Cable external damage prevention method and system based on non-uniform FDM-NLFM (Frequency Division Multiplexing-Non-Linear Frequency Modulation)

Abstract

The invention relates to a non-uniform FDM-NLFM-based cable external damage prevention method and system, and relates to the field of data processing, and the method comprises the steps: through optimizing a fitness function, population updating times and updating influence regulatory factors, comprehensively considering the influence of power attenuation and phase distortion on signal propagation, remarkably improving the adaptability of signals in different soil environments, and improving the signal transmission efficiency. The signal can be stably propagated in a complex underground environment; and the non-uniform FDM-NLFM signal parameters are optimized, so that the signal can better resist the influence caused by the change of soil characteristics in the propagation process. Selecting the signal parameter with the minimum fitness value to ensure that the signal has minimum attenuation and distortion in the propagation process; the signal parameters are continuously adjusted to gradually approach to the optimal solution, and the finally obtained signal parameters can keep stable propagation characteristics in different soil partitions through multiple times of loop optimization, so that the technical problem of low precision in the signal propagation process is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a cable external damage prevention method and system based on non-uniform FDM-NLFM. Background Art

[0002] As an important carrier of power transmission, the safe operation of cables is of vital importance. During the cable laying process, especially underground laying, cables are easily damaged by external factors, such as construction excavation, natural disasters, etc. In order to ensure the safe operation of cables, traditional cable damage prevention methods mainly rely on monitoring the cable status and analyzing the cable status by detecting the reflected signal of the injected signal.

[0003] Traditional cable damage prevention methods have certain limitations. The soil environment of underground cables is complex and diverse, and soil characteristics vary significantly in different regions. Such differences will cause signals to be affected to varying degrees during propagation, which in turn affects the accuracy of detection results. Traditional methods often ignore such regional differences, resulting in low adaptability of the injected signal to the actual scene, making it difficult to accurately reflect the true state of the cable. Summary of the invention

[0004] In view of the technical problem that the signals emitted by the cables laid underground are affected by regional differences, the reflected signals are affected and the cable status cannot be accurately analyzed, the present invention provides a cable external damage prevention method and system based on non-uniform FDM-NLFM to solve the problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, a cable external damage prevention method based on non-uniform FDM-NLFM comprises:

[0007] Receive soil detection information in the cable laying area, divide the laying soil into zones, and sort the zones based on the signal injection point and the detection direction to obtain a soil zone sequence;

[0008] Traversing the soil partition sequence to perform mode statistics of soil conductivity of historical samples to obtain a soil conductivity sequence;

[0009] Traversing the soil partition results to perform mode statistics of historical sample dielectric constants to obtain a soil dielectric constant sequence;

[0010] Randomly generate a number of non-uniform FDM-NLFM signal parameters, combine the soil conductivity sequence and the soil dielectric constant sequence to perform signal propagation simulation, and obtain a number of power attenuation coefficients and a number of phase distortion coefficients;

[0011] Based on the several power attenuation coefficients and the several phase distortion coefficients, the several non-uniform FDM-NLFM signal parameters are optimized to obtain target non-uniform FDM-NLFM signal parameters for cable anti-external damage detection.

[0012] In a second aspect, the present invention provides a cable external damage prevention system based on non-uniform FDM-NLFM, comprising:

[0013] The soil partition sequence module is used to receive soil detection information of the cable laying area, partition the laying soil, and sort it based on the signal injection point as the starting point and the detection direction to obtain the soil partition sequence;

[0014] A soil conductivity sequence module, used to traverse the soil partition sequence to perform mode statistics of soil conductivity of historical samples to obtain a soil conductivity sequence;

[0015] A soil dielectric constant sequence module is used to traverse the soil partition results to perform mode statistics of historical sample dielectric constants to obtain a soil dielectric constant sequence;

[0016] The power attenuation and phase distortion coefficient module is used to randomly generate a number of non-uniform FDM-NLFM signal parameters, combine the soil conductivity sequence and the soil dielectric constant sequence to perform signal propagation simulation, and obtain a number of power attenuation coefficients and a number of phase distortion coefficients;

[0017] The cable anti-external damage detection module is used to optimize the several non-uniform FDM-NLFM signal parameters according to the several power attenuation coefficients and the several phase distortion coefficients to obtain target non-uniform FDM-NLFM signal parameters for cable anti-external damage detection.

[0018] The beneficial effects of the present invention are as follows: compared with the traditional method of analyzing the cable status based on the reflected signal by detecting the injected non-uniform FDM-NLFM composite signal, this method is greatly affected by regional differences. By introducing the optimization fitness function, the present invention can comprehensively consider the influence of power attenuation and phase distortion on signal propagation. The fitness function not only considers the power attenuation coefficient and the phase distortion coefficient, but also introduces the number of population updates and the update impact adjustment factor, making the optimization process of the signal parameters more dynamic and adaptable. This optimization method that comprehensively considers multiple factors significantly improves the adaptability of the signal in different soil environments and ensures that the signal can be stably propagated in a complex underground environment; further, the present invention optimizes the non-uniform FDM-NLFM signal parameters so that the signal can better resist the influence of changes in soil characteristics during the propagation process. By selecting the signal parameters with the smallest fitness value, it is possible to ensure that the signal has the smallest attenuation and distortion during the propagation process, thereby improving the accuracy of the detection results; further, by guiding variation and cyclic optimization, the present invention can continuously adjust the signal parameters so that they gradually approach the optimal solution. Through multiple cyclic optimizations, the signal parameters finally obtained can maintain stable propagation characteristics in different soil partitions, reducing the uncertainty in the signal propagation process, thereby achieving the technical effect of improving efficiency while taking into account the cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of a process flow of a cable external damage prevention method based on non-uniform FDM-NLFM provided by the present invention;

[0020] Figure 2 A schematic structural diagram of a cable external damage protection system based on non-uniform FDM-NLFM provided by the present invention.

[0021] Figure numerals: soil partition sequence module 11, soil conductivity sequence module 12, soil dielectric constant sequence module 13, power attenuation and phase distortion coefficient module 14, cable anti-external damage detection module 15. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0023] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0024] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0025] Embodiment 1:

[0026] like Figure 1 As shown, the embodiment of the present invention provides a cable external damage prevention method based on non-uniform FDM-NLFM, and the execution steps include:

[0027] S10: receiving soil detection information of the cable laying area, partitioning the laying soil, and sorting the soil based on the signal injection point as the starting point and the detection direction to obtain a soil partition sequence;

[0028] Specifically, soil detection information includes soil composition information, soil temperature information and soil moisture information, which are obtained through sensors or field detection and are used to reflect the physical and chemical properties of the soil in the cable laying area.

[0029] Furthermore, multi-level hierarchical cluster analysis is a data mining method that gradually aggregates similar samples into clusters by calculating the similarity or distance between samples. Multi-level hierarchical cluster analysis refers to clustering at multiple levels, first performing rough clustering, and then performing more detailed clustering in each rough cluster. The soil partition set is a set of several soil partitions obtained through cluster analysis, and the soil characteristics (composition, temperature and humidity) in each partition are relatively consistent. Soil partition sequence: The soil partition set is sorted according to the signal injection point and the detection direction, which is used for subsequent signal propagation simulation and optimization.

