Courtyard buried equipment damage point detection method based on intelligent sensor

Abstract

The invention belongs to the field of damage detection, and particularly relates to a courtyard buried equipment damage point detection method and system based on an intelligent sensor, and the method comprises the steps: generating a fundamental frequency and frequency multiplication coefficient dictionary composite pure composite signal through an intelligent sensor transmitting end, testing and determining an initial phase difference, and setting the initial phase difference in a receiving end; determining a transmitting power and a grounding position point by using a Cole-Cole model and a dual-mode adaptation rule in combination with the buried equipment distribution diagram and a grounding loop sweep frequency signal; after a signal is transmitted, a receiving end obtains a multi-frequency voltage signal sequence through three-stage adaptive gain, and then obtains a damage point coordinate and a grading evaluation result by combining the initial phase difference and applying a fast Fourier algorithm and a damage fixed-point evaluation model configured with a two-way positioning rule; according to the method, the micro-damage identification capability and the positioning precision in a complex courtyard environment are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of remote sensing monitoring, and in particular relates to a method and system for detecting damage points of buried equipment in courtyards based on intelligent sensors. Background Technology

[0002] The principle of detecting corrosion layer damage points is to apply an electrical signal to the exposed part of the buried pipeline. At the damage point, current leakage occurs, forming a current field distribution centered on the damage point, with potential extremes directly above it. Currently, commonly used detection instruments include RD-PCM and DM, both using low-frequency damage point detection, suitable for long-distance pipelines. However, for backyard pipelines, the short distance and high grounding resistance make signal loading more difficult, hindering the detection of corrosion layer damage points. For example, patent publication CN110873734A proposes high-frequency pulse reflection... Patent CN110873734A proposes DC voltage gradient localization. The former relies on the reflection and refraction of spontaneous high-frequency pulses at the damage point, requiring complex waveform recognition and speed calibration, and cannot actively excite minor damage. The latter requires manual measurement on foot with a probe, relying on the deflection and reversal of the voltmeter to determine the damage, and lacks dynamic phase analysis capabilities. In addition, the existing methods do not integrate high-frequency active excitation and multi-frequency synthesis technology, resulting in the inability to accurately locate the damage direction through phase difference, and relying on manual experience for judgment, leading to a high misjudgment rate. Therefore, this application provides a method and system for detecting damage points of buried equipment in courtyards based on intelligent sensors. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention proposes a method and system for detecting damage points of buried equipment in courtyards based on intelligent sensors. This method generates a pure composite signal combining the fundamental frequency and harmonic coefficients at the transmitting end of the intelligent sensor, tests to determine the initial phase difference, and embeds it into the receiving end. Combining the buried equipment distribution map and the grounding loop frequency sweep signal, the transmitting power and grounding location are determined using the Cole-Cole model and dual-mode adaptation rules. After transmitting the signal, the receiving end acquires a multi-frequency voltage signal sequence through a three-level adaptive gain, and then, combined with the initial phase difference, uses a fast Fourier transform algorithm and a damage point assessment model with configured bidirectional positioning rules to obtain the damage point coordinates and graded assessment results. This method significantly improves the ability to identify and locate micro-damage in complex courtyard environments.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A method for detecting damage points of buried equipment in courtyards based on intelligent sensors includes:

[0006] Configure a smart sensor, combine the pure composite transmission signal obtained by combining the base frequency and harmonic coefficient dictionary at the smart sensor transmitter, test and determine the initial phase difference, and embed the determined initial phase difference into the receiver.

[0007] Based on the distribution map of buried equipment and the frequency sweep signal of the grounding loop, the power and grounding location of the transmitter are determined by using the Cole-Cole model and dual-mode adaptation rules.

[0008] Based on the determined transmitter power and grounding location, a corresponding pure composite transmission signal is transmitted, and the receiver uses a preset three-level adaptive gain to determine and amplify the acquired information to obtain the amplified multi-frequency voltage signal sequence.

[0009] Based on the multi-frequency voltage signal sequence combined with the initial phase difference, the damage point coordinates and graded assessment results of buried equipment are obtained by using the Fast Fourier Transform algorithm and a damage point assessment model configured with bidirectional positioning rules.

[0010] Specifically, the process of acquiring a pure composite transmission signal includes:

[0011] The accuracy of historical transmission signal information and corresponding evaluation results under different depths, grounding resistivity and buried equipment distribution conditions in the historical courtyard is obtained and preprocessed using a filtering algorithm.

[0012] The fundamental frequency signal and harmonic frequency signal information are obtained by combining filtered historical transmission signal information with spectrum analysis algorithms;

[0013] Based on the fundamental frequency signal and harmonic signal components, the phase offset of each harmonic relative to the fundamental frequency is obtained through frequency domain analysis algorithms;

[0014] A frequency doubling coefficient dictionary is constructed based on the depth of the corresponding buried equipment, the grounding resistivity of the buried area, the distribution status of the buried equipment, the phase offset of each frequency doubling relative to the fundamental frequency, and the corresponding historical evaluation accuracy, combined with a graph neural network.

[0015] Specifically, the process of acquiring a pure composite transmission signal also includes:

[0016] The distribution status, depth, and grounding resistivity of buried equipment in the courtyard area corresponding to the current target area are obtained in real time. The fundamental frequency and harmonic frequency sequence of the current target area are obtained by combining the matching algorithm with the harmonic coefficient dictionary.

[0017] A multi-channel frequency doubling sequence is obtained by combining the fundamental frequency matched to the current target region with the frequency doubling coefficients obtained from the matching.

[0018] The fundamental frequency and multi-channel harmonic frequency sequences are discretized, and multi-channel data vector addition is performed on the frequency points corresponding to each discrete frequency at the same timing node to obtain the composite transmission signal.

[0019] A pure composite transmitted signal is obtained by interpolation correction based on the composite transmitted signal and Chebyshev algorithm.

[0020] Specifically, the process of obtaining the power and grounding location of the transmitting end includes:

[0021] Select the coordinates of the initial grounding candidate point according to the distribution map of buried equipment, and generate a sweep frequency current at the candidate point location according to the set sweep frequency current range;

[0022] The generated sweep frequency current is injected into the ground electrode through a power amplifier. At the same time, the voltage response is measured using the four-electrode method, and the amplitude and phase angle of the complex impedance are extracted using a lock-in amplifier to obtain impedance spectrum data.

[0023] Based on impedance spectrum data and burial depth, the distribution function of Cole-Cole parameters with depth and the impedance spectrum prediction function are obtained by fitting the Levenberg-Marquardt algorithm. The fitted function is then embedded into the Cole-Cole model to obtain a three-dimensional soil electrical property model.

[0024] Input the target area, the distribution map of buried equipment, the three-dimensional soil electrical property model, and the coordinates of the initial grounding candidate points into the modeling software. Use the modeling software to generate a buffer zone with a radius of d for the buried equipment. At the same time, generate grid points in the remaining area at 2m intervals, filter out points that exceed the boundary, and output a gridded candidate point set.

