Traveling wave cable measurement sensing method and system based on time-frequency domain combined reflection principle

Abstract

The invention provides a traveling wave cable measurement sensing method and system based on a time-frequency domain combined reflection principle, and relates to the technical field of measurement, and the method comprises the steps: arranging a high-frequency signal transmitting device and a signal receiving device at one end of a traveling wave cable; establishing a time domain reflection function and a frequency domain reflection function of the traveling wave cable; constructing a time-frequency domain joint reflection model; collecting a reflected signal of the traveling wave cable through the signal receiving device to obtain time domain data and frequency domain data of the reflected signal; carrying out conjoint analysis on the time domain data and the frequency domain data by utilizing a time-frequency domain joint reflection model, and extracting characteristic parameters of a reflected signal; calculating impedance change and signal propagation characteristics along the traveling wave cable based on the characteristic parameters in combination with a time-frequency domain joint reflection model; through analysis of impedance change and signal propagation characteristics, intrusion detection, leakage positioning and fault diagnosis along the traveling wave cable are realized.

Description

Technical Field

[0001] The present invention relates to measurement technologies, and particularly to a traveling wave cable measurement and sensing method and system based on the joint reflection principle in the time-frequency domain. Background Art

[0002] As an important power system device, traveling wave cables play a key role in power transmission, distribution networks, and power supply in special occasions. With the rapid development of smart grids, higher requirements are put forward for the condition monitoring and fault diagnosis of traveling wave cables. Traditional traveling wave cable measurement methods mainly include the time domain reflectometry (TDR) and the frequency domain reflectometry (FDR).

[0003] The time domain reflectometry (TDR) is a method of determining cable characteristics and fault locations by sending short pulse signals to the cable and then analyzing the time delay and amplitude change of the reflected signals. This method is simple to operate and can quickly locate the fault point, but its spatial resolution is limited by the bandwidth of the incident pulse, making it difficult to detect minor or gradual faults.

[0004] The frequency domain reflectometry (FDR) is a method of obtaining cable characteristics by sending swept-frequency signals to the cable and analyzing the reflection coefficients at different frequencies. The FDR method has a high frequency resolution and can provide more detailed cable impedance information, but its spatial resolution is generally not as good as the TDR method, and the measurement time is longer.

[0005] However, the existing traveling wave cable measurement technologies have the following main defects:

[0006] Separation of time domain and frequency domain information: Traditional TDR and FDR methods only focus on time domain or frequency domain information respectively, and cannot make full use of the comprehensive information contained in the cable reflection signals. This information separation leads to limitations in measurement accuracy and fault diagnosis capabilities, especially in the case of complex or multiple faults.

[0007] Insufficient ability to capture dynamic characteristics: Existing technologies mainly measure the static characteristics of cables and are difficult to effectively capture the dynamic changes of cables during actual operation. This limitation makes the system unable to detect and respond in a timely manner to instantaneous changes in cable states, such as partial discharges, external interferences, or early fault signs.

[0008] Poor anti-interference ability and environmental adaptability: Traditional measurement methods are easily affected by external interferences in complex electromagnetic environments, resulting in unstable measurement results or misjudgments. At the same time, these methods are not sensitive enough to changes in the cable surrounding environment (such as temperature, humidity, etc.) and are difficult to maintain high-precision measurement performance under various actual working conditions.

[0009] In view of the above deficiencies, there is an urgent need to develop a traveling wave cable measurement and sensing technology that can comprehensively utilize time-domain and frequency-domain information, has the ability to capture dynamic characteristics, and at the same time has good anti-interference and environmental adaptability. The purpose of the present invention is to break through the limitations of the existing technology through the principle of joint time-frequency domain reflection, and provide a more comprehensive, accurate and robust traveling wave cable measurement and sensing method. Summary of the Invention

[0010] The purpose of the present invention is to solve the problems existing in the prior art, such as the separation of time-domain and frequency-domain information, insufficient ability to capture dynamic characteristics, and poor anti-interference ability and environmental adaptability.

[0011] To solve the above problems, the present invention provides a traveling wave cable measurement and sensing method based on the principle of joint time-frequency domain reflection, including:

[0012] A high-frequency signal transmitting device and a signal receiving device are arranged at one end of the traveling wave cable. The high-frequency signal transmitting device transmits a swept-frequency signal within a preset frequency range to the traveling wave cable, and the signal receiving device receives the reflected signal transmitted by the traveling wave cable; using the time-domain reflection principle and the frequency-domain reflection principle, establish the time-domain reflection function and the frequency-domain reflection function of the traveling wave cable; combine the time-domain reflection function and the frequency-domain reflection function to construct a time-frequency domain joint reflection model, which includes the geometric parameters, electrical parameters and environmental parameters of the traveling wave cable.

[0013] Collect the reflected signal of the traveling wave cable through the signal receiving device, perform noise reduction processing and digital conversion on the collected reflected signal to obtain the time-domain data and frequency-domain data of the reflected signal; use the time-frequency domain joint reflection model to jointly analyze the time-domain data and frequency-domain data, and extract the characteristic parameters of the reflected signal, including signal amplitude, phase, time delay and frequency response.

[0014] Based on the characteristic parameters, combined with the time-frequency domain joint reflection model, calculate the impedance change and signal propagation characteristics along the traveling wave cable; through the analysis of the impedance change and signal propagation characteristics, realize intrusion detection, leakage location and fault diagnosis along the traveling wave cable; compare the detection result with a preset threshold, and when the detection result exceeds the preset threshold, trigger an alarm mechanism and output corresponding measurement and sensing information, including the intrusion location, leakage point location or fault point location.

[0015] In a preferred manner,

[0016] Using the time-domain reflection principle and the frequency-domain reflection principle, establish the time-domain reflection function and the frequency-domain reflection function of the traveling wave cable; combining the time-domain reflection function and the frequency-domain reflection function to construct a time-frequency domain joint reflection model includes:

[0017] Using the telegraph equation in transmission line theory, the time-domain reflection function of the traveling wave cable is solved. The time-domain reflection function includes an attenuation coefficient, cable position, signal propagation speed, multiple reflection coefficient, and Dirac function, and is used to describe the reflection characteristics of the traveling wave cable in the time domain;

[0018] The Fourier transform is performed on the time-domain reflection function to obtain the frequency-domain reflection function. The frequency-domain reflection function includes a propagation constant, attenuation constant, phase constant, and angular frequency, and is used to describe the reflection characteristics of the traveling wave cable in the frequency domain; A frequency-dependent complex characteristic impedance is introduced. The complex characteristic impedance consists of resistance, inductance, capacitance, and shunt conductance per unit length, and is used to improve the accuracy of the frequency-domain reflection function;

[0019] The short-time Fourier transform method is used to convert the time-domain reflection function into a time-frequency joint representation, which includes a time offset parameter; The time-frequency joint representation is combined with the frequency-domain reflection function to obtain a time-frequency domain joint reflection model. This model synthesizes time-domain and frequency-domain information and can more comprehensively describe the reflection characteristics of the traveling wave cable;

[0020] The selected wavelet basis function is used to perform wavelet transform on the time-domain reflection function to obtain wavelet coefficients including scale parameters and translation parameters; The wavelet coefficients are combined with the time-frequency domain joint reflection model to obtain an optimized time-frequency domain joint reflection model.

[0021] In the preferred manner,

[0022] Using the time-frequency domain joint reflection model, the time-domain data and frequency-domain data are jointly analyzed, and the characteristic parameters of the reflection signal are extracted, including:

[0023] The collected time-domain data is subjected to wavelet threshold denoising, baseline correction, and normalization processing; The frequency-domain data is applied with a Hanning window function, fast Fourier transform is performed, and the moving average method is used for spectrum smoothing;

[0024] Using the short-time Fourier transform method, a time-frequency spectrogram is generated for the preprocessed time-domain data; The time-frequency spectrogram is matched with the pre-constructed time-frequency domain joint reflection model, and a similarity function is calculated; Based on the similarity function, a threshold is set to determine potential abnormal points;

[0025] For the detected potential abnormal points, the Morlet wavelet basis function is selected for continuous wavelet transform, and the wavelet transform coefficients are calculated; Based on the wavelet transform coefficients, a wavelet scale map is constructed to analyze the signal characteristics at different scales;

[0026] According to the time-frequency spectrogram and the multi-resolution analysis results, the reflection peak position, reflection peak amplitude, phase information, time delay estimation, frequency response, and attenuation coefficient are extracted; The bispectrum analysis method is used to extract the non-linear characteristics of the signal;

[0027] Perform multi-scale fusion on the characteristic parameters to improve parameter stability; utilize the complementarity of time-domain and frequency-domain information to perform time-frequency joint correction on the characteristic parameters; calculate the statistics of the characteristic parameters, including mean, variance, skewness, and kurtosis.

