Power transmission line anomaly identification model construction method combined with edge computing

Abstract

The application relates to the technical field of power transmission line anomaly identification, and particularly discloses a power transmission line anomaly identification model construction method combined with edge computing, which realizes real-time collection of current and voltage signals through step-by-step increase of electromagnetic interference intensity testing, dynamically calculates the influence value of electromagnetic interference by combining high-precision sensor monitoring of electric field intensity and signal change, recursively estimates the dynamic influence of electromagnetic interference through a Kalman filter, updates the system state and measurement data, accurately models the influence of electromagnetic interference, and realizes real-time evaluation of the influence of interference on the system by calculating electromagnetic interference coefficients and electromagnetic shielding efficiency expected deviation; the application adopts a spectrum analysis and high-frequency current and voltage fluctuation feature extraction method, effectively distinguishes accidental interference from system faults by calculating a spectrum ratio, and can accurately distinguish real fault signals from electromagnetic interference signals through multi-level data analysis and optimization, thereby reducing false positives and false negatives.

Description

Technical Field

[0001] This invention relates to the field of power transmission line anomaly identification technology, and specifically to a method for constructing a power transmission line anomaly identification model that incorporates edge computing. Background Technology

[0002] In power systems, the normal operation of transmission lines is crucial; however, electromagnetic interference and system faults often severely impact their stability and safety. Electromagnetic interference typically originates from external environmental factors such as lightning, radio waves, and equipment switching operations. It can cause fluctuations in transmission line current and voltage, affecting signal quality and even triggering false alarms or missed alarms. System faults, such as short circuits, overloads, and grounding faults, are the main causes of power system damage and safety hazards. Because the signal characteristics of electromagnetic interference and system faults overlap, traditional fault detection systems struggle to accurately distinguish between these two types of signals, thus affecting fault diagnosis and stable system operation. Therefore, accurately identifying real fault signals from interference signals in complex electromagnetic environments, and avoiding false alarms and missed alarms, has become a significant technical challenge in current transmission line anomaly identification.

[0003] Existing power system anomaly identification methods largely rely on time-domain signal processing techniques. However, under conditions of strong electromagnetic interference or high-frequency current and voltage fluctuations, traditional methods often struggle to handle complex interference patterns. Especially under high electromagnetic interference intensity, existing systems find it difficult to effectively extract useful features and accurately distinguish between fault and interference signals. Furthermore, traditional fault identification algorithms lack dynamic analysis and optimization of the interference effects, leading to reduced identification accuracy in high electromagnetic interference environments. Therefore, a novel method is urgently needed that can comprehensively consider the dual effects of electromagnetic interference and system faults, improving system robustness and fault identification accuracy through precise models and algorithms. Summary of the Invention

[0004] The purpose of this invention is to provide a method for constructing a transmission line anomaly identification model that incorporates edge computing, in order to solve the problems mentioned above.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] The method for constructing a transmission line anomaly identification model based on edge computing includes the following steps:

[0007] S1: During the anomaly identification process of transmission lines, based on the types of anomalies detected in the transmission lines, including short circuit, overload, and grounding fault, the electromagnetic frequency range and electromagnetic intensity level are set to simulate actual electromagnetic interference conditions.

[0008] S2: Real-time acquisition of current and voltage signals of transmission lines; within a set electromagnetic frequency range, gradually increasing the electromagnetic interference intensity; recording the electric field strength and current and voltage changes at each electromagnetic frequency; and using an electromagnetic field detector to monitor the absorption, reflection, and transmission characteristics of the signal, and recording the impact of electromagnetic interference on the signal.

[0009] The signals include: current signals and voltage signals;

[0010] S3: The acquired signals are classified in real time and labeled as normal signals, abnormal signals, and signals to be determined. Signals to be determined are also marked and saved for subsequent analysis to determine whether they are false signals or real fault signals caused by electromagnetic interference, thus ensuring the accuracy of subsequent signal analysis.

[0011] S4: Initiate system diagnostics. Based on the electromagnetic interference source, signal characteristics, and system operation status, analyze whether the signal is due to accidental interference or system failure. If it is an accidental anomaly, repeat the test. If it is a systemic problem, stop the test and troubleshoot the electromagnetic shielding equipment or system to repair hidden faults and ensure the effectiveness of subsequent tests.