[0030] Details are as follows:

[0031] Further, according to the soil detection information of the cable laying area, the laying soil is partitioned, and the signal injection point is used as the starting point and the detection direction is used to sort the soil partition sequence, and the execution of step S10 includes:

[0032] S11: The soil detection information includes soil composition information, soil temperature information and soil moisture information;

[0033] For example, assume that the cable laying area is a 100m x 100m square area, the signal injection point is located at the lower left corner of the area (coordinate (0, 0)), and the detection direction is from left to right and from bottom to top. Soil detection information is collected through the sensor network, including soil composition (content of component A%), soil temperature (℃) and soil moisture (%RH).

[0034] Receive soil detection information in the cable laying area. In the cable laying area, collect soil detection information through the sensor network. Assuming that the sensors are distributed at the following points, the collected data is shown in the following table:

[0035]

[0036] S12: configuring soil composition deviation threshold, soil temperature deviation threshold and soil moisture deviation threshold;

[0037] Further, the soil composition deviation threshold, the soil temperature deviation threshold and the soil moisture deviation threshold are configured, and the execution of step S12 includes:

[0038] S121: extracting attributes to be analyzed from soil composition information, soil temperature information, and soil moisture information;

[0039] S122: taking the attribute to be analyzed as the only variable, setting the set of attributes not to be analyzed as a fixed value, taking conductivity and dielectric constant as follow-up variables, and collecting a set of recorded values ​​of the attribute to be analyzed, a set of recorded values ​​of conductivity, and a set of recorded values ​​of dielectric constant;

[0040] S123: performing pairwise fluctuation calculations on the attribute record value set to be analyzed, the conductivity record value set, and the dielectric constant record value set to obtain a deviation set of attribute record values ​​to be analyzed, a deviation set of conductivity record values, and a deviation set of dielectric constant record values;

[0041] S124: Based on the conductivity record value deviation set, a first selected set of attribute record value deviations to be analyzed whose conductivity record value deviations are greater than or equal to a conductivity deviation threshold is selected from the attribute record value deviation set to be analyzed;

[0042] S125: Based on the dielectric constant record value deviation set, screening a first selected set of attribute record value deviations to be analyzed whose dielectric constant record value deviations are greater than or equal to a dielectric constant deviation threshold from the attribute record value deviation set to be analyzed;

[0043] S126: After performing outlier removal on the first selected set of attribute record value deviations to be analyzed and the first selected set of attribute record value deviations to be analyzed, respectively, extract the minimum value of the first retained attribute record value deviation to be analyzed and the second retained attribute record value deviation to be analyzed, set it as the attribute deviation threshold to be analyzed, and add it into the soil composition deviation threshold, the soil temperature deviation threshold and the soil moisture deviation threshold.

[0044] For example, the attributes to be analyzed are extracted from the collected soil detection information. Assume that we select soil composition (content of component A%), soil temperature (℃) and soil moisture (%RH) as the attributes to be analyzed, and collect the corresponding conductivity and dielectric constant as follow-up variables. The collected data is shown in the following table:

[0045]

[0046] The attribute to be analyzed is the only variable, the non-to-be-analyzed attribute is fixed, and a set of recorded values ​​is collected. The soil composition is the attribute to be analyzed, and the soil temperature and humidity are set as fixed values ​​(for example, the soil temperature is 20°C and the soil humidity is 60%RH), and the corresponding soil composition, conductivity and dielectric constant recorded value sets are collected. The sorted data is shown in the following table:

[0047]

[0048] Perform pairwise fluctuation calculations to obtain deviation sets and fluctuation calculations:

[0049] Calculate the pairwise fluctuation deviation of soil composition, conductivity and dielectric constant. For example, the deviation between point 1 and point 3 is calculated as follows:

[0050] Deviation of soil composition: |31 - 30| = 1%; Deviation of conductivity: |11 - 10| = 1 mS / m; Deviation of dielectric constant: |5.2 - 5| = 0.2.

[0051] The deviations between all points are calculated to obtain the following deviation sets: soil composition deviation set: {1%, 1%, 0%}; conductivity deviation set: {1 mS / m, 1 mS / m, 0 mS / m}; dielectric constant deviation set: {0.2, 0.2, 0.0}.

[0052] Filter the recorded values ​​whose deviation is greater than or equal to the threshold, assuming that the conductivity deviation threshold is 1 mS / m and the dielectric constant deviation threshold is 0.2. Filter the recorded values ​​whose conductivity and dielectric constant deviations are greater than or equal to the threshold:

[0053] The set of recorded values ​​with conductivity value deviation greater than or equal to 1 mS / m: {1 mS / m, 1 mS / m};

[0054] The set of recorded values ​​with a dielectric constant recorded value deviation greater than or equal to 0.2: {0.2, 0.2};

[0055] Corresponding soil component record value deviation set: Selected set of first attribute record value deviation to be analyzed: {1%, 1%}.

[0056] Delete outliers, extract the minimum value as the deviation threshold, and delete outliers from the filtered record value set. Assuming there are no outliers, directly extract the minimum value as the deviation threshold:

[0057] Soil composition deviation threshold: 1%; soil temperature deviation threshold: 2°C (assuming pre-set); soil moisture deviation threshold: 2%RH (assuming pre-set).

[0058] Automatically configure the deviation threshold. To achieve automatic configuration of the deviation threshold, you can use the following methods:

[0059] Data fitting and regression analysis: Use a multivariate linear regression model to analyze the relationship between soil composition, temperature, humidity, conductivity and dielectric constant. According to literature research, factors such as soil moisture content, soil bulk density, soil temperature, and soil salinity have a significant effect on soil dielectric properties. A similar regression model can be established to determine the degree of influence of each factor on conductivity and dielectric constant.

[0060] Threshold determination based on fluctuation analysis: Through fluctuation calculation, determine the influence of the minimum deviation value of each attribute on conductivity and dielectric constant. For example, according to literature research, the changes in soil conductivity and dielectric constant are closely related to changes in soil composition, moisture and temperature. A fluctuation range can be set, and when the attribute deviation exceeds this range, the change in conductivity or dielectric constant is significant.

[0061] Automatic threshold calculation: Calculate the minimum deviation threshold for each attribute based on the fluctuation analysis results. For example, if the conductivity changes significantly when the minimum deviation of soil composition is 1%, the soil composition deviation threshold is set to 1%. Similarly, set the corresponding deviation threshold based on the fluctuation analysis results of soil temperature and moisture.

[0062] Through the above steps, we obtained the following deviation thresholds: soil composition deviation threshold: 1%; soil temperature deviation threshold: 2°C; soil moisture deviation threshold: 2%RH.

[0063] Through the above embodiments, it is possible to realize automatic configuration of deviation thresholds of soil composition, temperature and humidity. The setting of these thresholds is based on the actual collected data, and their rationality and effectiveness are ensured through fluctuation calculation and screening. The final deviation threshold will provide a scientific basis for soil zoning, thereby improving the accuracy and reliability of cable anti-external damage detection.

[0064] S13: Perform a multi-level hierarchical clustering analysis based on the soil composition deviation threshold, the soil temperature deviation threshold and the soil moisture deviation threshold, in combination with the soil composition information, the soil temperature information and the soil moisture information, to obtain a soil partition set, wherein the soil composition, soil temperature and humidity of any partition are consistent.

[0065] Furthermore, a multi-level hierarchical clustering analysis is performed based on the thresholds obtained in the above embodiment, as follows:

[0066] Calculate the similarity between points:

[0067] Use Euclidean distance or other similarity measurement methods to calculate the similarity between each point. The similarity calculation formula is:

[0068]

[0069] Among them, x, y, z represent the normalized values ​​of soil composition, temperature and humidity respectively, and D1 represents the distance.

[0070] Initialize clustering: Initialize each point as a separate cluster.

