Artificial intelligence-based deep geothermal target selection system

Abstract

The present application relates to geothermal exploration technical field, specifically to a kind of deep geothermal target area selection system based on artificial intelligence;Actual direction vector is obtained according to the first principal component vector of exploration point in the neighborhood of exploration point, unit vector between neighborhood exploration point and the centroid of neighborhood;Direction consistency index is obtained according to the unit vector between neighborhood exploration point and the centroid and actual direction vector;From the centroid, heat source characteristic sequence is constructed along the direction of actual direction vector and the rate of change is obtained;According to the rate of change, distance feature between neighborhood exploration point and the centroid, theoretical heat source feature, heat source feature of neighborhood of neighborhood exploration point, gradual consistency index is obtained.The present application corrects clustering measurement distance according to direction consistency index and gradual consistency, and clusters exploration point, obtains exploration point cluster;The integrity of heat source region structure analysis and the accuracy of geothermal target area selection are improved.

Description

Technical Field

[0001] This invention relates to the field of geothermal exploration technology, specifically to an artificial intelligence-based deep geothermal target area selection system. Background Technology

[0002] Deep geothermal target areas refer to deep regions with geothermal resources, development potential, and clearly defined objectives. Exploration of these deep regions yields different exploration points, and the heat source characteristics at these points are analyzed and evaluated to determine suitable geothermal target areas for development. Existing technologies cluster the coordinates and multidimensional heat source characteristics of different exploration points to identify candidate regions with similar spatial heat source characteristics. These candidate regions are then evaluated to determine the final geothermal target area. In clustering tasks involving geothermal exploration point data with complex spatial distribution and unknown cluster numbers, the existing DBSCAN clustering algorithm has become the preferred algorithm due to its adaptability to clusters of arbitrary shapes and the fact that it does not require determining the number of clusters.

[0003] The DBSCAN clustering algorithm determines core points by using neighborhood radius and minimum neighborhood number to divide data points. In this scenario, deep heat flow zones are often distributed along deep fault zones, and the heat flow migration path follows the fault zone direction, expanding linearly or in strips. Some areas are far from the central region. However, DBSCAN determines core points based on circular neighborhoods, making it difficult to perceive the characteristics of geological bodies with extension. This leads to the heat source areas distributed along fault zones being mistakenly divided into multiple clusters, fragmenting the actual geothermal reservoir space, or being mistakenly identified as isolated points and discarded. Ultimately, this affects the completeness of the heat source area structure analysis and the accuracy of geothermal target area selection. Summary of the Invention

[0004] To address the technical problem mentioned above, where the linear or strip-like expansion of deep geothermal source regions can cause DBSCAN to divide the same geothermal source region into different clusters, affecting the accuracy of geothermal target area selection, the present invention aims to provide an artificial intelligence-based deep geothermal target area selection system. The specific technical solution adopted is as follows:

[0005] The data acquisition module is used to acquire the coordinates and heat source characteristics of the exploration point;

[0006] The location analysis module is used to obtain a first principal component vector based on the distribution characteristics of exploration points within the neighborhood of an exploration point; to obtain the actual direction vector of the neighborhood of the exploration point based on the first principal component vector and the unit vector between the exploration points within the neighborhood and the centroid of the neighborhood; and to obtain a direction consistency index based on the direction difference characteristics between the unit vector between the exploration points outside the neighborhood of the exploration point and the centroid and the actual direction vector.

[0007] The heat source feature analysis module is used to construct a heat source feature sequence of exploration points starting from the centroid along the direction of the actual direction vector; obtain the rate of change based on the change characteristics of the heat source feature sequence; obtain the theoretical heat source features of the centroid based on the actual direction vector and the exploration points in the neighborhood; and obtain the gradual consistency index based on the rate of change, the distance characteristics between the exploration points outside the neighborhood and the centroid, the theoretical heat source features, and the heat source features of the neighborhood of the exploration points outside the neighborhood.

