Coal mine geological model dynamic updating method based on drilling data and related equipment

Abstract

The application relates to the technical field of geological data processing and prediction, and provides a coal mine geological model dynamic updating method based on drilling data and related equipment. Drilling data generated by an intelligent drilling machine in a drilling process is acquired, pretreated and feature-extracted to generate drilling response sequences indexed by drilling depths; data analysis is performed to generate a geological response feature vector composed of multiple statistical features; the geological response feature vector is subjected to data analysis and sampling processing by a multilayer Fourier neural operator to generate an initial calculation result; the continuity and consistency of the result are optimized by sequence recalibration technology to generate a high-credibility virtual geological profile, and finally, only a local spatial area affected by the drilling hole is subjected to accurate and incremental model updating based on the profile, so that the updating efficiency is greatly improved while the accuracy is ensured, the geological model can dynamically evolve with the drilling, and the urgent demand of a production site for real-time geological information is met.

Description

Technical Field

[0001] This application relates to the field of geological data processing and prediction technology, and more specifically, to a method and related equipment for dynamic updating of coal mine geological models based on drilling data. Background Technology

[0002] With the continuous advancement of smart mine construction, transparent geology is gradually becoming an important foundation for ensuring safe and efficient coal mining. Currently, three-dimensional geological models for coal mines are mainly constructed based on data obtained during the exploration phase, such as borehole and geophysical data. This type of data is usually sparsely distributed spatially, resulting in limited accuracy of the geological models and long update cycles. In actual production, as tunnels are excavated or working faces advance, geological structural changes such as small faults, changes in coal seam thickness, or weak interbedded rock, which were not accurately predicted during the exploration phase, are often encountered. Due to the lagging updates of existing geological models, they cannot reflect these local geological changes in a timely manner, which can easily have an adverse impact on on-site construction decisions and safe production.

[0003] In recent years, intelligent drilling rigs have been widely used in underground coal mines. During drilling, they can collect various drilling data in real time. However, current technologies primarily utilize drilling data for data display, recording and storage, or simple threshold alarms, and have not yet developed a systematic method to effectively combine real-time drilling data with three-dimensional geological models. Furthermore, traditional three-dimensional geological models are typically large in size. When new geological information is introduced, the model often needs to be reconstructed entirely or updated on a large scale, resulting in high computational costs and low update efficiency, making it difficult to meet the demands for real-time and locally refined updates of geological information during underground production. Summary of the Invention

[0004] This application provides a method and related equipment for dynamic updating of coal mine geological models based on drilling data, which can at least partially solve the problem of low updating efficiency of local three-dimensional geological models in coal mines, making it difficult to meet the needs of real-time and local fine-grained updating of geological information during underground production.

[0005] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part by practice of this application.

[0006] According to one aspect of this application, a method for dynamically updating a coal mine geological model based on drilling data is provided, comprising: acquiring drilling data generated by an intelligent drilling rig during drilling; preprocessing and extracting features from the drilling data to generate a drilling response sequence indexed by drilling depth; performing data analysis on the drilling response sequence to generate a geological response feature vector composed of multiple statistical features; performing data analysis and sampling processing on the geological response feature vector using a multi-layer Fourier neural operator to generate an initial estimation result; recalibrating the initial estimation result to generate a virtual geological profile corresponding to each drilling depth along the borehole trajectory; and updating the local spatial area affected by the current drilling activity based on the virtual geological profile to generate a coal mine geological model.

[0007] In this application, based on the aforementioned scheme, the drilling data includes drilling pressure, drilling torque, drill rod rotation speed, pump pressure, and drilling depth generated at each sampling time.

[0008] In this application, based on the aforementioned scheme, the step of preprocessing and feature extraction of the drilling data to generate a drilling response sequence indexed by drilling depth includes: representing the drilling data as a signal vector of a multidimensional drilling response that varies with time; smoothing the parameter components in the signal vector to generate a first signal; and mapping the first signal into a drilling response sequence indexed by drilling depth based on the mapping relationship between drilling time and drilling depth.

[0009] In this application, based on the aforementioned scheme, the step of performing data analysis on the drilling response sequence to generate a geological response feature vector composed of multiple statistical features includes: combining the drilling data to generate an original feature vector; calculating the statistical features of each parameter in the drilling response sequence at a preset depth window scale; generating drilling energy features and energy change at a preset depth based on each parameter in the drilling response sequence; and combining the original feature vector, the statistical features, the drilling energy features, and the energy change to generate a geological response feature vector.

[0010] In this application, based on the aforementioned scheme, the step of performing data analysis and sampling processing on the geological response feature vector using a multi-layer Fourier neural operator to generate an initial inference result includes: performing a layer-by-layer transformation on the geological response feature vector using a multi-layer Fourier neural operator to generate intermediate features; inputting the intermediate features into a pre-trained geological attribute operator inference model to output a continuous prediction result representing the geological attributes; and sampling the continuous prediction result based on a preset drilling depth to generate an initial inference result.

[0011] In this application, based on the aforementioned scheme, the step of recalibrating the initial estimation results to generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory includes: sorting and combining the initial estimation results according to the increasing drilling depth to generate a depth sequence; encoding and linearly projecting the depth sequence to generate sequence-aware features; and linearly projecting the sequence-aware features to generate recalibrated results corresponding to each drilling depth, which serve as virtual geological profiles along the borehole trajectory.

[0012] In this application, based on the aforementioned scheme, the step of updating the local spatial region affected by the current drilling activity based on the virtual geological profile to generate a coal mine geological model includes: determining the local spatial region affected by the current drilling activity based on the spatial location in the three-dimensional geological model and the sequence of spatial sampling points formed along the drilling trajectory; and incrementally updating the local spatial region based on the virtual geological profile to generate a coal mine geological model.

[0013] In this application, based on the aforementioned scheme, after updating the local spatial area affected by the current drilling activity based on the virtual geological profile to generate a coal mine geological model, the method further includes: using the coal mine geological model as a new prior model, and iteratively updating the prior model based on the newly generated virtual geological profile during subsequent drilling processes.

[0014] According to one aspect of this application, a dynamic updating device for a coal mine geological model based on drilling data is provided, comprising:

[0015] The acquisition module is used to acquire drilling data generated by the intelligent drilling rig during the drilling process;

[0016] The extraction module is used to preprocess and extract features from the drilling data to generate a drilling response sequence indexed by the drilling depth.

[0017] The statistics module is used to perform data analysis on the drilling response sequence and generate a geological response feature vector composed of multiple statistical features;

[0018] The estimation module is used to perform data analysis and sampling processing on the geological response feature vector through multi-layer Fourier neural operators to generate initial estimation results;

[0019] The calibration module is used to recalibrate the initial calculation results and generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory.

