Advanced geological forecasting method and system based on tunnel geological exploration

Abstract

The invention relates to the technical field of geological exploration, and discloses an advanced geological forecasting method and system based on tunnel geological exploration. The method comprises the following steps: performing data denoising on a multi-source data set pre-collected in a tunnel construction area to obtain an original data set; performing preprocessing and multi-stage data fusion on the original data set to obtain a fused data set; performing three-dimensional space modeling and feature extraction on the fused data set to obtain an intelligent geological prediction result; performing multi-factor risk assessment on the intelligent geological prediction result to obtain a geological risk distribution map in front of the tunnel; performing construction scheme optimization on the geological risk distribution map to obtain a construction guidance scheme; and carrying out comprehensive analysis and dynamic updating on the real-time monitoring data, the construction guidance scheme, the intelligent geological prediction result and the geological risk distribution map to obtain an advanced geological prediction result. According to the method and the system, the defects of insufficient prediction precision, reliability and accuracy of a traditional geological prediction method are overcome, and the safety risk is reduced.

Description

Technical Field

[0001] The present invention relates to the field of geological exploration technology, and in particular to an advanced geological prediction method and system based on tunnel geological exploration. Background Art

[0002] Advanced geological prediction in tunnel construction is a key technology to ensure construction safety and efficiency. Traditional advanced geological prediction methods mainly rely on a single detection method, such as geological radar, seismic wave method, etc., combined with experience and judgment to predict the geological conditions ahead. These methods can detect abnormal geological bodies ahead to a certain extent, but the prediction accuracy and reliability are limited.

[0003] However, the information obtained by a single detection method is often one-sided and difficult to fully reflect the complex geological conditions. At the same time, the traditional method lacks the ability to comprehensively analyze multi-source data and cannot fully utilize the advantages of various detection data. In addition, the traditional method lacks a dynamic update mechanism and cannot timely correct the prediction results based on real-time monitoring data during the construction process. These deficiencies limit the accuracy and timeliness of advanced geological forecasts and increase the safety risks of tunnel construction. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides an advanced geological prediction method and system based on tunnel geological detection, which is used to solve the defects of traditional geological prediction methods such as insufficient prediction accuracy, reliability and accuracy, and high safety risks, and significantly improves the comprehensiveness and accuracy of geological prediction.

[0005] The present invention provides an advanced geological prediction method based on tunnel geological detection, comprising: performing data denoising on a multi-source data set collected in advance in a tunnel construction area to obtain an original data set, wherein the multi-source data set includes historical geological data, real-time monitoring data, geological radar detection data, seismic wave detection data, advanced drilling data and infrared detection data; performing preprocessing and multi-level data fusion on the original data set to obtain a fused data set; performing three-dimensional spatial modeling and feature extraction on the fused data set to obtain an intelligent geological prediction result; performing multi-factor risk assessment on the intelligent geological prediction result to obtain a geological risk distribution map in front of the tunnel; performing construction plan optimization on the geological risk distribution map to obtain a construction guidance plan; performing comprehensive analysis and dynamic updating on the collected real-time monitoring data, the construction guidance plan, the intelligent geological prediction result and the geological risk distribution map to obtain an advanced geological prediction result.

[0006] The present invention also provides an advanced geological prediction system based on tunnel geological detection, comprising:

[0007] A denoising module is used to denoise the multi-source data set collected in advance in the tunnel construction area to obtain the original data set, wherein the multi-source data set includes historical geological data, real-time monitoring data, geological radar detection data, seismic wave detection data, advance drilling data and infrared detection data;

[0008] A fusion module, used for preprocessing and multi-level data fusion of the original data set to obtain a fused data set;

[0009] An extraction module, used for performing three-dimensional spatial modeling and feature extraction on the fused data set to obtain intelligent geological prediction results;

[0010] An evaluation module, used to perform a multi-factor risk evaluation on the intelligent geological prediction results to obtain a geological risk distribution map ahead of the tunnel;

[0011] An optimization module, used to optimize the construction plan based on the geological risk distribution map to obtain a construction guidance plan;

[0012] The updating module is used to comprehensively analyze and dynamically update the collected real-time monitoring data, the construction guidance plan, the intelligent geological prediction results and the geological risk distribution map to obtain advanced geological prediction results.

[0013] In the technical solution provided by the present invention, through multi-source data fusion technology, historical geological data, real-time monitoring data, geological radar detection data and other information are comprehensively utilized, which significantly improves the comprehensiveness and accuracy of geological prediction. The multi-level data processing and analysis process, including steps such as data cleaning, spatial registration and scale normalization, effectively improves the data quality and comparability, and lays a solid foundation for subsequent analysis. The application of three-dimensional spatial modeling and feature extraction technology makes the expression of geological structure more intuitive and accurate, which helps to identify potential risks under complex geological conditions. The probability distribution information contained in the intelligent geological prediction results provides a more scientific basis for risk assessment, making risk management more refined and quantitative. The multi-factor risk assessment method takes into account the interaction and weight of geological factors, and the generated geological risk distribution map is closer to the actual situation, which is conducive to the formulation of targeted construction plans. In the process of optimizing the construction plan, through steps such as mechanical analysis, support demand analysis and numerical simulation, the scientificity and economy of the excavation method and support parameters are ensured, and the safety and efficiency of tunnel construction are improved. The dynamic update mechanism is a major innovation of this method. Through the comparative analysis and correction of real-time monitoring data and prediction results, the geological forecast can be continuously optimized as the construction progresses, improving the timeliness and accuracy of the forecast. The introduction and weight distribution of prediction correction indicators make the update process more flexible and accurate, and can quickly respond to changes in actual geological conditions. Overlay analysis and dynamic risk assessment technology enable risk distribution maps to be updated in real time, providing a timely and reliable basis for construction decisions. The adaptability analysis and adjustment suggestions of the construction plan ensure that the construction process can flexibly respond to changes in geological conditions and improve the adaptability and risk resistance of the construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 It is a flow chart of an advanced geological prediction method based on tunnel geological detection in an embodiment of the present invention;

[0016] Figure 2 Schematic diagram of an advanced geological prediction system based on tunnel geological detection in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0019] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0020] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 , Figure 1 is a flow chart of an advanced geological prediction method based on tunnel geological detection according to an embodiment of the present invention, such as Figure 1 As shown, the following steps are included:

[0021] S101, performing data denoising on a multi-source data set collected in advance in a tunnel construction area to obtain an original data set, wherein the multi-source data set includes historical geological data, real-time monitoring data, geological radar detection data, seismic wave detection data, advance drilling data, and infrared detection data;

[0022] S102, preprocessing and multi-level data fusion are performed on the original data set to obtain a fused data set;

[0023] S103, performing three-dimensional spatial modeling and feature extraction on the fused data set to obtain intelligent geological prediction results;

[0024] S104, performing a multi-factor risk assessment on the intelligent geological prediction results to obtain a geological risk distribution map ahead of the tunnel;

[0025] S105, optimizing the construction plan based on the geological risk distribution map to obtain a construction guidance plan;

[0026] S106. Comprehensively analyze and dynamically update the collected real-time monitoring data, construction guidance plans, intelligent geological prediction results and geological risk distribution maps to obtain advanced geological forecast results.

