of the electromagnetic emission characteristics of the equipment, but also reveal the relationship between electrical characteristics and electromagnetic interference. Ultimately, this data-driven analysis strategy helps to improve test accuracy, optimize design, and ensure the stability and reliability of equipment in complex electromagnetic environments.

[0027] Figure 2 FIG. 1 is a block diagram of an electromagnetic compatibility test data processing module in an electromagnetic compatibility intelligent test and diagnosis system according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the electromagnetic compatibility test data processing module 120 includes: an electromagnetic emission spectrum feature extraction unit 121, used to perform feature extraction on the electromagnetic emission spectrum diagram of the test product to obtain the electromagnetic emission spectrum enhanced attention feature vector; a test product time domain waveform feature extraction unit 122, used to perform feature extraction on the current values ​​and voltage values ​​of the test product at multiple predetermined time points collected by the sensor to obtain the current-voltage time domain waveform feature vector of the test product.

[0028] It should be understood that the electromagnetic emission spectrum graph usually contains a large amount of data that shows the emission characteristics of the device in different frequency ranges. However, not all the emission data of the frequencies in the entire spectrum graph are equally important for interference analysis. The emission signals of certain specific frequencies may have a greater impact on the electromagnetic compatibility of the device, so they need to be focused on. Through feature extraction, the spectrum graph can be simplified into a set of feature vectors that enhance attention. These vectors focus on key frequency bands and emission intensities, making the analysis work more efficient and targeted. Feature extraction helps to filter out the main frequency components related to device performance and interference from the spectrum graph. These components are usually closely related to the design, working status or electromagnetic compatibility standards of the device. By extracting and analyzing these key features, engineers can more easily identify potential sources of interference and make targeted improvements or optimization measures. For example, if the feature vector shows that there is a significant increase in emission intensity at certain frequency points, engineers can further examine these frequency points to determine whether it is necessary to adjust the device design or take shielding measures.

[0029] Furthermore, the time domain waveforms of current and voltage provide detailed information about the electrical performance of the device during operation. By extracting feature vectors from these waveform data, the electrical characteristics of the device under different operating conditions, such as current fluctuations, transient response, and voltage stability, can be captured. These feature vectors can help identify whether the device has electrical anomalies or noise problems, which may affect the normal operation of the device or cause electromagnetic interference. Feature extraction can simplify and quantify complex time domain data. Raw current and voltage data usually contain a lot of noise and irregular fluctuations, which makes it difficult to analyze this data directly. By extracting key time domain features, such as peak value, RMS value, spectral characteristics, etc., complex waveform data can be converted into easy-to-process feature vectors. These feature vectors not only reduce the complexity of data processing, but also improve the ability to detect electrical performance problems.

[0030] Figure 3 FIG. 1 is a block diagram of an electromagnetic emission spectrum feature extraction unit in an electromagnetic compatibility intelligent test and diagnosis system according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the electromagnetic emission spectrum feature extraction unit 121 includes: an electromagnetic emission spectrum feature encoding subunit 1211, used to perform feature encoding on the electromagnetic emission spectrum graph of the test sample to obtain an electromagnetic emission spectrum enhanced attention feature graph; an electromagnetic emission spectrum pooling subunit 1212, used to perform maximum value pooling on the electromagnetic emission spectrum enhanced attention feature graph to obtain the electromagnetic emission spectrum enhanced attention feature vector.

[0031] It should be understood that electromagnetic emission spectrum diagrams show the emission intensity information of the device at different frequencies, but these spectrum diagrams usually contain a large amount of data, covering multiple frequencies and intensity levels. By feature encoding the spectrum diagram, the frequency components and intensity changes that are most critical to electromagnetic compatibility can be extracted and emphasized. This encoding process usually involves compressing the original spectrum data into an image or vector with specific characteristics, thereby highlighting the emission characteristics of the key frequency band, making the identification of interference sources more focused and clear. Feature encoding helps to simplify and standardize the spectrum data for further analysis. The encoding process usually involves converting the spectrum diagram into an enhanced attention feature diagram, a graphical representation that can convert complex data structures into a more concise form. The enhanced attention feature diagram makes it easier for engineers to focus on key issues by highlighting important frequency areas and emission intensity changes. This not only improves the efficiency of analysis, but also enhances the ability to detect potential interference sources.

[0032] Furthermore, the electromagnetic emission spectrum enhanced attention feature map is subjected to maximum pooling to extract the most representative features from the complex spectrum feature map, thereby improving the efficiency and accuracy of data analysis. Among them, the electromagnetic emission spectrum enhanced attention feature map shows the emission intensity information of the device at different frequencies after feature encoding. These feature maps usually contain a lot of detailed information, but for actual analysis, not all details are equally important. Max pooling is a dimensionality reduction technique that converts high-dimensional data into low-dimensional feature vectors by extracting the maximum value from each local area of ​​the feature map. The advantage of this is that it can retain the most significant features while reducing the complexity and computational burden of the data. Max pooling can effectively highlight the key features in the electromagnetic emission spectrum. There may be multiple frequency regions in the spectrum map, and the emission intensity of some frequency points is significantly higher than that of other parts. Through maximum pooling, the strongest emission signal can be extracted from each local area, so that these important features can be summarized into the feature vector. This method ensures that the most significant information of electromagnetic emission is retained in the feature vector, which helps to more accurately identify and analyze potential electromagnetic interference sources.