[0008] The exploration point clustering module is used to obtain a trend index based on the directional consistency index and the gradual consistency index; to correct the clustering metric distance between the exploration points outside the neighborhood and the exploration points based on the trend index, thereby obtaining a corrected distance; and to cluster the exploration points based on the corrected distance to obtain an exploration point cluster.

[0009] Furthermore, the step of obtaining the first principal component vector based on the distribution characteristics of exploration points in the neighborhood of the exploration point includes:

[0010] The first principal component vector is obtained by using principal component analysis based on the coordinates of exploration points in the neighborhood of the exploration point.

[0011] Further, the step of obtaining the actual direction vector of the neighborhood of the exploration point based on the first principal component vector and the unit vector between the exploration point and the centroid of the neighborhood includes:

[0012] Calculate the angle between the unit vector between the exploration point and the centroid of the neighborhood and the first principal component vector to obtain a first angle value; calculate the cosine of the first angle value and perform a positive correlation mapping to obtain an angle weight; calculate the product of the unit vector between the exploration point and the centroid of the neighborhood and the angle weight to obtain a weighted unit vector of the exploration point in the neighborhood; calculate the sum of the weighted unit vectors in all neighborhoods of the exploration point to obtain the actual direction vector.

[0013] Further, the step of obtaining the directional consistency index based on the directional difference characteristics between the unit vector between the exploration point's neighboring exploration points and the centroid and the actual direction vector includes:

[0014] Calculate and normalize the cosine of the angle between the unit vector between the exploration point outside the neighborhood and the centroid and the actual direction vector to obtain the direction consistency index between the exploration point outside the neighborhood and the exploration point.

[0015] Further, the step of obtaining the rate of change based on the change characteristics of the heat source feature sequence includes:

[0016] The rate of change is obtained by calculating the average of the ratio of the absolute value of the difference in heat source characteristics between adjacent exploration points in the heat source characteristic sequence to the corresponding spatial distance.

[0017] Furthermore, the step of obtaining the theoretical heat source characteristics of the centroid based on the actual direction vector and the exploration points in the neighborhood includes:

[0018] Calculate the angle between the unit vector between the exploration point in the neighborhood and the centroid and the actual direction vector to obtain a second angle value; take the exploration point in the neighborhood corresponding to the minimum value of the second angle value as a reference exploration point; calculate the Euclidean distance between the reference exploration point and the centroid to obtain a reference distance; calculate the product of the reference distance and the rate of change to obtain a reference change; calculate the sum of the reference change and the heat source characteristics of the reference exploration point to obtain the theoretical heat source characteristics of the centroid.

[0019] Further, the step of obtaining the gradual consistency index based on the rate of change, the distance characteristics between the exploration point outside the neighborhood and the centroid, the theoretical heat source characteristics, and the heat source characteristics of the neighborhood of the exploration point outside the neighborhood includes:

[0020] In the formula, V represents the gradient consistency index between the exploration point outside the neighborhood and the exploration point, exp() represents the exponential function with the natural constant as the base, N represents the number of types of heat source characteristics, D represents the Euclidean distance between the exploration point outside the neighborhood and the exploration point, and H n D*H represents the rate of change of the characteristic of the nth heat source. n A represents the characteristic change of the nth heat source. n R represents the nth heat source characteristic of the exploration point outside the neighborhood. n Describes the nth theoretical heat source characteristic of the centroid, |D*H n +A n -R n | Indicates the feature gradient difference value; S represents the gradual change eigenvalue. n This represents the variance of the characteristics of the nth heat source within the neighborhood of the exploration point outside the neighborhood. This indicates the stability of the characteristic distribution of the heat source.

[0021] Further, the step of obtaining the trend index based on the directional consistency index and the gradual consistency index includes:

[0022] Calculate the average of the directional consistency index and the gradual consistency index to obtain the trend index between the exploration point outside the neighborhood and the exploration point.

[0023] Further, the step of correcting the clustering metric distance between the exploration point outside the neighborhood and the exploration point based on the trend index to obtain the corrected distance includes:

[0024] Calculate the difference between constant 1 and the trend index to obtain the adjustment coefficient; calculate the product of the clustering metric distance between the exploration point outside the neighborhood and the exploration point and the adjustment coefficient to obtain the corrected distance.

