Test equipment data synchronization processing method and system

Abstract

The invention discloses a test equipment data synchronization processing method and system, particularly relates to the technical field of data synchronization, and is used for solving the problem of insufficient adaptive optimization of data synchronization. The method comprises the following steps: constructing a port data association graph, optimizing a connection relationship between ports based on a time-varying adjacency matrix, extracting port data representation by using a graph neural network, optimizing a data synchronization strategy, analyzing influence factors of a data synchronization error based on a causal relationship model, identifying a propagation path of data deviation between the ports, and calculating a data drift amount. The method comprises the following steps: optimizing a synchronization window by using causal inference, correcting data drift, constructing a unified time scale, extracting modal information based on feature coding, modeling cross-modal data association by using a self-attention mechanism, optimizing feature matching between ports, improving the precision of data fusion, and realizing high-precision synchronous alignment and fusion of port data. And higher-quality data input is provided for data analysis and anomaly detection of test equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of data synchronization, and more specifically, to a method and system for data synchronization processing of test equipment. Background Art

[0002] With the widespread application of intelligent testing equipment, multi-port test systems have become an important means of high-precision data acquisition and analysis, and are widely used in power system status monitoring, industrial automation detection, wireless communication testing, intelligent manufacturing and other fields. In these application scenarios, multiple test ports usually simultaneously collect multi-modal data such as voltage, current, spectrum, temperature, vibration, and image to comprehensively evaluate the system operation status.

[0003] Deficiencies in existing technologies: Current multi-port test equipment lacks adaptive optimization capabilities in data synchronization strategies, resulting in unstable data alignment accuracy. In actual applications, due to factors such as data transmission delays, clock drift, different sampling frequencies, and environmental interference at different ports, once the test environment changes, it is difficult to dynamically adjust the synchronization strategy, resulting in the gradual accumulation of time alignment errors between data, which in turn affects the accuracy of data fusion and subsequent analysis. In addition, current synchronization optimization methods mostly rely on simple rules or fixed threshold adjustments, lack in-depth analysis of data synchronization errors at different ports, resulting in synchronization effects relying on manual settings and poor adaptability, which in turn affects the original test process. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a test equipment data synchronization processing method and system to solve the problem of insufficient data synchronization adaptive optimization in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: A method for synchronously processing test equipment data comprises the following steps: A port data association graph is constructed based on the data of each port of the test equipment, and the connection relationship between ports is determined based on the time-varying adjacency matrix. The graph neural network is used to extract the port data representation and optimize the data synchronization strategy. Build a causal relationship model, analyze the factors affecting data synchronization errors, identify the propagation path of data deviations between ports, calculate the amount of data drift, and optimize the synchronization window based on causal inference to correct data drift; The port data is time-aligned to construct a unified time scale, and the key information of the modality is extracted based on feature encoding. The cross-modal data association is determined based on the self-attention mechanism, and the feature matching between ports is optimized.

[0006] In a preferred embodiment, a port data association graph is constructed based on the data of each port of the test device. The specific process is as follows: Construct a port data association graph for the test equipment, define the port set and edge set of the test equipment, the edge set E is determined by the correlation between the port data, and represents the dynamic connection relationship between the ports. The edge set is expressed as: ,in, is the data correlation function between ports i and j at time t, is the correlation threshold; Construct the time-varying adjacency matrix A(t), the expression is: ,in, Controls the rate at which correlation decays, is the association weight between ports i and j.

[0007] In a preferred embodiment, the connection relationship between ports is determined based on the time-varying adjacency matrix. The specific process is as follows: Calculate the time-varying covariance matrix of the ports to determine the correlation between the port data: The calculation expression of the time-varying covariance matrix is: ,in, and Represent the data of port i and j in time window T respectively.

[0008] In a preferred embodiment, a graph neural network is used to extract port data representation and optimize the data synchronization strategy. The specific steps are as follows: The graph neural network is used to extract the port data representation. The initial port data features are: , X(t) represents the original data collected by the multi-port test equipment at time t; Build a graph neural network model and use graph convolution for feature update: ,in, is the node feature of the lth layer, is the learned weight matrix, is a nonlinear activation function; The loss function used for training the graph neural network model is: , optimized by gradient descent , output port data characterization ; Calculate the optimal synchronization window: , where D is the data consistency measurement function; The synchronization parameters are calculated based on the port data characterization to minimize data drift, and the optimized data synchronization strategy is output for port optimization.