[0012] S5: Perform detailed analysis on the signal data that has been marked as qualified, compare it with the expected performance indicators, evaluate the system's performance under different electromagnetic interference intensities, optimize the identification process, improve the system's fault identification capability under high-frequency current and voltage fluctuation conditions, and ensure that it can accurately distinguish between real fault signals and electromagnetic interference signals.

[0013] As a further aspect of the present invention: acquire electric field strength and current / voltage change data at each electromagnetic frequency; after acquisition, calculate the impact value of electromagnetic interference by analyzing the electric field strength and current / voltage change data, and compare it with the expected electromagnetic shielding efficiency to calculate the expected deviation of electromagnetic shielding efficiency.

[0014] As a further aspect of the present invention: the process for obtaining the expected deviation of the electromagnetic shielding efficiency is as follows:

[0015] Linear equations are used to represent dynamic systems of electromagnetic interference;

[0016] The computational expression for the linear equation is as follows:

[0017] x k =Ax k-1 +Bu k +w k ;

[0018] In the formula, k represents the data collection time point, and x k This represents the changes in electric field strength, current, and voltage caused by electromagnetic interference at time point k. A represents the system state transition matrix, and u...k B represents the external control input, and W represents the input matrix. k Let x represent process noise with covariance Q. k-1 This indicates the effect of electromagnetic interference at time point k-1 on the changes in electric field strength, current, and voltage.

[0019] The current and voltage signals measured by the sensor are represented by measurement equations;

[0020] The calculation expression for the measurement equation is as follows:

[0021] z k =Hx k +v k ;

[0022] In the formula, z k This represents the changes in electric field intensity, current, and voltage measured by the sensor at time point k, where H represents the measurement matrix, and v k This represents process noise with covariance R;

[0023] At each new sample, based on state x k-1 and input u k Predict the state at the current time point k, and denote it as the predicted state;

[0024] The calculation expression for the predicted state is as follows:

[0025]

[0026] In the formula, It is a predicted state;

[0027] The error of the predicted state is estimated to obtain the predicted covariance matrix, and the calculation expression is as follows:

[0028]

[0029] In the formula, Let represent the predicted state covariance matrix, and Q represent the process noise covariance matrix.

[0030] Combined with actual measured value z k To update the state estimate, calculate the Kalman gain, expressed as:

[0031]

[0032] In the formula, K k H represents the Kalman gain between the measured data and the predicted data, H represents the measurement matrix, and R represents the measurement noise covariance matrix.

[0033] Calculate the updated state estimate, and compute the expression:

[0034]

[0035] In the formula, This represents the updated state estimate;

[0036] Based on the updated state estimate from the Kalman filter, the impact of electromagnetic interference on the transmission line signal is calculated. The expression for calculating the impact of electromagnetic interference is as follows:

[0037]

[0038] In the formula, I k This represents the impact value of electromagnetic interference at time point k;

[0039] The deviation value obtained by calculating the difference between the actual electromagnetic interference effect and the expected electromagnetic shielding efficiency, and taking the absolute value, is denoted as the expected deviation of electromagnetic shielding efficiency F. k ;

[0040] The expected electromagnetic shielding efficiency is a system-determined value.

[0041] As a further aspect of the present invention: the influence value of electromagnetic interference and the expected deviation of electromagnetic shielding efficiency at time point k are obtained, the influence value of electromagnetic interference and the expected deviation of electromagnetic shielding efficiency are standardized, and the electromagnetic interference coefficient is calculated.

[0042] As a further aspect of the present invention: the real-time classification of the acquired signals, labeling them as normal signals, abnormal signals, and signals to be determined, specifically includes:

[0043] The electromagnetic interference coefficient is compared with a first preset threshold to determine whether the electromagnetic interference coefficient is greater than or equal to the first preset threshold.

[0044] If so, it means that electromagnetic interference affects the signal so that it cannot be judged as normal or abnormal, and the signal is marked as a signal to be determined.

[0045] If not, it means that electromagnetic interference will not affect the normal or abnormal judgment of the signal. Then, the collected signal is subjected to anomaly identification to determine whether the current signal is a normal signal or an abnormal signal.

[0046] As a further aspect of the present invention: whether the analysis signal is due to accidental interference or caused by a system malfunction specifically includes:

[0047] In the process of identifying anomalies in power transmission lines, Fourier transform and spectral analysis are used to distinguish signals;

[0048] Converting the time-domain signal to the frequency-domain signal, the Fourier transform calculation expression for each time point k is as follows:

[0049]

[0050] In the formula, f represents frequency, X(f) represents the spectrum of the signal at different frequency components, x(k) represents the time-domain signal at time point k, j represents the imaginary unit, and e represents the logarithm of the natural number base.

