Virtual drilling based resource exploration method and apparatus

Abstract

The embodiment of the application provides a resource exploration method and device based on virtual drilling, which comprises the following steps: discretizing a target area into a regular grid model and associating historical exploration data with the regular grid model by acquiring the range and the historical exploration data of the target area; performing geological attribute dependency calculation in the regular grid model according to the geological attribute information between each known grid, determining a corresponding dependency value, performing weighted calculation on the geological attribute information of an unknown grid in the regular grid model according to the dependency value, determining a corresponding initial belief graph, constructing a Markov decision process model according to the initial belief graph, inputting the initial belief graph and the Markov decision process model into a set Monte Carlo tree search algorithm to perform virtual drilling simulation, and determining a corresponding recommended drilling scheme. The application can improve the efficiency and accuracy of real drilling.

Description

Technical Field

[0001] This application relates to the field of data processing, specifically to a resource exploration method and apparatus based on virtual drilling. Background Technology

[0002] In the field of mineral resource exploration, traditional methods often rely on the experience of geologists and limited prior information, especially in scenarios involving concealed ore bodies in covered areas and deep prospecting (depths exceeding 100 meters), which face multiple challenges such as insufficient information, high costs, and low efficiency.

[0003] Traditional exploration methods primarily rely on geological mapping, geophysical exploration, and limited borehole verification. While these methods can guide mineral exploration to some extent, they have the following limitations when dealing with complex geological conditions and deep exploration:

[0004] Incomplete information acquisition: The information provided by geological mapping and geophysical exploration is often indirect and cannot directly reflect the spatial distribution and grade changes of ore bodies.

[0005] High degree of subjectivity in decision-making: The decision-making process based on experience and prior knowledge is easily affected by human factors, leading to deviation from the exploration target.

[0006] Inefficient: Traditional methods often require a large number of boreholes for verification, resulting in long exploration cycles and high costs.

[0007] Difficulty in handling deep mineral exploration: Deep mineral exploration requires higher drilling accuracy and more effective decision support, which traditional methods cannot meet.

[0008] To overcome the limitations of traditional exploration methods, there is an urgent need for a resource exploration method based on virtual drilling that can improve the efficiency and accuracy of real drilling. Summary of the Invention

[0009] To address the problems in the existing technology, this application provides a resource exploration method and apparatus based on virtual drilling, which can improve the efficiency and accuracy of real drilling.

[0010] To solve at least one of the above problems, this application provides the following technical solution:

[0011] In a first aspect, this application provides a resource exploration method based on virtual drilling, comprising:

[0012] Historical exploration data of the target area is obtained, the target area is discretized into a regular grid model, and the historical exploration data is associated with the regular grid model. The historical exploration data includes borehole data and geological attribute information of the target resource. The borehole data includes borehole location and borehole results.

[0013] In the regular grid model, the geological attribute dependency is calculated based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted based on the dependency value to determine the corresponding initial belief map. The initial belief map represents the prior probability or attribute strength of the existence of the target resource in each grid.

[0014] The initial belief map is used to determine the corresponding state space, the candidate borehole location set in the regular grid model is used to determine the corresponding action space, the drilling results are used to determine the corresponding observation space, the cost-effectiveness of drilling is used to determine the corresponding reward function, and the state space, action space, observation space and reward function are used to determine the corresponding Markov decision process model. The initial belief map and the Markov decision process model are input into a set Monte Carlo tree search algorithm to perform virtual drilling simulation and determine the corresponding recommended drilling layout scheme.

[0015] Further, the step of acquiring historical exploration data of the target area, discretizing the target area into a regular grid model, and associating the historical exploration data with the regular grid model includes:

[0016] Acquire historical exploration data for the target area, including borehole data and geological attribute information of the target resources, wherein the borehole data includes borehole location and borehole results indicating whether minerals were encountered or not;

[0017] The target region is discretized into a regular grid model at a preset resolution;

[0018] The borehole location is mapped to the corresponding grid cell in the regular grid model, and the borehole results and geological attribute information of the grid cell are marked as known data.

[0019] Further, the step of mapping the borehole location to the corresponding grid cell in the regular grid model, and marking the borehole results and geological attribute information of that grid cell as known data, includes:

[0020] Construct a two-dimensional matrix with the same dimensions as the regular grid model;

[0021] For each grid cell, if it contains the borehole location, then fill in the borehole results and geological attribute information at the corresponding position in the matrix;

[0022] If the matrix does not contain the borehole location, a specific null value identifier is filled in the corresponding position of the matrix to identify the location as an unexplored area.

[0023] Further, in the regular grid model, the calculation of geological attribute dependencies based on the geological attribute information between each known grid, and the determination of the corresponding dependency values, includes:

[0024] In the regular grid model, all known grid point pairs are traversed, the spatial distance between each pair of grid points and the semivariance of the geological attribute values ​​are calculated, and the corresponding geological attribute scatter plot is determined.

[0025] The geological attribute scatter plot is fitted according to the model in the preset variogram model library. Based on the fitted variogram model, the dependency value between each grid point pair is determined. The dependency value is used to characterize the spatial dependency of geological attributes between grids.

[0026] Furthermore, the step of calculating the geological attribute dependency based on the geological attribute information between known grids in the regular grid model to determine the corresponding dependency value also includes:

[0027] In the regular grid model, based on the prior distribution characteristics of the geological attribute information, a preset covariance function is selected from the kernel function library as the prior kernel function of the Gaussian process;

[0028] With the goal of maximizing marginal likelihood, the hyperparameters of the covariance function are optimized using the geological attribute data between the known grids. Based on the optimized covariance function, the dependency values ​​between each pair of grid points are determined. The hyperparameters include at least one of variance scale and length scale.

[0029] Further, the step of weighting the geological attribute information of the unknown grids in the regular grid model based on the dependency value to determine the corresponding initial belief map includes:

[0030] For any unknown grid point in the regular grid model, determine all known data points within its search neighborhood, construct a Kriging equation system based on the dependency relationship values ​​between the known points, and solve the Kriging equation system to obtain the Kriging weight of each known data point in the search neighborhood.

[0031] The geological attribute information of the unknown grid point is calculated by multiplying the geological attribute information of all known data points within the search area of ​​the unknown grid point with their corresponding Kriging weights and summing the results. Based on the geological attribute information of all unknown grid points, the corresponding initial belief map is determined.

[0032] Furthermore, the step of inputting the initial belief map and the Markov decision process model into a pre-defined Monte Carlo tree search algorithm for virtual drilling simulation to determine the corresponding recommended drilling layout includes:

[0033] The root node of the preset search tree is initialized based on the belief graph, and the corresponding Monte Carlo tree search algorithm is determined. The Monte Carlo tree search algorithm uses UCT scoring and constructs development items and exploration items, which are used to balance exploration and development.

[0034] The initial belief map and the Markov decision process model are input into the Monte Carlo tree search algorithm. The Monte Carlo tree search algorithm selects a candidate borehole location from the action space of the Markov decision process model as the current simulation action, and performs simulation sampling based on the initial belief map to determine the geological realization of the resource distribution in the corresponding target area.

[0035] In the geological realization of the resource distribution in the target area, forward modeling is performed on the borehole location corresponding to the current simulation action to determine the corresponding virtual observation results;

[0036] Based on the virtual observation results, the expected cumulative reward of the corresponding candidate boreholes is determined, and one or more candidate boreholes with the highest expected cumulative reward are output. Based on the output candidate boreholes, the corresponding recommended drilling scheme is determined.

[0037] Secondly, this application provides a resource exploration device based on virtual drilling, comprising:

[0038] The target area grid determination module is used to acquire historical exploration data of the target area, discretize the target area into a regular grid model, and associate the historical exploration data with the regular grid model. The historical exploration data includes borehole data and geological attribute information of the target resources. The borehole data includes borehole location and borehole results.

[0039] The initial belief map determination module is used to perform geological attribute dependency calculation based on the geological attribute information between each known grid in the regular grid model, determine the corresponding dependency value, and perform weighted calculation on the geological attribute information of unknown grids in the regular grid model based on the dependency value to determine the corresponding initial belief map. The initial belief map represents the prior probability or attribute strength of the existence of target resources in each grid.

[0040] The virtual drilling module is used to determine the corresponding state space based on the initial belief map, the corresponding action space based on the candidate borehole location set in the regular grid model, the corresponding observation space based on the drilling results, the corresponding reward function based on the cost-effectiveness of the drilling, and the corresponding Markov decision process model based on the state space, the action space, the observation space, and the reward function. The initial belief map and the Markov decision process model are then input into a set Monte Carlo tree search algorithm to perform virtual drilling simulation and determine the corresponding recommended drilling layout scheme.