[0028] In a preferred manner,

[0029] Based on the characteristic parameters, in combination with the time-frequency domain joint reflection model, calculate the impedance change and signal propagation characteristics along the traveling-wave cable, including:

[0030] Utilize the time-frequency domain joint reflection model to extract the reflection coefficient at different positions and frequencies, where the reflection coefficient is calculated by the ratio of the frequency-domain reflection function at that position to the spectrum of the incident signal; based on the reflection coefficient, calculate the characteristic impedance at each position of the cable, where the characteristic impedance is jointly determined by the nominal characteristic impedance of the cable and the reflection coefficient; calculate the impedance change rate, where the impedance change rate is the ratio of the difference between the characteristic impedance and the nominal characteristic impedance at each position to the nominal characteristic impedance; map the calculated impedance change rate to the time-frequency plane to generate a time-frequency joint impedance map;

[0031] Utilize the time-frequency domain joint reflection model to estimate the propagation constant of the cable, where the propagation constant includes the attenuation constant and the phase constant, and is calculated from the frequency-domain reflection functions at different positions; based on the phase constant, calculate the group velocity of the signal, where the group velocity is the derivative of the angular frequency with respect to the phase constant; utilize the attenuation constant to calculate the transmission loss of the cable, where the transmission loss varies with distance and frequency; based on the group velocity, calculate the delay spread of the signal during propagation, where the delay spread is determined by the product of the propagation distance and the difference between the reciprocals of the minimum and maximum group velocities.

[0032] In a preferred manner,

[0033] The method further includes:

[0034] Based on the time-frequency joint impedance map, set a threshold to mark potential impedance anomaly points; analyze the change trends of the propagation constant, group velocity, transmission loss, and delay spread to identify abnormal regions; perform spatial correspondence and correlation analysis on the impedance anomaly points and the propagation characteristic abnormal regions; based on the above analysis results, in combination with the impedance change amplitude, the increase in transmission loss, and the degree of delay spread, establish a comprehensive scoring mechanism to quantify the anomaly degree of each section of the cable; compare the measurement results at different times, analyze the change trend of the cable state, and provide a basis for predictive maintenance.

[0035] In a preferred manner,

[0036] Through the analysis of impedance change and signal propagation characteristics, achieve intrusion detection, leakage location, and fault diagnosis along the traveling-wave cable, including:

[0037] Using a joint time-frequency domain reflection model, reflection coefficients are extracted at different positions and frequencies, where the reflection coefficient is calculated as the ratio of the frequency domain reflection function at that position to the spectrum of the incident signal; based on the reflection coefficient, the characteristic impedance of the cable at each position is calculated, and the characteristic impedance is jointly determined by the nominal characteristic impedance of the cable and the reflection coefficient; the impedance change rate is calculated, and the impedance change rate is the ratio of the difference between the characteristic impedance and the nominal characteristic impedance at each position to the nominal characteristic impedance;

[0038] The calculated impedance change rate is mapped to the time-frequency plane to generate a time-frequency joint impedance map; time-frequency domain feature extraction is performed on the time-frequency joint impedance map, including calculating the first and second derivatives of the impedance change rate on the time axis, performing wavelet packet decomposition on the impedance change rate to extract the energy distribution characteristics of different frequency bands, and calculating the instantaneous frequency and group delay of the impedance change rate using the Wigner-Ville distribution;

[0039] A sliding window algorithm is designed for local anomaly detection in the spatial and temporal dimensions, including setting the spatial window size, calculating the statistical characteristics of the impedance change rate within the window, setting the temporal window size, analyzing the temporal evolution characteristics of the impedance change rate, and combining the analysis results of the spatial and temporal windows to calculate the comprehensive anomaly index;

[0040] A Gaussian mixture model is used to model the comprehensive anomaly index in the normal state, a dynamic detection threshold is calculated based on the Gaussian mixture model, and a particle filter algorithm is introduced to optimize the dynamic detection threshold in real time; when the comprehensive anomaly index is greater than the dynamic detection threshold, an anomaly alarm is triggered, and the continuously triggered alarm points are clustered in space-time to screen out potential anomaly events; using the joint time-frequency domain reflection model, the propagation constant of the cable is estimated, and the propagation constant includes the attenuation constant and the phase constant, which are calculated from the frequency domain reflection functions at different positions;

[0041] Based on the phase constant, the group velocity of the signal is calculated, and the group velocity is the derivative of the angular frequency with respect to the phase constant; using the attenuation constant, the transmission loss of the cable is calculated, and the transmission loss varies with distance and frequency; the spatial gradient of the attenuation constant is calculated to identify regions where the attenuation increases abnormally, and the frequency dependence of the group velocity is analyzed to detect abnormal changes in the dispersion characteristics;

[0042] Wavelet multi-resolution analysis is performed on the attenuation constant, group velocity, and transmission loss to extract features at different scales, calculate the correlation coefficient matrix between the features at different scales, identify cross-scale abnormal patterns, and reconstruct the abnormal feature map based on the significant scale; local maximum points are searched in the abnormal feature map as potential abnormal positions, and the spatial centroid of the abnormal feature map is calculated to provide a preliminary estimate of the abnormal position;

[0043] Fuse the impedance characteristics, propagation characteristics, and temporal characteristics extracted by the long short-term memory network to construct a high-dimensional feature space; use principal component analysis and the t-SNE algorithm to reduce the dimension of the high-dimensional feature space, and use the DBSCAN algorithm for density clustering in the reduced-dimensional space to identify potential abnormal patterns; train a support vector machine classifier and a random forest model to identify and classify different types of anomalies, and evaluate the importance of each feature in anomaly classification; establish a fuzzy rule base based on expert knowledge to evaluate the severity of anomalies, and use historical data to establish a regression model between anomaly features and severity;

[0044] Combine the anomaly severity and historical data to predict the remaining service life of the cable, consider the anomaly type, severity, and system importance to prioritize maintenance tasks, and use reinforcement learning algorithms to optimize long-term maintenance strategies.

[0045] A traveling wave cable measurement and sensing system based on the joint time-frequency domain reflection principle, including:

[0046] The first unit is used to set a high-frequency signal transmitting device and a signal receiving device at one end of the traveling wave cable. The high-frequency signal transmitting device transmits a swept-frequency signal within a preset frequency range to the traveling wave cable, and the signal receiving device receives the reflected signal transmitted by the traveling wave cable; use the time-domain reflection principle and the frequency-domain reflection principle to establish the time-domain reflection function and frequency-domain reflection function of the traveling wave cable; combine the time-domain reflection function and the frequency-domain reflection function to construct a joint time-frequency domain reflection model, which includes the geometric parameters, electrical parameters, and environmental parameters of the traveling wave cable;

[0047] The second unit is used to collect the reflected signal of the traveling wave cable through the signal receiving device, perform noise reduction processing and digital conversion on the collected reflected signal to obtain the time-domain data and frequency-domain data of the reflected signal; use the joint time-frequency domain reflection model to perform joint analysis on the time-domain data and frequency-domain data, and extract the characteristic parameters of the reflected signal, including signal amplitude, phase, time delay, and frequency response;

[0048] The third unit is used to calculate the impedance change and signal propagation characteristics along the traveling wave cable based on the characteristic parameters and in combination with the joint time-frequency domain reflection model; through the analysis of the impedance change and signal propagation characteristics, realize intrusion detection, leakage location, and fault diagnosis along the traveling wave cable; compare the detection result with a preset threshold, and when the detection result exceeds the preset threshold, trigger an alarm mechanism and output corresponding measurement and sensing information, including the intrusion location, leakage point location, or fault point location.

[0049] An electronic device, including:

[0050] A processor;

[0051] A memory for storing processor-executable instructions;

[0052] Wherein, the processor is configured to call the instructions stored in the memory to execute the method described above.

[0053] A computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method described above is implemented.

[0054] Advantages of the present invention: By combining the time-domain reflection principle and the frequency-domain reflection principle, the present invention constructs a joint time-frequency domain reflection model. This innovative method makes full use of the time-domain and frequency-domain information in the reflected signal, significantly improving the measurement accuracy and reliability. The joint time-frequency domain analysis can more comprehensively capture the characteristic changes of the traveling-wave cable, effectively reducing the information loss that may be caused by single-domain analysis. At the same time, by performing noise reduction processing and digital conversion on the collected reflected signal, the signal quality is further improved, laying a solid foundation for subsequent analysis. This comprehensive method enables the system to more accurately detect and locate various abnormal conditions, such as intrusion, leakage, and faults, greatly enhancing the accuracy and credibility of measurement perception.

