A method for iron ore prediction based on magnetic gradient profile
Patent Information
- Application Number
- CN202610912142.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-23
- Publication Date
- 2026-09-04
AI Technical Summary
但目前仍缺乏系统化的、基于剖面磁梯度的铁矿预测方法，尤其是在复杂地形及多剖面数据下的稳健处理技术，难以充分发挥高精度航磁数据在铁矿勘查中的潜力
针对传统基于磁异常强度的解释方法对弱异常及叠加异常铁矿体分辨能力不足，以及网格化处理中因插值平滑效应导致微弱异常信息丢失等技术瓶颈，本发明提出一种基于剖面磁梯度的铁矿预测方法。该方法充分利用磁梯度对局部场源响应敏感、原始剖面数据可完整保留真实地球物理场信息的优势，以剖面磁梯度作为核心特征，有效响应铁矿体引起的局部磁异常。通过引入KD-tree高效空间索引结合动态容差匹配策略，实现多测线剖面数据的同源点高精度空间关联，构建多剖面协同的磁梯度特征集，从而实现对弱磁异常铁矿体的高精度预测。本发明实现了从剖面磁梯度计算、多源数据空间集成到铁矿体精准预测的全流程自动化处理，显著提升了对弱异常铁矿体的识别灵敏度与预测可靠性，有效克服了传统方法的信息损失与分辨能力不足问题，为复杂地质条件下的铁矿勘查提供了有力的技术支撑。此外，本方法系统性强，可无缝嵌入现有基于航磁数据的铁矿预测流程，适用于大面积高精度航磁数据的快速矿产预测，具有良好的实用性与推广价值。
Smart Images
Figure CN122690701A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of aeromagnetic data interpretation technology, and more specifically to a method for predicting iron ore deposits based on profile magnetic gradients. Background Technology
[0002] Iron ore deposits are often accompanied by significant magnetic anomalies, and aeromagnetic surveys, with their advantages of high efficiency and wide coverage, have become one of the core methods for iron ore resource prediction. By analyzing the characteristics of magnetic anomaly intensity and magnetic gradient changes in aeromagnetic data, the distribution of underground magnetic geological bodies can be effectively inferred, providing geophysical basis for the selection of iron ore target areas. In recent years, with the rapid development of high-precision aeromagnetic detection technology, the resolution of magnetic field data has been significantly improved. How to automatically and accurately identify mineral-induced anomalies from massive amounts of aeromagnetic data has become a research hotspot in this field.
[0003] However, most existing methods interpret aeromagnetic anomalies directly, which is insufficient for distinguishing weak or superimposed anomalies. Magnetic gradients, as the rate of change of a magnetic field in space, are more sensitive to local field source responses and can more effectively highlight the boundary information of shallow or small-scale iron ore bodies, but a systematic interpretation framework has not yet been established. Furthermore, traditional interpretations often rely on gridded data, and the interpolation process inevitably introduces a smoothing effect, leading to the attenuation of high-frequency information in the original profile data and the masking of local weak anomalies, thus blurring the boundaries of magnetic bodies and reducing the predictive ability for concealed iron ore bodies. In contrast, profile data is collected along the survey line, fully preserving the true variation characteristics of the geophysical field, making it suitable for magnetic gradient calculation and iron ore prediction. However, a systematic iron ore prediction method based on profile magnetic gradients is still lacking, especially robust processing techniques for complex terrain and multi-profile data, making it difficult to fully realize the potential of high-precision aeromagnetic data in iron ore exploration.
[0004] Therefore, providing an iron ore prediction method based on profile magnetic gradient to overcome the insufficient ability of traditional aeromagnetic interpretation to identify weak or superimposed anomalies in iron ore bodies is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention
[0005] In view of this, the present invention provides an iron ore prediction method based on profile magnetic gradient. Based on the original profile data, the method achieves high-precision spatial registration of multi-profile data through dynamic tolerance matching, calculates the profile magnetic gradient to highlight anomalies in shallow and small-scale iron ore bodies, and thus achieves accurate prediction of iron ore bodies.