[0027] Specifically, after pre-collecting multi-source data sets in the tunnel construction area, the data are first denoised. For historical geological data, digital processing technology is used to convert paper data into digital format and classify them in layers. Real-time monitoring data removes high-frequency noise through filtering algorithms, such as wavelet transform denoising method. Adaptive filtering technology is used to enhance effective signals for geological radar detection data. Spectral analysis is performed on seismic wave detection data to extract characteristic frequencies. Lead drilling data obtains lithology distribution information through core analysis. Temperature gradient analysis is performed on infrared detection data to identify abnormal areas. Preprocessing of the original data set and multi-level data fusion are key steps. First, spatial registration is performed on data from different sources to ensure that the data is in the same coordinate system. Then scale normalization is performed to unify data of different scales to the same dimension. Outlier detection and interpolation are performed on monitoring data to ensure data continuity. Stratum interface data are extracted through image segmentation algorithms and geometric correction is performed. Seismic velocity structure data are converted into lithology attributes through waveform inversion. Spatial statistical analysis of lithology distribution data is performed to obtain probability distribution. Finally, the Bayesian reasoning method is used to fuse multi-source data to obtain a comprehensive fused data set.

[0028] The three-dimensional spatial modeling of the fused data set uses the kriging interpolation method to generate a continuous geological attribute field, and then grid division is performed to form a discretized three-dimensional geological distribution field. The discretized data is classified into geological units and the boundaries are refined to construct a digital twin of the target geological structure. The random field of geological parameters is generated by the Monte Carlo simulation method to obtain multiple sets of geological parameter realizations. Statistical analysis is performed on multiple sets of realizations to extract the statistical characteristics and spatial correlation of geological parameters. Geostatistical methods such as variogram analysis and kriging interpolation are used to obtain the spatial distribution prediction map of geological parameters. Finally, cluster analysis is used to identify key risk areas, and geological conditions are inferred to obtain intelligent geological prediction results containing probability distribution information. When performing multi-factor risk assessment on the intelligent geological prediction results, the geological factors are first decomposed into multiple dimensions, such as lithology, structure, hydrology, etc. The analytic hierarchy process (AHP) is used to assign weights to each factor to obtain the factor weight matrix. The main control factor combination is extracted by principal component analysis. The main control factors are fuzzy comprehensively evaluated to obtain the risk level score. After the score is normalized, a continuous risk distribution field is generated by kriging interpolation. Finally, regional division and boundary optimization are carried out to obtain the geological risk distribution map in front of the tunnel.

[0029] When optimizing the construction plan based on the geological risk distribution map, the regions are first divided according to the risk level, and appropriate excavation methods are matched for different regions. Finite element analysis is performed on the preliminary excavation plan to calculate the stress distribution and safety factor. The support requirements are determined based on the analysis results, and the support parameters are optimized. The support effect is predicted by numerical simulation, and safety verification is performed. Comprehensive consideration is given to the excavation and support plans, and detailed construction steps and process flows are formulated. Finally, the construction cycle is estimated and resource allocation is optimized to obtain a complete construction guidance plan. In order to achieve dynamic updates, real-time monitoring data needs to be continuously collected. The monitoring data is cleaned and trended, compared with the intelligent geological prediction results, and the prediction deviation is calculated. Correction indicators are set according to the deviation, and the original prediction results are weighted and corrected. The updated prediction results are superimposed and analyzed with the risk distribution map to obtain dynamic risk assessment results. Based on the latest risk assessment results, the construction plan is adaptively analyzed and optimized. Finally, the optimized construction plan, updated geological prediction results and risk distribution are data fused to form the latest advanced geological prediction results.

[0030] For example, in the construction of a tunnel, reflected wave data was obtained through geological radar detection. First, the original data was denoised, and the wavelet transform method was used to remove high-frequency noise. The high-frequency coefficients were removed by threshold processing to obtain the denoised signal. Then the denoised signal was enhanced by using an adaptive filtering algorithm. The enhanced signal was obtained by iteratively optimizing the filter coefficients. The enhanced signal was image segmented to extract the stratum interface information. Finally, the processed radar data was fused with other detection data using the Dempster-Shafer evidence theory method. In this way, the fusion of multi-source data was achieved, providing a basis for subsequent geological prediction and risk assessment. In this case, through data processing and fusion, a fault zone was successfully identified 50 meters ahead of the tunnel. The fault zone was about 5 meters wide and had an inclination of 75 degrees. Based on this prediction result, the construction plan was adjusted accordingly, and short-footage excavation and advanced small-pipe support were used to effectively control the construction risks.

[0031] By executing the above steps, the comprehensiveness and accuracy of geological prediction are significantly improved through the use of multi-source data fusion technology, comprehensive use of historical geological data, real-time monitoring data, geological radar detection data and other information. The multi-level data processing and analysis process, including data cleaning, spatial registration and scale normalization, effectively improves the data quality and comparability, laying a solid foundation for subsequent analysis. The application of three-dimensional spatial modeling and feature extraction technology makes the expression of geological structure more intuitive and accurate, which helps to identify potential risks under complex geological conditions. The probability distribution information contained in the intelligent geological prediction results provides a more scientific basis for risk assessment, making risk management more refined and quantitative. The multi-factor risk assessment method takes into account the interaction and weight of geological factors, and the generated geological risk distribution map is closer to the actual situation, which is conducive to the formulation of targeted construction plans. In the process of optimizing the construction plan, through mechanical analysis, support demand analysis and numerical simulation, the scientificity and economy of the excavation method and support parameters are ensured, and the safety and efficiency of tunnel construction are improved. The dynamic update mechanism is a major innovation of this method. Through the comparative analysis and correction of real-time monitoring data and prediction results, the geological forecast can be continuously optimized as the construction progresses, improving the timeliness and accuracy of the forecast. The introduction and weight distribution of prediction correction indicators make the update process more flexible and accurate, and can quickly respond to changes in actual geological conditions. Overlay analysis and dynamic risk assessment technology enable risk distribution maps to be updated in real time, providing a timely and reliable basis for construction decisions. The adaptability analysis and adjustment suggestions of the construction plan ensure that the construction process can flexibly respond to changes in geological conditions and improve the adaptability and risk resistance of the construction.

[0032] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0033] (1) Digitally process historical geological data to obtain digital geological data, and classify the digital geological data into layers to obtain layered geological information;

[0034] (2) performing denoising and filtering processing on the real-time monitoring data to obtain cleaned monitoring data, and performing time series analysis on the cleaned monitoring data to obtain monitoring trend data;

[0035] (3) Performing signal enhancement processing on the geological radar detection data to obtain enhanced radar data, and performing image segmentation on the enhanced radar data to obtain segmented stratum interface data;

[0036] (4) Performing spectrum analysis on the seismic wave detection data to obtain spectrum characteristic data, and performing inversion calculation on the spectrum characteristic data to obtain formation velocity structure data;

[0037] (5) Conducting core analysis on the advanced drilling data to obtain lithology distribution data, and performing spatial interpolation on the lithology distribution data to obtain three-dimensional lithology distribution data;

[0038] (6) Perform temperature gradient analysis on the infrared detection data to obtain temperature anomaly area data, and perform cluster analysis on the temperature anomaly area data to obtain location information of potential geological anomaly points;

[0039] (7) Merge the layered geological information, monitoring trend data, segmented stratigraphic interface data, stratigraphic velocity structure data, three-dimensional lithology distribution data, and potential geological anomaly location information into the original data set.