[0041] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the virtual drilling-based resource exploration method.

[0042] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the resource exploration method based on virtual drilling.

[0043] Fifthly, this application provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the aforementioned resource exploration method based on virtual drilling.

[0044] As can be seen from the above technical solution, this application provides a resource exploration method and apparatus based on virtual drilling. By acquiring the scope and historical exploration data of the target area, the target area is discretized into a regular grid model, and the historical exploration data is associated with the regular grid model. In the regular grid model, the geological attribute dependency is calculated based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted according to the dependency value to determine the corresponding initial belief map. A Markov decision process model is constructed based on the initial belief map. The initial belief map and the Markov decision process model are input into a set Monte Carlo tree search algorithm to perform virtual drilling simulation and determine the corresponding recommended drilling scheme, thereby improving the efficiency and accuracy of real drilling. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1This is one of the flowcharts illustrating the resource exploration method based on virtual drilling in the embodiments of this application;

[0047] Figure 2 This is a structural diagram of the resource exploration device based on virtual drilling in the embodiments of this application;

[0048] Figure 3 This is a schematic diagram of the structure of the electronic device in the embodiments of this application.

[0049] Figure label:

[0050] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver storage unit 9144, antenna 9111, speaker 9131, microphone 9132. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0052] The acquisition, storage, use, and processing of data in this application comply with relevant laws and regulations.

[0053] Traditional exploration methods rely primarily on geological mapping, geophysical exploration, and limited borehole verification, resulting in long exploration cycles and high costs when dealing with complex geological conditions and deep mineral exploration. This application provides a resource exploration method and apparatus based on virtual drilling. By acquiring the scope and historical exploration data of the target area, the target area is discretized into a regular grid model, and the historical exploration data is correlated with this model. Within the regular grid model, geological attribute dependencies are calculated based on the geological attribute information between known grids to determine corresponding dependency values. Based on these dependency values, the geological attribute information of unknown grids in the regular grid model is weighted to determine the corresponding initial belief map. A Markov decision process model is constructed based on the initial belief map. The initial belief map and the Markov decision process model are then input into a pre-defined Monte Carlo tree search algorithm for virtual drilling simulation to determine the recommended drilling layout scheme. This improves the efficiency and accuracy of real drilling.

[0054] First, let me explain the application scenario of this application. In the field of real resource exploration, traditional resource exploration guides mineral exploration work through resource maps and actual drilling results. However, this approach is characterized by high costs and resource waste due to the large number of actual drilling points, and low efficiency due to the time required for actual drilling.

[0055] The overall approach of this application is as follows:

[0056] First, spatial modeling of mineralization indicators is performed using the Kriging method and Gaussian process regression to construct a "mineralization probability map," also known as an initial belief map. In other words, a guiding map with initial mineralization probabilities is first built in the target area. This map is constructed based on historical drilling data, including the locations of historical drilling points, drilling results, and surrounding geological properties.

[0057] Next, the resource exploration problem is transformed into a partially observable Markov decision process, and a Markov decision process model is constructed. Based on the initial belief map and the Markov decision process model, a Monte Carlo tree search algorithm is used to perform virtual drilling on this initial belief map, obtaining the virtual drilling results. The reward of the Markov decision process model that can be obtained from the virtual drilling results is output, and the initial belief map is adaptively updated using Bayesian algorithm based on the virtual drilling results to update the global mineralization probability, resulting in an updated belief map. Throughout the virtual drilling, the Monte Carlo tree search algorithm uses the UCT scoring method for simulated exploration.

[0058] Finally, by continuously iterating and updating the belief graph, the cumulative reward of the Markov decision process model is output. The virtual drilling point with the highest cumulative reward value is then output as the recommended drilling point. By drilling at this recommended point, drilling efficiency can be maximized. In this process, the reward function of the Markov decision process model is adaptively adjusted to the virtual drilling radiation intensity, allowing the reward value to be directly determined using the virtual radiation intensity value.

[0059] To improve the efficiency and accuracy of real drilling, this application provides an embodiment of a resource exploration method based on virtual drilling, see [link to embodiment]. Figure 1 The resource exploration method based on virtual drilling specifically includes the following:

[0060] Step S101: Obtain historical exploration data of the target area, discretize the target area into a regular grid model, and associate the historical exploration data with the regular grid model. The historical exploration data includes borehole data and geological attribute information of the target resource. The borehole data includes borehole location and borehole results.

[0061] Optionally, in this embodiment, the task of this step is to convert complex, continuous, and irregular geological exploration information in the real world into a structured data format.

[0062] Optionally, in this embodiment, all available historical exploration data are systematically extracted from past exploration reports, geological maps, borehole databases, and geophysical and geochemical measurement results of the target area.

[0063] Specifically, borehole data records the spatial coordinates (such as latitude, longitude, and elevation) of each drilled borehole and its drilling results, i.e. whether the target resource was found at that location (e.g., mineralization or no mineralization). In digital processing, this result is quantified into a binary variable, such as "mineralization found" being marked as 1 and "no mineralization found" being marked as 0.

[0064] Specifically, the geological attribute information of the target resource includes various indirect or direct evidence related to its mineralization potential. This can be grade data measured in boreholes (e.g., gold content in g / t), anomaly intensity from geophysical methods (e.g., magnetic and electrical methods), elemental enrichment concentrations from geochemical methods, linear tectonic or alteration zone information interpreted by remote sensing, and so on. These attributes collectively constitute the basis for inferring the resource potential of unknown areas. In this embodiment, the geological attribute information is merely an enumeration and not an exhaustive list; other types of attribute information related to mineralization potential do not affect the implementation of this embodiment.

[0065] In this embodiment, the integrity of the information is ensured by using direct drilling results and indirect geological attribute information. This not only allows us to learn where mineralization has occurred, but also to understand the geological patterns that suggest mineralization may occur here.

[0066] Optionally, in this embodiment, a regular spatial grid is defined for the entire target area based on the exploration accuracy requirements and the balance of computing resources. Preferably, a two-dimensional rectangular grid (or a cubic grid in three-dimensional applications) is typically used, with each grid cell (or pixel) having a uniform size (e.g., 25 meters × 25 meters).

[0067] Optionally, in this embodiment, each known data point (such as borehole location or geochemical sampling point) is assigned to a specific grid cell based on its geographical coordinates. A grid cell may contain multiple data points or none.

[0068] For each grid cell, assign a corresponding attribute value based on the known data falling into it.

[0069] For meshes with drilled holes, the drill result (1 or 0) is directly used as the most authoritative "label" for that mesh.

[0070] For grids with other geological attribute data, their measured values ​​(such as grade value, radiation intensity value) are recorded.

[0071] For a grid with no known data, its state is marked as "unknown" (often represented by NaN or a specific identifier in the program).

[0072] Understandably, the grid space constructed in this step precisely defines the knowledge status of the current exploration stage, intuitively showing the distribution of known mineral deposits, anomaly zones, and large blank areas, providing decision-makers with a clear spatial understanding.

[0073] Step S102: In the regular grid model, geological attribute dependency calculation is performed based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted based on the dependency value to determine the corresponding initial belief map. The initial belief map represents the prior probability or attribute strength of the existence of target resources in each grid.

[0074] Optionally, in this embodiment, the purpose of this step is to transform the discrete known point space constructed in step S101 into a continuous global probability (or attribute strength) distribution map, i.e., the initial belief map. Essentially, it is a process combining spatial interpolation and statistical inference.

[0075] Optionally, in this embodiment, the geological attribute dependency calculation is based on the fundamental assumption that geographically close points have similar geological attributes (such as grade and mineralization), and this similarity weakens as the distance between points increases.

[0076] First, iterate through all pairwise combinations of known grid points (i.e., points with historical borehole data). For each pair of points, calculate the spatial distance between them, and simultaneously calculate the square of the difference in their geological attribute values ​​(e.g., 1 for mineralization, 0 for no mineralization; directly using grade values, etc.). This process ultimately yields a dataset containing multiple sets of "distances h" and corresponding "variance values ​​γ(h)," i.e., a scatter plot of geological attributes.

[0077] Next, a continuous mathematical function is used to quantitatively describe the relationship between "distance" and "attribute dependency". Preferably, the mathematical function can be a function of variation or a function of covariance.

[0078] By fitting the calculated geological attribute scatter plot to the mathematical function model, we can determine the hyperparameters of the mathematical function model that are most suitable for the current target area. Thus, we obtain a mathematical function model that can quantify the geological attribute dependency between any two points. For any given distance h, we can use this function to calculate its covariance C(h) or variation γ(h).