[0071] Merge the most similar clusters: At each step, find the two most similar clusters and merge them until all points are merged into one cluster.

[0072] Apply Deviation Threshold: During the merging process, check whether the merged clusters meet the deviation threshold. If the deviation of a point in a cluster on a certain attribute exceeds the threshold, stop merging the cluster.

[0073] For example, according to the thresholds obtained in the above embodiment: soil composition deviation threshold: 1%; soil temperature deviation threshold: 2°C; soil moisture deviation threshold: 2%RH, a multi-level hierarchical clustering analysis is performed, as follows:

[0074] Calculate the distance between points:

[0075] For example, the distance between point 1 and point 2 is calculated as follows:

[0076]

[0077] Initialize clustering, with each point as a separate cluster:

[0078] Cluster 1: {1}; Cluster 2: {2}; Cluster 3: {3}; Cluster 4: {4}; Cluster 5: {5}; Cluster 6: {6}; Cluster 7: {7}; Cluster 8: {8}; Cluster 9: {9}.

[0079] Merge the most similar clusters:

[0080] Find the two closest clusters and merge them. For example, the distance between point 1 and point 3 is:

[0081]

[0082] Merge cluster 1 and cluster 3: Cluster 1: {1, 3}. Continue merging other clusters until all points are merged into one cluster.

[0083] Apply a deviation threshold and check whether the attribute deviation of each point in each cluster exceeds the threshold. For example, check points 1 and 3 in cluster 1:

[0084] Soil composition deviation: |31 - 30| = 1% (less than the threshold of 1%); soil temperature deviation: |20 - 20| = 0℃ (less than the threshold of 2℃); soil moisture deviation: |61 - 60| = 1%RH (less than the threshold of 2%RH).

[0085] Since all deviations are smaller than the threshold, cluster 1 can continue to merge other points.

[0086] After multi-level hierarchical clustering analysis, the following partition sets are finally obtained:

[0087] Partition 1: includes point 1, point 3, and point 8; Partition 2: includes point 2, point 6, and point 9; Partition 3: includes point 4, point 5, and point 7.

[0088] S14: Sort the soil partition set based on the signal injection point as the starting point and the detection direction to obtain a soil partition sequence.

[0089] Exemplarily, in order to more accurately describe the soil characteristics of each partition, the average soil composition, temperature and humidity in each partition are calculated based on the data point coordinates and soil characteristics collected in the above embodiment. The calculation formula is the average value of the corresponding soil composition data in each partition. The following is the calculation result of the soil composition of each partition:

[0090] Partition 1: includes point 1, point 3, and point 8.

[0091] Average soil composition: (30+31+30) / 3=30.33%; average soil temperature: (20+20+20) / 3=20℃; average soil moisture: (60+61+60) / 3=60.33%RH.

[0092] Similarly, partition 2 includes points 2, 6, and 9.

[0093] Average soil composition: 32%; average soil temperature: 21.33℃; average soil moisture: 62%RH.

[0094] Partition 3: includes points 4, 5, and 7.

[0095] Average soil composition: 31.33%; average soil temperature: 22°C; average soil moisture: 62.67%RH.

[0096] According to the above-mentioned refined soil characteristics, a number is assigned to each sub-area to more clearly indicate the soil characteristics of each sub-area, as follows:

[0097] According to the signal injection point (0, 0) and the detection direction (from left to right, from bottom to top), the partitions are sorted. The sorted partition sequence is: Partition 1, Partition 3, Partition 2.

[0098] The sorted partition results are shown in the following table:

[0099]

[0100] The soil partition sequence obtained in the above way can accurately reflect the differences in soil characteristics at different locations in the cable laying area, and provide a partition model that is highly matched with the actual scenario for subsequent signal propagation simulation, thereby improving the accuracy and reliability of signal propagation simulation.

[0101] S20: traversing the soil partition sequence to perform mode statistics of soil conductivity of historical samples to obtain a soil conductivity sequence.

[0102] Further, according to traversing the soil partition sequence and performing mode statistics of soil conductivity of historical samples to obtain a soil conductivity sequence, executing step S20 includes:

[0103] S21: extracting a first soil partition according to the soil partition sequence, wherein the first soil partition has a soil component identifier, a soil temperature identifier, and a soil moisture identifier;

[0104] S22: constructing a first index constraint rule according to the soil component identifier, the soil temperature identifier and the soil moisture identifier;

[0105] S23: Retrieving a first soil conductivity detection value set of a first historical sample set that satisfies the first index constraint rule, wherein the first historical sample set has a soil component label set, a soil temperature label set, and a soil moisture label set;

[0106] S24: constructing a second index constraint rule set according to the soil component label set, the soil temperature label set and the soil moisture label set, and respectively retrieving a plurality of groups of second soil conductivity detection values ​​of a plurality of groups of second historical samples satisfying the second index constraint rule set;

[0107] S25: performing mode aggregation on the first soil conductivity detection value set and the plurality of second soil conductivity detection value sets to obtain first partition soil conductivity, and adding the first partition soil conductivity to the soil conductivity sequence.

[0108] Exemplarily, the soil partition set obtained by the multi-level hierarchical clustering analysis in the above embodiment has been sorted according to the signal injection point and the detection direction, and each partition is processed in turn according to the sorted soil partition sequence:

[0109] Process partition 1 and extract the first soil partition:

[0110] Partition 1: includes points 1, 3, and 8; soil composition identification: 30.33%; soil temperature identification: 20°C; soil moisture identification: 60.33%RH.

[0111] Construct the first index constraint rule: soil composition identifier: 30.33%; soil temperature identifier: 20°C; soil moisture identifier: 60.33%RH.

[0112] Retrieve the first soil conductivity detection value set of the first historical sample set: search the historical sample set to find the historical samples that meet the above index constraint rules.

[0113] Assume that the historical sample set is as follows: Sample 1: soil composition 30%, temperature 20℃, humidity 60%RH, conductivity 10mS / m; Sample 2: soil composition 31%, temperature 20℃, humidity 61%RH, conductivity 11 mS / m; Sample 3: soil composition 30%, temperature 20℃, humidity 60%RH, conductivity 10 mS / m.

[0114] The first soil conductivity detection value set: {10, 11, 10}.

[0115] Constructing a second index constraint rule set: constructing a second index constraint rule set according to the soil composition label set, the soil temperature label set and the soil moisture label set.

[0116] Assume that the second index constraint rule set is as follows: Rule 1: soil composition 30%, temperature 20°C, humidity 60%RH; Rule 2: soil composition 31%, temperature 20°C, humidity 61%RH.

[0117] Retrieve multiple sets of second soil conductivity detection values ​​for multiple sets of second historical samples:

[0118] Retrieve historical samples that meet rule 1: Sample 4: soil composition 30%, temperature 20°C, humidity 60%RH, conductivity 10mS / m; Sample 5: soil composition 30%, temperature 20°C, humidity 60%RH, conductivity 10 mS / m.

[0119] The second soil conductivity detection value set 1: {10, 10}.

[0120] Retrieve historical samples that meet Rule 2: Sample 6: soil composition 31%, temperature 20°C, humidity 61%RH, conductivity 11mS / m; Sample 7: soil composition 31%, temperature 20°C, humidity 61%RH, conductivity 11 mS / m.

[0121] The second soil conductivity detection value set 2: {11, 11}.

[0122] Perform mode aggregation to merge the first soil conductivity detection value set and multiple sets of second soil conductivity detection values:

[0123] The merged conductivity set is: {10, 11, 10, 10, 10, 11, 11}; Calculate the mode: The mode is 10 mS / m (the most common one).

