Primary and secondary fusion centralized DTU equipment intelligent simulation test method and system

Abstract

The invention discloses an intelligent simulation test method and system for primary and secondary fusion centralized DTU (Data Transfer Unit) equipment, particularly relates to the field of automatic testing of a power distribution network, and is used for solving the problems of loop cross interference signal separation and traceability. Accurate separation and traceability of cross-loop cross interference signals are realized, and the accuracy of test data and the reliability of equipment evaluation are guaranteed; the method comprises the following steps: firstly, extracting frequency and phase characteristics of a loop signal through multi-scale analysis to form a basis for interference signal separation; then, a dynamic characteristic track is constructed, accurate separation of a target signal and an interference signal in a frequency domain and a time domain is realized, path tracking is carried out in combination with propagation delay and phase deviation, and a conduction path and a source of the interference signal are accurately positioned; finally, the separated signal features are input into a feature library, interference recognition standards are optimized by means of a real-time feedback mechanism, and it is ensured that the system has the self-adaptive recognition capability in a variable test environment.

Description

Technical Field

[0001] The present invention relates to the field of power distribution network automation testing, and more specifically, to an intelligent simulation test method and system for primary and secondary fusion centralized DTU equipment. Background Art

[0002] During the testing process of the centralized DTU equipment that integrates primary and secondary systems, in order to ensure the stability and safety of the operation of the power distribution network, the test equipment must have the ability to collect and monitor the signal status of different circuits in real time. Due to the high degree of integration of the distribution automation system, the signals between the primary and secondary devices are interrelated during the test, allowing the centralized DTU equipment to collect and analyze data from multiple dimensions. Especially in scenarios with dense wiring and highly complex power circuits, the interference between different circuits becomes particularly complex. In this scenario, various types of signals may be affected by electromagnetic interference, frequency disturbances, or phase shifts during transmission, resulting in cross-interference between signals in different circuits, further increasing the difficulty of determining the source of the signal.

[0003] Existing systems lack effective cross-interference signal recognition and separation capabilities. When processing multi-frequency tests, problems such as signal superposition or crosstalk often occur. Interference signals and target signals are difficult to distinguish accurately, resulting in limited authenticity and validity of test data. Especially in frequently changing test environments, signal conduction paths and interference sources are difficult to effectively track and locate. Equipment performance evaluation is prone to misjudgment or omissions due to signal mixing, affecting the reliability of test results. In addition, the processing methods of existing test equipment are often limited to simple filtering methods in the frequency or time domain, and cannot perform in-depth analysis of signal propagation characteristics in complex loops, thereby limiting the equipment's interference management capabilities and data traceability accuracy in high-density connection environments. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent simulation test method and system for a primary-secondary fusion centralized DTU device. Through multi-layer feature analysis and dynamic path tracking in the frequency domain and time domain, the accurate separation and tracing of cross-loop interference signals are achieved, and under the synergistic effect of each module, the accuracy of the test data and the reliability of the equipment evaluation are ensured.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A primary and secondary fusion centralized DTU equipment intelligent simulation test method, comprising:

[0007] Step 1: Perform multi-scale analysis to extract the features of different loop signals and form a preliminary feature division of the target signal and the cross-interference signal;

[0008] Step 2: Based on the preliminary feature division results, a dynamic feature trajectory based on frequency domain feature parameters and time domain feature parameters is constructed to remove the sudden change interference components and retain the frequency and phase integrity of the target signal to achieve separation of the target signal and the interference signal;

[0009] Step 3, feature matching is performed on the separated target signal and interference signal with the feature template in the feature library. If the signal feature matching is successful, the matched feature template is directly called to determine the specific conduction path and source loop of the interference signal; if the matching is unsuccessful, the specific conduction path and source loop of the interference signal are determined by analyzing the difference in propagation characteristics between the separated target signal and the interference signal;

[0010] Step 4, entering the separated signal features into a feature library, optimizing the processing standard for interference signal identification in real time and forming a feature template, and feeding back the updated feature library to step 3.

[0011] Preferably, the step 1 specifically includes:

[0012] Step 1-1, divide the original signal into multiple scales according to the time window, and divide it into long-scale signal and short-scale signal;

[0013] Step 1-2: Transform the long and short scale signals through wavelet transform to obtain the time-frequency feature matrix M(f, t), and generate the target signal feature matrix M through the time-frequency feature matrix g (f, t) and interference signal feature matrix M i (f, t), preliminarily determine the interference frequency segment of each loop; each element of the time-frequency feature matrix represents the feature intensity at frequency f and time point t, which is used to analyze the signal change pattern at each scale to distinguish the target signal of the loop from the potential interference components;

[0014] Step 1-3, perform double-layer frequency domain decomposition on the time-frequency feature matrix to calculate the signal distribution characteristics in the frequency domain, wherein the double-layer frequency domain includes the main frequency offset and phase transfer coefficient of the signal, the main frequency offset is used to measure the distribution characteristics of the target signal at the main frequency, and the phase transfer coefficient is compared between different loops to find the corresponding relationship with the interference frequency;

[0015] Step 1-4, by comparing the main frequency offset and the phase transfer coefficient, the different frequency band features in the time-frequency feature matrix are matched and screened, the frequency components belonging to cross-interference are identified and marked in the time-frequency feature matrix. After the screening is completed, the preliminary feature division of the target signal and the cross-interference signal is obtained.

[0016] Preferably, the time window in step 1-1 is determined by the frequency f and the time resolution T. relationship.