[0079] Optionally, in this embodiment, after obtaining a quantitative model describing spatial dependencies, the prediction of attribute values ​​for all unknown grids can begin. Specifically, this involves precise weighted calculations based on spatial dependencies.

[0080] Specifically, for each unknown grid point in the regular grid model, the system searches for all known grid points within a set search radius (usually related to the correlation length), centered on that unknown point. The goal is to assign appropriate weights to the attribute values ​​of a set of known points so that the estimated value of the unknown point calculated using these weights and known values ​​is optimal (unbiased and with minimal variance). The weight assignment is solved using the dependency values ​​(variance function / covariance function) determined in the above steps through the Kriging equations (a mathematical system).

[0081] Specifically, the core of the Kriging equations is:

[0082] The closer a known point is to an unknown point, the greater its weight should be.

[0083] If the known points are very close to each other (i.e., their information is redundant), then their respective weights will be reduced accordingly to avoid double counting;

[0084] The sum of all weights is 1 to ensure the unbiasedness of the estimate;

[0085] For each unknown point, after solving for the optimal weights of all known points around it, the estimated value of the unknown point is calculated by weighted summation. This estimated value is the belief value of the unknown point grid.

[0086] For binary problems, the belief value is the prior probability of the existence of the target resource, ranging from [0, 1].

[0087] For continuous numerical values, the belief value represents the strength of the attribute.

[0088] Once all unknown grid points have been calculated, a continuous initial belief map covering the entire target area is generated. This map visually illustrates which areas are most likely to be rich in the target resources based on existing data.

[0089] Step S103: Determine the corresponding state space based on the initial belief map, determine the corresponding action space based on the candidate borehole location set in the regular grid model, determine the corresponding observation space based on the drilling results, determine the corresponding reward function based on the cost-effectiveness of drilling, determine the corresponding Markov decision process model based on the state space, the action space, the observation space, and the reward function, input the initial belief map and the Markov decision process model into the set Monte Carlo tree search algorithm to perform virtual drilling simulation, and determine the corresponding recommended drilling layout scheme.

[0090] Optionally, in this embodiment, this step builds a decision engine based on the initial belief map described above. Specifically, the initial belief map (belief(x,y)) is transformed into a decision optimization problem, and a massive "virtual drilling" simulation is performed in a computer using an advanced search algorithm to finally output the drilling scheme with the best economic benefits.

[0091] Optionally, in this embodiment, a mathematical model framework is first established for the exploration decision problem, which closely integrates geological uncertainties with benefit objectives.

[0092] Specifically, POMDP seven-tuple modeling (Markov decision process model).

[0093] formula: As shown in the table below:

[0094]

[0095] Optionally, in this embodiment, the state space.

[0096] The state space represents the actual subsurface conditions, i.e., whether each grid cell contains economically exploitable ore bodies. In real exploration, this is a partially observable state, and we cannot directly know the whole picture. Therefore, in the POMDP model, "state" is defined as the "belief state" of the current understanding of the subsurface. This belief state is the initial belief map itself.

[0097] The initial belief map is a two-dimensional probability matrix that we have constructed, corresponding one-to-one with the regular grid. The value of each cell (between 0 and 1) represents the confidence that the target resource exists at that location. The entire state space is the set of all possible belief maps.

[0098] Optionally, in this embodiment, the action space.

[0099] An action represents an operation that the prospector can perform, and is a set of all candidate borehole locations. Candidate borehole locations correspond to grid points in the regular grid model that have not yet been verified by drilling. In the early stages of algorithm execution, the action space is large (almost entirely undrilled points); as virtual (and subsequent real) drilling proceeds, selected actions are removed from the candidate set, and the action space dynamically shrinks accordingly.

[0100] Optionally, in this embodiment, the observation space.

[0101] The observation space defines the possible outcomes after performing an action (drilling a hole). Preferably, to simplify the model and highlight the main contradictions, we define observations as binary: "mineralization observed" (denoted as 1) or "mineralization not observed" (denoted as 0).

[0102] Optionally, in this embodiment, a reward function.

[0103] The reward function acts as a compass driving the optimization of the entire system, directly transforming geological discoveries into economic value.

[0104] Specifically, in the reward function, we rewrite the POMDP reward function into a form that can be directly measured by virtual radiation intensity. The reward function is as follows:

[0105]

[0106] β: Economic conversion factor

[0107] C drill Drilling cost (RMB / hole)

[0108] The mathematical expression for the virtual drill radiation intensity I(a) is as follows:

[0109] Physical assumptions

[0110] Radiation intensity ∝ Grade

[0111] Radiation intensity ∝ mineralization thickness

[0112] Radiation intensity ∝ 1 / (burial depth + d0) (d0 is a small constant to prevent division by zero)

[0113]

[0114] I: Virtual drill radiation intensity (any unit, linearly comparable to the "radiation flux value" in the report)

[0115] C: Predicted grade (g / t)

[0116] T: Predicted mineralization thickness (m)

[0117] Z: Predicted burial depth (m)

[0118] α: Calibration coefficient (calculated by back-calculating from existing boreholes to minimize the RMS error between I and the measured flux value)

[0119] Regarding the resulting effect, the reward function can directly measure the probability of encountering minerals based on the virtual radiation intensity, and achieve the following:

[0120] Discovery Reward: A positive reward (e.g., +10) is given when drilling at a location and observing "mineralization". This value can be calibrated based on the estimated grade of the resource, the metal price, and the recovery rate.

[0121] Drilling cost: Each drilling operation, regardless of the outcome, generates a negative reward (i.e., cost, such as -1) to reflect the financial, time, and manpower costs of drilling.

[0122] This function forces the algorithm not only to find ore bodies but also to consider the cost of finding them. It opposes blind drilling and encourages obtaining the greatest geological discoveries and economic returns with the fewest drill holes and the lowest cost, thereby realizing the green exploration concept of "less drilling and more precise drilling".

[0123] The definitions of the four core components above clearly describe the core elements of the exploration problem: in an uncertain world (belief state), we obtain observations (drilling results) by performing actions (choosing borehole locations) and pursue the maximization of long-term cumulative rewards (net benefits).

[0124] Optionally, in this embodiment, in order to pursue the optimal solution of the POMDP model, we use Monte Carlo Tree Search (MCTS) for virtual drilling simulation.

[0125] Specifically, the selection steps for Monte Carlo Tree Search (MCTS).

[0126] Starting from the root node of the MCTS tree (the current belief state), the algorithm recursively selects the optimal child node and searches downwards. The selection criterion adopts the Upper Confidence Bound (UCT) formula, striking a balance between "utilization" and "exploration," where "utilization" is the "development" term in the formula, representing the development and utilization of the mineral resource.

[0127]

[0128] Where n is the node currently being evaluated (i.e., a candidate borehole location that we are considering).

[0129] np is the parent node of node n;

[0130] Q(n) is the total cumulative reward, which is the sum of the total rewards obtained from all subsequent simulations starting from this borehole position (node ​​n) (e.g., the sum of the rewards from all virtual drilling paths).

[0131] N(n) visit count, the total number of times this borehole location (node ​​n) is visited during the MCTS simulation;

[0132] N(np) represents the number of times the parent node was visited, which is the total number of times the parent node (i.e., the previous drilling decision) was visited.

[0133] c is the exploration constant, used to balance "development" and "exploration";

[0134] Development item: Prefer to select nodes with high average rewards in historical simulations (i.e., holes with good virtual drilling results in the past).

[0135] Exploration: Provide a "reward score" for nodes with relatively few visits to encourage the algorithm to try holes with unknown potential but possibly better.

[0136] Understandably, the UCT strategy ensures that the search process does not get stuck in local optima too early, and can systematically explore a wide decision space while focusing on decision paths that have already shown high returns.

[0137] Specifically, the extended steps of Monte Carlo Tree Search (MCTS).

[0138] When a node that has not yet been fully explored (i.e., a candidate borehole) is found, the algorithm creates one or more new child nodes for that node. These child nodes represent new belief states corresponding to different observations (such as "mineralized" or "mineralized") that may occur after the drilling action.

[0139] Specifically, the simulation steps of Monte Carlo Tree Search (MCTS).

[0140] Starting from the newly expanded node, the algorithm performs a "rapid virtual drilling" until the preset depth or the end of the simulation. During this process:

[0141] Based on the current belief map, a possible underground reality is randomly sampled.

[0142] In this sampled "virtual world," subsequent drilling actions are performed according to a certain strategy (such as random selection). For each virtual borehole, a virtual observation result is generated based on the "real" situation of that location in the virtual world (mineralized or unmineralized) and combined with a simple noise model. The total reward obtained for the entire simulated path is calculated cumulatively.