[0124] Add to soil conductivity sequence: Soil conductivity of first partition: 10 mS / m; Soil conductivity sequence:

[10] .

[0125] Repeat the above steps to process partition 2:

[0126] Partition 2: includes points 2, 6, and 9; soil composition identification: 32%; soil temperature identification: 21.33°C; soil moisture identification: 62%RH.

[0127] Assume that the historical sample set is as follows: Sample 8: soil composition 32%, temperature 21°C, humidity 62%RH, conductivity 12mS / m; Sample 9: soil composition 32%, temperature 22°C, humidity 62%RH, conductivity 12 mS / m; Sample 10: soil composition 32%, temperature 21°C, humidity 62%RH, conductivity 12 mS / m.

[0128] After mode aggregation, the soil conductivity of partition 2 is 12 mS / m; soil conductivity series: [10, 12].

[0129] Repeat the above steps to process partition 3:

[0130] Partition 3: includes points 4, 5, and 7; soil composition identification: 31.33%; soil temperature identification: 22°C; soil moisture identification: 62.67%RH.

[0131] Assume that the historical sample set is as follows: Sample 11: soil composition 31%, temperature 22°C, humidity 63%RH, conductivity 11mS / m; Sample 12: soil composition 33%, temperature 23°C, humidity 64%RH, conductivity 13 mS / m; Sample 13: soil composition 31%, temperature 21°C, humidity 61%RH, conductivity 11 mS / m.

[0132] After mode aggregation, the soil conductivity of partition 3 is 11 mS / m; soil conductivity sequence: [10, 12, 11].

[0133] Through the above steps, the soil partition sequence was successfully traversed, and the mode statistics of the soil conductivity of historical samples were performed, and finally the soil conductivity sequence was obtained. The results are shown in the following table:

[0134]

[0135] S30: traversing the soil partition results to perform mode statistics of historical sample dielectric constants to obtain a soil dielectric constant sequence;

[0136] Further, the soil partition results are traversed to perform mode statistics of the historical sample dielectric constants, and the method for obtaining the soil dielectric constant sequence executes step S20, and the results are shown in the following table:

[0137]

[0138] Through the above steps, the soil partition sequence is traversed and the mode statistics of the soil conductivity of the historical samples are performed to obtain the soil conductivity sequence. This method can ensure that the soil conductivity and dielectric constant of each partition are representative, providing more accurate basic data for subsequent signal propagation simulation and cable anti-break detection.

[0139] S40: randomly generating a number of non-uniform FDM-NLFM signal parameters, combining the soil conductivity sequence and the soil dielectric constant sequence to perform signal propagation simulation, and obtaining a number of power attenuation coefficients and a number of phase distortion coefficients;

[0140] Further, according to randomly generating a plurality of non-uniform FDM-NLFM signal parameters, combining the soil conductivity sequence and the soil dielectric constant sequence to perform signal propagation simulation, a plurality of power attenuation coefficients and a plurality of phase distortion coefficients are obtained, and executing step S40 includes:

[0141] S41: Any one of the plurality of non-uniform FDM-NLFM signal parameters includes an FDM frequency band parameter set, an FDM propagation frequency parameter, an FDM propagation power set, an NLFM propagation frequency curve, and an NLFM propagation power parameter;

[0142] Specifically, the FDM (Frequency Division Multiplexing) frequency band parameter set refers to the frequency range of each subcarrier in the frequency division multiplexing signal, which determines the distribution of the signal in the frequency domain; the FDM propagation frequency parameter refers to the center frequency of each subcarrier in the frequency division multiplexing signal, which is used to define the frequency position of the signal; the FDM propagation power set refers to the propagation power of each subcarrier in the frequency division multiplexing signal, and these power values ​​determine the energy distribution of the signal at different frequencies; the NLFM (Non-Linear Frequency Modulation) propagation frequency curve refers to the frequency change curve of the nonlinear frequency modulation signal over time, which defines the frequency modulation characteristics of the signal; the NLFM propagation power parameter refers to the propagation power of the nonlinear frequency modulation signal, which determines the overall energy level of the signal.

[0143] S42: performing unit distance signal propagation simulation according to the FDM frequency band parameter set, the FDM propagation frequency parameter, the FDM propagation power set, the NLFM propagation frequency curve and the NLFM propagation power parameter, in combination with the soil conductivity sequence and the soil dielectric constant sequence, to obtain a unit distance power attenuation rate sequence and a unit distance phase distortion rate sequence;

[0144] Further, according to the FDM frequency band parameter set, the FDM propagation frequency parameter, the FDM propagation power set, the NLFM propagation frequency curve and the NLFM propagation power parameter, the soil conductivity sequence and the soil dielectric constant sequence are combined to perform unit distance signal propagation simulation to obtain a unit distance power attenuation rate sequence and a unit distance phase distortion rate sequence, and executing step S42 includes:

[0145] S421: configuring FDM frequency band parameter set recording data, FDM propagation frequency parameter recording data, FDM propagation power set recording data, NLFM propagation frequency curve recording data, NLFM propagation power parameter recording data, unit distance power attenuation rate identification data and unit distance phase distortion rate identification data;

[0146] S422: Based on the unit distance power attenuation rate identification data as supervision, the FDM frequency band parameter set recording data, FDM propagation frequency parameter recording data, FDM propagation power set recording data, NLFM propagation frequency curve recording data, and NLFM propagation power parameter recording data are used as input to train the first BP neural network to obtain a unit distance power attenuation rate simulation channel.

[0147] Specifically, the BP neural network training model is a multi-layer feedforward neural network, which is trained by the back propagation algorithm. The steps are as follows:

[0148] Initialize the network structure and determine the network structure.

[0149] The number of nodes in the input layer is determined by the number of features of the input data. The number of nodes in the hidden layer is determined by the complexity of the problem. The number of nodes in the output layer is determined by the number of features of the output data.

[0150] Initialize weights and biases. Weights: Usually randomly initialized to small non-zero values ​​to avoid all neurons having the same output; Bias: Usually initialized to 0 or a small non-zero value.

[0151] Forward propagation, input data: pass the input data X to the input layer; calculate the hidden layer output: for each hidden layer node j, calculate its input zj: .

[0152] Apply activation function σ (such as Sigmoid, ReLU, etc.): .

[0153] Calculate the output layer output: For each output layer node k, calculate its input zk: , apply activation function σ (such as Sigmoid, ReLU, etc.): .

[0154] Calculate the loss function. Commonly used loss functions include mean square error (MSE) and cross entropy loss (Cross-Entropy).

[0155] Mean Square Error:

[0156] Cross Entropy Loss:

[0157] The output of the forward propagation Compare with the true value y and calculate the loss value. N represents the training sample.

[0158] Back propagation, then adjust the weights of the model parameters when the loss value is greater than the loss threshold;

[0159] Iterative training: Repeat the process of forward propagation, loss calculation, back propagation, and weight update until the loss value converges or reaches the preset number of iterations to obtain a simulated channel of power attenuation rate per unit distance.

[0160] Through the above model training, the BP neural network can learn the complex relationship between input data and output data, thereby achieving accurate prediction of signal propagation characteristics. This method performs well in dealing with complex nonlinear problems and is suitable for signal propagation simulation in cable anti-break detection.

