Panoramic imaging system for power transmission and transformation engineering automation underground panoramic survey

Abstract

The present application relates to power transmission and transformation engineering surveying technical field, disclose a kind of panoramic imaging system of power transmission and transformation engineering automation underground panorama surveying, the system includes data acquisition module, panoramic imaging module and real-time monitoring module.Data acquisition module collects the multi-source detection data of rock-soil medium, pipeline distribution, electromagnetic field data, wherein rock-soil medium data contains static stratum data and dynamic medium parameter.Panoramic imaging module resolves multi-source data by three-dimensional imaging engine, realizes spatial consistency alignment by data reconstruction unit, and the wave field synthesis layer, feature matching layer, correlation analysis layer and imaging layer of model fusion unit carry out multi-modal data fusion and three-dimensional space reorganization, and construct underground space three-dimensional panoramic imaging model.Imaging layer is equipped with multi-view interactive interface, supports user operation model.Real-time monitoring module marks abnormal electromagnetic field area, triggers risk early warning.The system realizes multi-source data fusion and three-dimensional panoramic imaging, improves surveying accuracy and safety.

Description

Technical Field

[0001] This invention relates to the field of power transmission and transformation engineering surveying technology, specifically to an automated underground panoramic surveying and imaging system for power transmission and transformation engineering. Background Technology

[0002] In the construction of power transmission and transformation projects, the survey of the underground space environment is a crucial and fundamental step. Underground space encompasses complex geological structures, the distribution of various pipelines, and electromagnetic field environments, among other factors. These factors are interconnected and mutually influential, directly and significantly impacting the planning, design, construction, and safe operation of power transmission and transformation projects.

[0003] Traditional underground surveying methods have many limitations. On the one hand, a single detection method can often only obtain one or a few types of information about the underground environment. For example, ground-penetrating radar is mainly used to detect the distribution of underground media and pipeline conditions, while electromagnetic methods are mainly used to study the distribution and changes of electromagnetic fields. This limitation leads to a lack of comprehensiveness and integration in the survey results, making it difficult to fully reflect the true condition of the underground space. It is prone to problems such as missing or one-sided information, thereby affecting the scientific nature and accuracy of engineering decisions.

[0004] On the other hand, traditional surveying methods are inefficient and lack precision in data processing and analysis. Different types of survey data are typically processed independently, lacking effective data fusion and collaborative analysis mechanisms. This makes it difficult to establish organic connections between various data sources, hindering the full extraction of the rich information contained within the data. For example, the spatial correlation and interaction between geological data, pipeline data, and electromagnetic field data are difficult to accurately identify and analyze, resulting in the inability to perform panoramic imaging and modeling of underground space.

[0005] Furthermore, traditional surveying techniques have significant limitations in terms of real-time and dynamic monitoring. During the construction and operation of power transmission and transformation projects, the underground environment may undergo dynamic changes, such as minor displacements of strata, pipeline leaks, or abnormal fluctuations in electromagnetic fields. Traditional methods often fail to capture these changes in a timely manner, making it difficult to promptly identify potential safety hazards and thus failing to provide effective guarantees for the safe operation of the project.

[0006] With the development of power transmission and transformation projects towards larger scale, greater complexity, and higher intelligence, higher demands are being placed on underground surveying technology. There is an urgent need for an automated surveying system capable of multi-source data fusion, panoramic imaging, and real-time monitoring to improve the efficiency, accuracy, and comprehensiveness of surveys and meet the needs of modern power transmission and transformation project construction and operation. Existing technologies are no longer adequate to meet these new challenges; therefore, developing an advanced automated underground panoramic surveying and imaging system for power transmission and transformation projects is of significant practical importance and urgency. Summary of the Invention

[0007] The purpose of this invention is to provide an automated underground panoramic survey and imaging system for power transmission and transformation projects to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: an automated underground panoramic survey and imaging system for power transmission and transformation projects, the system comprising:

[0009] The data acquisition module is used to collect underground multi-source detection data of the target area. The multi-source detection data includes a first detection dataset corresponding to the soil and rock media data, a second detection dataset corresponding to the pipeline distribution data, and a third detection dataset corresponding to the electromagnetic field data. The soil and rock media data includes static stratigraphic data retrieved from the geological database and dynamic media parameters obtained through the distributed detection device.

[0010] A panoramic imaging module is used to input the multi-source detection data into a three-dimensional imaging engine for analysis;

[0011] A three-dimensional panoramic imaging model of the underground space is constructed based on the output of the three-dimensional imaging engine. The three-dimensional imaging engine includes a data reconstruction unit and a model fusion unit. The data reconstruction unit is used to perform spatial consistency alignment on the multi-source detection data. The model fusion unit performs multi-modal data fusion based on a preset underground structure template and historical survey data. The model fusion unit includes a wavefield synthesis layer, a feature matching layer, a correlation analysis layer, and an imaging layer connected in sequence. The wavefield synthesis layer is used to perform frequency domain transformation on each dataset in the multi-source detection data to generate a wavefield feature map. The correlation analysis layer is used to model the spatial correlation between the wavefield feature maps corresponding to the multi-source detection data to generate multi-dimensional topological relationships. The imaging layer is used to perform three-dimensional spatial reconstruction based on the multi-dimensional topological relationships and the wavefield feature maps to generate a three-dimensional panoramic imaging model.

[0012] Preferably, the step of modeling the spatial correlation between wavefield feature maps corresponding to multi-source detection data to generate multi-dimensional topological relationships includes:

[0013] A signal feature extraction algorithm is used to identify abnormal response regions in the wavefield feature map, and feature coding sequences corresponding to each type of detection data are generated based on the spectral characteristics of each abnormal response region.

[0014] Calculate the correlation coefficient between anomalous response regions with the same frequency band identification in the feature coding sequences corresponding to any two types of detection data, and determine the multi-dimensional topological relationship between the two types of detection data based on the correlation coefficient; the calculation of the correlation coefficient between anomalous response regions with the same frequency band identification in the feature coding sequences corresponding to any two types of detection data includes:

[0015] When the number of abnormal responses in the feature coding sequences corresponding to any two types of detection data is inconsistent, frequency band expansion compensation is performed based on the distribution characteristics of the highest energy frequency band among those with fewer abnormal responses, and the correlation coefficient between abnormal response regions with the same frequency band identification is calculated based on the compensated sequence.

[0016] Preferably, the data reconstruction unit is specifically used for:

[0017] The first detection dataset, the second detection dataset, and the third detection dataset are time-frequency synchronized according to a preset reference coordinate system to obtain the aligned first detection dataset, the aligned second detection dataset, and the aligned third detection dataset.

[0018] An adaptive filtering method is used to suppress interference signals in the aligned first and second probe datasets, and a phase compensation method is used to correct waveform distortion in the aligned third probe dataset, resulting in a first optimized probe dataset, a second optimized probe dataset, and a third optimized probe dataset.

[0019] Preferably, the first optimized detection dataset includes processed static formation data and processed dynamic medium parameters, and the data reconstruction unit is further used for:

[0020] Analyze the energy decay curves of the processed static stratigraphic data and the processed dynamic medium parameters within historical exploration periods;

[0021] Based on the energy attenuation curve and the wave velocity parameters of the processed static formation data in the specified area, the equivalent impedance parameters of the processed dynamic medium parameters in the specified area are calculated.