[0020] The update module is used to update the local spatial area affected by the current drilling activity based on the virtual geological profile, and generate a coal mine geological model.

[0021] According to one aspect of this application, a computer-readable medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method for dynamically updating a coal mine geological model based on drilling data as described in the above embodiments.

[0022] According to one aspect of this application, an electronic device is provided, comprising: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the dynamic update method for a coal mine geological model based on drilling data as described in the above embodiments.

[0023] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the dynamic updating method for a coal mine geological model based on drilling data provided in the various optional implementations described above.

[0024] The main differences and technical effects of the technical solution of this application compared with the prior art are as follows:

[0025] On the one hand, by using real-time drilling data from intelligent drilling rigs as direct input, the geological information acquisition process is moved forward to the drilling process itself. This enables a shift from static modeling after drilling is completed to dynamic sensing and updating during drilling, based on the Industrial Internet of Things (IIoT). This allows the 3D geological model to continuously track the progress of mining operations, significantly improving the timeliness of geological information. By adaptively determining the borehole's influence range, incremental updates are only performed on local geological areas affected by drilling activities, avoiding repeated reconstruction of the global geological model. This effectively reduces computational complexity and resource consumption while ensuring model accuracy, meeting the actual needs of downhole production environments for rapid response and lightweight computing.

[0026] On the other hand, by introducing a drilling response-geological attribute modeling method based on artificial intelligence deep learning, the nonlinear correlation between drilling data and geological attributes is automatically mined, enabling intelligent inference and evaluation of geological attributes. This reduces reliance on human experience interpretation and improves the objectivity, consistency, and repeatability of geological prediction results. By using each local geological model update as a priori basis for subsequent drilling and model correction, a continuously evolving closed-loop update process is formed. This allows the 3D geological model to continuously absorb new drilling observation information, gradually approximating the real geological structure and enhancing the model's adaptability under complex geological conditions.

[0027] The technical solution proposed in this application can provide real-time and detailed local geological transparency within the mining face, and intuitively present key geological information such as faults, coal thickness changes and lithological abrupt changes. It provides a reliable basis for the layout of gas drainage holes, roadway support design and mining process adjustment, directly serves on-site construction decision-making and safety production management, and helps to improve the level of coal mine production safety and comprehensive management efficiency.

[0028] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0029] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0030] Figure 1 The flowchart illustrating a method for dynamically updating a coal mine geological model based on drilling data is shown in one embodiment of this application.

[0031] Figure 2 This illustration shows a schematic diagram of the incremental and closed-loop update process of the local transparent geological model in the borehole trajectory influence domain in one embodiment of this application.

[0032] Figure 3 The illustration shows a schematic diagram of the process of constructing a gas extraction hole in the roof of a coal mine using an intelligent drilling rig in one embodiment of this application.

[0033] Figure 4 The illustration shows a schematic diagram of a dynamic updating device for a coal mine geological model based on drilling data in one embodiment of this application.

[0034] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown. Detailed Implementation

[0035] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art.

[0036] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0037] It should be noted that the data acquisition or information collection in this embodiment is performed after authorization by the user or the object of collection, and its process and purpose strictly follow the relevant regulations.

[0038] The block diagrams shown in the attached figures are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, or in one or more hardware modules composed of smart chips, smart integrated circuits, or application-specific integrated circuits (ASICs), or in different network and / or processor devices and / or microcontroller devices.

[0039] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily need to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0040] The implementation details of the technical solution of this application are described below:

[0041] Figure 1 A flowchart illustrating a method for dynamically updating a coal mine geological model based on drilling data, according to an embodiment of this application, is shown. (Refer to...) Figure 1 As shown, the dynamic updating method for coal mine geological models based on drilling data includes at least steps S110 to S160, which are described in detail below:

[0042] S110 acquires drilling data generated by the intelligent drilling rig during the drilling process.

[0043] In this embodiment, it is achieved through an intelligent drilling rig system deployed in the underground mining face of a coal mine. The intelligent drilling rig system integrates a high-precision sensor array, an edge computing unit, and a data communication module. It is used to synchronously and continuously collect multi-dimensional drilling data reflecting the interaction between the drilling rig and the formation during the process of the drill bit breaking the strata. This data forms the original data source for the dynamic updating of the geological model, serving as the drilling data.

[0044] The technical solution of this application can be used for new technology services in fields such as coal mining and petroleum, such as mineral geological exploration services using high technology, as well as other natural science research and experimental development.

[0045] Specifically, the drilling data in this embodiment includes, but is not limited to, data such as drilling pressure, drilling torque, drill rod rotation speed, pump pressure, and drilling depth generated at each sampling time. All of the collected drilling data can be stored as industrial big data on a cloud platform, enabling the storage and management of industrial big data.

[0046] Specifically, the sensor array equipped with the intelligent drilling rig includes: a drill pressure sensor installed on the feed mechanism for real-time measurement of the axial pressure (F) of the drill bit; a torque sensor (T) and a speed encoder (RPM) integrated into the rotary power head for measuring the drive torque and drill rod rotation speed, respectively; a pump pressure sensor (P_p) installed on the flushing fluid circulation pipeline for monitoring the borehole pressure; and a depth measurement unit (D) connected to the feed device for accurately recording the real-time drilling depth of the drill bit from the borehole opening. Preferably, the system can also obtain the mechanical drilling rate (ROP) through depth signal differentiation or direct measurement.

[0047] During drilling operations, each intelligent sensor synchronously acquires analog signals at a preset fixed sampling frequency (e.g., 10 Hz to 100 Hz) through its built-in sensor chip. After the signal is amplified and preliminarily filtered by the onboard signal conditioning circuit, it is converted into a digital signal by the analog-to-digital converter. The edge computing controller built into the intelligent drilling rig receives the data from each channel and aligns and packages it based on a unified time reference to form a raw drilling data vector x_raw(t) = [F(t), T(t), RPM(t), P_p(t), D(t), ...] organized according to the sampling time (t).

[0048] Subsequently, the edge computing unit transmits the packaged real-time data stream to the data aggregation node located at the underground control station or the ground server via the mine's industrial ring network or a dedicated wireless transmission link. The data transmission process employs reliable communication protocols or mechanisms, such as IoT secure communication protocols or industrial internet-based secure communication protocols, to ensure data integrity, timeliness, and low latency.

[0049] Before the data is used in subsequent steps, the system performs preliminary data quality control, including but not limited to: physical mechanism-based parameter value range verification, monotonicity checks of the drilling depth sequence, and labeling and filtering of transient outliers. Data streams that pass quality verification are stored in a time-series database, providing standardized access interfaces for subsequent data preprocessing and feature extraction modules.