[0143] Specifically, the backhaul steps of Monte Carlo Tree Search (MCTS).

[0144] The cumulative reward obtained in the simulation steps is propagated back along the previously selected path to update the statistics of all nodes on the path from the leaf node to the root node (mainly the total reward and the number of visits).

[0145] After repeating the above four steps multiple times, the algorithm has fully evaluated all possible actions starting from the current state. Ultimately, the recommended drilling scheme selects the action (hole position) corresponding to the child node with the most visits under the root node. This is because the number of visits reflects the algorithm's overall trust in the value of that action after weighing exploration and utilization.

[0146] This example demonstrates how this embodiment uses historical exploration data to describe the spatial correlation of mineralization indicators using geostatistics, constructs an initial belief map, updates the belief map with global mineralization probability based on Bayesian methods, and finally uses MCTS to simulate the expected net returns of different borehole sequences to select the optimal borehole scheme.

[0147] As described above, the resource exploration method based on virtual drilling provided in this application can discretize the target area into a regular grid model by acquiring the scope and historical exploration data of the target area, and associate the historical exploration data with the regular grid model. In the regular grid model, the geological attribute dependency is calculated based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted according to the dependency value to determine the corresponding initial belief map. A Markov decision process model is constructed based on the initial belief map. The initial belief map and the Markov decision process model are input into a set Monte Carlo tree search algorithm to perform virtual drilling simulation and determine the corresponding recommended drilling scheme, thereby improving the efficiency and accuracy of real drilling.

[0148] In one embodiment of the resource exploration method based on virtual drilling in this application, it may further include the following:

[0149] Step S201: Obtain historical exploration data of the target area. The historical exploration data includes borehole data and geological attribute information of the target resources. The borehole data includes borehole location and borehole results indicating whether minerals were encountered or not.

[0150] Step S202: Discretize the target region into a regular grid model at a preset resolution;

[0151] Step S203: Map the borehole location to the corresponding grid cell in the regular grid model, and mark the borehole results and geological attribute information of the grid cell as known data.

[0152] Optionally, in this embodiment, two types of historical data are extracted from existing geological reports, databases, and field records:

[0153] Drilling data requires precise recording of the spatial coordinates (such as latitude and longitude) of each borehole and clarification of its drilling results—that is, the conclusive judgments drawn from core analysis, etc.—and standardized representation in a binary form of "mineralized material found" (assigned a value of 1) or "mineralized material not found" (assigned a value of 0).

[0154] The target resource's geological attributes include, but are not limited to, geochemical element grades, geophysical anomalies, lithological codes, and other mineralization-related attribute parameters. Data cleaning and normalization processes ensure the accuracy, consistency, and computability of all input information.

[0155] Optionally, in this embodiment, a standard computational framework is created for spatial analysis. Based on the exploration accuracy requirements and computational resources, a fixed preset resolution (e.g., a 50m × 50m grid cell) is determined, and the entire continuous, irregular target area is discretized into a regular grid model. This model covers the entire study area, and each grid cell becomes the basic unit of spatial analysis, possessing a unique index identifier.

[0156] Optionally, in this embodiment, the fusion of real-world data and the digital mesh model is completed. Through spatial coordinate matching, each borehole location obtained in S201 is accurately mapped to a specific mesh cell of the mesh created in S202.

[0157] Through step S203, this embodiment successfully obtained a regular grid model, dividing the target space into a unified spatial analysis unit, laying the foundation for the subsequent construction of the initial belief map.

[0158] In one embodiment of the resource exploration method based on virtual drilling in this application, it may further include the following:

[0159] Step S301: Construct a two-dimensional matrix with the same dimensions as the regular grid model;

[0160] Step S302: For each grid cell, if it contains the borehole location, fill in the borehole results and geological attribute information at the corresponding position in the matrix;

[0161] Step S303: If the matrix does not contain the borehole location, then fill in a specific null value identifier in the corresponding position of the matrix to mark the location as an unexplored area.

[0162] Optionally, in this embodiment, a two-dimensional matrix with the exact same dimensions as the regular grid model is created. As a digital container carrying all spatial information for the entire region, the number of rows and columns of the two-dimensional matrix directly corresponds to the number of vertical and horizontal divisions of the grid model, ensuring a one-to-one mapping of spatial locations.

[0163] Optionally, in this embodiment, each grid cell is traversed and assigned a value based on its data status. For grid cells containing known borehole locations, the empirical results of that borehole are filled in at the corresponding coordinate position in the matrix. This includes a binary "borehole result" (e.g., 1 for mineralization and 0 for non-mineralization) and related quantitative "geological attribute information" (such as grade values, mineralization thickness, or radiation intensity values). Essentially, this precisely anchors the "hard data" obtained from field exploration into the digital grid system.

[0164] Optionally, in this embodiment, for cells that do not contain drill holes, the system fills in a specific null value identifier at the corresponding position in the matrix. These areas are marked as "data missing" or "state unknown" to contrast with the grid that has been filled with actual data.

[0165] Through step S303, this embodiment realizes the normalization of spatial data, depicts the current exploration status, and lays a solid data foundation for subsequent spatial interpolation to generate the initial belief map.

[0166] In one embodiment of the resource exploration method based on virtual drilling in this application, it may further include the following:

[0167] Step S401: In the regular grid model, traverse all known grid point pairs, calculate the spatial distance and semivariance of geological attribute values ​​between each pair of grid points, and determine the corresponding geological attribute scatter plot;

[0168] Step S402: Fit the geological attribute scatter plot according to the model in the preset variogram model library, and determine the dependency value between each grid point pair according to the fitted variogram model. The dependency value is used to characterize the spatial dependency of geological attributes between grids.

[0169] Optionally, in this embodiment, this step is a fundamental step in geostatistical spatial analysis, and the core task of the variogram model is to extract the original spatial dependence pattern from the actual exploration data.

[0170] First, in the regular grid model, all points with known geological attribute values ​​are identified. Then, all point pairs formed by these known points are traversed. For each point pair, two key calculations are performed:

[0171] Calculate spatial distance: Determine the straight-line distance between two points.

[0172] Calculate the semivariance: Calculate half the square of the difference between the geological attribute values ​​of the two points.

[0173] All point pairs are grouped according to their distance h (i.e., placed into different "distance tolerance" intervals), and then the average of all semivariances within each distance group is calculated. Finally, a series of data points is output, consisting of "distance h" and the corresponding "average semivariance γ(h)". This "distance-semivariance" scatter plot is the geological attribute scatter plot, revealing the original statistical laws governing the variation of geological attribute differences with distance within the current region.

[0174] Optionally, in this embodiment, the preset variation function model library includes several standard mathematical function forms such as the spherical model, the exponential model, and the Gaussian model.

[0175] Preferably, in this embodiment, a Gaussian model is selected. The geological attribute scatter plot is compared and fitted with the Gaussian model, and the key parameters in the Gaussian model are adjusted through an optimization algorithm. After the fitting is completed, the variogram model is formally determined. For any two grid points in the regular grid model, regardless of whether they are known or not, as long as the spatial distance h between them is input, a definite dependency value can be calculated through this variogram model.

[0176] Specifically, the variation function model is as follows:

[0177]

[0178] in:

[0179] γ(h) is the semivariance at a distance of h.

[0180] σ² is the sill value, representing the population variance of the process.

[0181] It is a variable range, representing the scope of influence of spatial correlation.

[0182] Understandably, the variogram measures "difference"; the larger the value, the greater the expected difference between two points, meaning they are less similar.

[0183] Through step S402, this embodiment successfully summarizes discrete empirical data into a continuous and stable mathematical law, providing a solid data foundation for subsequent Kriging interpolation.

[0184] In one embodiment of the resource exploration method based on virtual drilling in this application, it may further include the following:

[0185] Step S501: In the regular grid model, based on the prior distribution characteristics of the geological attribute information, a preset covariance function is selected from the kernel function library as the prior kernel function of the Gaussian process;

[0186] Step S502: With the goal of maximizing the marginal likelihood, the hyperparameters of the covariance function are optimized using the geological attribute data between the known grids. Based on the optimized covariance function, the dependency values ​​between each pair of grid points are determined. The hyperparameters include at least one of variance scale and length scale.

[0187] Optionally, in this embodiment, this step is a fundamental step in geostatistical spatial analysis, parallel to step S402. The core lies in defining how the similarity of attribute values ​​between any two points in space changes with distance using a covariance function (also known as a kernel function). Unlike the variogram function in step S402, the covariance function in this step measures "co-variance"; the larger the value, the more likely the two points are to be simultaneously above or below the average, i.e., the more similar they are.