[0022] A comprehensive medium dataset is generated based on the processed dynamic medium parameters and their equivalent impedance parameters, and the detection data corresponding to the comprehensive medium dataset is used as the first optimized detection dataset.

[0023] Preferably, the wave field synthesis layer includes a spectrum analysis unit and an energy mapping unit;

[0024] The spectrum analysis unit is used to perform multi-scale time-frequency decomposition on each type of detection dataset contained in the multi-source detection data to extract the corresponding frequency domain feature components from each type of detection dataset.

[0025] The energy mapping unit is used to phase-align and superimpose the frequency domain feature components extracted from each type of detection dataset with the original detection dataset to generate a wavefield feature map.

[0026] Preferably, the wavefield synthesis layer further includes a noise suppression unit, which is used to perform blind source separation and noise reduction processing on the wavefield feature map;

[0027] The process of modeling the spatial correlation between wavefield feature maps corresponding to multi-source detection data includes: performing cross-validation analysis on the denoised wavefield feature maps corresponding to various types of detection data to generate the multi-dimensional topological relationship.

[0028] Preferably, the imaging layer includes a multi-view interactive interface, which allows users to switch profiles, filter signal intensity, and adjust resolution of the three-dimensional panoramic imaging model through control commands.

[0029] The multimodal data fusion based on the preset underground structure template and historical survey data includes: dynamically optimizing the physical property parameters of the underground structure template through an iterative inversion algorithm to adapt to the update of the multidimensional topological relationship;

[0030] The three-dimensional spatial reconstruction based on the multi-dimensional topological relationship and the wave field feature map includes: performing spatial interpolation calculation on the multi-dimensional topological relationship according to the preset imaging weight, and vector superimposing it with the wave field feature map.

[0031] Preferably, the feature matching layer includes a dynamic compensation unit for amplitude compensation of the wavefield feature spectrum based on the medium density gradient of the target region; the operation performed by the dynamic compensation unit includes: establishing a compensation model for the medium absorption coefficient based on the electromagnetic field intensity data of the third detection dataset, and updating the energy distribution of each frequency band in the wavefield feature spectrum through a backpropagation algorithm.

[0032] Preferably, the imaging layer further includes an error feedback unit, used to calculate the residual value between the three-dimensional panoramic imaging model and historical survey data in real time during the vector overlay process; when the residual value exceeds a preset threshold, the data reconstruction unit is triggered to re-execute the time-frequency synchronization processing and update the spatial interpolation parameters of the multi-dimensional topological relationship.

[0033] Preferably, the system further includes a real-time monitoring module for marking areas in the three-dimensional panoramic imaging model where the electromagnetic field strength exceeds a safety threshold; the real-time monitoring module is configured with an early warning triggering mechanism, which generates a risk warning signal with coordinate positioning when an abnormal field strength is detected in the spatially overlapping area of ​​pipeline distribution data and electromagnetic field data.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] The system's data acquisition module can collect multi-source underground detection data of the target area, covering a first detection dataset corresponding to soil and rock media data, a second detection dataset corresponding to pipeline distribution data, and a third detection dataset corresponding to electromagnetic field data. The soil and rock media data integrates static stratigraphic data with dynamic media parameters. This multi-source data acquisition mode breaks through the limitations of traditional single-detection methods, achieving comprehensive capture of multi-dimensional information such as underground geological structure, pipeline distribution, and electromagnetic field environment. This ensures that the survey results fully reflect the true condition of the underground space, providing a rich and comprehensive data foundation for subsequent imaging and analysis, and effectively solving the problem of information loss in traditional methods.

[0036] The 3D imaging engine in the panoramic imaging module significantly improves data processing accuracy and imaging quality through innovative designs of the data reconstruction unit and model fusion unit. The data reconstruction unit performs spatial consistency alignment on multi-source detection data. Through time-frequency synchronization processing, interference signal suppression, and waveform distortion correction, it eliminates spatiotemporal differences and noise interference between different datasets, ensuring data accuracy and consistency. For example, adaptive filtering is used to suppress interference signals in the first and second detection datasets, and phase compensation is performed to correct waveform distortion in the third detection dataset, enabling various types of data to be fused and analyzed under a unified benchmark. Simultaneously, by analyzing the energy attenuation curves of static stratigraphic data and dynamic medium parameters, equivalent impedance parameters are calculated and a comprehensive medium dataset is generated, further enhancing the reliability and completeness of the soil and rock medium data.

[0037] The wavefield synthesis layer of the model fusion unit performs multi-scale time-frequency decomposition and phase-aligned superposition of multi-source detection data through the spectrum analysis unit and energy mapping unit, generating a wavefield feature map that accurately reflects the data characteristics. Simultaneously, the noise suppression unit further improves the map quality through blind source separation and denoising. The dynamic compensation unit of the feature matching layer performs amplitude compensation on the wavefield feature map based on the medium density gradient, establishes a medium absorption coefficient compensation model, and updates the energy distribution through a backpropagation algorithm, effectively addressing the influence of different media on the propagation of the detection signal and improving the accuracy of feature matching. The correlation analysis layer fully explores the spatial correlation between multi-source data through signal feature extraction algorithms and correlation coefficient calculation, generating multi-dimensional topological relationships. Even when the number of anomalous responses is inconsistent, methods such as bandwidth extension compensation ensure the accuracy of correlation calculation, thereby establishing an organic connection between various types of data. The imaging layer performs three-dimensional spatial reconstruction based on multi-dimensional topological relationships and wavefield feature maps. Through preset imaging weights, it performs spatial interpolation calculations and vector superposition to generate a high-precision three-dimensional panoramic imaging model. Simultaneously, the multi-view interactive interface supports users in switching profiles, filtering signal strength, and adjusting resolution to meet the analysis needs of different scenarios. The error feedback unit monitors the residual value in real time during the vector superposition process. When the residual value exceeds the preset threshold, the data reconstruction unit is triggered to reprocess and update the interpolation parameters to ensure the accuracy and reliability of the imaging model.

[0038] The system also includes a real-time monitoring module that can mark areas where electromagnetic field strength exceeds safety thresholds in the 3D panoramic imaging model and is equipped with an early warning triggering mechanism. When abnormal field strength is detected in the spatial overlap area between pipeline distribution data and electromagnetic field data, a risk warning signal with coordinate positioning can be generated in a timely manner. This enables real-time dynamic monitoring of the underground environment and early warning of safety hazards, providing strong protection for the safe operation of power transmission and transformation projects and significantly improving the safety and reliability of the projects. Attached Figure Description

[0039] Figure 1 This is a schematic diagram illustrating the working principle of the automated underground panoramic survey and imaging system for power transmission and transformation engineering described in this invention.

