An active source gravity exploration method and system based on topological Euler characteristic regulation

Abstract

The application discloses a kind of active source gravity exploration method and system based on topological Chen number regulation.The method is by deploying active gravity field modulation device containing topological material array, the state of Chen number of topological material is changed using field regulation module, and active modulation gravity field signal is generated in exploration target area;Through gravity sensor array, the gravity response signal modulated after underground medium is collected;With modulation signal as reference, the gravity response component caused by underground medium is extracted by coherent detection, matched filtering or projection separation, and background interference is eliminated;Based on the extracted gravity response signal, the underground density distribution is reconstructed by forward modeling and inversion calculation.The system includes active gravity field modulation device, control and signal generation module, gravity sensor array, signal processing and interference elimination module and inversion imaging module.The application upgrades gravity exploration from passive measurement to active detection, signal-to-noise ratio is improved by more than 10 times, longitudinal resolution is improved by 3-5 times, detection depth is increased by more than 50%, and can be applied to mineral resources exploration, deep geological structure detection, tunnel advanced geological prediction and planetary underground structure detection and other scenes.

Description

Technical Field

[0001] This invention relates to the fields of geophysical exploration and gravity detection technology, specifically to an active source gravity exploration method and system based on topological Chern number control. This invention can be applied to mineral resource exploration (metal mines, oil and gas fields, key metal ore layers in coal-bearing systems), deep geological structure detection (fault zones, basin basement, orogenic belt structure), hydrogeological survey (groundwater storage monitoring, aquifer structure detection), engineering geological exploration (tunnel advanced geological prediction, underground cavity detection, goaf exploration), geothermal resource exploration, and lunar and planetary underground structure detection. Background Technology

[0002] Gravity exploration is one of the important methods of geophysical exploration. It infers the underground density distribution by measuring changes in the gravity field at the surface or underground, thereby identifying mineral resources and geological structures. Different geological bodies have different density characteristics: metallic minerals (iron, copper, tungsten, gold, etc.) have a higher density than the surrounding rock, oil and gas reservoirs have a lower density than the surrounding rock, and fault zones and cavities show abrupt changes in density. Existing technical solutions and their shortcomings (a) Passive gravity exploration method Traditional gravity exploration uses a passive measurement method, which involves deploying gravity sensors only on the surface or in boreholes to passively receive changes in the gravity field caused by underground density anomalies. This method has the following problems: (1) The signal strength is limited by the density difference between the underground density anomaly and the surrounding rock, and the signal is weak for small-scale or deep targets; (2) The volume superposition effect of the gravity field leads to insufficient vertical resolution, making it difficult to distinguish density anomalies at different depths; (3) It is greatly affected by interference factors such as surface topography and uneven shallow density, making data correction difficult; (4) The inversion results are highly ambiguous, and different density distributions may produce the same gravity anomaly. (II) Joint Inversion and Data Processing Methods Existing technologies improve interpretation accuracy through methods such as joint gravity and seismic inversion and three-dimensional gravity gradient inversion. These methods have the following problems: (1) they rely on the synchronous acquisition of multiple data sources, which is costly and time-consuming; (2) the inversion algorithm has a large computational load and slow iteration convergence; (3) they fail to fundamentally solve the signal-to-noise ratio problem of passive measurement, and only optimize it in the post-processing stage. (III) Borehole / Downhole Gravity Detection Methods In recent years, active gravity detection instruments have emerged, which place gravity sensors inside the borehole for measurement. This type of method has the following problems: (1) "Active" only refers to the activeness of the instrument deployment method, not the active control of the gravity field; (2) The number and arrangement of sensors are limited by the borehole space; (3) It is still a passive receiving mode and cannot actively improve the signal strength. The common limitation of existing technologies is that: All are passive measurement modes and cannot actively modulate the gravity field signal. Signal strength is limited by differences in underground density, making it difficult to detect deep and small targets. The lack of effective interference cancellation methods makes it difficult to separate shallow interference. Inversion involves large computational costs and suffers from severe ambiguity. The inability to detect active sources in the gravity field is similar to the difference between "active seismic source" and "passive monitoring" in seismic exploration. To address the aforementioned problems, this invention provides an active source gravity exploration method and system based on topological Chern number control. Summary of the Invention