[0040] Specifically, historical geological data are digitized, paper geological maps, drilling records, etc. are converted into digital formats, optical character recognition (OCR) technology is used to identify text information, and vector processing is used to extract geological boundaries and structural information. The digitized geological data are stratified and classified according to lithology, stratigraphic age and structural characteristics to form a stratified geological information database. Real-time monitoring data includes parameters such as displacement, stress and water pressure, which need to be denoised and filtered. The wavelet transform method is used to remove high-frequency noise, and the Kalman filter algorithm is used to eliminate random errors to obtain the cleaned monitoring data. The cleaned data is subjected to time series analysis, and the autoregressive integrated moving average (ARIMA) model is used to predict the changing trend of the monitoring parameters.

[0041] The signal-to-noise ratio of geological radar detection data is improved through signal enhancement processing. The radar signal is enhanced by frequency domain filtering and amplitude equalization technology, and then the formation interface is identified by the image segmentation algorithm based on regional growth to obtain the segmented formation interface data. The seismic wave detection data is subjected to spectrum analysis, and the spectrum features are extracted by fast Fourier transform (FFT). The spectrum feature data is used for inversion calculation, and the least squares method is used to invert the formation velocity structure data.

[0042] The advance drilling data obtains lithology distribution information through core analysis. The lithology distribution data is spatially interpolated, and the Kriging interpolation method is used to generate three-dimensional lithology distribution data to achieve spatial continuous expression of lithology information. Infrared detection data identifies abnormal areas through temperature gradient analysis. The spatial gradient of the temperature field is calculated, and the temperature anomaly area is determined using the threshold method. The K-means clustering algorithm is applied to the temperature anomaly area data to identify the location information of potential geological anomalies.

[0043] Finally, the layered geological information, monitoring trend data, segmented stratum interface data, stratum velocity structure data, three-dimensional lithology distribution data, and potential geological anomaly location information are integrated into a unified original data set. Geographic Information System (GIS) technology is used in the data integration process to establish a unified spatial reference system to ensure the consistency and comparability of various data in space.

[0044] Taking a tunnel project as an example, historical geological data include 1:5000 geological maps and 20 borehole data, which are digitally processed to form layered geological information including 5 stratigraphic units and 3 faults. Real-time monitoring data collected 1000 sets of displacement data, and 950 sets of valid data points were obtained after denoising and filtering. Geological radar detection was carried out along the axis of the tunnel, and 2000m long profile data were collected. 15 main stratigraphic interfaces were identified through signal enhancement and image segmentation. The seismic wave detection data was analyzed by spectrum analysis and inversion calculation to obtain the P wave velocity distribution within the depth range of 0-500m, and the velocity value varied between 3000-6000m / s. Five boreholes were drilled in front of the tunnel face in advance drilling, and the core analysis results were spatially interpolated to generate a three-dimensional lithology distribution model with a range of 100m×100m×50m. Infrared detection covered a range of 100m around the tunnel, temperature gradient analysis identified 3 temperature anomaly areas, and cluster analysis identified 5 potential geological anomalies.

[0045] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0046] (1) Performing spatial registration processing on the layered geological information to obtain registered geological data, and performing scale normalization on the registered geological data to obtain standardized geological data;

[0047] (2) performing outlier detection on the monitoring trend data to obtain outlier labeled data, and interpolating the outlier labeled data to obtain corrected monitoring data;

[0048] (3) extracting the segmented stratum interface data to obtain interface feature data, and geometrically correcting the interface feature data to obtain corrected stratum structure data;

[0049] (4) Performing waveform inversion on the formation velocity structure data to obtain velocity field data, and converting the velocity field data into geological parameters to obtain lithology attribute data;

[0050] (5) Performing spatial statistical analysis on the three-dimensional lithology distribution data to obtain a lithology probability distribution map, and performing threshold segmentation on the lithology probability distribution map to obtain the candidate geological body boundaries;

[0051] (6) Performing morphological processing on the location information of potential geological anomaly points to obtain the contour of the abnormal area, and performing feature matching on the contour of the abnormal area to obtain the target geological structure information;

[0052] (7) Multi-source data fusion is performed on the standardized geological data, the corrected monitoring data, the calibrated stratigraphic structure data, the lithology attribute data, the candidate geological body boundary and the target geological structure information to obtain a fused data set.

[0053] Specifically, in the advanced geological prediction method based on tunnel geological detection, data preprocessing and fusion are key steps. First, the layered geological information is spatially registered, and the geological data from different sources are unified into the same coordinate system using affine transformation and control point matching technology to obtain the registered geological data. Subsequently, the registered geological data is scale-normalized, and the minimum-maximum normalization method is used to unify the various geological parameters into the [0,1] interval to form standardized geological data. Next, the monitoring trend data is detected for outliers, and the moving window method and the 3σ criterion are used to identify abnormal data points to obtain outlier marked data. The marked outliers are interpolated using the linear interpolation method to ensure the continuity and reliability of the data, thereby obtaining the corrected monitoring data.

[0054] For the segmented formation interface data, the edge detection algorithm is used to extract the formation interface and obtain the interface feature data. Then, the polynomial fitting method is used to perform geometric correction on the interface feature data to eliminate the geometric deformation caused by the detection equipment or terrain, and finally the corrected formation structure data is obtained. The formation velocity structure data is processed by waveform inversion, and the full waveform inversion algorithm is used to solve the inverse problem of the wave equation through iterative optimization method to obtain detailed velocity field data. Subsequently, based on the rock physics model, the velocity field data is converted into lithology attribute data, including parameters such as porosity and permeability.

[0055] Perform spatial statistical analysis on the three-dimensional lithology distribution data, and use the variogram analysis and Kriging interpolation method to generate the lithology probability distribution map. By setting an appropriate probability threshold, segment the lithology probability distribution map to obtain the candidate geological body boundary. For the location information of potential geological anomalies, use morphological processing techniques such as expansion and erosion operations to extract the contour of the abnormal area. Perform feature matching on the extracted contours, and use methods such as template matching or feature descriptor comparison to identify the target geological structure information, such as faults, folds, etc.

[0056] Finally, the standardized geological data, the corrected monitoring data, the calibrated stratigraphic structure data, the lithology attribute data, the candidate geological body boundary and the target geological structure information are multi-sourced for data fusion. The Bayesian network model is used to generate a comprehensive fused data set by considering the uncertainty and correlation of each data source.