[0017] Preferably, the main frequency offset F in step 1-3 is o for:

[0018] F o =∑ f (f·M(f,t)) / ∑ f M(f,t).

[0019] Preferably, the formula for matching and screening different frequency band features in the time-frequency feature matrix in steps 1-4 is:

[0020]

[0021] Among them, ∈ is the frequency tolerance of the target signal, and δ is the phase deviation tolerance.

[0022] Preferably, the frequency domain characteristic parameters in step 2 include a frequency domain local variation consistency coefficient, and the calculation method of the frequency domain local variation consistency coefficient is:

[0023] Based on the distribution of each frequency in the time-frequency feature matrix M(f,t), the local fluctuation characteristics of the frequency component in time are extracted;

[0024] Calculate the local standard deviation of the frequency distribution on the time axis, that is, the local energy fluctuation E of each frequency component 波动 (f);

[0025] The local coefficient of variation C of each frequency component is calculated by using the ratio of the local standard deviation to the average intensity of the frequency component. 变异 (f);

[0026] The frequency domain local variation consistency coefficient is obtained by weighted summing of the local variation coefficients of all frequencies.

[0027] Preferably, the local standard deviation is:

[0028]

[0029] Where N is the total number of time samples, M(f,t) is the signal strength value of the corresponding frequency f at time t in the matrix, Represents the average value of the corresponding frequency component at all time points.

[0030] Preferably, the local coefficient of variation is:

[0031]

[0032] Among them, C 变异 (f) is the local coefficient of variation corresponding to frequency f.

[0033] Preferably, the frequency domain local variation consistency coefficient is:

[0034] FDLFCI=∑ f C 变异 (f)·exp(-α·|f-f0|);

[0035] Among them, f0 is the main frequency of the target signal, and α is the attenuation coefficient.

[0036] Preferably, the time domain characteristic parameter in step 2 includes a time-phase variation rate coefficient, and a method for calculating the time-phase variation rate coefficient includes:

[0037] Extract the phase information of each time point from the time-frequency feature matrix to calculate the phase change rate on the time axis;

[0038] Calculate the sum of squares of the difference between the phase change rate and the mean of the phase change rate to form a phase change rate fluctuation intensity matrix;

[0039] Combined with the offset weight on the time axis, the fluctuation intensity matrix is ​​weighted summed to obtain the time-phase variation rate coefficient.

[0040] Preferably, the phase change rate is:

[0041]

[0042] in, It is the time derivative of the phase, which represents the rate of change of the phase Φ(t) with respect to time t; Φ(t) is the phase of the frequency component in the time-frequency characteristic matrix.

[0043] Preferably, the time-phase variation rate coefficient is:

[0044]

[0045] Among them, t0 is the time center of the target signal, and β is the offset weight coefficient on the time axis.

[0046] Preferably, the step 2 comprises:

[0047] The frequency domain local variation consistency coefficient and the time-phase variation rate coefficient are comprehensively calculated to form the target interference identification value TIIV, TIIV=FDLFCI×ln(TPVRC+1), where FDLFCI is the frequency domain local variation consistency coefficient and TPVRC is the time-phase variation rate coefficient;

[0048] If the signal recognition value is higher than the preset threshold, the signal is considered to be an interference signal and is eliminated; otherwise, it is retained as a target signal.

[0049] Preferably, in step 3, by analyzing the difference in propagation characteristics between the separated target signal and the interference signal, determining the specific conduction path and source loop of the interference signal specifically includes:

[0050] Step 3-1, calculating the propagation delay differences of the interference signal at different times to establish a time delay difference matrix;

[0051] Step 3-2, for the phase value of the signal at different times and frequencies, calculate its phase change amount hour by hour, and accumulate the change amount to form a phase offset;

[0052] Step 3-3, combining the time delay difference matrix and the phase offset to generate a path weight coefficient;

[0053] Step 3-4: According to the path weight coefficient, identify the main propagation path of the signal by point-by-point tracking, select the point with the highest path weight coefficient as the initial propagation source of the interference signal, and determine the path nodes in turn according to the adjacent weights, and gradually restore the conduction path of the interference signal;

[0054] Step 3-5, the corresponding path identifies the propagation trajectory of the interference signal between different loops and locates the interference source loop.

[0055] Preferably, the path weight in step 3-3 is:

[0056]

[0057] Preferably, the step 4 specifically includes:

[0058] Step 4-1, normalizing the characteristic values ​​of the separated target signal and interference signal;

[0059] Step 4-2, based on the normalized feature data, establish a pattern feature vector of the interference signal, and combine the average value and standard deviation of different interference signal features into a feature template to describe the typical distribution characteristics of the corresponding class of signals;

[0060] Step 4-3, compare the similarity of the newly entered interference signal features with the existing templates in the feature library to determine whether they conform to the existing feature pattern; specifically, when the new feature highly matches a template, the data of the new feature is included in the statistical range of the corresponding template; if the similarity is low, a new pattern template is generated for the corresponding signal type; the new signal features that conform to the existing pattern characteristics are dynamically included in the feature library, and the feature template is updated in real time.

