A method for quality control of pile foundation drilling in complex karst geology

Abstract

This invention provides a method for quality control of pile foundation drilling under complex karst geological conditions. The method includes generating a set of karst cave distribution characteristics and a set of stratigraphic anomalies based on karst geological survey data. Next, a set of karst cave development trajectory sequences is determined, and risk zoning is performed to identify high-risk karst cave development zones. Based on this, a set of high-risk borehole locations and a set of high-risk stratigraphic nodes are determined, thereby constructing a karst geological risk map. Based on the map, a set of drilling process paths and a borehole risk heat map are generated and overlaid onto the pile foundation construction plan to form a pile foundation borehole risk heat map, which visually displays construction risks and guides pile foundation construction. This invention can improve the quality and safety of pile foundation drilling under complex karst geological conditions and reduce construction risks.

Description

Technical Field

[0001] This invention relates to the field of karst geological pile foundation construction technology, and more specifically, to a method for controlling the quality of pile foundation drilling under complex karst geological conditions. Background Technology

[0002] Construction of pile foundations under complex karst geological conditions is an extremely challenging task. Karst geology is typically characterized by widespread distribution of caves, complex geological structures, and frequent changes in rock layer interfaces. These characteristics expose the pile foundation drilling process to numerous risks, such as borehole collapse, grout leakage, and stuck drill bits. Traditional pile foundation construction methods rely primarily on geological survey reports and the experience of construction personnel, but this approach has several limitations. First, geological survey reports typically provide only limited geological information and cannot accurately reflect the distribution and development of caves. Second, while the experience of construction personnel is important, it is often difficult to accurately assess construction risks when facing complex karst geological conditions, leading to various problems during construction and affecting the construction progress and quality.

[0003] Existing pile foundation construction technologies face several challenges when applied to karst geological conditions. Firstly, traditional drilling processes often involve simplistic path planning, failing to adequately consider the development trajectory of karst caves and geological anomalies. This makes it easy for the drill bit to encounter karst caves or areas of geological abnormality during construction, increasing construction risks. Secondly, existing risk assessment methods are primarily based on historical data and empirical formulas, lacking the analysis and processing of real-time geological data, making it difficult to accurately predict risks during construction. Furthermore, existing construction equipment lacks effective automatic correction and risk warning functions when facing complex karst geological conditions. Once construction deviations or risks occur, manual intervention is often required, which not only reduces construction efficiency but may also lead to safety accidents.

[0004] The existing technology has at least the following problems or defects: First, it is impossible to accurately obtain the distribution characteristics and development trajectory of karst caves under karst geological conditions, making it difficult to effectively assess construction risks; second, there is a lack of drilling process path planning methods for complex karst geological conditions, making it easy to encounter karst caves or abnormal strata areas during construction; third, the existing risk assessment and early warning methods are not accurate and timely enough to meet the needs of pile foundation construction under complex karst geological conditions; fourth, the existing construction equipment lacks effective automatic correction and risk early warning functions, requiring manual intervention once deviations or risks occur during construction, reducing construction efficiency and increasing safety risks. Summary of the Invention

[0005] This invention provides a method for controlling the quality of pile foundation drilling in complex karst geology, comprising:

[0006] Based on multi-source karst geological exploration data, a three-dimensional geological anomaly model characterizing the spatial distribution of karst caves and stratigraphic anomaly points is constructed through multi-modal data collaborative optimization processing.

[0007] Based on the aforementioned three-dimensional geological anomaly model, the development trajectory sequence of karst caves is dynamically analyzed and reconstructed to form a set of karst cave development trajectory sequences containing information on dissolution zones and rock strata interfaces.

[0008] Based on the aforementioned set of karst cave development trajectory sequences, and combined with the karst cave density and stratigraphic anomaly point distribution characteristics in the depth dimension, a three-dimensional dynamic risk assessment model is established to identify and delineate high-risk karst cave development zones.

[0009] Based on the spatial distribution characteristics of the high-risk karst cave development area set and the preset karst cave development conditions, the high-risk pore location set is located and determined;

[0010] Based on the set of high-risk karst cave development areas, stratigraphic nodes within their influence range are determined through spatial correlation analysis, forming a set of high-risk stratigraphic nodes;

[0011] By integrating the set of karst cave distribution characteristics, the set of karst cave development trajectory sequences, the set of high-risk karst cave development areas, the set of high-risk pore locations, and the set of high-risk stratigraphic nodes, a multi-dimensional karst geological risk map is constructed.

[0012] Based on the karst geological risk map, an optimized set of drilling process paths is intelligently generated, and a drilling risk heat map representing the spatial distribution of construction risks is generated accordingly.

[0013] The borehole risk heat map is spatially overlaid and merged with the pile foundation construction plan to generate a pile foundation borehole risk heat map, which is then visualized.

[0014] Furthermore, the construction of a three-dimensional geological anomaly model characterizing the spatial distribution features of karst caves and stratigraphic anomalies based on multi-source karst geological exploration data and through multi-modal data collaborative optimization processing includes:

[0015] Acquire karst geological exploration data, including borehole coring data, ground-penetrating radar detection data, and cross-hole CT imaging data;

[0016] The karst geological exploration data were standardized to unify the spatial benchmark and data scale, resulting in standardized borehole data, radar data, and CT data.

[0017] Spatial registration and data fusion are performed on the standardized borehole data, radar data, and CT data to generate effective geological anomaly data;

[0018] Based on the effective geological anomaly data, a set of cave distribution features is extracted and generated;

[0019] Identify geological anomalies associated with each cave feature in the cave distribution feature set, and generate a stratigraphic anomaly point set.

[0020] Furthermore, the spatial registration and data fusion of standardized borehole data, radar data, and CT data to generate effective geological anomaly data includes:

[0021] Spatial coordinate matching is performed on the standardized borehole data, radar data, and CT data to generate overlapping anomaly data sets and independent anomaly datasets;

[0022] For each overlapping anomaly data group in the aforementioned overlapping anomaly data group set, a three-dimensional data fusion calculation is performed based on a preset spatial weight allocation strategy to generate a fused geological anomaly dataset.

[0023] The fused geological anomaly dataset and the independent anomaly dataset are jointly identified as effective geological anomaly data.

[0024] Furthermore, based on the three-dimensional geological anomaly model, the development trajectory sequence of the karst cave is dynamically analyzed and reconstructed to form a set of karst cave development trajectory sequences containing information on dissolution zones and rock strata interfaces, including:

[0025] By applying borehole trajectory positioning technology, the original spatial development point sequence of each karst cave in the karst cave distribution feature set is determined;

[0026] The original spatial development point sequence is subjected to interpolation optimization under geological constraints to complete the missing trajectory segments caused by the sparse exploration points, and generate a continuous complete development point sequence.

[0027] The completed development point sequence is subjected to geological noise filtering to remove interference points caused by measurement errors or local geological disturbances, thereby generating an accurate development point sequence.

[0028] Based on the precise development point sequence, a set of cave development trajectory sequences containing information on dissolution zone boundaries and rock layer interfaces is constructed through geological law modeling and spatial curve fitting.

[0029] Furthermore, based on the set of cave development trajectory sequences, combined with the cave density at depth and the distribution characteristics of stratigraphic anomalies, a three-dimensional dynamic risk assessment model is established to identify and delineate high-risk cave development zones, including:

[0030] The set of cave development trajectory sequences is sliced ​​at preset depth intervals to extract the target erosion zone set corresponding to each depth level.

[0031] For the target dissolution zone sets at each depth level:

[0032] Spatial grid-based risk zoning is carried out to form a set of risk zones for cave development that includes multiple target karst zones;

[0033] Calculate the number of target karst zones per unit volume in each risk zone to obtain the regional karst cave density;

[0034] Based on the set of stratigraphic anomalies, the number of associated stratigraphic anomalies within each risk zone is counted to obtain the number of regional anomalies.

[0035] By combining the density of karst caves in the region with the number of anomalies in the region, and applying preset multi-dimensional high-risk judgment conditions, high-risk karst cave development areas at this depth level are identified and marked.

[0036] High-risk karst cave development areas at various depth levels are compiled to form a three-dimensional set of high-risk karst cave development areas.

[0037] Furthermore, the karst geological exploration data also includes a dynamic change sequence of drilling pressure, a mud loss sequence, and a core fragmentation rate sequence; and the method further includes:

[0038] The drilling pressure dynamic change sequence, mud loss sequence, and core fragmentation rate sequence are defined as the first geological risk time series data.