[0025] Specifically, the process of obtaining the power and grounding location of the transmitting end also includes:

[0026] Based on the coordinates of the gridded candidate point set and the correlation distribution of each candidate point with the three-dimensional soil electrical property parameters in Cole-Cole with depth, the fitness function constructed by depth and signal-to-noise ratio, depth and power intensity and depth signal quality and the constraints constructed by signal-to-noise ratio, power and signal, the input sequence of the particle swarm algorithm is obtained.

[0027] The particle swarm optimization algorithm is integrated into the three-dimensional soil electrical property model of the modeling software. The input sequence of the particle swarm optimization algorithm is input into the particle swarm optimization algorithm and combined with the simulation algorithm. For each candidate point, the minimum transmission power and the maximum signal quality score are taken as the objectives. The signal transmission and reception simulation is carried out to obtain the coordinates of the candidate point with the minimum transmission power and the maximum signal quality score, as well as the corresponding transmission power.

[0028] Specifically, the steps for the amplified multi-frequency voltage signal sequence include:

[0029] The buried equipment detection voltage signal is acquired and initially amplified by a differential amplifier to obtain a preliminary differential voltage signal.

[0030] The signal strength assessment model, constructed based on the preliminary differential voltage signal and a comprehensive fuzzy algorithm, is used to obtain the assessment strength of the preliminary acquired signal.

[0031] Based on the evaluation intensity of the initially acquired signal and combined with the preset three-level gain discrimination interval, the signal with different evaluation intensity is amplified to obtain the multi-frequency voltage signal sequence after amplification.

[0032] The three-level gain discrimination interval is constructed by the signal strength interval and the set three-level gain coefficient, and each signal strength interval corresponds one-to-one with each level of gain coefficient.

[0033] Specifically, the process for obtaining the coordinates of the damage points and the grading assessment results of buried equipment includes:

[0034] Based on the multi-frequency voltage signal sequence after gain, the multi-frequency voltage signal is framed by moving average filtering and Hanning window function to obtain the framed multi-frequency voltage signal sequence.

[0035] Based on the multi-frequency voltage signal sequence after framing, the time-domain signal corresponding to each frame is converted into frequency-domain data through the Fourier transform algorithm, and the real and imaginary parts of each frequency component are calculated to obtain a composite spectrum matrix; the composite spectrum matrix includes the real part, imaginary part, amplitude and phase corresponding to each frequency in the multi-frequency voltage signal;

[0036] Based on the composite spectrum matrix, a Costas ring is constructed with the fundamental frequency as the reference to automatically track the fundamental frequency drift. Based on the locked fundamental frequency position, the frequency coordinates of each frequency in the multi-frequency voltage signal are obtained through the FFT spectrum matrix search algorithm, and the real and imaginary parts of the corresponding spectrum components are extracted to construct a complex set of four frequency points.

[0037] The original phase is calculated based on the real and imaginary parts of each frequency in the four-frequency complex set, and the phase angle at the corresponding frequency is determined by the arctangent function of the original phase.

[0038] Specifically, the process for obtaining the coordinates of the damage points and the grading assessment results of buried equipment also includes:

[0039] Based on the original phase combined with the corresponding initial phase difference, the calibrated phase value is obtained;

[0040] Based on the calibrated phase values, a normalized phase difference array is obtained. in This represents the phase difference between the phase corresponding to the q-th frequency and the phase corresponding to the fundamental frequency;

[0041] based on The consistency score S of all phase differences is obtained through a multi-frequency consistency check algorithm.

[0042] Based on the consistency score of all phase differences and the preset damage point discrimination threshold, the damage point discrimination direction is obtained; the damage point discrimination threshold includes: rear safety zone: S≥0.7(q-1); forward warning zone: S≤0.3(q-1);

[0043] Based on the direction of the acquired damage point, the phase change value between the maximum harmonic frequency and the fundamental frequency of adjacent sampling points, and the moving distance of the receiving end are collected;

[0044] The phase change rate is obtained based on the ratio of the phase change value to the movement distance of the A-frame. When the phase change rate corresponding to three consecutive adjacent sampling points is greater than the damage judgment threshold, the coordinates of the damage point are obtained by combining the configured GPS module.

[0045] Specifically, the process for obtaining the coordinates of the damage points and the grading assessment results of buried equipment also includes:

[0046] Based on phase difference array The amplitude of the signal at the corresponding frequency is used to obtain the amplitude attenuation rate after passing through the damage point through an attenuation ratio algorithm;

[0047] Based on the phase difference and moving distance of adjacent sampling points, the gradient of signal phase change near the damage point is obtained through a slope extraction algorithm.

[0048] Based on the time-domain signal corresponding to the maximum harmonic frequency, the stability index of the high-frequency signal is obtained through wavelet packet transform decomposition and energy entropy algorithm.

[0049] Based on the signal phase abrupt change gradient, amplitude attenuation rate, and high-frequency signal stability index corresponding to the damage point coordinates, the damage level and characteristic index are obtained through a logical rule matching algorithm; the logical rule matching algorithm is constructed by combining the amplitude attenuation rate, signal phase abrupt change gradient, and high-frequency signal stability index.

[0050] Based on the damage level, characteristic indicators, and preset maintenance strategies, an algorithm is generated to output an assessment report and maintenance strategy.

[0051] The intelligent sensor-based damage detection system for underground garden equipment includes: a transmission configuration module, a grounding selection module, a gain module, and a positioning evaluation module.

[0052] The transmission configuration module, based on the intelligent sensor transmitter, combines a pure composite transmission signal obtained by combining the base frequency and harmonic coefficient dictionary, and tests and determines the initial phase difference and embeds the determined initial phase difference into the receiver.

[0053] The grounding selection module, based on the distribution map of buried equipment and the frequency sweep signal of the grounding loop, determines the power and grounding location of the transmitter by combining the Cole-Cole model with the dual-mode adaptation rule.

[0054] The gain module transmits a corresponding pure composite transmission signal based on the determined transmitter power and grounding location, and uses the receiver to determine and amplify the acquired information in combination with a preset three-level adaptive gain to obtain a multi-frequency voltage signal sequence after amplification.

[0055] The positioning and evaluation module, based on a multi-frequency voltage signal sequence combined with an initial phase difference, uses a fast Fourier transform algorithm and a damage point assessment model configured with bidirectional positioning rules to obtain the coordinates of the damage point and the graded assessment results of the buried equipment.