[0025] Furthermore, the step of clustering the exploration points according to the corrected distance to obtain a cluster of exploration points includes:

[0026] Clustering is performed using the DBSCAN clustering algorithm based on the coordinates and heat source characteristics of the exploration points. When other exploration points are not in the neighborhood of the exploration point, the corrected distance is used as the clustering metric distance between the other exploration points and the exploration point. After clustering is completed, different clusters of exploration points are obtained.

[0027] The present invention has the following beneficial effects:

[0028] In this invention, obtaining the first principal component vector enables analysis of the main trend direction and heat flow extension direction of exploration points within the neighborhood of an exploration point. Obtaining the actual direction vector reduces the error of the first principal component vector, improving the accuracy of determining the heat flow extension direction and exploration point clustering. Since heat flow exhibits extension characteristics along fault zones within the Earth's crust, obtaining the direction consistency index determines whether the extension direction of an exploration point is consistent with that of exploration points outside the neighborhood, thereby determining whether they are located in the same heat source region and improving the accuracy of exploration point clustering. Obtaining the rate of change determines the degree of change in heat source characteristics that conforms to the heat source extension trend, thereby determining whether exploration points conform to the gradual change of heat source characteristics corresponding to heat flow migration along fault zones. Obtaining the theoretical heat source characteristics determines the heat source characteristics at the centroid. Obtaining the gradual consistency index characterizes whether the change trend of heat source characteristics between an exploration point and exploration points outside the neighborhood conforms to the heat flow extension characteristics of the same heat source region. Obtaining trend indices can determine the likelihood that an exploration point and an adjacent exploration point are located in the same heat source region. Clustering based on the corrected distance yields exploration point clusters, which can group exploration points corresponding to strip-shaped heat source regions into a cluster in the actual deep underground scenario of heat source distribution. This improves the completeness of heat source region structure analysis and the accuracy of geothermal target area selection. Attached Figure Description

[0029] To more clearly illustrate the technical solutions and advantages 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a block diagram of an artificial intelligence-based deep geothermal target area selection system provided in one embodiment of the present invention. Detailed Implementation

[0031] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an artificial intelligence-based deep geothermal target area selection system proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0033] The following description, in conjunction with the accompanying drawings, details a specific scheme for an artificial intelligence-based deep geothermal target area selection system provided by the present invention.

[0034] Please see Figure 1 The diagram illustrates a block diagram of an artificial intelligence-based deep geothermal target area selection system according to an embodiment of the present invention. The system includes the following modules:

[0035] The data acquisition module S1 is used to acquire the coordinates and heat source characteristics of the exploration point.

[0036] In this invention, the implementation scenario involves selecting deep geothermal target areas to improve the accuracy of target area selection. Exploration is conducted in the exploration area to obtain the coordinates and heat source characteristics of exploration points at different locations. The coordinates represent the position of the exploration point in the underground three-dimensional space; the heat source characteristics represent the possibility that the exploration point is a geothermal reservoir area. In this embodiment, the heat source characteristics include temperature and thermal conductivity. These characteristics are obtained through in-situ measurements of exposed rock strata or profiles. To facilitate analysis, all heat source characteristics of the exploration points are standardized to eliminate dimensional differences. The implementer can determine the type of heat source characteristics collected from the exploration points according to the implementation scenario.

[0037] The location analysis module S2 is used to obtain the first principal component vector based on the distribution characteristics of exploration points in the neighborhood of the exploration point; to obtain the actual direction vector of the neighborhood of the exploration point based on the first principal component vector and the unit vector between the exploration point in the neighborhood and the centroid of the neighborhood; and to obtain the direction consistency index based on the direction difference characteristics between the unit vector between the exploration point outside the neighborhood and the centroid and the actual direction vector.