[0051] The spectrum of the signal is obtained through Fourier transform, and the spectral amplitude is calculated using the following expression:

[0052]

[0053] In the formula, Let X(f) represent the real part of the spectrum. Let X(f) represent the imaginary part of the spectrum.

[0054] The dominant frequency is calculated from the peak frequency of the spectrum. The calculation expression is as follows:

[0055] f0 = argmax|X(f)|;

[0056] In the formula, f0 represents the dominant frequency, and argmax|X(f)| represents the maximum frequency of the given spectrum;

[0057] The spectral ratio is calculated using the following expression:

[0058]

[0059] In the formula, T represents the spectral ratio, f low f represents the frequency range of the low-frequency region. high f represents the frequency range of the high-frequency region. max Indicates the maximum frequency of the signal;

[0060] Determine whether the spectrum ratio is greater than or equal to the second preset threshold. If yes, record it as accidental interference; otherwise, record it as a system fault.

[0061] As a further aspect of the present invention: the detailed analysis of the signal data marked as qualified signals, the comparison with expected performance indicators, the evaluation of the system's performance under different electromagnetic interference intensities, and the calculation of the system's identification accuracy coefficient specifically include:

[0062] Obtain the dataset that has been marked as a qualified signal;

[0063] A qualified signal is a signal that has passed the initial screening and indicates that the system is not affected or misjudged under the current electromagnetic interference.

[0064] For each signal, record the waveforms of the current and voltage, along with the corresponding timestamps;

[0065] Extracting time-domain features from qualified signals includes: peak value, mean value, and root mean square value;

[0066] Obtain the dominant frequency of the qualified signal and calculate the spectral density. The expression for calculating the spectral density is as follows:

[0067]

[0068] In the formula, S(f) represents the spectral density, f represents the frequency, and X(f) represents the spectrum of the signal at different frequency components;

[0069] The ratio of spectral density to the dominant frequency is used to calculate the system's recognition accuracy coefficient.

[0070] As a further aspect of the present invention: accurately distinguishing between real fault signals and electromagnetic interference signals specifically includes: determining whether the system identification accuracy coefficient is greater than or equal to a third preset threshold. If the system identification accuracy coefficient is greater than or equal to the third preset threshold, it indicates that the corresponding system accurately distinguishes between real fault signals and electromagnetic interference signals, and can accurately determine fault signals based on qualified signals.

[0071] If the system identification accuracy coefficient is less than the third preset threshold, it indicates that the corresponding system is not accurate in distinguishing between real fault signals and electromagnetic interference signals, and cannot accurately judge fault signals based on qualified signals.

[0072] The beneficial effects of this invention are:

[0073] (1) This invention innovatively employs a test method that gradually increases the intensity of electromagnetic interference (EMI) and combines it with high-precision sensors to monitor changes in current and voltage signals in real time, comprehensively capturing the impact of EMI on system signals. It utilizes a Kalman filter to recursively estimate the dynamic impact value of EMI and accurately models and tracks the impact of EMI by fusing state prediction and actual measurement data. Simultaneously, the method for calculating the deviation between the EMI impact value and electromagnetic shielding efficiency proposed in this invention can dynamically evaluate the impact of EMI on system performance at different intensities. Furthermore, it optimizes the system's response strategy by standardizing the calculation of the EMI coefficient, ensuring good fault identification capabilities even in high-intensity interference environments and significantly reducing false alarms and missed alarms caused by accidental interference. Based on this, this invention achieves accurate differentiation between fault signals and EMI signals, and significantly improves the accuracy and robustness of system fault detection through multi-level data analysis and dynamic optimization. Especially in complex electromagnetic environments, this system can adapt to changes in various interference types and intensities, providing a strong guarantee for the stability and reliability of the power system, demonstrating the technological advancement and broad application prospects of this invention in intelligent response and real-time processing capabilities.

