Cable full life cycle state monitoring method based on electromagnetic harmonic characteristic analysis

Abstract

The invention belongs to the technical field of power equipment monitoring, and relates to a cable full life cycle state monitoring method based on electromagnetic harmonic characteristic analysis, which comprises the following steps: generating a discrete signal sequence; generating a denoised signal sequence through threshold filtering reconstruction; dividing the denoised signal sequence into segmented signal sets according to a preset segment length, wherein each segment of sub-signal corresponds to a local area of the physical length of the cable; extracting multi-dimensional feature parameters, and combining according to segments to generate global feature vectors; judging the aging degree of the cable through a multi-level threshold value, and triggering an alarm and positioning process according to a judgment result; generating a multi-node signal data set; calculating the distance range of the fault point relative to each monitoring node, and generating a candidate distance set; and carrying out weighted fusion on the candidate distance set, and outputting a final positioning result of the fault point. According to the method, the problem that a positioning algorithm does not fuse multi-dimensional features and space-time relevance in a traditional mode is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power equipment monitoring and relates to a method for monitoring the whole life cycle state of a cable based on electromagnetic harmonic characteristic analysis. Background Art

[0002] As the core carrier of power transmission, a cable is prone to faults such as partial discharge and breakdown due to insulation aging, mechanical stress or environmental corrosion during long-term operation, threatening the safety of the power grid. Traditional monitoring methods rely on regular manual inspections and partial discharge detection, making it difficult to capture early aging characteristics in real time and accurately locate hidden fault points, resulting in delayed fault warnings and low maintenance efficiency. Especially in complex electromagnetic environments or long-distance cable scenarios, problems such as noise interference and signal attenuation further reduce the monitoring reliability.

[0003] Conventional solutions mostly use a single sensor to monitor the signal at the cable head end and combine threshold alarm or impedance method to locate faults. For example, the fault distance is estimated by measuring the change in cable impedance, or the signal reflection wave is captured by the time domain reflectometry method for rough positioning. Although such methods can achieve basic monitoring, they are limited by narrow signal bandwidth and weak anti-interference ability, making it difficult to distinguish noise from real fault characteristics and prone to false alarms and missed detections. In addition, traditional positioning techniques rely on single-node data and cannot solve problems such as multi-path propagation and non-linear signal attenuation, and the positioning error often reaches the order of dozens of meters, making it difficult to meet the high-precision requirements.

[0004] Based on the above problems, the traditional method has poor result stability due to the lack of integration of multi-dimensional features and spatio-temporal correlation in the positioning algorithm. Summary of the Invention

[0005] To solve the above problems, the present invention provides a method for monitoring the whole life cycle state of a cable based on electromagnetic harmonic characteristic analysis.

[0006] The method for monitoring the whole life cycle state of a cable based on electromagnetic harmonic characteristic analysis includes the following steps: S1. Obtain the continuous electromagnetic harmonic signals naturally generated during the operation of the cable and perform discretization processing at a preset sampling frequency to generate a discrete signal sequence; S2. Decompose the discrete signal sequence, separate the noise component and the effective harmonic component, and reconstruct through threshold filtering to generate a denoised signal sequence; S3. Divide the denoised signal sequence into a segmented signal set according to a preset segment length, and each segment of the sub-signal corresponds to a local area of the physical length of the cable; S4. Perform joint time-domain and frequency-domain analysis on each segment of the sub-signal in the segmented signal set, extract multi-dimensional characteristic parameters, and generate a global characteristic vector by combining them segment by segment; S5. Based on the global feature vector, judge the aging degree of the cable through multi-level thresholds, and trigger the alarm and positioning process according to the judgment result; S6. When it is determined that the cable is in a severely aging state, activate multiple monitoring nodes along the cable to synchronously collect electromagnetic harmonic signals and generate a multi-node signal data set; S7. Analyze the attenuation law of the characteristic parameters and the phase propagation time delay difference of the multi-node signal data set, calculate the distance range of the fault point relative to each monitoring node, and generate a candidate distance set; S8. Perform weighted fusion on the candidate distance set, screen the optimal solution that satisfies the signal attenuation consistency, phase propagation time delay matching, and energy distribution correlation, and output the final positioning result of the fault point.

[0007] A further solution of the present invention for generating a discrete signal sequence includes the following steps: The electromagnetic harmonic sensor is arranged at the cable head or a preset monitoring point. The electromagnetic harmonic sensor is internally provided with a broadband induction coil for coupling the alternating electromagnetic field between the cable conductor and the shielding layer to generate a continuous electromagnetic harmonic signal; Use an analog-to-digital conversion circuit to periodically sample the continuous electromagnetic harmonic signal to generate the discrete signal sequence. The sampling frequency is determined according to the rated voltage level of the cable and the highest harmonic frequency.