[0055] The joint time-frequency domain reflection model of the present invention not only includes the geometric parameters and electrical parameters of the traveling-wave cable, but also considers the influence of environmental parameters. This comprehensive model design enables the system to better adapt to complex actual operating environments and effectively capture the dynamic changes of cable characteristics. By real-time analyzing impedance changes and signal propagation characteristics, the system can quickly respond to instantaneous changes in cable status, such as partial discharge, external interference, or early fault signs. This enhanced ability to capture dynamic characteristics and environmental adaptability enables the present invention to maintain high-efficiency measurement perception performance under various complex working conditions, greatly improving the practicality and reliability of the system. Description of the Drawings

[0056] Figure 1 It is a schematic flow chart of the traveling-wave cable measurement and perception method based on the joint time-frequency domain reflection principle in an embodiment of the present invention;

[0057] Figure 2 It is a schematic structural diagram of the traveling-wave cable measurement and perception system based on the joint time-frequency domain reflection principle in an embodiment of the present invention. Detailed Embodiments

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are only a part rather than all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0059] The technical solutions of the present invention will be described in detail below with specific embodiments. These specific embodiments may be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0060] Figure 1 It is a schematic flowchart of a traveling wave cable measurement and sensing method based on the time-frequency domain joint reflection principle in an embodiment of the present invention, as Figure 1 shown, the method includes:

[0061] S101. A high-frequency signal transmitting device and a signal receiving device are arranged at one end of the traveling wave cable. The high-frequency signal transmitting device transmits a swept-frequency signal within a preset frequency range to the traveling wave cable, and the signal receiving device receives the reflected signal transmitted by the traveling wave cable; using the time-domain reflection principle and the frequency-domain reflection principle, establish the time-domain reflection function and the frequency-domain reflection function of the traveling wave cable; combine the time-domain reflection function and the frequency-domain reflection function to construct a time-frequency domain joint reflection model, which includes the geometric parameters, electrical parameters, and environmental parameters of the traveling wave cable.

[0062] S102. The signal receiving device collects the reflected signal of the traveling wave cable, performs noise reduction processing and digital conversion on the collected reflected signal to obtain the time-domain data and frequency-domain data of the reflected signal; using the time-frequency domain joint reflection model, jointly analyze the time-domain data and frequency-domain data, and extract the characteristic parameters of the reflected signal, including signal amplitude, phase, time delay, and frequency response.

[0063] S103. Based on the characteristic parameters, combined with the time-frequency domain joint reflection model, calculate the impedance change and signal propagation characteristics along the traveling wave cable; through the analysis of the impedance change and signal propagation characteristics, realize intrusion detection, leakage location, and fault diagnosis along the traveling wave cable; compare the detection result with a preset threshold. When the detection result exceeds the preset threshold, trigger an alarm mechanism and output corresponding measurement and sensing information, including the intrusion location, leakage point location, or fault point location.

[0064] The specific implementation manner of the traveling wave cable measurement and sensing method based on the time-frequency domain joint reflection principle is as follows:

[0065] First, a high-frequency signal transmitting device and a signal receiving device are set at one end of the traveling-wave cable. The high-frequency signal transmitting device can adopt a vector network analyzer (VNA) or a spectrum analyzer, which can generate a swept-frequency signal in the range of 1 MHz - 3 GHz. The signal receiving device can adopt an oscilloscope or a data acquisition card, which is used to collect the reflected signal after being transmitted by the traveling-wave cable.

[0066] The high-frequency signal transmitting device transmits a swept-frequency signal within a preset frequency range to the traveling-wave cable, for example, a swept-frequency signal from 1 MHz to 1 GHz with a step of 1 MHz. The signal receiving device synchronously receives the reflected signal after being transmitted by the traveling-wave cable. Based on the received reflected signal, the time-domain reflection function and the frequency-domain reflection function of the traveling-wave cable are respectively established by using the time-domain reflection principle and the frequency-domain reflection principle.

[0067] The time-domain reflection function can represent the variation relationship of the reflected signal amplitude with time, and the frequency-domain reflection function can represent the variation relationship of the reflected signal amplitude and phase with frequency. The time-domain reflection function and the frequency-domain reflection function are combined to construct a time-frequency domain joint reflection model. This model includes the geometric parameters (such as cable length, cross-sectional area, etc.), electrical parameters (such as characteristic impedance, propagation constant, etc.) and environmental parameters (such as temperature, humidity, etc.) of the traveling-wave cable.

[0068] Next, the reflected signal of the traveling-wave cable is collected by the signal receiving device. The collected reflected signal is subjected to noise reduction processing, and methods such as wavelet transform or empirical mode decomposition can be used to remove noise. Then, analog-to-digital conversion is performed to convert the analog signal into a digital signal, and the time-domain data and frequency-domain data of the reflected signal are obtained. The time-domain data can be represented as a discrete voltage-time series, and the frequency-domain data can be represented as discrete amplitude-frequency and phase-frequency series.

[0069] Using the above-established time-frequency domain joint reflection model, the time-domain data and frequency-domain data are jointly analyzed to extract the characteristic parameters of the reflected signal. The characteristic parameters include signal amplitude, phase, time delay, and frequency response. For example, the peak amplitude and arrival time of the time-domain reflected signal, the amplitude-frequency characteristic and phase-frequency characteristic of the frequency-domain reflected signal, etc. can be extracted.

[0070] Based on the extracted characteristic parameters, combined with the time-frequency domain joint reflection model, the impedance change and signal propagation characteristics along the traveling-wave cable are calculated. The impedance change can be obtained by analyzing the amplitude and phase change of the reflected signal, and the signal propagation characteristics can be obtained by analyzing the time delay and frequency response of the signal. For example, if there is an impedance mutation at a certain place, a strong reflected signal will be generated there; if there is a leakage at a certain place, it will cause the signal attenuation to increase.

[0071] By analyzing impedance changes and signal propagation characteristics, intrusion detection, leakage location, and fault diagnosis along the traveling-wave cable can be achieved. Intrusion detection is mainly based on local impedance changes, leakage location is mainly based on signal attenuation characteristics, and fault diagnosis comprehensively considers impedance changes and propagation characteristics.

[0072] Finally, the detection results are compared with the preset thresholds. The preset thresholds can be set according to the actual application scenario. For example, the impedance change threshold for intrusion detection is 5%, and the signal attenuation threshold for leakage detection is 3 dB / m. When the detection results exceed the preset thresholds, the alarm mechanism is triggered and the corresponding measurement and perception information is output. The measurement and perception information includes the intrusion location, leakage point location, or fault point location, which can be obtained by calculating the time delay of the reflected signal and combining it with the cable propagation speed.

[0073] For example, assume a measurement is carried out on a 100-meter-long traveling-wave cable. First, a swept-frequency signal from 1 MHz to 1 GHz is transmitted into the cable, and the reflected signal is received and its characteristic parameters are extracted. It is analyzed that there is an obvious impedance mutation at 50 meters from the cable start end, and the amplitude change exceeds 10%, which is judged as an intrusion event. At the same time, the signal attenuation rate suddenly increases to 5 dB / m at 80 meters, which is judged as the leakage point. The system then triggers an alarm and outputs the measurement and perception information that the intrusion location is at 50 meters and the leakage point location is at 80 meters.

[0074] Through the above steps, a measurement and perception method for traveling-wave cables based on the joint reflection principle in the time-frequency domain is realized. This method combines time-domain and frequency-domain information, improves the accuracy and reliability of measurement, and can effectively achieve intrusion detection, leakage location, and fault diagnosis along the traveling-wave cable.

[0075] In an optional implementation, using the time-domain reflection principle and the frequency-domain reflection principle, the time-domain reflection function and the frequency-domain reflection function of the traveling-wave cable are established; combining the time-domain reflection function and the frequency-domain reflection function, a time-frequency domain joint reflection model is constructed, including:

[0076] Using the telegraph equation in the transmission line theory, the time-domain reflection function of the traveling-wave cable is solved. The time-domain reflection function includes the attenuation coefficient, cable position, signal propagation speed, multiple reflection coefficient, and Dirac function, and is used to describe the reflection characteristics of the traveling-wave cable in the time domain;

[0077] Performing a Fourier transform on the time-domain reflection function to obtain the frequency-domain reflection function. The frequency-domain reflection function includes the propagation constant, attenuation constant, phase constant, and angular frequency, and is used to describe the reflection characteristics of the traveling-wave cable in the frequency domain; introducing a frequency-dependent complex characteristic impedance, which is composed of the resistance, inductance, capacitance, and shunt conductance per unit length, to improve the accuracy of the frequency-domain reflection function;

[0078] Using the short-time Fourier transform method, the time-domain reflection function is converted into a time-frequency joint representation, which contains a time-offset parameter; the time-frequency joint representation is combined with the frequency-domain reflection function to obtain a time-frequency domain joint reflection model, which synthesizes time-domain and frequency-domain information and can more comprehensively describe the reflection characteristics of the traveling-wave cable;

[0079] Selecting a wavelet basis function, performing wavelet transform on the time-domain reflection function to obtain wavelet coefficients containing scale parameters and translation parameters; combining the wavelet coefficients with the time-frequency domain joint reflection model to obtain an optimized time-frequency domain joint reflection model.