[0033] In a specific embodiment of the present application, the electromagnetic emission spectrum feature encoding subunit 1211 includes: using a deep learning network model to reduce the noise of the electromagnetic emission spectrum graph of the test sample to obtain an electromagnetic emission spectrum enhanced image; passing the electromagnetic emission spectrum enhanced image through a convolutional neural network model based on the electromagnetic emission spectrum spatial attention mechanism to obtain the electromagnetic emission spectrum enhanced attention feature graph.

[0034] It should be understood that electromagnetic emission spectrograms are often affected by various noises in actual measurements, which may come from electronic noise of the measuring equipment, environmental interference, or loss during signal transmission. These noises can mask or distort the real signal of electromagnetic emission, making it difficult to analyze the spectrogram. Deep learning network models can automatically identify and suppress noise by learning the relationship between a large amount of noisy data and clean data, thereby generating clearer and more accurate enhanced images of electromagnetic emission spectra. Deep learning models, especially convolutional neural networks (CNNs), perform well when processing image data. These networks can automatically extract multi-level features from spectrum images to identify and remove different types of noise. For example, deep learning models can identify and remove background noise, smooth spikes in the spectrogram, and retain important spectral features. This adaptive noise suppression capability makes the final spectrum image more accurately reflect the actual electromagnetic emission characteristics of the device.

[0035] Furthermore, after the electromagnetic emission spectrum enhanced image is processed by noise reduction, it may still contain a large amount of spectrum information, some of which is crucial for interference analysis, while other parts may not be significant. The convolutional neural network model based on the electromagnetic emission spectrum spatial attention mechanism can dynamically adjust the degree of attention to different spectrum regions by introducing the attention mechanism. This mechanism automatically focuses on those areas that are most critical to the electromagnetic emission feature analysis by calculating the weights of each area, thereby improving the analysis accuracy of the model. The spatial attention mechanism highlights the important areas in the image by generating an attention map in CNN. Specifically, the attention mechanism analyzes the spatial distribution of the spectrum image and assigns different attention weights to each area. This processing method enables the model to better capture the significant features in the electromagnetic emission spectrum, such as frequency spikes or abnormal signals, while suppressing irrelevant background information or noise. This method ensures that the most representative electromagnetic emission features are included in the final feature map, thereby improving the effectiveness of data analysis. The convolutional neural network model based on the spatial attention mechanism can process spectrum images with complex spatial structures. Through the combination of multi-layer convolution operations and attention mechanisms, the model can extract and integrate features layer by layer, and gradually enhance the recognition ability of key spectrum information. This multi-level feature extraction and weighted processing enables the final electromagnetic emission spectrum enhanced attention feature map to more accurately reflect the actual situation of electromagnetic emission, and facilitates further interference source location and analysis. Specifically, the electromagnetic emission spectrum enhanced image is deeply convolutionally encoded using the convolutional coding part of the convolutional neural network model based on the electromagnetic emission spectrum spatial attention mechanism to obtain an initial convolutional feature map; the initial convolutional feature map is input into the spatial attention part of the convolutional neural network model based on the electromagnetic emission spectrum spatial attention mechanism to obtain a spatial attention map; the spatial attention map is passed through a Softmax activation function to obtain a spatial attention feature map; and the spatial attention feature map and the initial convolutional feature map are multiplied by the position points to obtain the electromagnetic emission spectrum enhanced attention feature map. More specifically, the initial convolution feature map is input into the spatial attention part of the convolution neural network model based on the electromagnetic emission spectrum spatial attention mechanism to obtain a spatial attention map, including: performing average pooling and maximum pooling along the channel dimension on the initial convolution feature map to obtain an average feature matrix and a maximum feature matrix; cascading and channel-adjusting the average feature matrix and the maximum feature matrix to obtain a channel feature matrix; and using the convolution layer of the spatial attention feature map to perform convolution encoding on the channel feature matrix to obtain the spatial attention map.

[0036] In a specific embodiment of the present application, a deep learning network model is used to reduce the noise of the electromagnetic emission spectrum diagram of the test object to obtain an electromagnetic emission spectrum enhanced image, including: passing the electromagnetic emission spectrum diagram of the test object through an electromagnetic emission spectrum denoising module based on an autoencoder to obtain the electromagnetic emission spectrum enhanced image.