[0008] A further solution of the present invention for generating a denoised signal sequence through threshold filtering reconstruction includes the following steps: Input the discrete signal sequence into a signal decomposition algorithm to decompose the signal into multiple sub-signal components according to different frequency bands; Set a dynamic threshold to judge the boundary between noise and effective harmonics. The sub-signal components higher than the threshold are determined as noise and set to zero, and the sub-signal components lower than the threshold are retained as effective harmonics. The dynamic threshold is adaptively calculated according to the noise standard deviation and signal length of the sub-signal components; All the processed sub-signal components are merged in reverse according to the original frequency band to generate a denoised signal sequence.

[0009] A further solution of the present invention for dividing the denoised signal sequence into a segmented signal set according to a preset segmentation length includes the following steps: Determine the physical length according to the ratio of the total length of the cable to the preset segmentation length; Calculate the number of sampling points included in each sub-signal in combination with the preset sampling frequency; Take the sampling points as the unit and intercept continuous data segments from the denoised signal sequence in chronological order to generate a segmented signal set corresponding to the local area of the cable physical length.

[0010] A further solution of the present invention for intercepting continuous data segments from the denoised signal sequence in chronological order includes the following steps: If the remaining signal length is less than one segment, it is extended to a complete segment through zero-padding operation; the zero-padding operation refers to the process of filling zero-valued data at the end of the signal to meet the segment length requirement. The zero-padding operation only fills zero values at the end of the signal, and the effective data segment is automatically recognized during feature extraction to exclude the interference of the zero-padded part on the analysis.

[0011] A further solution of the present invention is to generate a global feature vector, including the following steps: Each sub-signal in the segmented signal set converts the time-domain signal into a frequency-domain signal through fast Fourier transform, calculates the ratio of the sum of the squares of the amplitudes of each harmonic to the sum of the squares of the total harmonic amplitudes, and calculates the harmonic energy distribution characteristics; Statistically analyze the fluctuation degree of the signal amplitude in the time domain, and generate a signal fluctuation characteristic by calculating the ratio of the signal maximum value to the root mean square value; Analyze the phase difference between adjacent sub-signals at the same frequency to generate a phase consistency characteristic; The three types of characteristics of each sub-signal are spliced in the segmentation order to generate a global feature vector.

[0012] A further solution of the present invention is to judge the degree of cable aging through multi-level thresholds and trigger an alarm and positioning process according to the judgment result, including the following steps: The numerical values of each characteristic parameter in the global feature vector are sequentially compared with a preset safety threshold and a fault threshold; If the characteristic parameter does not exceed the preset safety threshold, it is determined that the cable is in a healthy state; If the characteristic parameter exceeds the safety threshold but is lower than the fault threshold, it is determined that the cable is in a mild aging state and a warning signal is generated; If any characteristic parameter exceeds the fault threshold, it is determined that the cable is in a severe aging state, a fault alarm signal is generated and the positioning process is activated.

[0013] A further solution of the present invention is to generate a multi-node signal data set, including the following steps: When a severe aging alarm signal is triggered, activation instructions are sent in combination with multiple monitoring nodes deployed at the head, middle, and end of the cable; After receiving the instruction, each monitoring node synchronously triggers each monitoring node to collect electromagnetic harmonic signals through a unified clock source, and generates a multi-node signal data set aligned by time stamps; the monitoring nodes are deployed at preset positions at the head, middle, and end of the cable.

[0014] A further solution of the present invention is to generate a candidate distance set, including the following steps: For the electromagnetic harmonic signals collected by each monitoring node, extract the harmonic energy distribution characteristics, signal fluctuation characteristics, and phase consistency characteristics; Based on the feature parameter attenuation model and the phase propagation model, calculate the fault distance intervals corresponding to each monitoring node, and generate the candidate distance set by cross - validating and screening the overlapping intervals; the attenuation model is an exponential decay function of the harmonic amplitude with distance, and the phase propagation model is a linear relationship between the phase difference and the distance.

[0015] A further solution of the present invention is to output the final positioning result of the fault point, including the following steps: For each candidate distance in the candidate distance set, calculate its signal attenuation consistency index, phase propagation time - delay matching index, and energy distribution correlation index with each monitoring node respectively; According to the weight coefficients of each set index, sum the three types of index values after weighting to generate a comprehensive score; Select the candidate distance with the highest comprehensive score as the optimal solution, and convert it into the actual fault location according to the physical coordinates of the cable head end.

[0016] In summary, the present invention includes the following beneficial technical effects: 1. By using electromagnetic harmonic sensors to capture the electromagnetic signals during the operation of the cable in real time, combined with signal decomposition and multi - dimensional feature extraction technologies, effectively separate the noise interference and extract the key harmonic features reflecting insulation deterioration, solving the problem of misjudgment caused by environmental noise in traditional methods. The decision - making mechanism based on multi - level thresholds can distinguish healthy states, mild aging, and severe aging, significantly improving the accuracy of aging degree assessment, providing a reliable basis for preventive maintenance, and avoiding power outages caused by sudden failures; 2. Adopt multi - node collaborative acquisition and spatio - temporal synchronization technologies. By analyzing the signal attenuation law and the phase propagation time - delay difference, construct a multi - dimensional constraint model to screen the candidate distance set, overcoming the defect that single - node positioning is easily affected by the non - linearity of signal attenuation. The weighted fusion algorithm further synthesizes the signal consistency, time - delay matching, and energy correlation indexes, especially suitable for complex fault scenarios of long - distance cables, and greatly reducing the cost of manual inspection; 3. The combination of dynamic threshold filtering, segmented signal fine - analysis, and feature parameter adaptive calibration technologies enables the system to adapt to the cable monitoring requirements of different voltage levels, material characteristics, and environmental interferences. The introduction of global feature vectors and multi - level decision models realizes the full - process automatic processing from the original signal to the fault decision, reducing the dependence on manual intervention. At the same time, it supports historical data backtracking and model iterative optimization, providing an expandable technical basis for the intelligent operation and maintenance of the power grid. Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. The accompanying drawings are used to provide a further understanding of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0018] Figure 1 The flowchart in the embodiments of the present application is disclosed.