[0050] Through the above implementation methods, this application ensures that multi-dimensional drilling data streams characterizing formation mechanical properties can be acquired in real time and reliably, providing an accurate and continuous data input foundation for subsequent data-driven dynamic inference of geological attributes and model updates.

[0051] S120, the drilling data is preprocessed and features are extracted to generate a drilling response sequence indexed by the drilling depth.

[0052] In this embodiment, the real-time acquired drilling data is preprocessed and features are extracted to construct a standardized drilling response sequence indexed by drilling depth. Specifically, the time-varying multidimensional drilling response signal is smoothed and filtered to suppress instantaneous noise caused by equipment disturbances or sensor anomalies, while retaining the continuous changing trend reflecting formation characteristics. Then, based on the mapping relationship between drilling time and drilling depth, the processed signal is converted from the time domain to the depth domain and aligned to a uniform depth interval through resampling. Finally, a continuous, consistent drilling response sequence with a clear spatial orientation along the borehole depth direction is formed, providing a structured input data foundation for subsequent analysis and extraction of geological response features.

[0053] In one embodiment of this application, the drilling data is preprocessed and feature extracted to generate a drilling response sequence indexed by drilling depth, including:

[0054] The drilling data is represented as a signal vector of a multidimensional drilling response that varies over time.

[0055] The parameter components in the signal vector are smoothed to generate a first signal;

[0056] Based on the mapping relationship between drilling time and drilling depth, the first signal is mapped into a drilling response sequence indexed by drilling depth.

[0057] During coal mine drilling operations, intelligent drilling rigs deployed underground collect multi-source drilling data on drilling pressure in real time. This drilling data reflects the mechanical and kinematic state of the drill bit interacting with the formation, and its changes are directly influenced by geological factors such as formation lithology, structural integrity, and degree of fragmentation. Therefore, the drilling data collected during the drilling process is represented as a signal vector of a multi-dimensional drilling response that varies over time. for:

[0058]

[0059] in, Indicates the sampling time of the drilling data; Indicates at time Drilling pressure at the location; Indicates at time Drilling torque at the point; Indicates at time The rotational speed of the drill pipe; Indicates at time Mechanical drilling speed at the location; Indicates at time Pump pressure at the location; Indicates at time The drilling depth corresponding to the drill bit.

[0060] In actual drilling operations, drilling data is not only affected by geological conditions but may also be interfered with by non-geological factors such as equipment start-up and shutdown, operational adjustments, and momentary sensor anomalies. To avoid interference from these non-geological disturbances on subsequent geological response analysis, this invention introduces a robust signal modeling algorithm to preprocess the drilling response signal. Within a time window... Within, for any drilling data component The first signal is obtained by smoothing the signal using a median-mean joint filtering method. for:

[0061]

[0062] in, Indicates the first Drilling data at time [time] The original value; This represents the filtered, robust drilling data, i.e., the first signal; Indicated by time The central time window; This indicates a value to be taken within the window. This indicates an operation to average the values ​​within the window; These are weighting coefficients used to balance the effects of median filtering and mean filtering. Through this process, transient anomalies caused by non-geological factors can be effectively suppressed, while preserving the continuous drilling response characteristics caused by changes in formation properties.

[0063] Since drilling proceeds gradually along the borehole depth, and geological models are typically constructed based on spatial coordinates, it is necessary to map the drilling response signal from the time domain to the depth domain. This is based on the drilling depth signal. The mapping relationship between drilling time and drilling depth is established as follows:

[0064]

[0065] Wherein, mapping function This represents the mapping relationship between drilling time and drilling depth, determined by the depth data recorded in real time by the drilling rig. Furthermore, the robust drilling response signal is resampled to a uniform depth step size. Construct a drilling response sequence indexed by depth. for:

[0066]

[0067] in, Indicates the drilling depth, i.e., the first... A discrete drilling depth position, where k represents the identifier of the drilling depth; Represents mapping function The inverse mapping; Indicates the depth at which the drill bit is located. The corresponding drilling response signal vector, i.e., the drilling response sequence.

[0068] Through the above-mentioned real-time drilling data stream acquisition and signal processing steps, standard drilling time-series data with unified structure and time and space alignment are formed, providing a stable data foundation for subsequent drilling response characteristic characterization and geological change perception.

[0069] S130, perform data analysis on the drilling response sequence to generate a geological response feature vector composed of multiple statistical features.

[0070] In this embodiment, a geological response feature vector that comprehensively reflects formation characteristics is constructed by performing multi-dimensional analysis on the drilling response sequence indexed by drilling depth. First, the drilling parameters are directly combined to form the original feature vector to retain the real-time status information of drilling. Then, at multiple preset depth window scales, the mean, variance, and other statistical characteristics of each parameter are calculated to characterize the continuous changing trend of the formation structure at different spatial scales. Simultaneously, the drilling energy characteristics per unit depth and their variation along the depth direction are calculated based on the drilling parameters to characterize the absolute magnitude and relative abrupt change of the formation's drill resistance. Finally, the original features, multi-scale statistical features, drilling energy characteristics, and their variations are fused to form a unified high-dimensional geological response feature vector, thereby providing the subsequent intelligent inference model with input features that are sensitive to geological changes and rich in information.

[0071] In one embodiment of this application, data analysis is performed on the drilling response sequence to generate a geological response feature vector composed of multiple statistical features, including:

[0072] The drilling data is combined to generate an original feature vector;

[0073] Calculate the statistical characteristics of each parameter in the drilling response sequence under a preset depth window scale;

[0074] Based on the parameters in the drilling response sequence, the drilling energy characteristics and energy change at the preset depth are generated.

[0075] The original feature vector, the statistical features, the drilling energy features, and the energy change are combined to generate a geological response feature vector.

[0076] In this embodiment, based on the robust, depth-aligned drilling response sequence obtained in the aforementioned steps, a drilling response feature that can sensitively reflect changes in the physical and mechanical properties of the formation is further constructed to characterize the differences in drilling behavior of the drill bit under different geological conditions.

[0077] At drilling depth At this point, the original feature vector generated during the initial drilling process is defined as:

[0078]

[0079] in, For drilling depth The original drilling response feature vector at the location, i.e., the original feature vector; They represent the depths respectively. The data includes drilling pressure, torque, rotational speed, mechanical drilling speed, and pump pressure after robust filtering.

[0080] Considering that geological structural changes at different scales will affect drilling response at different spatial scales, statistical response characteristics are constructed at multiple depth window scales. Based on depth... Centered on, window length is Depth window Inside, for the first The statistical characteristics of each drilling data point are calculated as follows:

[0081]

[0082] in, Represents different depth window scales; Indicates the first Depth window at various scales; Indicates the window length at the corresponding scale; Indicates the number of sample points within the window; and They represent the first The parameters in depth The data before and after processing; , They represent the first The parameters in the scale The mean and variance are calculated. By introducing multi-scale statistical features, it is possible to effectively distinguish between short-term fluctuations caused by random noise and persistent response changes caused by changes in stratigraphic structure.