[0040] Figure 2 A flowchart for modeling the spatial correlation of wavefield feature maps and generating multi-dimensional topological relationships;

[0041] Figure 3 Flowchart generated for the integrated media dataset;

[0042] Figure 4 This is a flowchart for multimodal data fusion in the imaging layer. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Please see Figures 1-4 The present invention relates to an automated underground panoramic survey and panoramic imaging system for power transmission and transformation engineering, the specific implementation scheme of which is as follows:

[0045] The data acquisition module is used to collect multi-source underground detection data of the target area. This multi-source detection data includes a first detection dataset corresponding to soil and rock media data, a second detection dataset corresponding to pipeline distribution data, and a third detection dataset corresponding to electromagnetic field data. The soil and rock media data includes static stratigraphic data retrieved from a geological database and dynamic media parameters acquired through a distributed detection device. The distributed detection device can be deployed at multiple detection points on the surface of the target area, using a sensor array to collect dynamic media parameters such as soil resistivity and dielectric constant in real time. This data is then combined with static stratigraphic data such as historical stratigraphic structure and lithological distribution stored in the geological database to form the first detection dataset. The second detection dataset is acquired through pipeline detection radar or electromagnetic induction equipment, recording information such as the spatial location, burial depth, and pipe diameter of power, communication, and water supply / drainage pipelines within the target area. The third detection dataset is collected using electromagnetic survey equipment and includes data on the intensity, distribution characteristics, and frequency characteristics of the underground electromagnetic field in the target area.

[0046] The panoramic imaging module is used to input multi-source detection data into the 3D imaging engine for analysis and to construct a 3D panoramic imaging model of the underground space based on the engine's output. The 3D imaging engine includes a data reconstruction unit and a model fusion unit. The data reconstruction unit performs spatial consistency alignment on the multi-source detection data. Specifically, the data reconstruction unit first performs time-frequency synchronization processing on the first, second, and third detection datasets according to a preset reference coordinate system (such as the National Geodetic Coordinate System). Through timestamp calibration and frequency domain matching, it ensures the consistency of data from different sources in spatial coordinates and time dimensions, resulting in three aligned datasets. The model fusion unit performs multimodal data fusion based on a preset underground structure template and historical survey data. It includes a wavefield synthesis layer, a feature matching layer, a correlation analysis layer, and an imaging layer connected sequentially. The wavefield synthesis layer performs frequency domain transformation on each dataset in the multi-source detection data to generate wavefield feature maps. The correlation analysis layer models the spatial correlation between the wavefield feature maps corresponding to the multi-source detection data, generating multi-dimensional topological relationships. The imaging layer reconstructs three-dimensional space based on multi-dimensional topological relationships and wavefield feature maps to generate a three-dimensional panoramic imaging model.

[0047] The present invention will be further described below with reference to Examples 1 to 5:

[0048] Example 1,

[0049] This embodiment focuses on the optimization of the data reconstruction unit, specifically including time-frequency synchronization processing, interference signal suppression and waveform correction, and static and dynamic data fusion. Each step is closely linked to achieve spatial consistency alignment and accuracy improvement of multi-source detection data, as detailed below:

[0050] The data reconstruction unit first performs time-frequency synchronization processing on the first detection dataset (soil and soil media data), the second detection dataset (pipeline distribution data), and the third detection dataset (electromagnetic field data). This processing uses a preset reference coordinate system (such as the National Geodetic Coordinate System) as a unified reference framework, and eliminates spatiotemporal misalignment between different datasets through timestamp calibration and frequency domain matching. For dynamic media parameters (such as soil resistivity and dielectric constant) collected by distributed detection devices, their spatial coordinates are calibrated in real time using the Global Navigation Satellite System (GNSS) to ensure that the position of each detection point is consistent with the reference coordinate system. At the same time, the clocks of each device are calibrated through a time synchronization protocol (such as Network Time Protocol NTP) to align the sampling time of the dynamic media parameters with the time window of the static stratigraphic data in the geological database. For the second detection dataset acquired by the pipeline detection radar, the radar device's built-in positioning module is matched with the reference coordinate system, and the data is framed according to the detection time series. Each frame of data corresponds to the spatial scan result at a specific time point, and its sampling frequency is unified to the same time interval as the first and third datasets (such as 100 sampling points per second) through a time interpolation algorithm. The time-frequency synchronization of electromagnetic field data needs to consider both time and frequency dimensions: in terms of time, the hardware clock of the electromagnetic survey equipment is synchronized with the reference clock to ensure the accuracy of the timestamp of the electromagnetic field strength data; in terms of frequency, the collected broadband signal is resampled to unify the frequency resolution of different devices to a preset value (such as 1Hz) for subsequent frequency domain analysis.

[0051] After time-frequency synchronization is completed, the data reconstruction unit performs interference signal suppression on the aligned first and second detection datasets. For the static stratigraphic data (such as historical stratigraphic structure and lithological distribution) and dynamic medium parameters in the first detection dataset, an adaptive filtering algorithm is used to remove background noise such as natural electric field noise and industrial electromagnetic interference. The adaptive filtering algorithm uses the minimum mean square error (LMS) as the criterion and recursively updates the filter coefficients to adapt to changes in signal statistical characteristics. For example, for the soil resistivity signal in the dynamic medium parameters, if there is 50Hz power frequency interference, the adaptive notch filter can automatically detect the interference frequency and generate a corresponding notch, attenuating the signal energy in that frequency band while retaining the effective signal components. For the pipeline distribution data in the second detection dataset, since radar echoes are easily scattered by surface metal objects (such as manhole covers and reinforcing bars), an adaptive filtering method based on independent component analysis (ICA) is used to decompose the radar echoes into independent signal components, separating the effective components containing pipeline characteristics from the noise components, thereby suppressing clutter interference.

[0052] For the third detection dataset (electromagnetic field data), the data reconstruction unit uses a phase compensation method to correct waveform distortion. When electromagnetic field signals propagate underground, the inhomogeneity of the dielectric constant and magnetic permeability of the soil and rock medium causes phase delays in signals of different frequency bands, resulting in waveform distortion. The specific steps of phase compensation are as follows: First, the electromagnetic field data is decomposed in the frequency domain to obtain the phase spectrum of each frequency band (e.g., low frequency 0-100Hz, mid-frequency 100-1000Hz, and high frequency above 1000Hz); then, the phase difference of each frequency band signal relative to the reference frequency band (e.g., mid-frequency 500Hz) is calculated to generate a phase correction function; finally, the phase of each frequency band signal is corrected by multiplication in the frequency domain to ensure that the corrected waveform remains consistent on the time axis. For example, if the high-frequency signal has a phase lag of π / 2 relative to the reference frequency band, the phase compensation algorithm will advance the phase of the high-frequency signal by π / 2, eliminating the waveform distortion caused by the phase difference.

[0053] After the above processing, the first optimized exploration dataset contains processed static stratigraphic data and dynamic medium parameters. The data reconstruction unit further performs fusion analysis on the two types of data: extracting static stratigraphic data and dynamic medium parameters from historical exploration periods, calculating signal energy attenuation values ​​at different depths, and plotting energy attenuation curves. This curve reflects the law of signal energy attenuation with increasing stratigraphic depth and can be fitted using an exponential function. ,in This represents the initial value of surface energy. The attenuation coefficient is... For example, if static stratigraphic data for a certain area shows that the signal energy decreases by 10% for every 1 meter increase in depth, then the attenuation coefficient is... .