[0039] Temporal features are extracted from the first geological risk time series data to obtain geological risk time series features that characterize the dynamic risks of the drilling process;

[0040] Based on the aforementioned geological risk time-series characteristics, the probability of potential borehole collapse during the drilling process is predicted;

[0041] Based on the set of high-risk karst cave development areas, the set of high-risk borehole locations, and real-time drilling parameters, a second geological risk spatial data is generated.

[0042] Multi-scale spatial feature extraction is performed on the second geological risk spatial data to obtain multi-scale geological risk features that reflect local and regional risks;

[0043] Multi-scale analysis was performed on the temporal characteristics of the geological risks to obtain multi-scale temporal characteristics of the geological risks;

[0044] The geological risk time series features and the geological risk time series multi-scale features are residually connected to preserve the original time series information and enhance the feature expressive power, thus obtaining the geological risk residual connection features.

[0045] By integrating the multi-scale characteristics of geological risk with the residual connectivity characteristics of geological risk, a fused geological risk characteristic that comprehensively represents geological risk is formed;

[0046] Based on the integrated geological risk characteristics, intelligent decision-making is made and borehole control commands are generated and transmitted to the pile foundation construction equipment for execution. The borehole control commands include dynamic adjustment commands for drilling speed, active reinforcement commands for wall protection, and adaptive compensation commands for grouting.

[0047] Furthermore, the intelligent generation of the optimized drilling process path set based on the karst geological risk map includes:

[0048] Based on the set of high-risk strata nodes in the karst geological risk map, a spatial node sequence for grouting reinforcement is planned.

[0049] Based on the spatial morphological characteristics of the set of karst development trajectory sequences, a set of drill bit avoidance paths is generated;

[0050] The grouting reinforcement node sequence is spatiotemporally optimized and matched with the drill bit avoidance path set to generate a hole forming process path set.

[0051] Furthermore, after visualizing the thermal map of the pile foundation drilling risk, the method further includes:

[0052] Real-time acquisition of drilling sensor data, including drill bit vibration spectrum characteristics, dynamic value of mud specific gravity, and changes in borehole inclination angle;

[0053] Calculate the dynamic deviation between the real-time drilling sensor data and the predicted state of the pile foundation hole formation risk heat map;

[0054] Based on the dynamic deviation value, a hole formation quality deviation alarm is generated;

[0055] When the hole formation quality deviation alarm exceeds the preset safety threshold, the closed-loop control correction protocol is automatically triggered.

[0056] Furthermore, the three-dimensional data fusion calculation based on the preset spatial weight allocation strategy includes:

[0057] Different spatial weight coefficients were assigned to borehole data, radar data, and CT data in the overlapping anomaly data group, with borehole data assigned the highest weight, followed by radar data, and CT data assigned the lowest weight.

[0058] Based on the assigned weight coefficients, the three-dimensional data of the overlapping areas are weighted and fused to generate fused geological anomaly data.

[0059] Furthermore, an area is identified as a high-risk karst cave development area if one of the following conditions is met:

[0060] The density of karst caves in the region reaches the first preset high density threshold and the number of anomalies in the region reaches the first preset high anomaly threshold.

[0061] Or the density of karst caves in the region reaches the second preset high density threshold and the number of abnormal points in the region reaches the second preset high abnormal point threshold.

[0062] Or the density of karst caves in the region reaches the third preset high density threshold and the number of abnormal points in the region reaches the third preset high abnormal point threshold.

[0063] The embodiments of the present invention have at least the following beneficial effects:

[0064] 1. This invention generates a set of karst cave distribution characteristics and a set of stratigraphic anomalies, determines a set of karst cave development trajectory sequences, and performs risk zoning, enabling accurate identification of high-risk areas under complex karst geological conditions. This allows construction personnel to understand the distribution and development of karst caves in advance, thereby taking targeted risk prevention measures during pile foundation construction, effectively reducing construction risks such as borehole collapse, grout leakage, and stuck drill bits, and improving the safety and reliability of construction.

[0065] 2. This invention determines a set of high-risk borehole locations and a set of high-risk stratigraphic nodes based on a set of high-risk karst cave development areas, and generates a karst geological risk map, which in turn generates a set of borehole drilling process paths and a borehole risk heat map. This series of steps achieves optimized planning of the borehole drilling process path, enabling the drill bit to effectively avoid karst caves and anomalous stratigraphic areas, reducing obstacles and risks during construction, improving borehole drilling efficiency, ensuring borehole quality, and helping to shorten the construction cycle and reduce construction costs.

[0066] 3. This invention overlays a borehole risk heat map onto the pile foundation construction plan for display, and acquires real-time drilling sensor data during construction. Based on the deviation value, it generates a borehole quality deviation alarm. When the alarm exceeds a preset threshold, it triggers an automatic correction control protocol. This real-time monitoring and automatic correction mechanism can promptly detect and correct deviations during construction, ensuring the construction process remains under control, further improving the accuracy and quality of pile foundation borehole formation, and enhancing the stability and reliability of the construction process. Attached Figure Description

[0067] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description taken in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated in the drawings by way of example, not limitation, in which:

[0068] Figure 1 This is a flowchart illustrating a method for controlling the quality of pile foundation drilling in complex karst geology, as provided in an embodiment of the present invention. Detailed Implementation

[0069] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following drawings indicate similar items; therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0070] In traditional pile foundation construction, the three-dimensional spatial distribution characteristics of karst caves under karst geological conditions are difficult to model accurately. Spatial coordinate deviations exist between borehole core data and ground-penetrating radar data, resulting in insufficient accuracy in reconstructing the development trajectory of karst caves. The spatial correlation of stratigraphic anomalies is not effectively quantified, and the delineation of anomaly area boundaries relies on empirical thresholds, leading to missed identification of high-risk borehole locations. The borehole drilling process path planning does not incorporate the dynamic evolution characteristics of the karst zone, and the drill bit movement trajectory lacks coordinated optimization with changes in the rock strata interface, resulting in unforeseen risks of karst cave penetration during drilling.

[0071] During implementation, in limestone strata at depths of 30 to 50 meters, traditional methods employed independent analytical borehole data and cross-hole CT data without establishing a spatial weighted fusion mechanism for multi-source data. This resulted in an error exceeding 15% in calculating the thickness of the cave roof. Stratigraphic anomalies were classified solely based on density thresholds, neglecting their aggregation characteristics along the karst zone, leading to the failure to identify concealed caves larger than 2 meters in diameter as high-risk areas. Drill bit path planning, based on two-dimensional planar projection, failed to consider the inclination changes of the cave development trajectory, causing the drill bit to become stuck at the 75-degree inclined karst zone interface, resulting in a sudden increase in mud loss to three times the normal value. If these problems are not addressed, errors in the cave spatial distribution model will lead to insufficient grouting reinforcement range, and the pile foundation bearing layer will be unable to effectively avoid fractured rock masses.

[0072] In response, this application proposes a method for controlling the quality of pile foundation drilling in complex karst geological conditions, such as... Figure 1 As shown, it includes:

[0073] S1. Based on multi-source karst geological exploration data, a three-dimensional geological anomaly model is constructed to characterize the spatial distribution features of karst caves and stratigraphic anomaly points through multi-modal data collaborative optimization processing.

[0074] S2. Based on the three-dimensional geological anomaly model, dynamically analyze and reconstruct the development trajectory sequence of the karst cave to form a set of karst cave development trajectory sequences containing information on the dissolution zone and rock strata interface;

[0075] S3. Based on the set of karst cave development trajectory sequences, and combined with the karst cave density and stratigraphic anomaly point distribution characteristics in the depth dimension, establish a three-dimensional dynamic risk assessment model to identify and delineate high-risk karst cave development zones.

[0076] S4. Based on the spatial distribution characteristics of the high-risk karst cave development area set and the preset karst cave development conditions, locate and determine the high-risk pore location set;

[0077] S5. Based on the set of high-risk karst cave development areas, determine the stratigraphic nodes within their influence range through spatial correlation analysis to form a set of high-risk stratigraphic nodes;

[0078] S6. By integrating the set of karst cave distribution characteristics, the set of karst cave development trajectory sequences, the set of high-risk karst cave development areas, the set of high-risk pore locations, and the set of high-risk stratigraphic nodes, a multi-dimensional karst geological risk map is constructed.

[0079] S7. Based on the karst geological risk map, intelligently generate an optimized set of drilling process paths, and generate a drilling risk heat map that characterizes the spatial distribution of construction risks.