[0074] (2) In practical applications, electromagnetic interference and system fault signals often have similar time-domain characteristics. Therefore, effectively distinguishing between the two is crucial for the stable operation of power systems. This invention performs detailed frequency domain analysis on the signal using Fourier transform and spectral analysis, and determines the interference nature of the signal based on the spectral ratio, thereby accurately distinguishing between random interference and system faults. By calculating the spectral amplitude, dominant frequency, and spectral ratio, the system can identify the different characteristics of low-frequency fault signals and high-frequency interference signals at the frequency domain level, providing a more accurate basis for subsequent fault diagnosis and processing. In addition, the system classifies and optimizes signal features using machine learning algorithms, and combined with actual measurement data, further enhances the ability to identify complex fault modes, ensuring efficient operation of the system under different electromagnetic interference intensities. This method not only improves the accuracy of identification but also enables a rapid response when random interference occurs, reducing the risk of misjudgment caused by random interference. Attached Figure Description

[0075] The invention will now be further described with reference to the accompanying drawings.

[0076] Figure 1 This is a flowchart illustrating the specific steps of the transmission line anomaly identification model construction method combining edge computing according to the present invention.

[0077] Figure 2 This is a flowchart illustrating the calculation process of the electromagnetic interference coefficient in this invention. Detailed Implementation

[0078] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0079] Please see Figure 1 As shown, this invention provides a method for constructing a transmission line anomaly identification model combining edge computing, comprising the following steps:

[0080] S1: During the anomaly identification process of transmission lines, based on the types of anomalies detected in the transmission lines, including short circuit, overload, and grounding fault, the electromagnetic frequency range and electromagnetic intensity level are set to simulate actual electromagnetic interference conditions.

[0081] S2: Real-time acquisition of current and voltage signals of transmission lines; within a set electromagnetic frequency range, gradually increasing the electromagnetic interference intensity; recording the electric field strength and current and voltage changes at each electromagnetic frequency; and using an electromagnetic field detector to monitor the absorption, reflection, and transmission characteristics of the signal, and recording the impact of electromagnetic interference on the signal.

[0082] The signals include: current signals and voltage signals;

[0083] S3: The acquired signals are classified in real time and marked as qualified signals, abnormal signals, and signals to be determined. Signals to be determined are marked and saved for subsequent analysis to determine whether they are false signals or real fault signals caused by electromagnetic interference, thus ensuring the accuracy of subsequent signal analysis.

[0084] S4: Initiate system diagnostics, and analyze whether the signal is due to accidental interference or system failure, based on the electromagnetic interference source, signal characteristics, and system operation status. If it is an accidental anomaly, repeat the test. If it is a systemic problem, stop the test and troubleshoot the electromagnetic shielding equipment or system to repair hidden faults and ensure the effectiveness of subsequent tests.

[0085] S5: Perform detailed analysis on the test data marked as qualified signals and compare them with the expected performance indicators to evaluate the system's performance under different electromagnetic interference intensities. Optimize the identification process to improve the system's fault identification capability under high-frequency current and voltage fluctuation conditions, ensuring accurate differentiation between real fault signals and electromagnetic interference signals, and reducing the possibility of false alarms and missed alarms.

[0086] In S1, during the transmission line anomaly identification process, based on the detected anomaly types (short circuit, overload, ground fault), the monitored electromagnetic frequency range and electromagnetic intensity level are set to simulate actual electromagnetic interference conditions. Specifically, this includes:

[0087] In the process of identifying anomalies in transmission lines, the electromagnetic frequency range and electromagnetic intensity level are first set according to the type of anomaly to be detected (such as short circuit, overload, ground fault).

[0088] The aforementioned anomaly types can cause current and voltage fluctuations in different frequency bands, therefore it is necessary to set an appropriate frequency range for each type of fault;

[0089] For example: the low-frequency band is used to simulate short circuits and ground faults, between 50 Hz and 500 Hz; while the high-frequency band is used to simulate electromagnetic interference, above 2 kHz;

[0090] Based on the current and voltage fluctuation frequencies of these fault types, select the frequency range and electromagnetic intensity level to ensure that different types of abnormal signals can be accurately identified and electromagnetic interference can be distinguished.

[0091] Next, after setting the electromagnetic frequency range, the electromagnetic interference intensity was gradually increased for testing; during this process, high-precision sensors were used to monitor the current and voltage signals in real time, and the changes in electric field intensity at each frequency were recorded.

[0092] Please participate Figure 2As shown, in S2, the electric field strength and current and voltage change data at each electromagnetic frequency are acquired. After acquisition, the impact value caused by electromagnetic interference is calculated by analyzing the electric field strength and current and voltage change data, and compared with the expected electromagnetic shielding efficiency to calculate the expected deviation of electromagnetic shielding efficiency.