[0056] Compared with the prior art, the beneficial effects of the present invention are:

[0057] This invention addresses the shortcomings of existing technologies by constructing a frequency doubling coefficient dictionary and combining it with real-time environmental parameters to generate a pure composite transmission signal, effectively enhancing signal stability and anti-interference capabilities. It utilizes the Cole-Cole model and particle swarm optimization algorithm to determine the transmission power and grounding location, achieving precise modeling of soil electrical properties and optimized matching of transmission parameters, thus reducing energy consumption while improving signal transmission quality. The receiver employs a three-level adaptive gain mechanism, dynamically adjusting the amplification factor according to signal strength to avoid signal distortion and ensure accurate data acquisition. Combining the Fast Fourier Transform algorithm, multi-frequency phase difference analysis, and bidirectional positioning rules, it can quickly and accurately pinpoint the location of damage points and achieve scientific classification of damage severity through multi-dimensional evaluation indicators. Finally, it outputs a detailed evaluation report and maintenance strategy, providing a comprehensive and reliable decision-making basis for the maintenance of buried equipment in courtyards. Attached Figure Description

[0058] Figure 1 This is a flowchart of the method for detecting damage points of buried courtyard equipment based on intelligent sensors, as described in Embodiment 1 of the present invention.

[0059] Figure 2 This is a block diagram of the damage detection system for underground garden equipment based on intelligent sensors, according to Embodiment 2 of the present invention. Detailed Implementation

[0060] Example 1

[0061] Please see Figure 1 The present invention provides an embodiment of a method for detecting damage points in buried equipment in courtyards based on intelligent sensors, applied to the detection of damage points in the anti-corrosion layer of buried metal pipes in courtyards, comprising the following steps:

[0062] S1. Configure a smart sensor, combine the pure composite transmission signal obtained by combining the base frequency and harmonic coefficient dictionary at the smart sensor transmitter, test and determine the initial phase difference, and embed the determined initial phase difference into the receiver.

[0063] S2. Based on the distribution map of buried equipment and the frequency sweep signal of the grounding loop, the power and grounding location of the transmitter are determined by the Cole-Cole model and the dual-mode adaptation rule.

[0064] S3. Based on the determined transmitter power and grounding location, transmit the corresponding pure composite transmission signal, and use the receiver to judge and amplify the acquired information in combination with the preset three-level adaptive gain to obtain the amplified multi-frequency voltage signal sequence.

[0065] S4. Based on the multi-frequency voltage signal sequence combined with the initial phase difference, the damage point coordinates and graded evaluation results of the buried equipment are obtained through the Fast Fourier Transform algorithm and the damage point assessment model configured with bidirectional positioning rules.

[0066] It should be noted that this embodiment also controls the change of the transmitter's signal frequency and the switching on and off of the transmission signal through the remote control module; the data collected by the receiver is sent to a designated computer indoors through the remote control module, which facilitates indoor expert diagnosis and reduces the technical requirements of field staff.

[0067] It should be further explained that the process of acquiring a pure composite transmission signal includes:

[0068] The accuracy of historical transmission signal information and corresponding evaluation results under different depths, grounding resistivity and buried equipment distribution conditions in the historical courtyard is obtained and preprocessed using a filtering algorithm.

[0069] The fundamental frequency signal and harmonic frequency signal information are obtained by combining filtered historical transmission signal information with spectrum analysis algorithms;

[0070] Based on the fundamental frequency signal and harmonic signal components, the phase offset of each harmonic relative to the fundamental frequency is obtained through frequency domain analysis algorithms;

[0071] A frequency doubling coefficient dictionary is constructed based on the depth of the corresponding buried equipment, the grounding resistivity of the buried area, the distribution status of the buried equipment, the phase offset of each frequency doubling relative to the fundamental frequency, and the corresponding historical evaluation accuracy, combined with a graph neural network.

[0072] The distribution status, depth, and grounding resistivity of buried equipment in the courtyard area corresponding to the current target area are obtained in real time. The fundamental frequency and harmonic frequency sequence of the current target area are obtained by combining the matching algorithm with the harmonic coefficient dictionary.

[0073] A multi-channel frequency doubling sequence is obtained by combining the fundamental frequency matched to the current target region with the frequency doubling coefficients obtained from the matching.

[0074] The fundamental frequency and multi-channel harmonic frequency sequences are discretized, and multi-channel data vector addition is performed on the frequency points corresponding to each discrete frequency at the same timing node to obtain the composite transmission signal.

[0075] A pure composite transmitted signal is obtained by interpolation correction based on the composite transmitted signal and Chebyshev algorithm.

[0076] To further illustrate the process of acquiring a pure composite transmission signal, the fundamental frequency is set to 1 Hz, and the harmonics include three different frequency levels: 2 Hz, 4 Hz, and 8 Hz. The specific implementation process includes:

[0077] Emission signal data were collected from historical courtyards under different burial depths, grounding resistivity, and equipment distribution conditions. Kalman filtering algorithm was used to preprocess the historical signals to eliminate environmental noise and power frequency interference. At the same time, the accuracy of the evaluation results was normalized to ensure data consistency.

[0078] Based on the filtered signal, a spectrum analysis is performed using Fast Fourier Transform combined with Hanning window function. A bandpass filter bank is constructed for the 1Hz fundamental frequency and the 2, 4, and 8Hz harmonics. The amplitude and phase information of each frequency component are identified by the peak detection algorithm, and the fundamental frequency signal and harmonic signals are extracted.

[0079] Based on the extraction of fundamental frequency and harmonic signals, a frequency domain phase dewinding algorithm is used to calculate the phase offset sequence of each harmonic relative to the fundamental frequency. Combined with the buried equipment depth, grounding resistivity, equipment distribution coordinates, and historical assessment accuracy, a graph neural network is constructed: using depth, resistivity, historical assessment accuracy, and the phase offset of each harmonic relative to the fundamental frequency as graph node features, and the equipment distribution relationship as edges, the parameter mapping relationship is learned through graph convolutional layers, and the association dictionary of harmonic phase offset and assessment accuracy is output to complete the construction of the harmonic coefficient dictionary.

[0080] The device distribution map, depth and resistivity of the current target area are collected in real time. The most similar combination of fundamental frequency and harmonic frequency of environmental parameters is retrieved in the digital dictionary through the Euclidean distance matching algorithm to obtain the matching fundamental frequency and harmonic frequency coefficients k2, k4 and k8.

[0081] Based on the matched fundamental frequency and harmonic coefficients k2, k4, and k8, combined with a direct digital frequency synthesis algorithm, a multi-channel harmonic sequence is generated.

[0082] The fundamental and harmonic sequences are discretized using 10kSPS sampling to obtain time-domain sequence points. Vector addition is then performed at the same time node n. Simultaneously, a Chebyshev polynomial interpolation algorithm is used to smooth out high-frequency abrupt changes in the composite signal. It should be further noted that in this embodiment, a genetic algorithm is used in conjunction with the high-frequency abrupt change information in the composite signal to select a Chebyshev node set to optimize the interpolation accuracy, eliminate sampling noise and phase discontinuities, generate a pure composite transmission signal, and ensure the phase consistency and amplitude stability of each frequency component.

[0083] It should be further explained that the process of obtaining the power and grounding location of the transmitting end in this embodiment includes:

[0084] Select the coordinates of the initial grounding candidate point according to the distribution map of buried equipment, and generate a sweep frequency current at the candidate point location according to the set sweep frequency current range;

[0085] The generated sweep frequency current is injected into the ground electrode through a power amplifier. At the same time, the voltage response is measured using the four-electrode method, and the amplitude and phase angle of the complex impedance are extracted using a lock-in amplifier to obtain impedance spectrum data.

