Blind ore body positioning method and system based on multi-component magnetic measurement data

Abstract

This application relates to the field of geophysical exploration, specifically to a method and system for locating blind ore bodies based on multi-component magnetic survey data. The method includes: acquiring multi-component magnetic survey data, gravity anomaly data, transient electromagnetic anomaly data, and stratigraphic attitude data; performing constrained inversion on the multi-component magnetic survey data to obtain the three-dimensional distribution of the magnetic source body; extracting the dip angle and boundary gradient direction to generate magnetic source body orientation constraints; performing orientation correction and position registration on the gravity anomaly data and transient electromagnetic anomaly data based on the magnetic source body orientation constraints, and constructing the target probability distribution of the blind ore body by combining the three-dimensional distribution of the magnetic source body; determining the three-dimensional location result of the blind ore body based on the target probability distribution and stratigraphic attitude data, and outputting the coordinates of the drilling target area and the drilling depth range. This application can improve the stability of blind ore body location and the targeting of drilling deployment.

Description

Technical Field

[0001] This application relates to the field of geophysical exploration, specifically to a method and system for locating blind ore bodies based on multi-component magnetic survey data. Background Technology

[0002] Blind ore bodies are characterized by deep burial, weak outcrops, and complex spatial distribution, directly impacting the accuracy of deep mineral exploration, drilling deployment efficiency, and the continuity of mineral resources. With the deepening application of quantum geomagnetism, multi-component magnetic surveying, gravity exploration, and transient electromagnetic detection, the industry has the capability to acquire multi-source anomaly information. However, existing interpretation methods often rely on single anomaly interpretation or simple superposition analysis, making them susceptible to background field interference, anomaly misalignment, and tectonic influences. This leads to inaccurate identification of the ore body's extension direction and unstable three-dimensional positioning, affecting the reliability and economy of deep mineral exploration. Therefore, establishing a blind ore body positioning method that can utilize multi-component magnetic survey direction information and achieve collaborative constraints from multiple anomalies has significant engineering value. Summary of the Invention

[0003] This application provides a method and system for locating blind ore bodies based on multi-component magnetic survey data, which is used to at least solve the problem of how to achieve stable three-dimensional positioning of blind ore bodies when there are misalignments and inconsistent directions in multi-source geophysical anomalies.

[0004] In a first aspect, this application provides a method for locating blind ore bodies based on multi-component magnetic survey data, the method comprising:

[0005] Acquire multi-component magnetic measurement data, gravity anomaly data, transient electromagnetic anomaly data, and stratigraphic attitude data. Perform constrained inversion on the multi-component magnetic measurement data based on the stratigraphic attitude data to obtain the three-dimensional distribution of the magnetic source body.

[0006] Based on the three-dimensional distribution of the magnetic source body, the tilt angle and boundary gradient direction are extracted to generate magnetic source body orientation constraints;

[0007] Based on the orientation constraint of the magnetic source body, the orientation correction and position registration of gravity anomaly data and transient electromagnetic anomaly data are performed, and the target probability distribution of the blind ore body is constructed by combining the three-dimensional distribution of the magnetic source body.

[0008] The three-dimensional positioning results of the blind ore body are determined based on the target probability distribution and stratigraphic occurrence data, and the coordinates of the drilling target area and the drilling depth range are output based on the three-dimensional positioning results of the blind ore body.

[0009] In one possible implementation, the constrained inversion includes: performing bandpass filtering and trend surface fitting on the multi-component magnetic measurement data to obtain the three-component vector magnetic field data; using the stratigraphic attitude data as a geometric constraint, iteratively inverting the three-component vector magnetic field data to obtain the three-dimensional distribution of the magnetic source body.

[0010] In one possible implementation, extracting the tumble angle and boundary gradient direction and generating magnetic source body orientation constraints includes: determining the principal axis direction of the magnetic source body and the tumble angle corresponding to the principal axis direction of the magnetic source body based on the three-dimensional distribution of the magnetic source body; determining the boundary gradient direction based on the magnetic field gradient at the boundary of the three-dimensional distribution of the magnetic source body; and generating magnetic source body orientation constraints based on the principal axis direction of the magnetic source body and the boundary gradient direction.

[0011] In one possible implementation, extracting the tumble angle and boundary gradient direction to generate magnetic source body orientation constraints further includes: performing depth layering on the multi-component magnetic measurement data based on the tumble angle to obtain the magnetic anomaly distribution of each depth layer; generating multiple orientation constraint vectors based on the magnetic anomaly distribution of each depth layer, the principal axis direction of the magnetic source body, and the boundary gradient direction; and the multiple orientation constraint vectors constituting the magnetic source body orientation constraints.

[0012] In one possible implementation, orientation correction and position registration include: determining the anomaly center position, anomaly distribution range, and anomaly principal axis direction corresponding to the gravity anomaly data and transient electromagnetic anomaly data, respectively; correcting the anomaly principal axis direction according to the magnetic source body orientation constraint; and registering the gravity anomaly data and transient electromagnetic anomaly data in position according to the corrected anomaly principal axis direction, anomaly center position, and anomaly distribution range.

[0013] In one possible implementation, the orientation correction includes: calculating the angle between the abnormal principal axis direction and the magnetic source body orientation constraint; if the angle is greater than a preset threshold, rotating and adjusting the abnormal principal axis direction with the abnormal center position as the rotation center so that the angle between the abnormal principal axis direction and the magnetic source body orientation constraint is not greater than the preset threshold.

[0014] In one possible implementation, constructing the target probability distribution includes: determining the contribution weights of the registered gravity anomaly data and the registered transient electromagnetic anomaly data based on the spatial overlap and directional consistency between the registered gravity anomaly data, the registered transient electromagnetic anomaly data, and the three-dimensional distribution of the magnetic source body; and then weighting and fusing the registered gravity anomaly data, the registered transient electromagnetic anomaly data, and the three-dimensional distribution of the magnetic source body according to the contribution weights to obtain the target probability distribution.

[0015] In one possible implementation, weighted fusion includes: mapping the location-registered gravity anomaly data, location-registered transient electromagnetic anomaly data, and the three-dimensional distribution of magnetic sources to a unified spatial grid; statistically analyzing the gravity anomaly response value, transient electromagnetic anomaly response value, and magnetic source response value for each spatial grid cell, and calculating the target probability value for each spatial grid cell based on the contribution weight; and generating the target probability distribution of the blind ore body based on the target probability value of each spatial grid cell.

[0016] In one possible implementation, determining the three-dimensional positioning result of the blind ore body and outputting the coordinates of the drilling target area and the drilling depth range includes: performing kernel density estimation on the target probability distribution to obtain candidate connected regions; determining the three-dimensional positioning result of the blind ore body based on the matching result of the geometric morphology parameters of the candidate connected regions and the stratigraphic occurrence data; outputting the coordinates of the drilling target area based on the spatial centroid of the three-dimensional positioning result of the blind ore body, and outputting the drilling depth range based on the top and bottom boundary depths of the three-dimensional positioning result of the blind ore body.