[0009] In a preferred implementation, a causal relationship model is constructed to analyze the influencing factors of data synchronization errors and identify the propagation path of data deviations between ports. The specific process is as follows: Define the set of causal variables: ,in, is the data flow of port i at time t, is the data flow of port N; is the time synchronization error, indicating the clock drift between ports; is the signal interference strength, indicating the degree of signal influence between ports; is the equipment error, which represents the systematic deviation of the sensor or measuring device; Use a structural learning algorithm based on a probabilistic graphical model to automatically learn causal relationships and obtain a causal relationship graph. The nodes of the causal relationship graph represent the data of each port and the influencing factors, and the edges represent the causal relationships. After constructing the causal relationship diagram, transfer entropy is used to calculate the impact of data deviation; The transfer entropy expression is: , where H(X) represents the information entropy of the data stream, Indicates the impact strength of the data of port i on the future value of port j, and are the data flows of ports i and j respectively; By calculating the transfer entropy between all ports, the data deviation propagation matrix is ​​constructed: , where Indicates the strength of the impact of the data at port k on the future value of port j.

[0010] In a preferred embodiment, the data drift is calculated, and the synchronization window is optimized based on causal inference to correct the data drift. The specific process is as follows: The data deviation propagation matrix constructed based on the transfer entropy between ports determines the propagation path, and the data drift is calculated using causal intervention: The deviation between the theoretical value and the actual observed value of the port data is taken as the data drift; The contribution of data deviation is calculated based on Bayesian inference, and the expression is: , is the posterior probability, which indicates the data drift of port i after the actual observation value X is known. The probability of it actually happening; is the data drift of port i The probability that the process observes the actual observation value X, is the data drift The prior probability of is the total probability of the actual observed value X occurring; Define the synchronization error function: ,in, It is the data synchronization window. It is a function of the relationship between window size and drift error; Use the optimization method to find the window size with the smallest synchronization error: ; The smallest synchronization window is chosen to minimize data drift errors.

[0011] In a preferred embodiment, the port data is time-aligned to construct a unified time scale, and the key information of the modality is extracted based on feature encoding, the cross-modal data association is determined based on the self-attention mechanism, and the feature matching between ports is optimized. The specific process is as follows: Time-align the port data and use linear interpolation for the timing signal data; keep the nearest value for the low-frequency data; and perform Fourier interpolation on the frequency domain data; Use Transformer position encoding to identify data at different time steps and construct a Token sequence; Construct a modal feature encoding network and use independent feature encoders to extract features from different modalities; Obtain the feature representation before fusion, calculate the feature vectors of all modalities, use the self-attention mechanism to model the cross-modal relationship, generate a fused representation, and optimize the feature matching between ports based on the fused representation.

[0012] A test equipment data synchronization processing system, used to implement the above-mentioned test equipment data synchronization processing method, comprising: The strategy determination module is used to construct a port data association graph based on the data of each port of the test equipment, determine the connection relationship between ports based on the time-varying adjacency matrix, use the graph neural network to extract the port data representation, and optimize the data synchronization strategy; The data correction analysis module is used to build a causal relationship model, analyze the influencing factors of data synchronization errors, identify the propagation path of data deviations between ports, calculate the data drift, and optimize the synchronization window based on causal inference to correct data drift; The port synchronization optimization module is used to time-align port data, build a unified time scale, extract modal key information based on feature encoding, determine cross-modal data association based on the self-attention mechanism, and optimize feature matching between ports.

[0013] Technical effects and advantages of the present invention: The present invention constructs a port data association graph, optimizes the connection relationship between ports based on a time-varying adjacency matrix, uses a graph neural network to extract port data representation, optimizes data synchronization strategies, improves the accuracy of data alignment, analyzes the influencing factors of data synchronization errors based on a causal relationship model, identifies the propagation path of data deviations between ports, calculates the data drift, uses causal inference to optimize the synchronization window, dynamically corrects data drift, improves data consistency, time-aligns port data, constructs a unified time scale, extracts modal key information based on feature encoding, uses a self-attention mechanism to model cross-modal data associations, optimizes feature matching between ports, and improves the accuracy of data fusion. The present invention can dynamically adjust data synchronization parameters, reduce data drift errors, achieve high-precision synchronous alignment and fusion of port data, improve data consistency, provide high-quality data input for data analysis, anomaly detection and prediction of test equipment, and improve the stability and intelligence level of the test equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flow chart of a method for synchronously processing test equipment data.

[0015] Figure 2 The present invention is a structural schematic diagram of a test equipment data synchronization processing system. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of 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.