[0080] S8. The hole formation risk heat map is spatially superimposed and fused with the pile foundation construction plan to generate a pile foundation hole formation risk heat map, and then visualized.

[0081] Constructing a three-dimensional geological anomaly model involves integrating borehole core data, ground-penetrating radar data, and cross-hole CT imaging data, and employing multimodal data collaborative optimization processing to form a three-dimensional geological anomaly model characterizing the spatial distribution features of karst caves and stratigraphic anomalies. This can be achieved using methods such as spatial coordinate matching and three-dimensional voxel fusion calculation. Dynamically analyzing and reconstructing the development trajectory sequence of karst caves involves using borehole trajectory positioning technology combined with geological interpolation optimization to reconstruct the dynamic development process of karst caves. This can be achieved through geological noise filtering and generating a complete sequence of development points, overcoming the limitations of static geological analysis in understanding the evolution of karst zones.

[0082] Establishing a three-dimensional dynamic risk assessment model refers to identifying and classifying high-risk karst development zones based on a set of karst cave development trajectory sequences, combined with the depth dimension of karst cave density and the distribution characteristics of stratigraphic anomalies. A comprehensive assessment is then conducted using multi-dimensional indicators such as regional karst cave density and the number of regional anomalies, overcoming the limitations of single-indicator risk assessment. A karst geological risk map is a visualized topological network formed by spatially mapping karst cave distribution, development trajectories, pore location risks, and stratigraphic nodes. It can be generated through data fusion and spatial path matching, providing dynamic geological support for process route planning.

[0083] Generating a borehole risk heatmap refers to the intelligent generation of an optimized set of borehole drilling process paths based on a karst geological risk map, and the subsequent generation of a borehole risk heatmap characterizing the spatial distribution of construction risks. By spatially overlaying and integrating the borehole risk heatmap with the pile foundation construction plan, the construction risks are visualized, providing real-time dynamic early warnings and guidance for the construction process.

[0084] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0085] First, karst geological exploration data, including borehole core data, ground-penetrating radar data, and cross-hole CT imaging data, were acquired. This data was then standardized to obtain standardized borehole, radar, and CT data. During data integration, spatial coordinate matching was performed on the standardized data to generate overlapping anomaly datasets and independent anomaly datasets. The overlapping anomaly datasets were then fused in 3D to generate a fused geological anomaly dataset. The fused geological anomaly dataset and the independent anomaly dataset were collectively identified as the effective geological anomaly data, thereby constructing a 3D geological anomaly model characterizing the spatial distribution of karst caves and stratigraphic anomaly points.

[0086] Secondly, based on borehole trajectory positioning technology, the original spatial development point sequence of each karst cave in the 3D geological anomaly model is determined. Interpolation optimization under geological constraints is performed on the original spatial development point sequence to fill in missing trajectory segments caused by sparse exploration points, generating a continuous and complete development point sequence. Further geological noise filtering is applied to the complete development point sequence to remove interference points caused by measurement errors or local geological disturbances, generating an accurate development point sequence. Based on the accurate development point sequence, a set of karst cave development trajectory sequences containing information on karst zone boundaries and rock layer interfaces is constructed through geological regularity modeling and spatial curve fitting.

[0087] Then, based on the set of karst development trajectory sequences, combined with the karst density and stratigraphic anomaly distribution characteristics at depth, a three-dimensional dynamic risk assessment model is established. The karst development trajectory sequence set is sliced ​​at preset depth intervals to extract the target karst zones corresponding to each depth level. For each depth level's target karst zone set, spatial gridding risk partitioning is performed to form a karst development risk partition set containing multiple target karst zones. The number of target karst zones per unit volume in each risk partition is calculated to obtain the regional karst density. Based on the stratigraphic anomaly set, the number of associated stratigraphic anomalies within each risk partition is counted to obtain the regional anomaly count. Combining the regional karst density and the regional anomaly count, preset multi-dimensional high-risk judgment conditions are applied to identify and mark high-risk karst development areas at that depth level. The high-risk karst development areas at each depth level are summarized to form a three-dimensional set of high-risk karst development areas.

[0088] Next, based on the spatial distribution characteristics of high-risk karst cave development zones and the pre-set karst cave development conditions, a set of high-risk pore locations is located and determined. Based on the high-risk karst cave development zones, stratigraphic nodes within their influence range are determined through spatial correlation analysis, forming a set of high-risk stratigraphic nodes.

[0089] Furthermore, by integrating the distribution feature set of karst caves, the sequence set of karst cave development trajectories, the set of high-risk karst cave development areas, the set of high-risk borehole locations, and the set of high-risk stratigraphic nodes, a multi-dimensional karst geological risk map is constructed. Based on the karst geological risk map, an optimized set of borehole drilling process paths is intelligently generated, and a borehole risk heat map characterizing the spatial distribution of construction risks is generated accordingly.

[0090] Finally, the borehole risk heat map is spatially overlaid and merged with the pile foundation construction plan to generate a pile foundation borehole risk heat map, which is then visualized.

[0091] This application further proposes a method for constructing a three-dimensional geological anomaly model characterizing the spatial distribution features of karst caves and stratigraphic anomalies based on multi-source karst geological exploration data and through multi-modal data collaborative optimization processing, including:

[0092] Acquire karst geological exploration data, including borehole coring data, ground-penetrating radar data, and cross-hole CT imaging data. Standardize this data to unify spatial references and data scales, resulting in standardized borehole, radar, and CT data. Standardization processes may include coordinate system unification, data format conversion, and scale normalization.

[0093] Standardized borehole, radar, and CT data are spatially registered and fused to generate effective geomorphic anomaly data. Specifically, spatial coordinate matching is first performed to map data from different sources to the same 3D spatial grid. Then, data from overlapping areas are fused, using a weighted average method where borehole data has a weight of 0.5, radar data 0.3, and CT data 0.2. For independent data regions without overlap, the original data is directly retained. This method yields effective geomorphic anomaly data.

[0094] Based on effective geological anomaly data, a feature set of karst cave distribution is extracted and generated. Clustering algorithms can be used to identify the spatial distribution patterns of anomaly data points and extract feature parameters such as the location, shape, and volume of the karst caves. In practice, the DBSCAN algorithm can be used to cluster anomaly points to obtain the spatial contours of each karst cave.

[0095] Identify geological anomalies associated with various karst features within a set of karst cave distribution characteristics to generate a stratigraphic anomaly set. A buffer zone of a certain radius can be established around each karst feature, and anomalies falling within the buffer zone can be associated with that karst feature, thus forming a stratigraphic anomaly group. In practice, a buffer zone is established centered on the karst cave boundary, and rock fracture points within a 5-meter radius are extracted as associated anomalies.

[0096] Specifically, the core fragmentation rate in borehole coring data is converted into a standardized value in the 0-1 range; the electromagnetic wave reflection time difference in ground-penetrating radar data is converted into spatial coordinate offset; and the density difference values ​​in cross-hole CT imaging data are converted into three-dimensional meshed data. During data integration, when different data sources have overlapping anomalies at the same spatial location, a weighted fusion algorithm is used to generate anomaly data with higher confidence. In practice, when borehole data and CT data simultaneously detect anomalies at a depth of 15 meters, that location is marked as a high-confidence anomaly area. The generation of karst cave distribution characteristics is achieved by extracting continuous anomaly areas from the fused data. In practice, anomaly areas with a volume greater than 2 cubic meters and a connectivity index higher than 0.7 are defined as independent karst caves. The generation of stratigraphic anomaly point groups is achieved by analyzing the geological parameters of karst cave boundaries. In practice, borehole points with a core fragmentation rate exceeding 60% within 3 meters of the karst cave edge are extracted as associated anomaly points.

[0097] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0098] Acquire karst geological exploration data, including borehole core data, ground-penetrating radar data, and cross-hole CT imaging data. Standardize these data to obtain standard borehole data, standard radar data, and standard CT data. Standardization may include steps such as coordinate system unification, data format conversion, and scale normalization.

[0099] Furthermore, the standardized data is integrated. First, spatial coordinate matching is performed to map data from different sources to the same 3D spatial grid. Then, data from overlapping areas are fused, using a weighted average method, with borehole data weighted at 0.5, radar data at 0.3, and CT data at 0.2. For independent data areas without overlap, the original data is directly retained. This method yields effective geological anomaly data.