[0038] The existing DBSCAN algorithm, based on neighborhood radius, may fail to identify distant exploration points within a heat source region, resulting in low accuracy in clustering heat source regions. To improve the accuracy of heat source region selection, the selection of neighborhood radius during clustering needs to be improved. Deep heat source regions typically form near fault zones, and heat flow often extends along these fault zones. Artificially placed exploration points follow local geological patterns, and their distribution can initially reflect the directionality of local deep heat sources. Therefore, the extension trend of exploration points can be determined, and the extension region of exploration points characterizes the extension characteristics of the heat source region. Thus, a first principal component vector is obtained based on the distribution characteristics of exploration points within the neighborhood of an exploration point. Preferably, in this embodiment of the invention, the step of obtaining the first principal component vector includes: obtaining the first principal component vector using principal component analysis based on the coordinates of exploration points within the neighborhood of the exploration point. It should be noted that principal component analysis is an existing technology, and this algorithm can select the direction with the largest variance after data projection as the principal component; the direction of the first principal component vector characterizes the distribution extension trend of all exploration points in the neighborhood space of the exploration point. The neighborhood of the exploration point is the range formed by the neighborhood radius in the DBSCAN algorithm. In the implementation of this invention, taking a common geothermal exploration grid with a side length of 2km as an example, the neighborhood radius is set to 200 meters and the minimum number of neighborhoods is set to 8; the implementer can determine it according to the implementation scenario.

[0039] Furthermore, since the first principal component vector obtained by principal component analysis cannot distinguish whether it is consistently contributed by most exploration points or formed by the torsion of a few strongly influential points such as isolated points, in order to improve the accuracy of determining the direction of heat flow extension, it can be further determined based on the angle relationship between the exploration points in the neighborhood of the exploration point and the first principal component vector, thereby reducing the interference of isolated points on the first principal component vector. Therefore, the actual direction vector of the neighborhood of the exploration point is obtained based on the first principal component vector and the unit vector between the exploration point in the neighborhood and the centroid of the neighborhood. Preferably, in this embodiment of the invention, the step of obtaining the actual direction vector includes: calculating the vector angle between the unit vector between the exploration point in the neighborhood and the centroid of the neighborhood and the first principal component vector to obtain a first angle value; the centroid reflects the center position of all exploration points in the neighborhood of the exploration point, and the unit vector characterizes the positional relationship of the exploration point in the neighborhood relative to the neighborhood center. The magnitude of the first angle value characterizes the degree of consistency between an exploration point in any neighborhood and the main trend direction. A smaller first angle value indicates a more consistent direction, meaning the direction of the exploration point and its centroid closely aligns with the main trend direction of that neighborhood and is more consistent with the overall trend. The cosine of the first angle value is calculated and positively correlated to obtain the angle weight. A smaller first angle value corresponds to a larger cosine value, resulting in a larger angle weight. Consequently, exploration points in a neighborhood that are more consistent with the main trend direction are more likely to dominate the determination of the true direction and are assigned a higher weight. The product of the unit vector between the exploration point and the centroid of the neighborhood and the angle weight is calculated to obtain the weighted unit vector of the exploration point in that neighborhood. The sum of the weighted unit vectors in all neighborhoods of the exploration point is calculated to obtain the actual direction vector. This actual direction vector reflects the main distribution direction of exploration points in all neighborhoods of the exploration point and the direction of heat flow extension in the fault zone.

[0040] After obtaining the actual direction vector corresponding to the neighborhood of the exploration point, since deep heat flow regions usually form long extension zones with fault extension regions, even though these extension zones are spatially distant, if the extension direction is highly consistent with the original heat flow extension direction, then the extension zone may belong to the extension region of the heat flow region. Therefore, the difference between the directional characteristics of the exploration points outside the neighborhood of the exploration point and the actual direction vector of the centroid of the neighborhood of the exploration point can be analyzed to determine whether they belong to the same heat flow region. Thus, a directional consistency index is obtained based on the directional difference between the unit vector and the actual direction vector between the exploration points outside the neighborhood and the centroid of the exploration point. Preferably, in this embodiment of the invention, the step of obtaining the directional consistency index includes: calculating and normalizing the cosine value of the angle between the unit vector and the actual direction vector between the exploration points outside the neighborhood and the centroid to obtain the directional consistency index between the exploration points outside the neighborhood and the exploration point; the smaller the angle, the more consistent the directions, and the larger the cosine value, the larger the directional consistency index. Since heat flow regions are typically distributed along fault zones, and the direction of heat flow migration aligns with the fault zone's strike, extending linearly or in stripes, the smaller the angle between these vectors, the larger the directional consistency index. This indicates that the exploration point outside the vicinity is more likely to belong to the extended region of the exploration point itself, and thus more likely to belong to the same heat flow region. Conversely, the smaller the directional consistency index, the less likely the exploration point outside the vicinity belongs to the same heat flow region as the exploration point itself.