[0019] Figure 2 The structural diagram in the embodiments of the present application is disclosed. Detailed implementation manners

[0020] In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0021] The following will make a preferred and detailed description of the present invention in conjunction with the attached Figure 1 - Figure 2 drawings.

[0022] Referring to the attached Figure 1 drawings, the present invention proposes a method for monitoring the entire life cycle state of a cable based on the analysis of electromagnetic harmonic characteristics, including the following steps: S1. Obtain the continuous electromagnetic harmonic signals naturally generated during the operation of the cable, and perform discretization processing at a preset sampling frequency to generate a discrete signal sequence; S2. Decompose the discrete signal sequence, separate the noise component and the effective harmonic component, and generate a denoised signal sequence through threshold filtering and reconstruction; S3. Divide the denoised signal sequence into a segmented signal set according to a preset segmented length, and each segmented signal corresponds to a local area of the physical length of the cable; S4. Perform joint time-domain and frequency-domain analysis on each segmented signal in the segmented signal set, extract multi-dimensional characteristic parameters, and generate a global characteristic vector by combining them segment by segment; S5. Based on the global characteristic vector, judge the aging degree of the cable through multiple-level thresholds, and trigger an alarm and positioning process according to the judgment result; S6. When it is determined that the cable is in a severely aging state, activate multiple monitoring nodes along the cable to synchronously collect electromagnetic harmonic signals and generate a multi-node signal data set; S7. Analyze the attenuation law of characteristic parameters and the phase propagation time delay difference of the multi-node signal dataset, calculate the distance range of the fault point relative to each monitoring node, and generate a candidate distance set; S8. Perform weighted fusion on the candidate distance set, screen the optimal solution that meets the signal attenuation consistency, phase propagation time delay matching, and energy distribution correlation, and output the final positioning result of the fault point.

[0023] In one embodiment of the present invention, step S1 includes the following steps: Real-time capture the continuous electromagnetic harmonic signals naturally generated during the operation of the cable through an electromagnetic harmonic sensor. The electromagnetic harmonic sensor discretizes the continuous electromagnetic harmonic signals at a preset sampling frequency to generate a discrete signal sequence.

[0024] Specifically, the electromagnetic harmonic sensor is fixedly arranged at the cable head or a preset monitoring point. The electromagnetic harmonic sensor is internally provided with a broadband induction coil for coupling the alternating electromagnetic field between the cable conductor and the shielding layer to generate continuous electromagnetic harmonic signals corresponding to the cable operation state. Subsequently, the continuous electromagnetic harmonic signals are periodically sampled by an analog-to-digital conversion circuit. The sampling frequency is determined according to the cable rated voltage level and the highest harmonic frequency to ensure that the time-frequency characteristics of the original signal are completely retained in the sampled discrete signal sequence.

[0025] Among them, the continuous electromagnetic harmonic signal refers to the periodic electromagnetic field fluctuation formed in the surrounding space due to the current change during the energized operation of the cable, and its frequency includes the fundamental wave and integer multiple harmonic components. The sampling frequency refers to the set value of the number of signal points collected by the analog-to-digital conversion circuit per second, and this value needs to meet the Nyquist sampling theorem, that is, at least twice the highest frequency component in the cable electromagnetic harmonic signal. The discrete signal sequence refers to the set of digital signals arranged in chronological order after sampling, and each number corresponds to the amplitude of the electromagnetic harmonic signal at the sampling moment.

[0026] Exemplarily, assume that an electromagnetic harmonic sensor of model CYZ-EMD-1000 is installed at the head of a 10 kV cable, the sampling frequency is set to 200 MHz, and the continuous electromagnetic harmonic signals captured by the sensor induction coil are converted into a discrete signal sequence by a 24-bit high-precision analog-to-digital converter. The amplitude range of each data point in the sequence is -5V to +5V, corresponding to the instantaneous electromagnetic field intensity during the cable operation. Technicians verify the integrity of the discrete signal sequence through an oscilloscope to ensure that the time-frequency characteristics of the original signal are completely retained in the sampled discrete signal sequence.

[0027] In one embodiment of the present invention, step S2 includes the following steps: Perform signal decomposition on the discrete signal sequence to separate the noise component and the effective harmonic component; eliminate the noise component through threshold filtering, and reconstruct the denoised signal sequence. The denoised signal sequence retains the electromagnetic harmonic characteristics related to the cable operating state.