[0083] To further characterize the relationship between input energy and formation resistance during drilling, a specific energy model from engineering mechanics is introduced to construct the drilling energy characteristics as follows:

[0084]

[0085] in, Indicates depth The specific energy at a given location, i.e., its energy characteristics; This represents the effective cross-sectional area of ​​the drill bit, used to normalize energy to a unit area and eliminate the influence of drill bit size. , , , These represent the drilling pressure, torque, rotational speed, and mechanical drilling speed at the corresponding depths, respectively.

[0086] Then, the energy change at adjacent depths is calculated as follows:

[0087]

[0088] in, This represents the change in specific energy between adjacent depth locations, reflecting abrupt changes in formation resistance. A significant increase in energy change usually corresponds to a harder formation, while a significant decrease in energy change usually corresponds to a softer formation.

[0089] Subsequently, the original response characteristics, multi-scale statistical characteristics, and drilling energy characteristics are fused to construct a unified geological response feature vector. for:

[0090]

[0091] in, It represents the original drilling response characteristics and reflects real-time drilling behavior; It represents multi-scale statistical characteristics, distinguishing noise from persistent changes in the formation; It represents the characteristics of drilling energy, and is expressed as the absolute and relative changes in formation resistance; Indicates depth The geological change-sensitive feature vector is used as input for subsequent dynamic inference of geological attributes and local geological model update algorithms.

[0092] By using the above-mentioned real-time characterization method of geological change sensitive features (geological response feature vector), the original drilling signal is transformed into a multi-level feature representation that is sensitive to geological changes, providing high-quality input features for subsequent preliminary inference models of geological attributes and sequence recalibration inference networks.

[0093] S140, the geological response feature vector is analyzed and sampled using a multi-layer Fourier neural operator to generate initial calculation results.

[0094] In this embodiment, the geological response feature vector is input into an inference model constructed based on multi-layer Fourier neural operators. This model first performs layer-by-layer nonlinear transformation and information fusion on the input features in the frequency and spatial domains through its multi-layered cascaded operator structure, effectively capturing and modeling the complex, particularly spatially correlated, global mapping relationship between drilling response and geological properties. Then, based on this learned relationship, the model outputs a geological property prediction curve continuously distributed along the borehole depth. Finally, this continuous prediction result is discretely sampled according to the actual drilling depth sequence, thereby generating initial inference results corresponding to each depth point, reflecting lithology or other key geological properties, providing basic inference data for subsequent sequence calibration steps.

[0095] In one embodiment of this application, the geological response feature vector is analyzed and sampled using a multi-layer Fourier neural operator to generate an initial estimation result, including:

[0096] The geological response feature vector is transformed layer by layer using a multi-layer Fourier neural operator to generate intermediate features;

[0097] The intermediate features are input into a pre-trained geological attribute operator inference model, which outputs continuous prediction results representing geological attributes.

[0098] The continuous prediction results are sampled based on a preset drilling depth to generate initial calculation results.

[0099] To achieve rapid initial inference of geological attributes in transparent geological modeling of coal mines, a preliminary inference method based on the Fourier Neural Operator (FNO) is designed, and a pre-trained geological attribute operator inference model is generated. The geological change sensitivity features (geological response feature vectors) calculated in the aforementioned steps are used as the basis for this method. As input, the spatial correlation features of geological data are captured through a global mapping from the frequency domain to the spatial domain, and the initial values ​​and confidence levels of coarse-precise geological attributes are output, providing a basis for subsequent sequence optimization.

[0100] The geological change sensitivity features obtained along the borehole trajectory are considered as multi-channel characteristic functions defined over the continuous drilling depth domain, denoted as... , representing the geological response feature vector The continuous feature field is reconstructed along the depth direction. The complex nonlinear relationship between drilling response features and geological properties is addressed by employing a deep learning-based deep neural network model as the geological property inference function. This model models the overall distribution of geological change-sensitive features along the depth direction, and the mapping relationship is expressed as follows:

[0101]

[0102] in, This indicates the continuous prediction results of geological properties along the borehole trajectory; The geological attribute operator inference model is represented by the FNO structure, and its parameter set is as follows: .

[0103] Subsequently, the model internally transforms the input feature field layer by layer through multi-layer FNO operator mapping, then the th The intermediate feature function of the layer is represented as:

[0104]

[0105] in, ; and These represent the learnable spectral weight matrix and the bias term, respectively. To represent a non-linear activation function, choose Sigmoid or its equivalent variant. This represents the number of layers in the FNO operator network. Its value can be set according to the complexity of the drilling data and the scale of geological structure changes. The initial value is set to 4.

[0106] in, This represents the Fourier integral operator, used to capture global spatial correlations in the frequency domain, and is implemented through the following process:

[0107]

[0108] in, and These represent the Fourier transform and the inverse Fourier transform, respectively. This represents element-wise multiplication.

[0109] Finally, the output layer will continuously predict the geological attributes. Sampling to discrete depth points The initial calculation results of geological properties were obtained. for:

[0110]

[0111] in, Indicates depth Sensitive feature vectors of geological changes at the location; For depth Preliminary inferences of the geological properties at the location; This represents the overall mapping relationship of the FNO model.

[0112] Optionally, considering the continuous spatial distribution characteristics of geological bodies, a depth direction continuity constraint is introduced during the geological attribute inference process to construct a joint optimization objective loss function:

[0113]

[0114] in, It represents the true geological properties; To predict and infer geological properties; This represents the prediction error term; Weights for continuity constraints; This is the prediction error term, representing the degree of fit between the model's predicted values ​​and the actual geological properties; As a continuity constraint term, it is weighted... By controlling the variation range of geological properties at adjacent depths, the results can better reflect the actual stratigraphic distribution.

[0115] Optionally, calculate the characteristic rate of change. for:

[0116]

[0117] in, When the value exceeds a preset threshold, the location is identified as a potential geological mutation point and used as a reference for subsequent local model partitioning updates.

[0118] Finally, the set of geological attributes G along the borehole trajectory is obtained as follows:

[0119]

[0120] in, A set of geological properties along the borehole trajectory; This is a preliminary inference of the geological properties at the corresponding depth; This represents the confidence level of the corresponding preliminary inference result. The higher the confidence level, the stronger the reliability of the inference. This set is directly used as the core input for sequence recalibration, providing basic constraints for accurate correction.

[0121] S150, the initial calculation results are recalibrated to generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory.