[0017] Secondly, this application provides a blind orebody positioning system based on multi-component magnetic survey data, used to implement a blind orebody positioning method based on multi-component magnetic survey data. The system includes:

[0018] The data inversion module is used to acquire multi-component magnetic measurement data, gravity anomaly data, transient electromagnetic anomaly data and stratigraphic attitude data, and to perform constrained inversion on the multi-component magnetic measurement data based on the stratigraphic attitude data to obtain the three-dimensional distribution of the magnetic source body.

[0019] The orientation constraint module is used to extract the tumble angle and boundary gradient direction based on the three-dimensional distribution of the magnetic source body, and generate orientation constraints for the magnetic source body.

[0020] The correction and fusion module is used to perform orientation correction and position registration on gravity anomaly data and transient electromagnetic anomaly data based on the orientation constraints of the magnetic source body, and to construct the target probability distribution of the blind ore body by combining the three-dimensional distribution of the magnetic source body.

[0021] The positioning output module is used to determine the three-dimensional positioning result of the blind ore body based on the target probability distribution and stratigraphic occurrence data, and output the coordinates of the drilling target area and the drilling depth range based on the three-dimensional positioning result of the blind ore body.

[0022] Compared with existing technologies, the advantages and beneficial effects of this application are as follows:

[0023] By employing multi-component magnetic inversion techniques constrained by stratigraphic attitude, the stable determination of the three-dimensional distribution of magnetic sources was achieved, enhancing the directional basis for interpreting deep anomalies. By extracting dip angles and boundary gradient directions, directional constraints on magnetic sources were constructed, improving the feasibility of unified correction of multi-source anomalies. Through directional correction, location registration, and probabilistic fusion techniques, the synergistic localization of gravity anomalies, transient electromagnetic anomalies, and the three-dimensional distribution of magnetic sources was achieved. Finally, by combining target probability distribution with stratigraphic attitude determination, three-dimensional localization results of blind ore bodies and drilling target area parameters were output, reducing drilling deployment deviations. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the method described in this application;

[0025] Figure 2This is a block diagram of the system modules in this application. Detailed Implementation

[0026] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0027] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0028] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0029] Multi-component magnetic survey data refers to information on multiple directional magnetic field components acquired at the same observation location. Compared to single-channel magnetic survey results, which only reflect overall strength variations, multi-component magnetic survey data can characterize not only the amplitude distribution of magnetic anomalies but also their directional changes, boundary transition characteristics, and spatial extension trends. Therefore, it is more suitable for analyzing the geometric morphology and burial state of underground magnetic bodies. With the development of quantum geomagnetic measurement technology, the sensitivity, stability, and component resolution of magnetic field observations have been further improved, providing a better data foundation for multi-component magnetic survey data in deep anomaly identification, weak anomaly extraction, and magnetic information expression in complex tectonic zones. Furthermore, by combining multi-component magnetic survey data with other geophysical and geological structural information for synergistic interpretation, it is possible to move beyond simple anomaly response analysis to a comprehensive determination of the spatial location and extension characteristics of underground targets. Based on this, this application proposes a method for locating blind ore bodies based on multi-component magnetic survey data.

[0030] like Figure 1 As shown, a method for locating blind ore bodies based on multi-component magnetic survey data is described. This method includes:

[0031] Acquire multi-component magnetic measurement data, gravity anomaly data, transient electromagnetic anomaly data, and stratigraphic attitude data. Perform constrained inversion on the multi-component magnetic measurement data based on the stratigraphic attitude data to obtain the three-dimensional distribution of the magnetic source body.

[0032] In one implementation, multi-component magnetic survey data, gravity anomaly data, transient electromagnetic anomaly data, and stratigraphic attitude data are first acquired in the target survey area. These data types are then aligned using unified coordinates, unified elevation datums, and sampling intervals to form a basic dataset under the same spatial reference. Subsequently, using the strike, dip, and dip angle information provided by the stratigraphic attitude data, constrained inversion is performed on the multi-component magnetic survey data to obtain the three-dimensional distribution of magnetic sources. The key processing point in this stage is to first eliminate reference differences between different data sources, and then introduce geological structural information into the inversion process. This ensures that the inversion results not only reflect the spatial location of local magnetic anomalies but also the extension direction, depth variation, and spatial boundaries of the magnetic sources, providing a stable foundation for subsequent extraction of dip angles, boundary gradient directions, and direction correction and location registration of multi-source anomalies.

[0033] The constrained inversion includes: performing bandpass filtering and trend surface fitting on the multi-component magnetic data to obtain the three-component vector magnetic field data; using the stratigraphic attitude data as geometric constraints, iteratively inverting the three-component vector magnetic field data to obtain the three-dimensional distribution of the magnetic source body.

[0034] In one embodiment, the constrained inversion process can be divided into two levels: data preprocessing and iterative inversion. First, the multi-component magnetic survey data is preprocessed. Preprocessing is not simple filtering, but rather a targeted organization based on the target depth and background conditions of the survey area. For the original magnetic survey records, obviously distorted jump points and missing points are first removed, and then points are smoothly added according to the survey line sequence to prevent discrete anomalies from directly entering the inversion stage. Next, bandpass filtering is performed, with the bandpass range set according to the expected depth of the target body in the survey area, the survey line spacing, and the sampling interval. When the expected depth is large, the low-frequency retention ratio is appropriately increased; when shallow interference is strong, the high-frequency cutoff position is appropriately lowered to reduce the local spike effects caused by scattered magnetic bodies on the surface.

[0035] After bandpass filtering, trend surface fitting is performed on the data to separate the regional background field from the local anomaly field. The order of the trend surface is set according to the smoothness of the background changes in the survey area; a lower order is used when the background changes are relatively stable, and a higher order is used when the background fluctuations are more obvious, but the principle is not to excessively weaken the target anomaly. After the above processing, vector magnetic field three-component data suitable for inversion solution are obtained. Here, vector magnetic field three-component data refers to magnetic field component data corresponding to three orthogonal directions in a unified coordinate system, used to characterize the directional information of magnetic field changes in the target survey area.

[0036] After preprocessing, the iterative inversion stage begins. First, a three-dimensional discrete grid is established based on the survey area, and the stratigraphic attitude data is converted into spatial geometric constraints. Specifically, strike information is used to limit the planar extension direction of the magnetic source body, and dip and dip angle information is used to limit the vertical expansion trend of the magnetic source body. For areas with significant fault development, the location of the main fault zones can also be used as boundary constraints to avoid the inversion results crossing obvious discontinuous structures. The initial model can be a uniform weakly magnetized body model or a partitioned initial value model set based on known mineralization information. Subsequently, the difference between the three components of the vector magnetic field data and the calculation results of the initial model is used to progressively correct the magnetization intensity and spatial distribution of each grid cell. After each correction, the model continuity, boundary stability, and consistency with stratigraphic attitude are checked.