[0188] Optionally, in this embodiment, the covariance function C(h) and the variation function γ(h) are closely related and are complements of each other. That is, the relationship between the covariance function C(h) and the variation function γ(h) is: γ(h) = C(0) - C(h).

[0189] Therefore, the covariance function model is as follows:

[0190]

[0191] The covariance model is selected from the prior distribution characteristics of the geological attribute information of the target area.

[0192] For example, if geological properties are considered to change gradually and have a wide range of influence, a function that produces a smooth interpolation surface, such as the squared exponential covariance function, will be preferred; if the property changes have obvious "inflection points" or linear trends, the Marton kernel or a combination of linear kernels may be considered.

[0193] Optionally, in this embodiment, after selecting the form of the covariance function, multiple hyperparameters of the covariance function are calibrated using actual observation data.

[0194] Hyperparameters mainly include variance scale and length scale. Variance scale describes the overall fluctuation range of geological properties within a region; length scale defines the influence range of property correlation and is the direct controller of "dependency value".

[0195] The optimization process aims to maximize the marginal likelihood. Marginal likelihood refers to the probability of observing existing known data given hyperparameters. The system iteratively adjusts the hyperparameters using algorithms such as gradient descent or Bayesian optimization to find the optimal set of values ​​that maximizes this probability. Once the optimal hyperparameters are found, the covariance function is completely determined. At this point, for any two grid points in the regular grid model, regardless of whether they already have data, their spatial distance can be substituted into the optimized covariance function to calculate an accurate dependency value.

[0196] It is understandable that the covariance function model in this step and the variation function model in step S402 are mathematically equivalent. The variation function is used to provide the optimal interpolation result for the subsequent steps, while the covariance function model is used to provide the posterior probability distribution for the subsequent steps.

[0197] Through step S502, this embodiment accurately quantifies the spatial autocorrelation structure of geological properties within the study area, providing a solid data foundation for subsequent Kriging interpolation.

[0198] In one embodiment of the resource exploration method based on virtual drilling in this application, it may further include the following:

[0199] Step S601: For any unknown grid point in the regular grid model, determine all known data points within its search neighborhood, construct a Kriging equation system based on the dependency relationship values ​​between the known points, and solve the Kriging equation system to obtain the Kriging weight of each known data point in the search neighborhood.

[0200] Step S602: Multiply the geological attribute information of all known data points within the search area of ​​the unknown grid point by their corresponding Kriging weights and sum them to calculate the geological attribute information of the unknown grid point. Based on the geological attribute information of all unknown grid points, determine the corresponding initial belief map.

[0201] Optionally, in this embodiment, for each unknown grid point in the regular grid model, the system searches for all known grid points within a set search radius (usually related to the correlation length). The goal is to assign appropriate weights to the attribute values ​​of a set of known points so that the estimated value of the unknown point calculated using these weights and known values ​​is optimal (unbiased and with minimal variance). The weight assignment is solved using the dependency values ​​(variance function / covariance function) determined in the above steps through the Kriging equations (a mathematical system).

[0202] Specifically, the core of the Kriging equations is:

[0203] The closer a known point is to an unknown point, the greater its weight should be.

[0204] If the known points are very close to each other (i.e., their information is redundant), then their respective weights will be reduced accordingly to avoid double counting;

[0205] The sum of all weights is 1 to ensure the unbiasedness of the estimate;

[0206] For each unknown point, after solving for the optimal weights of all known points around it, the estimated value of the unknown point is calculated by weighted summation. This estimated value is the belief value of the unknown point grid.

[0207] For binary problems, the belief value is the prior probability of the existence of the target resource, ranging from [0, 1].

[0208] For continuous numerical values, the belief value represents the strength of the attribute.

[0209] Once all unknown grid points have been calculated, a continuous initial belief map covering the entire target area is generated. This map visually illustrates which areas are most likely to be rich in the target resources based on existing data.

[0210] Through step S602, this embodiment successfully obtained the initial belief map, laying a solid data foundation for subsequent virtual drilling based on the initial belief map.

[0211] In one embodiment of the resource exploration method based on virtual drilling in this application, it may further include the following:

[0212] Step S701: Initialize the root node of the preset search tree according to the belief graph, determine the corresponding Monte Carlo tree search algorithm, the Monte Carlo tree search algorithm adopts UCT scoring and constructs development items and exploration items, the development items and exploration items are used to balance exploration and development;

[0213] Step S702: Input the initial belief map and the Markov decision process model into the Monte Carlo tree search algorithm. The Monte Carlo tree search algorithm selects a candidate borehole location from the action space of the Markov decision process model as the current simulation action, and performs simulation sampling according to the initial belief map to determine the geological realization of the resource distribution in the corresponding target area.

[0214] Step S703: In the geological realization of the resource distribution in the target area, perform forward modeling calculation on the borehole location corresponding to the current simulation action to determine the corresponding virtual observation results;

[0215] Step S704: Determine the expected cumulative reward of the corresponding candidate boreholes based on the virtual observation results, output one or more candidate boreholes with the highest expected cumulative reward, and determine the corresponding recommended drilling scheme based on the output candidate boreholes.

[0216] Optionally, in this embodiment, this step is the construction and application step of the Monte Carlo tree search algorithm.

[0217] First, the construction steps of the Monte Carlo tree search algorithm.

[0218] The search tree is initialized and its selection strategy is established. At the start of the algorithm, the "belief graph," representing the probability of the global resource distribution, is encapsulated as the root node of the search tree. This root node represents the starting point of the decision-making process, i.e., the initial state before any virtual drilling has been performed. Simultaneously, UCT (Upper Confidence Interval) is established as the standard scoring mechanism for tree node selection. The UCT score is composed of the sum of "development items" and "exploration items":

[0219]

[0220] The development item is calculated as the average reward a node receives (total cumulative reward divided by the number of visits). It tends to select nodes with high historical returns, guiding the algorithm to "cultivate" known high-potential areas, i.e., to maximize returns using existing knowledge. The exploration item, on the other hand, is proportional to the square root of the total number of visits to the parent node and inversely proportional to the number of visits to the node itself. Its purpose is to give extra "points" to nodes with relatively few visits, encouraging the algorithm to explore areas that have not yet been fully evaluated, thus avoiding getting trapped in local optima. By adjusting the weights of these two items with a constant coefficient, the UCT scoring mechanism cleverly balances the core contradiction between "developing" known high-value areas and "exploring" unknown potential areas.

[0221] Second, the application steps of the Monte Carlo tree search algorithm.

[0222] First, the MCTS algorithm selects a location from the Markov decision process model's action space—that is, from all preset candidate borehole locations—as the "simulated action" to be evaluated. This selection can be based on the tree search strategy from the previous step, or it can be randomly selected from a new simulation path.

[0223] Secondly, the simulation sampling based on the initial belief map involves treating the belief map as a probability distribution and randomly selecting a possible subsurface resource distribution scenario from this distribution, called a geological realization. By generating a large number of different geological realizations, the algorithm can explore all possible subsurface conditions that conform to current knowledge, thereby avoiding relying on a single model for decision-making. The geological realization provides a computational environment for subsequent virtual drilling.

[0224] Next, in the generated geological implementation, the algorithm will perform selected simulation actions to conduct virtual drilling.

[0225] Forward modeling of virtual drilling is a process of calculating observation results based on a known model (in this case, geological realization) and known physical laws. That is, along the set borehole trajectory, the geological attribute information of each grid passed through is queried and integrated, and finally one or more theoretical observation values ​​are output.

[0226] Finally, in actual exploration, any measurement inevitably contains errors and uncertainties. These uncertainties may stem from errors in laboratory analysis, sample depletion or dilution caused by core drilling, the subjectivity of geological logging, and the microscopic heterogeneity of the subsurface geological body itself. The theoretical observations obtained from the above steps are ideal values; to more realistically simulate reality, perturbations must be introduced.

[0227] In this embodiment, an observation noise model is introduced. The noise model is a mathematical probability distribution used to quantitatively describe the statistical characteristics of the observation error. The most common model is a Gaussian (normal) distribution with a mean of zero and a variance of σ², indicating that the error fluctuates around the true value, and the range of fluctuation is determined by σ². Specifically, the theoretical observation value is superimposed with a noise value randomly selected from the noise model, and the result of the superposition is the final virtual observation result.