[0037] It should be understood that the autoencoder is a neural network architecture that includes two main parts: an encoder and a decoder. The encoder is responsible for compressing the input electromagnetic emission spectrum into a low-dimensional feature representation, while the decoder reconstructs this low-dimensional representation into the original image. Through this compression and reconstruction process, the autoencoder can learn the main features of the spectrum and effectively remove those irrelevant noises. In the application of electromagnetic emission spectrum denoising, the autoencoder can remove the noise or interference in the measurement process by capturing the inherent structure and pattern of the spectrum data. The training process of the autoencoder is optimized based on the difference between the original spectrum and the clean spectrum. By minimizing the error between the input image and the reconstructed image, the autoencoder can gradually learn how to extract the most realistic signal from the noisy spectrum. This training method enables the autoencoder to effectively separate the noise from the actual electromagnetic emission signal, thereby generating a clearer electromagnetic emission spectrum enhanced image. Specifically, the electromagnetic emission spectrum diagram of the test object is input into the encoder of the electromagnetic emission spectrum denoising module based on the autoencoder, wherein the encoder uses a convolution layer to perform explicit spatial encoding on the electromagnetic emission spectrum diagram of the test object to obtain image features; and the image features are input into the decoder of the electromagnetic emission spectrum denoising module based on the autoencoder, wherein the decoder uses a deconvolution layer to perform deconvolution processing on the image features to obtain the electromagnetic emission spectrum enhanced image.

[0038] In a specific embodiment of the present application, the electromagnetic emission spectrum pooling subunit 1213 includes: using Spring Initializr to create a Spring Boot project and adding Apache Commons Math dependencies; create the class AttentionFeatureMap that represents the electromagnetic emission spectrum enhanced attention feature map and the class AttentionFeatureVector that represents the electromagnetic emission spectrum enhanced attention feature vector, and define their properties and methods; write the method maxPooling to receive the AttentionFeatureMap object, traverse each row of the electromagnetic emission spectrum enhanced attention feature map, find the maximum value of each row, store these maximum values ​​in an array, and finally return an AttentionFeatureVector object; use Spring's dependency injection mechanism to manage services and components, create a service class AttentionService, which contains the maximum pooling method maxPooling, and mark it with Spring's @Service annotation; create a REST controller AttentionController, define the HTTP request path through Spring's @RestController and @RequestMapping annotations, and use the @Autowired annotation to inject AttentionService in the controller, define the POST request method maxPooling through the @PostMapping annotation, receive HTTP requests and call service methods; use Gradle to build the project and deploy it on the server.

[0039] It should be understood that SSM (Spring, Spring MVC, MyBatis) is a popular framework combination for building Java enterprise applications. The Spring framework is responsible for providing comprehensive dependency injection and management functions to simplify the development of enterprise applications; Spring MVC handles requests and responses at the Web layer, supports the MVC (Model-View-Controller) design pattern, and realizes the decoupling of front-end requests and back-end business logic; MyBatis is responsible for the operation of the data persistence layer, providing flexible SQL mapping functions, making database operations simpler and more efficient. The integration of the SSM framework enables developers to build, maintain, and expand complex Java application systems more efficiently.

[0040] Specifically, some examples of deployment code are as follows.

[0041]

[0042]

[0043]

[0044] In this way, the maximum value pooling of the electromagnetic emission spectrum enhanced attention feature map is performed using the SSM mechanism, realizing the whole process from data structure definition to final deployment, ensuring the accuracy of the algorithm and the scalability of the system. First, the environment preparation module ensures that the development environment has the necessary tools and libraries, such as JDK, Maven or Gradle, Apache Commons Math and Spring Boot. These tools and libraries provide a solid foundation for subsequent development and deployment. Next, the project creation module uses SpringInitializr or manually to create a new Spring Boot project and add the necessary dependencies. This step simplifies the initialization process of the project, allowing developers to quickly enter the development of core functions. In the data structure definition module, two key classes are defined: AttentionFeatureMap and AttentionFeatureVector. These two classes represent the electromagnetic emission spectrum enhanced attention feature map and feature vector, respectively, and provide data structure support for subsequent algorithm implementation. The maximum value pooling implementation module is one of the core modules. By writing the maxPooling method, the maximum value pooling operation of the feature map is implemented. This method traverses each row of the feature map, finds the maximum value of each row, stores these maximum values ​​in an array, and finally returns a feature vector object. This process ensures the accuracy and efficiency of the algorithm. The SSM mechanism integration module uses Spring's dependency injection mechanism to manage services and components, creates the AttentionService class, and marks it with the @Service annotation. This step improves the maintainability and scalability of the code, making the system more flexible. The controller writing module creates the REST controller AttentionController, defines the HTTP request path through the @RestController and @RequestMapping annotations, and injects the AttentionService into the controller using the @Autowired annotation. The POST request method maxPooling is defined through the @PostMapping annotation to receive HTTP requests and call service methods. This step enables the system to provide services externally. Finally, the test and deployment module verifies the correctness of the algorithm by writing unit tests, builds the project using Maven or Gradle, and finally deploys it to the server. This step ensures the stability and reliability of the system. In this way, the maximum pooling function of the electromagnetic emission spectrum enhanced attention feature map is realized, and it has the characteristics of high efficiency, accuracy, and scalability. Whether in terms of development efficiency or system performance, the system has shown significant advantages.