[0017] Example 1: Figure 1 As shown, a method for synchronously processing test equipment data includes the following steps: A port data association graph is constructed based on the data of each port of the test equipment, and the connection relationship between ports is determined based on the time-varying adjacency matrix. The graph neural network is used to extract the port data representation and optimize the data synchronization strategy. Build a causal relationship model, analyze the factors affecting data synchronization errors, identify the propagation path of data deviations between ports, calculate the amount of data drift, and optimize the synchronization window based on causal inference to correct data drift; The port data is time-aligned to construct a unified time scale, and the key information of the modality is extracted based on feature encoding. The cross-modal data association is determined based on the self-attention mechanism, and the feature matching between ports is optimized.

[0018] In the data synchronization process of multi-port test equipment, the measurement data of each port often has complex spatial correlation and time dependence. These correlations may be caused by factors such as equipment layout, signal propagation path, correlation of measured physical quantities, environmental interference, etc. If the relationship between ports is not fully modeled, it may lead to insufficient optimization of data synchronization strategy and reduced synchronization accuracy.

[0019] Step 1: Perform multi-port data association modeling based on graph neural network. The specific steps are as follows: In the test equipment, multiple ports measure different data streams at the same time, such as current, voltage, spectrum, etc. The data of these ports may have certain correlations. For example, adjacent ports may measure different parts of the same signal, so the data streams of adjacent ports are highly correlated. Due to the different signal propagation paths, some ports may have strong or weak time correlations, and the correlation between ports may be time-varying, that is, the correlation between data in different time periods may change; Construct a port data association diagram for the test equipment. The specific process is as follows: Define a port set. Assume that the test device has N ports and the data flow measured by each port is , then define the port set: ,in, represents the i-th port, which is a time series signal; Define the edge set. The edge set E is determined by the correlation between the port data and represents the dynamic connection relationship between the ports: ,in, is the data correlation function between ports i and j at time t, that is, it represents the correlation between the data flows of ports i and j at time t. Commonly used methods include Pearson correlation coefficient, mutual information or cross-correlation function; is the correlation threshold; The correlation threshold is set to 0.7 or 0.8. Only when the correlation function value is greater than the correlation threshold, it is considered that there is a significant correlation between port i and port j, so that an edge connection is established during graph composition; Construct the time-varying adjacency matrix A(t). If there is a significant correlation between ports i and j, then set is the association weight between ports i and j, otherwise it is set to zero, that is, the edge weight is defined as: ,in, Controlling the rate at which correlations decay (e.g. ), used to amplify or suppress the impact of correlation on weight; By assigning lower attenuation values ​​(higher weights) to port pairs with high correlation (larger values) and directly setting them to zero, only the information of strong correlation is retained in the graph.

[0020] By calculating the correlation and covariance between ports, the time-varying adjacency matrix can reflect the real-time correlation degree of signals between ports. Ports with high correlation have larger weights in the graph, which means that their data changes tend to be consistent.

[0021] Complete the above steps to output the port data association graph G(V, E) and the time-varying adjacency matrix A(t).

[0022] Calculate the time-varying covariance matrix of the port. The correlation of multi-port data is not fixed, but changes with time. Therefore, a time-varying covariance matrix is ​​constructed to dynamically calculate the correlation (connection relationship) between port data and adjust the data synchronization strategy: Assume that the data of ports i and j in time window T is and , define the covariance function: , by calculating the covariance of data from different ports, we can get the time-varying covariance matrix ; If the time-varying covariance matrix changes greatly, adjust the adjacency matrix A(t): ; By comparing the covariances of adjacent moments, we can capture the trend of data correlation changing over time; when the change is large, it indicates that there may be a synchronization deviation between the two ports. The lower weight after the update reflects this inconsistency, which facilitates subsequent detection and compensation.

[0023] The output is an optimized time-varying adjacency matrix for graph neural network modeling.

[0024] Train a graph neural network (GNN) to extract port data representation; Construct the GNN model and set the initial port data features as , X(t) represents the raw data collected by the multi-port test equipment at time t, and graph convolution (GCN) is used for feature update: ,in, is the node feature of the lth layer, is the learned weight matrix, is a nonlinear activation function (such as ReLU); The loss function used for training the GNN model is: , optimized by gradient descent , output port data characterization , used for data synchronization optimization, optimizes the weights of each layer through gradient descent, so that the representation differences of ports with high weight (high correlation) connections are minimized.