[0080] In the specific implementation, first, the telegraph equation in the transmission line theory is used to solve the time-domain reflection function of the traveling-wave cable. This function contains parameters such as the attenuation coefficient, cable position, signal propagation speed, multiple reflection coefficient, and Dirac function, and is used to describe the reflection characteristics of the cable in the time domain. For example, for a cable with a length of 100m, the attenuation coefficient can be set to 0.01dB / m, the propagation speed to 2x10^8m / s, and the multiple reflection coefficient to 0.1. Through these parameters, a function curve representing the change of the cable reflection characteristics over time can be obtained.

[0081] Next, perform Fourier transform on the time-domain reflection function to obtain the frequency-domain reflection function. This function contains parameters such as the propagation constant, attenuation constant, phase constant, and angular frequency, and describes the reflection characteristics of the cable in the frequency domain. To improve the accuracy, a frequency-dependent complex characteristic impedance is introduced, which consists of the resistance, inductance, capacitance, and shunt conductance per unit length. For example, the resistance can be set to 10Ω / m, the inductance to 0.25μH / m, the capacitance to 100pF / m, and the shunt conductance to 0.1μS / m. In this way, a complex impedance function varying with frequency can be obtained.

[0082] Then, use the short-time Fourier transform method to convert the time-domain reflection function into a time-frequency joint representation. Specifically, a suitable window function (such as a Hanning window) can be selected, with a window length of 256 points and an overlap rate of 50%. By segmenting and windowing the time-domain signal and performing Fourier transform, a time-frequency representation containing a time-offset parameter is obtained. Combining this representation with the previously obtained frequency-domain reflection function, a time-frequency domain joint reflection model is constructed. This model synthesizes time-domain and frequency-domain information and can more comprehensively describe the reflection characteristics of the cable.

[0083] Finally, perform wavelet transform on the time-domain reflection function using the selected wavelet basis function. Daubechies wavelet can be selected, and the decomposition scale is 5 levels. Through wavelet transform, wavelet coefficients containing scale parameters and translation parameters can be obtained. Combining these wavelet coefficients with the previous time-frequency domain joint reflection model, an optimized model is obtained. This optimized model not only contains the overall characteristics of the time-frequency domain but also retains the local details of the signal, and can more accurately describe the reflection characteristics of the cable.

[0084] In practical applications, the optimized model can be used to analyze the measured cable reflection signals. For example, a 100m faulty cable is measured with a sampling rate of 1GHz and a measurement duration of 1μs. By processing the measured reflection signals through the above model, the location and nature of the fault point can be clearly identified. Compared with simply using time-domain or frequency-domain methods, this model can provide more comprehensive and accurate diagnostic results.

[0085] Through the above steps, the combination of time-domain reflection and frequency-domain reflection principles is achieved, and a time-frequency domain joint model that comprehensively describes the traveling wave cable reflection characteristics is constructed. This method combines multiple signal processing techniques and can effectively improve the accuracy and reliability of cable fault diagnosis.

[0086] In an optional implementation, the time-frequency domain joint reflection model is used to jointly analyze the time-domain data and frequency-domain data, and the characteristic parameters of the reflection signal are extracted, including:

[0087] Perform wavelet threshold denoising, baseline correction, and normalization processing on the collected time-domain data; apply the Hanning window function to the frequency-domain data, perform fast Fourier transform, and use the moving average method for spectrum smoothing;

[0088] Use the short-time Fourier transform method to generate a time-frequency spectrogram for the preprocessed time-domain data; match the time-frequency spectrogram with the pre-constructed time-frequency domain joint reflection model to calculate the similarity function; set a threshold based on the similarity function to determine potential abnormal points;

[0089] For the detected potential abnormal points, select the Morlet wavelet basis function for continuous wavelet transform to calculate the wavelet transform coefficients; construct a wavelet scale map based on the wavelet transform coefficients to analyze the signal characteristics at different scales;

[0090] According to the time-frequency spectrogram and the multi-resolution analysis results, extract the reflection peak position, reflection peak amplitude, phase information, delay estimation, frequency response, and attenuation coefficient; use the bispectrum analysis method to extract the non-linear characteristics of the signal;

[0091] Perform multi-scale fusion on the characteristic parameters to improve the parameter stability; use the complementarity of time-domain and frequency-domain information to perform time-frequency joint correction on the characteristic parameters; calculate the statistics of the characteristic parameters, including mean, variance, skewness, and kurtosis.

[0092] In the specific implementation, first, the collected time-domain data is preprocessed. The wavelet threshold denoising method is adopted. The selected wavelet basis function (such as db4 wavelet) and the decomposition level (usually 3 - 5 levels) are used to perform multi-scale decomposition on the signal. Soft threshold processing is carried out at each scale, and then the denoised signal is reconstructed. Next, baseline correction is performed. The baseline drift is estimated and removed by methods such as polynomial fitting or wavelet decomposition. Finally, the signal is normalized, and the signal amplitude is normalized to the range of [-1, 1].

[0093] For the frequency-domain data, first, the Hanning window function is applied to reduce spectral leakage. The window function length is usually selected to be 1 / 4 to 1 / 2 of the signal length. Then, the fast Fourier transform (FFT) is performed to obtain the spectrum of the signal. To smooth the spectrum, the moving average method is adopted, and the selected window length (such as 11 points) is used to smooth the spectrum.

[0094] The short-time Fourier transform (STFT) method is used to generate the time-frequency spectrogram for the preprocessed time-domain data. A suitable window function (such as Hamming window) and window length (such as 512 points) are selected. The signal is segmented and FFT is performed to obtain the time-frequency distribution map. The obtained time-frequency spectrogram is matched with the pre-constructed time-frequency domain joint reflection model, and the similarity function is calculated. The similarity function can adopt methods such as cross-correlation coefficient or Euclidean distance. Based on the similarity function, a threshold (such as 0.8) is set to determine potential abnormal points.

[0095] For the detected potential abnormal points, the Morlet wavelet basis function is selected for continuous wavelet transform (CWT). The Morlet wavelet has good time-frequency localization characteristics and is suitable for analyzing non-stationary signals. By adjusting the scale parameter (such as 1 - 64), the wavelet transform coefficients at different scales are calculated. Based on the wavelet transform coefficients, a wavelet scale map is constructed to analyze the signal characteristics at different scales, such as energy distribution, phase change, etc.

[0096] According to the time-frequency spectrogram and the multi-resolution analysis results, the characteristic parameters of the reflection signal are extracted. The reflection peak position can be determined by finding the local maximum points in the time-frequency spectrogram. The reflection peak amplitude is the amplitude at the corresponding position. The phase information can be obtained by calculating the phase angle of the STFT result. The time delay estimation adopts the cross-correlation method, calculates the cross-correlation function of the reference signal and the reflection signal, and takes the time delay corresponding to the maximum value. The frequency response is obtained by normalizing the FFT result. The attenuation coefficient can be obtained by fitting the attenuation curve of the reflection peak amplitude over time.

[0097] The bispectrum analysis method is used to extract the non-linear characteristics of the signal. First, the third-order cumulant of the signal is calculated, and then its two-dimensional Fourier transform is performed to obtain the bispectrum. Non-linear phase coupling and other characteristics can be extracted from the bispectrum.

[0098] Perform multi-scale fusion on the extracted feature parameters to improve parameter stability. The wavelet multi-resolution analysis method can be used to extract feature parameters at different scales and then fuse them through weighted averaging or other methods. Utilize the complementarity of time-domain and frequency-domain information to perform time-frequency joint correction on the feature parameters. For example, the results of time-domain peak detection and frequency-domain peak detection can be combined to improve the accuracy of peak positions.

[0099] Finally, calculate the statistics of the feature parameters, including mean, variance, skewness, and kurtosis. The mean reflects the overall level of the parameters, the variance characterizes the degree of dispersion, the skewness reflects the asymmetry of the distribution, and the kurtosis characterizes the sharpness of the distribution. These statistics can be used for subsequent classification or regression analysis.

[0100] Taking the reflection peak position parameter as an example, assume that 5 reflection peaks are detected in a certain measurement, and their positions are [1.2, 2.5, 3.8, 5.1, 6.4] ms. The calculated mean is 3.8 ms, the variance is 3.41 ms^2, the skewness is 0, and the kurtosis is 1.7. These statistics can be used to characterize the overall characteristics of the reflection peak distribution.