[0041] The heat source feature analysis module S3 is used to construct a heat source feature sequence of exploration points starting from the centroid along the direction of the actual direction vector; obtain the rate of change based on the change characteristics of the heat source feature sequence; obtain the theoretical heat source characteristics of the centroid based on the actual direction vector and exploration points in the neighborhood; and obtain the gradual consistency index based on the rate of change, the distance characteristics between exploration points outside the neighborhood and the centroid, the theoretical heat source characteristics, and the heat source characteristics of the neighborhood of exploration points outside the neighborhood.

[0042] In deep geothermal systems, the conduction and diffusion mechanism of heat energy in the Earth's crust involves diffusion from high-temperature regions to low-temperature regions, gradually weakening with distance. Especially in areas where heat flow extends along fault zones, these fault zones act as high-speed channels for heat transfer. Energy released from the central heat source propagates along the fault zones, causing the heat source characteristics of different exploration points along the fault zones to exhibit gradual changes. Simultaneously, influenced by the same geological structure, the heat source characteristics within the same region change relatively smoothly. Therefore, a heat source characteristic sequence for exploration points is first constructed starting from the centroid and along the actual direction vector. This sequence characterizes the changing trend of heat source characteristics from the centroid along the actual direction vector. Furthermore, the rate of change can be obtained based on the changing characteristics of the heat source feature sequence. Specifically, this includes: calculating the average of the ratios of the absolute values ​​of the differences in heat source features between adjacent exploration points in the heat source feature sequence to the corresponding spatial distances, thereby obtaining the rate of change of the heat source features. The spatial distance is the Euclidean distance between adjacent exploration points in space. This rate of change characterizes the normal degree of change of the heat source features in the direction of heat flow extension. The larger the rate of change, the greater the degree of change of the heat source features with distance, and the smaller the rate of change, the smaller the degree of change.

[0043] Furthermore, the rate of change can be used to analyze whether the degree of change in heat source characteristics from the exploration point outside the neighborhood conforms to the characteristics of the heat source extension trend. First, the theoretical heat source characteristics of the centroid are obtained based on the actual direction vector and the exploration points within the neighborhood. Since there may be no actual exploration point at the centroid, the heat source characteristics at the centroid are not collected. Therefore, in order to determine the heat source characteristics at the centroid for evaluating the changing trend of heat source characteristics of the exploration points outside the neighborhood, it is necessary to calculate the heat source characteristics based on the existing exploration points within the neighborhood. Preferably, in this embodiment of the invention, the step of obtaining the theoretical heat source characteristics includes: calculating the vector angle between the unit vector and the actual direction vector between the exploration point and the centroid in the neighborhood, and obtaining a second angle value; the smaller the second angle value, the closer the position of the exploration point is to the extension direction in the neighborhood, and the more the exploration point in the neighborhood conforms to the gradual change characteristics of the heat source characteristics on the fault zone; then, the exploration point in the neighborhood corresponding to the minimum value of the second angle value is taken as a reference exploration point; the Euclidean distance between the reference exploration point and the centroid is calculated to obtain a reference distance; the product of the reference distance and the rate of change is calculated to obtain a reference change; the reference change is the change in heat source characteristics from the reference exploration point to the centroid under normal gradual change characteristics. The sum of the reference change and the heat source characteristics of the reference exploration point is calculated to obtain the theoretical heat source characteristics of the centroid; the theoretical heat source characteristics characterize the degree to which the centroid conforms to the gradual change of normal heat source characteristics; it should be noted that if there is a real exploration point at the centroid, the real heat source characteristics are used as the theoretical heat source characteristics.