[0086] It should be further explained that one implementation method for obtaining impedance spectrum data in this embodiment is as follows:

[0087] First, the coordinates of the initial grounding candidate point are selected based on the distribution map of buried equipment. Direct digital frequency synthesis technology is used to generate a swept frequency current signal containing a 1Hz fundamental frequency and 2, 4, and 8Hz harmonic frequencies. The current amplitude is controlled within the range of 10mA-100mA by a digitally controlled attenuator.

[0088] After the signal is amplified by the configured power amplifier, it is injected into the soil through a copper grounding electrode. The electrode is a silver-plated copper rod with a diameter of 10mm and a length of 50cm, buried at a depth of 0.8m to avoid interference from surface stray currents.

[0089] When measuring voltage response using the four-electrode method, the electrodes are arranged in the Wenner arrangement. Preferably, the current electrodes are spaced 5m apart and the voltage electrodes are spaced 2m apart and located at the midpoint of the current electrodes. They are connected to the measurement system via insulated cables.

[0090] The preamplifier differentially amplifies the picked-up voltage signal, and after filtering out high-frequency noise by an 8th-order Butterworth low-pass filter, it is converted from analog to digital by an ADC.

[0091] The lock-in amplifier uses a 1Hz fundamental frequency as a reference signal and extracts the voltage response at each frequency through a phase-sensitive detection algorithm: First, the reference signal is frequency-multiplied to generate 2Hz, 4Hz, and 8Hz reference components, which are then mixed with the input signal. After low-pass filtering, the in-phase component (real part) and quadrature component (imaginary part) at each frequency point are obtained. The phase angle is calculated using the arctangent function, and the amplitude is obtained by taking the square root of the sum of the squares of the real and imaginary parts. To suppress environmental noise, this embodiment preferably uses a 50-times average sampling technique with a sampling interval of 100ms to ensure improved signal-to-noise ratio.

[0092] During the data processing stage, temperature drift compensation is performed on the real and imaginary data at each frequency point, and the influence of electrode contact resistance is eliminated by least squares fitting.

[0093] Finally, impedance spectrum data containing frequencies of 1, 2, 4, and 8 Hz are generated. For each frequency, the real part, imaginary part, amplitude, and phase angle of the complex impedance are recorded, forming a complete impedance spectrum matrix.

[0094] Based on impedance spectrum data and burial depth, the distribution function of Cole-Cole parameters with depth and the impedance spectrum prediction function are obtained by fitting the Levenberg-Marquardt algorithm. The fitted function is then embedded into the Cole-Cole model to obtain a three-dimensional soil electrical property model.

[0095] It should be further explained that one way to obtain the three-dimensional soil electrical property model in this embodiment is as follows:

[0096] After acquiring impedance spectrum data of 1Hz fundamental frequency and 2, 4 and 8Hz harmonic frequencies, the real and imaginary parts of the complex impedance at each frequency point are integrated with the corresponding burial depth information into a dataset, and the data is normalized to map the impedance amplitude and phase to the [0, 1] interval to eliminate the influence of dimensions.

[0097] Based on the Cole-Cole model, the Levenberg-Marquardt algorithm is used for parameter fitting. It should be further noted that the Cole-Cole model in this embodiment is used to describe the change of soil electrical properties with frequency, involving parameters such as DC resistance, polarizability, time constant and frequency exponent.

[0098] Using burial depth as the independent variable, initial parameter values ​​are first set. The objective function is constructed by minimizing the mean square error between the measured impedance data and the model prediction values. During the algorithm iteration, the damping parameters are dynamically adjusted to balance the optimization characteristics of the gradient descent method and the Gauss-Newton method. The relevant parameters are updated in each iteration until the mean square error converges to a set threshold, thereby obtaining the model parameters corresponding to each depth. It should be further noted that the initial parameter values ​​in this embodiment include, but are not limited to, the average impedance amplitude at each depth, polarizability, time constant, and frequency exponent.

[0099] Based on the fitted parameters, a distribution function of depth and model parameters is constructed. A cubic spline interpolation algorithm is used to smooth discrete depth points, generating continuous parameter variation curves. Simultaneously, the impedance spectrum prediction function for each frequency is combined with the depth distribution function to establish a three-dimensional parameter space with depth and frequency as variables. Finally, these functions and parameter relationships are embedded into the Cole-Cole model. Through gridding, a three-dimensional model containing electrical property parameters such as soil resistivity and polarization characteristics at different depths and frequencies is generated, visually presenting the variation patterns of soil electrical properties in spatial and frequency dimensions. For example, the gridding process involves a 2-meter spacing horizontally and a 0.5-meter spacing vertically.

[0100] Input the target area, the distribution map of buried equipment, the three-dimensional soil electrical property model, and the coordinates of the initial grounding candidate points into the modeling software. Use the modeling software to generate a buffer zone with a radius of d for the buried equipment. At the same time, generate grid points in the remaining area at 2m intervals, filter out points that exceed the boundary, and output a gridded candidate point set.

[0101] Based on the coordinates of the gridded candidate point set and the correlation distribution of each candidate point with the three-dimensional soil electrical property parameters in Cole-Cole with depth, the fitness function constructed by depth and signal-to-noise ratio, depth and power intensity and depth signal quality and the constraints constructed by signal-to-noise ratio, power and signal, the input sequence of the particle swarm algorithm is obtained.

[0102] The particle swarm optimization algorithm is integrated into the three-dimensional soil electrical property model of the modeling software. The input sequence of the particle swarm optimization algorithm is input into the particle swarm optimization algorithm and combined with the simulation algorithm. For each candidate point, the minimum transmission power and the maximum signal quality score are taken as the objectives. The signal transmission and reception simulation is carried out to obtain the coordinates of the candidate point with the minimum transmission power and the maximum signal quality score, as well as the corresponding transmission power.

[0103] It should be further explained that one method for obtaining the candidate point coordinates and corresponding transmission power in this embodiment is as follows:

[0104] First, the gridded candidate point set is spatially associated with the three-dimensional soil electrical property model, and a depth-electrical property parameter mapping table is constructed for each candidate point (x,y); the three-dimensional soil electrical property model contains resistivity and polarization parameters at different depths at frequencies of 1, 2, 4, and 8 Hz.

[0105] Meanwhile, a fitness function is constructed based on historical data. Specifically, the fitness function is constructed by comprehensively evaluating the impact of depth on signal-to-noise ratio and the nonlinear relationship between power intensity and signal quality, and constraints are set, specifically, an upper limit for power and a lower limit for signal-to-noise ratio.

[0106] When using the particle swarm optimization algorithm for optimization, each particle is encoded as a four-dimensional vector [(x, y), z, P], where (x, y) are the planar coordinates of the candidate point, z is the burial depth, and P is the emission power;

[0107] During particle initialization, (x, y) is randomly selected from the gridded candidate points, z follows a uniform distribution, P is preset with an initial value based on the soil electrical properties (a larger value is taken for high-resistivity soil), and the velocity vector is randomly generated in the range of [-1, 1].