[0161] S423: Based on the unit distance phase distortion rate identification data as supervision, the FDM frequency band parameter set record data, the FDM propagation frequency parameter record data, the FDM propagation power set record data, the NLFM propagation frequency curve record data, and the NLFM propagation power parameter record data are used as input to train the second BP neural network to obtain a unit distance phase distortion rate simulation channel;

[0162] S424: Merge the input layers of the unit distance power attenuation rate simulation channel and the unit distance phase distortion rate simulation channel, generate a unit distance signal propagation simulation function, perform unit distance signal propagation simulation, and obtain a unit distance power attenuation rate sequence and a unit distance phase distortion rate sequence.

[0163] Exemplarily, the soil conductivity series and soil dielectric constant series obtained previously are used:

[0164] Soil conductivity series: [10, 12, 11] mS / m; soil dielectric constant series: [5.0, 5.5, 5.3].

[0165] Configure signal parameter recording data and randomly generate several non-uniform FDM-NLFM signal parameters. Each parameter includes:

[0166] FDM frequency band parameter set; FDM propagation frequency parameter; FDM propagation power set; NLFM propagation frequency curve; NLFM propagation power parameter.

[0167] Assume that three sets of non-uniform FDM-NLFM signal parameters are randomly generated, as follows:

[0168]

[0169] Configure the following data for BP neural network training:

[0170] FDM frequency band parameter set record data; FDM propagation frequency parameter record data; FDM propagation power set record data; NLFM propagation frequency curve record data; NLFM propagation power parameter record data; unit distance power attenuation rate identification data; unit distance phase distortion rate identification data. Assume that the historical data is as shown in the following table:

[0171]

[0172] Train the first BP neural network (power attenuation rate), input data: FDM frequency band parameter set; FDM propagation frequency parameter; FDM propagation power set; NLFM propagation frequency curve; NLFM propagation power parameter.

[0173] Output data: power attenuation rate per unit distance; Training process: Use historical data to train the first BP neural network to predict the power attenuation rate per unit distance.

[0174] Training the second BP neural network (power attenuation rate), input data: FDM frequency band parameter set; FDM propagation frequency parameter; FDM propagation power set; NLFM propagation frequency curve; NLFM propagation power parameter.

[0175] Output data: power attenuation rate per unit distance; Training process: Use historical data to train the second BP neural network to predict the power attenuation rate per unit distance.

[0176] Generate a unit distance signal propagation simulation function and merge the input layer: merge the input layers of the first BP neural network and the second BP neural network to generate a unit distance signal propagation simulation function.

[0177] Perform unit distance signal propagation simulation: Use the generated simulation function, combined with the soil conductivity sequence and the soil dielectric constant sequence, to perform unit distance signal propagation simulation and obtain the unit distance power attenuation rate sequence and the unit distance phase distortion rate sequence.

[0178] Assume that the simulation results are shown in the following table:

[0179]

[0180] Through the above steps, the BP neural network was used to simulate the signal propagation per unit distance, and the power attenuation rate per unit distance and the phase distortion rate per unit distance were obtained. This method can effectively consider the complex relationship between soil characteristics and signal parameters, and provide more accurate basic data for subsequent signal propagation simulation and cable anti-external damage detection.

[0181] S43: obtaining a propagation distance sequence according to the soil partition sequence, and accumulating and calculating the unit distance power attenuation rate sequence and the unit distance phase distortion rate sequence respectively to obtain a first power attenuation coefficient and a first phase distortion coefficient;

[0182] Furthermore, according to the soil partition sequence, a propagation distance sequence is obtained, and the unit distance power attenuation rate sequence and the unit distance phase distortion rate sequence are accumulated and calculated respectively to obtain a first power attenuation coefficient and a first phase distortion coefficient.

[0183] For example, partition 1 accumulates the calculation results:

[0184] The first power attenuation coefficient: 0.1×10+0.2×20+0.15×15=1+4+2.25=7.25 dB; the first phase distortion coefficient: 0.01×10+0.02×20+0.015×15=0.1+0.4+0.225=0.725 rad.

[0185] Cumulative calculation results for partition 2:

[0186] The first power attenuation coefficient: 0.12×10+0.22×20+0.17×15=1.2+4.4+2.55=8.15 dB; the first phase distortion coefficient: 0.012×10+0.022×20+0.017×15=0.12+0.44+0.255=0.815 rad.

[0187] Cumulative calculation results of partition 3:

[0188] The first power attenuation coefficient: 0.11×10+0.21×20+0.16×15=1.1+4.2+2.4=7.7 dB; the first phase distortion coefficient: 0.011×10+0.021×20+0.016×15=0.11+0.42+0.24=0.77 rad.

[0189] Sum up the first power attenuation coefficient and the first phase distortion coefficient of all signal parameters:

[0190]

[0191] S44: Add the first power attenuation coefficient into the plurality of power attenuation coefficients, and add the first phase distortion coefficient into the plurality of phase distortion coefficients.

[0192] Furthermore, a power attenuation coefficient set is added: AttenuationCoefficients=[]; a phase distortion coefficient set is added: PhaseDistortionCoefficients=[].

[0193] The first power attenuation coefficient and the first phase distortion coefficient of each partition are respectively added to the corresponding set.

[0194] Add the coefficients of partition 1, power attenuation coefficient set: AttenuationCoefficients = [7.25]; phase distortion coefficient set: PhaseDistortionCoefficients = [0.725].

[0195] Add the coefficients of partition 2, power attenuation coefficient set: AttenuationCoefficients=[7.25,8.15]; phase distortion coefficient set: PhaseDistortionCoefficients=[0.725,0.815].

[0196] Add the coefficients of partition 3, power attenuation coefficient set: AttenuationCoefficients = [7.25, 8.15, 7.7]; phase distortion coefficient set: PhaseDistortionCoefficients = [0.725, 0.815, 0.77].

[0197] Specifically, through the above steps, the first power attenuation coefficient and the first phase distortion coefficient of each partition are added to the corresponding sets respectively. These sets summarize the signal propagation characteristics of all partitions, providing comprehensive data support for subsequent signal optimization and cable anti-break detection.

[0198] S50: Based on the plurality of power attenuation coefficients and the plurality of phase distortion coefficients, optimizing the plurality of non-uniform FDM-NLFM signal parameters to obtain target non-uniform FDM-NLFM signal parameters for cable external damage prevention detection;

[0199] Further, based on the plurality of power attenuation coefficients and the plurality of phase distortion coefficients, optimizing the plurality of non-uniform FDM-NLFM signal parameters to obtain target non-uniform FDM-NLFM signal parameters for cable external damage prevention detection, executing step S50 includes:

[0200] S51: Constructing the optimization fitness function:

[0201] ,in, Represents the fitness value, and Characterize the weight parameters, Characterize the power prediction attenuation coefficient, Characterize the phase prediction distortion coefficient, represents the number of population updates, Characterize the update effect regulator, >0, e represents a natural constant, Characterize the power attenuation coefficient, Characterize the phase distortion coefficient threshold;

[0202] S52: obtaining, according to the optimization fitness function, a plurality of fitness evaluation values ​​for the plurality of non-uniform FDM-NLFM signal parameters based on the plurality of power attenuation coefficients and the plurality of phase distortion coefficients;

[0203] S53: selecting first three non-uniform FDM-NLFM signal parameters from the plurality of non-uniform FDM-NLFM signal parameters based on the plurality of fitness evaluation values ​​from small to large;

[0204] S54: performing mean calculation on the quantized signal parameters of the first three non-uniform FDM-NLFM signal parameters, finding a union of the type signal parameters, and obtaining target signal parameters;

[0205] S55: taking the target signal parameter as the guiding target, guiding the mutation of the plurality of non-uniform FDM-NLFM signal parameters to obtain an update non-uniform FDM-NLFM signal parameter execution loop, wherein the mutation rule is a quantization parameter distance reduction mutation and a type parameter cross mutation;

[0206] S56: until a non-uniform FDM-NLFM signal parameter whose fitness value is less than or equal to the convergence threshold is obtained, the parameter is set as the target non-uniform FDM-NLFM signal parameter.