[0101] Through the above steps, the joint analysis of time-domain data and frequency-domain data can be realized, the feature parameters of the reflection signal can be extracted, and a basis can be provided for subsequent signal recognition and classification.

[0102] In an alternative embodiment, based on the feature parameters, in combination with the time-frequency domain joint reflection model, calculate the impedance change and signal propagation characteristics along the traveling-wave cable, including:

[0103] Using the time-frequency domain joint reflection model, extract the reflection coefficient at different positions and frequencies. The reflection coefficient is calculated by the ratio of the frequency-domain reflection function at that position to the spectrum of the incident signal; based on the reflection coefficient, calculate the characteristic impedance of the cable at each position, where the characteristic impedance is jointly determined by the nominal characteristic impedance of the cable and the reflection coefficient; calculate the impedance change rate, where the impedance change rate is the ratio of the difference between the characteristic impedance and the nominal characteristic impedance at each position to the nominal characteristic impedance; map the calculated impedance change rate to the time-frequency plane to generate a time-frequency joint impedance map;

[0104] Using the time-frequency domain joint reflection model, estimate the propagation constant of the cable. The propagation constant includes the attenuation constant and the phase constant, which are calculated from the frequency-domain reflection functions at different positions; based on the phase constant, calculate the group velocity of the signal, where the group velocity is the derivative of the angular frequency with respect to the phase constant; using the attenuation constant, calculate the transmission loss of the cable, where the transmission loss varies with distance and frequency; based on the group velocity, calculate the delay spread of the signal during propagation, where the delay spread is determined by the product of the propagation distance and the difference between the reciprocals of the minimum and maximum group velocities.

[0105] When implementing the calculation of the impedance change and signal propagation characteristics along a traveling-wave cable based on the time-frequency domain joint reflection model, it is first necessary to obtain the characteristic parameters of the cable, including the nominal characteristic impedance, resistance per unit length, inductance, capacitance, and conductance of the cable. These parameters can be obtained from the cable specification or through testing methods. For example, for a typical coaxial cable, its nominal characteristic impedance is 50 Ω, the resistance per unit length is 0.01 Ω / m, the inductance is 250 nH / m, the capacitance is 100 pF / m, and the conductance can be ignored.

[0106] Next, using the time-frequency domain joint reflection model, the reflection coefficient is extracted at different positions and frequencies of the cable. The specific method is to inject a broadband test signal at the input end of the cable and then measure the reflected signal at different positions. By performing time-frequency analysis on the reflected signal, the frequency-domain reflection function at that position can be obtained. Dividing the frequency-domain reflection function by the spectrum of the incident signal gives the reflection coefficient. For example, at a position 10 m from the cable entrance, the reflection coefficient measured at a frequency of 100 MHz is 0.05 + j0.02.

[0107] Based on the obtained reflection coefficient, the characteristic impedance of the cable at each position is calculated. The characteristic impedance is jointly determined by the nominal characteristic impedance of the cable and the reflection coefficient. When the reflection coefficient is zero, the characteristic impedance is equal to the nominal characteristic impedance; when the reflection coefficient is not zero, the characteristic impedance deviates from the nominal value. For example, for the above reflection coefficient, the calculated characteristic impedance at that position is 52.6 + j2.1 Ω.

[0108] Then the impedance change rate is calculated, which is the ratio of the difference between the characteristic impedance and the nominal characteristic impedance at each position to the nominal characteristic impedance. For the above example, the impedance change rate is 5.2% + j4.2%. Mapping the calculated impedance change rate to the time-frequency plane to generate a time-frequency joint impedance diagram can visually show the variation of the cable impedance with time and frequency.

[0109] When calculating the signal propagation characteristics, first use the time-frequency domain joint reflection model to estimate the propagation constant of the cable. The propagation constant includes the attenuation constant and the phase constant, which can be calculated from the frequency-domain reflection functions at different positions. For example, at a frequency of 100 MHz, the estimated attenuation constant is 0.05 Np / m, and the phase constant is 3.14 rad / m.

[0110] Based on the phase constant, the group velocity of the signal is calculated. The group velocity is the derivative of the angular frequency with respect to the phase constant and represents the speed at which the signal energy propagates. For example, at a frequency of 100 MHz, the calculated group velocity is 2×10^8 m / s. Using the attenuation constant, the transmission loss of the cable can be calculated. The transmission loss varies with distance and frequency and is usually expressed in dB / m. For example, at a frequency of 100 MHz, the transmission loss is 0.43 dB / m.

[0111] Finally, based on the group velocity, the delay spread of the signal during propagation is calculated. The delay spread is determined by the product of the propagation distance and the difference between the reciprocals of the minimum and maximum group velocities, reflecting the broadening degree of the signal during propagation. For example, for a cable with a length of 100 m, in the frequency band of 10 - 200 MHz, the calculated delay spread is 5 ns.

[0112] Through the above steps, the impedance changes and signal propagation characteristics along the traveling - wave cable can be comprehensively analyzed. These information are of great significance for cable fault diagnosis, signal integrity analysis, etc. For example, positions with a relatively large impedance change rate may have local damage or poor joints; too high transmission loss may indicate cable aging or moisture; too large delay spread may cause signal distortion. By analyzing these characteristics, problems existing in the cable can be detected in time and corresponding maintenance measures can be taken.

[0113] In an alternative embodiment, the method further includes:

[0114] Based on the time - frequency joint impedance diagram, set a threshold to mark potential impedance abnormal points; analyze the change trends of the propagation constant, group velocity, transmission loss, and delay spread to identify abnormal regions; perform spatial correspondence and correlation analysis on the impedance abnormal points and the abnormal regions of propagation characteristics; based on the above - mentioned analysis results, combined with the impedance change amplitude, the increase amount of transmission loss, and the degree of delay spread, establish a comprehensive scoring mechanism to quantify the abnormal degree of each section of the cable; compare the measurement results at different times, analyze the change trend of the cable state, and provide a basis for predictive maintenance.

[0115] In this embodiment, first, a threshold is set based on the time - frequency joint impedance diagram to mark potential impedance abnormal points. Specifically, by analyzing the amplitude distribution characteristics of the time - frequency joint impedance diagram, a suitable threshold can be set. For example, take the 95% quantile of the impedance amplitude as the threshold. Mark the points exceeding this threshold as potential impedance abnormal points, and these points may correspond to the fault or deterioration positions in the cable.

[0116] Next, analyze the change trends of the propagation constant, group velocity, transmission loss, and delay spread to identify abnormal regions. For the propagation constant, its first - order derivative along the cable length can be calculated. If the change rate of the propagation constant of a certain section of the cable is significantly higher than the average level, it is marked as an abnormal region. The group velocity should generally remain relatively stable. If the group velocity of a certain section suddenly decreases by more than 10%, it can also be considered abnormal. The transmission loss needs to consider frequency dependence. The average loss in different frequency bands can be calculated and compared with the theoretical value. The region with a deviation exceeding 3 dB / 100 m is regarded as abnormal. The delay spread generally increases slowly with distance. If there is a sharp increase in a certain section, such as an increase amplitude exceeding 50%, it is marked as an abnormal region.

[0117] Then, spatial correspondence and correlation analysis are performed on the impedance anomaly points and the propagation characteristic anomaly regions. This step aims to verify and complement the analysis results of the previous two steps. For example, if there are both impedance anomalies and transmission loss anomalies at a certain location, it can be more certain that there is a problem there. Or if there is only an impedance anomaly at a certain location but the propagation characteristics are normal, it may be a false alarm. Through this correlation analysis, the accuracy of fault location can be improved.

[0118] Based on the above analysis results, combined with the impedance change amplitude, the increase in transmission loss, and the degree of delay spread, a comprehensive scoring mechanism is established to quantify the anomaly degree of each section of the cable. Weights can be set for each index. For example, the impedance change accounts for 40%, the transmission loss accounts for 40%, and the delay spread accounts for 20%. For the impedance change, the density and amplitude of the anomaly points can be calculated; for the transmission loss, the excess amount relative to the theoretical value is calculated; for the delay spread, the increase ratio relative to the normal value is calculated. After normalizing these indexes, the comprehensive score is obtained by weighting according to the weights. The higher the score, the more serious the anomaly degree of that section of the cable.

[0119] Finally, compare the measurement results at different times, analyze the change trend of the cable state, and provide a basis for predictive maintenance. The comprehensive score of each measurement can be calculated and a change curve over time can be plotted. If the score of a certain section of the cable shows an upward trend, even if it does not reach the level of serious anomaly currently, attention should be paid and further inspections should be arranged. The change rate of each index can also be analyzed to predict the time when problems may occur in the future, so as to reasonably arrange the maintenance plan.

