A method and system for automatic management of electromagnetic compatibility test data

Abstract

The present invention discloses an automated management method and system for electromagnetic compatibility test data. The prior art lacks an effective data verification function, and cannot automatically detect and correct abnormal values ​​in the data, resulting in difficulty in ensuring the integrity and accuracy of the data. The method adopted by the present invention includes: constructing an automated data acquisition system, acquiring test data in real time, performing multi-level preprocessing on the test data, extracting features from the fused data, and obtaining nonlinear spatiotemporal features; using nonlinear spatiotemporal features to perform time series anomaly detection, detecting abnormal data points, using an adaptive interpolation correction method to perform correction, and introducing a nonlinear optimization method based on Lagrange multipliers to further optimize the correction effect of the data. The present invention improves the accuracy of anomaly detection by extracting complex patterns and potential abnormal information from the data, and ensures that the corrected data has good smoothness and consistency.

Description

Technical Field

[0001] The invention belongs to the technical field of electromagnetic compatibility testing and data management, and in particular to an electromagnetic compatibility test data automatic management method and system. Background Art

[0002] Electromagnetic compatibility (EMC) is an important factor that must be considered in the design and production of modern electronic equipment to ensure that the equipment can operate normally in an electromagnetic environment and does not cause harmful interference to other equipment. With the widespread application and increasing complexity of electronic equipment, electromagnetic interference (EMI) and electromagnetic susceptibility (EMS) problems have become more and more serious. Therefore, it is particularly important to carry out systematic electromagnetic compatibility testing.

[0003] Modern electronic equipment manufacturers and testing institutions are in urgent need of an electromagnetic compatibility test management system that can integrate multiple test equipment, automate test processes, collect and process test data in real time, and can intelligently analyze and warn. The system should be able to achieve automated control and coordinated operation of multiple test equipment through a unified control platform to ensure the efficiency and consistency of the test process. At the same time, the system should have strong data processing and analysis capabilities, be able to collect, store, process and analyze test data in real time, and provide accurate and comprehensive test results and analysis reports. In addition, the system should also be able to intelligently detect and correct abnormal values ​​in the data, and timely discover and prompt potential test anomalies through the early warning system to improve the safety and reliability of the test.

[0004] The existing technology lacks effective data verification functions and cannot automatically detect and correct outliers in the data, making it difficult to ensure the integrity and accuracy of the data. The means for anomaly detection are limited and complex abnormal patterns cannot be accurately identified, resulting in low accuracy in anomaly detection and difficulty in timely detection and response to potential abnormal situations. The safety and reliability of the system are therefore affected and cannot meet the high standards for modern electromagnetic compatibility test data management. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the defects of the above-mentioned prior art and provide an automated management method and system for electromagnetic compatibility test data to ensure the integrity and accuracy of the data; extract complex patterns and potential abnormal information in the data to improve the accuracy of abnormality detection; and ensure that the corrected data has good smoothness and consistency.

[0006] To this end, the present invention adopts the following technical solution.

[0007] In a first aspect, the present invention provides a method for automatic management of electromagnetic compatibility test data, comprising:

[0008] S1. Build an automated data acquisition system to acquire test data in real time, perform multi-level preprocessing on the test data, wherein the multi-level preprocessing includes standardization, smoothing, and spatiotemporal data fusion, and perform feature extraction on the fused data to obtain nonlinear spatiotemporal features;

[0009] S2. Use nonlinear spatiotemporal features to perform time series anomaly detection, detect abnormal data points, use adaptive interpolation correction methods to correct them, and introduce nonlinear optimization methods based on Lagrange multipliers to further optimize the correction effect of the data.

[0010] Furthermore, in S1, when constructing an automated data acquisition system, test data is acquired in real time through sensors and data acquisition cards, and the test data is stored in a database to ensure data security and traceability.

[0011] Furthermore, in S1, a piecewise multiple smoothing method is applied to smooth the data, and the formula used by the piecewise multiple smoothing method is as follows:

[0012]

[0013] in, is the smoothed data at time t; is the size of the first-level sliding window; is the size of the second-level sliding window, which is the window size for smoothing based on the first-level sliding window; is the size of the third-level sliding window, which is the window size for smoothing based on the second-level sliding window; It is the value of the original data at different time offsets, indicating the data value at multiple levels of time points; , , It is the index in the smoothing window, indicating the position index at different window levels.