[0003] (a) Purpose of the invention This invention aims to solve the technical problems in existing gravity exploration technology, such as weak passive measurement signals, low resolution, difficulty in eliminating interference, and severe inversion multiple solutions, and provides an active source gravity exploration method and system based on topological Chern number control. (II) Technical Solution This invention provides an active source gravity exploration method based on topological Chern number control, characterized by comprising the following steps: Step 1: Configure the active gravity field modulation device An active gravity field modulation device is deployed in the exploration target area. The device includes a topological material array and a field control module. The topological material array is composed of topological materials with non-zero Chern numbers, and the field control module actively modulates the local gravity field by changing the boundary conditions of the topological materials. The Chern number C of a topological material is determined by the following relationship: Where F is the Berry curvature of the topological material, and S is the closed surface of the Brillouin zone. Step 2: Generate an actively modulated gravity field signal By applying control signals to the topological material array through a field control module, the Chern number state of the topological material is changed, thereby generating an actively modulated gravity field signal in the exploration target area. The modulation method includes at least one of the following: Frequency modulation: Periodically changing the gravitational field strength at a preset frequency f (0.01Hz–100Hz). Amplitude modulation: The gravitational field strength is changed by a preset amplitude Δg (0.1μGal–100μGal). Phase modulation: altering the spatial distribution of the gravitational field using a preset phase sequence. Encoding modulation: Modulating the gravitational field with a preset encoding sequence (such as pseudo-random code, Chirp code). The actively modulated gravitational field signal Δg(t) is expressed as: Where Δg(t) is the actively modulated change in the gravitational field, A k f is the amplitude of the k-th modulation component. k For the modulation frequency, φ k This is the initial phase. Step 3: Acquire the modulated gravity response signal Gravity response signals modulated by the subsurface medium were acquired using a gravity sensor array (at least one of a ground gravimeter, a well gravimeter, an airborne gravity gradiometer, or a seabed gravimeter). The acquisition method involved simultaneously acquiring transmitted and received signals, recording the gravity field intensity G_i(t) of each receiving channel, along with the corresponding spatial coordinates and timestamp. Step 4: Interference cancellation and signal extraction based on coherent detection Using the modulated signal as a reference, the gravity response caused by the underground medium is extracted from the acquired signal. Background interference is eliminated through the following methods: Coherent detection: Utilizing prior information about the modulated signal, the gravity response component that is in phase and frequency with the modulated signal is extracted through coherent demodulation. Matched filtering: Using the modulated signal as a template, matched filtering is applied to the received signal to suppress uncorrelated interference. Projection separation: Projecting the received signal onto a subspace spanned by the modulated signal to separate the gravitational response caused by active modulation. The output signal R(t) of coherent detection is expressed as: Where G(t) is the received signal, S_ref(t) is the reference modulation signal, and T is the integration period. Step 5: Subsurface density structure inversion imaging Based on the extracted gravity response signal, the underground density structure is reconstructed through the following steps: (1) Establish a forward model of the density-gravity response of the exploration target area; (2) The gravity response signal extracted in step 4 is used as input, and physical information neural network (PINN) is used for fast inversion; (3) Output the three-dimensional density distribution image of the underground to delineate the spatial location and geometric shape of density anomalies (mineral resources, geological structures, cavities, etc.). The inversion objective function is: Where d_obs is the observed gravity response vector, F(ρ) is the forward modeling operator, ρ is the density distribution to be reconstructed, λ and μ are regularization parameters, and ρ_prior is the prior density distribution. (III) Beneficial Effects Active source detection improves signal-to-noise ratio: By actively modulating the gravity field, the weak underground gravity response is separated from the strong background, improving the signal-to-noise ratio by more than 10 times compared to passive measurement. Interference cancellation capability: Utilizing coherent detection and matched filtering, it effectively suppresses interference factors such as surface topography and shallow inhomogeneities. Significantly improved resolution: Active modulation and coherent detection improve vertical resolution by 3–5 times compared to traditional gravity exploration. Fast inversion imaging: Combining PINN for rapid inversion reduces inversion time from hours to seconds compared to traditional methods. Increased detection depth: The active modulation signal can be designed with low-frequency components, increasing the penetration depth by more than 50% compared to passive measurement. Multi-source data fusion: can be combined with seismic, electromagnetic, and other methods to form a comprehensive multi-parameter interpretation. Attached Figure Description

[0004] Figure 1 This is a schematic diagram of the overall structure of an active source gravity exploration system based on topological Chern number control, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the active gravity field modulation signal waveform provided in an embodiment of the present invention. Figure 3 This is a schematic diagram illustrating the principle of coherent detection and interference cancellation provided in an embodiment of the present invention. Figure 4 The control flowchart of the active source gravity exploration method provided in the embodiment of the present invention. Detailed Implementation