[0108] During the iteration process, the following operations are performed on each particle: the propagation of 1, 2, 4, and 8 Hz signals at (x, y, z) is simulated based on a three-dimensional soil model, the attenuation effect of soil impedance on each frequency point is calculated, and the signal-to-noise ratio at the receiver is predicted in combination with the transmit power; the energy loss of the signal penetrating the soil is evaluated based on the electrical characteristic parameters corresponding to depth z, and the signal-to-noise ratio, power consumption, depth construction difficulty, and other indicators are input into the fitness function to calculate the comprehensive score.

[0109] By comparing individual optimality and global optimality, the particle velocity and position are updated, and the coordinates and power that exceed the boundary are truncated. For example, if the depth exceeds 3m, it is set to 3m.

[0110] When the number of iterations reaches the preset value or the global optimal score converges, the (x, y, z) coordinates and power P corresponding to the optimal particle are extracted, and then verified by simulation algorithm: input the scheme into the three-dimensional soil model, simulate multi-frequency signal transmission and reception, and confirm that the signal-to-noise ratio of each frequency point meets the detection requirements. For example, the signal-to-noise ratio of 1Hz is ≥30dB, and the power consumption is the minimum value among the candidate schemes. Finally, the optimal grounding position coordinates and transmission power are output.

[0111] This process significantly improves the accuracy and efficiency of buried equipment detection through the synergistic effect of multi-level signal processing and intelligent optimization algorithms. At the signal generation stage: based on historical data spectral analysis and a frequency harmonic coefficient dictionary constructed using graph neural networks, intelligent mapping between environmental parameters and signal characteristics is achieved. Combined with discretized vector synthesis and Chebyshev interpolation correction, noise interference and phase distortion are effectively suppressed, ensuring that the composite transmitted signal maintains frequency purity and phase consistency in complex soil environments. At the power optimization stage: through swept-frequency impedance measurement and deep fitting of the Cole-Cole model, a three-dimensional soil electrical property model is constructed, accurately characterizing the dynamic changes in resistivity with depth / frequency. Furthermore, the particle swarm optimization algorithm is used to integrate spatially gridded candidate points and multi-objective constraints, achieving collaborative intelligent optimization of grounding location and transmitted power. The core effect of this dual-path technology framework is that: in the signal generation stage, historical knowledge-driven and real-time environment matching ensures the adaptive capability of the transmitted signal to soil electrical properties and equipment distribution; in the power configuration stage, electrical property modeling and multi-objective optimization maximize signal propagation quality while reducing energy consumption. The combination of these two approaches forms a closed loop, ultimately achieving comprehensive benefits such as improved detection accuracy, reduced system energy consumption, and enhanced environmental adaptability, providing technical support for the reliable detection of complex underground facilities.

[0112] Furthermore, the steps for obtaining the amplified multi-frequency voltage signal sequence include:

[0113] The buried equipment detection voltage signal is acquired and initially amplified by a differential amplifier to obtain a preliminary differential voltage signal.

[0114] The signal strength assessment model, constructed based on the preliminary differential voltage signal and a comprehensive fuzzy algorithm, is used to obtain the assessment strength of the preliminary acquired signal.

[0115] Based on the evaluation intensity of the initially acquired signal and combined with the preset three-level gain discrimination interval, the signal with different evaluation intensity is amplified to obtain the multi-frequency voltage signal sequence after amplification.

[0116] The three-level gain discrimination interval is constructed by the signal strength interval and the set three-level gain coefficient, and each signal strength interval corresponds one-to-one with each level of gain coefficient.

[0117] It should be further explained that one implementation method for obtaining the multi-frequency voltage signal sequence after gain in this embodiment is as follows:

[0118] The buried equipment's detected voltage signal is connected to a differential amplifier via an A-frame electrode. The 140dB common-mode rejection ratio eliminates soil potential difference interference, and the signal is initially amplified at 1x gain to output a preliminary differential voltage signal. After being filtered by an RC low-pass filter, the signal is input to a window comparator module consisting of two LM393 voltage comparators.

[0119] The window comparator sets three signal strength thresholds through a resistor network: extremely weak signal voltage range: <90μV, medium signal voltage range: 90μV-1mV, and strong signal voltage range: >1mV. A hysteresis mechanism is set in the 90-110μV range: when the signal fluctuates in this range, the comparator output remains unchanged for 5ms to prevent gain oscillation.

[0120] The STM32F407 microcontroller reads the comparator output status in real time and verifies the comparator status every 100ms via a hardware watchdog circuit. In case of an anomaly, the default 1000x gain channel is forcibly enabled.

[0121] Based on the signal strength level output by the comparator, the microcontroller controls the analog switch via the SPI interface to select the corresponding gain channel. Specifically, a very weak signal triggers the 10,000x gain channel; a medium signal switches to the 3,000x gain channel; and a strong signal maintains the 1,000x gain. During gain switching, a ramp signal controls the change in the resistance of the digital potentiometer to avoid transient interference caused by signal abrupt changes.

[0122] The amplified analog signal is limited to ±4.5V by a Schottky diode clamping circuit and then input to the ADC. The ADC has a built-in Sinc5 filter to suppress out-of-band noise. The sampled data is processed by a digital gain compensation module, specifically by multiplying the ADC output value by a compensation coefficient 1 / G according to the current analog gain multiple G, thereby normalizing the dynamic range and finally generating a multi-frequency voltage signal sequence containing time-domain voltage values ​​at frequencies of 1, 2, 4, and 8 Hz.

[0123] Furthermore, the process for obtaining the coordinates of the damaged points and the grading assessment results of the buried equipment in this embodiment includes:

[0124] Based on the multi-frequency voltage signal sequence after gain, the multi-frequency voltage signal is framed by moving average filtering and Hanning window function to obtain the framed multi-frequency voltage signal sequence.

[0125] Based on the multi-frequency voltage signal sequence after framing, the time-domain signal corresponding to each frame is converted into frequency-domain data through the Fourier transform algorithm, and the real and imaginary parts of each frequency component are calculated to obtain a composite spectrum matrix; the composite spectrum matrix includes the real part, imaginary part, amplitude and phase corresponding to each frequency in the multi-frequency voltage signal.

[0126] Based on the composite spectrum matrix, a Costas ring is constructed with the fundamental frequency as the reference to automatically track the fundamental frequency drift. Based on the locked fundamental frequency position, the frequency coordinates of each frequency in the multi-frequency voltage signal are obtained through the FFT spectrum matrix search algorithm, and the real and imaginary parts of the corresponding spectrum components are extracted to construct a complex set of four frequency points.

[0127] The original phase is calculated based on the real and imaginary parts of each frequency in the four-frequency complex set, and the phase angle at the corresponding frequency is determined by the arctangent function of the original phase.