[0061] A primary and secondary fusion centralized DTU equipment intelligent simulation test system, including:

[0062] The feature analysis module performs multi-scale analysis to extract the features of different loop signals, forms a preliminary feature division of the target signal and the cross-interference signal, and sends the division result to the signal separation module;

[0063] The signal separation module constructs a dynamic feature trajectory based on the frequency domain and time domain, removes the sudden interference components and retains the frequency and phase integrity of the target signal, realizes the separation of the target signal and the interference signal, and transmits the separated target signal and interference signal data to the path tracking module and the feature feedback module;

[0064] The path tracking module performs feature matching between the separated target signal and interference signal and the feature template in the feature library. If the signal feature matches successfully, the matched feature template is directly called to determine the specific conduction path and source loop of the interference signal. If the match fails, the specific conduction path and source loop of the interference signal are determined by analyzing the difference in propagation characteristics between the target signal and the interference signal.

[0065] The feature feedback module enters the separated signal features into the feature library, provides real-time feedback to optimize the processing standards for interference signal identification and forms a feature template, and feeds back the updated template to the path tracking module.

[0066] Compared with the prior art, the significant advantages of the primary and secondary fusion centralized DTU equipment intelligent simulation test system of the present invention are:

[0067] The present invention realizes the precise separation and tracing of cross-loop interference signals through multi-layer feature analysis and dynamic path tracking in the frequency domain and time domain, ensuring the accuracy of test data and the reliability of equipment evaluation. The frequency and phase characteristics of the loop signal are extracted through multi-scale analysis, laying the foundation for the subsequent separation of interference signals; the construction of dynamic feature trajectories in the frequency domain and time domain further realizes the precise separation of target signals and interference signals. Based on the separated signal characteristics, path tracking is performed in combination with propagation delay and phase offset to accurately locate the conduction path and source of the interference signal. The separated signal characteristics are entered into the feature library, and the interference identification standard is optimized through a real-time feedback mechanism, so that the system can maintain the ability to identify and adapt to interference signals in a changing test environment. Each module synergistically improves the accuracy of signal processing and traceability. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 The present invention is a structural schematic diagram of a primary-secondary fusion centralized DTU equipment intelligent simulation test system.

[0069] Figure 2 The present invention is a flowchart of a feature feedback module of a primary-secondary fusion centralized DTU equipment intelligent simulation test system. DETAILED DESCRIPTION

[0070] 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.

[0071] The present invention provides a primary and secondary fusion centralized DTU equipment intelligent simulation test system, which is used to solve the problem of loop cross-interference signal separation and tracing. It realizes accurate separation and tracing of cross-loop interference signals through frequency domain and time domain feature analysis and path tracing, so as to ensure the accuracy of test data and the reliability of equipment evaluation. The system first extracts the frequency and phase characteristics of the loop signal through multi-scale analysis to form the basis for interference signal separation. Then, a dynamic feature trajectory is constructed to realize accurate separation of target signals and interference signals in the frequency domain and time domain, and path tracing is performed in combination with propagation delay and phase offset to accurately locate the conduction path and source of the interference signal. Finally, the separated signal features are entered into a feature library, and the interference identification standard is optimized with the help of a real-time feedback mechanism to ensure that the system has adaptive identification capabilities in a variable test environment.

[0072] Figure 1 The invention provides a primary-secondary fusion centralized DTU equipment intelligent simulation test system, comprising: a feature analysis module, a signal separation module, a path tracking module and a feature feedback module.

[0073] Feature analysis module: Perform multi-scale analysis to extract the features of different loop signals, form a preliminary feature division of the target signal and the cross-interference signal, and send the division results to the signal separation module.

[0074] Signal separation module: Based on the dynamic feature trajectories in the frequency domain and time domain, it separates the target signal from the interference signal, eliminates the sudden interference components and retains the frequency and phase integrity of the target signal. The separated target signal data is transmitted to the path tracking module and the feature feedback module.

[0075] Path tracking module: The separated target signal and interference signal are matched with the feature templates in the feature library. If the signal feature match is successful, the matched feature template is directly called to determine the specific conduction path and source loop of the interference signal; if the match is unsuccessful, the specific conduction path and source loop of the interference signal are determined by analyzing the difference in propagation characteristics between the target signal and the interference signal.

[0076] Feature feedback module: The separated signal features are entered into the feature library, real-time feedback is provided to optimize the processing standard for interference signal identification and form a feature template, and the updated template is fed back to the path tracking module.

[0077] In a preferred embodiment, the feature analysis module includes the following contents:

[0078] In the test environment of the primary and secondary integrated centralized DTU equipment, the high-density closed wiring with dense signals often leads to cross-interference between different loops. These interference signals originate from complex electromagnetic characteristics, which make the signals of each loop easily interfere with each other during the test, thus affecting the accurate evaluation of the performance of the equipment. To ensure the accuracy of the test data, it is necessary to perform multi-level analysis on the signals of each loop, and to achieve accurate division of the target signal and the cross-interference signal by decomposing the characteristics of the interference signal layer by layer.

[0079] In order to effectively separate the target signal and the cross-interference signal, a layer-by-layer analysis approach is adopted to extract the characteristic differences of each loop signal without interfering with the overall test signal. The specific processing method is as follows:

[0080] The original signal is divided into multiple scales according to the time window, which is divided into a long-scale signal with a larger time window and a short-scale signal with a smaller time window. Long-scale analysis is used to capture the macroscopic characteristics of the signal and reflect the overall frequency and phase distribution; short-scale analysis is used to observe instantaneous signal changes and focus on interference characteristics such as frequency mutation and waveform mutation. The selected time window is determined by the frequency f and the time resolution T, which can meet the requirements of the original signal. to ensure that the scale division has sufficient sensitivity in capturing rapidly changing signal details.