[0028] Specifically, the discrete signal sequence is input into a signal decomposition algorithm, and the signal is decomposed into multiple sub-signal components according to different frequency bands. The high-frequency sub-signal components contain random noise introduced by the cable external environment, and the low-frequency sub-signal components contain regular harmonics generated by the cable body. For each sub-signal component, a dynamic threshold is set to judge the boundary between noise and effective harmonics. The sub-signal components above the threshold are determined as noise and set to zero, and the sub-signal components below the threshold are retained as effective harmonics. All processed sub-signal components are merged in reverse according to the original frequency bands to generate the denoised signal sequence.

[0029] Among them, the signal decomposition algorithm includes, but is not limited to, wavelet transform and empirical mode decomposition widely used in the field of signal processing. Wavelet transform is a time-frequency analysis method. By decomposing the signal into wavelet basis functions of different scales and frequencies, and by decomposing the signal and removing the high-frequency noise components, the denoised signal sequence is reconstructed. Empirical mode decomposition is an adaptive signal decomposition method that decomposes a complex signal into several intrinsic mode functions, and each intrinsic mode function represents an inherent vibration mode in the signal.

[0030] Reverse merging refers to the operation of recombining the filtered sub-signal components into a complete signal sequence according to the reverse process during decomposition, and reconstructing the denoised signal that retains the effective harmonic characteristics of the cable.

[0031] The dynamic threshold refers to the filtering critical value adaptively adjusted according to the noise energy. Its value is related to the statistical characteristics of the sub-signal component, and is calculated adaptively according to the noise energy of each sub-signal component, and satisfies the following formula:

[0032] Among them, represents the dynamic threshold. represents the noise standard deviation, which is a statistic measuring the degree of data dispersion and reflects the fluctuation degree of the noise signal. N represents the signal length, which is the number of data points included in the signal.

[0033] In one embodiment of the present invention, step S3 includes the following steps: According to the preset segmentation length, the denoised signal sequence is divided into several segments of continuous sub-signals. Each segment of sub-signal corresponds to a local area of the cable physical length, and a segmented signal set is generated to realize the refined analysis of the local area of the cable state.

[0034] Specifically, determine the physical length corresponding to each sub-signal according to the ratio of the total cable length to the preset segmentation length. Combine the sampling frequency set in step S1 to calculate the number of sampling points included in each sub-signal. Intercept continuous data segments from the denoised signal sequence in chronological order based on the number of sampling points as the unit to generate a set of sub-signals corresponding to the local area of the cable physical length. If the remaining signal length is less than one segment, extend it to a complete segment through zero-padding operation. The zero-padding operation refers to the processing method of filling zero-value data at the end of the signal to meet the segmentation length requirement. The zero-padding operation only fills zero values at the end of the signal, and the effective data segment is automatically recognized during feature extraction to exclude the interference of the zero-padding part on the analysis.

[0035] Among them, the preset segmentation length refers to the total amount of segments preset according to the cable length and the monitoring accuracy requirement, and its value is proportional to the total cable length.

[0036] The number of sampling points refers to the amount of discrete signal data included in each sub-signal, which is determined by the ratio of the physical segmentation length of the cable to the signal propagation speed and satisfies the following formula:

[0037] Among them, N is the number of sampling points of each sub-signal; is the physical length of each section of the cable, determined according to the ratio of the total cable length to the preset segmentation length; is the sampling frequency, obtained by combining step S1; is the signal propagation speed, determined by combining the cable material characteristics.

[0038] Exemplarily, for a 10 kV cable with a total length of 100 meters and a preset segmentation length of M = 10 meters, the physical length value of each section of the cable is 10; the signal propagation speed value is , and the sampling frequency value is . Substitute into the formula to calculate the number of sampling points of each sub-signal:

[0039] It can be seen that one sub-signal is intercepted every 10 points in the denoised signal sequence, and a total of 10 sub-signals are generated.

[0040] In one embodiment of the present invention, step S4 includes the following steps: Perform joint time-domain and frequency-domain analysis on each sub-signal in the segmented signal set, extract multi-dimensional characteristic parameters reflecting the cable operating state, and combine the multi-dimensional characteristic parameters by segment to generate a global feature vector.

[0041] Specifically, each sub-signal performs the following operations respectively: converting the time-domain signal into a frequency-domain signal through fast Fourier transform, calculating the ratio of the sum of the squares of the amplitudes of each harmonic to the sum of the squares of the total harmonic amplitudes, and generating a harmonic energy distribution feature. The fluctuation degree of the signal amplitude is statistically analyzed in the time domain, and a signal fluctuation feature is generated by calculating the ratio of the signal maximum value to the root mean square value. The phase difference between adjacent sub-signals at the same frequency is analyzed for consistency to generate a phase consistency feature.

[0042] Finally, the three types of features of each sub-signal are spliced according to the segmentation order to generate a global feature vector.