[0054] Based on the processed static formation data, wave velocity parameters (such as P-wave velocity) in the specified area transverse wave velocity The equivalent impedance parameter is used to deduce the dynamic medium parameters. Equivalent impedance Defined as dielectric density With wave speed The product ( Wave velocity (PV) is used to characterize the ability of a medium to impede wave propagation. For static stratigraphic data, wave velocity parameters can be obtained from geological exploration reports. For example, if the P-wave velocity of a clay layer is 1500 m / s and the S-wave velocity is 800 m / s, combined with the density data of that layer (e.g., 1800 kg / m³), the P-wave impedance can be calculated as follows: The transverse wave impedance is For dynamic medium parameters (such as the dielectric constant acquired in real time). ), through the empirical relationship between dielectric constant and density (such as ,in The density is estimated by taking a constant value, and then the equivalent impedance is calculated by combining the wave velocity parameters of static formation data.

[0055] The data reconstruction unit fuses the processed dynamic medium parameters with their equivalent impedance parameters to generate a comprehensive medium dataset. Specifically, it attaches a corresponding equivalent impedance value to each dynamic medium parameter sample, forming a dataset containing multi-dimensional features. For example, if the dynamic medium parameters of a detection point are a dielectric constant of 20 and a conductivity of 0.01 S / m, its equivalent impedance calculated using the above method is... Then, this sample is represented in the comprehensive dielectric dataset as [dielectric constant = 20, conductivity = 0.01, equivalent impedance = 3.0 × 10⁻⁶]. 6 This comprehensive dataset, serving as the first optimized detection dataset, retains real-time dynamic parameters while incorporating impedance characteristics derived from static data. This enhances the spatial consistency and physical coherence of the soil and rock media data, providing a more accurate input basis for subsequent 3D imaging.

[0056] Throughout the processing flow, the data reconstruction unit achieves spatial alignment and noise suppression of multi-source detection data through multiple steps, including benchmark coordinate system unification, time-frequency synchronization, adaptive filtering, phase compensation, and dynamic-static data fusion. This solves the fusion challenges caused by inconsistent spatiotemporal benchmarks, noise interference, and differences in medium properties among different types of data, laying a data foundation for the construction of the 3D model of the panoramic imaging module. Each processing step is based on signal processing theory and geophysical principles. By combining mathematical algorithms with physical models, the scientific nature and accuracy of data processing are ensured, avoiding the introduction of subjective assumptions or fictitious effect descriptions.

[0057] Example 2,

[0058] This embodiment focuses on the structure and processing flow of the wavefield synthesis layer, covering the specific implementation of the spectrum analysis unit, energy mapping unit, and noise suppression unit. Through operations such as multi-scale time-frequency decomposition, phase-aligned superposition, and blind source separation and denoising, a high signal-to-noise ratio wavefield feature map is generated and a multi-dimensional topological relationship is established, as detailed below:

[0059] The spectral analysis unit of the wavefield synthesis layer performs multi-scale time-frequency decomposition on each type of dataset in the multi-source probe data to extract frequency domain feature components. For the first probe dataset (rock and soil media data), which includes static stratigraphic data (such as lithological distribution and stratigraphic stratification) and dynamic medium parameters (such as soil resistivity and dielectric constant), a continuous wavelet transform algorithm is used for decomposition. Wavelet transform achieves multi-resolution analysis of the signal by convolving wavelet basis functions with the signal at different scales. For example, for the seismic wave velocity curve in the static stratigraphic data, the Mexican cap wavelet is selected as the basis function to analyze the overall structural trend of the stratigraphy at a large scale (low frequency) and to identify subtle features such as local interlayers or fractures at a small scale (high frequency); for the soil resistivity time series in the dynamic medium parameters, the db4 wavelet is used for three-level decomposition to obtain low-frequency approximate components (reflecting the long-term trend of resistivity) and high-frequency detail components (reflecting short-term disturbances). For the second detection dataset (pipeline distribution data), short-time Fourier transform (STFT) is used for time-frequency analysis. The radar echo signal is divided into multiple time windows using a fixed window function. A Fourier transform is performed on the signal within each window to generate a time-frequency matrix, where the horizontal axis represents frequency and the vertical axis represents time (corresponding to spatial location). Matrix element values ​​represent the energy intensity of the frequency component at the corresponding location, thus clearly showing the frequency characteristics and spatial distribution of the pipeline reflection signal. For the third detection dataset (electromagnetic field data), empirical mode decomposition (EMD) is used to decompose the complex electromagnetic field signal into multiple intrinsic mode functions (IMFs). Each IMF corresponds to a different frequency component, achieving adaptive time-frequency decomposition for non-stationary signals.

[0060] The energy mapping unit performs phase-aligned superposition of frequency domain feature components with the original detection dataset to generate a wavefield feature map. The core of phase alignment is to eliminate the phase difference between the frequency domain components and the original data, ensuring the temporal consistency of the superimposed signal. The specific steps are as follows: First, perform an inverse time-frequency transform on the frequency domain feature components extracted from each dataset to restore the time domain signal; then, calculate the cross-correlation function between the time domain signal and the original dataset to determine the time delay corresponding to the maximum cross-correlation value, i.e., the phase difference; next, adjust the phase of the frequency domain feature components to compensate for this time delay, ensuring precise alignment on the time axis; finally, superimpose the adjusted frequency domain feature components with the original dataset point by point to generate a wavefield feature map. Taking pipeline distribution data as an example, assuming that the time delay between a certain high-frequency feature component extracted by STFT (corresponding to the strong reflection of the pipeline's metal material) and the original radar echo after inverse transformation is 20ns, phase compensation advances the phase of this feature component by 20ns to the corresponding radian value (e.g., phase difference). ,in: Indicates phase difference; Indicates frequency; This represents the time interval, where f is the characteristic frequency. (For the delay time), and then superimposed with the original echo, so that the high-frequency features are accurately mapped to the actual spatial location of the pipeline, forming a bright area of ​​energy focusing in the wave field feature spectrum.

[0061] The noise suppression unit performs blind source separation denoising on the wavefield feature map, employing the Independent Component Analysis (ICA) algorithm to separate the signal from the noise. Blind source separation assumes that the source signals are independent and the observed signal is a linear combination of the source signals. The wavefield feature map is considered as an observed signal composed of a mixture of effective signal sources (such as pipeline reflections and formation interface responses) and noise sources (such as environmental electromagnetic interference and instrument noise). The ICA algorithm optimizes the separation matrix to maximize the statistical independence of each separated component, thereby decomposing the map into independent effective signal and noise components. Specifically, the map is first standardized to have a mean of zero and a variance of one. Then, the separation matrix is ​​initialized, and the matrix parameters are iteratively updated using the FastICA algorithm, calculating the negative entropy criterion as a measure of independence until convergence yields the independent components. Finally, the effective components containing the main energy are retained, while the independent components dominated by noise are removed, resulting in the denoised wavefield feature map. For example, in the wave field feature map of electromagnetic field data, the ICA algorithm can separate power frequency interference (manifested as periodic noise at a specific frequency) from the electromagnetic response signal of the underground target. By eliminating the independent components corresponding to power frequency interference, the signal-to-noise ratio of the map is improved.