[0207] In the embodiment of the present application, preferably, an optimization fitness function is used to obtain a fitness evaluation value. The fitness function evaluation value increases as the power attenuation coefficient and the phase distortion coefficient increase, and decreases as the power attenuation coefficient and the phase distortion coefficient are continuously iterated. The closer to the optimal value, the smaller the fitness value, indicating that the attenuation and distortion of the signal parameters during the propagation process are smaller, and the signal quality is higher. In order to test convergence, a convergence value is set. This convergence value is set according to actual use. The specific steps are as follows:

[0208] The first three non-uniform FDM-NLFM signal parameters are selected as the preferred parameters, which are more likely to guide the inferior parameters. The first three non-uniform FDM-NLFM signal parameters selected are further processed to obtain the target signal parameters: the quantized signal parameters (such as the propagation power set) are averaged to obtain a comprehensive quantized parameter value; the type signal parameters (such as the frequency band parameter set) are unioned to obtain a comprehensive type parameter set.

[0209] Furthermore, the target signal parameters are used as the guiding targets to guide the variation of all non-uniform FDM-NLFM signal parameters:

[0210] Quantization parameter distance reduction mutation: adjust the quantization parameter to make it closer to the quantization parameter of the target signal parameter and reduce the distance with the target parameter; type parameter crossover mutation: through the crossover operation, the type parameter of the target signal parameter is combined with the type parameters of other signal parameters to generate a new type parameter set.

[0211] Furthermore, the above process is repeated for multiple cycles of optimization until a non-uniform FDM-NLFM signal parameter with a fitness value less than or equal to a convergence threshold is obtained. The convergence threshold is a preset value used to determine whether the optimization process has reached a satisfactory level. Once the fitness value of a signal parameter reaches or is lower than the threshold, the parameter is considered to be the target non-uniform FDM-NLFM signal parameter.

[0212] Through the above optimization process, the most suitable signal parameters for cable anti-breakage detection are found from multiple non-uniform FDM-NLFM signal parameters. This method comprehensively considers the attenuation and distortion of the signal during propagation, and ensures the optimality of the signal parameters through dynamic adjustment and optimization, thereby improving the accuracy and reliability of cable anti-breakage detection.

[0213] The embodiment of the present invention provides a cable external damage prevention method based on non-uniform FDM-NLFM, which has at least the following technical effects:

[0214] By introducing the optimization fitness function, the present invention can comprehensively consider the influence of power attenuation and phase distortion on signal propagation. The fitness function not only considers the power attenuation coefficient and the phase distortion coefficient, but also introduces the number of population updates and the update influence adjustment factor, so that the optimization process of signal parameters is more dynamic and adaptable. This optimization method that comprehensively considers multiple factors significantly improves the adaptability of the signal in different soil environments and ensures that the signal can be stably propagated in a complex underground environment; further, the present invention optimizes the non-uniform FDM-NLFM signal parameters so that the signal can better resist the influence of soil property changes during the propagation process. By selecting the signal parameter with the smallest fitness value, it can ensure that the signal has the smallest attenuation and distortion during the propagation process, thereby improving the accuracy of the detection results; further, by guiding the variation and cyclic optimization, the present invention can continuously adjust the signal parameters so that it gradually approaches the optimal solution. Through multiple cyclic optimizations, the signal parameters finally obtained can maintain stable propagation characteristics in different soil partitions, reducing the uncertainty in the signal propagation process, thereby achieving the technical effect of improving efficiency and taking into account the cost.

[0215] Embodiment 2, as Figure 2As shown, based on the same inventive concept as the cable external damage prevention method based on non-uniform FDM-NLFM provided in Embodiment 1, the embodiment of the present invention further provides a cable external damage prevention system based on non-uniform FDM-NLFM, including:

[0216] The soil partition sequence module 11 is used to receive soil detection information of the cable laying area, partition the laying soil, and sort the soil based on the signal injection point as the starting point combined with the detection direction to obtain a soil partition sequence;

[0217] A soil conductivity sequence module 12 is used to traverse the soil partition sequence to perform mode statistics of soil conductivity of historical samples to obtain a soil conductivity sequence;

[0218] A soil dielectric constant sequence module 13 is used to traverse the soil partition results to perform mode statistics of historical sample dielectric constants to obtain a soil dielectric constant sequence;

[0219] The power attenuation and phase distortion coefficient module 14 is used to randomly generate a number of non-uniform FDM-NLFM signal parameters, combine the soil conductivity sequence and the soil dielectric constant sequence to perform signal propagation simulation, and obtain a number of power attenuation coefficients and a number of phase distortion coefficients;

[0220] The cable anti-breakage detection module 15 is used to optimize the several non-uniform FDM-NLFM signal parameters according to the several power attenuation coefficients and the several phase distortion coefficients to obtain target non-uniform FDM-NLFM signal parameters for cable anti-breakage detection.

[0221] Furthermore, the soil partition sequence module 11 executes the following steps:

[0222] The soil detection information includes soil composition information, soil temperature information and soil moisture information;

[0223] Configure soil composition deviation threshold, soil temperature deviation threshold, and soil moisture deviation threshold;

[0224] According to the soil composition deviation threshold, the soil temperature deviation threshold and the soil moisture deviation threshold, a multi-level hierarchical clustering analysis is performed in combination with the soil composition information, the soil temperature information and the soil moisture information to obtain a soil partition set, wherein the soil composition, soil temperature and moisture of any partition are consistent;

[0225] The soil partition set is sorted based on the signal injection point as the starting point and the detection direction to obtain a soil partition sequence.

[0226] Furthermore, the soil partition sequence module 11 executes the following steps:

[0227] Extracting attributes to be analyzed from soil composition information, soil temperature information and soil moisture information;

[0228] Taking the attribute to be analyzed as the only variable, setting the set of attributes not to be analyzed as a fixed value, taking conductivity and dielectric constant as follow-up variables, collecting the set of recorded values ​​of the attribute to be analyzed, the set of recorded values ​​of conductivity, and the set of recorded values ​​of dielectric constant;

[0229] Performing pairwise fluctuation calculations on the attribute record value set to be analyzed, the conductivity record value set, and the dielectric constant record value set to obtain a deviation set of attribute record values ​​to be analyzed, a deviation set of conductivity record values, and a deviation set of dielectric constant record values;

[0230] Based on the conductivity record value deviation set, screening a first selected set of attribute record value deviations to be analyzed whose conductivity record value deviations are greater than or equal to a conductivity deviation threshold from the attribute record value deviation set to be analyzed;

[0231] Based on the dielectric constant record value deviation set, screening a first selected set of attribute record value deviations to be analyzed whose dielectric constant record value deviations are greater than or equal to a dielectric constant deviation threshold from the attribute record value deviation set to be analyzed;

[0232] After outlier removal is performed on the first selected set of attribute record value deviations to be analyzed and the first selected set of attribute record value deviations to be analyzed, the minimum value of the first retained attribute record value deviation to be analyzed and the second retained attribute record value deviation to be analyzed is extracted and set as the attribute deviation threshold to be analyzed, and the soil composition deviation threshold, the soil temperature deviation threshold and the soil moisture deviation threshold are added.