[0043] Among them, the harmonic energy distribution feature refers to the proportion of the energy of each harmonic in the total energy of the signal, which is used to quantify the degree of deterioration of the cable insulation performance. An imbalance in the proportion indicates local discharge or carbonization of the insulation material. The signal fluctuation feature refers to the degree to which the amplitude of the time-domain signal deviates from the average level, and an increase in its value reflects that the cable is deformed due to mechanical stress or environmental corrosion.

[0044] The phase consistency feature represents the comparison of the phase differences of adjacent sub-signals at the same frequency. If the difference exceeds the preset threshold, it is determined to be inconsistent. The preset threshold is obtained based on the statistical value of the maximum phase difference of the same type of cable in good condition.

[0045] In one embodiment of the present invention, step S5 includes the following steps: Based on the generated global feature vector, the aging degree of the cable is judged through multi-level thresholds, and the corresponding alarm and positioning processes are triggered according to the judgment results.

[0046] Specifically, the numerical values of each feature parameter in the global feature vector of step S4 are sequentially compared with the preset safety threshold and the fault threshold: If the feature parameter does not exceed the preset safety threshold, it is determined that the cable is in a healthy state; if the feature parameter exceeds the safety threshold but is lower than the fault threshold, it is determined that the cable is in a mild aging state and a warning signal is generated; if any feature parameter exceeds the fault threshold, it is determined that the cable is in a severe aging state, a fault alarm signal is generated and the positioning process is activated.

[0047] Among them, the preset safety threshold corresponds to the normal aging boundary of the cable, which refers to the maximum allowable value of the feature parameter when the cable is allowed to operate normally, and is obtained based on the long-term monitoring data statistics of the same type of cable under standard working conditions. The fault threshold corresponds to the cable fault critical point, which refers to the minimum abnormal value of the feature parameter when the cable suffers irreversible damage, and is calibrated through accelerated aging experiments and fault simulation data.

[0048] The warning signal represents text or audible and visual prompt information indicating that the cable is entering the early aging stage. The fault alarm signal represents an alarm instruction indicating that the cable has serious defects and needs to be processed immediately. The positioning process represents a program trigger instruction for subsequent multi-node collaborative fault positioning.

[0049] Exemplarily, the global feature vector generated in step S4 includes harmonic energy distribution characteristics , signal fluctuation characteristics , phase consistency characteristics degrees; Preset safety threshold: ≤15%, ≤3.5, ≤20 degrees; Fault threshold: ≥25%, ≥5.0, ≥30 degrees; Since exceeds the safety threshold but does not reach the fault threshold, the degree exceeds the safety threshold but does not reach the fault threshold, exceeds the safety threshold but does not reach the fault threshold, it is determined that the cable is in a mild aging state and a warning signal is generated. If , , any one of the characteristic parameters exceeds the preset fault threshold, it is directly determined that the cable is in a serious aging state, a fault alarm signal is generated, and the positioning process is activated.

[0050] In one embodiment of the present invention, step S6 includes the following steps: When it is determined that the cable is in a serious aging state, by activating multiple monitoring nodes deployed along the cable, the electromagnetic harmonic signals of each monitoring node are synchronously collected to generate a multi-node signal dataset.

[0051] Specifically, when the serious aging alarm signal is triggered in step S5, activation instructions are sent in combination with multiple monitoring nodes deployed at the head, middle, and end of the cable. After receiving the instructions, each monitoring node synchronously starts the electromagnetic harmonic sensor based on a unified clock source and collects the electromagnetic harmonic signals at the location of this node in real time according to the sampling frequency of step S1. The sampling data of all nodes are aligned by time stamps to form a multi-node signal dataset.

[0052] Among them, the monitoring nodes are used for local signal acquisition, integrating the hardware devices of sensors, communication modules, and processing units, and are fixedly installed at preset positions along the cable. The unified clock source refers to a timing system that synchronizes the sampling moments of each monitoring node through GPS or Network Time Protocol, ensuring the time alignment of multi-node signals. The multi-node signal dataset refers to a structured data set that includes node numbers, physical positions, timestamps, and discrete signal sequences.

[0053] N monitoring nodes are deployed along the cable. The acquisition time of each monitoring node is T seconds. The total data volume of the multi-node signal dataset is the product of N, T, and the sampling frequency.

[0054] Exemplarily, assume that 5 monitoring nodes are deployed at equal intervals along a 100-meter-long cable (at positions 0 meters, 25 meters, 50 meters, 75 meters, and 100 meters). When a severe aging alarm signal is triggered, each monitoring node starts acquisition at the whole second moment through the GPS synchronization module, and continuously acquires electromagnetic harmonic signals at a frequency of Hz for 10 seconds. The total data volume is .

[0055] After the multi-node signal dataset is aligned by timestamps, the differences in signals of each monitoring node at the same time point can be compared, similar to multiple cameras shooting the same event from different angles and analyzing the object's movement trajectory through frame synchronization.

[0056] In one embodiment of the present invention, step S7 includes the following steps: Based on the multi-node signal dataset generated in step S6, by analyzing the attenuation law of the characteristic parameters and the phase propagation time delay difference of the signals of each monitoring node, calculate the distance range of the fault point relative to each monitoring node, and generate a candidate distance set for subsequent precise positioning.