[0005] Example 1: Active Source Surface Gravity Exploration System Based on Topological Material Array Scenario Description: Deploy an active gravity field modulation device and a gravity sensor array in the target exploration area (metal mining area) to conduct active source gravity exploration. Step 1: Configure an active gravity field modulation device An active gravity field modulation device is deployed at the center of the exploration area. The device consists of a topological material array (composed of topological insulators or topological semimetals, Chern number C ≠ 0), with an array size of 10m × 10m and composed of 100 topological material units (1m × 1m / unit). Each unit is connected to the control module through an independent field modulation electrode. The field control module actively modulates the local gravitational field by changing the electric or magnetic field parameters applied to the topological material unit. The modulation parameters include Chern number control accuracy ΔC ≤ 0.01 and response time ≤ 1ms. Step 2: Generate an actively modulated gravity field signal The control module applies an coded modulation signal to the topological material array. A pseudo-random code sequence (length 1023, chip rate 100Hz) is used to modulate the local gravity field, with a modulation depth Δg = 10μGal. The active modulation signal is continuously transmitted for T = 60 seconds, covering the entire exploration area (500m in diameter). The center frequency of the modulation signal is designed to be 0.1Hz–10Hz, balancing penetration depth and signal-to-noise ratio. Step 3: Acquire the modulated gravity response signal A gravity sensor array consisting of 36 high-precision gravimeters (resolution ≤1μGal) was deployed in the exploration area, arranged in a 6×6 grid with a grid spacing of 100m. Each gravimeter synchronously acquired gravity field data at a sampling rate of 1000Hz and a time synchronization accuracy of ≤1μs. Simultaneously record the three-dimensional spatial coordinates (accuracy ≤ 1 cm) and acquisition timestamps of each gravimeter. Step 4: Coherent Detection and Interference Cancellation Using the transmitted pseudo-random code modulated signal as a reference template, coherent detection and matched filtering are performed on the gravity data from each receiving channel. Through projection separation, the gravity response component caused by active modulation is extracted, suppressing unrelated interferences such as surface vibrations and atmospheric pressure changes. After coherent detection, the signal-to-noise ratio of each receiving channel is improved by approximately 15 dB compared to the original data. Step 5: Subsurface density structure inversion imaging The extracted gravity response signal is input into the PINN inversion engine (8 network layers, 256 neurons per layer). The network input consists of the spatial coordinates of each receiving channel and the extracted gravity response value, and the output is the three-dimensional density distribution of the subsurface (imaging range 500m×500m×200m, grid resolution 5m×5m×5m). The inversion time is ≤10 seconds, and a three-dimensional density distribution image is output. Density anomalies (metallic ore bodies) appear as high-density areas in the image (density ≥3.5g / cm³, background surrounding rock density approximately 2.7g / cm³), and the spatial location, geometry, and scale of the anomalies are readily apparent. Performance test results: Example 2: Deep Mineral Resource Exploration Based on Gravity Shielding Scenario Description: Conduct deep (>500m) mineral resource exploration in a known metallogenic belt, and combine gravity shielding devices to eliminate shallow interference. Step 1: Configure the gravity shielding device and active modulation system Gravity shielding devices (gravity shielding devices based on topological Chern number control) are deployed in the exploration area to shield against gravity interference caused by shallow density inhomogeneities from the surface to a depth of 200m. Simultaneously, an active gravity field modulation device (as in Example 1) is deployed. Step 2: Active modulation and shielding work together The active modulation device transmits a coded modulated gravity signal (frequency 0.05Hz–5Hz, pseudo-random code sequence). The gravity shielding device works synchronously to shield shallow interference at the receiving end, so that the sensor mainly receives the modulated response signal from deep (>200m). Step 3: Deep Signal Acquisition and Extraction Data were collected using a combination of deep-well gravimeters (placed inside the borehole at depths of 100m–500m) and surface