[0100] More specifically, clustering algorithms can be used to identify the spatial distribution patterns of outlier data points and extract characteristic parameters such as the location, shape, and volume of caves. In practice, the DBSCAN algorithm can be used to cluster outlier points to obtain the spatial outline of each cave.

[0101] Identify the stratigraphic anomalies associated with each karst cave feature in the set of karst cave distribution features, and generate a set of stratigraphic anomalies. A buffer zone of a certain radius can be set around each karst cave feature, and anomalies falling within the buffer zone can be associated with that karst cave feature, thus forming a group of stratigraphic anomalies.

[0102] This application further proposes a method for spatial registration and data fusion of standardized borehole data, radar data, and CT data to generate effective geological anomaly data, including:

[0103] Spatial coordinate matching is performed on standardized borehole data, radar data, and CT data to generate overlapping anomaly datasets and independent anomaly datasets.

[0104] Spatial coordinate matching employs an iterative nearest-point algorithm to unify the spatial reference of multi-source data, with point cloud registration errors controlled within ±0.05 meters. Specifically, a unified three-dimensional coordinate system is first established, and then GPS positioning technology is used to determine the spatial coordinates of each data acquisition point. Further, a coordinate transformation algorithm is used to convert the coordinate systems of different data sources to a unified coordinate system. In implementation, a seven-parameter coordinate transformation method can be used for coordinate system unification.

[0105] For each overlapping anomaly dataset, a 3D data fusion calculation is performed based on a preset spatial weighting strategy to generate a fused geological anomaly dataset. During the 3D fusion process, borehole data, radar data, and CT data are assigned weight coefficients of 0.5, 0.3, and 0.2, respectively, and fused voxel data are generated using a weighted average algorithm. Independent anomaly datasets are filtered by setting isolated anomaly judgment thresholds, where anomaly areas with a signal-to-noise ratio exceeding 8 dB in a single data source are retained. During data integration, the spatial correlation established by spatial coordinate matching of multi-source data synergizes with the complementary mechanism of 3D fusion. The high-resolution characteristics of borehole data are combined with the deep detection capabilities of CT data to effectively eliminate multipath interference from radar data in the shallow layer.

[0106] The fused geological anomaly dataset, together with the independent anomaly dataset, is identified as effective geological anomaly data. This approach preserves high-confidence anomaly information obtained from cross-validation of multi-source data while ensuring that no unique anomalies detected by a single method are overlooked, thus guaranteeing the completeness and accuracy of the anomaly data.

[0107] As a preferred embodiment, spatial coordinate matching is performed on standard borehole data, standard radar data, and standard CT data. Specifically, a unified three-dimensional coordinate system is first established, and then GPS positioning technology is used to determine the spatial coordinates of each data acquisition point. Further, a coordinate transformation algorithm is used to convert the coordinate systems of different data sources to a unified coordinate system. In practice, a seven-parameter coordinate transformation method can be used for coordinate system unification.

[0108] After spatial coordinate matching is completed, an overlapping anomaly dataset and an independent anomaly dataset are generated. The overlapping anomaly dataset contains anomalies detected simultaneously by multiple survey methods at the same spatial location, while the independent anomaly dataset contains anomalies detected by only a single survey method.

[0109] Three-dimensional fusion is performed on each overlapping anomaly data set to generate fused geological anomaly data. The fusion process can employ a weighted average method, assigning weights of 0.5, 0.3, and 0.2 to borehole data, radar data, and CT data, respectively. A three-dimensional voxel fusion algorithm integrates information from different data sources into the same spatial unit, thus obtaining the fused geological anomaly dataset.

[0110] By combining the fused geological anomaly dataset and the independent anomaly dataset, effective geological anomaly data is identified. This approach preserves high-confidence anomaly information obtained from cross-validation of multi-source data while ensuring that no unique anomalies detected by a single method are overlooked, thus guaranteeing the completeness and accuracy of the anomaly data.

[0111] This application further proposes a method for dynamically analyzing and reconstructing the development trajectory sequence of karst caves based on the aforementioned three-dimensional geological anomaly model, forming a set of karst cave development trajectory sequences containing information on dissolution zones and rock strata interfaces, including:

[0112] By applying borehole trajectory positioning technology, the original spatial development point sequence of each karst cave in the distribution feature set is determined. The borehole trajectory positioning technology can be implemented using a multi-source sensor fusion approach, combining gyroscope azimuth data with borehole inclinometer tilt data to form a three-dimensional spatial coordinate positioning. More specifically, when obtaining the original development points through borehole trajectory positioning, a high-precision inertial navigation unit is used to collect borehole azimuth data, ensuring that the spatial coordinate error of each development point is controlled within ±5 cm.

[0113] The original spatial development point sequence is optimized by interpolation under geological constraints to fill in missing segments of the trajectory caused by the sparseness of exploration points, generating a continuous sequence of completed development points. Geological interpolation optimization can be implemented based on a geostatistical model, using Kriging interpolation to predict the development trend of karst caves between adjacent boreholes. The interpolation interval can be set within the range of 0.5 meters to 2 meters. When supplementing missing points through geological interpolation optimization, an interpolation model is established based on the dip angle data of the rock strata exposed by two adjacent boreholes. During implementation, when the rock strata strike 30 degrees north of east and the dip angle is 15 degrees, development points are inserted along this direction at 1-meter intervals.

[0114] The completed development point sequence is subjected to geological noise filtering to remove interfering points caused by measurement errors or local geological disturbances, generating an accurate development point sequence. Geological noise filtering can be combined with morphological filtering algorithms. During implementation, a dissolution rate threshold is set; points where the dissolution volume change rate between adjacent development points exceeds 0.01 cubic meters per second are identified as anomalies and removed. When performing geological noise filtering, a density anomaly threshold is set; data points where the resistivity value exceeds three times the standard deviation of the surrounding rock are identified as instrument interference and removed.

[0115] Based on the precise sequence of development points, a set of karst cave development trajectory sequences, including information on dissolution zone boundaries and rock strata interfaces, is constructed through geological pattern modeling and spatial curve fitting. During the generation of the karst cave development trajectory sequence, filtered development points can be smoothly connected using a Bezier curve algorithm to form continuous trajectories. In the final generated karst cave development trajectory sequence, the dissolution zone set is automatically divided by calculating the rate of change of dissolution volume between adjacent development points, while the rock strata interface set is determined by identifying resistivity abrupt changes and core fragmentation peak points. The resulting trajectory sequence accurately reflects the extension direction of karst caves in three-dimensional space. In practice, a certain karst cave development trajectory shows that it extends along a northwest-southeast direction and is accompanied by three dissolution zones, providing a precise basis for subsequent assessment of the collapse risk in this area.

[0116] As a preferred embodiment, the development trajectory sequence of each karst cave in the karst cave distribution feature set is determined to obtain a karst cave development trajectory sequence set. First, based on borehole trajectory positioning technology, the original development point sequence of each karst cave distribution feature in the karst cave distribution feature set is determined to obtain an original development point sequence set. During implementation, a magnetic field positioning system is used to track the borehole trajectory in real time, record the drill bit position coordinates, and spatially match the detected karst cave feature points with the borehole trajectory to generate the original development point sequence.

[0117] Secondly, geological interpolation optimization is performed on each original development point sequence in the original development point sequence set to generate a complete development point sequence, resulting in a complete development point sequence set. In practice, Kriging interpolation is used to estimate the missing regions based on the spatial distribution characteristics of known development points, generating a continuous complete development point sequence.

[0118] Then, geological noise filtering is applied to each completed development point sequence in the complete development point sequence set to generate a development point sequence, resulting in a development point sequence set. During implementation, a mean-mode filtering algorithm is applied to remove outliers caused by equipment errors or local geological anomalies, retaining development points that conform to the development patterns of karst caves.

[0119] Finally, the cave development trajectory sequence corresponding to each development point sequence in the development point sequence set is determined, resulting in a cave development trajectory sequence set. Each cave development trajectory sequence in the set contains a set of dissolution zones and a set of rock layer interfaces. During implementation, a three-dimensional curve fitting algorithm is used to transform the development point sequence into a continuous cave development trajectory, and the spatial distribution of dissolution zones and rock layer interfaces is identified based on geological principles.

[0120] This application further proposes a method for identifying and classifying high-risk karst development zones by establishing a three-dimensional dynamic risk assessment model based on the aforementioned set of karst development trajectory sequences, combined with the karst density and stratigraphic anomaly distribution characteristics in the depth dimension. The method includes:

[0121] The set of karst development trajectory sequences is sliced ​​at preset depth intervals to extract the target karst zone set corresponding to each depth level. During implementation, the sequence is divided longitudinally at intervals of one meter depth, and karst zones at the same depth are grouped into the corresponding target depth karst zone set.