[0081] The long and short scale signals are transformed by wavelet to obtain the time-frequency feature matrix M(f,t). Each element of the time-frequency feature matrix represents the feature intensity at frequency f and time point t. This feature matrix is ​​used to analyze the signal change pattern at each scale to distinguish the target signal of the loop from potential interference components. The time-frequency feature matrix can also generate the target signal feature matrix M g (f, t) and interference signal feature matrix M i (f, t), preliminarily determine the interference frequency segment of each loop.

[0082] The time-frequency feature matrix is ​​decomposed in the double-layer frequency domain to calculate the signal distribution characteristics in the frequency domain. The double-layer frequency domain features include the main frequency offset and phase shift coefficient θ of the signal to reflect the unique influence of frequency on the signal characteristics. The main frequency offset is calculated by formula F o =∑ f (f·M(f,t)) / ∑ f M(f,t) is calculated, the main frequency offset is used to measure the distribution characteristics of the target signal at the main frequency, and the phase transfer coefficient is compared between different loops to find the corresponding relationship with the interference frequency. The double-layer decomposition of frequency domain features can distinguish the steady-state characteristics of the target signal from the fluctuation characteristics of the interference signal.

[0083] Finally, the different frequency band features in the time-frequency feature matrix are matched and screened. By comparing the main frequency offset and the phase transfer coefficient, the frequency components belonging to cross interference are identified and marked in the time-frequency feature matrix. The screening formula for forming the time-frequency feature matrix is ​​as follows: Where ∈ is the frequency tolerance of the target signal, and δ is the phase deviation tolerance. After the screening is completed, the preliminary feature division of the target signal and the cross-interference signal is obtained, providing an accurate basis for subsequent separation.

[0084] The feature analysis module realizes the refined extraction and differentiation of different loop signal features through multi-scale analysis, so that the target signal and the cross-interference signal can be effectively distinguished in terms of key features such as frequency and phase, forming a feature matrix with multi-level information. Through the construction of the time-frequency feature matrix and the double-layer frequency domain feature decomposition, the dynamic changes of the signal on different time scales are accurately captured to ensure the feature integrity of long-scale and short-scale signals, thereby improving the sensitivity and accuracy of interference signal identification. The matching screening of the main frequency offset and the phase transfer coefficient further effectively isolates the target signal, ensures the reliability of the identification of cross-interference signals in high-density wiring environments, and lays a solid feature foundation for the subsequent separation and tracing of interference signals.

[0085] When separating the target signal from the cross-interference signal, the signal separation module analyzes the differences between the frequency domain and the time domain based on the dynamic feature trajectory to enhance the accuracy of interference identification. This step designs two parameters: the frequency domain local change consistency coefficient and the time-phase variation rate coefficient. Starting from the local consistency of the frequency domain and the phase change rate of the time axis, the interference signal is eliminated by analyzing the different characteristic trajectories of the signal to ensure the integrity of the target signal. The frequency domain local change consistency coefficient is used to quantify the stability of the signal at a specific frequency and identify the high-frequency fluctuation characteristics of the interference signal; the time-phase variation rate coefficient measures the phase variation of the signal on the time axis, captures the phase jump, and then eliminates the interference component. Through the complementary analysis of the two, a dynamic feature trajectory model is constructed to achieve accurate signal separation.

[0086] In a preferred embodiment, the signal separation module includes the following contents:

[0087] In signal separation, the dynamic characteristic trajectory of the target signal is significantly different from that of the cross-interference signal. This difference is reflected in the local fluctuation characteristics in the frequency domain and the phase change rate in the time domain. Usually, the target signal changes relatively smoothly in the frequency domain, while the interference signal exhibits abruptness or instability. At the same time, the phase change rate on the time scale is significantly different, and the interference signal often has a significant phase jump. Therefore, by calculating the characteristic parameters of the frequency domain and the time domain and constructing a dynamic characteristic trajectory, the target signal and the interference signal can be effectively distinguished. The characteristic parameters in the frequency domain include the local change consistency coefficient in the frequency domain; the characteristic parameters in the time domain include the time-phase variation rate coefficient;

[0088] Frequency domain local variation consistency coefficient: Evaluates the local consistency of the signal in the frequency domain to identify stable target signals and wildly fluctuating interference signals.

[0089] Time-phase variation rate coefficient: measures the rate of phase change on the time axis and is used to identify interference signals with sudden phase changes.

[0090] The frequency domain local variation consistency coefficient is based on the time-frequency feature matrix obtained by the feature analysis module, which quantifies the consistency of each frequency component in time to identify the stability characteristics of the signal. The following is the detailed calculation process:

[0091] First, based on the distribution of each frequency in the feature matrix, the local fluctuation characteristics of the frequency component in time are extracted. Let the feature matrix be a two-dimensional matrix with the horizontal axis representing time and the vertical axis representing frequency. Each element in the matrix represents the intensity of a certain frequency component at a certain time point.

[0092] Calculate the local standard deviation of the frequency distribution on the time axis to obtain the local energy fluctuation of each frequency component to identify the stability of the frequency component. The local standard deviation formula is as follows: Where N is the total number of time samples, M(f,t) is the signal strength value of the corresponding frequency f at time t in the matrix, Represents the average value of the corresponding frequency component at all time points. The local coefficient of variation of each frequency component is calculated using the ratio of the local standard deviation to the average intensity of the frequency component to quantify the consistency of frequency changes: The coefficient of variation characterizes the volatility of the frequency component over time. The larger the value, the more violent the fluctuation, which usually corresponds to the characteristics of the interference signal; the smaller the value, the smaller the fluctuation, which usually corresponds to the stable characteristics of the target signal. Finally, by weighted summing the local coefficients of variation of all frequencies, the frequency domain local variation consistency coefficient FDLFCI is obtained, which further highlights the interference characteristics that are far away from the main frequency of the target signal. The formula is as follows: FDLFCI = ∑ f C 变异(f)·exp(-α·|f-f0|); where f0 is the main frequency of the target signal and α is the attenuation coefficient, which is used to reduce the interference effect far from the main frequency. This coefficient combines the fluctuation consistency of each frequency component to more accurately identify the presence of the interference signal in the frequency domain.