[0045] In a specific embodiment of the present application, the test product time domain waveform feature extraction unit 122 includes: extracting the test product current-voltage time domain waveform diagram from the current values ​​and voltage values ​​of the test product at multiple predetermined time points collected by the sensor; passing the test product current-voltage time domain waveform diagram through the test product current-voltage time domain waveform feature encoder to obtain the test product current-voltage time domain waveform feature vector.

[0046] It should be understood that the current-voltage time domain waveform can provide information about the current and voltage changes of the device at different time points. By recording the current and voltage values ​​at multiple predetermined time points, a current-voltage relationship diagram of the device over time can be drawn. This time domain waveform intuitively shows the changing trend, fluctuation amplitude and mutual relationship of the current and voltage of the device during operation, which helps to identify the working status and potential problems of the device. For example, the time domain waveform can reveal the current-voltage response characteristics of the device during startup, operation or load change, helping engineers understand the behavior mode of the device under actual working conditions. The current-voltage time domain waveform is very important for detecting the stability and performance of the device. By analyzing the current and voltage fluctuations in the waveform, it can be found whether the device has abnormalities, such as overcurrent, overvoltage or oscillation. These abnormalities may indicate that the device has design defects, aging problems or unsuitable working environment. Timely identification of these problems can prevent equipment failures and improve the reliability and life of the equipment. The time domain waveform can also be used to optimize the performance of the device. By comparing the current-voltage waveforms under different working conditions, the performance of the device under different loads or operating modes can be evaluated, and the design and control strategy of the device can be optimized. For example, during the debugging phase, by adjusting the parameters of the equipment and observing the changes in the time domain waveform, engineers can find the optimal operating parameters to improve the efficiency and stability of the equipment.

[0047] Furthermore, the current-voltage time-domain waveform contains the current and voltage change information of the device at different time points, which may be very complex and lengthy. Directly processing these original waveforms may lead to inefficient calculation and analysis. Through the feature encoder, these time-domain waveforms are converted into feature vectors, which can compress the complex time-domain data into a relatively concise, high-dimensional representation. This feature vector can retain the key information in the original waveform while greatly reducing the complexity of the data, making subsequent analysis and model training more efficient and manageable. The feature encoder can highlight the key current and voltage change patterns by learning and extracting important features in the waveform. Specifically, each layer of the test product current-voltage time-domain waveform feature encoder is used to perform convolution processing, mean pooling processing based on the local feature matrix, and nonlinear activation processing on the input data in the forward transfer of the layer to output the test product current-voltage time-domain waveform feature vector by the last layer of the test product current-voltage time-domain waveform feature encoder, wherein the input of the test product current-voltage time-domain waveform feature encoder is the test product current-voltage time-domain waveform.

[0048] In the above-mentioned electromagnetic compatibility intelligent test and diagnosis system 100, the electromagnetic interference level label analysis module 130 is used to obtain the electromagnetic interference level classification label based on the electromagnetic emission spectrum enhanced attention feature vector and the current-voltage time domain waveform feature vector of the test object. It should be understood that the electromagnetic emission spectrum enhanced attention feature vector and the current-voltage time domain waveform feature vector provide detailed information on the equipment operation status and electromagnetic interference from different perspectives. The electromagnetic emission spectrum enhanced feature vector captures the key features of the equipment in electromagnetic interference by analyzing the electromagnetic emission spectrum diagram, such as spikes in the spectrum, noise distribution, etc. These features are crucial for evaluating the intensity and characteristics of electromagnetic interference. The current-voltage time domain waveform feature vector reflects the current and voltage response characteristics of the equipment under actual working conditions, including the dynamic behavior of the equipment, load changes, etc. These two feature vectors provide a comprehensive perspective for electromagnetic interference analysis. Through the classification label based on the feature vector, targeted interference suppression and management strategies can be formulated. For example, by identifying equipment or environments with high interference levels, maintenance, design adjustment or shielding measures can be prioritized. This strategy helps to improve the reliability and performance of equipment and reduce the negative impact of electromagnetic interference on the system.

[0049] Figure 4 FIG. 1 is a block diagram of an electromagnetic interference level label analysis module in an electromagnetic compatibility intelligent test and diagnosis system according to an embodiment of the present application. Figure 4As shown, in a specific embodiment of the present application, the electromagnetic interference level label analysis module 130 includes: an electromagnetic interference data feature fusion unit 131, which is used to perform perspective projection matching on the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector based on feature fine-grained association to obtain an electromagnetic interference level label classification feature vector; an electromagnetic interference level label classification unit 132, which is used to pass the electromagnetic interference level label classification feature vector through a classifier to obtain a classification result, and the classification result is used to represent the electromagnetic interference level classification label.

[0050] It should be understood that by fusing the electromagnetic emission spectrum enhanced attention feature vector and the current-voltage time domain waveform feature vector, the electromagnetic interference level label classification feature vector can be obtained, which can provide a multi-dimensional and comprehensive interference analysis tool. This method not only improves the accuracy of electromagnetic interference assessment, but also supports more efficient interference management and equipment optimization strategies.