[0122] For the target karst zones at each depth level, spatial gridding risk partitioning is performed to form a karst development risk partition set containing multiple target karst zones. Karst zones at the same depth level are divided into multiple rectangular or hexagonal units according to planar coordinates, and the number of karst zones contained in each unit is quantified as the regional karst density.

[0123] The number of target karst zones per unit volume in each risk zone is calculated to obtain the regional karst cave density. Based on the stratigraphic anomaly point set, the number of associated stratigraphic anomaly points within each risk zone is counted to obtain the regional anomaly point count. The calculation of the stratigraphic anomaly point count requires association with geological anomaly data within the same spatial unit. In practice, the stratigraphic anomaly point set is matched with the karst cave development risk zones through coordinate mapping.

[0124] By combining the regional cave density and the number of anomalies, a multi-dimensional high-risk determination condition is applied to identify and mark high-risk cave development areas at this depth level. The determination condition uses three sets of fixed threshold combinations: the first set of thresholds is a regional cave density greater than or equal to 5 per cubic meter and a formation anomaly number greater than or equal to 20; the second set of thresholds is a regional cave density greater than or equal to 3 per cubic meter and a formation anomaly number greater than or equal to 30; and the third set of thresholds is a regional cave density greater than or equal to 8 per cubic meter and a formation anomaly number greater than or equal to 10. When the spatial spacing of the erosion zones within a grid cell is detected to be less than a preset aggregation threshold, a dynamic reduction mechanism for the grid cell size is activated until the spacing of the erosion zones meets the analysis accuracy requirements. If the regional cave density meets the high-risk threshold but the number of formation anomalies does not meet the standard, a cave connectivity index is introduced for supplementary determination. This cave connectivity index is calculated by measuring the proportion of connecting paths between cavities using cross-hole CT data. When the cave connectivity index is greater than or equal to 0.7, the grid cell is still marked as a high-risk cave development area.

[0125] High-risk karst cave development areas at various depth levels are aggregated to form a three-dimensional set of high-risk karst cave development areas. During implementation, at a depth of 20 meters, all karst zones at that depth are extracted and formed into a planar distribution dataset. Next, the karst zones at this layer are gridded, with each grid cell covering an area of ​​10 square meters. The number of karst zones within each cell is counted as the regional karst cave density. Simultaneously, data from the stratigraphic anomaly point set is mapped to corresponding grid cells through coordinate matching, and the number of anomalies within each cell is counted as the regional anomaly point count. A dual-indicator threshold combination is used to pre-determine high-risk conditions. During implementation, a grid cell is identified as a high-risk area when its karst cave density reaches 8 per cubic meter and the number of anomalies exceeds 10. By performing this determination layer by layer, the high-risk areas at each depth level are ultimately integrated into a three-dimensional set of high-risk karst cave development areas.

[0126] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0127] Each karst erosion zone at the same depth within the karst development trajectory sequence set is identified as a target depth karst erosion zone, resulting in a target depth karst erosion zone set. This is then used to generate various target depth karst erosion zone sets. During implementation, for the three target depths of 10 meters, 20 meters, and 30 meters, corresponding target depth karst erosion zone sets are generated respectively.

[0128] For each set of karst zones at each target depth, perform the following steps to generate a set of high-risk karst cave development areas at each target depth, thus obtaining the set of high-risk karst cave development areas at each target depth:

[0129] First, risk zoning is performed on each target depth karst zone within the target depth karst zone set, resulting in a karst development risk zone set. Each karst development risk zone within this set contains multiple target depth karst zones. During implementation, the target depth karst zone set is divided into multiple 10m × 10m grids, with each grid serving as a karst development risk zone.

[0130] The number of karst zones at the target depth within each karst development risk zone set is determined as the regional karst density, resulting in a regional karst density set. During implementation, the number of karst zones within each 10m × 10m grid is calculated and used as the regional karst density for that grid.

[0131] Based on the stratigraphic anomaly set, the number of stratigraphic anomalies corresponding to each karst development risk zone in the karst development risk zone set is determined as the regional anomaly count, resulting in the regional anomaly count set. During implementation, the number of stratigraphic anomalies within each 10m × 10m grid is counted as the regional anomaly count for that grid.

[0132] Based on the regional cave density set and the regional anomaly point count set, each cave development risk zone that meets the preset high-risk conditions is identified as a high-risk cave development zone at the target depth, thus obtaining the target depth high-risk cave development zone set. During implementation, when the regional cave density of a certain grid is greater than or equal to 5 caves / cubic meter and the number of regional anomalies is greater than or equal to 20, that grid is identified as a high-risk cave development zone at the target depth.

[0133] Each target depth's high-risk karst cave development area set is identified as a high-risk karst cave development area set. During implementation, the high-risk karst cave development areas at depths of 10 meters, 20 meters, and 30 meters are merged to form the final high-risk karst cave development area set.

[0134] This application further proposes defining the drilling pressure dynamic change sequence, mud loss sequence, and core fragmentation rate sequence contained in karst geological exploration data as the first geological risk time series data; extracting time series features from the first geological risk time series data to obtain geological risk time series features characterizing the dynamic risks of the drilling process; predicting the probability of potential borehole collapse during the drilling process based on the geological risk time series features; generating second geological risk spatial data according to the high-risk karst cave development area set, the high-risk borehole location set, and real-time drilling parameters; and extracting multi-scale spatial features from the second geological risk spatial data to obtain multi-scale geological risk features reflecting local and regional risks. The process involves: 1) performing multi-scale analysis on the temporal characteristics of the geological risk to obtain multi-scale temporal features of the geological risk; 2) performing residual connection between the temporal features of the geological risk and the multi-scale temporal features of the geological risk to retain the original temporal information and enhance the feature expressive power, resulting in geological risk residual connection features; 3) fusing the multi-scale features of the geological risk and the geological risk residual connection features to form a fused geological risk feature that comprehensively characterizes the geological risk; 4) making intelligent decisions and generating drilling control commands based on the fused geological risk features, transmitting them to the pile foundation construction equipment for execution, wherein the drilling control commands include dynamic adjustment commands for drilling speed, active wall reinforcement commands, and adaptive compensation commands for grouting.

[0135] The dynamic change sequence of drilling pressure is acquired by a pressure sensor at a sampling frequency of once per second, the mud loss sequence is monitored in real time by a flow meter, and the core breakage rate sequence is calculated based on the volume ratio of broken cores in the core barrel.

[0136] Temporal feature extraction employs a sliding window analysis method, with a window size of five or ten seconds, to capture the frequency of parameter fluctuations and the correlation of abrupt change points. Hole collapse probability prediction is achieved through a trained LSTM neural network model; the input layer receives the standardized temporal feature vector, and the output layer generates probability values.

[0137] In the generation of the second geological risk data, high-risk karst cave development areas are spatially matched with the drill bit position coordinates in the real-time drilling parameters. Dynamic data acquisition is triggered when the drill bit enters a high-risk borehole location. Multi-scale feature extraction uses convolutional kernels of different sizes for local anomaly detection and regional trend separation. In practice, a 3x3 convolutional kernel is used to capture local details, and a 7x7 convolutional kernel is used to extract macroscopic trends. Residual connections add the original temporal features to the multi-scale temporal features through a skip connection structure, preserving high-frequency components in the original signal. Feature fusion employs a channel attention mechanism to weightedly concatenate the multi-scale spatial features and the residual-enhanced temporal features.

[0138] The generation of borehole control commands is based on the comparison between fusion features and preset risk thresholds. When the probability of borehole collapse exceeds 70%, a drilling speed adjustment command is triggered. The wall reinforcement command dynamically adjusts the grouting pressure according to the rate of increase of core breakage. The flow rate parameter of the grouting compensation command is linearly related to the mud loss.

[0139] Specifically, during pile foundation drilling, pressure sensors monitor the axial pressure of the drill rod in real time. When the drill bit contacts the edge of the karst cave, a positive pulse signal exceeding 20 kN appears in the drill pressure abrupt change sequence. Mud loss sensors detect a loss exceeding 0.5 cubic meters per minute, marking it as a high-risk leakage event. Core fragmentation rate is calculated using image recognition algorithms; a fragmentation rate exceeding 40% triggers the feature extraction process. The time-series analysis module identifies three parameters rising synchronously within a five-second window, determining it as a composite risk pattern, and inputs the time-series features into the neural network model.