[0044] Furthermore, after obtaining the theoretical heat source characteristics at the centroid, the gradual consistency index can be obtained based on the rate of change, the distance characteristics between the exploration points outside the neighborhood and the centroid, the theoretical heat source characteristics, and the heat source characteristics of the neighborhood of the exploration points outside the neighborhood; preferably, in this embodiment of the invention, the step of obtaining the gradual consistency index includes:

[0045]

[0046] In the formula, V represents the gradient consistency index between the exploration point outside the neighborhood and the exploration point, exp() represents the exponential function with the natural constant as the base, used for negative correlation mapping, N represents the number of heat source characteristics, D represents the Euclidean distance between the exploration point outside the neighborhood and the exploration point, and H n D*H represents the rate of change of the characteristic of the nth heat source. n A represents the characteristic change of the nth heat source, that is, the change of heat source characteristic from the exploration point outside the neighborhood at the extension direction to the centroid inside the neighborhood. n R represents the characteristic of the nth heat source at an exploration point outside the neighborhood. n This represents the characteristic of the nth theoretical heat source at the center of mass. |D*H n +A n -R n | represents the feature gradient difference value, if A n With D*H n The closer the sum of values ​​is to the theoretical heat source characteristics at the centroid, the smaller the characteristic gradient difference value, which means that the trend of heat source characteristic change from the centroid to the exploration point outside the neighborhood is more consistent with the gradual trend of heat source characteristics caused by the extension of heat flow on the fault zone, and the more likely the exploration point outside the neighborhood belongs to the same heat source region. This represents a gradual change in characteristic value; the larger the gradual change in characteristic value, the more likely that the exploration points in that neighborhood belong to the same heat source region. n This represents the variance of the nth heat source characteristic within the neighborhood of the exploration point outside the neighborhood. The smaller the variance, the more stable the heat source characteristics in the neighborhood of the exploration point outside the neighborhood, and the more likely they belong to the influence of the same geological structure. This indicates the stability of the heat source characteristic distribution. The greater the stability of the heat source characteristic distribution, the more likely the exploration point outside the neighborhood is located in the direction of heat flow extension. Furthermore, the greater the gradient consistency index, the more likely the exploration point outside the neighborhood belongs to the same direction of heat flow extension as the exploration point itself, and the more similar the heat source regions they represent.

[0047] The exploration point clustering module S4 is used to obtain a trend index based on the directional consistency index and the gradual consistency index; to correct the clustering metric distance between exploration points outside the neighborhood based on the trend index, and to obtain a corrected distance; and to cluster the exploration points based on the corrected distance to obtain exploration point clusters.

[0048] After obtaining the directional consistency index and gradual consistency index of all neighboring exploration points of the exploration point, a trend index can be obtained based on the directional consistency index and gradual consistency index. Preferably, in this embodiment of the invention, the step of obtaining the trend index includes: calculating the average of the directional consistency index and the gradual consistency index to obtain the trend index between the neighboring exploration points and the exploration point. The larger the trend index, the more similar the heat flow extension trend and heat source change trend between the neighboring exploration points and the exploration point, and the more likely they are to characterize the features of the same heat source region. Even if the spatial interval is far, they should be clustered into the same cluster.