[0128] Based on the original phase combined with the corresponding initial phase difference, the calibrated phase value is obtained;

[0129] Based on the calibrated phase values, a normalized phase difference array is obtained. in This represents the phase difference between the phase corresponding to the q-th frequency and the phase corresponding to the fundamental frequency;

[0130] based on The consistency score S of all phase differences is obtained through a multi-frequency consistency check algorithm.

[0131] Based on the consistency score of all phase differences and the preset damage point discrimination threshold, the damage point discrimination direction is obtained; the damage point discrimination threshold includes: rear safety zone: S≥0.7(q-1); forward warning zone: S≤0.3(q-1);

[0132] Based on the direction of the acquired damage point, the phase change value between the maximum harmonic frequency and the fundamental frequency of adjacent sampling points, and the moving distance of the receiving end are collected;

[0133] The phase change rate is obtained based on the ratio of the phase change value to the movement distance of the A-frame. When the phase change rate corresponding to three consecutive adjacent sampling points is greater than the damage judgment threshold, the coordinates of the damage point are obtained by combining the configured GPS module.

[0134] It should be further explained that one method for obtaining the coordinates of the damaged point in this embodiment is as follows:

[0135] The multi-frequency voltage signal sequence after gaining is preprocessed. A moving average filtering algorithm with a window size of 10 sampling points is used to eliminate random noise. Then, the signal is framed using a Hanning window function, for example, with 1024 sampling points per frame and 50% overlap between frames to reduce spectral leakage. A 1024-point Fast Fourier Transform is performed on the framed time-domain signal to calculate the real and imaginary parts of each frequency component, generating a composite spectrum matrix containing amplitude and phase information.

[0136] A Costas loop is constructed based on a 1Hz fundamental frequency. In-phase and quadrature local oscillator signals are generated by a numerically controlled oscillator (CNC). These signals are compared with the input signal in phase, and the error is smoothed by a loop filter before being fed back to the CNC oscillator to achieve automatic frequency drift tracking. Based on the locked fundamental frequency position, the coordinates of the 2nd, 4th, and 8th octaves are searched in the FFT spectrum matrix, and the corresponding real and imaginary parts are extracted to construct a complex set of four frequency points. The original phase is calculated using the arctangent function and calibrated using the initial phase difference built in before transmission to obtain the precise phase value of each frequency point. Then, a normalized phase difference array is calculated.

[0137] Using a multi-frequency consistency check algorithm, the consistency score S of the phase difference array is determined based on a preset threshold. Combined with the rules for determining the rear safety zone and the forward warning zone, the direction of the A-frame relative to the damage point is determined. If it is determined to be close to the damage point, the system collects the phase change values ​​of the 8Hz harmonic and 1Hz fundamental frequencies from adjacent sampling points in real time. Combined with the movement distance of the A-frame, the phase change rate is calculated. When the phase change rate of three consecutive adjacent sampling points exceeds the preset threshold, a high-precision RTK differential GPS module is triggered to acquire three-dimensional coordinates (longitude, latitude, and elevation). Combined with the buried depth estimated by the soil electrical property model, the precise location of the damage point is finally determined. The coordinates and related parameters are stored in JSON format, and a purple crosshair is displayed on the screen.

[0138] It should be further explained that, in addition to providing the above-mentioned method for locating damage points, this embodiment can also use an A-frame to locate damage points in the anti-corrosion layer. When the two grounding points of the A-frame are behind the damage point, the measured phase difference relative to the high-frequency signal is -180° to 180°, and a forward arrow indicates the direction of the damage point; when the two grounding points of the A-frame are in front of the damage point, the measured phase difference relative to the high-frequency signal is 180° to 270°, and a backward arrow indicates the direction of the damage point; the position of the damage point relative to the A-frame detection point is determined by judging the phase difference in the harmonic signal relative to the high-frequency signal.

[0139] Based on phase difference array The amplitude of the signal at the corresponding frequency is used to obtain the amplitude attenuation rate after passing through the damage point through an attenuation ratio algorithm;

[0140] Based on the phase difference and moving distance of adjacent sampling points, the gradient of signal phase change near the damage point is obtained through a slope extraction algorithm.

[0141] Based on the time-domain signal corresponding to the maximum harmonic frequency, the stability index of the high-frequency signal is obtained through wavelet packet transform decomposition and energy entropy algorithm.

[0142] Based on the signal phase abrupt change gradient, amplitude attenuation rate, and high-frequency signal stability index corresponding to the damage point coordinates, the damage level and characteristic index are obtained through a logical rule matching algorithm; the logical rule matching algorithm is constructed by combining the amplitude attenuation rate, signal phase abrupt change gradient, and high-frequency signal stability index.

[0143] Based on the damage level, characteristic indicators, and preset maintenance strategies, an algorithm is generated to output an assessment report and maintenance strategy.

[0144] First, for the 1Hz fundamental frequency and 8Hz harmonics, their corresponding amplitudes A1 and A8 are extracted from the composite spectrum matrix. The amplitude attenuation rate is obtained by calculating the attenuation ratio A8 / A1, which directly reflects the degree of energy loss of the signal after passing through the damage point. Then, based on the adjacent sampling points... The phase difference data, combined with the A-frame movement distance information, is used to fit the phase difference change with distance using the least squares method. The value at the point with the maximum slope in the fitted curve is extracted to determine the signal phase change gradient near the damage point.

[0145] For high-frequency stability assessment, a time-domain signal at an 8Hz harmonic is selected. Wavelet packet transform is used to decompose it into multiple frequency bands, and the energy proportion of each band is calculated. Then, the energy entropy value is calculated using the information entropy formula. A higher entropy value indicates a more disordered energy distribution in that frequency band, meaning poorer high-frequency signal stability. Simultaneously, regarding... Phase difference data is used to calculate its standard deviation. The smaller the standard deviation, the higher the phase consistency of the multi-frequency signal and the more significant the damage point characteristics.

[0146] Finally, data such as amplitude attenuation rate, phase abrupt change gradient, high-frequency stability index, and multi-frequency consistency standard deviation are input into a logical matching algorithm model constructed based on preset rules. By setting different threshold ranges, each index is matched and compared with the judgment criteria for severe damage, moderate damage, slight corrosion, and coating defects to determine the damage level of the buried equipment and output characteristic indicators including parameters such as amplitude attenuation rate, phase abrupt change gradient, and high-frequency stability, providing data support for subsequent maintenance strategy formulation. Furthermore, in this embodiment, the judgment criteria for severe damage, moderate damage, slight corrosion, and coating defects are specifically set by those skilled in the art based on historical information and assessment needs, and will not be elaborated here.