[0140] After the model outputs the borehole collapse probability value, it combines the distribution data of high-risk karst development areas at the drill bit location to generate second geological risk data containing spatial coordinates. Multi-scale convolutional layers extract local dissolution features within a three-meter radius of the current borehole point and regional seepage trends within a ten-meter radius, respectively. Residual connections retain information on sudden fluctuations in the original time-series data. The feature fusion module combines spatial multi-scale features with temporal residual features through adaptive weight allocation to form a four-dimensional risk tensor. The control command generator, based on the maximum value and threshold relationship of each dimension in the tensor, coordinates and issues composite commands to reduce the drilling speed to 0.5 meters per minute, increase the wall grouting pressure to 2 MPa, and increase the mud compensation flow rate to 0.8 cubic meters per minute, achieving dynamic risk control.

[0141] This application further proposes a method for intelligently generating an optimized set of drilling process paths based on karst geological risk maps, including:

[0142] Based on the high-risk strata node set in the karst geological risk map, a spatial node sequence for grouting reinforcement is planned. The grouting reinforcement node sequence is determined by density analysis of the high-risk strata node set using a spatial clustering algorithm. During implementation, the DBSCAN algorithm is used to mark dense areas with node spacing less than 0.5 meters as the core reinforcement area. Within the high-risk strata node set of the karst geological risk map, the geological risk quantification results are converted into specific coordinate points for grouting reinforcement operations using spatial coordinate mapping technology. During implementation, areas with a solution zone density exceeding 5 nodes / cubic meter are automatically marked as reinforcement nodes.

[0143] Based on the spatial morphological characteristics of the cave development trajectory sequence set, a set of drill bit avoidance paths is generated. The generation of the drill bit avoidance path set can be combined with the three-dimensional curvature characteristics of the cave trajectory. When the radius of curvature of the cave trajectory is less than 3 meters, a circular avoidance path is generated; when the radius of curvature is greater than or equal to 3 meters, a straight detour path is generated. The cave development trajectory sequence set generates obstacle avoidance rules through a trajectory analysis algorithm, in which the boundary point set of the cave cavity is converted into a three-dimensional obstacle avoidance surface, and the drill bit's travel path is constrained within a safe range of 0.5 meters outside this surface.

[0144] The grouting reinforcement node sequence and the drill bit avoidance path set are spatiotemporally optimized and matched to generate a drilling process path set. The spatial path matching process employs a three-dimensional collision detection algorithm, where the reinforcement radius of the grouting reinforcement node sequence is set to 1.2 times the drill bit diameter, and the minimum safe distance between the avoidance path and the reinforcement area is set to 0.8 meters. During the spatial path matching stage, the grouting reinforcement node sequence and the drill bit avoidance path set are dynamically coupled through vector superposition in a three-dimensional coordinate system. The spatial coverage of the reinforcement area and the obstacle avoidance area of ​​the drill bit's path form a complementary layout, ensuring that the drill bit remains within the protection range of the reinforced area when traversing high-risk strata, while simultaneously preventing spatial interference between the drill bit trajectory and unreinforced karst areas. This matching process calculates the time difference between the solidification time of the reinforcement area and the drill bit's travel speed in real time, ensuring that the grouting material has reached its design strength when the drill bit reaches the reinforcement area. In practice, when the drill bit's travel speed is 0.3 meters per minute, the reinforcement operation must be completed 15 minutes in advance.

[0145] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0146] In the high-risk stratum node set of the karst geological risk map, the geological risk quantification results are converted into specific coordinate points for grouting reinforcement operations using spatial coordinate mapping technology. During implementation, areas with a solution zone density exceeding 5 nodes / cubic meter are automatically marked as reinforcement nodes. The set of karst cave development trajectory sequences generates obstacle avoidance rules through trajectory analysis algorithms. The boundary point set of karst cave cavities is converted into a three-dimensional obstacle avoidance surface, and the drill bit's travel path is constrained within a safe range of 0.5 meters outside this surface. In the spatial path matching stage, the grouting reinforcement node sequence and the drill bit avoidance path set are dynamically coupled through vector superposition in a three-dimensional coordinate system. The spatial coverage of the reinforcement area and the obstacle avoidance area of ​​the drill bit's travel path form a complementary layout, ensuring that the drill bit remains within the protection range of the reinforced area when traversing high-risk strata, while avoiding spatial interference between the drill bit trajectory and unreinforced karst cave areas. This matching process calculates the time difference between the solidification time of the reinforced area and the drill bit's travel speed in real time, ensuring that the grouting material has reached the design strength when the drill bit reaches the reinforced area. When the drill bit travels at a speed of 0.3 meters per minute, the reinforcement operation must be completed 15 minutes in advance.

[0147] This application further proposes to acquire real-time drilling sensor data, which includes drill bit vibration spectrum characteristics, dynamic values ​​of mud specific gravity, and changes in borehole inclination angle; calculate the dynamic deviation value between the real-time drilling sensor data and the predicted state of the pile foundation borehole risk heat map; generate a borehole quality deviation alarm based on the dynamic deviation value; and automatically trigger a closed-loop control correction protocol when the borehole quality deviation alarm exceeds a preset safety threshold.

[0148] The sampling frequency of the drill bit vibration spectrum was set to the range of 2000Hz to 5000Hz. Energy distribution characteristics in the 0.5kHz to 2kHz frequency band were extracted using Fast Fourier Transform to capture the impact vibration signal when the drill bit contacts the karst cave. The mud specific gravity was monitored using an online densitometer with a measurement accuracy controlled within ±0.01g / cm³. A mud specific gravity drop exceeding 5% within 30 seconds was considered a karst cave leakage event. The borehole inclination angle was acquired in real-time using a dual-axis tilt sensor with an angular resolution of 0.01 degrees. Trajectory correction was triggered when the horizontal offset exceeded 0.3% of the designed trajectory.

[0149] In the coordinate system of the pile foundation construction plan, real-time drilling sensor data is converted into a three-dimensional spatial coordinate point set and spatially matched with the predicted risk areas in the karst geological risk map. By calculating the Euclidean distance deviation between the real-time data points and the heat map grid cells, a level one alarm is generated when the average deviation value of three consecutive sampling periods exceeds 0.25, and automatic correction is triggered when it exceeds 0.3.

[0150] The deviation control protocol comprises three parallel execution modules: a drill pressure adjustment module dynamically adjusts the rotation speed based on vibration spectrum characteristics, controlling the impact vibration energy within the range of 50J to 80J; a mud compensation module increases the mud specific gravity to a preset safety value at a rate of 0.5% per second; and a trajectory correction module applies a correction torque of 0.5kN·m to 1.2kN·m via a hydraulic deviation corrector, restoring the borehole inclination angle to within the allowable deviation range within five drilling cycles. This multi-parameter collaborative control mechanism ensures a dynamic match between the actual borehole trajectory and the predicted path in the risk heat map, effectively suppressing the accumulation of trajectory deviations caused by karst cave collapse.

[0151] For example, during the pile foundation drilling stage, a triaxial accelerometer is used to collect the vibration spectrum data of the drill bit in real time, with the vibration spectrum sampling frequency set to 2000Hz. At the same time, a mud density sensor is used to obtain the mud specific gravity value at a sampling rate of twice per second, and the hole inclination angle is monitored in real time with a resolution of 0.1 degrees by a high-precision tilt sensor.

[0152] The aforementioned sensor data is transmitted to the data processing center via an industrial bus and dynamically compared with a pre-generated risk heat map. The deviation calculation employs a weighted Euclidean distance algorithm, with the vibration spectrum deviation weight set at 0.6, the mud specific gravity deviation weight at 0.3, and the borehole inclination angle deviation weight at 0.1. When the overall deviation value exceeds the 0.85 threshold, the system automatically activates the deviation correction control module. This module comprises three parallel execution units: the first unit controls the hydraulic system to adjust the drilling pressure distribution for trajectory correction; the second unit starts the high-pressure grouting pump to perform wall reinforcement; and the third unit adjusts the main winch speed via a frequency converter to optimize the drilling speed.