[0122] In this embodiment, the initial inference results generated in the aforementioned steps, distributed along the depth, are organized into a complete sequence according to their corresponding drilling depths in ascending order. This sequence simultaneously includes the attribute inference values ​​and their confidence information for each depth point. Subsequently, a recalibrated inference network based on an advanced sequence modeling architecture is used to encode and perform depth analysis on this sequence. This network can efficiently model and utilize the long-range dependencies and spatial continuity of geological attributes along the borehole trajectory to globally optimize and context-awarely correct the initial inference results. The network progressively extracts and fuses the correlation features between different depth points through its core sequence processing layer, generating an enhanced sequence-aware representation. Finally, this representation is linearly projected, directly outputting the precisely corrected geological attribute values ​​at each drilling depth, thereby forming a set of spatially continuous, logically consistent, and highly reliable virtual geological profiles, serving as the core constraint data driving the incremental updates of the local geological model.

[0123] In one embodiment of this application, the initial calculation results are recalibrated to generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory, including:

[0124] The initial calculation results are sorted and combined according to the increasing drilling depth to generate a depth sequence;

[0125] The depth sequence is encoded and linearly projected to generate sequence-aware features;

[0126] Linear projection is performed on the sequence sensing features to generate recalibration results corresponding to each drilling depth, which serve as virtual geological profiles along the borehole trajectory.

[0127] To correct potential local abrupt changes and sequence inconsistencies in preliminary inference results, and to fully utilize the long-range dependence and evolutionary continuity of geological attributes along the borehole depth direction, a recalibrated inference network based on the Mamba state-space structure, GeoMamba, is proposed. This network accurately calibrates the preliminary inference sequence, achieves global context modeling of long sequences with linear computational complexity, and outputs geological attribute results with high consistency and high reliability.

[0128] First, the preliminary inference set of geological attributes calculated and output in the above steps is as follows: According to drilling depth Organized in ascending order, the resulting multivariable depth sequence S is:

[0129]

[0130] in, , respectively representing the first Preliminary inferred values ​​and their confidence levels at various depth locations; K and K represent the identifier and total number of depth sampling points, respectively. The depth sequence generated through the above process directly reflects the evolution trend of geological attributes with depth and the reliability of the inference.

[0131] Then, the entire sequence is encoded using an Attribute Embedding Encoder (AEE) to output the sequence features of a single branch. for:

[0132]

[0133] in, Indicates attribute encoding, used to encode two-dimensional... Mapped to high-dimensional features; Used for dimensional splitting of high-dimensional features; Used for long-range dependency modeling of split features; Indicates linear projection, used to... The output data blocks are mapped back to a unified dimension; This indicates the output features.

[0134] Secondly, the data blocks are stacked to obtain the final sequence-aware features. for:

[0135]

[0136] In this context, superscripts (1), (2), and (3) indicate the 1st, 2nd, and 3rd layer stacks, respectively. For each data block, the core is the Mamba layer, combined with residual connections, to achieve state-space model (SSM) based capture. Long-range dependency, while outputting local enhancement features for:

[0137]

[0138] in, and These represent the current input features and the state at the previous time step, respectively. .

[0139] Finally, based on sequence-aware features Depth is generated through linear projection Recalibration results of geological properties at the location for:

[0140]

[0141] in, This represents the final geological attribute result obtained after recalibrating the inference network. It directly reflects the real stratigraphic attributes (such as lithology, coal seam thickness, etc.) at each depth along the borehole trajectory and is the core basic data for subsequent model updates.

[0142] Optionally, to support adaptive determination of subsequent local model update regions, the difference between the preliminary inference and the final calibration result is quantified. for:

[0143]

[0144] in, This represents the L2 norm difference between the initial inferred value and the recalibrated inferred value. The larger the value, the more significant the correction to the geological properties at that depth.

[0145] Finally, the calibration results and calibration differences at each depth are mapped to the three-dimensional spatial coordinates of the borehole, forming a virtual geological profile along the borehole trajectory. This profile serves as a high-confidence constraint for local geological model updates.

[0146]

[0147] in, Virtual geological profiles, as highly reliable geological constraint information, contain accurate geological attribute data and quantify the reliability of the data. They are directly input into the local implicit representation and incremental update steps of the three-dimensional geological model to drive the dynamic correction of the subsequent locally transparent geological model.

[0148] S160, Based on the virtual geological profile, the local spatial area affected by the current drilling activity is updated to generate a coal mine geological model.

[0149] In this embodiment, firstly, based on the three-dimensional spatial coordinates corresponding to each sampling point in the virtual geological profile, a local spatial region affected by the current drilling activity is adaptively determined along the borehole trajectory. The extent of this region is dynamically adjusted according to the significance of changes in geological attributes. Subsequently, within this local spatial region, the high-confidence geological attribute observation data provided by the virtual geological profile is fused with the prior information of the existing global three-dimensional geological model in this region. By introducing adaptive weights related to spatial distance and observation confidence, the original model is refined and incrementally corrected and updated. At the same time, spatial continuity constraints are applied to ensure the rationality of the updated geological structure. Finally, a coal mine geological model that reflects the latest geological knowledge is generated, which is updated only in the local spatial region while the rest remains unchanged. This updated model will serve as new prior knowledge to support continuous iterative updates in the subsequent drilling process, thus forming a dynamic evolutionary closed loop.

[0150] like Figure 2As shown, in this embodiment, a virtual geological profile along the borehole trajectory is obtained. Subsequently, based on the existing three-dimensional geological model, the virtual geological profile generated in real time during drilling is used as a new high-confidence geological constraint to construct local observation correction terms. Local incremental updates are only performed on the local update areas affected by the current drilling activities. By introducing locality, incrementality and closed-loop feedback mechanisms in the model update process, the three-dimensional geological model can continuously evolve with the progress of drilling operations, and finally realize the dynamic and transparent expression of the local geological structure of the coal mine. The update results are used as prior inputs for the next round of drilling work.

[0151] In one embodiment of this application, based on the virtual geological profile, a local spatial region affected by the current drilling activity is updated to generate a coal mine geological model, including:

[0152] Based on the spatial location in the three-dimensional geological model and the sequence of spatial sampling points formed along the drilling trajectory, the local spatial area affected by the current drilling activity is determined.

[0153] Based on the virtual geological profile, incremental updates are performed on the local spatial region to generate a coal mine geological model.