[0147] This embodiment significantly improves the accuracy and reliability of buried equipment damage detection through the synergistic optimization of multi-level dynamic gain control and intelligent damage analysis algorithms. In the signal conditioning stage: based on an adaptive amplification mechanism with a three-level gain discrimination interval, combined with hardware hysteresis protection and digital gain compensation, a distortion-free dynamic range extension from microvolts to millivolts is achieved, effectively overcoming saturation or missed detection problems caused by drastic signal strength fluctuations in the soil environment, ensuring the integrity and comparability of multi-frequency voltage signal sequences. In the damage analysis stage: through frame-based spectrum processing and Costas ring fundamental frequency tracking technology, the accurate improvement of multi-frequency signal phase information is ensured. Furthermore, by utilizing multi-frequency phase consistency verification and direction discrimination algorithms, combined with the correlation analysis of phase change rate and spatial movement, dual verification of direction and distance for damage point location was achieved, significantly reducing the false alarm rate. In addition, this process, based on the synergistic analysis of amplitude attenuation rate, phase abrupt change gradient, and high-frequency stability indicators, dynamically correlates physical characteristics with damage level through logical rule matching. This design breaks through the limitations of traditional single-parameter criteria. Specifically, the phase gradient reflects the electromagnetic field distortion intensity at the damage edge, the amplitude attenuation rate quantifies the degree of energy loss, and the high-frequency energy entropy characterizes the signal distortion dispersion characteristics; the three together construct the characteristic fingerprint of the damage state. Ultimately, a closed-loop system of signal gain, precise positioning, and quantitative classification is formed, achieving comprehensive benefits such as measurable weak signals under complex interference, traceable damage location, and assessable damage level, providing reliable technical support for intelligent operation and maintenance of underground facilities.

[0148] Example 2

[0149] Please see Figure 2 Another embodiment of the present invention provides: a damage detection system for buried equipment in courtyards based on intelligent sensors, comprising: a transmission configuration module, a grounding selection module, a gain module and a positioning evaluation module;

[0150] The transmission configuration module, based on the intelligent sensor transmitter, combines a pure composite transmission signal obtained by combining the base frequency and harmonic coefficient dictionary, and tests and determines the initial phase difference and embeds the determined initial phase difference into the receiver.

[0151] The grounding selection module, based on the distribution map of buried equipment and the frequency sweep signal of the grounding loop, determines the power and grounding location of the transmitter by combining the Cole-Cole model with the dual-mode adaptation rule.

[0152] The gain module transmits a corresponding pure composite transmission signal based on the determined transmitter power and grounding location, and uses the receiver to determine and amplify the acquired information in combination with a preset three-level adaptive gain to obtain a multi-frequency voltage signal sequence after amplification.

[0153] The positioning and evaluation module, based on a multi-frequency voltage signal sequence combined with an initial phase difference, uses a fast Fourier transform algorithm and a damage point assessment model configured with bidirectional positioning rules to obtain the coordinates of the damage point and the graded assessment results of the buried equipment.

[0154] Example 3

[0155] An electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement a method for detecting damage points of buried devices in courtyards based on intelligent sensors.

[0156] A computer-readable storage medium storing computer instructions that, when executed, perform a method for detecting damage points in buried garden equipment based on intelligent sensors.

[0157] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments under the guidance of the present invention without departing from the spirit and scope of the claims. All of these variations are within the protection scope of the present invention.

[0158] If the technical solution disclosed herein involves personal information, the product using this technical solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution disclosed herein involves sensitive personal information, the product using this technical solution has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set up to inform users that they have entered the scope of personal information collection and that personal information will be collected. If an individual voluntarily enters the collection scope, it is deemed that they have agreed to the collection of their personal information; or on the personal information processing device, with clear signs / information informing users of the personal information processing rules, authorization is obtained from the individual through pop-up information or by asking the individual to upload their personal information; wherein, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

Claims

1. A method for detecting damage points of buried equipment in courtyards based on intelligent sensors, characterized in that, include: Configure a smart sensor, combine the pure composite transmission signal obtained by combining the base frequency and harmonic coefficient dictionary at the smart sensor transmitter, test and determine the initial phase difference, and embed the determined initial phase difference into the receiver. Based on the distribution map of buried equipment and the frequency sweep signal of the grounding loop, the power and grounding location of the transmitter are determined by using the Cole-Cole model and dual-mode adaptation rules. Based on the determined transmitter power and grounding location, a corresponding pure composite transmission signal is transmitted, and the receiver uses a preset three-level adaptive gain to determine and amplify the acquired information to obtain the amplified multi-frequency voltage signal sequence. Based on the multi-frequency voltage signal sequence combined with the initial phase difference, the damage point coordinates and graded assessment results of buried equipment are obtained by using the Fast Fourier Transform algorithm and a damage point assessment model configured with bidirectional positioning rules.

2. The method for detecting damage points of buried equipment in courtyards based on intelligent sensors as described in claim 1, characterized in that, The process of acquiring the pure composite transmission signal includes: The accuracy of historical transmission signal information and corresponding evaluation results under different depths, grounding resistivity and buried equipment distribution conditions in the historical courtyard is obtained and preprocessed using a filtering algorithm. The fundamental frequency signal and harmonic frequency signal information are obtained by combining filtered historical transmitted signal information with spectrum analysis algorithms. Based on the fundamental frequency signal and harmonic signal components, the phase offset of each harmonic relative to the fundamental frequency is obtained through frequency domain analysis algorithms; A frequency doubling coefficient dictionary is constructed based on the depth of the corresponding buried equipment, the grounding resistivity of the buried area, the distribution status of the buried equipment, the phase offset of each frequency doubling relative to the fundamental frequency, and the corresponding historical evaluation accuracy, combined with a graph neural network.

3. The method for detecting damage points of buried equipment in courtyards based on intelligent sensors as described in claim 2, characterized in that, The process of acquiring the pure composite transmission signal also includes: The distribution status, depth, and grounding resistivity of buried equipment in the courtyard area corresponding to the current target area are obtained in real time. The fundamental frequency and harmonic frequency sequence of the current target area are obtained by combining the matching algorithm with the harmonic coefficient dictionary. A multi-channel frequency doubling sequence is obtained by combining the fundamental frequency matched to the current target region with the frequency doubling coefficients obtained from the matching. The fundamental frequency and multi-channel harmonic frequency sequences are discretized, and multi-channel data vector addition is performed on the frequency points corresponding to each discrete frequency at the same timing node to obtain the composite transmission signal. A pure composite transmitted signal is obtained by interpolation correction based on the composite transmitted signal and Chebyshev algorithm.

4. The method for detecting damage points of buried equipment in courtyards based on intelligent sensors as described in claim 3, characterized in that, The process of obtaining the power and grounding location of the transmitting end includes: Select the coordinates of the initial grounding candidate point according to the distribution map of buried equipment, and generate a sweep frequency current at the candidate point location according to the set sweep frequency current range; The generated sweep frequency current is injected into the ground electrode through a power amplifier. At the same time, the voltage response is measured using the four-electrode method, and the amplitude and phase angle of the complex impedance are extracted using a lock-in amplifier to obtain impedance spectrum data. Based on impedance spectrum data and burial depth, the distribution function of Cole-Cole parameters with depth and the impedance spectrum prediction function are obtained by fitting the Levenberg-Marquardt algorithm. The fitted function is then embedded into the Cole-Cole model to obtain a three-dimensional soil electrical property model. Input the target area, the distribution map of buried equipment, the three-dimensional soil electrical property model, and the coordinates of the initial grounding candidate points into the modeling software. Use the modeling software to generate a buffer zone with a radius of d for the buried equipment. At the same time, generate grid points in the remaining area at 2m intervals, filter out points that exceed the boundary, and output a gridded candidate point set.