[0062] After denoising, the wavefield synthesis layer performs cross-validation analysis on the wavefield feature maps corresponding to various types of detection data to generate multi-dimensional topological relationships. Cross-validation analysis judges the consistency and correlation of data by comparing the signal characteristics of different types of maps at the same spatial location. The specific operation is as follows: First, the three types of maps (soil and rock media, pipeline distribution, and electromagnetic field) are spatially registered based on a reference coordinate system to ensure that the same geographical coordinates are consistent in the positions of each map; then, a region of interest (ROI) is selected in the spatially registered map, such as a suspected pipeline anomaly area, and the feature parameters of the area in the three types of maps are extracted (such as the equivalent impedance value of the soil and rock media map, the radar reflection intensity of the pipeline map, and the field strength value of the electromagnetic field map); next, the correlation index between each feature parameter (such as covariance and mutual information) is calculated to evaluate the degree of correlation between different types of data in the area; finally, the degree of correlation of the entire area is represented in the form of a graph structure, with nodes representing spatial location points and edges representing the correlation of feature parameters, forming a multi-dimensional topological relationship. For example, at a certain coordinate point (X,Y,Z), if the geotechnical medium spectrum shows a sudden change in equivalent impedance, the pipeline spectrum shows a strong reflection signal, and the electromagnetic field spectrum detects an anomaly in field strength, then these three types of data are significantly correlated at this point. In the topological relationship diagram, the edge weight (correlation coefficient) between this node and its adjacent nodes is high, indicating the possibility of an anomaly in the underground structure at this point.

[0063] The entire processing flow of the wavefield synthesis layer strictly adheres to the mathematical principles of signal processing. It extracts frequency features from the data through multi-scale time-frequency decomposition, achieves spatial localization of features using phase alignment and energy superposition, suppresses noise with a blind source separation algorithm, and establishes spatial correlations between multi-source data through cross-validation. Each step is based on deterministic algorithms and physical models, avoiding the introduction of fictitious experimental effects and ensuring the accuracy of the wavefield feature map and the reliability of multi-dimensional topological relationships. This provides high-quality input data for subsequent feature matching and correlation analysis layers.

[0064] Example 3,

[0065] This embodiment details the modeling process of spatial correlation between wavefield feature maps in the correlation analysis layer, covering steps such as anomaly response region identification, feature encoding sequence generation, correlation coefficient calculation, and multi-dimensional topological relationship construction. Spatial correlation analysis of multi-source data is achieved through signal processing algorithms and mathematical modeling, as detailed below:

[0066] The correlation analysis layer first employs a signal feature extraction algorithm to identify anomalous response regions in the wavefield feature map. The signal feature extraction algorithm is based on a combination of threshold detection and edge detection logic: for the energy distribution matrix in the wavefield feature map, a global energy threshold is set. (e.g., twice the average energy value), with an energy value greater than Pixels are marked as potential anomalies; then, the Canny edge detection algorithm is used to extract the contour edges of these potential anomalies, forming continuous anomaly response regions. For example, in the radar wavefield feature map corresponding to pipeline distribution data, the reflection signal of metal pipelines exhibits local energy peaks, which are identified by threshold detection as having energy higher than [a certain value]. The pixel clusters are then used to outline the spatial contour of the pipeline through edge detection, thus determining the geometric range of the abnormal response area. For the wave field feature map of electromagnetic field data, the abnormal response area may be caused by leakage in underground cables, which manifests as an abnormal increase in field strength at a specific frequency (such as 50Hz power frequency). The location of the abnormal field source is located by frequency domain threshold detection and spatial contour extraction.

[0067] After identifying anomalous response regions, feature coding sequences are generated for each type of probe data based on the spectral characteristics of each region. The feature coding sequences are in multi-dimensional vector form, containing parameters such as frequency band identifier, energy level, and spatial coordinate offset. The specific coding rules are as follows:

[0068] Frequency band identification: Dividing the frequency range of the wavefield characteristic spectrum into Each frequency band (e.g., low frequency, mid frequency, and high frequency bands) corresponds to a unique identifier. ( ).

[0069] Energy Level: Normalize the energy values ​​of the abnormal response region to the [1, M] interval (e.g., M=5), corresponding to the energy level. ( (1 is the lowest level, and 5 is the highest level).

[0070] Spatial coordinate offset: Calculate the coordinates of the centroid of the abnormal response region with reference to the origin of the reference coordinate system. And convert it into an offset relative to the center of the current detection area. .

[0071] Taking a certain anomalous response region in the first detection dataset (rock and soil medium data) as an example, if its main energy is concentrated in the medium frequency ( The normalized energy value is 4 ( The centroid coordinate offset is ( If ), then its feature encoding is The feature codes of all abnormal response regions are arranged in spatial coordinate order, forming a feature code sequence for this type of data. ,in This represents the number of abnormal response areas.

[0072] In calculating any two types of detection data (such as the first...) Class and the When determining the correlation between the feature encoding sequences of two classes, the first step is to determine whether the number of abnormal responses is consistent. If Then, based on the one with fewer abnormal responses (let's assume it's the first one)... Class, quantity Bandwidth spread compensation is performed on the distribution characteristics of the highest energy frequency band in the spectrum. The specific steps of bandwidth spread compensation are as follows:

[0073] Identify the first The highest energy level in the class feature encoding sequence (e.g.) The abnormal response area was identified, and its frequency band identifier distribution was statistically analyzed to determine the dominant frequency band. (e.g., the frequency band with the highest frequency).

[0074] In the Insertion in class sequence A virtual anomaly response point, the frequency band identifier of the virtual point is set to... Energy level set to (Second highest level) Spatial coordinate offset is generated based on the coordinate interpolation of adjacent real points to ensure that the distribution of virtual points conforms to the spatial trend of the original sequence.

[0075] After compensation, the length of the feature encoding sequence for both types of data is [length missing]. , recorded as and Next, the correlation coefficients between anomalous response regions with the same frequency band identifier are calculated. For frequency bands... ,extract and All frequency bands are identified as Feature encoding to form subsequences and Each subsequence contains the energy level and spatial coordinate offset for that frequency band. The Pearson correlation coefficient is used to calculate the correlation between subsequences, as shown in the formula:

[0076] ,

[0077] in, For the first Class and the Class data in frequency band The correlation coefficient under; For subsequence and The length of the frequency band (i.e., the number of anomalous responses in that frequency band); , They are respectively , The Middle The energy level of an abnormal response; , They are respectively , The average value of medium energy levels.

[0078] This formula measures the linear correlation of energy levels using standardized covariance, with values ​​ranging from [-1, 1]. A larger absolute value indicates a stronger correlation. For example, if two types of data are in the mid-frequency range... The energy level sequence below shows a significant positive correlation ( This indicates that the anomalous responses of the two in this frequency band have similar energy distribution characteristics, and may correspond to the same underground target (such as the electromagnetic coupling effect between metal pipelines and the surrounding rock and soil media).