[0006] To achieve the above objectives, the present invention adopts the following technical solution: A method for predicting iron ore deposits based on profile magnetic gradients, comprising: Step 1: Obtain the raw aeromagnetic profile data of the target area and perform preprocessing; Step 2: Perform survey line data segmentation on the preprocessed raw aeromagnetic profile data to obtain multiple ordered survey line datasets; Step 3: Perform dynamic tolerance matching of source sampling points for each set of survey line datasets to obtain a set of source sampling point pairs; Step 4: Extract the magnetic anomaly value and height of each pair of sampling points from the same source in each survey line dataset and calculate the corresponding magnetic gradient value to obtain the profile magnetic gradient dataset of the target area; Step 5: Based on the profile magnetic gradient dataset of the target area, with high magnetic gradient data as the core indicator, filter the iron ore magnetic gradient data and delineate the iron ore prediction target area.
[0007] Preferably, the original aeromagnetic profile data includes two sets of parallel survey lines at different flight altitudes, each survey line including position coordinates, altitude value, and magnetic anomaly value.
[0008] Preferably, the survey line data segmentation involves grouping two sets of survey line data with the same survey line number into one group, and sorting all groups of survey line data according to spatial coordinates to obtain multiple ordered survey line datasets.
[0009] Preferably, the dynamic tolerance matching of the same-source sampling points adopts the KD-tree algorithm, with the tolerance distance as the threshold, specifically including: The eastward and northward distances of the two sets of survey lines in each survey line dataset are used as spatial coordinates to construct a spatial coordinate matrix of the sampling points; A spatial index tree for survey line data B is constructed based on the KD-tree algorithm. For each sampling point in survey line data A, the nearest sampling point in survey line data B is queried, and the matching distance and index are recorded. A dynamic tolerance adjustment strategy is adopted to determine the relationship between the current number of matching points and the number of sampling points in the two survey lines. The tolerance is gradually increased by a fixed step size until the number of matching points reaches the target value. If the target is not met even after reaching the maximum tolerance, the current maximum set of matching points is used. The matching point set is deduplicated to ensure that each sampling point in the survey line data B is matched only once, thus obtaining a set of sampling point pairs from the same source.
[0010] Preferably, the formula for calculating the magnetic gradient value in step 4 is as follows:
[0011] in, and These are the magnetic anomaly and altitude of the low-altitude flight survey line data in the same sampling point pair. and It represents the magnetic anomaly and altitude of the high-altitude flight survey line data in the same sampling point pair, where K is a constant.
[0012] As can be seen from the above technical solution, compared with the prior art, the present invention provides an iron ore prediction method based on profile magnetic gradient, with the following beneficial effects: To address the limitations of traditional magnetic anomaly intensity-based interpretation methods in distinguishing weak and superimposed anomalies in iron ore bodies, and the loss of subtle anomaly information due to interpolation smoothing effects during gridding, this invention proposes an iron ore prediction method based on profile magnetic gradients. This method fully leverages the advantages of magnetic gradients' sensitivity to local source responses and the ability of original profile data to fully preserve true geophysical field information. Using profile magnetic gradients as the core feature, it effectively responds to local magnetic anomalies caused by iron ore bodies. By introducing efficient KD-tree spatial indexing combined with a dynamic tolerance matching strategy, high-precision spatial correlation of source points from multiple survey line profiles is achieved, constructing a multi-profile collaborative magnetic gradient feature set, thereby enabling high-precision prediction of weak magnetic anomaly iron ore bodies. This invention achieves fully automated processing from profile magnetic gradient calculation and multi-source data spatial integration to accurate iron ore body prediction, significantly improving the sensitivity and reliability of identifying weak anomaly iron ore bodies. It effectively overcomes the information loss and insufficient resolution problems of traditional methods, providing strong technical support for iron ore exploration under complex geological conditions. Furthermore, this method is highly systematic and can be seamlessly integrated into existing iron ore prediction workflows based on aeromagnetic data. It is suitable for rapid mineral prediction using large-area, high-precision aeromagnetic data and has good practicality and promotion value. Attached Figure Description
[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0014] Figure 1 This is a schematic diagram of the method flow provided by the present invention.
[0015] Figure 2 This is a magnetic anomaly difference map provided in an embodiment of the present invention.