[0037] When the overall change in the results of two adjacent inversion rounds falls below a preset convergence threshold, the iteration stops, and the three-dimensional distribution of the magnetic source body is output. The convergence threshold is typically set based on the noise level and resolution requirements of the survey area; it is appropriately relaxed when noise is high and appropriately tightened when high precision positioning is required. This processing method combines the directional information from multi-component magnetic survey data with the structural information from stratigraphic attitude data to obtain a three-dimensional distribution of the magnetic source body with spatial constraints, providing directly usable foundational results for subsequent steps.

[0038] Based on the three-dimensional distribution of the magnetic source body, the tilt angle and boundary gradient direction are extracted to generate magnetic source body orientation constraints;

[0039] After determining the three-dimensional distribution of magnetic sources in the target survey area, the process continues by extracting the dip angle and boundary gradient direction based on this distribution, and generating directional constraints for the magnetic sources. The purpose of this stage is to convert the spatial morphological information already formed in the three-dimensional distribution of the magnetic sources into directional information that can be directly used for subsequent multi-source anomaly correction. Specifically, the main extension direction of the magnetic sources is first identified, and then the inclination of this main extension direction relative to the horizontal plane is determined to form the dip angle. Subsequently, the boundary gradient direction is extracted in the boundary area between the magnetic sources and the surrounding rock to form boundary control information. Finally, the main extension direction, dip angle, and boundary gradient direction are combined to generate directional constraints for the magnetic sources. This process transforms the three-dimensional distribution of the magnetic sources from a static spatial result into a constraint result with directional guidance, providing a basis for subsequent directional correction, position registration, and joint positioning of gravity anomalies and transient electromagnetic anomalies.

[0040] Extracting the tumble angle and boundary gradient direction and generating magnetic source body orientation constraints includes: determining the principal axis direction of the magnetic source body and the corresponding tumble angle based on the three-dimensional distribution of the magnetic source body; determining the boundary gradient direction based on the magnetic field gradient at the boundary of the three-dimensional distribution of the magnetic source body; and generating magnetic source body orientation constraints based on the principal axis direction and boundary gradient direction.

[0041] In one embodiment, the process of extracting the pledging angle and boundary gradient direction and generating magnetic source body orientation constraints focuses on separately identifying the morphological features and boundary features in the three-dimensional distribution of the magnetic source body, and then organizing them uniformly. First, the three-dimensional distribution of the magnetic source body is divided into connected regions, and spatially continuous regions with consistent magnetization intensity and stable burial depth are grouped into the same magnetic source body candidate region.

[0042] When dividing connected regions, either a voxel adjacency method or a aggregation method based on distance and magnetization change rate can be used. The core principle is to avoid mistakenly merging separate anomalous volumes. After dividing the candidate regions, the extension length, extension width, and extension thickness of each magnetic source candidate region are calculated, and the direction with the largest extension length is taken as the principal axis direction of the magnetic source. To avoid interference from local spikes or edge noise in determining the principal axis direction, the overall distribution characteristics are preferred when calculating the principal axis direction, rather than the position of a single extreme point. After the principal axis direction is determined, the pledging angle is determined based on the vertical variation of the principal axis direction. The pledging angle can be understood as the degree of inclination of the principal axis direction relative to the horizontal plane; a larger value is taken when the downward extension is obvious, and a smaller value is taken when the distribution is approximately horizontal.

[0043] For magnetic sources with complex extended morphologies, local plunging angles can be calculated segment by depth, and then a weighted average can be taken as the plunging angle of the current candidate magnetic source region. The weights are determined by the anomaly intensity and connectivity stability of each depth segment. Subsequently, boundary gradient directions are extracted from the three-dimensional distribution boundary regions of the magnetic source. The boundary region is identified by the obvious change in magnetization intensity as it transitions from a high-value area to the background area. During processing, the magnetic field change trend is calculated point by point along the outer contour of the magnetic source, and the direction with the most significant change is selected as the gradient direction of the current boundary point. The gradient directions of adjacent boundary points are then smoothed to form a set of boundary gradient directions. These boundary gradient directions reflect the main change direction of the anomaly boundary transitioning from the inside to the outside, and are used to describe the pointing characteristics of the magnetic source on the spatial boundary. After extracting the principal axis direction, plunging angle, and boundary gradient direction, these three are correlated.

[0044] The specific approach involves using the principal axis direction as the main directional information, the dip angle as the vertical variation information, and the boundary gradient direction as the boundary control information. These are stored uniformly within the same magnetic source candidate region to generate magnetic source direction constraints. These constraints are not abstract markers but rather directional results that correspond one-to-one with specific magnetic source candidate regions. When performing multi-source anomaly alignment later, these directional results can be directly invoked to constrain the main extension direction, boundary trends, and spatial positions of gravity anomalies and transient electromagnetic anomalies. To ensure the stability of the magnetic source direction constraints, a consistency check can be added after generation. When the angle between the principal axis direction and the boundary gradient direction is significantly abnormal, or when the dip angle varies excessively across different depth segments within the same magnetic source candidate region, the process reverts to the 3D distribution of the magnetic source to re-check the connected region partitioning results before deciding whether to re-extract the direction parameters. This process establishes a clear source and usage relationship between the magnetic source direction constraints and the 3D distribution of the magnetic source, facilitating subsequent invocation and preventing directional information from existing independently of the original inversion results.

[0045] Extracting the tumble angle and boundary gradient direction to generate magnetic source body orientation constraints also includes: performing depth layering on multi-component magnetic measurement data based on the tumble angle to obtain the magnetic anomaly distribution of each depth layer; generating multiple orientation constraint vectors based on the magnetic anomaly distribution of each depth layer, the principal axis direction of the magnetic source body, and the boundary gradient direction; and the multiple orientation constraint vectors constitute the magnetic source body orientation constraints.

[0046] In one embodiment, after extracting the principal axis direction, dip angle, and boundary gradient direction of the magnetic source body, depth layering processing is performed on the multi-component magnetic measurement data, and multiple directional constraint vectors are generated based on the layering results. These multiple directional constraint vectors together constitute the magnetic source body's directional constraints. Compared to using only a single overall directional result, this embodiment adds processing content that refines directional information by depth segment. The purpose is to reflect the possible extension, deflection, dip changes, and boundary changes of the magnetic source body at different burial depths, so that subsequent directional correction and position registration have a layered constraint basis, rather than relying solely on a single overall directional result.