[0025] According to the port data representation learned by GNN, the data synchronization strategy can be optimized to dynamically adjust the synchronization window size and improve the data synchronization accuracy; Calculate the optimal synchronization window: , where D is the data consistency measurement function, according to the port data representation Calculate synchronization parameters to minimize data drift and output optimized data synchronization strategy; For example, the data synchronization strategy dynamically adjusts the synchronization window (optimizes the synchronization window size). A fixed synchronization window may cause a large data alignment error. If the window is too large, irrelevant data may be introduced. If the window is too small, valid data may be lost. The optimal synchronization window is calculated based on the port data characterization, so that the synchronization window can dynamically adapt to changes in the data flow and improve synchronization accuracy. It can also automatically adjust the timestamp of the data to align the data more accurately and reduce synchronization errors. A specific embodiment is: assuming that a test device has 4 ports, the specific data and parameter settings are as follows: Port data settings for each port is a time series with a length of 100 (e.g., a voltage signal), a sampling frequency of 100 Hz, and a correlation threshold of , attenuation parameter , the time window T is ten seconds and the step size is one second; At a certain moment, the correlation between the ports is calculated using the Pearson correlation coefficient as follows (assuming that the data has been normalized), and the port correlation value is: , , , , , ; According to the correlation threshold analysis, only when When a connection is established, for ports 1 and 2: ,but ; Ports 1 and 3: ; Ports 1 and 4: ; For ports 2 and 3: , ; Ports 2 and 4 are 0; Ports 3 and 4: ,but ; Finally, we get the time-varying adjacency matrix (symmetric matrix): ; The graph neural network is processed by taking the 100-dimensional time series of each port as a row vector to form a 4×100 matrix. Assume that the first-layer weight matrix The size of the graph convolution is 100×64 (randomly initialized), and the activation function is ReLU. Then, a 4×64 matrix is ​​obtained, and each row is the representation of the port after graph convolution. Design multiple layers (for example, 2 or 3 layers in total) of graph convolution, and finally obtain the fused port features. ,at this time Each row of contains the comprehensive information of the port after considering its time-varying association with other ports.

[0026] based on The data consistency information between ports reflected in the sync window can guide the adjustment of the synchronization window (for example, selecting the best synchronization window, dynamically compensating the timestamps of each port, and aligning the multi-port data in time, thereby achieving overall data synchronization.

[0027] Through the graph convolution layer, the raw data of each port and its dynamic relationship are integrated into a unified feature representation. This representation contains both the local information of the single-port signal and the global relationship between ports. The final port representation It can be used to evaluate the synchronization status between ports - if highly correlated ports have large differences in the representation of the port characterization, it indicates that there is a data drift or delay problem. Based on this information, the system can dynamically adjust the synchronization window and timestamp correction amount to achieve high-precision alignment and synchronization of data.

[0028] Step 2: Conduct causal inference driven data consistency analysis; After completing step 1, a dynamic correlation graph of the data of each port of the test equipment has been constructed, and the topological representation of the multi-port data has been extracted through GNN. However, in the process of multi-port data fusion, data consistency deviation may still occur, that is, the data of some ports may drift, mismatch, error accumulation and other problems. These problems may be caused by clock drift, data transmission delay, equipment error, signal interference and other factors; Construct a causal relationship model between the test equipment ports and define a set of causal variables. Suppose the test equipment has N ports and the data flow measured by each port is When building a causal relationship model, you need to define a set of causal variables: ,in, is the data flow of port i at time t, is the data flow of port N; is the time synchronization error, which indicates the synchronization deviation caused by the clock drift between ports; Signal interference strength, which measures the degree of signal influence between ports; is the equipment error, which represents the systematic bias or noise level of the sensor or measuring device; Construct a causal relationship graph G=(V, E), and use a structural learning algorithm based on a probabilistic graphical model (such as the PC algorithm and the GIES algorithm) to automatically learn the causal relationship to obtain the causal relationship graph G(V, E), where the node V represents the data of each port and the influencing factor, and the edge E represents the causal relationship, that is, how the data of a certain port affects other ports, such as how the data change of a certain port is affected by time error, interference intensity or equipment error, or how the data anomaly of a port will be transmitted and affect other ports.