[0014] Furthermore, in S1, the smoothed data is subjected to spatiotemporal data fusion, the spatiotemporal data collected by different sensors are fused, and the reliability of the data is improved by a weighted average method. The formula for spatiotemporal data fusion is as follows:

[0015]

[0016] in, is the fused data, is the smoothed data of the kth sensor, is the weight of the kth sensor data, is the number of sensors.

[0017] Furthermore, in S1, the formula for extracting features from the fused data is as follows:

[0018]

[0019] in, is the extracted nonlinear spatiotemporal features; , is the position index in the feature matrix, indicating the first Row and List; is the weight matrix, which represents the convolution operation in the Row and The weight of the column position, , is the index of the convolution kernel; The fused data is indexed The value at is the bias term, is the hyperbolic tangent activation function; represents the mean of the fused data; Represents the standard deviation of the fused data.

[0020] Furthermore, in S2, the extracted nonlinear spatiotemporal features and historical time series data are used To detect and correct time series anomalies, we consider the autocorrelation characteristics and nonlinear spatiotemporal characteristics of the time series and adopt a multivariate composite autoregressive model, the formula of which is as follows:

[0021]

[0022] in, is the predicted value of the time series data at time t, For the front The autoregressive parameters for time steps, is the time lag index in the autoregressive model, For the front time step input feature weights, is the extracted nonlinear spatiotemporal feature The time delay index of Indicates that time series data is The actual observed value at time; The nonlinear spatiotemporal features extracted are The value of the moment; is the space-time coupling coefficient; is the second-order derivative of the historical time series data at time t-1; is the error term; is the activation function; is the time delay order of the autoregressive part, indicating that time steps; is the time delay order of the extracted nonlinear spatiotemporal features, indicating that time steps;

[0023] If the predicted value The difference between the actual observed value If the preset threshold is exceeded, it is considered that there are abnormal data points in the data.

[0024] Furthermore, in S2, the formula used in the adaptive interpolation correction method is as follows:

[0025]

[0026] in, is the data at time t after correction; and These are calibration parameters, which are adjusted based on historical data; is the preset threshold; is the historical time series data at time t-1; is the predicted value of the time series data at time t+1; is the predicted value of the time series data at time t.

[0027] Furthermore, in S2, the optimization objective function used in the nonlinear optimization method is as follows:

[0028]

[0029] in, is the optimization objective function, is the data at time t after correction, is the original test data at time t, is the standard deviation of the data at time t, is the regularization parameter, is the second-order derivative of the original test data, is the constraint function, For the p Lagrange multipliers, one for each constraint condition, is the number of time points, is the total number of constraints;

[0030] Find the optimal solution of the optimization objective function and calculate the optimization objective function For variables Gradient ,gradient The calculation formula is:

[0031]

[0032] Iterative update using gradient descent To minimize the objective function , the update formula is:

[0033]

[0034] in, is the learning rate, which can be adjusted according to the specific situation; Indicates the calibration data before updating; Represents the updated correction data; repeat the gradient calculation and update steps until the objective function is optimized Convergence means that the change in the objective function value is less than the preset threshold.

[0035] Furthermore, in S2, in the process of solving the optimization objective function, each iteration obtains are all correction values ​​of the current data. When the optimization objective function converges, the final That is, the optimized correction data is used for the judgment of the intelligent early warning system;

[0036] The intelligent early warning system automatically detects and prompts potential test abnormalities based on real-time data analysis. The early warning judgment formula is as follows:

[0037]

[0038] in, is the warning judgment value, is the warning threshold coefficient, Represents the expected value of the time series data at time t.