[0057] Specifically, for the electromagnetic harmonic signals collected by each monitoring node, respectively extract the harmonic energy distribution characteristics, signal fluctuation characteristics, and phase consistency characteristics defined in step S4; construct a characteristic parameter attenuation model according to the distribution parameters of the cable to describe the attenuation law of the characteristic parameters with the cable transmission distance. At the same time, establish a phase propagation model to reflect the linear relationship between the signal phase difference and the cable transmission distance. The characteristic parameters of each monitoring node are input into the above models to calculate the possible distance intervals of the fault point, and the candidate distance set that meets the multi-node consistency constraints is screened through cross-validation.

[0058] Among them, cross-validation refers to overlapping and comparing the distance intervals calculated by different nodes, and only retaining the distance values that simultaneously meet the amplitude attenuation and phase time delay constraints.

[0059] The candidate distance set refers to a set of discrete distance values of all possible fault positions, and each distance value corresponds to a physical coordinate on the cable.

[0060] The characteristic parameter attenuation model is a mathematical expression of the change of characteristic parameters with distance when electromagnetic harmonic signals propagate in a cable. The characteristic parameters are calibrated through cable material characteristics and historical data and satisfy the following formula:

[0061] Among them, represents the harmonic amplitude of node j; represents the harmonic amplitude at the cable head; is the attenuation coefficient, calibrated through cable material parameters and experiments; is the distance from the fault point to node j. The farther the fault point is from the node, the more significant the amplitude attenuation.

[0062] The phase propagation model is a linear relationship between the signal phase difference and the transmission distance, determined by the electromagnetic wave propagation speed of the cable and satisfies the following formula:

[0063] Among them, represents the phase difference of node j; f represents the signal frequency; represents the distance from the fault point to node j; represents the signal propagation speed.

[0064] Exemplarily, for the multi-node signal dataset collected by 5 monitoring nodes in step S6, the 3rd harmonic amplitude of node 1 (0 m) is 5 V, the 3rd harmonic amplitude of node 2 (25 m) is 3.8 V, and the 3rd harmonic amplitude of node 3 (50 m) is 2.9 V; According to the fitting of the characteristic parameter attenuation model , when the distance from the fault point to node 3 is x = 50, the theoretical amplitude should be 2.5 V, and the actual measured value = 2.9 V, indicating that the fault point is downstream of node 3; Combined with the phase propagation model , the calculated distance of the phase difference of node 4 (75 m) is 72 m, and that of node 5 (100 m) is 98 m. After cross-validation, the candidate distance set {70 - 75 m, 95 - 100 m} is screened out.

[0065] In one embodiment of the present invention, step S8 includes the following steps: Based on the weighted fusion of the candidate distance set generated in step S7, the optimal solution that meets multiple constraint conditions is screened out. The multiple constraint conditions include signal attenuation consistency, phase propagation delay matching, and energy distribution correlation, and the final positioning result of the fault point relative to the cable head is output.

[0066] Specifically, for each candidate distance in the candidate distance set, calculate its compliance with the signal attenuation consistency index, phase propagation delay matching index, and energy distribution correlation index of each monitoring node; generate a comprehensive score by weighted summation of the three types of index values according to the set weight coefficients of each index; screen out the candidate distance with the highest comprehensive score as the optimal solution, and convert it to the actual fault location according to the physical coordinates of the cable head end.

[0067] Among them, the signal attenuation consistency index is the matching degree between the theoretical signal attenuation value corresponding to the candidate distance and the actual measurement value. The smaller the value, the more consistent the attenuation law. The phase propagation delay matching index is the deviation degree between the theoretical phase difference corresponding to the candidate distance and the actual phase difference. The smaller the value, the more accurate the delay model. The energy distribution correlation index is the correlation degree between the energy characteristics at the candidate distance and the cable aging model. The larger the value, the more significant the fault characteristics. The weight coefficient is a weighted factor allocated according to the index reliability and is determined by expert experience.

[0068] Exemplarily, the candidate distance set is {70 meters, 75 meters, 95 meters, 98 meters}; 1. Calculate for the 70-meter candidate distance: Signal attenuation consistency index = 0.12 (theoretical attenuation 3.2V, actual measurement 3.1V); Phase delay matching index = 0.08 (theoretical phase difference 15 degrees, actual measurement 14 degrees); Energy correlation index = 0.85 (complies with the characteristics of the aging model); Determined by expert experience, weight coefficients: = 0.3, = 0.5, = 0.2; Comprehensive score = 0.12×0.3 + 0.08×0.5 + 0.85×0.2 = 0.23.

[0069] 2. Calculate for the 98-meter candidate distance: Signal attenuation consistency index = 0.25; Phase delay matching index = 0.05; Energy correlation index = 0.90; Determined by expert experience, weight coefficients: = 0.3, = 0.5, = 0.2; Comprehensive score = 0.25×0.3 + 0.05×0.5 + 0.90×0.2 = 0.265.

[0070] Compare the comprehensive scores of the 70-meter candidate distance and the 98-meter candidate distance, and thus determine that 98 meters is the optimal solution, and output the positioning result that the fault point is 98 meters from the head end.