[0029] After constructing the causal relationship graph, the propagation direction and impact range of data deviations between ports are further quantified. For example, if the data anomaly of a port affects multiple other ports, the data consistency problem of this port is more serious and needs to be optimized first. Therefore, the transfer entropy is used to calculate the impact of data deviations. The specific steps are as follows: Calculate the information flow between data flows, that is, assume that the data flows of ports i and j are and , the transfer entropy is defined as: ,in, represents the information entropy of the data stream, Indicates the impact strength of the data of port i on the future value of port j; If the uncertainty is greatly reduced after adding the data flow of port i, it means right There is a stronger information contribution, that is, there is a stronger causal influence; By calculating the transfer entropy between all ports, the data deviation propagation matrix is ​​constructed: ,This matrix is ​​used to identify the key influencing ports, i.e., the ports where data deviation propagates most seriously. The larger the transfer entropy, the more serious the data deviation, e.g., It indicates the relative contribution of port i in influencing the future state of port j. The larger the transfer entropy, the easier it is for data deviation to be propagated by this port, thus providing a basis for identifying key influencing ports.

[0030] After identifying the propagation path of data deviation, quantitatively evaluate the real impact of data drift in order to compensate for it. Use causal intervention to calculate the amount of data drift. The specific steps are as follows: Define the data drift amount, and assume that the data drift amount of port i is: , which means that in eliminating the time error In the case of , the deviation between the theoretical expected value of the port data and the actual observed value; It means that human intervention (i.e. eliminating clock drift, making The theoretical expectation value under is the expected value actually observed; The true contribution value of data deviation is calculated based on Bayesian inference. The contribution value expression of data deviation is: ,in, is the posterior probability, which indicates the data drift of port i after the actual observation value X is known. The probability of it actually happening; The amount of data drift at port i The probability of observing data X is is the data drift The prior probability of is the total probability of the actual observed value X occurring; the credibility of each port drift is quantitatively analyzed through Bayesian inference, so as to perform subsequent synchronization compensation.

[0031] Data from different ports may have different optimal alignment windows. If the synchronization window is too large, irrelevant data may be aligned, increasing data noise; if the synchronization window is too small, data loss may occur. Therefore, it is necessary to calculate the optimal synchronization window size to minimize data drift errors. Assume that the data drift of port i is , then the synchronization error function can be defined as: ,in, is the data synchronization window size, is the relationship function between window size and drift error (obtained by fitting historical data); Use the optimization method to find the window size that minimizes the synchronization error: ,The synchronization window is selected to minimize the data drift error, thus enabling optimal data alignment and synchronization.

[0032] Furthermore, after adjusting the synchronization window, some ports may still have clock drift, and the optimal time compensation amount can be further calculated to correct the data timestamp to make the data alignment more accurate.

[0033] For example, a test device has 4 ports, the signal collected by each port is voltage data, the sampling rate is 100Hz, and each signal has 1000 sampling points. The specific implementation steps are as follows: The voltage time series obtained from the four ports ; Time synchronization error Set about 0.05 seconds, signal interference intensity 0.2, equipment error is 0.01 (after normalization); synchronization window candidate range: from 2 seconds to 10 seconds; fitting function is a linear relationship ; After preliminary statistics of the port data and related factors, the PC algorithm is used to automatically learn the causal structure, and the following example relationship is obtained: and Affected greatly by time synchronization error; and They influence each other, and are also affected by the signal interference intensity; The entropy between the ports is calculated as follows: ; ; ; Other combinations are slightly lower; The propagation matrix T is constructed as follows: , and so on; it can be identified from the matrix that port 2 has the greatest impact on port 3, indicating that the data synchronization accuracy of these two ports is particularly critical; The Bayesian formula calculates the posterior probability of data drift for each port. For example, if the probability for port 2 is higher, it means that the data drift problem of port 2 is more serious and needs to be compensated first; According to the aforementioned data drift and fitting function, the synchronization error function is calculated. In this example, if w Seconds and The error can be as small as 1 second. According to other synchronization requirements (such as data coverage, real-time requirements, etc.), you can choose seconds as the optimal synchronization window; Data synchronization is achieved through causal inference methods. First, structural learning is used to determine the causal relationship between the data of each port and its influencing factors, and to identify which factors (such as time error, signal interference, and equipment error) have the greatest impact on data consistency. Subsequently, through quantitative calculation of transfer entropy and data drift, the propagation path of data deviation and the inconsistency in data synchronization of each port can be captured. Then, the synchronization error function and optimization method are used to automatically select the optimal synchronization window and time compensation amount, so as to finely align and compensate the data, achieve consistency in time and value of multi-port data, and thus achieve high-precision and real-time data synchronization.