[0039] In a second aspect, the present invention provides an electromagnetic compatibility test data automatic management system, which is used to implement the above-mentioned electromagnetic compatibility test data automatic management method, and comprises:

[0040] Nonlinear spatiotemporal feature acquisition unit: construct an automated data acquisition system to acquire test data in real time, perform multi-level preprocessing on the test data, the multi-level preprocessing includes standardization, smoothing and spatiotemporal data fusion, perform feature extraction on the fused data, and obtain nonlinear spatiotemporal features;

[0041] Data detection, correction and optimization unit: Use nonlinear spatiotemporal features to detect time series anomalies, detect abnormal data points, use adaptive interpolation correction methods to correct them, and introduce nonlinear optimization methods based on Lagrange multipliers to further optimize the correction effect of the data.

[0042] The present invention has the following beneficial effects:

[0043] 1. By integrating various types of sensors and using edge computing devices for preliminary processing, data transmission delays are reduced, ensuring the accuracy and real-time nature of data collection, laying a solid foundation for subsequent data processing and analysis; automated data collection systems and advanced database technology ensure data security and traceability.

[0044] 2. The piecewise multiple smoothing method is used to smooth the data, which minimizes the random noise in the data. Through spatiotemporal data fusion, the data collected by different sensors are fused, and the reliability and consistency of the data are improved through the weighted average method. Combining nonlinear spatiotemporal feature extraction and high-dimensional space embedding technology, complex patterns and potential abnormal information in the data are extracted, thereby improving the accuracy of anomaly detection.

[0045] 3. By designing a new algorithm based on a multivariate composite autoregressive model, the extracted nonlinear spatiotemporal features and historical time series data are used to detect and correct time series anomalies. The correction effect of the data is further optimized through the adaptive interpolation correction method and the nonlinear optimization method based on Lagrange multipliers to ensure that the corrected data has good smoothness and consistency. The intelligent early warning system automatically detects and prompts potential test anomalies based on real-time data analysis, and detects anomalies in time by real-time monitoring of data changes, thereby improving the safety and reliability of the system. The early warning judgment formula ensures that the early warning system can respond quickly and avoid potential risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of an automatic management method for electromagnetic compatibility test data according to the present invention;

[0047] Figure 2 This is a structural diagram of an electromagnetic compatibility test data automation management system described in the present invention. DETAILED DESCRIPTION

[0048] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0050] Example 1

[0051] See attached Figure 1 , which shows a flow chart of an electromagnetic compatibility test data automatic management method provided by an embodiment of the present invention, the method comprising the following steps:

[0052] S1. Build an automated data acquisition system to acquire test data in real time, perform multi-level preprocessing on the test data, including standardization, smoothing, and spatiotemporal data fusion, extract features from the fused data, and obtain nonlinear spatiotemporal features.

[0053] In the establishment of the electromagnetic compatibility automated test platform, multiple test equipment, including power frequency magnetic field, static electricity, constant magnetic field and other test devices, were first integrated into a unified platform. By developing a unified control system, the work of each test equipment is coordinated to realize the automated test process. In the automated test process, various types of sensors are used, such as electromagnetic field sensors, temperature sensors, humidity sensors, etc. The sensors are preliminarily processed by edge computing devices to reduce data transmission delays and ensure the accuracy and real-time nature of data collection.

[0054] Build an automated data acquisition system, obtain test data in real time through sensors and data acquisition cards, and store the test data in a database to ensure data security and traceability.

[0055] Specifically, the test data is preprocessed at multiple levels, including standardization, smoothing, and spatiotemporal data fusion. The data standardization can be done using existing technology, and the piecewise multiple smoothing method is used to smooth the data to reduce noise. The piecewise multiple smoothing formula is as follows:

[0056]

[0057] in, is the smoothed data at time t, is the size of the first-level sliding window; is the size of the second-level sliding window, which is the window size for smoothing based on the first-level sliding window; is the size of the third-level sliding window, which is the window size for smoothing based on the second-level sliding window; It is the value of the original data at different time offsets, indicating the data value at multiple levels of time points; , , is the index in the smoothing window, indicating the position index at different window levels. Through segmented processing and multiple smoothing, the random noise in the data is minimized.