[0154] The impact of drilling operations on the formation structure exhibits significant spatial locality. Firstly, based on virtual geological profiles... Adaptively determine the local model region that needs updating, represented by the spatial sampling point sequence formed along the borehole trajectory. .in, It is derived from borehole geometry information (spatial coordinates) ) and depth Mapped three-dimensional spatial position , It is a planar coordinate system. It's about depth. Based on this spatial sampling point sequence, the local model update region is defined as a set of points that satisfy spatial proximity relationships, generating the local spatial region affected by the borehole data. for:

[0155]

[0156] in, Represents the set of locations in the entire three-dimensional space. Represents any spatial location in a three-dimensional geological model; For depth point The corresponding radius of influence is determined by the magnitude of changes in geological attributes within the virtual geological profile. Adaptive settings The larger the value, the larger the radius of influence. The range of values ​​is 5~20m to ensure that the model update coverage of the geological change area is reasonable.

[0157] After determining the local update area, the geological attribute information in the virtual geological profile will be... Mapped into a local three-dimensional space, corresponding to the original three-dimensional geological model in the same spatial position. Geological attribute prediction results at the location By combining the results and constructing the difference, the local observation correction term is obtained. Based on this, the geological model within the local spatial region is incrementally updated:

[0158]

[0159] in, Three-dimensional spatial coordinates; To adaptively update the weights, their values ​​are related to spatial location. The distance to the borehole trajectory is related to the corresponding virtual geological attributes. The closer the distance, the higher the credibility and the greater the weight. This is used to balance the influence of the prior information of the original model and the new observation information. Representative of virtual geological profiles Provided observation correction terms; To determine the spatial location of the existing three-dimensional geological model Geological attribute prediction results at the location; The updated 3D geological model for a local spatial region in spatial location The geological properties of the location.

[0160] The update process operates only on a local spatial region, while the original model remains unchanged in other regions, thus achieving local incremental updates. To ensure the rationality and continuity of the updated local geological model in terms of spatial structure, spatial consistency constraints are introduced within the local spatial region. Based on the K-nearest neighbor relationship of the 3D mesh, a set of spatial adjacency relationships is constructed in the local spatial region. And define space consistency constraints. for:

[0161]

[0162] in, For local spatial regions The set of spatial adjacency relationships is constructed based on the K-nearest neighbor relationship of the 3D mesh, that is, the pairs of adjacent spatial points in the 3D model; , This represents the geological attributes of adjacent spatial points after the update; this spatial consistency constraint ensures the spatial continuity of geological attributes within a local spatial region by minimizing the differences in attributes between adjacent points.

[0163] This constraint is used to suppress non-physical abrupt changes introduced by local observations, while maintaining the ability to express real geological structures such as faults and coal thickness variations. It is also incorporated as a regularization term into the update objective function, participating in the local model update process. The updated local geological model is then integrated and displayed transparently with the original global model, highlighting areas where significant changes have occurred. (Exceeding the threshold) is highlighted to achieve an intuitive and visual representation of the local geological structure of the coal mine.

[0164] In addition, in one embodiment of this application, after updating the local spatial area affected by the current drilling activity based on the virtual geological profile and generating a coal mine geological model, the method further includes: using the coal mine geological model as a new prior model, and iteratively updating the prior model based on the newly generated virtual geological profile during subsequent drilling processes.

[0165] After completing a local geological model update, the updated model results will be... As a new prior model input into subsequent drilling processes, it continuously receives newly generated virtual geological profile constraint information. Its closed-loop recursive update relationship is expressed as follows:

[0166]

[0167] in, The model after the nth update is used as prior information. This represents a virtual geological profile generated during a new round of drilling, providing new observational information. Indicates by , This collaborative update mechanism integrates prior models with new observational information. Through this closed-loop update system, the 3D geological model can continuously absorb new observational information as drilling progresses, achieving a dynamic evolution process of continuous updating and gradually approximating the actual geological structure.

[0168] The updated local model is integrated with the original global geological model in a unified spatial coordinate system, highlighting areas of significant change and making geological changes revealed during drilling readily apparent. When critical geological conditions such as faults, abrupt changes in coal thickness, or significant lithological variations are detected, corresponding change alerts are generated to remind relevant personnel to pay attention to potential geological risks. Furthermore, the update results and change feedback serve as a reference for subsequent drilling operations and continuous updates to the geological model, enabling the 3D geological model to be continuously corrected and improved during drilling progress. This forms a closed loop of dynamic updates to the locally transparent geological model driven by real-time drilling data.

[0169] like Figure 3As shown, taking the construction of a roof gas extraction hole by a smart drilling rig in a coal mine as an example, in practical applications, before conducting online real-time inference, a training sample set needs to be constructed using historical drilling data of the mining area and geological attribute labels measured from rock cores. The historical data is then processed using the signal processing method for subsequent S1 drilling response feature extraction and the feature representation method for S2 geological change sensitive features to generate historical feature vectors. These vectors are paired with corresponding geological attribute labels to train the S3 geological attribute inference model and the S4 recalibration inference network, respectively. After training, the model is deployed to the underground edge nodes for the S5 local transparent geological model update. The specific steps are explained below:

[0170] (1) Drilling data acquisition: During the drilling operation, the intelligent drilling rig collects drilling data such as drilling pressure, torque, rotation speed, mechanical drilling speed, pump pressure and drilling depth in real time, and forms a continuous drilling data stream.

[0171] (2) Drilling data processing and feature construction: The collected drilling data is filtered, aligned and feature extracted to construct a drilling response feature sequence indexed by drilling depth.

[0172] (3) Dynamic inference of borehole geological properties: Based on the drilling response feature sequence, the geological properties at different depths along the borehole trajectory are dynamically predicted using the geological property inference model to form a virtual geological profile along the borehole.

[0173] (4) Local geological model incremental update: The virtual geological profile is introduced into the original three-dimensional geological model as new geological constraint information. The local update area is adaptively determined according to the spatial location of the borehole, and only the affected local geological model is incrementally updated.

[0174] (5) Local model display and feedback: The updated local geological model is displayed in a lightweight manner so that on-site operators can intuitively obtain the latest geological information and use it as a reference for adjusting subsequent drilling construction parameters and extraction plans.

[0175] (6) Model closed-loop iterative update: The updated geological model is used as a new prior basis to continuously participate in the geological attribute inference and model update in the subsequent drilling process, so as to realize the dynamic closed-loop update of the three-dimensional geological model as drilling progresses.

[0176] This application's technical solution involves acquiring drilling data generated by an intelligent drilling rig during the drilling process; preprocessing and extracting features from the drilling data to generate a drilling response sequence indexed by drilling depth; analyzing the drilling response sequence to generate a geological response feature vector composed of multiple statistical features; performing data analysis and sampling processing on the geological response feature vector using a multi-layer Fourier neural network operator to generate initial estimation results; recalibrating the initial estimation results to generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory; and updating the local spatial areas affected by the current drilling activity based on the virtual geological profiles to generate a coal mine geological model. By utilizing neural network operators to mine the complex mapping relationship between data and geological attributes, preliminary intelligent inference is achieved. Then, sequence recalibration technology is used to optimize the continuity and consistency of the results, generating a highly reliable virtual geological profile. Finally, based on this profile, precise and incremental model updates are performed only on the local spatial areas affected by the borehole, thereby significantly improving update efficiency while ensuring accuracy. This allows the geological model to dynamically evolve with drilling, meeting the urgent need for real-time geological information at the production site.