[0034] Step 3: Multimodal data fusion based on cross-modal Transformer; After completing the causal inference-driven data consistency analysis in step 2, the source of data consistency deviation has been identified, and the data synchronization strategy has been adjusted based on causal inference. However, in the application scenario of multi-port test equipment, the data may contain different modes (such as voltage, current, spectrum, image, temperature, displacement, etc.), and there are nonlinear correlations and time-series dependencies between these modes. It is difficult to effectively capture the complex relationship between different modal data, resulting in information loss or fusion distortion; The characteristic spaces of different modal data are different. For example, the voltage and current signals have different physical units in numerical value and cannot be directly numerically calculated. The data are at different time scales, for example, some sensors (such as temperature sensors) are sampled at a low frequency, while other sensors (such as spectrum analyzers) are sampled at a high frequency, making it difficult to time-align the data.

[0035] The data of each port usually have large differences in time dimension and numerical scale. For example, the data of some ports are high-frequency signals, while the data of some ports are low-frequency measurements, which need to be aligned; For time series signal data (such as voltage and current), linear interpolation is used; for low-frequency data, the nearest value is kept; for frequency domain data, Fourier interpolation is performed; Token is a data unit used for Transformer processing. In different tasks and data types, the data Token structure of each time step is: ,in, The data values ​​after time alignment are represented by interpolation or resampling to obtain the standard time points. For example, for high-frequency and low-frequency signals, different interpolation strategies (such as linear interpolation, Fourier interpolation or keeping the nearest value) are used to ensure that the data are expressed in the same time base; For time position coding, it is generated using sine and cosine functions; It is a modal index, which is used to distinguish different data modalities (for example, voltage is 1, current is 2, spectrum is 3, image is 4, etc.). Modal information can be added in the form of an embedded vector so that the subsequent network can distinguish data from different sources. Use Transformer position encoding to identify data at different time steps and construct Token sequences , M is the total number of time steps, and is processed.

[0036] The data features of different modalities are distributed differently, so the data is feature encoded to extract key information within the modality; Construct a modal feature encoding network using an independent feature encoder Feature extraction for different modalities: ; For example, for a timing signal, A 1D convolutional network (a convolutional neural network used to process one-dimensional sequence data) is used to extract local temporal patterns; for image data, Convolutional neural network (CNN) is used to extract features and analyze spectrum data. Fourier transform (FFT) is used to extract frequency domain features.

[0037] Obtain the feature representation before fusion and calculate the feature vectors of all modalities: , used for Transformer to perform cross-modal information fusion.

[0038] Use the self-attention mechanism to model cross-modal relationships and generate fusion representations, that is, perform cross-modal fusion on these tokens and calculate the self-attention weights: ,in, is the Query vector, is the Key vector, is the normalization factor of the Key dimension, To query the transformation matrix, the input features are projected into the query space so that the queries of different data points have appropriate feature expressions; is the key transformation matrix, which projects the input features into the key space so that the keys of different data points have appropriate feature expressions; The Query key (query vector) represents the information that the current data point wants to ask about other data points. The Key (key vector) represents the characteristics of the current data point and is used for matching queries of other data points. The Value (value vector) stores the content information of the data point and is used to calculate the weighted output in the end.

[0039] Calculated by weighted sum: ,in, is the Value vector, is the value transformation matrix.

[0040] Update the cross-modal representation of the port data and calculate the new fused data representation as: , where H is the original fused data representation, which is reconstructed using decoder g: ;in is the fused data output, and then, represents the difference in different time windows according to the fusion; The fused data representation is an optimized feature vector or high-dimensional data representation containing multi-port and multi-modal information. The fused data representation has the characteristics of time alignment, complementary modal features, and information enhancement. By analyzing the data, we can detect abnormal patterns, mutation points or abnormal modal relationships in the data, and improve the accuracy of abnormal detection. For example, in power system monitoring, if the current signal of a port is abnormal, while the spectrum signal and temperature signal remain normal, the fused data representation can reveal the deviation between these modes and help accurately locate the cause of the fault. Since the fused data representation combines data from different modes, it can provide more comprehensive information for the data prediction model and improve the prediction accuracy; After tokenizing data from different ports and different modalities, the self-attention mechanism of Transformer can automatically capture the nonlinear and time-series correlation between the modalities. Each modal data carries the original signal and time position information. During the fusion process, the model can learn which time steps and modalities have complementary or redundant information, thereby forming a unified fusion representation. The cross-modal Transformer can fuse redundant information from different signal sources together. Even if a modal data has anomalies or noise, it can be compensated by other modal data, which improves the robustness of the data synchronization process as a whole. By designing the specific formulas and parameters for Token composition, position encoding, linear transformation matrix, attention weight calculation, and synchronization window optimization in detail, the paper illustrates how to use the cross-modal Transformer to fuse information from different data sources, thereby providing a reliable basis for dynamically adjusting data synchronization strategies and ultimately achieving high-precision, real-time data synchronization processing.