[0047] In practice, the depth stratification method is first determined based on the obtained dip angle. When the dip angle is small, the magnetic source extends relatively gently, and the thickness of the depth segment can be appropriately increased; when the dip angle is large, the magnetic source changes more significantly along the depth direction, and the thickness of the depth segment should be appropriately decreased. The division of depth segments is also determined in conjunction with the sampling interval of the survey area, the target burial depth range, and the strength of anomaly changes. In areas with denser sampling and faster changes in target burial depth, the number of layers is appropriately increased; in areas with sparser sampling or gentler anomaly changes, the number of layers is appropriately reduced. The stratification principle is to preserve the anomaly differences in the depth direction while avoiding excessively fine segments that lead to instability in single-layer data.

[0048] After completing the depth stratification, the magnetic anomaly distribution is statistically analyzed within each depth segment. The magnetic anomaly distribution includes the anomaly center location, anomaly boundary range, and main extension trend within the current segment. The anomaly center location reflects the concentrated location of the magnetic anomaly response in that segment; the anomaly boundary range reflects the spatial coverage of the anomaly body in that segment; and the main extension trend reflects the planar orientation change of the magnetic anomaly within that segment. If multiple separate anomaly regions exist within a certain depth segment, the distribution results of each anomaly region are extracted separately and then merged according to their correspondence with the three-dimensional distribution of the magnetic source body. Subsequently, the magnetic anomaly distribution of each depth segment is jointly analyzed with the previously determined principal axis direction and boundary gradient direction of the magnetic source body to generate multiple directional constraint vectors. Each directional constraint vector corresponds to a specific depth segment and contains information on the main extension direction, dip continuation, and boundary control information of that depth segment.

[0049] The primary extension direction information is determined by the magnetic anomaly distribution and the principal axis direction of the magnetic source body within the layer. The plunging continuation information is determined by the projection relationship of the plunging angle within the layer. The boundary control information is determined by the pointing characteristics of the boundary gradient direction on the anomaly boundary of the layer. If the magnetic anomaly distribution of a certain layer deviates little from the overall principal axis direction, the direction constraint vector of that layer inherits more of the overall direction result. If a certain layer exhibits significant bending, deflection, or boundary changes, the magnetic anomaly distribution of that layer is used as the correction basis to locally adjust the direction constraint vector. To ensure the continuity between the direction constraint vectors of each layer, interlayer smoothing checks are also performed between adjacent layers.

[0050] When the directional change between two adjacent depth segments exceeds a preset threshold, the system first checks whether the magnetic anomaly distribution in the current segment is affected by local noise before deciding whether to use interpolation correction between adjacent segments. The preset threshold is set based on the structural complexity and anomaly continuity, with smaller values ​​for structurally stable regions and larger values ​​for regions with significant structural transitions. After multiple directional constraint vectors are generated, they are organized in depth order to form magnetic source body directional constraints. The magnetic source body directional constraints formed in this way are no longer single-directional parameters, but a set of directional constraints with depth resolution, which can more accurately characterize the extension and variation characteristics of the magnetic source body in three-dimensional space, and provide a more specific basis for subsequent multi-source anomaly layer direction correction and layer position registration.

[0051] Based on the orientation constraint of the magnetic source body, the orientation correction and position registration of gravity anomaly data and transient electromagnetic anomaly data are performed, and the target probability distribution of the blind ore body is constructed by combining the three-dimensional distribution of the magnetic source body.

[0052] In one implementation, after obtaining the directional constraints of the magnetic source bodies, the gravity anomaly data and transient electromagnetic anomaly data are further directionally corrected and spatially registered based on the magnetic source body directional constraints, and a target probability distribution of the blind ore body is constructed by combining the three-dimensional distribution of the magnetic source bodies. The key processing point in this stage is to first unify the spatial morphological information related to the target ore body in the gravity anomalies and transient electromagnetic anomalies into the directional framework corresponding to the three-dimensional distribution of the magnetic source bodies, and then fuse the corrected multi-source anomalies to form a spatial distribution result that can characterize the possibility of the existence of the target ore body. In specific implementation, the center position, distribution boundary and extension direction of various anomalies are first extracted, and then the directional adjustment and spatial alignment are completed using the magnetic source body directional constraints. Subsequently, the target probability distribution is constructed based on the spatial correspondence between the multi-source anomalies and the three-dimensional distribution of the magnetic source bodies, thereby providing a unified, continuous and interpretable data foundation for the subsequent three-dimensional positioning of the blind ore body.

[0053] Orientation correction and position registration include: determining the anomaly center position, anomaly distribution range, and anomaly principal axis direction corresponding to gravity anomaly data and transient electromagnetic anomaly data, respectively; correcting the anomaly principal axis direction according to the magnetic source body orientation constraint; and registering the gravity anomaly data and transient electromagnetic anomaly data in position according to the corrected anomaly principal axis direction, anomaly center position, and anomaly distribution range.

[0054] In one embodiment, orientation correction and position registration primarily focus on the spatial geometric features of the anomaly. A new constraint is that the original anomaly maps are not directly used for fusion. Instead, morphological parameters are first extracted from the gravity anomaly data and transient electromagnetic anomaly data, and then spatial alignment is completed based on the same set of orientation information. The purpose of this approach is to unify the representation methods of different geophysical methods, avoiding inconsistencies in subsequent fusion standards caused by one data source using grid boundary descriptions and another using center point descriptions.

[0055] In practice, the process begins by identifying local high- or low-value anomaly regions in the gravity anomaly data. The anomaly center location, anomaly distribution range, and anomaly principal axis direction are then determined based on the anomaly's closed boundary. The anomaly center location can be determined using the spatial centroid after weighting the anomaly amplitude; the anomaly distribution range can be determined using the outer boundary of the anomaly's closed contour; and the anomaly principal axis direction can be determined using the direction of the long side of the anomaly contour. For transient electromagnetic anomaly data, the processing method is consistent with that for gravity anomaly data: first, the region with the most concentrated anomaly response is identified, and then the corresponding anomaly center location, anomaly distribution range, and anomaly principal axis direction are extracted. Since transient electromagnetic anomalies are often affected by shallow conductive layers, response segments that correspond to the target burial depth are preferentially retained when determining the anomaly boundary, while boundary portions that significantly deviate from the target depth are weakened. After extracting the geometric parameters for both types of anomalies, directional correction is performed based on the magnetic source body's directional constraints.

[0056] Orientation correction targets the principal axis direction of the anomaly, aiming to align the main extension directions of gravity anomalies and transient electromagnetic anomalies as closely as possible with the main extension direction of the magnetic source body. After orientation correction, position registration is performed. Position registration is not simply translating the anomaly center; rather, it involves comprehensively considering the anomaly principal axis direction after orientation correction, the anomaly center position, and the anomaly distribution range to make corresponding adjustments to the gravity anomaly and transient electromagnetic anomaly. Specifically, the projected area of ​​the magnetic source body's three-dimensional distribution at the current depth position can be used as a reference. Then, based on the offset between the gravity anomaly and the transient electromagnetic anomaly and this projected area, the center positions and boundary ranges of the two types of anomalies are corrected respectively.