[0057] Taking a tunnel project as an example, the layered geological information contains 5 main stratigraphic units. After spatial registration processing, it achieves accurate correspondence with the engineering coordinate system, and the registration error is controlled within 0.5m. The scale normalization process unifies the lithology strength index to the 0-1 interval, which is convenient for subsequent analysis. 15 outlier points were found in the monitoring trend data. After correction by interpolation processing, the data continuity was significantly improved. The stratigraphic interface extraction identified 7 main interfaces, and the geometric correction eliminated the 2-3% deformation caused by the terrain undulation. The velocity field data was inverted to obtain the P-wave velocity distribution within 500m around the tunnel, with an accuracy of ±50m / s, and converted into lithology attribute data, including porosity (range 10-30%) and permeability (10^-3-10^-6m / s). The lithology probability distribution map generated by spatial statistical analysis shows that within the range of ±50m of the tunnel axis, sandstone accounts for 60%, mudstone accounts for 30%, and limestone accounts for 10%. Morphological processing and feature matching identified two major faults and one fold structure. After multi-source data fusion, a comprehensive data set including geological structure, physical properties and abnormal characteristics was formed, providing comprehensive and reliable data support for subsequent tunnel construction risk assessment and construction plan optimization.

[0058] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0059] (1) Perform spatial interpolation processing on the fused data set to obtain a continuously distributed geological attribute field, and then grid the continuously distributed geological attribute field to obtain a discretized three-dimensional geological distribution field;

[0060] (2) Classifying the geological units of the discretized three-dimensional geological distribution field to obtain a geological unit zoning map, and refining the boundaries of the geological unit zoning map to obtain a target geological structure twin;

[0061] (3) Assigning geological parameters to the target geological structure twin to obtain a parameterized geological twin, and quantifying the uncertainty of the parameterized geological twin to obtain a parameter probability density function;

[0062] (4) generating random fields for parameter probability density functions to obtain multiple sets of geological parameter realizations, and statistically analyzing the multiple sets of geological parameter realizations to obtain statistical characteristics of the geological parameters;

[0063] (5) Perform spatial correlation analysis on the statistical characteristics of geological parameters to obtain the geological parameter variation function, and perform geostatistical interpolation on the geological parameter variation function to obtain a geological parameter spatial distribution prediction map;

[0064] (6) Extracting abnormal features from the geological parameter spatial distribution prediction map to obtain potential geological abnormality areas, and calculating the risk probability of the potential geological abnormality areas to obtain a geological risk probability distribution map;

[0065] (7) Perform cluster analysis on the geological risk probability distribution map to obtain key risk areas, and infer the geological conditions of the key risk areas to obtain intelligent geological prediction results.

[0066] Specifically, the Kriging interpolation method is used to perform spatial interpolation processing on the fused data set to obtain a continuously distributed geological attribute field. This process takes into account spatial autocorrelation and ensures the smoothness and accuracy of the interpolation results. Subsequently, the octree algorithm is used to adaptively grid the continuously distributed geological attribute field to obtain a discretized three-dimensional geological distribution field. The accuracy of the grid division is dynamically adjusted according to the changing gradient of the geological attributes. The grid density is higher in complex areas, while the grid is sparser in uniform areas, which not only ensures the calculation accuracy but also improves the efficiency.

[0067] Next, the discretized three-dimensional geological distribution field is classified into geological units using the support vector machine (SVM) algorithm to obtain a geological unit zoning map. The SVM algorithm achieves accurate division of different geological units by establishing the optimal separation hyperplane. Then, the boundary of the geological unit zoning map is refined using the level set method to obtain the target geological structure twin. The level set method can effectively capture the complex geological interface morphology and improve the precision of geological structure expression. Assigning geological parameters to the target geological structure twin is the next key step. Here, the Bayesian inference method is used to comprehensively consider prior knowledge and observational data to obtain a parameterized geological twin. In order to quantify uncertainty, the Monte Carlo simulation method is used to sample the parameterized geological twin multiple times to generate a parameter probability density function.

[0068] Based on the parameter probability density function, the sequential Gaussian simulation algorithm is used to generate multiple sets of geological parameter realizations. This random field generation method can maintain the spatial structural characteristics of the original data while reflecting the uncertainty of the parameters. Statistical analysis is performed on multiple sets of geological parameter realizations, and statistical characteristics such as mean, variance, skewness and kurtosis are calculated to obtain the statistical characteristics of the geological parameters. Subsequently, the spatial correlation analysis of the statistical characteristics of the geological parameters is performed, and the spatial autocorrelation structure of the geological parameters is calculated using the variogram method to obtain the geological parameter variogram. The variogram describes the spatial variation law of the geological parameters and provides a theoretical basis for subsequent interpolation. The ordinary kriging method is used to perform geostatistical interpolation on the geological parameter variogram to obtain a high-precision prediction map of the spatial distribution of geological parameters.

[0069] Extracting abnormal features from the prediction map of the spatial distribution of geological parameters is the key to identifying potential geological risks. Here, the local anomaly factor (LOF) algorithm is used to identify potential geological anomaly areas by calculating the local density ratio of the data point and its neighborhood. For the identified potential geological anomaly areas, the logistic regression model is used to calculate the risk probability of each abnormal area in combination with historical data and expert knowledge, thereby obtaining a geological risk probability distribution map. Finally, the DBSCAN density clustering algorithm is applied to the geological risk probability distribution map to identify key risk areas. The DBSCAN algorithm can effectively handle clusters of irregular shapes and is suitable for identifying risk areas under complex geological conditions. For the identified key risk areas, the decision tree algorithm is used to infer geological conditions, and a variety of geological factors are comprehensively considered to obtain the final intelligent geological prediction results.

[0070] For example, in a tunnel project with a length of 5 kilometers, the fused data set was first interpolated by Kriging to generate a continuous geological attribute field with a resolution of 100m×100m×10m. The adaptive grid division was performed by the octree algorithm, and the grid accuracy in complex areas such as faults was improved to 10m×10m×1m, while the resolution of 100m×100m×10m was maintained in the uniform rock area, and finally a discretized three-dimensional geological distribution field containing about 5 million grid cells was formed. The SVM algorithm classified these grid cells and identified five major geological units, including sandstone, mudstone, limestone, granite and metamorphic rock. The level set method refined the unit boundaries and accurately portrayed the geometric morphology of two major faults and a complex fold structure. Bayesian inference combined the measured data and the regional geological background to assign parameters such as lithology strength, porosity, and permeability to each geological unit. Monte Carlo simulation generated 1,000 sets of parameter samples and constructed the parameter probability density function. Sequential Gaussian simulation generates 100 sets of geological parameter realizations based on these probability density functions. Statistical analysis shows that the average uniaxial compressive strength of sandstone along the tunnel is 50 MPa with a standard deviation of 5 MPa; the average permeability of mudstone is 1×10 -6 m / s, showing a log-normal distribution.

[0071] The variogram analysis showed that the correlation distance of lithology strength in the horizontal direction was about 200m and in the vertical direction was about 20m. Ordinary Kriging interpolation generated a prediction map of the spatial distribution of geological parameters with a resolution of 10m×10m×1m. The LOF algorithm identified 15 potential geological anomaly areas in the prediction map, of which 3 were located near faults and 2 were located in the core of the fold. The logistic regression model calculated that the geological risk probability of these abnormal areas was distributed between 0.1 and 0.8. The DBSCAN algorithm finally clustered 4 key risk areas, and the decision tree analysis inferred that these areas may have geological disaster risks such as water inrush, karst collapse and rock burst.