[0041] In equipment health monitoring, the prediction model can be trained based on the fused data representation to predict abnormal conditions that may occur in the equipment in the future. In the intelligent decision-making system, the fused data can be used to determine whether the current status meets expectations or to provide optimization suggestions.

[0042] This step uses the self-attention mechanism to model the nonlinear relationship between different modalities, automatically learns the correlation between data, realizes information interaction and feature fusion, time-aligns and tokenizes multimodal data, maps data from different ports to a unified time axis, and converts it into a unified Token representation. Modal feature encoding is used to extract key information from different modalities, and an adaptive attention mechanism is used to calculate the weights between different data points, so that the Transformer automatically pays attention to important features, improves fusion accuracy, and outputs a unified cross-modal representation, ensuring that data from all ports can be optimized and matched in a high-dimensional feature space, providing high-precision input for subsequent data analysis and synchronization strategies.

[0043] It should be noted that the thresholds involved in the embodiments can be determined according to specific scenarios and requirements.

[0044] The present invention constructs a port data association graph, optimizes the connection relationship between ports based on a time-varying adjacency matrix, uses a graph neural network to extract port data representation, optimizes data synchronization strategies, improves the accuracy of data alignment, analyzes the influencing factors of data synchronization errors based on a causal relationship model, identifies the propagation path of data deviations between ports, calculates the data drift, uses causal inference to optimize the synchronization window, dynamically corrects data drift, improves data consistency, time-aligns port data, constructs a unified time scale, extracts modal key information based on feature encoding, uses a self-attention mechanism to model cross-modal data associations, optimizes feature matching between ports, and improves the accuracy of data fusion. The present invention can dynamically adjust data synchronization parameters, reduce data drift errors, achieve high-precision synchronous alignment and fusion of port data, improve data consistency, provide high-quality data input for data analysis, anomaly detection and prediction of test equipment, and improve the stability and intelligence level of the test equipment.

[0045] Embodiment 2: A test equipment data synchronization processing system, such as Figure 2 As shown, specifically including: The strategy determination module is used to construct a port data association graph based on the data of each port of the test equipment, determine the connection relationship between ports based on the time-varying adjacency matrix, use the graph neural network to extract the port data representation, and optimize the data synchronization strategy; The data correction analysis module is used to build a causal relationship model, analyze the influencing factors of data synchronization errors, identify the propagation path of data deviations between ports, calculate the data drift, and optimize the synchronization window based on causal inference to correct data drift; The port synchronization optimization module is used to time-align port data, build a unified time scale, extract modal key information based on feature encoding, determine cross-modal data association based on the self-attention mechanism, and optimize feature matching between ports.

[0046] The above formulas are all dimensionless and calculated numerically. Specific dimension removal can be achieved by various means such as standardization, which will not be elaborated here. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0047] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, an ATA hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state ATA hard disk.

[0048] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0049] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0050] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0051] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0052] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0053] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for synchronously processing test equipment data, characterized in that: The steps include: A port data association graph is constructed based on the data of each port of the test equipment, and the connection relationship between ports is determined based on the time-varying adjacency matrix. The graph neural network is used to extract the port data representation and optimize the data synchronization strategy. Build a causal relationship model, analyze the factors affecting data synchronization errors, identify the propagation path of data deviations between ports, calculate the amount of data drift, and optimize the synchronization window based on causal inference to correct data drift; The port data is time-aligned to construct a unified time scale, and the key information of the modality is extracted based on feature encoding. The cross-modal data association is determined based on the self-attention mechanism, and the feature matching between ports is optimized.

2. A method for synchronously processing test equipment data according to claim 1, characterized in that: Construct a port data association diagram based on the data of each port of the test equipment. The specific process is as follows: Construct a port data association graph for the test equipment, define the port set and edge set of the test equipment, the edge set E is determined by the correlation between the port data, and represents the dynamic connection relationship between the ports. The edge set is expressed as: ,in, is the data correlation function between ports i and j at time t, is the correlation threshold; Construct the time-varying adjacency matrix A(t), the expression is: ,in, Controls the rate at which correlation decays, is the association weight between ports i and j.