[0049] Furthermore, the clustering metric distance between the exploration points outside the neighborhood and the exploration points can be corrected based on the trend index to obtain the corrected distance. Preferably, in this embodiment of the invention, the step of obtaining the corrected distance includes: calculating the difference between the constant 1 and the trend index to obtain the adjustment coefficient; the smaller the adjustment coefficient, the more similar the heat source region characteristics represented by the exploration point outside the neighborhood is to the heat source region characteristics represented by the exploration point. The corrected distance is obtained by multiplying the clustering metric distance between the exploration points outside the neighborhood and the exploration point by the adjustment coefficient; if the heat source region characteristics represented by the exploration point outside the neighborhood are more similar to the heat source region characteristics represented by the exploration point, the corrected distance is smaller than the corresponding clustering metric distance, thereby avoiding splitting the two during the clustering process. It should be noted that the clustering metric distance is obtained by weighted summation of the spatial distance between exploration points and the difference distance of heat source characteristics, and is used by DBSCAN to cluster exploration points; the specific steps are not elaborated here. The smaller the clustering metric distance, the more similar the location characteristics and heat source characteristics represented by the two exploration points are.

[0050] After obtaining the corrected distance between each exploration point and its corresponding neighboring exploration points, the exploration points can be clustered based on the corrected distance to obtain exploration point clusters. Specifically, this involves clustering exploration points using the DBSCAN clustering algorithm based on their coordinates and heat source characteristics. When other exploration points are not within the neighborhood of a given exploration point, the corresponding corrected distance is used as the clustering metric between that other exploration point and the given exploration point. After clustering, different exploration point clusters are obtained. This clustering method can group exploration points that are far apart but belong to the same heat source region into the same cluster in actual geothermal exploration scenarios, thus avoiding the splitting of the same heat source region. Finally, geothermal target areas can be selected based on different exploration point clusters, improving the completeness of heat source region structure analysis and the accuracy of geothermal target area selection.

[0051] In summary, this invention provides an artificial intelligence-based deep geothermal target area selection system. It obtains the actual direction vector based on the first principal component vector of exploration points within the neighborhood of an exploration point and the unit vector between the exploration point and the centroid of the neighborhood. It obtains a directional consistency index based on the unit vector between the exploration point outside the neighborhood and the centroid, and the actual direction vector. Starting from the centroid, it constructs a heat source feature sequence along the direction of the actual direction vector and obtains the rate of change. It obtains a gradual consistency index based on the rate of change, the distance characteristics between the exploration point outside the neighborhood and the centroid, theoretical heat source characteristics, and heat source characteristics of the neighborhood of the exploration point outside the neighborhood. This invention corrects the clustering metric distance based on the directional consistency index and gradual consistency, and clusters exploration points to obtain exploration point clusters; thus improving the completeness of heat source region structure analysis and the accuracy of geothermal target area selection.

[0052] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0053] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A deep geothermal target area selection system based on artificial intelligence, characterized in that, The system includes the following modules: The data acquisition module is used to acquire the coordinates and heat source characteristics of the exploration point; The location analysis module is used to obtain a first principal component vector based on the distribution characteristics of exploration points within the neighborhood of an exploration point; to obtain the actual direction vector of the neighborhood of the exploration point based on the first principal component vector and the unit vector between the exploration points within the neighborhood and the centroid of the neighborhood; and to obtain a direction consistency index based on the direction difference characteristics between the unit vector between the exploration points outside the neighborhood of the exploration point and the centroid and the actual direction vector. The heat source feature analysis module is used to construct a heat source feature sequence of the exploration point starting from the centroid and along the direction of the actual direction vector; and to obtain the rate of change based on the changing characteristics of the heat source feature sequence. The theoretical heat source characteristics of the centroid are obtained based on the actual direction vector and the exploration points in the neighborhood; the gradual consistency index is obtained based on the rate of change, the distance characteristics between the exploration points outside the neighborhood and the centroid, the theoretical heat source characteristics, and the heat source characteristics of the neighborhood of the exploration points outside the neighborhood. The exploration point clustering module is used to obtain a trend index based on the directional consistency index and the gradual consistency index; to correct the clustering metric distance between the exploration points outside the neighborhood and the exploration points based on the trend index, thereby obtaining a corrected distance; and to cluster the exploration points based on the corrected distance to obtain an exploration point cluster.

2. The deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of obtaining the first principal component vector based on the distribution characteristics of exploration points in the neighborhood of an exploration point includes: The first principal component vector is obtained by using principal component analysis based on the coordinates of exploration points in the neighborhood of the exploration point.