[0079] After calculating the correlation coefficients for all frequency bands, the weights of each frequency band are then used. (Preset values, such as those set based on frequency band resolution or data reliability), the comprehensive correlation coefficient of any two types of data is obtained through weighted averaging:

[0080] ,

[0081] in Comprehensive correlation coefficient This reflects the spatial correlation between the two types of data across the entire frequency band.

[0082] Finally, the multi-dimensional topological relationship between any two types of probe data is determined based on the comprehensive correlation coefficient. This multi-dimensional topological relationship is represented by an undirected graph. express:

[0083] Among them, nodes For each abnormal response region of the detected data, each node contains spatial coordinates and frequency band information from the feature encoding; edges Connect nodes that are related, with the weight of each edge being the corresponding comprehensive correlation coefficient. The larger the weight value, the stronger the correlation between the two types of data in that region.

[0084] For example, if the comprehensive correlation coefficient between soil and rock media data and pipeline distribution data in a certain area is 0.7, then in the topological relationship diagram, the corresponding nodes of the two are connected by an edge with a weight of 0.7, indicating that there is a strong correlation between soil and rock media anomalies and pipeline distribution in this area, which may be caused by changes in soil compaction or corrosion around the pipeline, resulting in anomalies in media parameters.

[0085] The correlation analysis layer, through the aforementioned process, extracts spatial correlation features from multi-source data from the wavefield feature map, constructing multi-dimensional topological relationships encompassing frequency bands, energy, and spatial location. This provides a quantitative correlation basis for the multimodal data fusion of subsequent model fusion units. The entire process is based on signal processing algorithms and statistical analysis theory, achieving correlation modeling through rigorous mathematical definitions and logical derivations. This avoids introducing fictitious experimental effects and ensures the scientific validity and interpretability of the topological relationships.

[0086] Example 4,

[0087] This embodiment focuses on the functions of the imaging layer, covering aspects such as multi-view interactive interface, iterative inversion algorithm optimization, spatial interpolation calculation and vector superposition, and error feedback mechanism. It details the construction and optimization process of the 3D panoramic imaging model in conjunction with specific application scenarios, as follows:

[0088] The multi-view interactive interface of the imaging layer provides users with interactive analysis tools for 3D panoramic imaging models. Taking the underground survey scenario of a substation in a power transmission and transformation project as an example, users can input control commands through the graphical user interface (GUI): sliding the mouse wheel adjusts the display resolution of the model, switching from macroscopic geological structure (1m resolution) to pipeline details (0.1m resolution); clicking interface buttons switches the profile direction, such as displaying a horizontal profile along the X-axis to observe geological stratification, or displaying a vertical profile along the Z-axis to view the vertical distribution of pipelines. The signal strength filtering function allows users to set energy thresholds using a slider, for example, displaying only areas with electromagnetic field strength greater than 50μT, quickly locating high field strength anomalies. The interactive interface renders model changes in real time, allowing users to intuitively compare underground structural features under different perspectives, resolutions, and signal strengths. For example, if an anomaly in geological impedance is found in a certain area on the horizontal profile, switching to the vertical profile reveals that the anomaly spatially overlaps with a 2-meter-deep buried metal pipeline, thus indicating a possible correlation between the two.

[0089] In the multimodal data fusion stage, the imaging layer dynamically optimizes the pre-set underground structure template through an iterative inversion algorithm. The underground structure template is constructed based on the geological survey report of the target area, initially containing three strata (silty clay, sand, and bedrock) and known pipeline distribution (such as an east-west power cable buried at a depth of 1.5 meters). After the wavefield synthesis layer generates a new wavefield feature map, the iterative inversion algorithm compares the anomalous response areas in the map with the template model: for example, the wavefield feature map shows a high-frequency, high-energy response at coordinates (X=10m, Y=5m, Z=3m), while the template model labels this location as a sand layer with no pipeline distribution. The algorithm calculates the difference between the model's predicted values ​​and the measured data using the least squares method, adjusts the medium parameters of this area in the template (e.g., adjusting the dielectric constant of the sand layer from 8 to 12 to match the measured high-frequency response), and adds a virtual node for an unknown pipeline. After five iterations, the sand layer distribution range of the template model was reduced, and a new suspected pipeline running north-south was added. Its spatial location completely matched the abnormal area in the wave field feature map, realizing the dynamic adaptation of the model to the measured data.

[0090] During the 3D spatial reconstruction process, the imaging layer performs spatial interpolation calculations on the multi-dimensional topological relationships according to preset imaging weights. The imaging weights are set based on the reliability of the data types. For example, electromagnetic field data is assigned a weight of 0.4 due to its high positioning accuracy (error ±0.2m), pipeline radar data is assigned a weight of 0.3 due to its high resolution (0.1m), and soil and rock media data is assigned a weight of 0.3 due to its wide coverage. Taking a node in the multi-dimensional topological relationship as an example, the node corresponds to coordinates (X=15m, Y=8m, Z=2.5m), and is associated with the equivalent impedance anomaly of the soil and rock media data, the weak reflection signal of the pipeline radar, and the moderate intensity response of the electromagnetic field (weights of 0.3, 0.3, and 0.4, respectively). The spatial interpolation calculation uses the Kriging interpolation method, estimating the 3D spatial attribute values ​​of the point based on the correlation coefficients of the 10 neighboring nodes around the node (e.g., 0.6 with the node to the north and 0.7 with the node to the south), generating a continuous correlation field distribution. Subsequently, the correlated field distribution and the wavefield feature map are vector-superimposed: for example, in the wavefield feature map of a pipeline radar, the reflection intensity at this point is 15 dB. After superimposing the weighted contributions of the electromagnetic field data in the correlated field, the final comprehensive intensity value at this point in the imaging model is... (Assuming the electromagnetic field strength is 10), it is displayed as an orange gradient block, representing a medium level of anomaly.

[0091] The error feedback unit monitors model accuracy in real time during vector overlay. Taking a set of electromagnetic field intensity profiles from historical survey data as an example, this data records field strength values ​​every 1m along the X-axis from 0-20m (Z=2m). The error feedback unit calculates the absolute error between the predicted value of the 3D panoramic imaging model at the corresponding location and the historical data. For example, at X=5m, the predicted value is 45μT, the historical data is 48μT, and the residual value is 3μT. When the residual value of a certain area exceeds a preset threshold (e.g., 5μT) for five consecutive calculations, the data reconstruction unit is triggered to re-execute time-frequency synchronization processing. For example, if a systematic 100ms delay is found in the timestamp of the third detection dataset (electromagnetic field data), causing time-frequency synchronization error, the delay effect is eliminated by recalibrating the device clock and adjusting the phase compensation parameters. Simultaneously, the imaging layer updates the spatial interpolation parameters of the multi-dimensional topological relationships, such as reducing the search radius of Kriging interpolation from 3m to 2m to improve the interpolation accuracy in local areas. After feedback optimization, the residual value between the model's predicted value and the historical data is reduced to 2μT, meeting the accuracy requirements.