gravimeters. By comparing the signal differences before and after shielding, the shallow and deep gravity responses were separated. Step 4: Deep density structure inversion Using the extracted deep gravity response as input, the density distribution at depths of 500m–2000m is inverted. Geological targets such as deep concealed ore bodies and fault structures are identified. The beneficial effects of this embodiment are that the gravity shielding device eliminates shallow interference, enabling the extraction of weak signals from deep depths, and increasing the detection depth by more than double compared to traditional methods. Example 3: Fast Gravity Data Inversion Based on Physical Information Neural Network Scenario Description: Rapid gravity data inversion is performed in a large-scale exploration area (10km×10km) to generate real-time images of underground density distribution. Step 1: Large-scale active source data acquisition Active gravity field modulation devices (multiple units operating in tandem) and mobile gravity sensor arrays (vehicle-mounted or UAV-mounted) are deployed in a 10km×10km exploration area. The active modulation signal adopts frequency division multiple access (FDMA) mode, with each modulation device operating at a different frequency to achieve regional synchronous modulation. Step 2: Real-time transmission and preprocessing of massive amounts of data The collected data is transmitted to the data center in real time via a wireless network. The data volume is approximately 100GB per hour. Preprocessing includes noise reduction, time synchronization, and coordinate correction. Step 3: Fast Inversion Based on PINN A Physical Information Neural Network (PINN) is used for fast inversion. PINN encodes the Eikonal equation and the forward gravity field equation as part of the loss function, satisfying both physical laws and data fitting during training. The network input consists of the spatial coordinates of each sensor and the extracted gravity response values, and the output is the three-dimensional density distribution of the subsurface. PINN inversion is performed on a GPU cluster, which improves processing speed by more than 10 times compared to traditional iterative inversion methods. Step 4: Real-time imaging and dynamic updates The inversion results are displayed in real time as a three-dimensional density distribution image, supporting dynamic updates and on-site decision-making during the exploration process. Example 4: Advanced geological prediction of tunnels based on active source gravity exploration Scenario Description: During tunnel excavation, active source gravity exploration technology is used to predict geological anomalies ahead (faults, fracture zones, karst caves, water-rich areas, etc.). Step 1: Deployment of active gravity field modulation device inside the tunnel A miniaturized active gravity field modulation device is installed at an appropriate location behind the tunnel face (10m–30m away from the face). The device is ≤0.5m×0.5m×0.5m in size and can be moved with the tunnel boring machine. Step 2: Active Modulation and Forward Detection The modulation device transmits coded modulated gravity signals (frequency 1Hz–50Hz, detection distance 50m–100m) forward (towards the tunnel axis). A gravity sensor array (deployed near the tunnel face) receives the gravity response signals modulated by the geological body in front. Step 3: Rapid identification of geological anomalies ahead The PINN fast inversion engine (deployed on edge devices, inference latency ≤1 second) is used to reconstruct the density distribution within a range of 50m–100m in front of the tunnel face in real time. Density anomaly areas (faults, fracture zones, karst caves, water-rich areas) appear as low-density or abrupt density changes in the density images. Step 4: Geological Anomaly Early Warning and Tunneling Recommendations The system automatically generates geological anomaly warnings based on the inversion results: - Faults / fracture zones: abrupt density changes, accompanied by low-density zones - Karst caves: localized low-density anomalies - Water-rich areas: decreased density - Hard rock interlayers: increased density Early warning information is transmitted to the tunneling machine control panel in real time to guide the adjustment of tunneling parameters and the deployment of safety measures. The beneficial effects of this embodiment are: it enables real-time, non-destructive detection of geological conditions ahead during tunnel excavation, which is more efficient, less costly, and does not interrupt construction compared to traditional borehole advance prediction methods. Terminology Explanation