[0057] If both types of anomalies have a stable overlap with the projection area of ​​the magnetic source, the original boundaries are maintained, with only minor adjustments to the center position. If there is only partial overlap, the boundary closest to the projection area of ​​the magnetic source is retained first, and the portion farther away is truncated. After location registration, the gravity anomaly data and transient electromagnetic anomaly data establish a correspondence with the three-dimensional distribution of the magnetic source in both direction and spatial location, providing conditions for subsequent unified calculation of spatial overlap and directional consistency. Through this step-by-step processing method, direction correction and location registration form a clear sequential connection, avoiding anomaly misalignment caused by direct fusion and providing a unified spatial reference for multi-source anomaly results.

[0058] Direction correction includes: calculating the angle between the abnormal principal axis direction and the magnetic source body direction constraint; if the angle is greater than a preset threshold, rotating and adjusting the abnormal principal axis direction with the abnormal center position as the rotation center so that the angle between the abnormal principal axis direction and the magnetic source body direction constraint is not greater than the preset threshold.

[0059] In one embodiment, orientation correction is further refined into two actions: angle calculation and rotation adjustment. The added limitation is that a clear criterion is provided for determining whether an abnormal principal axis orientation needs adjustment, and implementable control rules are given for the adjustment method. The purpose of this approach is to avoid orientation correction relying too heavily on experience-based judgment, enabling those skilled in the art to complete orientation correction according to uniform rules in different measurement areas.

[0060] In practice, the angles between the principal axis of the gravity anomaly and the magnetic source body direction constraint, as well as the angles between the principal axis of the transient electromagnetic anomaly and the magnetic source body direction constraint, are calculated first. The angle calculations are based on the projection direction of the same plane, prioritizing the comparison of direction results within the current depth segment to avoid mixing directions between deep and shallow layers. After the angle calculations are completed, the angles are compared with preset thresholds. These preset thresholds distinguish between normal directional deviations and directional deviations requiring correction. Their setting principles are related to the complexity of the survey area's structure, sampling density, and the stability of the anomaly boundary. In areas with relatively stable structures and dense survey lines, the preset threshold can be set to a smaller value; in areas with many structural transitions and large fluctuations in the anomaly boundary, the preset threshold can be appropriately relaxed.

[0061] In practical applications, the preset threshold can be determined by combining statistical results from known ore body sample areas, or it can be manually set based on historical interpretation results. If the included angle is not greater than the preset threshold, the current anomaly's principal axis direction is considered to have acceptable consistency with the magnetic source body's direction constraint, and no further rotation adjustment is performed. If the included angle is greater than the preset threshold, the anomaly's principal axis direction is rotated using the anomaly's center position as the rotation center. The rotation adjustment process is not a one-time forced alignment, but rather a gradual rotation with small steps, recalculating the included angle after each rotation until the included angle is no greater than the preset threshold. Using the anomaly's center position as the rotation center ensures that the main body position of the anomaly remains stable during the direction adjustment process, avoiding overall positional drift caused by improper selection of the rotation center. To prevent significant distortion of the anomaly boundary after rotation adjustment, the aspect ratio change of the anomaly's distribution range can be checked synchronously after each rotation. When the aspect ratio change exceeds the allowable range, the current step size is reduced, and the adjustment is re-executed.

[0062] The same rules can be used for orientation correction for both gravity anomalies and transient electromagnetic anomalies. However, in practical applications, transient electromagnetic anomalies are more susceptible to the influence of shallow conductors, so their allowable rotation step size can be slightly smaller than that for gravity anomalies. By introducing angle determination and a step-by-step rotation mechanism, orientation correction is transformed from an empirical process into a verifiable and repeatable process. This also ensures that subsequent position registration is based on consistent orientation, reducing misregistration where the anomaly centers are close but their extension directions are significantly different.

[0063] The construction of the target probability distribution includes: determining the contribution weights of the registered gravity anomaly data and the registered transient electromagnetic anomaly data based on the spatial overlap and directional consistency between the registered gravity anomaly data, the registered transient electromagnetic anomaly data, and the three-dimensional distribution of the magnetic source body; and then weighting and fusing the registered gravity anomaly data, the registered transient electromagnetic anomaly data, and the three-dimensional distribution of the magnetic source body according to the contribution weights to obtain the target probability distribution.

[0064] In one embodiment, the key to constructing the target probability distribution lies in transforming the spatial correspondence between the registered gravity anomaly data, the registered transient electromagnetic anomaly data, and the three-dimensional distribution of the magnetic source into a unified contribution weight, and then performing weighted fusion based on the contribution weight. The added limitation of this embodiment is that specific participation methods are set for the degree of spatial overlap and directional consistency, so that the target probability distribution is no longer a simple superposition result, but a comprehensive result reflecting the correspondence between multiple anomalies and magnetic source bodies.

[0065] In practice, the spatial overlap between the registered gravity anomaly data and the three-dimensional distribution of the magnetic source body is first calculated within the same depth segment. The spatial overlap between the registered transient electromagnetic anomaly data and the three-dimensional distribution of the magnetic source body is also calculated. The degree of spatial overlap can be obtained as the ratio of the overlapping area to the reference area; a larger ratio indicates a closer spatial correspondence between the anomaly and the magnetic source body. Subsequently, the directional consistency between the registered gravity anomaly data and the registered transient electromagnetic anomaly data and the three-dimensional distribution of the magnetic source body is calculated. Directional consistency characterizes the proximity of the anomaly's principal axis direction to the magnetic source body's directional constraint; the closer the directions, the higher the directional consistency.

[0066] After calculating the above two types of indicators, the contribution weight is further determined by combining the normalized outlier amplitude. For ease of implementation, the contribution weight can be calculated using the following formula:

[0067]

[0068] in, To contribute weight, The ratio of spatial overlap areas. For consistency of direction, To normalize the outlier amplitude, , , These are the weighting coefficients, and .

[0069] The spatial overlap area ratio reflects the degree of overlap between the anomaly and the magnetic source body in the same space after location registration; directional consistency reflects the closeness between the anomaly extension direction and the directional constraint of the magnetic source body; and normalized anomaly amplitude reflects the relative significance of the current anomaly in its data source. The setting principle of the weighting coefficients is related to the differences in physical properties of the survey area. If the density difference of the target ore body is more significant, the influence ratio of the spatial overlap area ratio and the anomaly amplitude corresponding to the gravity anomaly should be appropriately increased; if the conductivity difference of the target ore body is more significant, the influence ratio of the indicators corresponding to the transient electromagnetic anomaly should be appropriately increased.