[0120] For example, assume that a 1000-meter-long power cable is measured and the following results are obtained: an impedance anomaly point is found at 500 meters, and the amplitude exceeds the threshold by 20%; in the interval of 450 - 550 meters, the transmission loss is 4 dB / 100m higher than the theoretical value; the delay spread increases by 60% in this interval. The comprehensive score is 0.4×20% + 0.4×(4 / 3) + 0.2×60% = 0.61. If the comprehensive score of this section rises to 0.75 when measured again three months later, it indicates that the problem is intensifying and maintenance should be arranged as soon as possible.

[0121] Through this method, the health status of the cable can be comprehensively evaluated, potential problems can be discovered in a timely manner, predictive maintenance can be achieved, and the reliability and economy of the power system can be improved.

[0122] In an alternative embodiment, through the analysis of impedance changes and signal propagation characteristics, intrusion detection, leakage location, and fault diagnosis along the traveling wave cable are achieved, including:

[0123] Using a time-frequency domain joint reflection model, reflection coefficients are extracted at different positions and frequencies, and the reflection coefficient is calculated as the ratio of the frequency-domain reflection function at that position to the spectrum of the incident signal; based on the reflection coefficient, the characteristic impedance of the cable at each position is calculated, and the characteristic impedance is jointly determined by the nominal characteristic impedance of the cable and the reflection coefficient; the impedance change rate is calculated, and the impedance change rate is the ratio of the difference between the characteristic impedance and the nominal characteristic impedance at each position to the nominal characteristic impedance;

[0124] The calculated impedance change rate is mapped to the time-frequency plane to generate a time-frequency joint impedance map; time-frequency domain feature extraction is performed on the time-frequency joint impedance map, including calculating the first and second derivatives of the impedance change rate on the time axis, performing wavelet packet decomposition on the impedance change rate to extract the energy distribution characteristics of different frequency bands, and calculating the instantaneous frequency and group delay of the impedance change rate using the Wigner-Ville distribution;

[0125] A sliding window algorithm is designed for local anomaly detection in the spatial and temporal dimensions, including setting the spatial window size, calculating the statistical characteristics of the impedance change rate within the window, setting the temporal window size, analyzing the temporal evolution characteristics of the impedance change rate, and combining the analysis results of the spatial and temporal windows to calculate the comprehensive anomaly index;

[0126] The Gaussian mixture model is used to model the comprehensive anomaly index in the normal state, the dynamic detection threshold is calculated based on the Gaussian mixture model, and the particle filter algorithm is introduced to optimize the dynamic detection threshold in real time; when the comprehensive anomaly index is greater than the dynamic detection threshold, an anomaly alarm is triggered, and the continuously triggered alarm points are clustered in space-time to screen out potential anomaly events; using the time-frequency domain joint reflection model, the propagation constant of the cable is estimated, and the propagation constant includes the attenuation constant and the phase constant, which are calculated from the frequency-domain reflection functions at different positions;

[0127] Based on the phase constant, the group velocity of the signal is calculated, and the group velocity is the derivative of the angular frequency with respect to the phase constant; using the attenuation constant, the transmission loss of the cable is calculated, and the transmission loss varies with distance and frequency; the spatial gradient of the attenuation constant is calculated to identify regions where the attenuation increases abnormally, and the frequency dependence of the group velocity is analyzed to detect abnormal changes in the dispersion characteristics;

[0128] Wavelet multi-resolution analysis is performed on the attenuation constant, group velocity, and transmission loss to extract features at different scales, calculate the correlation coefficient matrix between features at different scales, identify cross-scale abnormal patterns, and reconstruct the abnormal feature map based on the significant scale; local maximum points are searched for in the abnormal feature map as potential abnormal positions, and the spatial centroid of the abnormal feature map is calculated to provide a preliminary estimate of the abnormal position;

[0129] Fuse the impedance characteristics, propagation characteristics, and temporal characteristics extracted by the long short-term memory network to construct a high-dimensional feature space; use principal component analysis and t-SNE algorithm to reduce the dimension of the high-dimensional feature space, and use the DBSCAN algorithm for density clustering in the reduced-dimensional space to identify potential abnormal patterns; train a support vector machine classifier and a random forest model to identify and classify different types of anomalies, and evaluate the importance of each feature in anomaly classification; establish a fuzzy rule base based on expert knowledge to evaluate the severity of anomalies, and use historical data to establish a regression model between anomaly features and severity;

[0130] Combine the anomaly severity and historical data to predict the remaining service life of the cable, consider the anomaly type, severity, and system importance to prioritize the maintenance tasks, and use the reinforcement learning algorithm to optimize the long-term maintenance strategy.

[0131] In the specific implementation, first use the time-frequency domain joint reflection model to analyze the reflection characteristics of the traveling wave cable. By extracting the reflection coefficients at different positions and frequencies, the impedance distribution information along the cable can be obtained. The calculation of the reflection coefficient is obtained by dividing the frequency-domain reflection function at that position by the incident signal spectrum. Based on the reflection coefficient, the characteristic impedance of the cable at each position can be further calculated, and the characteristic impedance is jointly determined by the nominal characteristic impedance of the cable and the reflection coefficient. In order to quantify the degree of impedance change, it is necessary to calculate the impedance change rate, that is, the difference between the characteristic impedance and the nominal characteristic impedance at each position divided by the nominal characteristic impedance.

[0132] Mapping the calculated impedance change rate to the time-frequency plane can generate a time-frequency joint impedance map, which intuitively shows the time and frequency characteristics of impedance change. Further feature extraction is performed on the time-frequency joint impedance map, including calculating the first and second derivatives of the impedance change rate on the time axis to reflect the speed and acceleration of impedance change. At the same time, wavelet packet decomposition is performed on the impedance change rate to extract the energy distribution characteristics of different frequency bands. In addition, using the Wigner-Ville distribution to calculate the instantaneous frequency and group delay of the impedance change rate can obtain the frequency characteristics of impedance change.

[0133] In order to achieve local anomaly detection of the traveling wave cable, a sliding window algorithm is designed. The algorithm analyzes simultaneously in the spatial and temporal dimensions. First, set the spatial window size, such as 10 meters, and calculate the statistical characteristics of the impedance change rate within this window, such as mean, variance, skewness, and kurtosis. Then set the time window size, such as 1 hour, and analyze the temporal evolution characteristics of the impedance change rate during this time period, including trends, periodicity, and mutations. Combining the analysis results of the spatial and temporal windows, a comprehensive anomaly index can be calculated, which can reflect the anomaly degree of the local area within a specific time period.

[0134] To accurately judge anomalies, it is necessary to establish a statistical model of the comprehensive anomaly index under normal conditions. Here, the Gaussian mixture model is used to model the normal state, and this model can describe complex multi-modal distributions. Based on the Gaussian mixture model, a dynamic detection threshold can be calculated. To enable the detection threshold to adapt to changes in the system state, the particle filter algorithm is introduced to optimize the threshold in real time. When the comprehensive anomaly index exceeds the dynamic detection threshold, an anomaly alarm is triggered. For continuously triggered alarm points, spatio-temporal clustering analysis is performed to screen out potential anomaly events.

[0135] In addition to impedance characteristics, the propagation characteristics of signals also contain rich diagnostic information. Using the time-frequency domain joint reflection model, the propagation constant of the cable can be estimated, including the attenuation constant and the phase constant. Based on the phase constant, the group velocity of the signal can be calculated, which is the derivative of the angular frequency with respect to the phase constant. Using the attenuation constant, the transmission loss of the cable can be calculated, and its variation law with distance and frequency can be analyzed. By calculating the spatial gradient of the attenuation constant, the area where the attenuation increases abnormally can be identified. At the same time, by analyzing the frequency dependence of the group velocity, abnormal changes in the dispersion characteristics can be detected.

[0136] To analyze the propagation characteristics more comprehensively, wavelet multi-resolution analysis is performed on the attenuation constant, group velocity, and transmission loss to extract features at different scales. By calculating the correlation coefficient matrix between features at different scales, cross-scale abnormal patterns can be identified. Based on the significant scale, the abnormal feature map is reconstructed, and local maximum points are searched in this map as potential abnormal positions. Calculating the spatial centroid of the abnormal feature map can provide a preliminary estimate of the abnormal position.

[0137] To comprehensively utilize various feature information, the impedance features, propagation features, and temporal features extracted using long short-term memory networks are fused to construct a high-dimensional feature space. The principal component analysis and t-SNE algorithms are used to reduce the dimension of the high-dimensional feature space, and the DBSCAN algorithm is used for density clustering in the reduced-dimensional space to identify potential abnormal patterns. A support vector machine classifier and a random forest model are trained to identify and classify different types of anomalies, and the importance of each feature in anomaly classification is evaluated.