[0016] Figure 3 The magnetic gradient map provided in the embodiments of the present invention. Detailed Implementation
[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0018] like Figure 1 As shown in the figure, this invention discloses a method for predicting iron ore based on profile magnetic gradient, comprising: Step 1: Obtain the raw aeromagnetic profile data of the target area and perform preprocessing; Step 2: Perform survey line data segmentation on the preprocessed raw aeromagnetic profile data to obtain multiple ordered survey line datasets; Step 3: Perform dynamic tolerance matching of source sampling points for each set of survey line datasets to obtain a set of source sampling point pairs; Step 4: Extract the magnetic anomaly value and height of each pair of sampling points from the same source in each survey line dataset and calculate the corresponding magnetic gradient value to obtain the profile magnetic gradient dataset of the target area; Step 5: Based on the profile magnetic gradient dataset of the target area, with high magnetic gradient data as the core indicator, filter the iron ore magnetic gradient data and delineate the iron ore prediction target area.
[0019] Specifically, the original aeromagnetic profile data includes two sets of parallel survey lines at different flight altitudes. Each survey line includes position coordinates (eastward distance, northward distance), altitude value, and magnetic anomaly value.
[0020] Specifically, the survey line data segmentation involves grouping two sets of survey line data with the same survey line number into one group, and sorting all groups of survey line data according to spatial coordinates to ensure the spatial continuity of the data, thereby obtaining multiple ordered survey line datasets.
[0021] Specifically, the dynamic tolerance matching of the same-source sampling points adopts the KD-tree algorithm, using the tolerance distance as a threshold, to automatically search and match the data points with the closest spatial coordinates on the two sets of survey line data in each group, ensuring that the data between different profiles are strictly aligned in space, specifically including: The eastward and northward distances of the two sets of survey lines in each survey line dataset are used as spatial coordinates to construct a spatial coordinate matrix of the sampling points; A spatial index tree for survey line data B is constructed based on the KD-tree algorithm. For each sampling point in survey line data A, the nearest sampling point in survey line data B is queried, and the matching distance and index are recorded. A dynamic tolerance adjustment strategy is adopted to determine the relationship between the current number of matching points and the number of sampling points in the two survey lines. The tolerance is gradually increased in fixed step sizes until the number of matching points reaches the target value. If the target is not met even after reaching the maximum tolerance, the current maximum set of matching points is used. For example, with an initial tolerance of 2.5m as the baseline, if the current number of matching points is less than the number of sampling points in the two survey lines with fewer points, the tolerance is gradually increased in step sizes of 0.5m (maximum tolerance not exceeding 20m) until the number of matching points reaches the target value. If the target is not met even after reaching the maximum tolerance, the current maximum set of matching points is used. The initial tolerance and step size parameters can be set manually. The obtained set of matching points is deduplicated to ensure that each sampling point in the survey line data B is matched only once, thus obtaining a set of sampling point pairs from the same source.
[0022] In a specific embodiment of the present invention, the KD-tree is a nearest neighbor search data structure oriented towards high-dimensional space, the core of which lies in the use of recursive search and pruning strategies. The algorithm starts from the root node, compares the coordinates of the query point based on the partition dimension to determine the search path, and dynamically maintains the current nearest neighbor and the minimum distance during the process. In the backtracking phase, pruning is performed by calculating the vertical distance from the query point to the partitioning hyperplane: if this distance is less than the current minimum distance, the other subtree needs to be searched; otherwise, the branch is directly pruned. This spatial partitioning-based pruning mechanism significantly reduces the number of node visits and has significant efficiency.
[0023] Specifically, the formula for calculating the magnetic gradient value in step 4 is as follows:
[0024] in, and These are the magnetic anomaly and altitude of the low-altitude flight survey line data in the same sampling point pair. and These are the magnetic anomalies and altitudes of high-altitude flight survey lines in a pair of sampling points from the same source, where K is a constant, typically taken as... The average value is used to avoid gradient interference caused by excessively small altitude differences between two flights due to terrain such as mountains during actual flight measurements. Figures 2-3 As shown, the obtained high magnetic gradient and the magnetic anomaly difference in the formula are plotted. Figure 2 Represents poor magnetic anomaly ; Figure 3 This is the final magnetic gradient diagram. The low-altitude and high-altitude flights refer to two flights over the same area: one at low altitude and one at high altitude.
[0025] In a specific embodiment of the present invention, step 5 is as follows: Step 5.1: Calculate the magnetic gradient amplitude and statistically analyze its distribution characteristics. Based on the magnetic gradient values of each pair of sampling points from the same source obtained in Step 4, calculate the mean and standard deviation of the magnetic gradient values for the entire region, and statistically analyze the characteristics of the magnetic gradient across the entire region.