[0058] The smoothed data is subjected to spatiotemporal data fusion, and the spatiotemporal data collected by different sensors are fused to improve the reliability of the data through the weighted average method. The spatiotemporal data fusion formula is as follows:

[0059]

[0060] in, is the fused data, For the k The smoothed data of each sensor is For the k The weight of the sensor data, is the number of sensors. By comprehensively considering the data from different sensors, the reliability and accuracy of the overall data are improved.

[0061] The fused data is used for nonlinear spatiotemporal feature extraction. Nonlinear spatiotemporal feature extraction can capture complex patterns and potential abnormal information in the data. The fused data contains comprehensive information from multiple sensors. In order to discover patterns hidden in different time and space dimensions and enhance the expressiveness of features, a nonlinear spatiotemporal feature extraction method is designed. The feature extraction formula is as follows:

[0062]

[0063] in, is the nonlinear spatiotemporal features extracted. , is the position index in the feature matrix, indicating the first Row and List, is the weight matrix, which represents the convolution operation in the Row and The weight of the column position, , is the index of the convolution kernel, The fused data is indexed The value at is the bias term, is the hyperbolic tangent activation function, represents the mean value of the fused data, Represents the standard deviation of the fused data. By embedding in high-dimensional space, the nonlinear characteristics of spatiotemporal data are extracted, which improves the accuracy of anomaly detection.

[0064] S2. Use nonlinear spatiotemporal features to perform time series anomaly detection, detect abnormal data points, use adaptive interpolation correction methods to correct them, and introduce nonlinear optimization methods based on Lagrange multipliers to further optimize the correction effect of the data.

[0065] The extracted nonlinear spatiotemporal features represent the characteristic values ​​at specific time and space points, which contain complex patterns and potential abnormal information of the test data. By extracting nonlinear spatiotemporal features, abnormal patterns in the data can be identified. Using the extracted nonlinear spatiotemporal features and historical time series data To detect and correct time series anomalies, the autocorrelation characteristics and nonlinear spatiotemporal characteristics of the time series are considered, and a multivariate composite autoregressive model is adopted, the formula of which is as follows:

[0066]

[0067] in, is the predicted value of the time series data at time t, For the front The autoregressive parameters for time steps, is the time lag index in the autoregressive model, For the front time step input feature weights, is the input feature The time delay index of Indicates that time series data is The actual observed value at time; The nonlinear spatiotemporal features extracted are The value of the moment, is the space-time coupling coefficient, is the second-order derivative of the time series data at time t-1, is the error term, is the activation function, is the time delay order of the autoregressive part, indicating that time steps, is the time delay order of the input feature, indicating that By considering the autocorrelation characteristics and nonlinear spatiotemporal characteristics of the time series, the ability to identify and correct outliers is improved.

[0068] If the predicted value The difference between the actual observed value If the preset threshold is exceeded, it is considered that there are abnormal data points in the data.

[0069] For the detected abnormal data points, an adaptive interpolation correction method is used to correct them, and the overall consistency of the data is improved through nonlinear optimization.

[0070] Specifically, adaptive interpolation correction is performed, and the formula is as follows:

[0071]

[0072] in, is the data at time t after correction; and To calibrate the parameters, adjustments are made based on historical data; is the preset threshold; is the historical time series data at time t-1; is the predicted value of the time series data at time t+1. Through the interpolation method, the outliers are corrected to ensure the continuity of the data.

[0073] The corrected data is used to introduce a nonlinear optimization method based on Lagrange multipliers to further optimize the correction effect of the data. The optimization objective function is as follows:

[0074]

[0075] in, is the optimization objective function, is the original test data at time t, is the standard deviation of the data at time t, is the regularization parameter, is the second-order derivative of the original test data, is the constraint function, is the pth Lagrange multiplier, each constraint condition corresponds to a Lagrange multiplier, is the number of time points, is the total number of constraints. By minimizing the data error and regularization term, the corrected data is ensured to have good smoothness and consistency.

[0076] Find the optimal solution of the optimization objective function and calculate the optimization objective function For variables The gradient of . The gradient is calculated as:

[0077]

[0078] Iterative update using gradient descent To minimize the optimization objective function The update formula is:

[0079]

[0080] in, It is the learning rate, which needs to be adjusted according to the specific situation; Indicates the calibration data before updating; Represents the updated correction data. Repeat the gradient calculation and update steps until the objective function is optimized. Convergence means that the change in the objective function value is less than the preset threshold.