[0228] Next, firstly, the immediate reward of the Markov decision process model reward function corresponding to the virtual observation results is obtained. Simultaneously, the initial belief map is updated using Bayesian methods based on the virtual observation results. This Bayesian-updated belief map, representing the latest geological knowledge, is then used as the current state input for the Partially Observable Markov Decision Process (POMDP). Based on this new belief state, subsequent virtual drilling actions are selected, virtual drilling is performed again, new virtual observation results are obtained, and the belief map is updated accordingly. Each recursion receives an immediate reward from the reward function based on its virtual observation results. This recursive process continues until the simulation termination conditions are met (e.g., reaching a preset simulation depth, budget exhaustion, or exploration reaching the boundary).

[0229] Ultimately, all immediate rewards generated by this series of future actions are discounted and summed according to a preset discount factor to calculate the "subsequent reward". The discount factor (γ, usually less than 1) ensures that the algorithm favors recent gains, which is in line with the economic principle of "earning money earlier is better", and also guarantees that the sum of rewards over an infinite time series converges.

[0230] After simulating future paths and calculating subsequent rewards, the value is summarized and backtracked. The immediate reward directly generated by this virtual drilling (i.e., the action currently being evaluated) is added to the sum of all calculated future rewards to obtain the expected cumulative reward for this candidate borehole under the current specific simulated path.

[0231] One or more candidate boreholes with the highest expected cumulative reward are output as the recommended drilling scheme. At this time, the candidate boreholes balance factors such as environmental protection and economic benefits and are the optimal drilling points after evaluation.

[0232] Through step S704, this embodiment successfully obtained the optimal drilling scheme based on the construction and application of Monte Carlo tree search.

[0233] To improve the efficiency and accuracy of real drilling, this application provides an embodiment of a virtual drilling-based resource exploration apparatus for implementing all or part of the aforementioned virtual drilling-based resource exploration method. See [link to embodiment]. Figure 2 The virtual drilling-based resource exploration device specifically includes the following components:

[0234] The target area grid determination module 10 is used to acquire historical exploration data of the target area, discretize the target area into a regular grid model, and associate the historical exploration data with the regular grid model. The historical exploration data includes borehole data and geological attribute information of the target resource. The borehole data includes borehole location and borehole results.

[0235] The initial belief map determination module 20 is used to perform geological attribute dependency calculation based on the geological attribute information between each known grid in the regular grid model, determine the corresponding dependency value, and perform weighted calculation on the geological attribute information of the unknown grid in the regular grid model based on the dependency value to determine the corresponding initial belief map. The initial belief map represents the prior probability or attribute strength of the existence of target resources in each grid.

[0236] The virtual drilling module 30 is used to determine the corresponding state space based on the initial belief map, the corresponding action space based on the candidate borehole location set in the regular grid model, the corresponding observation space based on the drilling results, the corresponding reward function based on the cost-effectiveness of the drilling, and the corresponding Markov decision process model based on the state space, the action space, the observation space, and the reward function. The initial belief map and the Markov decision process model are input into a set Monte Carlo tree search algorithm to perform virtual drilling simulation and determine the corresponding recommended drilling layout scheme.

[0237] As described above, the resource exploration device based on virtual drilling provided in this application can discretize the target area into a regular grid model by acquiring the scope and historical exploration data of the target area, and associate the historical exploration data with the regular grid model. In the regular grid model, the geological attribute dependency is calculated based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted according to the dependency value to determine the corresponding initial belief map. A Markov decision process model is constructed based on the initial belief map. The initial belief map and the Markov decision process model are input into a set Monte Carlo tree search algorithm to perform virtual drilling simulation and determine the corresponding recommended drilling scheme, thereby improving the efficiency and accuracy of real drilling.

[0238] To further illustrate this solution, this application also provides a specific application example of using the aforementioned virtual drilling-based resource exploration device to implement the virtual drilling-based resource exploration method, which specifically includes the following:

[0239] The core physical principles of this scheme include:

[0240] 1.1 Spatial correlation modeling: Geostatistics (such as Kriging and Gaussian processes) are used to describe the spatial correlation of mineralization indicators and construct a "belief map".

[0241] 1.2 Bayesian Update: For each hole drilled, the global mineralization probability is updated using the observation results (mineralization found / mineralization not found) through the Bayesian formula.

[0242] 1.3 Maximizing Information Value: MCTS Simulation of Expected Net Gains (Discovery Rewards) for Different Drilling Sequences (Drilling cost), select the optimal location for the next hole.

[0243] The nuclear-related mathematical formulas used in this scheme include:

[0244] 1. POMDP 7-tuple

[0245] S: State space (whether each grid contains minerals)

[0246] A: Action space (candidate borehole location)

[0247] O: Observation space (drilling results)

[0248] T: State transition (only measurement state is updated, geological state remains unchanged)

[0249] Z: Observation probability (the probability of finding the mineral is given by the belief map)

[0250] R:

[0251] γ: Discount factor (0.9)

[0252] 2. Belief Renewal (Bayes)

[0253]

[0254] Formula: b′(s′) ∝ Observational Likelihood Z(o∣s′,a) Predicted Prior ∑s T(s′∣s,a) b(s)

[0255]

[0256] 3. UCT rating (MCTS selection)

[0257]

[0258]

[0259] 4. Spatial covariance (Kriging / GPR kernel function)

[0260] ; Relevance length (Gaussian kernel)

[0261] Numerator: 2 (empirical multiple, adjustable) Denominator: 1 (implicit) Unit: meter (m)

[0262]

[0263]

[0264] This solution can be applied in the following situations:

[0265]

[0266] The implementation steps and code examples for this solution are as follows:

[0267]

[0268] The concept of virtual drilling and its application in this scheme:

[0269] Virtual drilling refers to "drilling" a hole in a computer without actually drilling, using numerical simulation and rapid inversion to immediately obtain possible observations (lithology, mineralization, elemental content, etc.). These virtual observations are then fed into the POMDP belief update module, instantly refreshing the global mineralization probability map and updating the initial belief map, thereby assessing "what the expected net benefit would be if we actually drilled here."

[0270] Essentially, it equals differentiable / fast forward modeling + Bayesian inversion + observation noise model, returning results in milliseconds. This eliminates the need for simple random sampling in the "rollout" phase of MCTS. Instead, it uses "virtual real diamond" feedback based on physical-statistical coupling, significantly improving policy accuracy.

[0271] The underlying physical framework of virtual drilling is as follows:

[0272]

[0273] Virtual drilling in the closed-loop integration of POMDP-MCTS includes:

[0274] Tree node expansion → Candidate action a (hole position + hole depth)

[0275] Virtual Drill Rollout

[0276] Sample a geological realization si from the current belief b(s).

[0277] In this implementation, action a is executed, which calls the fast forward model ovirt.

[0278] Update belief using ovirt → new belief b′

[0279] Immediately calculate and accumulate the single-step reward r.

[0280] Backpropagation → Send back the accumulated rewards and update the UCT statistics.

[0281] Repeat this process hundreds of times, then select the node with the "most virtual visits and the highest average Q" as the next real borehole.

[0282] The steps to implement virtual drilling:

[0283]

[0284] The three technical forms of virtual drilling include:

[0285] 0D Quick Version - Only returns "Ore Found / Not Found" + Radiation Intensity (This version is the one with the uploaded MATLAB script).

[0286] 1D Logging Version – Returns grade-depth curves along the borehole track for updating the 3D block model.

[0287] 2D / 3D Differentiable Version – Uses differentiable earth modeling or deep learning proxy models, supports gradient backpropagation, and can directly perform end-to-end optimization.

[0288] Formula for virtual drill radiation intensity:

[0289] Physical assumptions

[0290] Radiation intensity ∝ Grade

[0291] Radiation intensity ∝ mineralization thickness

[0292] Radiation intensity ∝ 1 / (burial depth + d0) (d0 is a small constant to prevent division by zero)

[0293] Mathematical expression

[0294]

[0295] I: Virtual drill radiation intensity (any unit, linearly comparable to the "radiation flux value" in the report)

[0296] C: Predicted grade (g / t)

[0297] T: Predicted mineralization thickness (m)

[0298] Z: Predicted burial depth (m)

[0299] α: Calibration coefficient (calculated by back-calculating from existing boreholes to minimize the RMS error between I and the measured flux value)

[0300] Global Fast Calculation Process

[0301] Prior model

[0302] Grade surface C(x,y): Interpolation of the grade of observed boreholes using indicator kriging or GPR.

[0303] Thickness surface T(x,y): Abnormal vertical extension = Length of abnormal depth interval × Probability weight

[0304] Burial depth Z(x,y): The depth of the abnormal top plate is taken (the report has given the range of 5–145 m).