[0051] In particular, in the technical solution of the present application, considering that the electromagnetic emission spectrum enhanced attention feature vector is extracted from the electromagnetic emission spectrum diagram of the test object, the main focus is on the frequency domain characteristics of the electromagnetic emission. The test object current-voltage time domain waveform feature vector is extracted from the current value and voltage value of the test object, and the main focus is on the time domain characteristics of the current and voltage. Since the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector are extracted from different physical properties and processing paths, they represent the information of the test object in different aspects. Due to the essential differences in these features and the different extraction methods, these two feature vectors have significant feature heterogeneity. When these two heterogeneous feature vectors are fused, the feature manifold and feature space mismatch caused by heterogeneity will cause the electromagnetic interference level label classification feature vector to fail to have feature manifold smoothness, thereby affecting the certainty of its feature expression. Specifically, the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector may exist on different manifolds, which means that they have different distribution laws in their respective spaces. Directly splicing them together may cause the final feature vector to have no clear manifold in the new space, that is, the feature manifold smoothness is insufficient. Since the two feature vectors come from different sources, they may not be well mapped to each other in the fused feature space, that is, one type of feature may not accurately reflect the feature changes of another type, resulting in increased uncertainty in feature expression. In response to the above technical problems, in the technical solution of the present application, the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector are matched based on feature fine-grained association of perspective projection.

[0052] Specifically, in an embodiment of the present application, the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector are matched with each other based on feature fine-grained association to obtain an electromagnetic interference level label classification feature vector, including: constructing an electromagnetic interference distance topology matrix between the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector; constructing an electromagnetic interference eigenvalue granularity association matrix between the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector; and matching the electromagnetic interference based on the electromagnetic interference distance topology matrix. The eigenvalue granularity association matrix is ​​topologically associated modulated to obtain an electromagnetic interference topological modulation association matrix; the electromagnetic interference topological modulation association matrix is ​​used as the perspective co-projection space, and the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector are projected into the perspective co-projection space to obtain the perspective modulated electromagnetic emission spectrum enhanced attention feature vector and the perspective modulated test object current-voltage time domain waveform feature vector; the perspective modulated electromagnetic emission spectrum enhanced attention feature vector and the perspective modulated test object current-voltage time domain waveform feature vector are fused to obtain the electromagnetic interference level label classification feature vector.

[0053] In an embodiment of the present application, another implementation expression of performing perspective projection matching based on feature fine-grained association on the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector to obtain the electromagnetic interference level label classification feature vector can be: the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector are processed by the following formula to obtain the electromagnetic interference level label classification feature vector; wherein the formula is:

[0054] M t =D(V1,V2)

[0055]

[0056]

[0057] V c =α·V1′+β·V2′

[0058] Wherein, V1 represents the electromagnetic emission spectrum enhanced attention feature vector, V2 represents the current-voltage time domain waveform feature vector of the test object, and D(V1, V2) is the distance topological matrix between the electromagnetic emission spectrum enhanced attention feature vector and the current-voltage time domain waveform feature vector of the test object, that is, d i,j =d(v 1i ,v 2j ), d i,jis the eigenvalue of the position (i, j) in the distance topology matrix, d(v 1i ,v 2j ) represents the distance between the eigenvalue of the i-th position of the electromagnetic emission spectrum enhanced attention feature vector and the eigenvalue of the j-th position of the test sample current-voltage time domain waveform feature vector, and V1 and V2 are both column vectors, M t represents the distance topology matrix, T represents the transpose of the vector, Represents matrix multiplication, M v represents the eigenvalue granularity correlation matrix, exp represents the natural exponential function, M c represents the topological modulation correlation matrix, V1' represents the feature vector of the electromagnetic emission spectrum enhanced attention by visual angle modulation, V2' represents the feature vector of the current-voltage time domain waveform of the test object modulated by visual angle, α and β represent weighted hyperparameters, V c Represents the electromagnetic interference level label classification feature vector.

[0059] That is, in the technical solution of the present application, the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector are matched by perspective projection based on feature fine-grained association, and the electromagnetic interference distance topological matrix between the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector is first constructed. This step utilizes the idea of ​​graph theory, taking the eigenvalues ​​of each position in the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector as nodes, and using the distance metric function to determine the low-dimensional embedding expression of the edges between nodes, so as to construct a feature expression that can quantify the topological association relationship between the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector. This step is crucial to understanding the spatial distribution of data points and their mutual connections, and also provides basic structural information for subsequent steps.