[0070] After calculating the contribution weights of various anomalies, the gravity anomaly data, transient electromagnetic anomaly data, and the 3D distribution of magnetic sources after location registration are weighted and fused according to their contribution weights. During weighted fusion, the contributions of anomalies from different data sources within the same grid cell are accumulated based on the spatial grid where the 3D distribution of magnetic sources is located, to obtain the target probability value for each spatial cell. Regions with higher target probability values ​​indicate that multi-source anomalies and magnetic sources have strong consistency in spatial location and direction; regions with lower target probability values ​​indicate that at least one anomaly lacks a stable correspondence with the magnetic source. Through this processing method, the originally scattered multi-source anomaly results can be integrated into a unified probability expression result, providing a direct basis for subsequently determining the 3D location range of blind ore bodies.

[0071] The weighted fusion includes: mapping the location-registered gravity anomaly data, location-registered transient electromagnetic anomaly data, and the three-dimensional distribution of magnetic sources to a unified spatial grid; statistically analyzing the gravity anomaly response value, transient electromagnetic anomaly response value, and magnetic source response value for each spatial grid cell, and calculating the target probability value for each spatial grid cell based on the contribution weight; and generating the target probability distribution of the blind ore body based on the target probability value of each spatial grid cell.

[0072] In one embodiment, the weighted fusion stage is performed after direction correction and position registration. The added constraint is that anomaly results from different sources are first converted to the same spatial grid, and then the target probability value is calculated based on the spatial grid cells, ultimately forming the target probability distribution of the blind ore body. This constraint is introduced because the sampling interval, data form, and spatial resolution of gravity anomaly data, transient electromagnetic anomaly data, and the three-dimensional distribution of magnetic sources are usually inconsistent. Direct superposition can easily lead to local layering, boundary fractures, or probability peak shifts.

[0073] In practice, a unified spatial grid covering the target survey area is first established. This unified spatial grid is constructed in three dimensions, with the grid size determined based on the survey line spacing, sampling density, target burial depth range, and the expected scale of the anomaly. For areas with denser sampling and smaller target bodies, a smaller grid size is used; for areas with sparser sampling and larger target bodies, a larger grid size is used. The principle is to preserve the spatial variation characteristics of the anomaly while avoiding excessively fine grids that could lead to discrete calculation results. After establishing the unified spatial grid, the registered gravity anomaly data, registered transient electromagnetic anomaly data, and the three-dimensional distribution of magnetic sources are mapped onto the unified spatial grid.

[0074] During mapping, spatial interpolation is preferentially used to determine the anomalous response values ​​within each grid cell for gravity anomaly data and transient electromagnetic anomaly data. For the three-dimensional distribution of magnetic sources, the magnetic source response value is determined based on the proportion of the magnetic source in the grid and the magnetization distribution. If a grid cell intersects with multiple anomaly regions simultaneously, the primary response with the highest amplitude and closest to the current location is retained, and the remaining responses are used as auxiliary information in subsequent weighting. After mapping, the gravity anomaly response value, transient electromagnetic anomaly response value, and magnetic source response value are statistically analyzed for each spatial grid cell. Here, the gravity anomaly response value is used to characterize the degree of density difference within the current grid cell, the transient electromagnetic anomaly response value is used to characterize the degree of conductivity anomaly within the current grid cell, and the magnetic source response value is used to characterize the correspondence between the current grid cell and the three-dimensional distribution of magnetic sources. Subsequently, the contribution weights determined in the previous steps are introduced, and the three response values ​​are jointly calculated to obtain the target probability value for each spatial grid cell.

[0075] In actual calculations, anomaly types with larger contribution weights have a higher proportion of influence in the target probability value, while anomaly types with smaller contribution weights have a correspondingly lower proportion of influence. This ensures that the fusion result reflects the true correspondence between multi-source anomalies and magnetic source bodies. For grid cells with only a single anomaly response and lacking support from other anomalies, the target probability value is retained but appropriately reduced to avoid isolated local anomalies directly forming high-probability false target areas. For grid cells with simultaneously high responses from gravity anomalies, transient electromagnetic anomalies, and magnetic source bodies, and whose spatial locations are consistent, a higher target probability value is assigned. After the target probability values ​​of all spatial grid cells are calculated, they are reorganized according to spatial location to generate a continuous target probability distribution for blind ore bodies.

[0076] To ensure the interpretability of the target probability distribution, a continuity check can be performed on the probability changes between adjacent grid cells. When the probability difference between adjacent grid cells is too large and there is a lack of corresponding anomaly support in the surrounding area, the current grid cell is marked as a local anomaly cell and adjusted using a neighborhood correction method. After the above processing, the target probability distribution can simultaneously reflect the spatial overlap, directional consistency, and spatial constraints of the magnetic source body of multi-source anomalies, providing a unified input for subsequent extraction of candidate connected regions and determination of the three-dimensional positioning results of blind ore bodies.

[0077] The three-dimensional positioning results of the blind ore body are determined based on the target probability distribution and stratigraphic occurrence data, and the coordinates of the drilling target area and the drilling depth range are output based on the three-dimensional positioning results of the blind ore body.

[0078] In one implementation, after obtaining the target probability distribution, the three-dimensional positioning result of the blind ore body is determined by combining stratigraphic occurrence data, and the coordinates of the drilling target area and the drilling depth range are output. During processing, continuous high-value regions in the target probability distribution are first identified within a unified spatial grid. Then, the consistency between each continuous high-value region and the geological structure is judged by combining the stratigraphic strike, dip, and dip angle, and target areas that match the direction of mineralization are selected. Subsequently, the spatial center of the target area is used as the coordinates of the drilling target area, and the depth range corresponding to the upper and lower vertical boundaries of the target area is used as the drilling depth range. Through this processing, the probability expression result can be converted into a spatial positioning result that can be directly used for borehole deployment.

[0079] The process of determining the three-dimensional location of the blind ore body and outputting the coordinates of the drilling target area and the drilling depth range includes: estimating the kernel density of the target probability distribution to obtain candidate connected regions; determining the three-dimensional location of the blind ore body based on the matching results of the geometric morphology parameters of the candidate connected regions and the stratigraphic occurrence data; outputting the coordinates of the drilling target area based on the spatial centroid of the three-dimensional location of the blind ore body, and outputting the drilling depth range based on the top and bottom boundary depths of the three-dimensional location of the blind ore body.

[0080] In one embodiment, determining the three-dimensional location of the blind ore body and outputting the coordinates of the drilling target area and the drilling depth range can be carried out in three levels: candidate region extraction, structural matching determination, and engineering parameter output. First, kernel density estimation is performed on the target probability distribution to transform discrete high-value cells into a continuous probability field. The purpose of kernel density estimation here is to reduce the influence of accidental high values ​​of individual grid cells on the location results and to form continuous regions from spatially adjacent and probabilistically similar anomalous cells. During kernel density estimation, the smoothing range is set according to the grid size, sampling density, and the expected scale of the anomalous body.