[0138] A fuzzy rule base based on expert knowledge is established to evaluate the severity of anomalies. At the same time, a regression model of anomaly features and severity is established using historical data. Combining the anomaly severity and historical data, the remaining service life of the cable is predicted. Considering the anomaly type, severity, and system importance, the maintenance tasks are prioritized. Finally, the long-term maintenance strategy is optimized using the reinforcement learning algorithm to achieve a balance between cost and reliability.

[0139] In practical applications, a 100-meter-long traveling wave cable can be selected for testing. First, different types of defects, such as slight indentations, local corrosion, and insulation layer damage, are artificially created at different positions of the cable (e.g., at 10 meters, 30 meters, 50 meters, 70 meters, and 90 meters). Then, the network analyzer is used to scan the reflection characteristics of the cable in the frequency range from 1 MHz to 1 GHz, with a sampling interval of 1 MHz. Based on the measured reflection data, the reflection coefficient and characteristic impedance at each position are calculated.

[0140] Taking the local corrosion at 50 meters as an example, the calculated impedance change rate at this position is 5%. Performing time-frequency analysis on the impedance change rate, it is found that there is obvious energy concentration near 500 MHz. Calculating the first derivative of the impedance change rate, the maximum value appears at 49.8 meters, indicating a sharp impedance change at this location. Using a 60-cm spatial sliding window and a 1-hour time window for anomaly detection, the calculated comprehensive anomaly index for this area is 0.85, exceeding the dynamic detection threshold of 0.7, triggering an anomaly alarm.

[0141] Analyzing the signal propagation characteristics, it is found that the attenuation constant at 50 meters increases from the original 0.2 dB / m to 0.3 dB / m, and the spatial gradient reaches 0.1 dB / m². The group velocity decreases from the original 2×10^8 m / s to 1.8×10^8 m / s at the 500-MHz frequency point, indicating a change in the dispersion characteristics. Inputting these features into the trained random forest model, the probability that this anomaly belongs to the "local corrosion" type is 92%. According to expert rule evaluation, the severity of this anomaly is "medium".

[0142] The life prediction model trained based on historical data shows that if no repair is carried out, this defect will develop into a serious fault within 3 months. Considering the high importance of this section of the cable, the system sets the priority of its maintenance task to "high" and recommends completing the repair within 2 weeks. The reinforcement learning algorithm further optimizes the maintenance strategy and proposes a suggestion to conduct a comprehensive inspection of the entire cable while repairing this defect to improve long-term reliability.

[0143] Figure 2 This is a schematic structural diagram of the traveling wave cable measurement and perception system based on the joint reflection principle in the time-frequency domain according to the embodiment of the present invention. As Figure 2 shown, the system includes:

[0144] The first unit is used to set a high-frequency signal transmitting device and a signal receiving device at one end of a traveling-wave cable. The high-frequency signal transmitting device transmits a swept-frequency signal within a preset frequency range to the traveling-wave cable, and the signal receiving device receives the reflected signal after being transmitted by the traveling-wave cable. Using the time-domain reflection principle and the frequency-domain reflection principle, the time-domain reflection function and the frequency-domain reflection function of the traveling-wave cable are established. Combining the time-domain reflection function and the frequency-domain reflection function, a time-frequency domain joint reflection model is constructed, which includes the geometric parameters, electrical parameters, and environmental parameters of the traveling-wave cable.

[0145] The second unit is used to collect the reflected signal of the traveling-wave cable through the signal receiving device, perform noise reduction processing and digital conversion on the collected reflected signal to obtain the time-domain data and frequency-domain data of the reflected signal. Using the time-frequency domain joint reflection model, the time-domain data and frequency-domain data are jointly analyzed to extract the characteristic parameters of the reflected signal, including signal amplitude, phase, time delay, and frequency response.

[0146] The third unit is used to calculate the impedance change and signal propagation characteristics along the traveling-wave cable based on the characteristic parameters and in combination with the time-frequency domain joint reflection model. Through the analysis of the impedance change and signal propagation characteristics, intrusion detection, leakage location, and fault diagnosis of the traveling-wave cable along the line are realized. Comparing the detection result with a preset threshold, when the detection result exceeds the preset threshold, an alarm mechanism is triggered and corresponding measurement perception information is output, including the intrusion location, leakage point location, or fault point location.

[0147] In the third aspect of the embodiments of the present invention,

[0148] A kind of electronic device is provided, including:

[0149] A processor;

[0150] A memory for storing instructions executable by the processor;

[0151] Wherein, the processor is configured to call the instructions stored in the memory to execute the method described above.

[0152] In the fourth aspect of the embodiments of the present invention,

[0153] A computer-readable storage medium is provided, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method described above is realized.

[0154] The present invention can be a method, device, system, and / or computer program product. The computer program product can include a computer-readable storage medium, on which computer-readable program instructions for executing various aspects of the present invention are uploaded.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A traveling wave cable measurement and perception method based on the time-frequency domain joint reflection principle is characterized by: include: A high-frequency signal transmitting device and a signal receiving device are arranged at one end of the traveling wave cable, wherein the high-frequency signal transmitting device transmits a sweeping frequency signal within a preset frequency range to the traveling wave cable, and the signal receiving device receives a reflected signal after the traveling wave cable is transmitted; a time domain reflection function and a frequency domain reflection function of the traveling wave cable are established by using the time domain reflection principle and the frequency domain reflection principle; the time domain reflection function and the frequency domain reflection function are combined to construct a time-frequency domain joint reflection model, which includes geometric parameters, electrical parameters and environmental parameters of the traveling wave cable; The reflected signal of the traveling wave cable is collected by the signal receiving device, and the collected reflected signal is subjected to noise reduction processing and digital conversion to obtain time domain data and frequency domain data of the reflected signal; the time domain data and the frequency domain data are jointly analyzed by using a time-frequency domain joint reflection model to extract characteristic parameters of the reflected signal, including signal amplitude, phase, delay and frequency response; Based on the characteristic parameters and combined with the time-frequency domain joint reflection model, the impedance change and signal propagation characteristics along the traveling wave cable are calculated; through the analysis of impedance change and signal propagation characteristics, intrusion detection, leakage location and fault diagnosis along the traveling wave cable are realized; the detection result is compared with the preset threshold. When the detection result exceeds the preset threshold, the alarm mechanism is triggered and the corresponding measurement perception information is output, including the intrusion location, leakage point location or fault point location.

2. The traveling wave cable measurement and perception method based on the time-frequency domain joint reflection principle according to claim 1 is characterized in that: By using the time domain reflection principle and the frequency domain reflection principle, a time domain reflection function and a frequency domain reflection function of the traveling wave cable are established; and the time domain reflection function and the frequency domain reflection function are combined to construct a time-frequency domain joint reflection model, including: The telegraph equation in the transmission line theory is used to solve the time domain reflection function of the traveling wave cable, which includes the attenuation coefficient, the cable position, the signal propagation speed, the multiple reflection coefficient and the Dirac function, and is used to describe the reflection characteristics of the traveling wave cable in the time domain. Performing Fourier transform on the time domain reflection function to obtain the frequency domain reflection function, wherein the frequency domain reflection function includes a propagation constant, an attenuation constant, a phase constant and an angular frequency, and is used to describe the reflection characteristics of the traveling wave cable in the frequency domain; introducing a frequency-dependent complex characteristic impedance, wherein the complex characteristic impedance is composed of resistance, inductance, capacitance and parallel conductance per unit length, and is used to improve the accuracy of the frequency domain reflection function; The short-time Fourier transform method is used to convert the time-domain reflection function into a time-frequency joint representation, which includes a time offset parameter; the time-frequency joint representation is combined with the frequency-domain reflection function to obtain a time-frequency domain joint reflection model, which integrates the time-domain and frequency-domain information and can more comprehensively describe the reflection characteristics of the traveling wave cable; The selected wavelet basis function performs wavelet transform on the time domain reflection function to obtain wavelet coefficients including scale parameters and translation parameters; the wavelet coefficients are combined with the time-frequency domain joint reflection model to obtain an optimized time-frequency domain joint reflection model.