[0060] Next, an electromagnetic interference eigenvalue granularity association matrix is ​​constructed between the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector. This step focuses on exploring the correlation between the internal attributes of the feature vector. In a specific example, the product between the electromagnetic emission spectrum enhanced attention feature vector and the transposed vector of the test object current-voltage time domain waveform feature vector is calculated to obtain the electromagnetic interference eigenvalue granularity association matrix. Furthermore, the electromagnetic interference eigenvalue granularity association matrix is ​​topologically modulated based on the electromagnetic interference distance topology matrix to obtain the electromagnetic interference topology modulation association matrix. That is, the eigenvalue granularity association information of the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector is driven to wander on the distance topology map to construct a low-dimensional modulation perspective co-projection space domain for mapping the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector to a continuous high-dimensional regression space attribute.

[0061] Next, the electromagnetic interference topological modulation correlation matrix is ​​used as the perspective co-projection space, and the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector are projected into the perspective co-projection space to obtain the perspective modulated electromagnetic emission spectrum enhanced attention feature vector and the perspective modulated test object current-voltage time domain waveform feature vector. That is, the key to this technical solution is to find an implicit third space with modulation capability, in which as much information of the original high-dimensional data as possible is retained, so that the converted new feature vector not only reduces the dimension, but also maintains the relative position relationship between the original features.

[0062] Finally, the perspective-modulated electromagnetic emission spectrum enhanced attention feature vector and the perspective-modulated test sample current-voltage time domain waveform feature vector are fused to obtain the electromagnetic interference level label classification feature vector. Fusion can be achieved in a variety of forms, ranging from simple arithmetic averages to complex ensemble learning algorithms, depending on the application scenario and the goals pursued. The final electromagnetic interference level label classification feature vector combines the main features of the two sets of input data and provides a more comprehensive data representation, which is not only conducive to improving the performance of machine learning models, but also promotes the ability of cross-modal or multi-source data analysis, thereby significantly enhancing the understanding and utilization efficiency of complex data structures.

[0063] Furthermore, the electromagnetic interference level label classification feature vector integrates a variety of electromagnetic interference related information, including electromagnetic emission spectrum characteristics and current-voltage time domain response. These feature vectors provide a comprehensive perspective on the electromagnetic environment and interference characteristics of the device. However, the simple feature vector itself is often difficult to use directly for actual decision-making, and further processing is required to generate actionable classification results. By training a large amount of labeled sample data, the classifier can grasp the characteristic patterns and distribution characteristics of different interference levels. The electromagnetic interference level label classification feature vector is input into the classifier, which analyzes and processes the input features according to its learned pattern and finally outputs a classification result. This classification result corresponds to a specific level of electromagnetic interference, such as low, medium, and high. This process not only automates the identification of interference levels, but also improves the accuracy and efficiency of evaluation. The results generated by the classifier can clearly identify the severity of electromagnetic interference, support more scientific and effective interference management strategies, and thus improve the performance of the equipment and the overall stability of the system.

[0064] In summary, the embodiment of the present application first obtains the electromagnetic emission spectrum diagram of the test product, the current value and voltage value of the test product at multiple predetermined time points collected by the sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally uses a classifier to obtain an electromagnetic interference level classification label, thereby reducing site restrictions and increasing test flexibility, thereby providing valuable insights into product design and improvement, and promoting design optimization and performance improvement.

[0065] As described above, the electromagnetic compatibility intelligent test and diagnosis system 100 according to the embodiment of the present application can be implemented in various terminal devices. In one example, the electromagnetic compatibility intelligent test and diagnosis system 100 can be integrated into the terminal device as a software module and / or a hardware module. For example, the electromagnetic compatibility intelligent test and diagnosis system 100 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the electromagnetic compatibility intelligent test and diagnosis system 100 can also be one of the many hardware modules of the terminal device.

[0066] Alternatively, in another example, the electromagnetic compatibility intelligent test and diagnosis system 100 and the terminal device may also be separate devices, and the electromagnetic compatibility intelligent test and diagnosis system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in a predetermined data format.

[0067] Figure 5 FIG. 1 is a flow chart of an electromagnetic compatibility intelligent test and diagnosis method according to an embodiment of the present application. Figure 5As shown, according to the electromagnetic compatibility intelligent testing and diagnosis method of the embodiment of the present application, it includes: S110, obtaining the electromagnetic emission spectrum diagram of the test product, the current value and the voltage value of the test product at multiple predetermined time points collected by the sensor; S120, extracting the electromagnetic emission spectrum enhanced attention feature vector and the test product current-voltage time domain waveform feature vector from the electromagnetic emission spectrum diagram of the test product and the current value and voltage value of the test product at multiple predetermined time points collected by the sensor; S130, obtaining the electromagnetic interference level classification label based on the electromagnetic emission spectrum enhanced attention feature vector and the test product current-voltage time domain waveform feature vector.

[0068] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned electromagnetic compatibility intelligent test and diagnosis method have been referred to above. Figures 1 to 4 The electromagnetic compatibility intelligent test and diagnosis system has been described in detail, and therefore, its repeated description will be omitted.

[0069] Below, reference Figure 6 To describe an electronic device according to an embodiment of the present application.

[0070] Figure 6 is a block diagram of an electronic device according to an embodiment of the present application. Figure 6 , which shows a schematic diagram of the structure of an electronic device suitable for implementing a terminal device or a server in an embodiment of the present application.