[0081] In the process of solving the optimization objective function, each iteration obtains are all correction values ​​of the current data. When the optimization objective function converges, the final That is, the optimized correction data, which is used for the judgment of the intelligent early warning system.

[0082] The intelligent early warning system automatically detects and prompts potential test abnormalities based on real-time data analysis. The early warning judgment formula is as follows:

[0083]

[0084] in, is the warning judgment value, is the warning threshold coefficient, It represents the expected value of time series data at time t. By monitoring data changes in real time, abnormal situations can be discovered in time to improve the security and reliability of the system.

[0085] Application Examples

[0086] The method for automatic management of electromagnetic compatibility test data described in Example 1 is used as follows.

[0087] Assume that the following devices and sensors are integrated on an electromagnetic compatibility test platform: power frequency magnetic field test device, electrostatic test device, constant magnetic field test device, electromagnetic field sensor, temperature sensor, humidity sensor; during the automated testing process, the sensor is preliminarily processed by the edge computing device to reduce data transmission delays and ensure the accuracy and real-time nature of data collection.

[0088] The test data is acquired in real time through sensors and data acquisition cards. Data is collected once every second. The collected data include: electromagnetic field strength (unit: A / m), temperature (unit: ℃), humidity (unit: %);

[0089] Assume that the data collected every second is as follows:

[0090] Electromagnetic field strength: 1.5A / m, 1.6A / m, 1.4A / m, 1.5A / m, 1.7A / m;

[0091] Temperature: 25℃, 25.1℃, 25.2℃, 25℃, 24.9℃; Humidity: 50%, 51%, 50%, 49%, 50%.

[0092] The data is standardized and converted into zero mean and unit standard deviation. The mean and standard deviation of the data are:

[0093] Mean electromagnetic field strength: 1.54A / m, standard deviation: 0.11A / m;

[0094] Temperature mean: 25.04℃, standard deviation: 0.1℃;

[0095] Humidity mean: 50%, standard deviation: 0.7%.

[0096] The standardized data is as follows:

[0097] Electromagnetic field strength: (-0.36, 0.55, -1.27, -0.36, 1.45);

[0098] Temperature: (-0.4, 0.6, 1.6, -0.4, -1.4);

[0099] Humidity: (0,1.43,0,-1.43,0).

[0100] The data is smoothed using piecewise multiple smoothing method, assuming that the window sizes are , , , the smoothing formula is as follows:

[0101]

[0102] Taking electromagnetic field strength as an example, the smoothed data is as follows:

[0103]

[0104] The smoothed data from different sensors are fused, assuming the weights All are 1, and the fusion formula is as follows:

[0105]

[0106] The calculation results are:

[0107] Electromagnetic field strength: 0.095A / m;

[0108] Temperature: -0.2℃;

[0109] Humidity: 0.2%;

[0110] Assume that the mean and standard deviation of the fused data are:

[0111] The mean value of electromagnetic field strength is 0.036A / m, standard deviation is 0.233A / m;

[0112] Temperature mean: 0.15℃, standard deviation: 0.207℃;

[0113] Humidity mean: 0.125%, standard deviation: 0.316%.

[0114] Use the fused data for nonlinear spatiotemporal feature extraction, assuming that the weight matrix , bias term , the feature extraction formula is as follows:

[0115]

[0116] Assume that the result of feature extraction of the fused data (0.095, -0.2, 0.2) is:

[0117]

[0118] A multivariate composite autoregressive model is used for time series anomaly detection and correction. The model formula is as follows:

[0119]

[0120] Assume the parameter values ​​are: , , , , .

[0121] The correction formula is as follows:

[0122]

[0123] Assumptions , , , the calculation result is:

[0124]

[0125] Using the nonlinear optimization method based on Lagrange multipliers, the optimization objective function is as follows:

[0126]

[0127] Assume that the regularization parameter , the Lagrange multiplier , constraints , the optimized result is:

[0128]

[0129] Calculate the optimization objective function For variables The gradient of . The gradient calculation formula is:

[0130]

[0131] Iterative update using gradient descent To minimize the optimization objective function The update formula is:

[0132]

[0133] in, is the learning rate, assuming Initial value .