[0026] Step 5.2: Set a high magnetic gradient anomaly screening threshold. Magnetic gradients can effectively highlight the boundary information of shallow or small-scale iron ore bodies. Therefore, a high magnetic gradient anomaly screening threshold is set to guide mineral exploration.
[0027] Step 5.3: Delineate the iron ore prediction target area. Based on the high magnetic gradient anomaly screening threshold, the outer envelope of the selected high magnetic gradient anomaly segments is used as the target area boundary. Specifically, for each continuous anomaly segment on a survey line, the planar coordinates of the starting and ending points are taken, and the corresponding anomaly endpoints on adjacent survey lines are connected to form a polygonal target area. Within the target area, the most favorable ore-forming center is further identified based on the peak position of the magnetic gradient amplitude.
[0028] Step 5.4: Conduct auxiliary verification in conjunction with geological background. Overlay analysis is performed on the geological map of the delineated target area and the distribution of known faults or rock masses in the study area. Target areas that are obviously located within non-magnetic overburden or known non-mineral anomalies (such as volcanic rocks) are eliminated, and target areas that match the metallogenic geological conditions are retained as the final prediction results.
[0029] In summary, this invention provides an iron ore prediction method based on profile magnetic gradients to address the shortcomings of traditional methods in distinguishing weak anomalies and the loss of information due to gridding. By combining KD-tree with dynamic tolerance to achieve high-precision spatial matching of multiple survey lines, a multi-profile collaborative magnetic gradient feature body is constructed, effectively improving the sensitivity and prediction accuracy of identifying weak anomaly iron ore bodies, thus providing technical support for iron ore exploration.
[0030] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.
[0031] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for predicting iron ore deposits based on profile magnetic gradient, characterized in that, include: Step 1: Obtain the raw aeromagnetic profile data of the target area and perform preprocessing; Step 2: Perform survey line data segmentation on the preprocessed raw aeromagnetic profile data to obtain multiple ordered survey line datasets; Step 3: Perform dynamic tolerance matching of source sampling points for each set of survey line datasets to obtain a set of source sampling point pairs; Step 4: Extract the magnetic anomaly value and height of each pair of sampling points from the same source in each survey line dataset and calculate the corresponding magnetic gradient value to obtain the profile magnetic gradient dataset of the target area; Step 5: Based on the profile magnetic gradient dataset of the target area, with high magnetic gradient data as the core indicator, filter the iron ore magnetic gradient data and delineate the iron ore prediction target area.
2. The iron ore prediction method based on profile magnetic gradient according to claim 1, characterized in that, The original aeromagnetic profile data includes two sets of parallel survey lines at different flight altitudes. Each survey line includes position coordinates, altitude values, and magnetic anomaly values.
3. The iron ore prediction method based on profile magnetic gradient according to claim 1, characterized in that, The survey line data segmentation involves grouping two sets of survey line data with the same survey line number into one group, and sorting all the survey line data in all groups according to spatial coordinates to obtain multiple ordered survey line datasets.
4. The iron ore prediction method based on profile magnetic gradient according to claim 1, characterized in that, The dynamic tolerance matching of the same sampling points adopts the KD-tree algorithm, with the tolerance distance as the threshold, specifically including: The eastward and northward distances of the two sets of survey lines in each survey line dataset are used as spatial coordinates to construct a spatial coordinate matrix of the sampling points; A spatial index tree for survey line data B is constructed based on the KD-tree algorithm. For each sampling point in survey line data A, the nearest sampling point in survey line data B is queried, and the matching distance and index are recorded. A dynamic tolerance adjustment strategy is adopted to determine the relationship between the current number of matching points and the number of sampling points in the two survey lines. The tolerance is gradually increased by a fixed step size until the number of matching points reaches the target value. If the target is not met even after reaching the maximum tolerance, the current maximum set of matching points is used. The matching point set is deduplicated to ensure that each sampling point in the survey line data B is matched only once, thus obtaining a set of sampling point pairs from the same source.
5. The iron ore prediction method based on profile magnetic gradient according to claim 1, characterized in that, The formula for calculating the magnetic gradient value in step 4 is as follows: in, and These are the magnetic anomaly and altitude of the low-altitude flight survey line data in the same sampling point pair. and It represents the magnetic anomaly and altitude of the high-altitude flight survey line data in the same sampling point pair, where K is a constant.