[0092] In practical applications of power transmission and transformation engineering, the imaging layer processing workflow can completely reproduce the complex structure of underground space. For example, in the survey of a power transmission line route, the 3D panoramic imaging model shows: the surface layer (0-1.5 meters) is a silty clay layer containing an east-west oriented communication pipeline (buried 1.2 meters deep, 50mm diameter); the layer (1.5-3 meters) is a sand layer with localized water-rich areas (identified by equivalent impedance abrupt changes in soil and rock media data); below 3 meters is bedrock, above which is a north-south oriented power cable (buried 2.8 meters deep, located by high field strength areas in electromagnetic field data). A multi-view interactive interface allows engineers to observe the spatial relationship between the pipeline and the strata from different angles, such as rotating the model to view the intersection of the power cable and the communication pipeline, and analyzing the electromagnetic field coupling effect at the intersection through cross-section analysis. The iterative inversion algorithm automatically optimizes the undulation of the bedrock interface to match seismic wave velocity data; the error feedback mechanism continuously monitors the model accuracy, ensuring that the location error of water-rich areas is less than 0.5 meters.

[0093] The imaging layer achieves high-precision 3D modeling of underground space through multi-view interaction, dynamic model optimization, spatial interpolation fusion, and error closed-loop feedback. Each step is based on actual data processing logic and engineering application scenarios, avoiding fictitious effect descriptions and ensuring the reliability and engineering practicality of the imaging results. This provides intuitive and accurate decision-making basis for underground surveying, route planning, and risk assessment in power transmission and transformation projects.

[0094] Example 5,

[0095] This embodiment involves the specific implementation of the dynamic compensation unit and real-time monitoring module of the feature matching layer. Combined with the underground survey scenario of power transmission and transformation engineering, it details the amplitude compensation based on the medium density gradient and the process of marking and risk warning of electromagnetic field anomalies.

[0096] The dynamic compensation unit of the feature matching layer performs amplitude compensation on the wavefield feature spectrum based on the medium density gradient of the target area. Taking the underground survey of a substation as an example, the surface of the target area is a backfill soil layer (density 1.8×10³kg / m³), below which are clay layers (density 2.0×10³kg / m³) and sandstone layers (density 2.3×10³kg / m³), with the medium density increasing stepwise with depth. The dynamic compensation unit first constructs a medium absorption coefficient compensation model based on the electromagnetic field strength data of the third detection dataset. By analyzing the electromagnetic field strength attenuation at different depths, it is found that the signal attenuation rate is 0.5dB / m in the backfill soil layer, 1.2dB / m in the clay layer, and 1.8dB / m in the sandstone layer. Based on this, a piecewise linear absorption coefficient model is established: absorption coefficient α=0.5 for depths of 0-2 meters (backfill soil), α=1.2 for depths of 2-5 meters (clay), and α=1.8 for depths below 5 meters (sandstone).

[0097] The dynamic compensation unit updates the energy distribution of each frequency band in the wavefield feature map using a backpropagation algorithm. Taking a power cable buried 3 meters deep as an example, its energy response in the wavefield feature map of electromagnetic field data is in the 50Hz frequency band. Since this location is in a clay layer (α=1.2), the energy attenuates by approximately 3.6 dB (1.2×3) after the signal propagates 3 meters, leading to an underestimation of the energy value in the map. The backpropagation algorithm, based on the absorption coefficient model, performs reverse compensation on the energy value of this frequency band according to the attenuation, i.e., increasing it by 3.6 dB. This restores the cable's energy response in the map to a level close to the true intensity of the transmitter at the ground surface, thus more clearly showing the cable's spatial trajectory. Similarly, for pipelines traversing different media layers, the dynamic compensation unit calculates and compensates for the attenuation of each media layer segment by segment, ensuring consistent amplitude of the target body at different depths in the wavefield feature map, facilitating feature matching and identification.

[0098] The real-time monitoring module marks areas in the 3D panoramic imaging model where the electromagnetic field strength exceeds the safety threshold. The safety threshold is set according to the "Electromagnetic Environment Control Limits" (GB8702-2014), for example, the public exposure limit for power frequency electromagnetic fields is 4000V / m for electric field strength and 100μT for magnetic induction intensity. In the underground cable survey of a transmission line, the 3D model showed a magnetic induction intensity of 120μT near the cable terminal joint (coordinates X=20m, Y=15m, Z=1.8m), exceeding the safety threshold of 100μT. The real-time monitoring module automatically marks this area in red and generates a detailed information box on the right side of the interface, displaying the field strength value, the exceedance range (+20%), and the surrounding medium layer (backfill soil), alerting engineers that there is an electromagnetic radiation risk in this area and that shielding measures or adjustments to the construction plan are necessary.

[0099] The real-time monitoring module's early warning trigger mechanism is activated when abnormal field strength is detected in the spatial overlap area between pipeline distribution data and electromagnetic field data. For example, a natural gas pipeline buried at a depth of 1.5 meters (second detection dataset) and a running 10kV power cable (third detection dataset) are only 0.3 meters apart at coordinates (X=10m, Y=8m, Z=1.6m), which is lower than the safe distance (0.5 meters) stipulated in the "Regulations on the Protection of Power Facilities". The real-time monitoring module detects a sudden increase in electromagnetic field strength in this overlapping area to 150μT (normal operating condition is 80μT), and determines that there may be leakage caused by cable insulation damage. The system immediately generates a risk warning signal with coordinate location, including the following information:

[0100] Warning type: Abnormal electromagnetic coupling in pipeline

[0101] Coordinates: X=10m, Y=8m, Z=1.5-1.6m

[0102] Anomaly Description: A sudden increase in electric field strength in the vicinity of power cables and natural gas pipelines.

[0103] Recommendation: Immediately cease excavation work in the area and conduct cable insulation testing.

[0104] Warning signals are pushed through multiple channels, including system pop-ups, SMS messages (sent to the project manager's mobile phone), and on-site alarms (audible and visual alarms), ensuring timely communication of risk information. After receiving the alarm, engineers can access historical wave field characteristic maps of the area through a multi-view interactive interface, compare and analyze the time series changes in electromagnetic field intensity, confirm that the anomaly is a sudden increase rather than equipment noise, and thus initiate the emergency investigation process.

[0105] In another application scenario, a power transmission and transformation project needed to assess the electromagnetic environment impact of aging underground pipelines. The real-time monitoring module marked the electromagnetic field intensity distribution of multiple abandoned cables in a 3D model. The field strength value of an abandoned cable buried 4 meters deep (already de-energized) was close to the background value, while another abandoned cable buried 3 meters deep (possibly not completely de-energized) had a field strength of 60 μT. Although this did not exceed the safety threshold, it was significantly higher than the surrounding area (background value 20 μT). The monitoring module marked the latter as a yellow warning area, indicating to engineers that the cable might still be energized and further testing was needed. Through the dynamically compensated wavefield characteristic spectrum, the relationship between the cable's trajectory and the change in medium density could be clearly observed: when crossing a sandstone layer (high-density medium), due to the amplitude enhancement by the dynamic compensation unit, the cable's energy response continuity in the spectrum was good; while when crossing a backfill layer, due to the smaller compensation amount, the energy response was slightly weaker but still higher than the background value, aiding in the assessment of cable integrity.