[0072] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0073] (1) Decomposing the intelligent geological prediction results into geological factors to obtain multidimensional geological factor data, and assigning weights to the multidimensional geological factor data to obtain a factor weight matrix;

[0074] (2) Perform hierarchical analysis on the factor weight matrix to obtain the factor importance ranking, and perform threshold screening on the factor importance ranking to obtain the set of key influencing factors;

[0075] (3) Conduct correlation analysis on the key influencing factor set to obtain the factor correlation coefficient matrix, and extract the principal components of the factor correlation coefficient matrix to obtain the main control factor combination;

[0076] (4) Perform fuzzy comprehensive evaluation on the combination of main control factors to obtain risk level scores, and normalize the risk level scores to obtain standardized risk indexes;

[0077] (5) Perform spatial interpolation on the standardized risk index to obtain a continuous risk distribution field, and divide the continuous risk distribution field into regions to obtain a risk zoning map;

[0078] (6) The risk zoning map is optimized to obtain refined risk areas, and risk levels are assigned to the refined risk areas to obtain a geological risk distribution map in front of the tunnel.

[0079] Specifically, the intelligent geological prediction results are decomposed into geological factors, and the principal component analysis (PCA) method is used to decompose the complex geological conditions into multiple independent geological factors to obtain multidimensional geological factor data. Subsequently, the analytic hierarchy process (AHP) is used to assign weights to these geological factors and construct a factor weight matrix. The AHP method ensures the scientificity and reliability of weight assignment through expert scoring and consistency testing.

[0080] Perform hierarchical analysis on the factor weight matrix, use the eigenvalue method to calculate the relative importance of each factor, and get the factor importance ranking. Set the importance threshold λ, filter out factors with importance greater than λ, and form a set of key influencing factors. This process can be expressed by the following formula:

[0081]

[0082] Among them, W i represents the weight of the i-th factor, a ijIt represents the importance ratio of the i-th factor to the j-th factor, and m and n are the number of rows and columns respectively. Perform correlation analysis on the key influencing factor set, calculate the Pearson correlation coefficient, and obtain the factor correlation coefficient matrix R. Then perform principal component analysis on R, extract the principal components with eigenvalues ​​greater than 1, and obtain the main control factor combination. The calculation formula of principal component analysis is as follows:

[0083]

[0084] Here, F k represents the kth principal component, l ki is the i-th component of the eigenvector corresponding to the k-th principal component, X i is the original variable, and p is the total number of variables. The combination of main control factors is evaluated comprehensively, and the membership function and fuzzy evaluation matrix are established. The risk level score is obtained through fuzzy synthesis operation. The score result is processed by maximum-minimum normalization to obtain the standardized risk index.

[0085] The standardized risk index is spatially interpolated, and the inverse distance weighted method (IDW) is used to generate a continuous risk distribution field. Then, the K-means clustering algorithm is used to divide the continuous risk distribution field into regions to obtain a preliminary risk zoning map. In order to optimize the boundaries of the risk zoning, the active contour model (Snake algorithm) is used to refine the boundaries and obtain refined risk areas. Finally, based on the numerical range of the risk index, the risk level is assigned to the refined risk area to generate the final geological risk distribution map ahead of the tunnel.

[0086] For example, in a tunnel project with a length of 10 kilometers, the intelligent geological prediction results were decomposed into 8 main geological factors through PCA, including lithology strength, groundwater content, fault density, karst development degree, etc. The AHP analysis results show that the weight of lithology strength is 0.3, groundwater content is 0.25, fault density is 0.2, karst development degree is 0.15, and other factors account for 0.1. The importance threshold λ=0.1 is set, and the top 5 factors are selected as the key influencing factor set. Correlation analysis shows that the correlation coefficient between lithology strength and fault density is -0.6, showing a strong negative correlation. Principal component analysis extracts 2 principal components, and the cumulative explained variance reaches 85%. The first principal component mainly reflects the integrity of the rock mass, and the second principal component mainly reflects the hydrogeological conditions.

[0087] Fuzzy comprehensive evaluation established five risk levels (very low, low, medium, high, and very high), and the risk level score was obtained through expert scoring and fuzzy synthesis. After normalization, the risk index was distributed between 0 and 1. Spatial interpolation generated a continuous risk distribution field with a resolution of 25m×25m×5m. K-means clustering (K=5) preliminarily divided five risk areas, and the risk boundary that was more in line with geological reality was obtained after optimization of the Snake algorithm. The final risk distribution map showed that there were three high-risk areas along the tunnel, located in the 2km-3km, 5.5km-6km and 8km-9km sections, which were mainly affected by fault fracture zones and karst development. The medium-risk area accounted for 30% of the total length, mainly distributed in the 4km-5km and 7km-8km sections, corresponding to the groundwater-rich areas. The low-risk and very low-risk areas accounted for 50% of the total length, mainly complete hard rock sections.

[0088] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0089] (1) Divide the geological risk distribution map into regions to obtain risk level zones, and match the excavation methods to the risk level zones to obtain a preliminary excavation plan;

[0090] (2) Conduct mechanical analysis on the preliminary excavation plan to obtain stress distribution data, calculate the safety factor of the stress distribution data, and obtain the excavation stability assessment result;

[0091] (3) Analyze the support demand based on the excavation stability assessment results, obtain support type recommendations, and optimize the parameters of the support type recommendations to obtain a preliminary support plan;

[0092] (4) Conduct numerical simulation on the preliminary support scheme to obtain support effect prediction data, and perform safety verification on the support effect prediction data to obtain optimized support parameters;

[0093] (5) Comprehensively analyze the optimized support parameters and preliminary excavation plan to obtain the construction step arrangement, and optimize the construction step arrangement to obtain the preliminary construction process;

[0094] (6) Estimate the construction period of the preliminary construction process to obtain a construction schedule, and optimize the resource allocation of the construction schedule to obtain a construction guidance plan.