[0134] Repeat the gradient calculation and update steps until the objective function is optimized Convergence means that the change in the objective function value is less than the preset threshold. For example, after 10 iterations, assuming that the final optimization result is .

[0135] In the process of solving the optimization objective function, each iteration obtains are all correction values ​​of the current data. When the optimization objective function converges, the final This is the optimized correction data.

[0136] Intelligent early warning system based on optimized calibration data Make early warning judgment, the early warning judgment formula is as follows:

[0137]

[0138] Assuming the corrected data , real-time data mean , standard deviation , warning threshold coefficient The early warning judgment steps are as follows:

[0139] According to the corrected data , calculate the real-time mean and standard deviation ;Compare and If , then trigger the warning, set ; Otherwise, no warning is triggered, set .

[0140] In this application example, the calculation is as follows:

[0141]

[0142] because ,so .

[0143] In summary, an automated management method for electromagnetic compatibility test data is completed.

[0144] Example 2

[0145] This embodiment provides an electromagnetic compatibility test data automatic management system. Figure 2 As shown, the method for automatically managing electromagnetic compatibility test data described in Example 1 includes:

[0146] Nonlinear spatiotemporal feature acquisition unit: construct an automated data acquisition system to acquire test data in real time, perform multi-level preprocessing on the test data, the multi-level preprocessing includes standardization, smoothing and spatiotemporal data fusion, perform feature extraction on the fused data, and obtain nonlinear spatiotemporal features;

[0147] Data detection, correction and optimization unit: Use nonlinear spatiotemporal features to detect time series anomalies, detect abnormal data points, use adaptive interpolation correction methods to correct them, and introduce nonlinear optimization methods based on Lagrange multipliers to further optimize the correction effect of the data.

[0148] It should be noted that each module in the above-mentioned electromagnetic compatibility test data automation management system can be fully or partially implemented by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. For the specific definition of an electromagnetic compatibility test data automation management system, please refer to the definition of an electromagnetic compatibility test data automation management method mentioned above. The two have the same functions and effects, which will not be repeated here.

[0149] The sequence of the embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0150] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0151] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for automatic management of electromagnetic compatibility test data, characterized in that: include: S1. Build an automated data acquisition system to acquire test data in real time, perform multi-level preprocessing on the test data, wherein the multi-level preprocessing includes standardization, smoothing, and spatiotemporal data fusion, and perform feature extraction on the fused data to obtain nonlinear spatiotemporal features; S2. Use nonlinear spatiotemporal features to detect time series anomalies, detect abnormal data points, use adaptive interpolation correction methods to correct them, and introduce nonlinear optimization methods based on Lagrange multipliers to further optimize the correction effect of data; In S2, the extracted nonlinear spatiotemporal features and historical time series data are used To detect and correct time series anomalies, we consider the autocorrelation characteristics and nonlinear spatiotemporal characteristics of the time series and adopt a multivariate composite autoregressive model, the formula of which is as follows: ， in, is the predicted value of the time series data at time t, For the front The autoregressive parameters for time steps, is the time lag index in the autoregressive model, For the front time step input feature weights, is the extracted nonlinear spatiotemporal feature The time delay index of Indicates that time series data is The actual observed value at time; The nonlinear spatiotemporal features extracted are The value of the moment; is the space-time coupling coefficient, is the second-order derivative of the historical time series data at time t-1; is the error term; is the activation function; is the time delay order of the autoregressive part, indicating that time steps; is the time delay order of the extracted nonlinear spatiotemporal features, indicating that time steps; If the predicted value The difference between the actual observed value If the preset threshold is exceeded, it is considered that there are abnormal data points in the data.

2. The method for automatic management of electromagnetic compatibility test data according to claim 1, characterized in that: In S1, when constructing an automated data acquisition system, test data is acquired in real time through sensors and data acquisition cards, and the test data is stored in a database to ensure data security and traceability.