3. The deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of obtaining the actual direction vector of the neighborhood of the exploration point based on the first principal component vector and the unit vector between the exploration point and the centroid of the neighborhood includes: Calculate the angle between the unit vector between the exploration point and the centroid of the neighborhood and the first principal component vector to obtain a first angle value; calculate the cosine of the first angle value and perform a positive correlation mapping to obtain an angle weight; calculate the product of the unit vector between the exploration point and the centroid of the neighborhood and the angle weight to obtain a weighted unit vector of the exploration point in the neighborhood; calculate the sum of the weighted unit vectors in all neighborhoods of the exploration point to obtain the actual direction vector.

4. The deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of obtaining the directional consistency index based on the directional difference characteristics between the unit vector and the actual direction vector between the neighboring exploration points and the centroid of the exploration point includes: Calculate and normalize the cosine of the angle between the unit vector between the exploration point outside the neighborhood and the centroid and the actual direction vector to obtain the direction consistency index between the exploration point outside the neighborhood and the exploration point.

5. The deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of obtaining the rate of change based on the change characteristics of the heat source feature sequence includes: The rate of change is obtained by calculating the average of the ratio of the absolute value of the difference in heat source characteristics between adjacent exploration points in the heat source characteristic sequence to the corresponding spatial distance.

6. The deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of obtaining the theoretical heat source characteristics of the centroid based on the actual direction vector and the exploration points in the neighborhood includes: Calculate the angle between the unit vector between the exploration point in the neighborhood and the centroid and the actual direction vector to obtain a second angle value; take the exploration point in the neighborhood corresponding to the minimum value of the second angle value as a reference exploration point; calculate the Euclidean distance between the reference exploration point and the centroid to obtain a reference distance; calculate the product of the reference distance and the rate of change to obtain a reference change; calculate the sum of the reference change and the heat source characteristics of the reference exploration point to obtain the theoretical heat source characteristics of the centroid.

7. The deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of obtaining the gradual consistency index based on the rate of change, the distance characteristics between the exploration point outside the neighborhood and the centroid, the theoretical heat source characteristics, and the heat source characteristics of the neighborhood of the exploration point outside the neighborhood includes: In the formula, V represents the gradient consistency index between the exploration point outside the neighborhood and the exploration point, exp() represents the exponential function with the natural constant as the base, N represents the number of types of heat source characteristics, D represents the Euclidean distance between the exploration point outside the neighborhood and the exploration point, and H n D*H represents the rate of change of the characteristic of the nth heat source. n A represents the characteristic change of the nth heat source. n R represents the nth heat source characteristic of the exploration point outside the neighborhood. n Describes the nth theoretical heat source characteristic of the centroid, |D*H n +A n -R n | Indicates the feature gradient difference value; S represents the gradual change eigenvalue. n This represents the variance of the characteristics of the nth heat source within the neighborhood of the exploration point outside the neighborhood. This indicates the stability of the characteristic distribution of the heat source.

8. The deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of obtaining the trend index based on the directional consistency index and the gradual consistency index includes: Calculate the average of the directional consistency index and the gradual consistency index to obtain the trend index between the exploration point outside the neighborhood and the exploration point.

9. The deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of correcting the clustering metric distance between the exploration point outside the neighborhood and the exploration point based on the trend index to obtain the corrected distance includes: Calculate the difference between constant 1 and the trend index to obtain the adjustment coefficient; calculate the product of the clustering metric distance between the exploration point outside the neighborhood and the exploration point and the adjustment coefficient to obtain the corrected distance.

10. A deep geothermal target area selection system based on artificial intelligence according to claim 1, characterized in that, The step of clustering the exploration points according to the modified distance to obtain exploration point clusters includes: Clustering is performed using the DBSCAN clustering algorithm based on the coordinates and heat source characteristics of the exploration points. When other exploration points are not in the neighborhood of the exploration point, the corrected distance is used as the clustering metric distance between the other exploration points and the exploration point. After clustering is completed, different clusters of exploration points are obtained.

Images