5. The method for detecting damage points of buried equipment in courtyards based on intelligent sensors as described in claim 4, characterized in that, The process of obtaining the power and grounding location of the transmitting end also includes: Based on the coordinates of the gridded candidate point set and the correlation distribution of each candidate point with the three-dimensional soil electrical property parameters in Cole-Cole with depth, the fitness function constructed by depth and signal-to-noise ratio, depth and power intensity and depth signal quality and the constraints constructed by signal-to-noise ratio, power and signal, the input sequence of the particle swarm algorithm is obtained. The particle swarm optimization algorithm is integrated into the three-dimensional soil electrical property model of the modeling software. The input sequence of the particle swarm optimization algorithm is input into the particle swarm optimization algorithm and combined with the simulation algorithm. For each candidate point, the minimum transmission power and the maximum signal quality score are taken as the objectives. The signal transmission and reception simulation is carried out to obtain the coordinates of the candidate point with the minimum transmission power and the maximum signal quality score, as well as the corresponding transmission power.

6. The method for detecting damage points of buried equipment in courtyards based on intelligent sensors as described in claim 5, characterized in that, The steps for obtaining the amplified multi-frequency voltage signal sequence include: The buried equipment detection voltage signal is acquired and initially amplified by a differential amplifier to obtain a preliminary differential voltage signal. The signal strength assessment model, constructed based on the preliminary differential voltage signal and a comprehensive fuzzy algorithm, is used to obtain the assessment strength of the preliminary acquired signal. Based on the evaluation intensity of the initially acquired signal and combined with the preset three-level gain discrimination interval, the signal with different evaluation intensity is amplified to obtain the multi-frequency voltage signal sequence after amplification. The three-level gain discrimination interval is constructed by the signal strength interval and the set three-level gain coefficient, and each signal strength interval corresponds one-to-one with each level of gain coefficient.

7. The method for detecting damage points of buried equipment in courtyards based on intelligent sensors as described in claim 6, characterized in that, The process for obtaining the coordinates of the damaged points and the graded assessment results of the buried equipment includes: Based on the multi-frequency voltage signal sequence after gain, the multi-frequency voltage signal is framed by moving average filtering and Hanning window function to obtain the framed multi-frequency voltage signal sequence. Based on the multi-frequency voltage signal sequence after framing, the time-domain signal corresponding to each frame is converted into frequency-domain data through the Fourier transform algorithm, and the real and imaginary parts of each frequency component are calculated to obtain a composite spectrum matrix; the composite spectrum matrix includes the real part, imaginary part, amplitude and phase corresponding to each frequency in the multi-frequency voltage signal; Based on the composite spectrum matrix, a Costas ring is constructed with the fundamental frequency as the reference to automatically track the fundamental frequency drift. Based on the locked fundamental frequency position, the frequency coordinates of each frequency in the multi-frequency voltage signal are obtained through the FFT spectrum matrix search algorithm, and the real and imaginary parts of the corresponding spectrum components are extracted to construct a complex set of four frequency points. The original phase is calculated based on the real and imaginary parts of each frequency in the four-frequency complex set, and the phase angle at the corresponding frequency is determined by the arctangent function of the original phase.

8. The method for detecting damage points of buried equipment in courtyards based on intelligent sensors as described in claim 7, characterized in that, The process for obtaining the coordinates of the damaged points and the grading assessment results of the buried equipment also includes: Based on the original phase combined with the corresponding initial phase difference, the calibrated phase value is obtained; Based on the calibrated phase values, a normalized phase difference array is obtained. in This represents the phase difference between the phase corresponding to the q-th frequency and the phase corresponding to the fundamental frequency; based on The consistency score S of all phase differences is obtained through a multi-frequency consistency check algorithm. Based on the consistency score of all phase differences and the preset damage point discrimination threshold, the damage point discrimination direction is obtained; the damage point discrimination threshold includes: rear safety zone: S≥0.7(q-1); forward warning zone: S≤0.3(q-1); Based on the direction of the acquired damage point, the phase change value between the maximum harmonic frequency and the fundamental frequency of adjacent sampling points, and the moving distance of the receiving end are collected; The phase change rate is obtained based on the ratio of the phase change value to the movement distance of the A-frame. When the phase change rate corresponding to three consecutive adjacent sampling points is greater than the damage judgment threshold, the coordinates of the damage point are obtained by combining the configured GPS module.

9. The method for detecting damage points of buried equipment in courtyards based on intelligent sensors as described in claim 8, characterized in that, The process for obtaining the coordinates of the damaged points and the grading assessment results of the buried equipment also includes: Based on phase difference array The amplitude of the signal at the corresponding frequency is used to obtain the amplitude attenuation rate after passing through the damage point through an attenuation ratio algorithm; Based on the phase difference and moving distance of adjacent sampling points, the gradient of signal phase change near the damage point is obtained through a slope extraction algorithm. Based on the time-domain signal corresponding to the maximum harmonic frequency, the stability index of the high-frequency signal is obtained through wavelet packet transform decomposition and energy entropy algorithm. Based on the signal phase abrupt change gradient, amplitude attenuation rate, and high-frequency signal stability index corresponding to the damage point coordinates, the damage level and characteristic index are obtained through a logical rule matching algorithm; the logical rule matching algorithm is constructed by combining the amplitude attenuation rate, signal phase abrupt change gradient, and high-frequency signal stability index. Based on the damage level, characteristic indicators, and preset maintenance strategies, an algorithm is generated to output an assessment report and maintenance strategy.

10. A damage detection system for buried courtyard equipment based on intelligent sensors, implemented according to any one of claims 1-9, characterized in that, include: Transmit configuration module, ground selection module, gain module, and positioning evaluation module; The transmission configuration module, based on the intelligent sensor transmitter, combines a pure composite transmission signal obtained by combining the base frequency and harmonic coefficient dictionary, and tests and determines the initial phase difference and embeds the determined initial phase difference into the receiver. The grounding selection module, based on the distribution map of buried equipment and the frequency sweep signal of the grounding loop, determines the power and grounding location of the transmitter by combining the Cole-Cole model with the dual-mode adaptation rule. The gain module transmits a corresponding pure composite transmission signal based on the determined transmitter power and grounding location, and uses the receiver to determine and amplify the acquired information in combination with a preset three-level adaptive gain to obtain a multi-frequency voltage signal sequence after amplification. The positioning and evaluation module, based on a multi-frequency voltage signal sequence combined with an initial phase difference, uses a fast Fourier transform algorithm and a damage point assessment model configured with bidirectional positioning rules to obtain the coordinates of the damage point and the graded assessment results of the buried equipment.

Images