[0081] When the grid is dense, a smaller smoothing range is used to preserve local details; when the grid is sparse, the smoothing range is appropriately increased to avoid dividing continuous areas into multiple scattered micro-regions. After kernel density estimation, candidate connected regions that reach a preset probability threshold are extracted by classifying them from high to low according to the target probability value. The preset probability threshold is primarily determined based on the background probability distribution within the survey area and the statistical results of known anomalous sample areas. In principle, it should be able to exclude large areas of low-probability noise without losing the main high-probability clusters. If the survey area lacks prior sample areas, the high quantile value of the target probability distribution can be used as the initial threshold, and then corrected by combining the number and area of ​​connected regions. After extracting the candidate connected regions, the geometric morphological parameters of each candidate connected region are calculated. The geometric morphological parameters include the aspect ratio, extension direction, thickness variation trend, spatial connectivity, and vertical extension range. Among them, the aspect ratio of the plane is used to reflect whether the candidate connected region is strip-shaped or block-shaped, the extension direction is used to reflect the main extension trend of the candidate connected region on the plane, the thickness variation trend is used to reflect whether the candidate connected region gradually converges or gradually expands along the depth direction, the spatial connectivity is used to reflect whether there are obvious breaks and discrete voids inside the candidate connected region, and the vertical extension range is used to reflect the distribution of the top and bottom boundary depths of the candidate connected region.

[0082] These geometric parameters are then matched with stratigraphic attitude data. During matching, the focus is on comparing whether the planar extension direction of the candidate connected region is close to the stratigraphic strike, whether the vertical extension trend of the candidate connected region is consistent with the stratigraphic dip and angle, and whether the overall shape of the candidate connected region conforms to the basic laws of ore body development along strata or structures. If the main extension direction of the candidate connected region deviates too much from the stratigraphic strike, or if the vertical extension trend is significantly opposite to the stratigraphic dip direction, the priority of the candidate connected region is reduced. If the extension direction, dip trend, and spatial continuity of the candidate connected region are highly consistent with the stratigraphic attitude, the candidate connected region is determined as the three-dimensional location result of the blind ore body. After determining the location result, the engineering deployment parameters are output. In specific processing, the spatial centroid of the three-dimensional location result of the blind ore body is first calculated, and the planar coordinates corresponding to the spatial centroid are used as the coordinates of the drilling target area.

[0083] The reason for using a spatial centroid instead of a single maximum probability point is that the spatial centroid can simultaneously consider the distribution range of the entire target area, resulting in more stable output results. Subsequently, the depth boundaries of the three-dimensional positioning results of the vertically statistical blind ore body are determined. The shallowest position in the continuous target area is taken as the top boundary depth, and the deepest position in the continuous target area is taken as the bottom boundary depth. The interval between the two is taken as the drilling depth interval. For target areas with obvious vertical stratification, priority drilling intervals can also be output according to the depth segment where the main probability peak is located, making the on-site implementation more targeted. Through the above processing, the target probability distribution not only completes the geological interpretation but also transforms into borehole coordinates and depth ranges with direct engineering application value.

[0084] like Figure 2 As shown, a blind orebody location system based on multi-component magnetic survey data is used to implement a blind orebody location method based on multi-component magnetic survey data. The system includes:

[0085] The data inversion module is used to acquire multi-component magnetic measurement data, gravity anomaly data, transient electromagnetic anomaly data and stratigraphic attitude data, and to perform constrained inversion on the multi-component magnetic measurement data based on the stratigraphic attitude data to obtain the three-dimensional distribution of the magnetic source body.

[0086] The hardware foundation of the data inversion module mainly includes multi-component magnetic surveying and acquisition equipment, gravity surveying and acquisition equipment, transient electromagnetic surveying and acquisition equipment, stratigraphic attitude recording terminal, data synchronization interface, industrial computer or high-performance processor, and memory. The multi-component magnetic surveying and acquisition equipment outputs three-component magnetic field observation data; the gravity surveying and acquisition equipment outputs gravity anomaly observation data; the transient electromagnetic surveying and acquisition equipment outputs electromagnetic response data; and the stratigraphic attitude recording terminal inputs or imports parameters such as stratigraphic strike, dip, and dip angle. The data synchronization interface is used to align the time and coordinates of sampling results from different devices. The industrial computer or high-performance processor performs preprocessing and constrained inversion calculations, and the memory caches raw observation data, intermediate inversion results, and three-dimensional distribution results of magnetic sources. Overall, this module is a combination of "multi-source acquisition hardware + computing hardware."

[0087] The orientation constraint module is used to extract the tumble angle and boundary gradient direction based on the three-dimensional distribution of the magnetic source body, and generate orientation constraints for the magnetic source body.

[0088] The hardware foundation of the orientation constraint module mainly includes a computational processing unit, a graphics processing unit or parallel acceleration unit, a memory unit, and an intermediate result caching unit. This module generally no longer directly connects to external sensors, but primarily receives the 3D distribution results of the magnetic source body from the data inversion module. The computational processing unit is used to identify the principal axis orientation of the magnetic source body, extract the tumble angle, and calculate the boundary gradient direction. When the number of 3D voxels is large, a graphics processing unit or parallel acceleration unit can be configured to improve the efficiency of 3D geometric analysis. The memory unit is used to temporarily store the 3D distribution mesh, boundary extraction results, and orientation parameters. The caching unit is used to store the orientation constraint results of the magnetic source body for subsequent use by the correction and fusion module. This module is essentially a 3D spatial geometric analysis hardware unit.

[0089] The correction and fusion module is used to perform orientation correction and position registration on gravity anomaly data and transient electromagnetic anomaly data based on the orientation constraints of the magnetic source body, and to construct the target probability distribution of the blind ore body by combining the three-dimensional distribution of the magnetic source body.

[0090] The hardware foundation of the correction and fusion module mainly includes a high-performance central processing unit (CPU), a graphics processing unit (GPU), shared memory, a high-speed bus, and large-capacity storage units. This module needs to simultaneously process gravity anomaly data, transient electromagnetic anomaly data, magnetic source orientation constraints, and the three-dimensional distribution of magnetic sources; therefore, it requires high data throughput and parallel computing capabilities. The CPU is used to execute orientation correction, position registration, and probabilistic fusion control processes. The GPU is suitable for handling large-scale mesh alignment, spatial overlap calculations, and probability field generation tasks. Shared memory reduces latency caused by repeated scheduling of multiple types of data within the module. The high-speed bus ensures rapid data exchange between the processor, memory, and external database. Large-capacity storage units are used to store position registration results and target probability distributions. This module can be understood as the core computing hardware of the entire system.

[0091] The positioning output module is used to determine the three-dimensional positioning result of the blind ore body based on the target probability distribution and stratigraphic occurrence data, and output the coordinates of the drilling target area and the drilling depth range based on the three-dimensional positioning result of the blind ore body.