[0305] Grid-by-grid computation

[0306] The grid step size is 25 m (consistent with the minimum spacing of actual boreholes).

[0307] Substituting each (xi, yj) into the above equation yields the two-dimensional matrix Iij.

[0308] Cross-validation with satellite radiative flux

[0309] Linear regression: Ivirt = k Isat + b

[0310] If R² ≥ 0.8, accept the formula; otherwise, rewind and recalibrate α or adjust the prior.

[0311] For the adaptive application of virtual radiation intensity in this scheme:

[0312] 1. We rewrite the POMDP reward function into a form that can be directly measured by virtual radiation intensity.

[0313]

[0314] β: Economic conversion factor

[0315] Cdrill: Drilling cost (RMB / hole)

[0316] When MCTS nodes are expanded

[0317] Instead of randomly rolling out "mineral found / not found", it directly reads I(a).

[0318] 2. We simplify our belief to: High virtual strength → High probability of finding minerals.

[0319]

[0320] This eliminates the need for each "sampling-forward modeling" step, enabling a single preprocessing step followed by second-level optimization across the entire domain.

[0321] Finally, the "Ultimate Drilling Ranking" is output in one go.

[0322] Algorithm objective:

[0323]

[0324] step

[0325] Global computation I(x,y)

[0326] Sort by I in descending order and select the top 3 × N candidate holes.

[0327] Use MCTS (or simply a greedy algorithm with distance constraints) to further refine the selection of N elements, ensuring the space does not cluster.

[0328] Output the final Top-N hole positions, expected profit, and cumulative metal quantity.

[0329] Let's take one example to illustrate:

[0330] (Already used α=1, β=0.3 million yuan·t) -1 Cdrill = 50,000 yuan, grid 25 m, distance ≥ 100 m (cluster removal)

[0331]

[0332] From a hardware perspective, in order to improve the efficiency and accuracy of real drilling, this application provides an embodiment of an electronic device for implementing all or part of the resource exploration method based on virtual drilling, wherein the electronic device specifically includes the following:

[0333] The system comprises a processor, memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to realize information transmission between the virtual drilling-based resource exploration method and core business systems, user terminals, and related databases and other related equipment; the logic controller can be a desktop computer, tablet computer, or mobile terminal, etc., and this embodiment is not limited to these. In this embodiment, the logic controller can be implemented with reference to the embodiments of the virtual drilling-based resource exploration method in the present embodiment, and the contents of the embodiments of the virtual drilling-based resource exploration method are incorporated herein, and repeated details will not be described again.

[0334] It is understood that the user terminal may include smartphones, tablet computers, network set-top boxes, portable computers, desktop computers, personal digital assistants (PDAs), in-vehicle devices, smart wearable devices, etc. Among these, the smart wearable devices may include smart glasses, smartwatches, smart bracelets, etc.

[0335] In practical applications, parts of the virtual drilling-based resource exploration method can be executed on the electronic device side as described above, or all operations can be completed in the client device. The choice can be made based on the processing power of the client device and the limitations of the user's usage scenario. This application does not impose any limitations on this. If all operations are completed in the client device, the client device may further include a processor.

[0336] The aforementioned client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side; in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a distributed server structure.

[0337] Figure 3 This is a schematic block diagram illustrating the system configuration of the electronic device 9600 according to an embodiment of this application. Figure 3 As shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It is worth noting that... Figure 3 This is an example; other types of structures can also be used to supplement or replace this structure to achieve telecommunications functions or other functions.

[0338] In one embodiment, the resource exploration method based on virtual drilling can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following controls:

[0339] Step S101: Obtain historical exploration data of the target area, discretize the target area into a regular grid model, and associate the historical exploration data with the regular grid model. The historical exploration data includes borehole data and geological attribute information of the target resource. The borehole data includes borehole location and borehole results.

[0340] Step S102: In the regular grid model, geological attribute dependency calculation is performed based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted based on the dependency value to determine the corresponding initial belief map. The initial belief map represents the prior probability or attribute strength of the existence of target resources in each grid.

[0341] Step S103: Determine the corresponding state space based on the initial belief map, determine the corresponding action space based on the candidate borehole location set in the regular grid model, determine the corresponding observation space based on the drilling results, determine the corresponding reward function based on the cost-effectiveness of drilling, determine the corresponding Markov decision process model based on the state space, the action space, the observation space, and the reward function, input the initial belief map and the Markov decision process model into the set Monte Carlo tree search algorithm to perform virtual drilling simulation, and determine the corresponding recommended drilling layout scheme.

[0342] As described above, the electronic device provided in this application acquires the range and historical exploration data of the target area, discretizes the target area into a regular grid model, and associates the historical exploration data with the regular grid model. In the regular grid model, the geological attribute dependency is calculated based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted according to the dependency value to determine the corresponding initial belief map. A Markov decision process model is constructed based on the initial belief map. The initial belief map and the Markov decision process model are input into a set Monte Carlo tree search algorithm to perform virtual drilling simulation and determine the corresponding recommended drilling scheme, thereby improving the efficiency and accuracy of real drilling.

[0343] In another embodiment, the virtual drilling-based resource exploration method can be configured separately from the central processing unit 9100. For example, the virtual drilling-based resource exploration method can be configured as a chip connected to the central processing unit 9100, and the function of the virtual drilling-based resource exploration method can be realized through the control of the central processing unit.

[0344] like Figure 3 As shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily need to include these components. Figure 3 All components shown; in addition, the electronic device 9600 may also include Figure 3 For components not shown, please refer to existing technologies.

[0345] like Figure 3 As shown, the central processing unit 9100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device, which receives inputs and controls the operation of various components of the electronic device 9600.

[0346] The memory 9140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 9100 may execute the program stored in the memory 9140 to perform information storage or processing, etc.

[0347] Input unit 9120 provides input to central processing unit 9100. Input unit 9120 may be, for example, a keypad or touch input device. Power supply 9170 provides power to electronic device 9600. Display 9160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0348] The memory 9140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs or processes for executing the operation of the electronic device 9600 via the central processing unit 9100.

[0349] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for the electronic device for communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0350] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which is the same as in a conventional mobile communication terminal.

[0351] Based on different communication technologies, multiple communication modules 9110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby realizing typical telecommunications functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. Furthermore, the audio processor 9130 is also coupled to a central processing unit 9100, enabling on-device recording via the microphone 9132 and on-device playback of stored sound via the speaker 9131.

[0352] Embodiments of this application also provide a computer-readable storage medium capable of implementing all steps of the virtual drilling-based resource exploration method with a server or client as the execution subject in the above embodiments. The computer-readable storage medium stores a computer program that, when executed by a processor, implements all steps of the virtual drilling-based resource exploration method with a server or client as the execution subject in the above embodiments. For example, when the processor executes the computer program, it implements the following steps:

[0353] Step S101: Obtain historical exploration data of the target area, discretize the target area into a regular grid model, and associate the historical exploration data with the regular grid model. The historical exploration data includes borehole data and geological attribute information of the target resource. The borehole data includes borehole location and borehole results.

[0354] Step S102: In the regular grid model, geological attribute dependency calculation is performed based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted based on the dependency value to determine the corresponding initial belief map. The initial belief map represents the prior probability or attribute strength of the existence of target resources in each grid.

[0355] Step S103: Determine the corresponding state space based on the initial belief map, determine the corresponding action space based on the candidate borehole location set in the regular grid model, determine the corresponding observation space based on the drilling results, determine the corresponding reward function based on the cost-effectiveness of drilling, determine the corresponding Markov decision process model based on the state space, the action space, the observation space, and the reward function, input the initial belief map and the Markov decision process model into the set Monte Carlo tree search algorithm to perform virtual drilling simulation, and determine the corresponding recommended drilling layout scheme.

[0356] As described above, the computer-readable storage medium provided in this application embodiment acquires the range and historical exploration data of the target area, discretizes the target area into a regular grid model, and associates the historical exploration data with the regular grid model. In the regular grid model, the geological attribute dependency is calculated based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted according to the dependency value to determine the corresponding initial belief map. A Markov decision process model is constructed based on the initial belief map. The initial belief map and the Markov decision process model are input into a set Monte Carlo tree search algorithm for virtual drilling simulation to determine the corresponding recommended drilling scheme, thereby improving the efficiency and accuracy of real drilling.

[0357] Embodiments of this application also provide a computer program product capable of implementing all steps in the virtual drilling-based resource exploration method described above, where the execution subject is a server or client. When executed by a processor, this computer program / instruction implements the steps of the virtual drilling-based resource exploration method. For example, the computer program / instruction implements the following steps:

[0358] Step S101: Obtain historical exploration data of the target area, discretize the target area into a regular grid model, and associate the historical exploration data with the regular grid model. The historical exploration data includes borehole data and geological attribute information of the target resource. The borehole data includes borehole location and borehole results.