[0093] The process for obtaining the expected deviation of the electromagnetic shielding efficiency is as follows:

[0094] Linear equations are used to represent dynamic systems of electromagnetic interference;

[0095] The computational expression for the linear equation is as follows:

[0096] x k =Ax k-1 +Bu k +w k ;

[0097] In the formula, k represents the data collection time point, and x k This represents the changes in electric field strength, current, and voltage caused by electromagnetic interference at time point k. A represents the system state transition matrix, and u... k B represents the external control input, and W represents the input matrix. k Let x represent process noise with covariance Q. k-1 This indicates the effect of electromagnetic interference at time point k-1 on the changes in electric field strength, current, and voltage.

[0098] The current and voltage signals measured by the sensor are represented by measurement equations;

[0099] The calculation expression for the measurement equation is as follows:

[0100] z k =Hx k +v k ;

[0101] In the formula, z k This represents the changes in electric field intensity, current, and voltage measured by the sensor at time point k, where H represents the measurement matrix, and v k This represents process noise with covariance R;

[0102] At each new sample, based on state x k-1 and input u k Predict the state at the current time point k, and denote it as the predicted state;

[0103] The calculation expression for the predicted state is as follows:

[0104]

[0105] In the formula, It is a predicted state;

[0106] The error of the predicted state is estimated to obtain the predicted covariance matrix, and the calculation expression is as follows:

[0107]

[0108] In the formula, Let represent the predicted state covariance matrix, and Q represent the process noise covariance matrix.

[0109] Combined with actual measured value z k To update the state estimate, calculate the Kalman gain, expressed as:

[0110]

[0111] In the formula, K k H represents the Kalman gain between the measured data and the predicted data, H represents the measurement matrix, and R represents the measurement noise covariance matrix.

[0112] Calculate the updated state estimate, and compute the expression:

[0113]

[0114] In the formula, This represents the updated state estimate;

[0115] Based on the updated state estimate from the Kalman filter, the impact of electromagnetic interference on the transmission line signal is calculated. The expression for calculating the impact of electromagnetic interference is as follows:

[0116]

[0117] In the formula, I k This represents the impact value of electromagnetic interference at time point k;

[0118] The deviation value obtained by calculating the difference between the actual electromagnetic interference effect and the expected electromagnetic shielding efficiency, and taking the absolute value, is denoted as the expected deviation of electromagnetic shielding efficiency F. k ;

[0119] Wherein, the expected electromagnetic shielding efficiency is a system-determined value;

[0120] It should be noted that processing the impact of electromagnetic interference using the Kalman filter algorithm can effectively estimate the deviation of electromagnetic shielding efficiency and provide an accurate basis for subsequent fault detection and signal processing. By recursively estimating the system state and combining actual measurements with predictions, the Kalman filter accurately captures the impact of electromagnetic interference on current and voltage, thereby ensuring the stability and accuracy of the power system in the electromagnetic environment.

[0121] If the anomaly is occasional, repeat the test; if it is a systemic problem, stop the test and troubleshoot the electromagnetic shielding equipment or system to fix the hidden fault and ensure the effectiveness of subsequent tests.

[0122] Obtain the electromagnetic interference impact value and the expected deviation of electromagnetic shielding efficiency at time point k, standardize the electromagnetic interference impact value and the expected deviation of electromagnetic shielding efficiency, and calculate the electromagnetic interference coefficient.

[0123] The calculation expression for the electromagnetic interference coefficient is as follows:

[0124]

[0125] In the formula, G k I represents the electromagnetic interference coefficient. k F represents the impact value of electromagnetic interference at time point k. k This represents the expected deviation of the electromagnetic shielding efficiency at time point k, where a1 and a2 are preset proportional coefficients.

[0126] It should be noted that the electromagnetic interference coefficient reflects the degree of interference of electromagnetic interference on the identification of transmission line faults. The larger the value of the electromagnetic interference coefficient, the higher the degree of electromagnetic interference. Random interference refers to the occasional, non-continuous electromagnetic interference signal in the power system or communication system. It is usually caused by external factors, equipment failure or instantaneous events, and has randomness, transience and irregularity.

[0127] In S3, the real-time classification of the acquired signals, labeling them as normal signals, abnormal signals, and signals to be determined, specifically includes:

[0128] The electromagnetic interference coefficient is compared with a first preset threshold to determine whether the electromagnetic interference coefficient is greater than or equal to the first preset threshold.