[0093] The frequency domain local variation consistency coefficient is used to indicate the stability and consistency of the signal within a specific frequency range, reflecting the stability of the signal energy in the frequency domain over time. The larger the value of this coefficient, the greater the fluctuation of the frequency component over time, which usually corresponds to the characteristics of the interference signal; the smaller the value, the signal changes smoothly within the frequency band, the energy distribution is uniform, and it meets the stability characteristics of the target signal. This coefficient makes the signal stability judgment more accurate by quantifying the intensity of the local frequency variation, which helps to identify and distinguish abnormal fluctuation signals in the frequency domain.

[0094] The time-phase variation rate coefficient is based on the time-frequency feature matrix obtained by the feature analysis module, which quantifies the phase change rate of the signal on the time axis to identify the stability and mutation characteristics of the phase in the time domain. The specific calculation process is as follows:

[0095] Extract the phase information at each time point from the feature matrix to calculate the phase change rate on the time axis. Assume that each element in the matrix contains phase information, and define the instantaneous phase change rate P(t) at the time point, which is calculated as follows: in, It is the time derivative of the phase, which represents the rate of change of the phase Φ(t) with respect to time t; Φ(t) is the phase of the frequency component in the time-frequency feature matrix. The rate of change of the phase is calculated at each time point to capture the phase mutation characteristics that may be generated by the interference signal.

[0096] In order to identify the interference characteristics in the frequency domain, the square sum of the phase change rate is further calculated to form the phase change rate fluctuation intensity matrix Q(t) to characterize the concentration degree of phase variation: in, is the average of the phase change rate at all time points, indicating the overall stable trend of the signal. The larger the value of the fluctuation intensity matrix, the more drastic the phase change, which usually points to a sudden change in the interference signal.

[0097] Combined with the offset weights on the time axis, the fluctuation intensity matrix is ​​weighted and summed to obtain the time-phase variation rate coefficient TPVRC, which highlights the interference components far from the time center: Where t0 is the time center of the target signal, and β is the weight coefficient of the time offset, which is used to reduce the interference weight that deviates from the target signal characteristics. The time-phase variation rate coefficient can capture the characteristics of sudden phase change signals and help distinguish the phase mutations of interference signals.

[0098] The time-phase variation rate coefficient is used to indicate the phase change rate of the signal on the time axis, reflecting the phase stability and mutation trend of the signal at different times. The larger the value of this coefficient, the faster the phase change rate of the signal is, and there are significant jumps or irregular fluctuations, which usually point to the characteristics of the interference signal; the smaller the value, the smoother the phase change is, which conforms to the continuity characteristics of the target signal. This coefficient provides sensitivity to signal fluctuations in time by quantifying the phase variation rate, which helps to separate the interference components of short-term mutations and realize accurate identification of time domain characteristics.

[0099] Finally, the frequency domain local variation consistency coefficient and the time-phase variation rate coefficient parameters are comprehensively calculated to form the target interference identification value TIIV as the basis for removing interference signals. Through the complementary characteristics of the frequency domain and the time domain, the target interference identification value integrates the dynamic change information of the signal: TIIV = FDLFCI × ln (TPVRC + 1); where the logarithmic operation is used to smooth the differences between parameters to avoid the influence of outliers. If the target interference identification value is higher than the preset threshold, the signal is considered to be an interference signal and is removed; otherwise, it is retained as a target signal.

[0100] After the interference signal is removed, the integrity of the remaining target signal is reviewed to check its frequency and phase continuity to confirm that the structure of the target signal is not destroyed during the removal process. The dynamic compensation algorithm is used to restore the tiny faults caused by interference removal to ensure the integrity of the target signal. The target signal data set finally generated retains all the key features of the loop signal, providing an accurate signal basis for subsequent traceability analysis.

[0101] The signal separation module achieves effective separation of target signals and interference signals through dynamic feature trajectories, and constructs a complementary feature discrimination system in the frequency domain and time domain. The local variation consistency coefficient in the frequency domain captures the stability of the frequency component and has a sensitive recognition capability for sudden interference signals; the time-phase variation rate coefficient highlights the phase mutations occurring on the time axis by quantifying the phase change rate. The combination of the two enables signal separation to have high accuracy in both frequency and time domain features, reduces the probability of misjudgment of target signals, ensures the integrity and reliability of target signals in complex environments, and provides a stable foundation for subsequent signal analysis.

[0102] After the separation processing of the signal separation module, the characteristic matrix of the target signal and the interference signal has been obtained. Based on the differences in the time-frequency propagation characteristics of the interference signal, the path tracking module constructs the delay and phase offset model of the interference signal, and accurately locates the interference source and its propagation path through the path analysis method.