[0106] The feature matching layer and real-time monitoring module solve the problem of feature ambiguity caused by signal attenuation in complex underground media environments, as well as the challenge of identifying electromagnetic safety risks in power transmission and transformation projects, through medium density compensation and real-time electromagnetic field monitoring. The dynamic compensation unit achieves physical consistency correction of wave field characteristics based on actual medium parameters and electromagnetic field propagation laws; the real-time monitoring module provides quantitative risk assessment and early warning according to national standards and engineering specifications. The entire process is based on measured data and engineering logic, avoiding fictitious effect descriptions and ensuring the safety and practicality of the system in underground surveying of power transmission and transformation projects.

[0107] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0108] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A panoramic imaging system for automated underground surveying of power transmission and transformation projects, characterized in that, include: The data acquisition module is used to collect underground multi-source detection data of the target area. The multi-source detection data includes a first detection dataset corresponding to the soil and rock media data, a second detection dataset corresponding to the pipeline distribution data, and a third detection dataset corresponding to the electromagnetic field data. The soil and rock media data includes static stratigraphic data retrieved from the geological database and dynamic media parameters obtained through the distributed detection device. A panoramic imaging module is used to input the multi-source detection data into a three-dimensional imaging engine for analysis; A three-dimensional panoramic imaging model of the underground space is constructed based on the output of the three-dimensional imaging engine. The three-dimensional imaging engine includes a data reconstruction unit and a model fusion unit. The data reconstruction unit is used to perform spatial consistency alignment on the multi-source detection data. The model fusion unit performs multi-modal data fusion based on a preset underground structure template and historical survey data. The model fusion unit includes a wavefield synthesis layer, a feature matching layer, a correlation analysis layer, and an imaging layer connected in sequence. The wavefield synthesis layer is used to perform frequency domain transformation on each dataset in the multi-source detection data to generate a wavefield feature map. The correlation analysis layer is used to model the spatial correlation between the wavefield feature maps corresponding to the multi-source detection data to generate multi-dimensional topological relationships. The imaging layer is used to perform three-dimensional spatial reconstruction based on the multi-dimensional topological relationships and the wavefield feature maps to generate a three-dimensional panoramic imaging model.

2. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 1, characterized in that, The process of modeling the spatial correlation between wavefield feature maps corresponding to multi-source detection data and generating multi-dimensional topological relationships includes: A signal feature extraction algorithm is used to identify abnormal response regions in the wavefield feature map, and feature coding sequences corresponding to each type of detection data are generated based on the spectral characteristics of each abnormal response region. Calculate the correlation coefficient between anomalous response regions with the same frequency band identification in the feature coding sequences corresponding to any two types of detection data, and determine the multi-dimensional topological relationship between the two types of detection data based on the correlation coefficient; the calculation of the correlation coefficient between anomalous response regions with the same frequency band identification in the feature coding sequences corresponding to any two types of detection data includes: When the number of abnormal responses in the feature coding sequences corresponding to any two types of detection data is inconsistent, frequency band expansion compensation is performed based on the distribution characteristics of the highest energy frequency band among those with fewer abnormal responses, and the correlation coefficient between abnormal response regions with the same frequency band identification is calculated based on the compensated sequence.

3. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 1, characterized in that, The data reconstruction unit is specifically used for: The first detection dataset, the second detection dataset, and the third detection dataset are time-frequency synchronized according to a preset reference coordinate system to obtain the aligned first detection dataset, the aligned second detection dataset, and the aligned third detection dataset. An adaptive filtering method is used to suppress interference signals in the aligned first and second probe datasets, and a phase compensation method is used to correct waveform distortion in the aligned third probe dataset, resulting in a first optimized probe dataset, a second optimized probe dataset, and a third optimized probe dataset.

4. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 3, characterized in that, The first optimized detection dataset includes processed static formation data and processed dynamic medium parameters. The data reconstruction unit is further used for: Analyze the energy decay curves of the processed static stratigraphic data and the processed dynamic medium parameters within historical exploration periods; Based on the energy attenuation curve and the wave velocity parameters of the processed static formation data in the specified area, the equivalent impedance parameters of the processed dynamic medium parameters in the specified area are calculated. A comprehensive medium dataset is generated based on the processed dynamic medium parameters and their equivalent impedance parameters, and the detection data corresponding to the comprehensive medium dataset is used as the first optimized detection dataset.

5. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 1, characterized in that, The wave field synthesis layer includes a spectrum analysis unit and an energy mapping unit; The spectrum analysis unit is used to perform multi-scale time-frequency decomposition on each type of detection dataset contained in the multi-source detection data to extract the corresponding frequency domain feature components from each type of detection dataset. The energy mapping unit is used to phase-align and superimpose the frequency domain feature components extracted from each type of detection dataset with the original detection dataset to generate a wavefield feature map.

6. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 5, characterized in that, The wavefield synthesis layer further includes a noise suppression unit, which is used to perform blind source separation and noise reduction processing on the wavefield feature map. The process of modeling the spatial correlation between wavefield feature maps corresponding to multi-source detection data includes: performing cross-validation analysis on the denoised wavefield feature maps corresponding to various types of detection data to generate the multi-dimensional topological relationship.

7. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 1, characterized in that, The imaging layer includes a multi-view interactive interface, which allows users to switch profiles, filter signal intensity, and adjust resolution of the three-dimensional panoramic imaging model through control commands. The multimodal data fusion based on the preset underground structure template and historical survey data includes: dynamically optimizing the physical property parameters of the underground structure template through an iterative inversion algorithm to adapt to the update of the multidimensional topological relationship; The three-dimensional spatial reconstruction based on the multi-dimensional topological relationship and the wave field feature map includes: performing spatial interpolation calculation on the multi-dimensional topological relationship according to the preset imaging weight, and vector superimposing it with the wave field feature map.

8. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 1, characterized in that, The feature matching layer includes a dynamic compensation unit, which is used to perform amplitude compensation on the wavefield feature map according to the medium density gradient of the target region. The operations performed by the dynamic compensation unit include: establishing a compensation model for the medium absorption coefficient based on the electromagnetic field intensity data of the third detection dataset, and updating the energy distribution of each frequency band in the wave field feature spectrum through a backpropagation algorithm.

9. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 7, characterized in that, The imaging layer also includes an error feedback unit, which is used to calculate the residual value between the three-dimensional panoramic imaging model and the historical survey data in real time during the vector superposition process; when the residual value exceeds a preset threshold, the data reconstruction unit is triggered to re-execute the time-frequency synchronization processing and update the spatial interpolation parameters of the multi-dimensional topological relationship.

10. The automated underground panoramic surveying and imaging system for power transmission and transformation engineering as described in claim 1, characterized in that, The system also includes a real-time monitoring module, which is used to mark areas in the three-dimensional panoramic imaging model where the electromagnetic field strength exceeds the safety threshold. The real-time monitoring module is equipped with an early warning triggering mechanism, which generates a risk warning signal with coordinate positioning when an abnormal field strength is detected in the spatial overlap area between pipeline distribution data and electromagnetic field data.

Images