[0359] Step S102: In the regular grid model, geological attribute dependency calculation is performed based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted based on the dependency value to determine the corresponding initial belief map. The initial belief map represents the prior probability or attribute strength of the existence of target resources in each grid.

[0360] Step S103: Determine the corresponding state space based on the initial belief map, determine the corresponding action space based on the candidate borehole location set in the regular grid model, determine the corresponding observation space based on the drilling results, determine the corresponding reward function based on the cost-effectiveness of drilling, determine the corresponding Markov decision process model based on the state space, the action space, the observation space, and the reward function, input the initial belief map and the Markov decision process model into the set Monte Carlo tree search algorithm to perform virtual drilling simulation, and determine the corresponding recommended drilling layout scheme.

[0361] As described above, the computer program product provided in this application obtains the scope and historical exploration data of the target area, discretizes the target area into a regular grid model, and associates the historical exploration data with the regular grid model. In the regular grid model, the geological attribute dependency is calculated based on the geological attribute information between each known grid to determine the corresponding dependency value. The geological attribute information of the unknown grid in the regular grid model is weighted according to the dependency value to determine the corresponding initial belief map. A Markov decision process model is constructed based on the initial belief map. The initial belief map and the Markov decision process model are input into a set Monte Carlo tree search algorithm to perform virtual drilling simulation and determine the corresponding recommended drilling scheme, thereby improving the efficiency and accuracy of real drilling.

[0362] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0363] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0364] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0365] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0366] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for resource exploration based on virtual drilling, characterized in that, The method comprises: acquiring historical exploration data of a target area, discretizing the target area into a regular grid model, and associating the historical exploration data with the regular grid model, wherein the historical exploration data comprises drilling data and target resource geological attribute information, and the drilling data comprises drilling positions and drilling results; in the regular grid model, performing geological attribute dependency calculation according to the geological attribute information between each known grid, determining a corresponding dependency relationship value, performing weighted calculation on the geological attribute information of unknown grids in the regular grid model according to the dependency relationship value, and determining a corresponding initial belief graph, wherein the initial belief graph represents a prior probability or attribute intensity of the existence of a target resource in each grid; determining a corresponding state space according to the initial belief graph, determining a corresponding action space according to a set of candidate drilling positions in the regular grid model, determining a corresponding observation space according to drilling results, determining a corresponding reward function based on the cost-effectiveness of drilling, determining a corresponding Markov decision process model according to the state space, the action space, the observation space, and the reward function, initializing a root node of a preset search tree according to the belief graph, determining a corresponding Monte Carlo tree search algorithm, wherein the Monte Carlo tree search algorithm adopts UCT scoring and constructs development items and exploration items, and the development items and the exploration items are used to balance exploration and development; inputting the initial belief graph and the Markov decision process model into the Monte Carlo tree search algorithm, selecting a candidate drilling position from the action space of the Markov decision process model as a current simulation action by the Monte Carlo tree search algorithm, and performing simulation sampling according to the initial belief graph to determine a corresponding geological realization of a target area resource distribution; performing forward calculation on a drilling position corresponding to the current simulation action in the geological realization of the target area resource distribution to determine a corresponding virtual observation result; determining a corresponding expected cumulative reward of a candidate drilling according to the virtual observation result, outputting one or more candidate drillings with the highest expected cumulative reward, and determining a corresponding recommended drilling scheme according to the output candidate drillings.

2. The virtual drilling based resource exploration method as claimed in claim 1, wherein, The acquiring historical exploration data of a target area, discretizing the target area into a regular grid model, and associating the historical exploration data with the regular grid model comprises: acquiring historical exploration data of a target area, wherein the historical exploration data comprises drilling data and target resource geological attribute information, and the drilling data comprises drilling positions and drilling results representing the presence or absence of ore; discretizing the target area into a regular grid model according to a preset resolution; mapping the drilling positions to corresponding grid cells in the regular grid model, and marking the drilling results and the geological attribute information of the grid cells as known data.

3. The virtual drilling based resource exploration method of claim 2, wherein, The mapping the drilling positions to corresponding grid cells in the regular grid model, and marking the drilling results and the geological attribute information of the grid cells as known data comprises: constructing a two-dimensional matrix with the same dimension as the regular grid model; For each grid cell, if it contains the drilling position, fill in the drilling result and geological attribute information of the drilling position in the corresponding position of the matrix; If it does not contain the drilling position, fill in the null identifier in the corresponding position of the matrix to identify the position as an unexplored area.

4. The virtual drilling based resource exploration method as claimed in claim 1, wherein, The geological attribute dependency calculation in the regular grid model according to the geological attribute information between each known grid point includes: In the regular grid model, all known grid point pairs are traversed, the spatial distance and the semi-variance of the geological attribute value between each pair of grid points are calculated, and the corresponding geological attribute scatter plot is determined; According to the model in the preset variogram model library, the geological attribute scatter plot is fitted, and according to the fitted variogram model, the dependency relationship value between each grid point pair is determined, and the dependency relationship value is used to represent the spatial dependency of the geological attribute between the grids.

5. The virtual drilling based resource exploration method as claimed in claim 1, wherein, The geological attribute dependency calculation in the regular grid model according to the geological attribute information between each known grid point also includes: In the regular grid model, according to the prior distribution characteristics of the geological attribute information, a preset covariance function is selected from a kernel function library as a core function of a Gaussian process prior; The hyperparameters of the covariance function are optimized by using the geological attribute data between the known grid points, and the dependency relationship value between each grid point pair is determined according to the optimized covariance function, and the hyperparameters include at least one of the variance scale and the length scale.

6. The virtual drilling based resource exploration method as claimed in claim 1, wherein, The weighted calculation of the geological attribute information of the unknown grid in the regular grid model according to the dependency relationship value includes: For any unknown grid point in the regular grid model, all known data points in the search neighborhood range of the unknown grid point are determined, and a Kriging equation set is constructed based on the dependency relationship value between the known points and solved to obtain the Kriging weight of each known data point in the search neighborhood. The geological attribute information of all known data points in the search range of the unknown grid point is multiplied by the corresponding Kriging weight and summed to calculate the geological attribute information of the unknown grid point, and the initial belief map corresponding to all unknown grid points is determined according to the geological attribute information of the unknown grid points.

7. A virtual drilling based resource exploration apparatus, characterized by, The device includes: A target area grid determination module is configured to obtain historical exploration data of a target area, discretize the target area into a regular grid model, and associate the historical exploration data with the regular grid model, wherein the historical exploration data includes drilling data and target resource geological attribute information, and the drilling data includes drilling positions and drilling results; The device includes: An initial belief map determination module is configured to determine a corresponding dependency relationship value by calculating a geological attribute dependency according to the geological attribute information between each known grid in the rule grid model, and to determine a corresponding initial belief map by performing a weighted calculation on the geological attribute information of an unknown grid in the rule grid model according to the dependency relationship value, wherein the initial belief map represents a prior probability or attribute strength of the existence of a target resource in each grid. A virtual drilling module is configured to determine a corresponding state space according to the initial belief map, to determine a corresponding action space according to a candidate drilling position set in the rule grid model, to determine a corresponding observation space according to a drilling result, to determine a corresponding reward function based on the cost-effectiveness of drilling, to determine a corresponding Markov decision process model according to the state space, the action space, the observation space, and the reward function, to initialize a root node of a preset search tree according to the belief map, to determine a corresponding Monte Carlo tree search algorithm, wherein the Monte Carlo tree search algorithm adopts a UCT score and constructs exploitation items and exploration items, and the exploitation items and the exploration items are used to balance exploration and exploitation, to input the initial belief map and the Markov decision process model into the Monte Carlo tree search algorithm, to select a candidate drilling position from the action space of the Markov decision process model as a current simulation action by the Monte Carlo tree search algorithm, and to determine a corresponding geological realization of a target area resource distribution by performing simulation sampling according to the initial belief map, to perform forward calculation on a drilling position corresponding to the current simulation action in the geological realization of the target area resource distribution to determine a corresponding virtual observation result, to determine a corresponding expected cumulative reward of a candidate drilling according to the virtual observation result, to output one or more candidate drillings with the highest expected cumulative reward, and to determine a corresponding recommended drilling plan according to the output candidate drillings.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the steps of the resource exploration method based on virtual drilling according to any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the resource exploration method based on virtual drilling according to any one of claims 1 to 6.

Images