[0233] Furthermore, the soil conductivity sequence module 12 executes the following steps:

[0234] Extracting a first soil partition according to the soil partition sequence, wherein the first soil partition has a soil component identifier, a soil temperature identifier, and a soil moisture identifier;

[0235] Constructing a first index constraint rule according to the soil component identifier, the soil temperature identifier and the soil moisture identifier;

[0236] Retrieving a first soil conductivity detection value set of a first historical sample set that satisfies the first index constraint rule, wherein the first historical sample set has a soil component label set, a soil temperature label set, and a soil moisture label set;

[0237] Constructing a second index constraint rule set according to the soil component label set, the soil temperature label set and the soil moisture label set, and respectively retrieving a plurality of groups of second soil conductivity detection values ​​of a plurality of groups of second historical samples that satisfy the second index constraint rule set;

[0238] The first soil conductivity detection value set and the plurality of second soil conductivity detection value sets are subjected to mode aggregation to obtain first partition soil conductivity, which is added to the soil conductivity sequence.

[0239] Furthermore, the execution steps of the soil dielectric constant sequence module 13 are the same as the execution steps of the soil conductivity sequence module 12 .

[0240] Furthermore, the power attenuation and phase distortion coefficient module 14 executes the following steps:

[0241] Any one of the several non-uniform FDM-NLFM signal parameters includes an FDM frequency band parameter set, an FDM propagation frequency parameter, an FDM propagation power set, an NLFM propagation frequency curve and an NLFM propagation power parameter;

[0242] According to the FDM frequency band parameter set, the FDM propagation frequency parameter, the FDM propagation power set, the NLFM propagation frequency curve and the NLFM propagation power parameter, in combination with the soil conductivity sequence and the soil dielectric constant sequence, a unit distance signal propagation simulation is performed to obtain a unit distance power attenuation rate sequence and a unit distance phase distortion rate sequence;

[0243] According to the soil partition sequence, a propagation distance sequence is obtained, and the unit distance power attenuation rate sequence and the unit distance phase distortion rate sequence are accumulated and calculated respectively to obtain a first power attenuation coefficient and a first phase distortion coefficient;

[0244] The first power attenuation coefficient is added to the plurality of power attenuation coefficients, and the first phase distortion coefficient is added to the plurality of phase distortion coefficients.

[0245] Furthermore, the power attenuation and phase distortion coefficient module 14 executes the following steps:

[0246] Configure FDM frequency band parameter set recording data, FDM propagation frequency parameter recording data, FDM propagation power set recording data, NLFM propagation frequency curve recording data, NLFM propagation power parameter recording data, unit distance power attenuation rate identification data and unit distance phase distortion rate identification data;

[0247] According to the unit distance power attenuation rate identification data as supervision, the FDM frequency band parameter set record data, the FDM propagation frequency parameter record data, the FDM propagation power set record data, the NLFM propagation frequency curve record data, and the NLFM propagation power parameter record data are used as input to train the first BP neural network to obtain a unit distance power attenuation rate simulation channel;

[0248] According to the unit distance phase distortion rate identification data as supervision, the FDM frequency band parameter set record data, the FDM propagation frequency parameter record data, the FDM propagation power set record data, the NLFM propagation frequency curve record data, and the NLFM propagation power parameter record data are used as input to train the second BP neural network to obtain a unit distance phase distortion rate simulation channel;

[0249] The input layers of the unit distance power attenuation rate simulation channel and the unit distance phase distortion rate simulation channel are combined to generate a unit distance signal propagation simulation function, perform unit distance signal propagation simulation, and obtain a unit distance power attenuation rate sequence and a unit distance phase distortion rate sequence.

[0250] Furthermore, the cable anti-external damage detection module 15 executes the following steps:

[0251] Construct an optimization fitness function:

[0252] ,

[0253] in, Represents the fitness value, and Characterize the weight parameters, Characterize the power prediction attenuation coefficient, Characterize the phase prediction distortion coefficient, represents the number of population updates, Characterize the update effect regulator, >0, e represents a natural constant, Characterize the power attenuation coefficient, Characterize the phase distortion coefficient threshold;

[0254] According to the optimization fitness function, based on the several power attenuation coefficients and the several phase distortion coefficients, obtaining several fitness evaluation values ​​for the several non-uniform FDM-NLFM signal parameters;

[0255] Based on the plurality of fitness evaluation values ​​from small to large, selecting first three non-uniform FDM-NLFM signal parameters from the plurality of non-uniform FDM-NLFM signal parameters;

[0256] Performing mean calculation on the quantized signal parameters of the first three non-uniform FDM-NLFM signal parameters, finding the union of the type signal parameters, and obtaining the target signal parameters;

[0257] Taking the target signal parameter as the guiding target, guiding mutation of the plurality of non-uniform FDM-NLFM signal parameters to obtain an update non-uniform FDM-NLFM signal parameter execution cycle, wherein the mutation rule is quantization parameter distance reduction mutation and type parameter cross mutation;

[0258] Until a non-uniform FDM-NLFM signal parameter whose fitness value is less than or equal to the convergence threshold is obtained, it is set as the target non-uniform FDM-NLFM signal parameter.

[0259] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0260] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0261] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0262] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0263] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0264] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0265] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A cable damage prevention method based on non-uniform FDM-NLFM, characterized in that: include: Receive soil detection information in the cable laying area, divide the laying soil into zones, and sort the zones based on the signal injection point and the detection direction to obtain a soil zone sequence; Traversing the soil partition sequence to perform mode statistics of soil conductivity of historical samples to obtain a soil conductivity sequence; Traversing the soil partition results to perform mode statistics of historical sample dielectric constants to obtain a soil dielectric constant sequence; Randomly generate a number of non-uniform FDM-NLFM signal parameters, combine the soil conductivity sequence and the soil dielectric constant sequence to perform signal propagation simulation, and obtain a number of power attenuation coefficients and a number of phase distortion coefficients; Based on the several power attenuation coefficients and the several phase distortion coefficients, the several non-uniform FDM-NLFM signal parameters are optimized to obtain target non-uniform FDM-NLFM signal parameters for cable anti-external damage detection.

2. The method according to claim 1, characterized in that Receive soil detection information of the cable laying area, divide the laying soil into zones, and sort them based on the signal injection point and the detection direction to obtain a soil zone sequence, including: The soil detection information includes soil composition information, soil temperature information and soil moisture information; Configure soil composition deviation threshold, soil temperature deviation threshold, and soil moisture deviation threshold; According to the soil composition deviation threshold, the soil temperature deviation threshold and the soil moisture deviation threshold, a multi-level hierarchical clustering analysis is performed in combination with the soil composition information, the soil temperature information and the soil moisture information to obtain a soil partition set, wherein the soil composition, soil temperature and moisture of any partition are consistent; The soil partition set is sorted based on the signal injection point as the starting point and the detection direction to obtain a soil partition sequence.