[0153] This application further proposes to acquire drilling sensor data in real time, including drill bit vibration spectrum characteristics, dynamic value of mud specific gravity, and hole inclination angle change; calculate the dynamic deviation value between the real-time drilling sensor data and the predicted state of the pile foundation hole formation risk heat map; generate a hole formation quality deviation alarm based on the dynamic deviation value; and automatically trigger a closed-loop control correction protocol when the hole formation quality deviation alarm exceeds a preset safety threshold.

[0154] The sampling frequency of the drill bit vibration spectrum was set to the range of 2000Hz to 5000Hz. Energy distribution characteristics in the 0.5kHz to 2kHz frequency band were extracted using Fast Fourier Transform to capture the impact vibration signal when the drill bit contacts the karst cave. The mud specific gravity was monitored using an online densitometer with a measurement accuracy controlled within ±0.01g / cm³. A mud specific gravity drop exceeding 5% within 30 seconds was considered a karst cave leakage event. The borehole inclination angle was acquired in real-time using a dual-axis tilt sensor with an angular resolution of 0.01 degrees. Trajectory correction was triggered when the horizontal offset exceeded 0.3% of the designed trajectory.

[0155] In the coordinate system of the pile foundation construction plan, real-time drilling sensor data is converted into a three-dimensional spatial coordinate point set and spatially matched with the predicted risk areas in the karst geological risk map. By calculating the Euclidean distance deviation between the real-time data points and the heat map grid cells, a level one alarm is generated when the average deviation value of three consecutive sampling periods exceeds 0.25, and automatic correction is triggered when it exceeds 0.3.

[0156] The deviation control protocol comprises three parallel execution modules: a drill pressure adjustment module dynamically adjusts the rotation speed based on vibration spectrum characteristics, controlling the impact vibration energy within the range of 50J to 80J; a mud compensation module increases the mud specific gravity to a preset safety value at a rate of 0.5% per second; and a trajectory correction module applies a correction torque of 0.5kN·m to 1.2kN·m via a hydraulic deviation corrector, restoring the borehole inclination angle to within the allowable deviation range within five drilling cycles. This multi-parameter collaborative control mechanism ensures a dynamic match between the actual borehole trajectory and the predicted path in the risk heat map, effectively suppressing the accumulation of trajectory deviations caused by karst cave collapse.

[0157] For example, during the pile foundation drilling stage, a triaxial accelerometer is used to collect the vibration spectrum data of the drill bit in real time, with the vibration spectrum sampling frequency set to 2000Hz. At the same time, a mud density sensor is used to obtain the mud specific gravity value at a sampling rate of twice per second, and the hole inclination angle is monitored in real time with a resolution of 0.1 degrees by a high-precision tilt sensor.

[0158] The aforementioned sensor data is transmitted to the data processing center via an industrial bus and dynamically compared with a pre-generated risk heat map. The deviation calculation employs a weighted Euclidean distance algorithm, with the vibration spectrum deviation weight set at 0.6, the mud specific gravity deviation weight at 0.3, and the borehole inclination angle deviation weight at 0.1. When the overall deviation value exceeds the 0.85 threshold, the system automatically activates the deviation correction control module. This module comprises three parallel execution units: the first unit controls the hydraulic system to adjust the drilling pressure distribution for trajectory correction; the second unit starts the high-pressure grouting pump to perform wall reinforcement; and the third unit adjusts the main winch speed via a frequency converter to optimize the drilling speed.

[0159] This application further proposes a method for 3D data fusion calculation based on a preset spatial weight allocation strategy, including:

[0160] Different spatial weight coefficients were assigned to borehole data, radar data, and CT data in overlapping anomaly data groups, with borehole data receiving the highest weight, followed by radar data, and CT data receiving the lowest weight. Specifically, after completing spatial coordinate matching of standard borehole data, standard radar data, and standard CT data, weight allocation and fusion calculations were performed for each overlapping anomaly data group. The high weight of borehole data made it the primary source for cave boundary localization. In practice, when a borehole indicated the presence of a cave at a certain coordinate, that coordinate was still marked as a potential risk area even if radar and CT data did not detect any anomalies. The second-highest weight of radar data enhanced the ability to identify the direction of cave extension. In practice, in areas where the borehole spacing exceeded 10 meters, radar data could supplement the identification of laterally developed erosion zones. The lowest weight of CT data was used to correct local anomaly features. In practice, in densely borehole areas, CT data could assist in identifying micro-cavities with a diameter of less than 0.5 meters.

[0161] In a preferred embodiment, the borehole data, radar data, and CT data are normalized to a spatial coordinate system, unifying their spatial sampling rates to a 0.1 m × 0.1 m × 0.1 m voxel grid. In overlapping anomaly regions, core strength values ​​from the borehole data, reflected wave amplitude values ​​from the radar data, and density inversion values ​​from the CT data are extracted for each voxel unit. A linear weighted algorithm is used to superimpose the borehole data (0.5 weight), radar data (0.3 weight), and CT data (0.2 weight) to generate fused anomaly values. For non-overlapping regions, the original independent anomaly data are retained. An anomaly threshold is set during the fusion process; when the weighted calculation result exceeds the threshold, the voxel is determined to be a valid geological anomaly, ultimately constructing a three-dimensional fused geological anomaly model.

[0162] This application further proposes that an area meeting one of the following combinations of conditions should be identified as a high-risk karst cave development area:

[0163] The density of karst caves in the region reaches the first preset high density threshold and the number of anomalies in the region reaches the first preset high anomaly threshold.

[0164] Or the density of karst caves in the region reaches the second preset high density threshold and the number of abnormal points in the region reaches the second preset high abnormal point threshold.

[0165] Or the density of karst caves in the region reaches the third preset high density threshold and the number of abnormal points in the region reaches the third preset high abnormal point threshold.

[0166] Specifically, during the geological risk zoning process before pile foundation drilling, three-dimensional karst cave distribution data and stratigraphic anomaly detection results are obtained for the target area. The construction area is then layered by depth and divided into several cubic units. For each unit, the number of karst caves per unit volume is counted as the regional karst cave density. At the same time, the sum of rock fracture zones, fissure development points, and groundwater seepage points within that unit is counted as the regional anomaly number.

[0167] For example, in a certain area unit with a depth of 15-20 meters, the number of karst caves is measured to be 7 per cubic meter and the number of abnormal points is 25. At this time, the first high-risk judgment condition is triggered, the area is automatically marked as a red warning zone, and a corresponding grouting reinforcement plan is generated.

[0168] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0169] During the geological risk zoning process before pile foundation drilling, three-dimensional karst cave distribution data and stratigraphic anomaly detection results of the target area are obtained. The construction area is then layered by depth and divided into several cubic units. For each unit, the number of karst caves per unit volume is counted as the regional karst cave density. At the same time, the sum of rock fracture zones, fissure development points, and groundwater seepage points within the unit is counted as the regional anomaly number.

[0170] The system uses three preset threshold conditions to determine whether a unit is a high-risk karst cave development area. A unit is marked as such when it meets one of the following conditions:

[0171] The density of karst caves in the area is greater than or equal to 5 per cubic meter and the number of anomalies in the area is greater than or equal to 20.

[0172] Or the density of karst caves in the area is greater than or equal to 3 per cubic meter and the number of anomalies in the area is greater than or equal to 30;

[0173] Or the density of karst caves in the area is greater than or equal to 8 per cubic meter and the number of abnormal points in the area is greater than or equal to 10.

[0174] The embodiments of the present invention have at least the following beneficial effects:

[0175] By generating a set of karst cave distribution characteristics and a set of stratigraphic anomalies, determining a set of karst cave development trajectory sequences, and performing risk zoning, high-risk areas under complex karst geological conditions can be accurately identified. This allows construction personnel to understand the distribution and development of karst caves in advance, enabling them to take targeted risk prevention measures during pile foundation construction, effectively reducing construction risks such as borehole collapse, grout leakage, and stuck drill bits, and improving the safety and reliability of construction. Based on the set of high-risk karst cave development areas, a set of high-risk borehole locations and a set of high-risk stratigraphic nodes are determined, and a karst geological risk map is generated, which in turn generates a set of borehole drilling process paths and a borehole risk heat map. This series of steps achieves optimized planning of the borehole drilling process path, enabling the drill bit to effectively avoid karst caves and stratigraphic anomaly areas, reducing obstacles and risks during construction, improving borehole drilling efficiency, and ensuring borehole quality, which helps to shorten the construction cycle and reduce construction costs. The borehole risk heat map is overlaid on the pile foundation construction plan for display, and real-time drilling sensor data is acquired during construction. Borehole quality deviation alarms are generated based on deviation values, and an automatic correction control protocol is triggered when the alarm exceeds a preset threshold. This real-time monitoring and automatic correction mechanism can promptly detect and correct deviations during the construction process, ensuring that the construction process remains under control. This further improves the accuracy and quality of pile foundation drilling and enhances the stability and reliability of the construction process.