[0071] See the appendix Figure 2 , the present invention also proposes a cable full-life cycle condition monitoring system based on electromagnetic harmonic characteristic analysis, including the following modules: A signal acquisition module, configured to acquire continuous electromagnetic harmonic signals naturally generated during the operation of the cable, and perform discretization processing at a preset sampling frequency to generate a discrete signal sequence; A signal processing module, which decomposes the discrete signal sequence, separates the noise component and the effective harmonic component, and reconstructs through threshold filtering to generate a denoised signal sequence; A segmented analysis module, which divides the denoised signal sequence into a set of segmented signals according to a preset segmented length, and each segmented signal corresponds to a local area of the physical length of the cable; A feature extraction module, which performs joint time-domain and frequency-domain analysis on each segmented signal in the set of segmented signals, extracts multi-dimensional feature parameters, and generates a global feature vector by combining them segment by segment; An aging determination module, based on the global feature vector, determines the aging degree of the cable through multi-level thresholds, and triggers an alarm and positioning process according to the determination result; A multi-node cooperation module, when it is determined that the cable is in a severely aged state, activates multiple monitoring nodes along the cable to synchronously collect electromagnetic harmonic signals, and generates a multi-node signal data set; A distance calculation module, which analyzes the attenuation law of the feature parameters and the phase propagation time delay difference of the multi-node signal data set, calculates the distance range of the fault point relative to each monitoring node, and generates a set of candidate distances; A positioning optimization module, which performs weighted fusion on the set of candidate distances, screens the optimal solutions that meet the signal attenuation consistency, phase propagation time delay matching, and energy distribution correlation, and outputs the final positioning result of the fault point.

[0072] It should be noted that: for the formulas mentioned above, through the principle of dimensional consistency and mathematical standardization means (such as normalization processing, dimensionless parameter conversion, or unit system unification), physical quantities with different attributes can be translated into dimensionless standard values or superimposable parameters with the same dimension, so as to eliminate the interference of different dimensions on the operation logic, and make the formulas have mathematical operation rationality and objective law adaptability while retaining the original data distribution characteristics. The above is only an exemplary embodiment of the present invention, and the scope of the present invention cannot be limited thereby.

[0073] All or part of the above-mentioned modules can be implemented by software, hardware, and their combinations, support being embedded in the processor of the computer device in hardware form or being independent of it, and also support being stored in the memory of the computer device in software form, so as to facilitate the processor to call and execute the operations corresponding to the above-mentioned modules.

[0074] It should be noted that the human body information (including but not limited to human body device information and personal information, etc.) and data (including but not limited to data for analysis, stored data, and displayed data, etc.) involved in the present invention are all information and data authorized by the human body or fully authorized by all parties. The collection, use, and processing of relevant data require relevant legal standards.

[0075] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; 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 recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for monitoring the full life cycle status of a cable based on the analysis of electromagnetic harmonic characteristics, characterized in that It includes the following steps: S1. Obtain the continuous electromagnetic harmonic signals naturally generated during the operation of the cable, and perform discretization processing at a preset sampling frequency to generate a discrete signal sequence; S2. Decompose the discrete signal sequence, separate the noise component and the effective harmonic component, and generate a denoised signal sequence through threshold filtering and reconstruction; S3. Divide the denoised signal sequence into a set of segmented signals according to the preset segmentation length, and each segmented signal corresponds to a local area of the physical length of the cable; S4. Perform joint time-domain and frequency-domain analysis on each segmented signal in the set of segmented signals, extract multi-dimensional characteristic parameters, and generate a global characteristic vector by segment combination; S5. Based on the global characteristic vector, judge the aging degree of the cable through multi-level thresholds, and trigger an alarm and positioning process according to the judgment result; S6. When it is determined that the cable is in a serious aging state, activate multiple monitoring nodes along the cable to synchronously collect electromagnetic harmonic signals and generate a multi-node signal data set; S7. Analyze the attenuation law of the characteristic parameters and the phase propagation time delay difference of the multi-node signal data set, calculate the distance range of the fault point relative to each monitoring node, and generate a candidate distance set; S8. Perform weighted fusion on the candidate distance set, screen the optimal solution that meets the signal attenuation consistency, phase propagation time delay matching, and energy distribution correlation, and output the final positioning result of the fault point.

2. The method for monitoring the whole life cycle state of a cable based on electromagnetic harmonic characteristic analysis according to claim 1, wherein Generating a discrete signal sequence includes the following steps: An electromagnetic harmonic sensor is arranged at the head end of the cable or a preset monitoring point. The electromagnetic harmonic sensor is internally provided with a broadband induction coil for coupling the alternating electromagnetic field between the cable conductor and the shielding layer to generate continuous electromagnetic harmonic signals; Use an analog-to-digital conversion circuit to periodically sample the continuous electromagnetic harmonic signals to generate the discrete signal sequence, and the sampling frequency is determined according to the rated voltage level of the cable and the highest harmonic frequency.