[0071] like Figure 6As shown, the electronic device includes one or more processors, a communication unit, etc. The one or more processors are, for example: one or more central processing units (CPUs), and / or one or more graphics processing units (GPUs), etc. The processor can perform various appropriate actions and processes according to executable instructions stored in a read-only memory (ROM) or executable instructions loaded from a storage part to a random access memory (RAM). The communication unit may include but is not limited to a network card, and the network card may include but is not limited to an IB (Infiniband) network card. The processor may communicate with a read-only memory and / or a random access memory to execute executable instructions, and is connected to the communication unit through a bus and communicates with other target devices through the communication unit, thereby completing the operation corresponding to any method provided in the embodiments of the present application, for example, the primary shard manager shards the features in the feature queue; for any shard, the primary shard manager stores the any shard in the corresponding shard feature library, and records the mapping relationship between the any shard and the corresponding shard feature library in a mapping table; in response to a failure of the primary shard manager, a shard manager is reselected from at least one slave shard manager as the new primary shard manager; the new primary shard manager obtains shard information of the shard currently being written from all shard feature libraries, and recreates the mapping relationship of the shard currently being written based on the shard information of the shard currently being written in all shard feature libraries.

[0072] In addition, various programs and data required for the operation of the device can also be stored in the RAM. The CPU, ROM and RAM are connected to each other via a bus. In the case of RAM, the ROM is an optional module. The RAM stores executable instructions, or writes executable instructions into the ROM at runtime, and the executable instructions enable the processor to perform operations corresponding to any of the above methods of the present invention. The input / output (I / O) interface is also connected to the bus. The communication unit can be integrated or can be configured to have multiple submodules (e.g., multiple IB network cards) and be linked on the bus.

[0073] The following components are connected to the I / O interface: an input section including a keyboard, a mouse, etc.; an output section including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section including a hard disk, etc.; and a communication section including a network interface card such as a LAN card, a modem, etc. The communication section performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as needed. Removable media, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., are installed on the drive as needed so that a computer program read therefrom is installed into the storage section as needed.

[0074] If necessary, Figure 6The architecture shown is only an optional implementation method. In the specific practice, the above can be modified according to actual needs. Figure 6 The number and type of components can be selected, deleted, added or replaced; different functional components can be set up by separate settings or integrated settings, for example, the GPU and CPU can be set up separately or the GPU can be integrated on the CPU, the communication unit can be set up separately or integrated on the CPU or GPU, etc. These replaceable implementations all fall within the protection scope disclosed in the present invention.

[0075] In addition, an embodiment of the present invention further provides a computer program, including computer-readable code. When the computer-readable code runs on a device, a processor in the device executes instructions for implementing each step in the data processing method of any of the above embodiments of the present invention.

[0076] In addition, an embodiment of the present invention further provides a computer-readable storage medium for storing computer-readable instructions, which, when executed, implement the operations of each step in the data processing method of any of the above embodiments of the present invention.

[0077] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0078] The method and apparatus of the present invention may be implemented in many ways. For example, the method and apparatus of the present invention may be implemented by software, hardware, firmware, or any combination of software, hardware, firmware. The above order of steps for the method is only for illustration, and the steps of the method of the present invention are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present invention may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present invention. Thus, the present invention also covers a recording medium storing a program for executing the method according to the present invention.

[0079] The description of the present invention is given for the purpose of illustration and description, and is not intended to be exhaustive or to limit the invention to the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for particular uses.

Claims

1. An electromagnetic compatibility intelligent test and diagnosis system, characterized in that: include: The electromagnetic compatibility test data acquisition module is used to obtain the electromagnetic emission spectrum of the test product, and the current value and voltage value of the test product at multiple predetermined time points collected by the sensor; An electromagnetic compatibility test data processing module, used to extract an electromagnetic emission spectrum enhanced attention feature vector and a test product current-voltage time domain waveform feature vector from the electromagnetic emission spectrum diagram of the test product and the current values ​​and voltage values ​​of the test product at multiple predetermined time points collected by the sensor; The electromagnetic interference level label analysis module is used to obtain an electromagnetic interference level classification label based on the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector.

2. The electromagnetic compatibility intelligent test and diagnosis system according to claim 1, characterized in that: The electromagnetic compatibility test data processing module comprises: An electromagnetic emission spectrum feature extraction unit, used for extracting features from the electromagnetic emission spectrum of the test object to obtain an electromagnetic emission spectrum enhanced attention feature vector; The test product time domain waveform feature extraction unit is used to extract features from the current values ​​and voltage values ​​of the test product at multiple predetermined time points collected by the sensor to obtain the test product current-voltage time domain waveform feature vector.

3. The electromagnetic compatibility intelligent test and diagnosis system according to claim 2, characterized in that: The electromagnetic emission spectrum feature extraction unit comprises: An electromagnetic emission spectrum feature encoding subunit, used for performing feature encoding on the electromagnetic emission spectrum graph of the test object to obtain an electromagnetic emission spectrum enhanced attention feature graph; The electromagnetic emission spectrum pooling subunit is used to perform maximum value pooling on the electromagnetic emission spectrum enhanced attention feature map to obtain the electromagnetic emission spectrum enhanced attention feature vector.