[0176] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for controlling the quality of pile foundation drilling in complex karst geology, characterized in that, include: Based on multi-source karst geological exploration data, a three-dimensional geological anomaly model characterizing the spatial distribution features of karst caves and stratigraphic anomalies is constructed through multi-modal data collaborative optimization processing. This includes: acquiring karst geological exploration data containing borehole core data, ground-penetrating radar data, and cross-hole CT imaging data; standardizing the karst geological exploration data to unify spatial benchmarks and data scales, resulting in standardized borehole data, radar data, and CT data; spatially registering and fusing the standardized borehole data, radar data, and CT data to generate effective geological anomaly data; extracting and generating a karst cave distribution feature set based on the effective geological anomaly data; and identifying geological anomalies associated with each karst cave feature in the karst cave distribution feature set to generate a stratigraphic anomaly point set. Based on the aforementioned three-dimensional geological anomaly model, the development trajectory sequence of karst caves is dynamically analyzed and reconstructed to form a set of karst cave development trajectory sequences containing information on dissolution zones and rock strata interfaces. Based on the aforementioned set of karst cave development trajectory sequences, and combined with the karst cave density and stratigraphic anomaly point distribution characteristics in the depth dimension, a three-dimensional dynamic risk assessment model is established to identify and delineate high-risk karst cave development zones. Based on the spatial distribution characteristics of the high-risk karst cave development area set and the preset karst cave development conditions, the high-risk pore location set is located and determined; Based on the set of high-risk karst cave development areas, stratigraphic nodes within their influence range are determined through spatial correlation analysis, forming a set of high-risk stratigraphic nodes; By integrating the set of karst cave distribution characteristics, the set of karst cave development trajectory sequences, the set of high-risk karst cave development areas, the set of high-risk pore locations, and the set of high-risk stratigraphic nodes, a multi-dimensional karst geological risk map is constructed. Based on the karst geological risk map, an optimized set of drilling process paths is intelligently generated, and a drilling risk heat map representing the spatial distribution of construction risks is generated accordingly. The borehole risk heat map is spatially overlaid and merged with the pile foundation construction plan to generate a pile foundation borehole risk heat map, which is then visualized.

2. The method according to claim 1, characterized in that, The process involves spatial registration and data fusion of standardized borehole data, radar data, and CT data to generate effective geological anomaly data, including: Spatial coordinate matching is performed on the standardized borehole data, radar data, and CT data to generate overlapping anomaly data sets and independent anomaly datasets; For each overlapping anomaly data group in the aforementioned overlapping anomaly data group set, a three-dimensional data fusion calculation is performed based on a preset spatial weight allocation strategy to generate a fused geological anomaly dataset. The fused geological anomaly dataset and the independent anomaly dataset are jointly identified as effective geological anomaly data.

3. The method according to claim 1, characterized in that, Based on the three-dimensional geological anomaly model, the development trajectory sequence of the karst caves is dynamically analyzed and reconstructed to form a set of karst cave development trajectory sequences containing information on dissolution zones and rock strata interfaces, including: By applying borehole trajectory positioning technology, the original spatial development point sequence of each karst cave in the karst cave distribution feature set is determined; The original spatial development point sequence is subjected to interpolation optimization under geological constraints to complete the missing trajectory segments caused by the sparse exploration points, and generate a continuous complete development point sequence. The completed development point sequence is subjected to geological noise filtering to remove interference points caused by measurement errors or local geological disturbances, thereby generating an accurate development point sequence. Based on the precise development point sequence, a set of cave development trajectory sequences containing information on dissolution zone boundaries and rock layer interfaces is constructed through geological law modeling and spatial curve fitting.

4. The method according to claim 1, characterized in that, Based on the aforementioned set of karst cave development trajectory sequences, and combined with the karst cave density and stratigraphic anomaly distribution characteristics at depth, a three-dimensional dynamic risk assessment model is established to identify and delineate high-risk karst cave development zones, including: The set of cave development trajectory sequences is sliced ​​at preset depth intervals to extract the target erosion zone set corresponding to each depth level. For the target dissolution zone sets at each depth level: Spatial grid-based risk zoning is carried out to form a set of risk zones for cave development that includes multiple target karst zones; Calculate the number of target karst zones per unit volume in each risk zone to obtain the regional karst cave density; Based on the set of stratigraphic anomalies, the number of associated stratigraphic anomalies within each risk zone is counted to obtain the number of regional anomalies. By combining the density of karst caves in the region with the number of anomalies in the region, and applying preset multi-dimensional high-risk judgment conditions, high-risk karst cave development areas at this depth level are identified and marked. High-risk karst cave development areas at various depth levels are compiled to form a three-dimensional set of high-risk karst cave development areas.

5. The method according to claim 1, characterized in that, The karst geological exploration data also includes drilling pressure dynamic change sequences, mud loss sequences, and core fragmentation rate sequences; and the method further includes: The drilling pressure dynamic change sequence, mud loss sequence, and core fragmentation rate sequence are defined as the first geological risk time series data. Temporal features are extracted from the first geological risk time series data to obtain geological risk time series features that characterize the dynamic risks of the drilling process; Based on the aforementioned geological risk time-series characteristics, the probability of potential borehole collapse during the drilling process is predicted; Based on the set of high-risk karst cave development areas, the set of high-risk borehole locations, and real-time drilling parameters, a second geological risk spatial data is generated. Multi-scale spatial feature extraction is performed on the second geological risk spatial data to obtain multi-scale geological risk features that reflect local and regional risks; Multi-scale analysis was performed on the temporal characteristics of the geological risks to obtain multi-scale temporal characteristics of the geological risks; The geological risk time series features and the geological risk time series multi-scale features are residually connected to preserve the original time series information and enhance the feature expressive power, thus obtaining the geological risk residual connection features. By integrating the multi-scale characteristics of geological risk with the residual connectivity characteristics of geological risk, a fused geological risk characteristic that comprehensively represents geological risk is formed; Based on the integrated geological risk characteristics, intelligent decision-making is made and borehole control commands are generated and transmitted to the pile foundation construction equipment for execution. The borehole control commands include dynamic adjustment commands for drilling speed, active reinforcement commands for wall protection, and adaptive compensation commands for grouting.

6. The method according to claim 1, characterized in that, The intelligently generated optimized drilling process path set based on karst geological risk maps includes: Based on the set of high-risk strata nodes in the karst geological risk map, a spatial node sequence for grouting reinforcement is planned. Based on the spatial morphological characteristics of the set of karst development trajectory sequences, a set of drill bit avoidance paths is generated; The grouting reinforcement spatial node sequence is spatiotemporally optimized and matched with the drill bit avoidance path set to generate a hole forming process path set.

7. The method according to claim 1, characterized in that, After visualizing the thermal map of the pile foundation drilling risk, the method further includes: Real-time acquisition of drilling sensor data, including drill bit vibration spectrum characteristics, dynamic value of mud specific gravity, and changes in borehole inclination angle; Calculate the dynamic deviation between the drilling sensor data and the predicted state of the pile foundation borehole risk thermal map; Based on the dynamic deviation value, a hole formation quality deviation alarm is generated; When the hole formation quality deviation alarm exceeds the preset safety threshold, the closed-loop control correction protocol is automatically triggered.

8. The method according to claim 2, characterized in that, The three-dimensional data fusion calculation based on the preset spatial weight allocation strategy includes: Different spatial weight coefficients were assigned to borehole data, radar data, and CT data in the overlapping anomaly data group, with borehole data assigned the highest weight, followed by radar data, and CT data assigned the lowest weight. Based on the assigned weight coefficients, the three-dimensional data of the overlapping areas are weighted and fused to generate fused geological anomaly data.

9. The method according to claim 4, characterized in that, A region is identified as a high-risk karst cave development area if it meets one of the following conditions: The density of karst caves in the region reaches the first preset high density threshold and the number of anomalies in the region reaches the first preset high anomaly threshold. Or the density of karst caves in the region reaches the second preset high density threshold and the number of abnormal points in the region reaches the second preset high abnormal point threshold. Or the density of karst caves in the region reaches the third preset high density threshold and the number of abnormal points in the region reaches the third preset high abnormal point threshold.

Images