[0095] Specifically, the cluster analysis method is used to divide the geological risk distribution map into regions to obtain risk level zones. Specifically, the K-means clustering algorithm is used to cluster regions with similar risk indexes into one category to form different risk level zones. Subsequently, based on the risk level zones and geological conditions, the expert system method is used to match the excavation methods of each zone to obtain a preliminary excavation plan. Mechanical analysis of the preliminary excavation plan is the next important step. The finite element method is used to perform numerical simulation of the excavation process to obtain stress distribution data. The calculation formula for stress distribution is:

[0096]

[0097] Among them, σ ij represents the stress tensor, u i and u j represent the displacement components in the i and j directions respectively, x j and x i is the spatial coordinate. Based on the stress distribution data, the safety factor of the excavation surface is calculated. The safety factor is calculated using the Mohr-Coulomb strength criterion:

[0098]

[0099] Here, FS is the safety factor, c is the rock mass cohesion, σ n is the normal stress, φ is the internal friction angle, and τ is the shear stress. By comparing the calculated safety factor with the set critical value, the excavation stability assessment result is obtained. The support demand analysis is carried out on the excavation stability assessment result. The fuzzy reasoning system is used to consider factors such as rock mass quality, groundwater conditions, and excavation section size to obtain support type recommendations. Then, the particle swarm optimization algorithm is used to optimize the support parameters and obtain a preliminary support plan. The objective function of particle swarm optimization can be expressed as:

[0100] f(x)=w1·C(x)+w2·S(x)

[0101] Among them, f(x) represents the support parameter vector, C(x) is the support cost function, S(x) is the support safety function, and w1 and w2 are weight coefficients. The preliminary support scheme is numerically simulated, and the three-dimensional finite difference method is used to simulate the interaction between the support structure and the surrounding rock to obtain the support effect prediction data. The prediction data includes the stress, deformation of the support structure, and displacement of the surrounding rock. The predicted data is safety checked, and the limit state method is used to calculate the bearing capacity of the support structure and the stability of the surrounding rock to obtain the optimized support parameters. The optimized support parameters are comprehensively analyzed with the preliminary excavation plan, and the critical path method (CPM) is used to determine the construction sequence. The CPM method considers the duration and interdependence of each process and calculates the optimal construction sequence. The construction sequence is optimized, and the value engineering method is used to weigh the construction efficiency, safety and economy to obtain the preliminary construction process.

[0102] Finally, the construction period of the preliminary construction process is estimated using the PERT (Project Evaluation and Review Technique) method, taking into account the optimal, most likely, and worst times for each process, and calculating the expected construction period and standard deviation. The PERT estimation formula is:

[0103]

[0104] Among them, T e is the expected construction period, T o is the optimal time, T m is the most likely time, T pThe worst time is the construction schedule. Based on the estimation results, a construction schedule is formulated, and the resources are optimally allocated using the linear programming method, and finally a detailed construction guidance plan is obtained. For example, in a tunnel project with a length of 5 kilometers, the geological risk distribution map is divided into five risk level areas by K-means clustering (K=5). The high-risk area (1.5km-2km section) is excavated by the step method, the medium-risk area (0-1.5km and 2km-3.5km section) is excavated by the full section method, and the low-risk area (3.5km-5km section) is excavated by the smooth blasting method. Mechanical analysis shows that the maximum principal stress in the high-risk area reaches 25MPa, and the safety factor is 1.2, which is lower than the set critical value of 1.5, and the support needs to be strengthened. The fuzzy reasoning system recommends the use of steel arch frame combined with anchor support in high-risk areas. The preliminary design is an H200 steel arch frame with a spacing of 0.8m, combined with a system anchor with L=4m and a spacing of 1m×1m. After optimization by particle swarm optimization algorithm, the spacing of steel arch frames was adjusted to 0.6m, and the length of anchor rods was increased to 4.5m. Numerical simulation results show that the optimized support scheme reduces the maximum deformation of surrounding rock from 15cm to 8cm, meeting the safety requirements. CPM analysis determined the key construction steps of "advanced geological prediction-face excavation-initial support-secondary lining". The total construction period of the tunnel estimated by PERT method is 550 days, with a standard deviation of 30 days. After optimal resource allocation, a plan for simultaneous advancement of three construction working faces was determined, which is expected to shorten the construction period to 500 days.

[0105] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0106] (1) Perform data cleaning on real-time monitoring data to obtain effective monitoring data, and perform trend analysis on the effective monitoring data to obtain the change trend of monitoring parameters;

[0107] (2) Compare and analyze the monitoring parameter change trend with the intelligent geological prediction results to obtain the prediction deviation data, and make threshold judgment on the prediction deviation data to obtain the prediction correction index;

[0108] (3) weighting the prediction correction indicators to obtain a correction weight matrix, and weightedly fusion the correction weight matrix with the intelligent geological prediction results to obtain updated geological prediction results;

[0109] (4) Superimpose and analyze the updated geological prediction results and the geological risk distribution map to obtain dynamic risk assessment results, and classify the dynamic risk assessment results to obtain an updated risk distribution map;

[0110] (5) Conduct adaptability analysis on the updated risk distribution map and construction guidance plan to obtain plan adjustment suggestions, and conduct feasibility assessment on the plan adjustment suggestions to obtain an optimized construction plan;

[0111] (6) The optimized construction plan, the updated geological prediction results and the updated risk distribution map are integrated to obtain the advanced geological prediction results.

[0112] Specifically, the real-time monitoring data is cleaned, and anomaly detection algorithms such as the local outlier factor (LOF) method are used to identify and remove abnormal data points. At the same time, moving average filtering is used to remove high-frequency noise to obtain effective monitoring data. Then, time series analysis methods such as the autoregressive integrated moving average (ARIMA) model are applied to the effective monitoring data to perform trend analysis on the monitoring parameters to obtain the changing trend of the monitoring parameters. The changing trend of the monitoring parameters is compared and analyzed with the intelligent geological prediction results. The dynamic time warping (DTW) algorithm is used to calculate the similarity between the two, and the prediction deviation data is obtained by the difference method. A threshold is set for the prediction deviation data, such as using the 3σ criterion, and the deviation exceeding the threshold is marked as an item that needs to be corrected, so as to obtain the prediction correction index. The next key step is to assign weights to the prediction correction index. The analytic hierarchy process (AHP) is used to consider the importance and reliability of each indicator, construct a judgment matrix, calculate the eigenvector, and obtain the correction weight matrix. The correction weight matrix is ​​weighted and fused with the intelligent geological prediction results, and the Bayesian update method is used to comprehensively consider the prior prediction results and new observation data to obtain the updated geological prediction results.

[0113] Subsequently, the updated geological prediction results are superimposed and analyzed with the original geological risk distribution map. The spatial analysis function of the geographic information system (GIS) is used to perform raster operations to obtain dynamic risk assessment results. Cluster analysis methods, such as the fuzzy C-means (FCM) clustering algorithm, are applied to the dynamic risk assessment results to divide the risk values ​​into different levels and generate an updated risk distribution map. The updated risk distribution map is analyzed for adaptability with the existing construction guidance plan. The decision tree algorithm is used to evaluate the applicability of the current construction plan based on the new risk distribution, and suggestions for plan adjustment are obtained. Then, a multi-criteria decision analysis (MCDA) method, such as the TOPSIS method, is used to conduct a feasibility assessment of the plan adjustment suggestions, and the optimized construction plan is obtained by comprehensively considering factors such as technical feasibility, economy, and safety.

[0114] Finally, the optimized construction plan, updated geological prediction results and updated risk distribution map are fused. The Dempster-Shafer evidence theory is used to consider the reliability and uncertainty of each data source, and the final advanced geological prediction results are obtained through evidence synthesis rules.