[0092] The hardware foundation of the positioning output module mainly includes a result determination processor, a storage unit, a display terminal, a communication interface, and an external output interface. The result determination processor calculates the 3D positioning results based on the target probability distribution and stratigraphic occurrence data, and generates the coordinates of the drilling target area and the drilling depth range. The storage unit saves the final positioning results, historical positioning records, and output reports. The display terminal graphically displays the 3D location of the blind ore body, the distribution of the drilling target area, and the depth range. The communication interface transmits the positioning results to a host computer, geological interpretation platform, or drilling platform. The external output interface can connect to printing equipment, mobile terminals, or network servers to achieve result publishing and sharing. This module focuses more on result solidification, display, and downstream equipment integration.

[0093] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0094] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for locating blind ore bodies based on multi-component magnetic survey data, characterized in that, The method includes: Acquire multi-component magnetic measurement data, gravity anomaly data, transient electromagnetic anomaly data, and stratigraphic attitude data. Perform constrained inversion on the multi-component magnetic measurement data based on the stratigraphic attitude data to obtain the three-dimensional distribution of the magnetic source body. Based on the three-dimensional distribution of the magnetic source body, the tilt angle and boundary gradient direction are extracted to generate magnetic source body orientation constraints; Based on the orientation constraint of the magnetic source body, the gravity anomaly data and the transient electromagnetic anomaly data are oriented and registered, and the target probability distribution of the blind ore body is constructed by combining the three-dimensional distribution of the magnetic source body. The three-dimensional positioning results of the blind ore body are determined based on the target probability distribution and the stratigraphic occurrence data, and the coordinates of the drilling target area and the drilling depth range are output based on the three-dimensional positioning results of the blind ore body.

2. The method according to claim 1, characterized in that, The constrained inversion includes: Bandpass filtering and trend surface fitting are performed on the multi-component magnetic measurement data to obtain three-component vector magnetic field data; Using the stratigraphic attitude data as geometric constraints, the three-component data of the vector magnetic field are iteratively inverted to obtain the three-dimensional distribution of the magnetic source body.

3. The method according to claim 1, characterized in that, The step of extracting the tilt angle and boundary gradient direction and generating magnetic source volume orientation constraints includes: The principal axis direction of the magnetic source body and the corresponding tumble angle are determined based on the three-dimensional distribution of the magnetic source body. The direction of the boundary gradient is determined based on the magnetic field gradient at the three-dimensional distribution boundary of the magnetic source body. The magnetic source body orientation constraint is generated based on the principal axis direction of the magnetic source body and the boundary gradient direction.

4. The method according to claim 3, characterized in that, The step of extracting the tilt angle and boundary gradient direction and generating magnetic source body orientation constraints also includes: The multi-component magnetic measurement data is divided into depth layers based on the tilt angle to obtain the magnetic anomaly distribution of each depth layer. Multiple directional constraint vectors are generated based on the magnetic anomaly distribution of each depth segment, the principal axis direction of the magnetic source body, and the boundary gradient direction; The multiple directional constraint vectors constitute the directional constraint of the magnetic source body.

5. The method according to claim 1, characterized in that, The orientation correction and position registration include: Determine the anomaly center location, anomaly distribution range, and anomaly main axis direction corresponding to the gravity anomaly data and the transient electromagnetic anomaly data, respectively. The abnormal principal axis direction is corrected according to the magnetic source body direction constraint. Based on the directionally corrected main axis direction of the anomaly, the location of the anomaly center, and the anomaly distribution range, the gravity anomaly data and the transient electromagnetic anomaly data are registered in position.

6. The method according to claim 5, characterized in that, The orientation correction includes: Calculate the angle between the abnormal principal axis direction and the direction constraint of the magnetic source body; If the included angle is greater than a preset threshold, the abnormal main axis direction is rotated and adjusted with the abnormal center position as the rotation center so that the included angle between the abnormal main axis direction and the magnetic source body direction constraint is not greater than the preset threshold.

7. The method according to claim 5, characterized in that, The constructed target probability distribution includes: Based on the degree of spatial overlap and directional consistency between the registered gravity anomaly data, the registered transient electromagnetic anomaly data and the three-dimensional distribution of the magnetic source body, the contribution weights of the registered gravity anomaly data and the registered transient electromagnetic anomaly data are determined. The gravity anomaly data after location registration, the transient electromagnetic anomaly data after location registration, and the three-dimensional distribution of the magnetic source are weighted and fused according to the contribution weight to obtain the target probability distribution.

8. The method according to claim 7, characterized in that, The weighted fusion includes: The gravity anomaly data, transient electromagnetic anomaly data, and three-dimensional distribution of the magnetic source body after location registration are mapped to a unified spatial grid. The gravitational anomaly response value, transient electromagnetic anomaly response value, and magnetic source response value of each spatial grid cell are statistically analyzed, and the target probability value of each spatial grid cell is calculated in combination with the contribution weight. The target probability distribution of the blind ore body is generated based on the target probability value of each of the spatial grid cells.

9. The method according to claim 1, characterized in that, The process of determining the three-dimensional positioning results of the blind ore body and outputting the coordinates of the drilling target area and the drilling depth range includes: Kernel density estimation is performed on the target probability distribution to obtain candidate connected regions; the three-dimensional positioning result of the blind ore body is determined based on the matching result of the geometric morphology parameters of the candidate connected regions and the stratigraphic occurrence data. The coordinates of the drilling target area are output based on the spatial centroid of the three-dimensional positioning result of the blind ore body, and the drilling depth range is output based on the top and bottom boundary depths of the three-dimensional positioning result of the blind ore body.

10. A blind orebody positioning system based on multi-component magnetic survey data, used to implement the blind orebody positioning method based on multi-component magnetic survey data as described in any one of claims 1-9, characterized in that, The system includes: The data inversion module is used to acquire multi-component magnetic measurement data, gravity anomaly data, transient electromagnetic anomaly data and stratigraphic attitude data, and to perform constrained inversion on the multi-component magnetic measurement data based on the stratigraphic attitude data to obtain the three-dimensional distribution of the magnetic source body. The orientation constraint module is used to extract the tumble angle and boundary gradient direction based on the three-dimensional distribution of the magnetic source body, and generate orientation constraints for the magnetic source body. The correction and fusion module is used to perform orientation correction and position registration on the gravity anomaly data and the transient electromagnetic anomaly data based on the orientation constraint of the magnetic source body, and to construct the target probability distribution of the blind ore body in combination with the three-dimensional distribution of the magnetic source body. The positioning output module is used to determine the three-dimensional positioning result of the blind ore body based on the target probability distribution and the stratigraphic occurrence data, and to output the coordinates of the drilling target area and the drilling depth range based on the three-dimensional positioning result of the blind ore body.

Images