3. The method according to claim 2, characterized in that Configure soil composition deviation threshold, soil temperature deviation threshold, and soil moisture deviation threshold, including: Extracting attributes to be analyzed from soil composition information, soil temperature information and soil moisture information; Taking the attribute to be analyzed as the only variable, setting the set of attributes not to be analyzed as a fixed value, taking conductivity and dielectric constant as follow-up variables, collecting the set of recorded values ​​of the attribute to be analyzed, the set of recorded values ​​of conductivity, and the set of recorded values ​​of dielectric constant; Performing pairwise fluctuation calculations on the attribute record value set to be analyzed, the conductivity record value set, and the dielectric constant record value set to obtain a deviation set of attribute record values ​​to be analyzed, a deviation set of conductivity record values, and a deviation set of dielectric constant record values; Based on the conductivity record value deviation set, screening a first selected set of attribute record value deviations to be analyzed whose conductivity record value deviations are greater than or equal to a conductivity deviation threshold from the attribute record value deviation set to be analyzed; Based on the dielectric constant record value deviation set, screening a first selected set of attribute record value deviations to be analyzed whose dielectric constant record value deviations are greater than or equal to a dielectric constant deviation threshold from the attribute record value deviation set to be analyzed; After outlier removal is performed on the first selected set of attribute record value deviations to be analyzed and the first selected set of attribute record value deviations to be analyzed, the minimum value of the first retained attribute record value deviation to be analyzed and the second retained attribute record value deviation to be analyzed is extracted and set as the attribute deviation threshold to be analyzed, and the soil composition deviation threshold, the soil temperature deviation threshold and the soil moisture deviation threshold are added.

4. The method according to claim 1, characterized in that Traversing the soil partition sequence to perform mode statistics of soil conductivity of historical samples, and obtaining a soil conductivity sequence, including: Extracting a first soil partition according to the soil partition sequence, wherein the first soil partition has a soil component identifier, a soil temperature identifier, and a soil moisture identifier; Constructing a first index constraint rule according to the soil component identifier, the soil temperature identifier and the soil moisture identifier; Retrieving a first soil conductivity detection value set of a first historical sample set that satisfies the first index constraint rule, wherein the first historical sample set has a soil component label set, a soil temperature label set, and a soil moisture label set; Constructing a second index constraint rule set according to the soil component label set, the soil temperature label set and the soil moisture label set, and respectively retrieving a plurality of groups of second soil conductivity detection values ​​of a plurality of groups of second historical samples that satisfy the second index constraint rule set; The first soil conductivity detection value set and the plurality of second soil conductivity detection value sets are subjected to mode aggregation to obtain first partition soil conductivity, which is added to the soil conductivity sequence.

5. The method according to claim 1, characterized in that A number of non-uniform FDM-NLFM signal parameters are randomly generated, and signal propagation simulation is performed in combination with the soil conductivity sequence and the soil dielectric constant sequence to obtain a number of power attenuation coefficients and a number of phase distortion coefficients, including: Any one of the several non-uniform FDM-NLFM signal parameters includes an FDM frequency band parameter set, an FDM propagation frequency parameter, an FDM propagation power set, an NLFM propagation frequency curve and an NLFM propagation power parameter; According to the FDM frequency band parameter set, the FDM propagation frequency parameter, the FDM propagation power set, the NLFM propagation frequency curve and the NLFM propagation power parameter, in combination with the soil conductivity sequence and the soil dielectric constant sequence, a unit distance signal propagation simulation is performed to obtain a unit distance power attenuation rate sequence and a unit distance phase distortion rate sequence; According to the soil partition sequence, a propagation distance sequence is obtained, and the unit distance power attenuation rate sequence and the unit distance phase distortion rate sequence are accumulated and calculated respectively to obtain a first power attenuation coefficient and a first phase distortion coefficient; The first power attenuation coefficient is added to the plurality of power attenuation coefficients, and the first phase distortion coefficient is added to the plurality of phase distortion coefficients.

6. The method according to claim 5, characterized in that According to the FDM frequency band parameter set, the FDM propagation frequency parameter, the FDM propagation power set, the NLFM propagation frequency curve and the NLFM propagation power parameter, a unit distance signal propagation simulation is performed in combination with the soil conductivity sequence and the soil dielectric constant sequence to obtain a unit distance power attenuation rate sequence and a unit distance phase distortion rate sequence, including: Configure FDM frequency band parameter set recording data, FDM propagation frequency parameter recording data, FDM propagation power set recording data, NLFM propagation frequency curve recording data, NLFM propagation power parameter recording data, unit distance power attenuation rate identification data and unit distance phase distortion rate identification data; According to the unit distance power attenuation rate identification data as supervision, the FDM frequency band parameter set record data, the FDM propagation frequency parameter record data, the FDM propagation power set record data, the NLFM propagation frequency curve record data, and the NLFM propagation power parameter record data are used as input to train the first BP neural network to obtain a unit distance power attenuation rate simulation channel; According to the unit distance phase distortion rate identification data as supervision, the FDM frequency band parameter set record data, the FDM propagation frequency parameter record data, the FDM propagation power set record data, the NLFM propagation frequency curve record data, and the NLFM propagation power parameter record data are used as input to train the second BP neural network to obtain a unit distance phase distortion rate simulation channel; The input layers of the unit distance power attenuation rate simulation channel and the unit distance phase distortion rate simulation channel are combined to generate a unit distance signal propagation simulation function, perform unit distance signal propagation simulation, and obtain a unit distance power attenuation rate sequence and a unit distance phase distortion rate sequence.

7. The method according to claim 1, characterized in that Based on the plurality of power attenuation coefficients and the plurality of phase distortion coefficients, optimizing the plurality of non-uniform FDM-NLFM signal parameters to obtain target non-uniform FDM-NLFM signal parameters for cable external damage prevention detection, including: Construct an optimization fitness function: ， in, Represents the fitness value, and Characterize the weight parameters, Characterize the power prediction attenuation coefficient, Characterize the phase prediction distortion coefficient, represents the number of population updates, Characterize the update effect regulator, >0, e represents a natural constant, Characterize the power attenuation coefficient, Characterize the phase distortion coefficient threshold; According to the optimization fitness function, based on the several power attenuation coefficients and the several phase distortion coefficients, obtaining several fitness evaluation values ​​for the several non-uniform FDM-NLFM signal parameters; Based on the plurality of fitness evaluation values ​​from small to large, selecting first three non-uniform FDM-NLFM signal parameters from the plurality of non-uniform FDM-NLFM signal parameters; Calculating the mean of the quantized signal parameters of the first three non-uniform FDM-NLFM signal parameters, finding the union of the type signal parameters, and obtaining the target signal parameters; Taking the target signal parameter as the guiding target, guiding mutation of the plurality of non-uniform FDM-NLFM signal parameters to obtain an update non-uniform FDM-NLFM signal parameter execution cycle, wherein the mutation rule is quantization parameter distance reduction mutation and type parameter cross mutation; Until a non-uniform FDM-NLFM signal parameter whose fitness value is less than or equal to the convergence threshold is obtained, it is set as the target non-uniform FDM-NLFM signal parameter.

8. A cable external damage prevention system based on non-uniform FDM-NLFM, characterized in that: include: The soil partition sequence module is used to receive soil detection information of the cable laying area, partition the laying soil, and sort it based on the signal injection point as the starting point and the detection direction to obtain the soil partition sequence; A soil conductivity sequence module, used to traverse the soil partition sequence to perform mode statistics of soil conductivity of historical samples to obtain a soil conductivity sequence; A soil dielectric constant sequence module is used to traverse the soil partition results to perform mode statistics of historical sample dielectric constants to obtain a soil dielectric constant sequence; The power attenuation and phase distortion coefficient module is used to randomly generate a number of non-uniform FDM-NLFM signal parameters, combine the soil conductivity sequence and the soil dielectric constant sequence to perform signal propagation simulation, and obtain a number of power attenuation coefficients and a number of phase distortion coefficients; The cable anti-external damage detection module is used to optimize the several non-uniform FDM-NLFM signal parameters according to the several power attenuation coefficients and the several phase distortion coefficients to obtain target non-uniform FDM-NLFM signal parameters for cable anti-external damage detection.

Images