[0115] For example, in a tunnel project with a length of 8 kilometers, the real-time monitoring system collects data once an hour, including parameters such as surrounding rock displacement, groundwater level and borehole stress. During the data cleaning process, 15 outliers were identified by the LOF algorithm, and the moving average filter reduced the noise level by 60%. The ARIMA model analysis showed that in the 3km-4km section, the surrounding rock displacement showed an accelerated growth trend, with the growth rate rising from 0.5mm / day to 1.2mm / day. The similarity between the prediction results calculated by the DTW algorithm and the actual monitoring data was 0.85, and the differential analysis found that there was a significant deviation at 3.5km, exceeding the set 2σ threshold. The AHP analysis results showed that the weights of surrounding rock displacement, groundwater level and borehole stress were 0.5, 0.3 and 0.2 respectively. After the Bayesian update, the geological prediction results at 3.5km were adjusted from the original "medium risk" to "high risk". The results of GIS spatial analysis showed that the area of ​​high-risk areas increased by 15%. FCM clustering reclassified the risk level into 5 levels, among which the 3.4km-3.6km section was classified as the highest risk level.

[0116] The decision tree analysis suggested that the advance small pipe should be used to strengthen the support in the 3.4km-3.6km section. The TOPSIS evaluation showed that the feasibility score of this adjustment suggestion was 0.82, which was higher than other options. After integrating various data, the Dempster-Shafer evidence theory finally determined that there was a fault fracture zone in the 3.4km-3.6km section, and the predicted water inflow could reach 300-500m 3 / h, curtain grouting and double-layer support are recommended.

[0117] The embodiment of the present invention also provides an advanced geological prediction system based on tunnel geological detection, such as Figure 2 As shown, the advanced geological prediction system based on tunnel geological detection specifically includes:

[0118] De-noising module 201, used for performing data denoising on a multi-source data set collected in advance in a tunnel construction area to obtain an original data set, wherein the multi-source data set includes historical geological data, real-time monitoring data, geological radar detection data, seismic wave detection data, advance drilling data and infrared detection data;

[0119] A fusion module 202 is used to perform preprocessing and multi-level data fusion on the original data set to obtain a fused data set;

[0120] An extraction module 203 is used to perform three-dimensional spatial modeling and feature extraction on the fused data set to obtain intelligent geological prediction results;

[0121] An evaluation module 204 is used to perform a multi-factor risk evaluation on the intelligent geological prediction result to obtain a geological risk distribution map ahead of the tunnel;

[0122] An optimization module 205 is used to optimize the construction plan based on the geological risk distribution map to obtain a construction guidance plan;

[0123] The updating module 206 is used to comprehensively analyze and dynamically update the collected real-time monitoring data, the construction guidance plan, the intelligent geological prediction results and the geological risk distribution map to obtain advanced geological prediction results.

[0124] Through the collaborative work of the above modules, through multi-source data fusion technology, comprehensive use of historical geological data, real-time monitoring data, geological radar detection data and other information, the comprehensiveness and accuracy of geological prediction have been significantly improved. The multi-level data processing and analysis process, including data cleaning, spatial registration and scale normalization, effectively improves the data quality and comparability, laying a solid foundation for subsequent analysis. The application of three-dimensional spatial modeling and feature extraction technology makes the expression of geological structure more intuitive and accurate, which helps to identify potential risks under complex geological conditions. The probability distribution information contained in the intelligent geological prediction results provides a more scientific basis for risk assessment, making risk management more refined and quantitative. The multi-factor risk assessment method takes into account the interaction and weight of geological factors, and the generated geological risk distribution map is closer to the actual situation, which is conducive to the formulation of targeted construction plans. In the process of optimizing the construction plan, through mechanical analysis, support demand analysis and numerical simulation, the scientificity and economy of the excavation method and support parameters are ensured, and the safety and efficiency of tunnel construction are improved. The dynamic update mechanism is a major innovation of this method. Through the comparative analysis and correction of real-time monitoring data and prediction results, the geological forecast can be continuously optimized as the construction progresses, improving the timeliness and accuracy of the forecast. The introduction and weight distribution of prediction correction indicators make the update process more flexible and accurate, and can quickly respond to changes in actual geological conditions. Overlay analysis and dynamic risk assessment technology enable risk distribution maps to be updated in real time, providing a timely and reliable basis for construction decisions. The adaptability analysis and adjustment suggestions of the construction plan ensure that the construction process can flexibly respond to changes in geological conditions and improve the adaptability and risk resistance of the construction.

[0125] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the embodiments, a person skilled in the art should understand that the specific implementation modes of the present invention can still be modified or replaced by equivalents, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. An advanced geological prediction method based on tunnel geological detection, characterized in that: include: De-noising the multi-source data set collected in advance in the tunnel construction area to obtain the original data set, where the multi-source data set includes historical geological data, real-time monitoring data, geological radar detection data, seismic wave detection data, advance drilling data and infrared detection data; Preprocessing and multi-level data fusion are performed on the original data set to obtain a fused data set; Performing three-dimensional spatial modeling and feature extraction on the fused data set to obtain intelligent geological prediction results; Performing a multi-factor risk assessment on the intelligent geological prediction results to obtain a geological risk distribution map ahead of the tunnel; Optimizing the construction plan based on the geological risk distribution map to obtain a construction guidance plan; The collected real-time monitoring data, the construction guidance plan, the intelligent geological prediction results and the geological risk distribution map are comprehensively analyzed and dynamically updated to obtain advanced geological prediction results.

2. The advanced geological prediction method based on tunnel geological detection according to claim 1 is characterized in that: The step of performing data denoising on a multi-source data set collected in advance in the tunnel construction area to obtain an original data set, wherein the multi-source data set includes historical geological data, real-time monitoring data, geological radar detection data, seismic wave detection data, advance drilling data and infrared detection data, comprises: Digitally process historical geological data to obtain digital geological data, and classify the digital geological data into layers to obtain layered geological information; Perform denoising and filtering on the real-time monitoring data to obtain cleaned monitoring data, and perform time series analysis on the cleaned monitoring data to obtain monitoring trend data; Perform signal enhancement processing on geological radar detection data to obtain enhanced radar data, and perform image segmentation on the enhanced radar data to obtain segmented stratum interface data; Perform spectrum analysis on the seismic wave detection data to obtain spectrum characteristic data, and perform inversion calculation on the spectrum characteristic data to obtain formation velocity structure data; Conduct core analysis on the advanced drilling data to obtain lithology distribution data, and perform spatial interpolation on the lithology distribution data to obtain three-dimensional lithology distribution data; Conduct temperature gradient analysis on infrared detection data to obtain temperature anomaly area data, and perform cluster analysis on temperature anomaly area data to obtain location information of potential geological anomaly points; The layered geological information, the monitoring trend data, the segmented stratum interface data, the stratum velocity structure data, the three-dimensional lithology distribution data and the potential geological anomaly point location information are combined into the original data set.