[0103] In a preferred embodiment, the path tracking module includes the following contents:

[0104] Before processing, the target signal and interference signal obtained from the signal separation module are first feature matched, and the separated signal features are compared with the feature templates stored in the feature library for similarity to determine whether the signal conforms to the existing feature pattern. If the signal feature matches successfully, the path information or processing rules in the corresponding feature template are directly called to quickly locate the interference source and conduction path; if the match is unsuccessful, the propagation delay difference of the interference signal at different times is calculated to establish a time delay difference matrix. The delay difference matrix quantifies the delay distribution of the signal by measuring the energy or eigenvalue difference of the signal at adjacent times, and captures the delay difference caused by the propagation distance or path characteristics of the signal in different paths. The area with a higher delay difference value in the delay difference matrix represents a more significant delay change in signal propagation, which usually corresponds to the main path of the interference signal.

[0105] On the basis of delay difference, the path offset characteristics of the interference signal are further captured by analyzing the phase change of the interference signal. For the phase value of the signal at different times and frequencies, the phase change is calculated hour by hour, and the change is accumulated to form a phase offset index. The phase offset describes the phase variation of the signal in the time dimension and can reveal the different characteristics caused by the phase offset in the propagation path. The interference signal usually manifests as a sudden phase change, while the target signal is more stable in phase, so the propagation characteristics of the interference signal can be better characterized by the phase offset.

[0106] Combine the calculation results of delay differential and phase offset to generate a path weight coefficient. This path weight coefficient normalizes the delay differential and phase offset to give it a joint weight effect of time delay and phase offset to highlight the main propagation path of the interference signal. The path weight coefficient locks the main nodes in the interference propagation path with a relatively high weight, making the high-weight points on the path more prominent and helping to determine the propagation direction of the signal.

[0107] For example, the calculation formula of the path weight coefficient is:

[0108] According to the path weight coefficient, the main propagation path of the signal is identified by point-by-point tracking. The point with the highest path weight coefficient is selected as the initial propagation source of the interference signal, and the path nodes are determined in turn according to the adjacent weights, and the conduction path of the interference signal is gradually restored. Finally, the corresponding path identifies the propagation trajectory of the interference signal between different loops, locates the interference source loop, and provides a clear path basis for signal separation and subsequent processing.

[0109] The path tracking module uses refined path tracking methods to accurately capture the delay and phase offset characteristics of the interference signal during the transmission process, and realizes the clear positioning of the interference signal propagation path. The joint analysis of the delay difference matrix and the phase offset not only improves the accuracy of path tracking, but also effectively reduces the misidentification of the target signal. The path tracking weight coefficient further enhances the significance of the path node, making the source positioning of the interference signal high-precision and high-reliability, providing a scientific basis for tracing the interference problem in a complex loop environment.

[0110] like Figure 2 As shown, in a preferred embodiment, the feature feedback module includes the following contents:

[0111] First, the eigenvalues ​​of the separated target signal and interference signal are normalized to ensure the stability of the signal characteristics under different sampling environments. Normalization calculates the difference between the eigenvalue and its average value and standardizes the result, thereby unifying the values ​​of different signal characteristics into a comparable range. The normalized eigenvalue can reduce the impact of environmental differences and lay a unified feature foundation for subsequent model modeling.

[0112] Based on the normalized feature data, the pattern feature vector of the interference signal is established, and the average value and standard deviation of different interference signal features are combined into a feature template to describe the typical distribution characteristics of the corresponding type of signal. The feature vector generates a representative pattern template by extracting the mean and variation range of different dimensions such as time, frequency and phase to help identify the matching degree of the new signal. Feature pattern modeling provides a clear standard for the structured description of interference signals, and can identify and distinguish different types of interference signal features.

[0113] Through the feedback mechanism, the newly entered interference signal features are compared with the existing templates in the feature library for similarity to determine whether they conform to the existing feature pattern. When the new feature is highly matched with a template, the data of the new feature is included in the statistical range of the template; if the similarity is low, a new pattern template is generated for the corresponding signal type. The feedback mechanism continuously adjusts and expands the recognition criteria of the feature library through real-time judgment of signal similarity to ensure the adaptability of the template in different environments.

[0114] As feedback optimization progresses, new signal features that meet existing pattern characteristics are dynamically incorporated into the feature library, feature templates are updated in real time, and the mean and standard deviation of the pattern vector are adjusted to expand the scope of application of the template. The feature library can dynamically adapt to the changes in the characteristics of new signals after each feedback and data entry, and continuously enhance the accuracy and robustness of the identification of interference signals. The dynamic update process ensures that the feature library can maintain the ability to identify new interference signals in a constantly changing environment, providing continuously optimized feature support for subsequent processing and signal management.

[0115] The feature feedback module gradually improves the accuracy and adaptability of interference signal identification by entering the separated signal features into the feature library and dynamically updating and optimizing them. Normalization ensures the uniformity of feature values ​​and reduces the impact of environmental differences; feature pattern modeling provides a clear structured description for different types of interference signals; the feedback mechanism continuously adjusts the recognition criteria through real-time similarity judgment, making the feature library adaptive; dynamic updates further expand the scope of application of the template, allowing the feature library to remain highly sensitive to changes in the signal environment and quickly adapt to new interference features. In this process, the feature library continuously strengthens its own recognition capabilities, realizes continuous optimization of interference signals, and provides robust and accurate feature support for interference management in complex environments.

[0116] The above formulas are all dimensionless and numerical calculations. 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.

[0117] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a primary-secondary fusion centralized DTU device intelligent simulation test method, which realizes the precise separation and tracing of cross-loop cross-interference signals through multi-layer feature analysis and dynamic path tracking in the frequency domain and time domain, and ensures the accuracy of test data and the reliability of equipment evaluation through synergistic effect.