3. The cable full-life cycle condition monitoring method based on electromagnetic harmonic characteristic analysis according to claim 1, wherein Generating a denoised signal sequence through threshold filtering and reconstruction includes the following steps: Input the discrete signal sequence into a signal decomposition algorithm to decompose the signal into multiple sub-signal components according to different frequency bands; Set a dynamic threshold to judge the boundary between noise and effective harmonics. The sub-signal components higher than the threshold are determined as noise and set to zero, and the sub-signal components lower than the threshold are retained as effective harmonics. The dynamic threshold is adaptively calculated according to the noise standard deviation and signal length of the sub-signal components; All processed sub-signal components are reversely combined according to the original frequency band to generate a denoised signal sequence.

4. The method for monitoring the whole life cycle state of a cable based on the analysis of electromagnetic harmonic characteristics according to claim 1, wherein Dividing the denoised signal sequence into a set of segmented signals according to the preset segmentation length includes the following steps: Determine the physical length according to the ratio of the total length of the cable to the preset segmentation length; Calculate the number of sampling points included in each segmented signal in combination with the preset sampling frequency; Taking the number of sampling points as the unit, intercept continuous data segments from the denoised signal sequence in chronological order to generate a set of segmented signals corresponding to the local area of the physical length of the cable.

5. The method for monitoring the cable full life cycle state based on electromagnetic harmonic characteristic analysis according to claim 3, wherein Intercepting continuous data segments from the denoised signal sequence in chronological order includes the following steps: If the remaining signal length is less than one segment, it is extended to a complete segment through zero-padding operation; the zero-padding operation refers to the process of filling zero-valued data at the end of the signal to meet the segment length requirement. The zero-padding operation only fills zero values at the end of the signal, and the effective data segment is automatically recognized during feature extraction to exclude the interference of the zero-padded part on the analysis.

6. The method for monitoring the whole life cycle state of a cable based on electromagnetic harmonic characteristic analysis according to claim 1, wherein Generate a global feature vector, including the following steps: Each sub-signal in the segmented signal set converts the time-domain signal into a frequency-domain signal through fast Fourier transform, calculates the ratio of the sum of the squares of the amplitudes of each harmonic to the sum of the squares of the total harmonic amplitudes, and calculates the harmonic energy distribution characteristics; Statistically analyze the fluctuation degree of the signal amplitude in the time domain, and generate a signal fluctuation characteristic by calculating the ratio of the signal maximum value to the root mean square value; Analyze the phase difference between adjacent sub-signals at the same frequency to generate a phase consistency characteristic; The three types of characteristics of each sub-signal are concatenated in the segment order to generate a global feature vector.

7. The cable full-life cycle status monitoring method based on electromagnetic harmonic characteristic analysis according to claim 1, wherein Judge the degree of cable aging through multi-level thresholds, and trigger an alarm and location process according to the judgment result, including the following steps: The numerical values of each characteristic parameter in the global feature vector are sequentially compared with the preset safety threshold and the fault threshold; If the characteristic parameter does not exceed the preset safety threshold, it is determined that the cable is in a healthy state; If the characteristic parameter exceeds the safety threshold but is lower than the fault threshold, it is determined that the cable is in a mild aging state and a warning signal is generated; If any characteristic parameter exceeds the fault threshold, it is determined that the cable is in a severe aging state, a fault alarm signal is generated and the location process is activated.

8. The cable full-life cycle status monitoring method based on electromagnetic harmonic characteristic analysis according to claim 1, characterized in that Generate a multi-node signal data set, including the following steps: When a severe aging alarm signal is triggered, send activation instructions in combination with multiple monitoring nodes deployed at the head, middle, and end of the cable; After receiving the instructions, each monitoring node synchronously triggers each monitoring node to collect electromagnetic harmonic signals through a unified clock source, and generates a multi-node signal data set aligned by time stamps; the monitoring nodes are deployed at preset positions at the head, middle, and end of the cable.

9. The method for monitoring the whole life cycle state of a cable based on electromagnetic harmonic characteristic analysis according to claim 6, wherein Generate a candidate distance set, including the following steps: For the electromagnetic harmonic signals collected by each monitoring node, extract the harmonic energy distribution characteristics, signal fluctuation characteristics, and phase consistency characteristics; Based on the characteristic parameter attenuation model and the phase propagation model, calculate the fault distance interval corresponding to each monitoring node, and generate the candidate distance set by cross-validating and screening the overlapping intervals; the attenuation model is an exponential attenuation function of the harmonic amplitude with respect to the distance, and the phase propagation model is a linear relationship between the phase difference and the distance.

10. The method for monitoring the whole life cycle state of a cable based on electromagnetic harmonic characteristic analysis according to claim 9, characterized in that, Output the final location result of the fault point, including the following steps: For each candidate distance in the candidate distance set, calculate its signal attenuation consistency index, phase propagation time delay matching index, and energy distribution correlation index with each monitoring node; According to the weight coefficients of each set index, weight and sum the three types of index values to generate a comprehensive score; Select the candidate distance with the highest comprehensive score as the optimal solution, and convert it into the actual fault location according to the physical coordinates of the cable head.

Images