3. The method for automatic management of electromagnetic compatibility test data according to claim 1, characterized in that: In S1, a piecewise multiple smoothing method is applied to smooth the data. The formula used by the piecewise multiple smoothing method is as follows: ， in, is the smoothed data at time t; is the size of the first-level sliding window; is the size of the second-level sliding window, which is the window size for smoothing based on the first-level sliding window; is the size of the third-level sliding window, which is the window size for smoothing based on the second-level sliding window; It is the value of the original data at different time offsets, indicating the data value at multiple levels of time points; , , It is the index in the smoothing window, indicating the position index at different window levels.

4. The method for automatic management of electromagnetic compatibility test data according to claim 1, characterized in that: In S1, the smoothed data is subjected to spatiotemporal data fusion, the spatiotemporal data collected by different sensors are fused, and the reliability of the data is improved by the weighted average method. The formula for spatiotemporal data fusion is as follows: ， in, is the fused data, is the smoothed data of the kth sensor, is the weight of the kth sensor data, is the number of sensors.

5. The method for automatic management of electromagnetic compatibility test data according to claim 1, characterized in that: In S1, the formula for feature extraction of the fused data is as follows: ， in, is the extracted nonlinear spatiotemporal features; , is the position index in the feature matrix, indicating the first Row and List; is the weight matrix, which represents the convolution operation in the Row and The weight of the column position, , is the index of the convolution kernel; The fused data is indexed The value at is the bias term, is the hyperbolic tangent activation function; represents the mean of the fused data; Represents the standard deviation of the fused data.

6. The method for automatic management of electromagnetic compatibility test data according to claim 1, characterized in that: In S2, the formula used in the adaptive interpolation correction method is as follows: ， in, is the data at time t after correction; and These are calibration parameters, which are adjusted based on historical data; is the preset threshold; is the historical time series data at time t-1; is the predicted value of the time series data at time t+1; is the predicted value of the time series data at time t.

7. The method for automatic management of electromagnetic compatibility test data according to claim 1, characterized in that: In S2, the optimization objective function used in the nonlinear optimization method is as follows: ， in, is the optimization objective function, is the data at time t after correction, is the original test data at time t, is the standard deviation of the data at time t, is the regularization parameter, is the second-order derivative of the original test data, is the constraint function, For the p Lagrange multipliers, one for each constraint condition, is the number of time points, is the total number of constraints; Find the optimal solution of the optimization objective function and calculate the optimization objective function For variables Gradient ,gradient The calculation formula is: ， Iterative update using gradient descent To minimize the optimization objective function , the update formula is: ， in, is the learning rate, which can be adjusted according to the specific situation; Indicates the calibration data before updating; Represents the updated correction data; repeat the gradient calculation and update steps until the objective function is optimized Convergence means that the change in the objective function value is less than the preset threshold.

8. The method for automatic management of electromagnetic compatibility test data according to claim 7, characterized in that: In S2, during the optimization objective function solution process, each iteration obtains are all correction values ​​of the current data. When the optimization objective function converges, the final That is, the optimized correction data is used for the judgment of the intelligent early warning system; The intelligent early warning system automatically detects and prompts potential test abnormalities based on real-time data analysis. The early warning judgment formula is as follows: ， in, is the warning judgment value, is the warning threshold coefficient, Represents the expected value of the time series data at time t.

9. An electromagnetic compatibility test data automatic management system, which is used to implement the electromagnetic compatibility test data automatic management method according to any one of claims 1 to 8, characterized in that: include: Nonlinear spatiotemporal feature acquisition unit: construct an automated data acquisition system to acquire test data in real time, perform multi-level preprocessing on the test data, the multi-level preprocessing includes standardization, smoothing and spatiotemporal data fusion, perform feature extraction on the fused data, and obtain nonlinear spatiotemporal features; Data detection, correction and optimization unit: Use nonlinear spatiotemporal features to detect time series anomalies, detect abnormal data points, use adaptive interpolation correction methods to correct them, and introduce nonlinear optimization methods based on Lagrange multipliers to further optimize the correction effect of the data.

Images