[0118] Firstly, the frequency and phase characteristics of each loop signal are extracted through multi-scale analysis to form a preliminary feature division of the target signal and the cross-interference signal, which provides a clear feature basis for the subsequent separation of the interference signal.

[0119] Then, by constructing dynamic feature trajectories, the target signal and the interference signal are accurately separated in the frequency domain and time domain, retaining the integrity of the target signal and effectively eliminating the mutation interference component. This separation process lays the foundation for high-precision signal input for subsequent interference signal path tracking;

[0120] In path tracing, the interference signal is traced based on the separated signal characteristics, combined with the propagation delay and phase offset, and its conduction path and source position are accurately locked, providing support for comprehensive signal tracing;

[0121] Finally, the separated signal features are entered into the feature library, and the interference identification criteria are continuously optimized through real-time feedback to form an adaptive feature template, ensuring that the system maintains a keen ability to identify new interference signals in a changing test environment.

[0122] The close cooperation of the modules of the present invention enables the present invention to have a high degree of recognition accuracy, dynamic adaptability and source positioning accuracy, thereby effectively ensuring the authenticity of data and the scientific nature of equipment evaluation in a complex power circuit environment to solve the problems raised in the above-mentioned background technology.

[0123] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0124] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0125] 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 intelligent simulation test of primary and secondary fusion centralized DTU equipment, characterized in that: include: Step 1: Perform multi-scale analysis to extract the features of different loop signals and form a preliminary feature division of the target signal and the cross-interference signal; Step 2: Based on the preliminary feature division results, a dynamic feature trajectory based on frequency domain feature parameters and time domain feature parameters is constructed to remove the sudden change interference components and retain the frequency and phase integrity of the target signal to achieve separation of the target signal and the interference signal; Step 3, feature matching is performed on the separated target signal and interference signal with the feature template in the feature library. If the signal feature matching is successful, the matched feature template is directly called to determine the specific conduction path and source loop of the interference signal; if the matching is unsuccessful, the specific conduction path and source loop of the interference signal are determined by analyzing the difference in propagation characteristics between the separated target signal and the interference signal; Step 4, entering the separated signal features into a feature library, optimizing the processing standard for interference signal identification in real time and forming a feature template, and feeding back the updated feature library to step 3.

2. According to claim 1, a primary and secondary fusion centralized DTU equipment intelligent simulation test method is characterized in that: The step 1 specifically includes: Step 1-1, divide the original signal into multiple scales according to the time window, and divide it into long-scale signal and short-scale signal; Step 1-2: Transform the long and short scale signals through wavelet transform to obtain the time-frequency feature matrix M(f, t), and generate the target signal feature matrix M through the time-frequency feature matrix g (f, t) and interference signal feature matrix M i (f, t), preliminarily determine the interference frequency segment of each loop; each element of the time-frequency feature matrix represents the feature intensity at frequency f and time point t, which is used to analyze the signal change pattern at each scale to distinguish the target signal of the loop from the potential interference components; Step 1-3, perform double-layer frequency domain decomposition on the time-frequency feature matrix to calculate the signal distribution characteristics in the frequency domain, wherein the double-layer frequency domain includes the main frequency offset and phase transfer coefficient of the signal, the main frequency offset is used to measure the distribution characteristics of the target signal at the main frequency, and the phase transfer coefficient is compared between different loops to find the corresponding relationship with the interference frequency; Step 1-4, by comparing the main frequency offset and the phase transfer coefficient, the different frequency band features in the time-frequency feature matrix are matched and screened, the frequency components belonging to cross-interference are identified and marked in the time-frequency feature matrix. After the screening is completed, the preliminary feature division of the target signal and the cross-interference signal is obtained.

3. According to claim 2, a primary and secondary fusion centralized DTU equipment intelligent simulation test method is characterized in that: The time window in step 1-1 is determined by the frequency f and the time resolution T. relationship.

4. According to claim 2, a primary and secondary fusion centralized DTU equipment intelligent simulation test method is characterized in that: The main frequency offset F in step 1-3 o for: F o =∑ f (f·M(f,t)) / ∑ f M(f,t)。 5. According to claim 2, a primary and secondary fusion centralized DTU equipment intelligent simulation test method is characterized in that: The formula for matching and screening different frequency band features in the time-frequency feature matrix in steps 1-4 is: Among them, ∈ is the frequency tolerance of the target signal, and δ is the phase deviation tolerance.

6. According to claim 2, a primary and secondary fusion centralized DTU equipment intelligent simulation test method is characterized in that: The frequency domain characteristic parameters in step 2 include the frequency domain local variation consistency coefficient, and the calculation method of the frequency domain local variation consistency coefficient is: Based on the distribution of each frequency in the time-frequency feature matrix M(f,t), the local fluctuation characteristics of the frequency component in time are extracted; Calculate the local standard deviation of the frequency distribution on the time axis, that is, the local energy fluctuation E of each frequency component 波动 (f); The local coefficient of variation C of each frequency component is calculated by using the ratio of the local standard deviation to the average intensity of the frequency component. 变异 (f); The frequency domain local variation consistency coefficient is obtained by weighted summing of the local variation coefficients of all frequencies.

7. According to claim 6, a primary and secondary fusion centralized DTU equipment intelligent simulation test method is characterized in that: The local standard deviation is: Where N is the total number of time samples, M(f,t) is the signal strength value of the corresponding frequency f at time t in the matrix, Represents the average value of the corresponding frequency component at all time points.