3. A method for synchronously processing test equipment data according to claim 2, characterized in that: And the connection relationship between ports is determined based on the time-varying adjacency matrix. The specific process is as follows: Calculate the time-varying covariance matrix of the ports to determine the correlation between the port data: The calculation expression of the time-varying covariance matrix is: ,in, and Represent the data of port i and j in time window T respectively.

4. A method for synchronously processing test equipment data according to claim 3, characterized in that: Use graph neural network to extract port data representation and optimize data synchronization strategy. The specific steps are as follows: The graph neural network is used to extract the port data representation. The initial port data features are: , X(t) represents the original data collected by the multi-port test equipment at time t; Build a graph neural network model and use graph convolution for feature update: ,in, is the node feature of the lth layer, is the learned weight matrix, is a nonlinear activation function; The loss function used for training the graph neural network model is: , optimized by gradient descent , output port data characterization ; Calculate the optimal synchronization window: , where D is the data consistency measurement function; The synchronization parameters are calculated based on the port data characterization to minimize data drift, and the optimized data synchronization strategy is output for port optimization.

5. A method for synchronously processing test equipment data according to claim 4, characterized in that: Construct a causal relationship model, analyze the influencing factors of data synchronization errors, and identify the propagation path of data deviations between ports. The specific process is as follows: Define the set of causal variables: ,in, is the data flow of port i at time t, is the data flow of port N; is the time synchronization error, indicating the clock drift between ports; is the signal interference strength, indicating the degree of signal influence between ports; is the equipment error, which represents the systematic deviation of the sensor or measuring device; Use a structural learning algorithm based on a probabilistic graphical model to automatically learn causal relationships and obtain a causal relationship graph. The nodes of the causal relationship graph represent the data of each port and the influencing factors, and the edges represent the causal relationships. After constructing the causal relationship diagram, transfer entropy is used to calculate the impact of data deviation; The transfer entropy expression is: , where H(X) represents the information entropy of the data stream, Indicates the impact strength of the data of port i on the future value of port j, and are the data flows of ports i and j respectively; By calculating the transfer entropy between all ports, the data deviation propagation matrix is ​​constructed: , where Indicates the strength of the impact of the data at port k on the future value of port j.

6. A method for synchronously processing test equipment data according to claim 5, characterized in that: Calculate the data drift, optimize the synchronization window based on causal inference, and correct the data drift. The specific process is as follows: The propagation path is determined by the data deviation propagation matrix constructed based on the transfer entropy between ports, and the data drift is calculated using causal intervention: The deviation between the theoretical value and the actual observed value of the port data is taken as the data drift; The contribution of data deviation is calculated based on Bayesian inference, and the expression is: , is the posterior probability, which indicates the data drift of port i after the actual observation value X is known. The probability of it actually happening; is the data drift of port i The probability that the process observes the actual observation value X, is the data drift The prior probability of is the total probability of the actual observed value X occurring; Define the synchronization error function: ,in, It is the data synchronization window. It is a function of the relationship between window size and drift error; Use the optimization method to find the window size with the smallest synchronization error: ; The smallest synchronization window is chosen to minimize data drift errors.

7. A method for synchronously processing test equipment data according to claim 6, characterized in that: Time-align the port data, build a unified time scale, extract modal information based on feature encoding, determine cross-modal data association based on the self-attention mechanism, and optimize feature matching between ports. The specific process is as follows: Time-align the port data and use linear interpolation for the timing signal data; keep the nearest value for the low-frequency data; and perform Fourier interpolation on the frequency domain data; Use Transformer position encoding to identify data at different time steps and construct a Token sequence; Construct a modal feature encoding network and use independent feature encoders to extract features from different modalities; Obtain the feature representation before fusion, calculate the feature vectors of all modalities, use the self-attention mechanism to model the cross-modal relationship, generate a fused representation, and optimize the feature matching between ports based on the fused representation.

8. A test equipment data synchronization processing system, used to implement a test equipment data synchronization processing method according to any one of claims 1 to 7, characterized in that: include: The strategy determination module is used to construct a port data association graph based on the data of each port of the test equipment, determine the connection relationship between ports based on the time-varying adjacency matrix, use the graph neural network to extract the port data representation, and optimize the data synchronization strategy; The data correction analysis module is used to build a causal relationship model, analyze the influencing factors of data synchronization errors, identify the propagation path of data deviations between ports, calculate the data drift, and optimize the synchronization window based on causal inference to correct data drift; The port synchronization optimization module is used to time-align port data, build a unified time scale, extract modal key information based on feature encoding, determine cross-modal data association based on the self-attention mechanism, and optimize feature matching between ports.

Images