3. The traveling wave cable measurement and perception method based on the time-frequency domain joint reflection principle according to claim 1 is characterized in that: The time domain data and the frequency domain data are jointly analyzed using a time-frequency domain joint reflection model to extract characteristic parameters of the reflection signal, including: The collected time domain data were subjected to wavelet threshold denoising, baseline correction and normalization processing; the frequency domain data were subjected to Hanning window function, fast Fourier transform, and spectrum smoothing using moving average method; Using the short-time Fourier transform method, a time-frequency spectrum is generated for the preprocessed time-domain data; the time-frequency spectrum is matched with a pre-constructed time-frequency domain joint reflection model to calculate a similarity function; a threshold is set based on the similarity function to determine potential abnormal points; For the detected potential abnormal points, the Morlet wavelet basis function is selected to perform continuous wavelet transform and the wavelet transform coefficients are calculated; a wavelet scalogram is constructed based on the wavelet transform coefficients to analyze the signal characteristics at different scales; According to the time-frequency spectrum and multi-resolution analysis results, the reflection peak position, reflection peak amplitude, phase information, delay estimation, frequency response and attenuation coefficient are extracted; the nonlinear characteristics of the signal are extracted using the bispectral analysis method; Perform multi-scale fusion of characteristic parameters to improve parameter stability; utilize the complementarity of time domain and frequency domain information to perform time-frequency joint correction of characteristic parameters; calculate the statistics of characteristic parameters, including mean, variance, skewness and kurtosis.

4. The traveling wave cable measurement and perception method based on the time-frequency domain joint reflection principle according to claim 1 is characterized in that: Based on the characteristic parameters and combined with the time-frequency domain joint reflection model, the impedance change and signal propagation characteristics along the traveling wave cable are calculated, including: Using a time-frequency domain joint reflection model, the reflection coefficient is extracted at different positions and frequencies, and the reflection coefficient is calculated by the ratio of the frequency domain reflection function at the position to the incident signal spectrum; based on the reflection coefficient, the characteristic impedance of the cable at each position is calculated, and the characteristic impedance is determined by the nominal characteristic impedance of the cable and the reflection coefficient; the impedance change rate is calculated, and the impedance change rate is the ratio of the difference between the characteristic impedance at each position and the nominal characteristic impedance to the nominal characteristic impedance; the calculated impedance change rate is mapped to the time-frequency plane to generate a time-frequency joint impedance diagram; The propagation constant of the cable is estimated by using a time-frequency domain joint reflection model. The propagation constant includes an attenuation constant and a phase constant, which are calculated by frequency domain reflection functions at different positions. The group velocity of the signal is calculated based on the phase constant. The group velocity is the derivative of the angular frequency with respect to the phase constant. The transmission loss of the cable is calculated by using the attenuation constant. The transmission loss varies with distance and frequency. The time delay spread of the signal during propagation is calculated based on the group velocity. The time delay spread is determined by the product of the propagation distance and the difference between the reciprocal of the minimum group velocity and the maximum group velocity.

5. The traveling wave cable measurement and perception method based on the time-frequency domain joint reflection principle according to claim 4 is characterized in that: The method further comprises: Based on the time-frequency joint impedance diagram, a threshold is set to mark potential impedance anomaly points; the changing trends of the propagation constant, group velocity, transmission loss and delay spread are analyzed to identify abnormal areas; the impedance anomaly points and propagation characteristic abnormal areas are spatially corresponded and correlated; based on the above analysis results, a comprehensive scoring mechanism is established in combination with the impedance change amplitude, transmission loss increase and delay spread degree to quantify the degree of abnormality of each section of the cable; the measurement results of different periods are compared, the changing trend of the cable status is analyzed, and a basis for predictive maintenance is provided.

6. The traveling wave cable measurement and perception method based on the time-frequency domain joint reflection principle according to claim 1 is characterized in that: Through the analysis of impedance changes and signal propagation characteristics, intrusion detection, leakage location and fault diagnosis along the traveling wave cable are realized, including: Using a time-frequency domain joint reflection model, the reflection coefficient is extracted at different positions and frequencies, and the reflection coefficient is calculated by the ratio of the frequency domain reflection function at the position to the incident signal spectrum; based on the reflection coefficient, the characteristic impedance of the cable at each position is calculated, and the characteristic impedance is determined by the nominal characteristic impedance of the cable and the reflection coefficient; the impedance change rate is calculated, and the impedance change rate is the ratio of the difference between the characteristic impedance at each position and the nominal characteristic impedance to the nominal characteristic impedance; The calculated impedance change rate is mapped to the time-frequency plane to generate a time-frequency joint impedance diagram; time-frequency domain feature extraction is performed on the time-frequency joint impedance diagram, including calculating the first-order and second-order derivatives of the impedance change rate on the time axis, performing wavelet packet decomposition on the impedance change rate to extract energy distribution characteristics of different frequency bands, and calculating the instantaneous frequency and group delay of the impedance change rate using Wigner-Ville distribution; Design a sliding window algorithm to perform local anomaly detection in spatial and temporal dimensions, including setting the spatial window size, calculating the statistical characteristics of the impedance change rate within the window, setting the temporal window size, analyzing the temporal evolution characteristics of the impedance change rate, and combining the analysis results of the spatial and temporal windows to calculate the comprehensive anomaly index; The Gaussian mixture model is used to model the comprehensive abnormality index under normal conditions, the dynamic detection threshold is calculated based on the Gaussian mixture model, and the particle filter algorithm is introduced to optimize the dynamic detection threshold in real time; when the comprehensive abnormality index is greater than the dynamic detection threshold, an abnormal alarm is triggered, and the continuously triggered alarm points are clustered in time and space to screen out potential abnormal events; the propagation constant of the cable is estimated using the time-frequency domain joint reflection model, and the propagation constant includes an attenuation constant and a phase constant, which are calculated from the frequency domain reflection function at different positions; Based on the phase constant, the group velocity of the signal is calculated, and the group velocity is the derivative of the angular frequency with respect to the phase constant; using the attenuation constant, the transmission loss of the cable is calculated, and the transmission loss varies with distance and frequency; the spatial gradient of the attenuation constant is calculated, the area where the attenuation increases abnormally is identified, the frequency dependence of the group velocity is analyzed, and the abnormal changes in the dispersion characteristics are detected; Perform wavelet multi-resolution analysis on the attenuation constant, group velocity and transmission loss, extract features of different scales, calculate the correlation coefficient matrix between features of different scales, identify cross-scale abnormal patterns, and reconstruct abnormal feature maps based on significance scales; find local maximum points in the abnormal feature maps as potential abnormal locations, calculate the spatial centroid of the abnormal feature maps, and provide a preliminary estimate of the abnormal location; The impedance characteristics, propagation characteristics and time series characteristics extracted by long short-term memory network are integrated to construct a high-dimensional feature space; the high-dimensional feature space is reduced in dimension using principal component analysis and t-SNE algorithm, and the DBSCAN algorithm is used to perform density clustering in the reduced-dimensional space to identify potential abnormal patterns; support vector machine classifiers and random forest models are trained to identify and classify different types of anomalies, and the importance of each feature in anomaly classification is evaluated; a fuzzy rule base based on expert knowledge is established to evaluate the severity of anomalies, and a regression model of abnormal characteristics and severity is established using historical data; Combine anomaly severity and historical data to predict the remaining service life of the cable, prioritize maintenance tasks considering anomaly type, severity, and system importance, and optimize long-term maintenance strategies using reinforcement learning algorithms.

7. A traveling wave cable measurement and perception system based on the time-frequency domain joint reflection principle, used to implement the method described in any one of claims 1 to 6, characterized in that: include: The first unit is used to set a high-frequency signal transmitting device and a signal receiving device at one end of the traveling wave cable, wherein the high-frequency signal transmitting device transmits a sweeping frequency signal within a preset frequency range to the traveling wave cable, and the signal receiving device receives a reflected signal after the traveling wave cable is transmitted; using the time domain reflection principle and the frequency domain reflection principle, a time domain reflection function and a frequency domain reflection function of the traveling wave cable are established; combining the time domain reflection function and the frequency domain reflection function to construct a time-frequency domain joint reflection model, which includes geometric parameters, electrical parameters and environmental parameters of the traveling wave cable; The second unit is used to collect the reflected signal of the traveling wave cable through the signal receiving device, perform noise reduction processing and digital conversion on the collected reflected signal, and obtain time domain data and frequency domain data of the reflected signal; use the time-frequency domain joint reflection model to jointly analyze the time domain data and the frequency domain data, and extract characteristic parameters of the reflected signal, including signal amplitude, phase, delay and frequency response; The third unit is used to calculate the impedance change and signal propagation characteristics along the traveling wave cable based on the characteristic parameters and the time-frequency domain joint reflection model; through the analysis of the impedance change and signal propagation characteristics, intrusion detection, leakage location and fault diagnosis along the traveling wave cable are realized; the detection result is compared with the preset threshold. When the detection result exceeds the preset threshold, the alarm mechanism is triggered and the corresponding measurement perception information is output, including the intrusion location, leakage point location or fault point location.

8. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images