[0129] If so, it means that electromagnetic interference affects the signal so that it cannot be judged as normal or abnormal, and the signal is marked as a signal to be determined.

[0130] If not, it means that electromagnetic interference will not affect the normal or abnormal judgment of the signal. Then, the collected signal will be identified as normal or abnormal.

[0131] In S4, the analysis of whether the signal is due to accidental interference or a system malfunction specifically includes:

[0132] In the process of identifying anomalies in power transmission lines, Fourier transform and spectral analysis are used to distinguish signals;

[0133] Converting the time-domain signal to the frequency-domain signal, the Fourier transform calculation expression for each time point k is as follows:

[0134]

[0135] In the formula, f represents frequency, X(f) represents the spectrum of the signal at different frequency components, x(k) represents the time-domain signal at time point k, j represents the imaginary unit, and e represents the logarithm of the natural number base.

[0136] The spectrum of the signal is obtained through Fourier transform, and the spectral amplitude is calculated using the following expression:

[0137]

[0138] In the formula, Let X(f) represent the real part of the spectrum. Let x(f) represent the imaginary part of the spectrum.

[0139] The dominant frequency is calculated from the peak frequency of the spectrum. The calculation expression is as follows:

[0140] f0 = argmax|X(f)|;

[0141] In the formula, f0 represents the dominant frequency, and argmax|X(f)| represents the maximum frequency of the given spectrum;

[0142] The spectral ratio is calculated using the following expression:

[0143]

[0144] In the formula, T represents the spectral ratio, f low f represents the frequency range of the low-frequency region. high f represents the frequency range of the high-frequency region. max Indicates the maximum frequency of the signal;

[0145] Determine whether the spectrum ratio is greater than or equal to the second preset threshold. If yes, record it as accidental interference; otherwise, record it as a system fault.

[0146] It should be noted that by calculating the spectrum ratio, random interference and system faults can be accurately identified, and the larger the spectrum ratio value, the more likely it is to be random interference.

[0147] In S5, the detailed analysis of the signal data marked as qualified signals and the comparison with expected performance indicators are performed to evaluate the system's performance under different electromagnetic interference intensities and calculate the system's identification accuracy coefficient. Specifically, this includes:

[0148] Obtain the dataset that has been marked as a qualified signal;

[0149] A qualified signal is a signal that has passed the initial screening and indicates that the system is not affected or misjudged under the current electromagnetic interference.

[0150] For each signal, record the waveforms of the current and voltage, along with the corresponding timestamps;

[0151] Extracting time-domain features from qualified signals includes: peak value, mean value, and root mean square value;

[0152] Obtain the dominant frequency of the qualified signal and calculate the spectral density. The expression for calculating the spectral density is as follows:

[0153]

[0154] In the formula, S(f) represents the spectral density, f represents the frequency, and X(f) represents the spectrum of the signal at different frequency components;

[0155] The ratio of spectral density to the dominant frequency is used to calculate the system's recognition accuracy coefficient.

[0156] The accurate distinction between real fault signals and electromagnetic interference signals specifically includes:

[0157] Determine whether the system identification accuracy coefficient is greater than or equal to the third preset threshold. If the system identification accuracy coefficient is greater than or equal to the third preset threshold, it indicates that the corresponding system can accurately distinguish between real fault signals and electromagnetic interference signals, and can accurately determine fault signals based on qualified signals.

[0158] If the system identification accuracy coefficient is less than the third preset threshold, it indicates that the corresponding system is not accurate in distinguishing between real fault signals and electromagnetic interference signals, and cannot accurately judge fault signals based on qualified signals.