4. The electromagnetic compatibility intelligent test and diagnosis system according to claim 3, characterized in that: The electromagnetic emission spectrum characteristic encoding subunit comprises: Using a deep learning network model to reduce noise on the electromagnetic emission spectrum of the test object to obtain an electromagnetic emission spectrum enhanced image; The electromagnetic emission spectrum enhanced image is passed through a convolutional neural network model based on the electromagnetic emission spectrum spatial attention mechanism to obtain the electromagnetic emission spectrum enhanced attention feature map.

5. The electromagnetic compatibility intelligent test and diagnosis system according to claim 4, characterized in that: The deep learning network model is used to reduce noise on the electromagnetic emission spectrum of the test object to obtain an electromagnetic emission spectrum enhanced image, including: The electromagnetic emission spectrum diagram of the test object is passed through an electromagnetic emission spectrum denoising module based on an autoencoder to obtain the electromagnetic emission spectrum enhanced image.

6. The electromagnetic compatibility intelligent test and diagnosis system according to claim 5, characterized in that: The electromagnetic emission spectrum pooling subunit comprises: Create a Spring Boot project using Spring Initializr and add the Apache Commons Math dependency; Create the class AttentionFeatureMap that represents the electromagnetic emission spectrum enhanced attention feature map and the class AttentionFeatureVector that represents the electromagnetic emission spectrum enhanced attention feature vector, and define their properties and methods; Write a method maxPooling that receives an AttentionFeatureMap object, traverses each row of the electromagnetic emission spectrum to enhance the attention feature map, finds the maximum value of each row, stores these maximum values ​​in an array, and finally returns an AttentionFeatureVector object; Use Spring's dependency injection mechanism to manage services and components, create a service class AttentionService, which contains the maximum pooling method maxPooling and is marked with Spring's @Service annotation; Create a REST controller AttentionController, define the HTTP request path through Spring's @RestController and @RequestMapping annotations, inject AttentionService into the controller using the @Autowired annotation, define the POST request method maxPool ing through the @PostMapping annotation, receive HTTP requests and call service methods; Use Gradle to build the project and deploy it to the server.

7. The electromagnetic compatibility intelligent test and diagnosis system according to claim 6, characterized in that: The test sample time domain waveform feature extraction unit comprises: Extracting a current-voltage time domain waveform of the test product from the current values ​​and voltage values ​​of the test product collected by the sensor at a plurality of predetermined time points; The current-voltage time-domain waveform diagram of the test product is passed through a current-voltage time-domain waveform feature encoder of the test product to obtain a current-voltage time-domain waveform feature vector of the test product.

8. The electromagnetic compatibility intelligent test and diagnosis system according to claim 7, characterized in that: The electromagnetic interference level label analysis module includes: An electromagnetic interference data feature fusion unit, used for performing perspective projection matching based on feature fine-grained association on the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector to obtain an electromagnetic interference level label classification feature vector; The electromagnetic interference level label classification unit is used to pass the electromagnetic interference level label classification feature vector through a classifier to obtain a classification result, and the classification result is used to represent the electromagnetic interference level classification label.

9. The electromagnetic compatibility intelligent test and diagnosis system according to claim 8, characterized in that: The electromagnetic interference data feature fusion unit is used to: Constructing an electromagnetic interference distance topology matrix between the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector; Constructing an electromagnetic interference eigenvalue granularity correlation matrix between the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector; Based on the electromagnetic interference distance topology matrix, the electromagnetic interference eigenvalue granularity correlation matrix is ​​topologically modulated to obtain an electromagnetic interference topology modulation correlation matrix; Taking the electromagnetic interference topology modulation correlation matrix as the perspective co-projection space, projecting the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector into the perspective co-projection space to obtain the perspective modulated electromagnetic emission spectrum enhanced attention feature vector and the perspective modulated test object current-voltage time domain waveform feature vector; The electromagnetic interference level label classification feature vector is obtained by fusing the visual angle modulated electromagnetic emission spectrum enhanced attention feature vector and the visual angle modulated test object current-voltage time domain waveform feature vector.

10. An electromagnetic compatibility intelligent test and diagnosis method, characterized in that: include: Obtaining the electromagnetic emission spectrum of the test object, and the current values ​​and voltage values ​​of the test object collected by the sensor at multiple predetermined time points; Extracting an electromagnetic emission spectrum enhanced attention feature vector and a test product current-voltage time domain waveform feature vector from the electromagnetic emission spectrum diagram of the test product and the current values ​​and voltage values ​​of the test product at multiple predetermined time points collected by the sensor; Based on the electromagnetic emission spectrum enhanced attention feature vector and the test object current-voltage time domain waveform feature vector, an electromagnetic interference level classification label is obtained.

Images