[0006] Terminology Meaning The topological Chern number is the global topological invariant of the band structure of the topological material, C = (1 / 2π)∮F·dS, taking integer values. Active source gravity exploration is a method of underground exploration that actively modulates the gravity field rather than passively receiving gravity signals. The coding modulation method modulates the gravity field signal using a preset coding sequence (including a pseudo-random code sequence or a Chirp code sequence), with a coding sequence length ≥ 128 and a chip rate ≥ 10 Hz. Coherent detection is a signal processing method that uses a known modulating signal as a reference to extract the in-phase and co-frequency components from the received signal, outputting R(t) = (1 / T)∫G(t)·S_ref(t)dt. Matched filtering is a signal processing method that uses a known signal as a template to filter the received signal, maximizing the signal-to-noise ratio. Projection separation is a signal processing method that projects the received signal into a signal subspace and a noise subspace to separate the useful signal from the interference. Physical Information Neural Network (PINN) is a neural network architecture that encodes physical equations (including the Eikonal equation and the forward modeling equation of the gravitational field) as a loss function. It is used to quickly solve forward and inverse modeling problems, and simultaneously satisfies physical laws and data fitting during training. Gravity shielding is a technique that actively shields the gravitational field in a specific region by controlling the topological Chern number.

Claims

1. Claim 1 (Independent claim, method) An active source gravity exploration method based on topological Chern number control, characterized by comprising the following steps: (1) An active gravity field modulation device is deployed in the exploration target area. The device includes a topological material array and a field control module. The topological material has a non-zero Chern number. The field control module achieves active modulation of the local gravity field by changing the boundary conditions of the topological material. (2) By applying a control signal to the topological material array through the field control module, the Chern number state of the topological material is changed, and an actively modulated gravity field signal is generated in the exploration target area. The modulation method includes at least one of frequency modulation, amplitude modulation, phase modulation or coded modulation. (3) Collect gravity response signals modulated by underground medium through gravity sensor array, and synchronously record the transmitted modulation signal and the received gravity response signal; (4) Using the modulated signal as a reference, extract the gravity response component caused by the underground medium from the acquired signal by at least one of coherent detection, matched filtering or projection separation, and eliminate background interference; (5) Based on the extracted gravity response signal, the underground density distribution is reconstructed through forward modeling and inversion calculation, and the spatial location and geometric shape of the density anomaly are identified.

2. Claim 2 (dependent claim) The method according to claim 1, characterized in that: The Chern number C of the topological material in step (1) is determined by the following relationship: Where F is the Berry curvature and S is the closed surface of the Brillouin zone; the field control module controls the Chern number state of the topological material through electric or magnetic fields, with a Chern number control accuracy ΔC ≤ 0.01 and a response time ≤ 1ms.

3. Claim 3 (dependent claim) The method according to claim 1, characterized in that: The actively modulated gravitational field signal Δg(t) mentioned in step (2) is expressed as: Where Δg(t) is the actively modulated change in the gravitational field, A k f is the amplitude of the k-th modulation component. k For the modulation frequency, φ k The initial phase is defined as follows: the modulation frequency range is 0.01Hz–100Hz, and the modulation amplitude range is 0.1μGal–100μGal.

4. Claim 4 (dependent claim) The method according to claim 1, characterized in that: The gravity sensor array mentioned in step (3) includes at least one of a ground gravimeter, a well gravimeter, an airborne gravity gradiometer, or a seabed gravimeter; the sampling rate of each sensor is ≥100Hz, the time synchronization accuracy is ≤1μs, and the spatial positioning accuracy is ≤1cm.

5. Claim 5 (dependent claim) The method according to claim 1, characterized in that: The output signal R(t) of the coherent detection in step (4) is expressed as: Where G(t) is the received signal, S_ref(t) is the reference modulation signal, and T is the integration period; the signal-to-noise ratio after coherent detection is improved by ≥10dB compared with passive measurement.

6. Claim 6 (dependent claim) The method according to claim 1, characterized in that: The inversion calculation in step (5) uses a physical information neural network for inversion, and the inversion objective function is: Where d_obs is the observed gravity response vector, F(ρ) is the forward modeling operator, ρ is the density distribution to be reconstructed, λ and μ are regularization parameters, and ρ_prior is the prior density distribution.

7. Claim 7 (dependent claim) The method according to claim 1, characterized in that: The method is applicable to at least one of the following scenarios: mineral resource exploration, deep geological structure detection, hydrogeological survey, tunnel advanced geological prediction, underground cavity detection, geothermal resource exploration, or planetary underground structure detection.

8. Claim 8 (Independent claim, system) An active source gravity exploration system based on topological Chern number control, characterized in that, include: An active gravity field modulation device includes a topological material array and a field control module. The topological material has a non-zero Chern number, and the field control module achieves active modulation of the local gravity field by changing the boundary conditions of the topological material. A control and signal generation module is connected to the active gravity field modulation device to generate control signals and apply them to the topological material array to generate an actively modulated gravity field signal in the exploration target area. A gravity sensor array, comprising multiple gravity sensor units, is used to acquire gravity response signals modulated by the underground medium and to simultaneously record the transmitted modulation signals and the received gravity response signals. A signal processing and interference cancellation module is connected to the gravity sensor array and is used to extract the gravity response component caused by the underground medium by at least one of coherent detection, matched filtering or projection separation, with the modulated signal as a reference. An inversion imaging module, connected to the signal processing and interference cancellation module, is used to reconstruct the underground density distribution based on the extracted gravity response signal and output the spatial location and geometry of the density anomaly.

9. Claim 9 (dependent claim) The system according to claim 8 is characterized in that: The system also includes a gravity shielding device connected to the gravity sensor array, used to shield the gravity interference caused by shallow density inhomogeneities within a preset depth range from the ground surface at the receiving end.

Images