8. According to claim 6, a method for intelligent simulation test of a centralized DTU device with primary and secondary integration, characterized in that: The local coefficient of variation is: Among them, C 变异 (f) is the local coefficient of variation corresponding to frequency f.

9. According to claim 6, a primary and secondary fusion centralized DTU equipment intelligent simulation test method is characterized in that: The frequency domain local variation consistency coefficient is: FDLFCI=∑ f C 变异 (f)·exp(-α·|f-f0|); Among them, f0 is the main frequency of the target signal, and α is the attenuation coefficient.

10. According to claim 6, a primary and secondary fusion centralized DTU equipment intelligent simulation test method is characterized in that: The time domain characteristic parameter in step 2 includes a time-phase variation rate coefficient, and a calculation method of the time-phase variation rate coefficient includes: Extract the phase information of each time point from the time-frequency feature matrix to calculate the phase change rate on the time axis; Calculate the sum of squares of the difference between the phase change rate and the mean of the phase change rate to form a phase change rate fluctuation intensity matrix; Combined with the offset weight coefficient on the time axis, the fluctuation intensity matrix is ​​weighted summed to obtain the time-phase variation rate coefficient.

11. According to claim 9, a method for intelligent simulation test of a centralized DTU device with primary and secondary integration, characterized in that: The phase change rate is: in, It is the time derivative of the phase, which represents the rate of change of the phase Φ(t) with respect to time t; Φ(t) is the phase of the frequency component in the time-frequency characteristic matrix.

12. According to claim 9, a method for intelligent simulation test of a centralized DTU device with primary and secondary integration, characterized in that: The time-phase variation rate coefficient is: Among them, t0 is the time center of the target signal, and β is the offset weight coefficient on the time axis.

13. The intelligent simulation test method for a primary and secondary fusion centralized DTU device according to claim 10 is characterized in that: The step 2 comprises: The frequency domain local variation consistency coefficient and the time-phase variation rate coefficient are comprehensively calculated to form the target interference identification value TIIV, TIIV=FDLFCI×ln(TPVRC+1), where FDLFCI is the frequency domain local variation consistency coefficient and TPVRC is the time-phase variation rate coefficient; If the signal recognition value is higher than the preset threshold, the signal is considered to be an interference signal and is eliminated; otherwise, it is retained as a target signal.

14. According to claim 1, a method for intelligent simulation test of a centralized DTU device with primary and secondary integration, characterized in that: In step 3, by analyzing the difference in propagation characteristics between the separated target signal and the interference signal, the specific conduction path and source loop of the interference signal are determined, which specifically includes: Step 3-1, calculating the propagation delay differences of the interference signal at different times to establish a time delay difference matrix; Step 3-2, for the phase value of the signal at different times and frequencies, calculate its phase change amount hour by hour, and accumulate the change amount to form a phase offset; Step 3-3, combining the time delay difference matrix and the phase offset to generate a path weight coefficient; Step 3-4: According to the path weight coefficient, identify the main propagation path of the signal by point-by-point tracking, select the point with the highest path weight coefficient as the initial propagation source of the interference signal, and determine the path nodes in turn according to the adjacent weights, and gradually restore the conduction path of the interference signal; Step 3-5, the corresponding path identifies the propagation trajectory of the interference signal between different loops and locates the interference source loop.

15. The method for intelligent simulation test of primary and secondary fusion centralized DTU equipment according to claim 14 is characterized in that: The path weight in step 3-3 is:

16. The method for intelligent simulation test of a centralized DTU device with primary and secondary integration according to claim 1 is characterized in that: The step 4 specifically includes: Step 4-1, normalizing the characteristic values ​​of the separated target signal and interference signal; Step 4-2, based on the normalized feature data, establish a pattern feature vector of the interference signal, and combine the average value and standard deviation of different interference signal features into a feature template to describe the typical distribution characteristics of the corresponding class of signals; Step 4-3, compare the similarity of the newly entered interference signal features with the existing templates in the feature library to determine whether they conform to the existing feature pattern; specifically, when the new feature highly matches a template, the data of the new feature is included in the statistical range of the corresponding template; if the similarity is low, a new pattern template is generated for the corresponding signal type; the new signal features that conform to the existing pattern characteristics are dynamically included in the feature library, and the feature template is updated in real time.

17. A primary and secondary fusion centralized DTU equipment intelligent simulation test system for implementing any of the methods described in claims 1-16, characterized in that: include: The feature analysis module performs multi-scale analysis to extract the features of different loop signals, forms a preliminary feature division of the target signal and the cross-interference signal, and sends the division result to the signal separation module; The signal separation module constructs a dynamic feature trajectory based on the frequency domain and time domain, removes the sudden interference components and retains the frequency and phase integrity of the target signal, realizes the separation of the target signal and the interference signal, and transmits the separated target signal and interference signal data to the path tracking module and the feature feedback module; The path tracking module performs feature matching between the separated target signal and interference signal and the feature template in the feature library. If the signal feature matches successfully, the matched feature template is directly called to determine the specific conduction path and source loop of the interference signal. If the match fails, the specific conduction path and source loop of the interference signal are determined by analyzing the difference in propagation characteristics between the target signal and the interference signal. The feature feedback module enters the separated signal features into the feature library, provides real-time feedback to optimize the processing standards for interference signal identification and forms a feature template, and feeds back the updated template to the path tracking module.

Images