[0177] The following describes embodiments of the dynamic updating device for coal mine geological models based on drilling data according to this application, which can be used to execute the dynamic updating method for coal mine geological models based on drilling data in the above embodiments of this application. It is understood that the dynamic updating device for coal mine geological models based on drilling data can be a computer program (including program code) running on a computer device. For example, the dynamic updating device for coal mine geological models based on drilling data can install industrial application software or industrial control management software to realize industrial cloud computing of industrial big data generated during the drilling exploration and production process through an industrial cloud platform. The dynamic updating device for coal mine geological models based on drilling data can be used to execute the corresponding steps in the method provided in the embodiments of this application. For details not disclosed in the embodiments of the dynamic updating device for coal mine geological models based on drilling data of this application, please refer to the embodiments of the dynamic updating method for coal mine geological models based on drilling data described above in this application.

[0178] Figure 4 A block diagram of a coal mine geological model dynamic update device based on drilling data according to an embodiment of this application is shown.

[0179] Reference Figure 4 As shown, a dynamic updating device for a coal mine geological model based on drilling data according to an embodiment of this application includes:

[0180] The acquisition module 310 is used to acquire drilling data generated by the intelligent drilling rig during the drilling process;

[0181] Extraction module 320 is used to preprocess and extract features from the drilling data to generate a drilling response sequence indexed by drilling depth;

[0182] The statistics module 330 is used to perform data analysis on the drilling response sequence and generate a geological response feature vector composed of multiple statistical features;

[0183] The estimation module 340 is used to perform data analysis and sampling processing on the geological response feature vector through a multi-layer Fourier neural operator to generate initial estimation results;

[0184] The calibration module 350 is used to recalibrate the initial calculation results and generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory.

[0185] The update module 360 ​​is used to update the local spatial area affected by the current drilling activity based on the virtual geological profile, and generate a coal mine geological model.

[0186] In this application, based on the aforementioned scheme, the drilling data includes drilling pressure, drilling torque, drill rod rotation speed, pump pressure, and drilling depth generated at each sampling time.

[0187] In this application, based on the aforementioned scheme, the step of preprocessing and feature extraction of the drilling data to generate a drilling response sequence indexed by drilling depth includes: representing the drilling data as a signal vector of a multidimensional drilling response that varies with time; smoothing the parameter components in the signal vector to generate a first signal; and mapping the first signal into a drilling response sequence indexed by drilling depth based on the mapping relationship between drilling time and drilling depth.

[0188] In this application, based on the aforementioned scheme, the step of performing data analysis on the drilling response sequence to generate a geological response feature vector composed of multiple statistical features includes: combining the drilling data to generate an original feature vector; calculating the statistical features of each parameter in the drilling response sequence at a preset depth window scale; generating drilling energy features and energy change at a preset depth based on each parameter in the drilling response sequence; and combining the original feature vector, the statistical features, the drilling energy features, and the energy change to generate a geological response feature vector.

[0189] In this application, based on the aforementioned scheme, the step of performing data analysis and sampling processing on the geological response feature vector using a multi-layer Fourier neural operator to generate an initial inference result includes: performing a layer-by-layer transformation on the geological response feature vector using a multi-layer Fourier neural operator to generate intermediate features; inputting the intermediate features into a pre-trained geological attribute operator inference model to output a continuous prediction result representing the geological attributes; and sampling the continuous prediction result based on a preset drilling depth to generate an initial inference result.

[0190] In this application, based on the aforementioned scheme, the step of recalibrating the initial estimation results to generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory includes: sorting and combining the initial estimation results according to the increasing drilling depth to generate a depth sequence; encoding and linearly projecting the depth sequence to generate sequence-aware features; and linearly projecting the sequence-aware features to generate recalibrated results corresponding to each drilling depth, which serve as virtual geological profiles along the borehole trajectory.

[0191] In this application, based on the aforementioned scheme, the step of updating the local spatial region affected by the current drilling activity based on the virtual geological profile to generate a coal mine geological model includes: determining the local spatial region affected by the current drilling activity based on the spatial location in the three-dimensional geological model and the sequence of spatial sampling points formed along the drilling trajectory; and incrementally updating the local spatial region based on the virtual geological profile to generate a coal mine geological model.

[0192] In this application, based on the aforementioned scheme, after updating the local spatial area affected by the current drilling activity based on the virtual geological profile to generate a coal mine geological model, the method further includes: using the coal mine geological model as a new prior model, and iteratively updating the prior model based on the newly generated virtual geological profile during subsequent drilling processes.

[0193] This application's technical solution involves acquiring drilling data generated by an intelligent drilling rig during the drilling process; preprocessing and extracting features from the drilling data to generate a drilling response sequence indexed by drilling depth; analyzing the drilling response sequence to generate a geological response feature vector composed of multiple statistical features; performing data analysis and sampling processing on the geological response feature vector using a multi-layer Fourier neural network operator to generate initial estimation results; recalibrating the initial estimation results to generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory; and updating the local spatial areas affected by the current drilling activity based on the virtual geological profiles to generate a coal mine geological model. By utilizing neural network operators to mine the complex mapping relationship between data and geological attributes, preliminary intelligent inference is achieved. Then, sequence recalibration technology is used to optimize the continuity and consistency of the results, generating a highly reliable virtual geological profile. Finally, based on this profile, precise and incremental model updates are performed only on the local spatial areas affected by the borehole, thereby significantly improving update efficiency while ensuring accuracy. This allows the geological model to dynamically evolve with drilling, meeting the urgent need for real-time geological information at the production site.

[0194] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown.

[0195] It should be noted that the computer system of the electronic device in this embodiment is only an example and should not impose any limitations on the function and scope of use of the embodiments of this application.

[0196] In this embodiment, the computer system includes a central processing unit 401, which can perform various appropriate actions and processes based on programs stored in read-only memory 402 or programs loaded from storage section 408 into random access memory 403, such as executing the dynamic update method for coal mine geological models based on drilling data described in the above embodiment. The random access memory 403 also stores various programs and data required for system operation, thereby realizing big data storage and big data management. The central processing unit 401, read-only memory 402, and random access memory 403 are interconnected via bus 404. Input / output interface 405 is also connected to bus 404.