[0159] The working principle of this invention is as follows: Through multi-level analysis and optimization steps, the fault identification capability under electromagnetic interference (EMI) environments is improved. First, during the transmission line anomaly identification process, appropriate electromagnetic frequency ranges and electromagnetic intensity levels are set according to different anomaly types, including short circuits, overloads, and grounding faults, to simulate actual EMI conditions. Next, by real-time acquisition of current and voltage signals and gradually increasing the EMI intensity, the electric field strength and current / voltage changes at each frequency are recorded. The absorption, reflection, and transmission characteristics of the signals are monitored using an electromagnetic field detector to calculate the impact of EMI on the signal. Then, the impact of EMI is recursively estimated using a Kalman filter algorithm, and compared with the expected electromagnetic shielding efficiency to calculate the expected deviation from the shielding efficiency. The EMI coefficient is then calculated through standardization. During the classification process, the acquired signals are classified in real-time as qualified signals, abnormal signals, or signals to be determined. Based on the comparison of the EMI coefficient with a preset threshold, it is determined whether the signal is affected by interference and further analysis is performed. To determine whether a signal belongs to accidental interference or a system fault, frequency features are extracted using Fourier transform and spectral analysis, and the spectral ratio is calculated to accurately distinguish between accidental interference and fault signals. Finally, the system's performance under different electromagnetic interference intensities is evaluated by comparing the labeled qualified signals with expected performance indicators, and the fault identification algorithm is optimized. By calculating the system's identification accuracy coefficient, it is ensured that the system can effectively distinguish between real fault signals and electromagnetic interference signals, reducing false alarms and false negatives, and improving the accuracy of fault identification and the robustness of the system. This method, through comprehensive analysis, algorithm optimization, and real-time monitoring, enhances the fault detection capability of transmission lines in complex electromagnetic environments, providing an important guarantee for the stability and safety of power systems.

[0160] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0161] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as 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, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0162] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0163] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply 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 this application.