3. The advanced geological prediction method based on tunnel geological detection according to claim 2 is characterized in that: The step of preprocessing the original data set and fusing the multi-level data to obtain a fused data set includes: Performing spatial registration processing on the layered geological information to obtain registered geological data, and performing scale normalization on the registered geological data to obtain standardized geological data; Performing outlier detection on the monitoring trend data to obtain outlier marked data, and performing interpolation processing on the outlier marked data to obtain corrected monitoring data; Extracting the segmented stratum interface data to obtain interface feature data, and geometrically correcting the interface feature data to obtain corrected stratum structure data; Performing waveform inversion on the formation velocity structure data to obtain velocity field data, and performing geological parameter conversion on the velocity field data to obtain lithology attribute data; Performing spatial statistical analysis on the three-dimensional lithology distribution data to obtain a lithology probability distribution map, and performing threshold segmentation on the lithology probability distribution map to obtain candidate geological body boundaries; Performing morphological processing on the position information of the potential geological anomaly point to obtain the contour of the abnormal area, and performing feature matching on the contour of the abnormal area to obtain target geological structure information; The standardized geological data, the modified monitoring data, the corrected stratigraphic structure data, the lithology attribute data, the candidate geological body boundary and the target geological structure information are subjected to multi-source data fusion to obtain the fused data set.

4. The advanced geological prediction method based on tunnel geological detection according to claim 1 is characterized in that: The step of performing three-dimensional spatial modeling and feature extraction on the fused data set to obtain intelligent geological prediction results includes: Performing spatial interpolation processing on the fused data set to obtain a continuously distributed geological attribute field, and performing grid division on the continuously distributed geological attribute field to obtain a discretized three-dimensional geological distribution field; Classifying the discretized three-dimensional geological distribution field into geological units to obtain a geological unit zoning map, and refining the boundaries of the geological unit zoning map to obtain a target geological structure twin; Assigning geological parameters to the target geological structure twin to obtain a parameterized geological twin, and quantifying uncertainty of the parameterized geological twin to obtain a parameter probability density function; Performing random field generation on the parameter probability density function to obtain multiple groups of geological parameter realizations, and performing statistical analysis on the multiple groups of geological parameter realizations to obtain statistical characteristics of the geological parameters; Performing spatial correlation analysis on the statistical characteristics of the geological parameters to obtain a geological parameter variation function, and performing geostatistical interpolation on the geological parameter variation function to obtain a geological parameter spatial distribution prediction map; Extracting abnormal features from the geological parameter spatial distribution prediction map to obtain potential geological abnormality areas, and calculating risk probability for the potential geological abnormality areas to obtain a geological risk probability distribution map; Cluster analysis is performed on the geological risk probability distribution map to obtain key risk areas, and geological conditions are inferred for the key risk areas to obtain the intelligent geological prediction results.

5. The advanced geological prediction method based on tunnel geological detection according to claim 1 is characterized in that: The step of performing a multi-factor risk assessment on the intelligent geological prediction result to obtain a geological risk distribution map ahead of the tunnel includes: Decomposing the intelligent geological prediction results into geological factors to obtain multidimensional geological factor data, and assigning weights to the multidimensional geological factor data to obtain a factor weight matrix; Performing hierarchical analysis on the factor weight matrix to obtain the factor importance ranking, and performing threshold screening on the factor importance ranking to obtain a set of key influencing factors; Performing correlation analysis on the key influencing factor set to obtain a factor correlation coefficient matrix, and performing principal component extraction on the factor correlation coefficient matrix to obtain a main control factor combination; Performing fuzzy comprehensive evaluation on the combination of the main control factors to obtain a risk level score, and normalizing the risk level score to obtain a standardized risk index; Performing spatial interpolation on the standardized risk index to obtain a continuous risk distribution field, and performing regional division on the continuous risk distribution field to obtain a risk zoning map; The risk zoning map is subjected to boundary optimization to obtain refined risk areas, and risk levels are assigned to the refined risk areas to obtain a geological risk distribution map in front of the tunnel.

6. The method for advanced geological prediction based on tunnel geological detection according to claim 1, characterized in that: The step of optimizing the construction plan for the geological risk distribution map to obtain a construction guidance plan includes: Divide the geological risk distribution map into regions to obtain risk level zones, and match the risk level zones with excavation methods to obtain a preliminary excavation plan; Performing mechanical analysis on the preliminary excavation plan to obtain stress distribution data, and performing safety factor calculation on the stress distribution data to obtain an excavation stability assessment result; Perform support demand analysis on the excavation stability assessment results to obtain support type recommendations, and perform parameter optimization on the support type recommendations to obtain a preliminary support plan; Performing numerical simulation on the preliminary support scheme to obtain support effect prediction data, and performing safety verification on the support effect prediction data to obtain optimized support parameters; Comprehensively analyzing the optimized support parameters and the preliminary excavation plan to obtain a construction step arrangement, and optimizing the construction step arrangement to obtain a preliminary construction process; The construction period of the preliminary construction process is estimated to obtain a construction schedule, and resource allocation optimization is performed on the construction schedule to obtain the construction guidance plan.

7. The advanced geological prediction method based on tunnel geological detection according to claim 6 is characterized in that: The step of comprehensively analyzing and dynamically updating the collected real-time monitoring data, the construction guidance plan, the intelligent geological prediction results and the geological risk distribution map to obtain the advanced geological prediction results includes: Performing data cleaning on the real-time monitoring data to obtain effective monitoring data, and performing trend analysis on the effective monitoring data to obtain a monitoring parameter change trend; Comparing and analyzing the change trend of the monitoring parameters with the intelligent geological prediction results to obtain prediction deviation data, and performing threshold judgment on the prediction deviation data to obtain prediction correction indicators; Performing weight allocation on the prediction correction index to obtain a correction weight matrix, and weighted fusion of the correction weight matrix and the intelligent geological prediction result to obtain an updated geological prediction result; Performing a superposition analysis on the updated geological prediction results and the geological risk distribution map to obtain a dynamic risk assessment result, and classifying the dynamic risk assessment result to obtain an updated risk distribution map; Performing adaptability analysis on the updated risk distribution map and the construction guidance plan to obtain plan adjustment suggestions, and performing feasibility assessment on the plan adjustment suggestions to obtain an optimized construction plan; The optimized construction plan, the updated geological prediction result and the updated risk distribution map are subjected to data fusion to obtain the advanced geological prediction result.

8. An advanced geological prediction system based on tunnel geological detection, used to execute the advanced geological prediction method based on tunnel geological detection as claimed in any one of claims 1 to 7, characterized in that: include: A denoising module is used to denoise the multi-source data set collected in advance in the tunnel construction area to obtain the original data set, wherein the multi-source data set includes historical geological data, real-time monitoring data, geological radar detection data, seismic wave detection data, advance drilling data and infrared detection data; A fusion module, used for preprocessing and multi-level data fusion of the original data set to obtain a fused data set; An extraction module, used for performing three-dimensional spatial modeling and feature extraction on the fused data set to obtain intelligent geological prediction results; An evaluation module, used to perform a multi-factor risk evaluation on the intelligent geological prediction results to obtain a geological risk distribution map ahead of the tunnel; An optimization module, used to optimize the construction plan based on the geological risk distribution map to obtain a construction guidance plan; The updating module is used to comprehensively analyze and dynamically update the collected real-time monitoring data, the construction guidance plan, the intelligent geological prediction results and the geological risk distribution map to obtain advanced geological prediction results.

Images