[0197] The following components are connected to the input / output interface 405: an input section 406 including a keyboard, mouse, etc.; an output section 407 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the input / output interface 405 as needed. A removable medium 411, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 410 as needed so that computer programs read from it can be installed into the storage section 408 as needed.

[0198] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 409, and / or installed from removable medium 411. When the computer program is executed by central processing unit 401, it performs various functions defined in the system of this application.

[0199] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. The transmitted data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0200] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0201] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0202] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various alternative implementations described above.

[0203] In another aspect, this application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to implement the dynamic updating method for coal mine geological models based on drilling data described in the above embodiments.

[0204] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0205] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the methods according to the embodiments of this application.

[0206] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0207] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for dynamically updating a coal mine geological model based on drilling data, characterized in that, include: Acquire drilling data generated by the intelligent drilling rig during the drilling process; The drilling data is preprocessed and features are extracted to generate a drilling response sequence indexed by drilling depth; Data analysis is performed on the drilling response sequence to generate a geological response feature vector composed of multiple statistical features; The geological response feature vector is analyzed and sampled using a multi-layer Fourier neural operator to generate initial estimation results; The initial calculation results are recalibrated to generate virtual geological profiles at each drilling depth along the borehole trajectory; Based on the virtual geological profile, the local spatial areas affected by the current drilling activities are updated to generate a coal mine geological model; Specifically, based on the virtual geological profile, local spatial areas affected by the current drilling activity are updated to generate a coal mine geological model, including: Based on the spatial location in the three-dimensional geological model and the sequence of spatial sampling points formed along the drilling trajectory, the local spatial area affected by the current drilling activity is determined. Based on the virtual geological profile, incremental updates are performed on the local spatial region to generate a coal mine geological model; specifically, the geological attribute information in the virtual geological profile is... Mapped into a local three-dimensional space, corresponding to the original three-dimensional geological model in the same spatial position. Geological attribute prediction results at the location By combining the results, a local observation correction term is obtained through constructing the difference. The geological model within a local spatial region is incrementally updated using the following formula: in, Three-dimensional spatial coordinates; To adaptively update the weights; Representative of virtual geological profiles Provided observation correction terms; To determine the spatial location of the existing three-dimensional geological model Geological attribute prediction results at the location; The updated 3D geological model for a local spatial region in spatial location The geological properties of the location.

2. The method for dynamically updating a coal mine geological model based on drilling data according to claim 1, characterized in that, The drilling data includes drilling pressure, drilling torque, drill rod speed, pump pressure, and drilling depth generated at each sampling time. The drilling data is preprocessed and feature extracted to generate a drilling response sequence indexed by drilling depth, including: The drilling data is represented as a signal vector of a multidimensional drilling response that varies over time. The parameter components in the signal vector are smoothed to generate a first signal; Based on the mapping relationship between drilling time and drilling depth, the first signal is mapped into a drilling response sequence indexed by drilling depth.

3. The method for dynamically updating a coal mine geological model based on drilling data according to claim 1, characterized in that, Data analysis is performed on the drilling response sequence to generate a geological response feature vector composed of multiple statistical features, including: The drilling data is combined to generate an original feature vector; Calculate the statistical characteristics of each parameter in the drilling response sequence under a preset depth window scale; Based on the parameters in the drilling response sequence, the drilling energy characteristics and energy change at the preset depth are generated. The original feature vector, the statistical features, the drilling energy features, and the energy change are combined to generate a geological response feature vector.

4. The method for dynamically updating a coal mine geological model based on drilling data according to claim 1, characterized in that, The geological response feature vector is analyzed and sampled using a multi-layer Fourier neural operator to generate initial estimation results, including: The geological response feature vector is transformed layer by layer using a multi-layer Fourier neural operator to generate intermediate features; The intermediate features are input into a pre-trained geological attribute operator inference model, which outputs continuous prediction results representing geological attributes. The continuous prediction results are sampled based on a preset drilling depth to generate initial calculation results.

5. The method for dynamically updating a coal mine geological model based on drilling data according to claim 1, characterized in that, The initial calculation results are recalibrated to generate virtual geological profiles along the borehole trajectory at each drilling depth, including: The initial calculation results are sorted and combined according to the increasing drilling depth to generate a depth sequence; The depth sequence is encoded and linearly projected to generate sequence-aware features; Linear projection is performed on the sequence sensing features to generate recalibration results corresponding to each drilling depth, which serve as virtual geological profiles along the borehole trajectory.

6. The method for dynamically updating a coal mine geological model based on drilling data according to claim 1, characterized in that, Based on the virtual geological profile, after updating the local spatial areas affected by the current drilling activity and generating the coal mine geological model, the process further includes: The coal mine geological model is used as a new prior model. During subsequent drilling, the prior model is iteratively updated based on the newly generated virtual geological profile.

7. A dynamic updating device for a coal mine geological model based on drilling data, characterized in that, include: The acquisition module is used to acquire drilling data generated by the intelligent drilling rig during the drilling process; The extraction module is used to preprocess and extract features from the drilling data to generate a drilling response sequence indexed by the drilling depth. The statistics module is used to perform data analysis on the drilling response sequence and generate a geological response feature vector composed of multiple statistical features; The estimation module is used to perform data analysis and sampling processing on the geological response feature vector through multi-layer Fourier neural operators to generate initial estimation results; The calibration module is used to recalibrate the initial calculation results and generate virtual geological profiles corresponding to each drilling depth along the borehole trajectory. The update module is used to update the local spatial area affected by the current drilling activity based on the virtual geological profile, and generate a coal mine geological model. Specifically, based on the virtual geological profile, local spatial areas affected by the current drilling activity are updated to generate a coal mine geological model, including: Based on the spatial location in the three-dimensional geological model and the sequence of spatial sampling points formed along the drilling trajectory, the local spatial area affected by the current drilling activity is determined. Based on the virtual geological profile, incremental updates are performed on the local spatial region to generate a coal mine geological model; specifically, the geological attribute information in the virtual geological profile is... Mapped into a local three-dimensional space, corresponding to the original three-dimensional geological model in the same spatial position. Geological attribute prediction results at the location By combining the results, a local observation correction term is obtained through constructing the difference. The geological model within a local spatial region is incrementally updated using the following formula: in, Three-dimensional spatial coordinates; To adaptively update the weights; Representative of virtual geological profiles Provided observation correction terms; To determine the spatial location of the existing three-dimensional geological model Geological attribute prediction results at the location; The updated 3D geological model for a local spatial region in spatial location The geological properties of the location.

8. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for dynamically updating a coal mine geological model based on drilling data as described in any one of claims 1 to 6.

9. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method for dynamically updating a coal mine geological model based on drilling data as described in any one of claims 1 to 6.

Images