[0164] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1.A method for constructing a power transmission line anomaly identification model combined with edge computing, characterized in that, Comprise the following steps: S1: In the process of power line anomaly identification, according to the type of power line detection anomaly including: short circuit, overload, ground fault, set the monitoring electromagnetic frequency range and electromagnetic intensity level, to simulate the actual electromagnetic interference situation; S2: Real-time acquisition of power line current signal and voltage signal, in the set electromagnetic frequency range, gradually increase the electromagnetic interference intensity, record the electric field intensity and current voltage change data under each electromagnetic frequency, and use electromagnetic field detector to monitor the absorption, reflection and transmission characteristics of the signal, and record the influence of electromagnetic interference on the signal; Wherein, the signal includes: current signal and voltage signal; S3: Real-time classification of the collected signal, marked as normal signal, abnormal signal and to be determined signal, and mark the to be determined signal for subsequent analysis whether it is a false signal or a real fault signal caused by electromagnetic interference, to ensure the accuracy of subsequent signal analysis; S4: Start system diagnosis, combined with electromagnetic interference source, signal characteristics and system operation, analyze whether the signal belongs to accidental interference or due to system failure, if it is accidental anomaly, repeat the test; If it is a systematic problem, stop testing and troubleshoot the electromagnetic shielding equipment or system to repair the hidden fault and ensure the effectiveness of subsequent testing; S5: Detailed analysis of the signal data marked as qualified signal, and compared with the expected performance index, evaluate the performance of the system under different electromagnetic interference intensity, and optimize the identification process, improve the fault identification ability of the system under high frequency current and voltage fluctuation conditions, and ensure that the real fault signal and electromagnetic interference signal can be accurately distinguished; In S5, the detailed analysis of the signal data marked as qualified signal and the comparison with the expected performance index to evaluate the performance of the system under different electromagnetic interference intensity, specifically includes: Obtain the data set marked as qualified signal; The qualified signal is the signal after preliminary screening, which indicates that the system is not disturbed or misjudged under the current electromagnetic interference; For each signal, record the waveform of current and voltage, and the corresponding timestamp; Extract the time domain features from the qualified signal, including peak value, mean value and root mean square value; Obtain the main frequency of the qualified signal and calculate the spectral density, the calculation expression of the spectral density is: In the formula, S(f) represents the spectral density, f represents the frequency, and X(f) represents the frequency spectrum of the signal at different frequency components; Calculate the ratio of the spectral density and the main frequency to obtain the system identification accuracy coefficient; The accurate distinction between real fault signal and electromagnetic interference signal specifically includes: judging whether the system identification accuracy coefficient is greater than or equal to the third preset threshold value, if the system identification accuracy coefficient is greater than or equal to the third preset threshold value, it means that the corresponding system can accurately distinguish the real fault signal and the electromagnetic interference signal, and can accurately judge the fault signal according to the qualified signal; If the system identification accuracy coefficient is less than the third preset threshold value, it means that the corresponding system cannot accurately distinguish the real fault signal and the electromagnetic interference signal, and cannot accurately judge the fault signal according to the qualified signal. 2.The method of claim 1, wherein, The electric field intensity and current voltage change data at each electromagnetic frequency are acquired, and after acquisition, the influence value of electromagnetic interference is calculated by analyzing the electric field intensity and current voltage change data, and compared with the expected electromagnetic shielding efficiency, so as to calculate the expected deviation of the electromagnetic shielding efficiency. 3.The method of claim 2, wherein, The acquisition process of the expected deviation of the electromagnetic shielding efficiency is: A linear equation is used to represent the dynamic system of electromagnetic interference; The calculation expression of the linear equation is: x k = Ax k-1 + Bu k + w k ; where k denotes the time point of collection, x k denotes the change in the electromagnetic interference-affected electric field intensity, current, and voltage at the time point k, A denotes a system state transition matrix, u k denotes an external control input, B denotes an input matrix, w k denotes process noise with a covariance of Q, x k-1 denotes the change in the electromagnetic interference-affected electric field intensity, current, and voltage at the time point k-1; A measurement equation is used to represent the current and voltage signals measured by the sensor; The calculation expression of the measurement equation is: z k = Hx k + v k ; where z k represents the change data of the electric field intensity, current and voltage measured by the sensor at the time point k, H represents the measurement matrix, v k represents the process noise with the covariance R; At each new sample, the state x k-1 and input u k is predicted for the current time point k, denoted as predicted state The calculation expression of the predicted state is: In the formula, is the predicted state; The error of the predicted state is estimated to obtain a predicted covariance matrix, and the calculation expression is: wherein denotes the predicted state covariance matrix, Q denotes the process noise covariance matrix; combining the actual measured values z k to update the state estimate, the Kalman gain is calculated, and the expression is calculated as In the formula, K k indicates the Kalman gain of the measurement data and the prediction data, H indicates the measurement matrix, and R indicates the measurement noise covariance matrix. The updated state estimation is calculated, and the calculation expression is: In the formula, denotes the updated state estimate; According to the updated state estimation of the Kalman filter, the influence value of electromagnetic interference on the power line signal is calculated, and the calculation expression of the influence value of electromagnetic interference is: In the formula, I k represents the influence value of electromagnetic interference at time point k; By calculating the difference between the actual electromagnetic interference effect and the expected electromagnetic shielding efficiency, and taking the absolute value, a deviation value is obtained, denoted as electromagnetic shielding efficiency expected deviation F k ; The expected electromagnetic shielding efficiency is a system determined value. 4.The method of claim 3, wherein, The influence value of electromagnetic interference and the expected deviation of the electromagnetic shielding efficiency at time point k are acquired, and the influence value of electromagnetic interference and the expected deviation of the electromagnetic shielding efficiency are standardized to calculate the electromagnetic interference coefficient. 5.The method of claim 1, wherein, The collected signals are classified in real time and marked as normal signals, abnormal signals and signals to be determined, specifically including: The electromagnetic interference coefficient is compared with the first preset threshold value to determine whether the electromagnetic interference coefficient is greater than or equal to the first preset threshold value; If yes, it means that the electromagnetic interference affected signal cannot be normally or abnormally judged, and the signal is marked as a signal to be determined; If not, it means that the electromagnetic interference will not affect the signal to be normally or abnormally judged, and the collected signal is abnormally identified to determine whether the current signal is a normal signal or an abnormal signal. 6.The method of claim 1, wherein, The signal is analyzed to determine whether it belongs to accidental interference or is caused by system failure, specifically including: In the process of power line abnormal identification, the signals are distinguished by Fourier transform and spectrum analysis; The time domain signal is converted into a frequency domain signal, and for each time point k, the calculation expression of the Fourier transform is: In the formula, f represents the frequency, X(f) represents the frequency spectrum of the signal at different frequency components, x(k) represents the time domain signal at time point k, j represents the imaginary unit, and e represents the natural number base logarithm; The frequency spectrum of the signal is obtained by Fourier transform, the spectrum amplitude is calculated, and the calculation expression is: wherein denotes the real part of the spectrum X(f), denotes the imaginary part of the spectrum X(f); The main frequency is calculated by the peak frequency of the spectrum, and the calculation expression is: f0=argmax|X(f)|; In the formula, f0 represents the main frequency, and argmax|X(f)| represents the maximum frequency of the spectrum; The spectrum ratio is calculated, and the calculation expression is: where T represents the spectral ratio, f low represents the frequency range of the low frequency region, f high represents the frequency range of the high frequency region, f max represents the maximum frequency of the signal; It is judged whether the spectrum ratio is greater than or equal to the second preset threshold value, if yes, it is recorded as accidental interference, and if not, it is recorded as system failure.

Images