Subway engineering rotary excavating pile construction risk source identification method and system based on BIM

Abstract

The invention relates to the technical field of risk identification, and discloses a BIM-based subway engineering rotary excavating pile construction risk source identification method and system. The method comprises the following steps: acquiring construction data through field acquisition equipment, performing digital processing, and generating a construction information data set; performing space-time dimension decoupling reconstruction on the data, and establishing a multi-dimensional risk database; performing dynamic coupling analysis and interactive verification to obtain a risk identification data set; performing entropy calculation and weight distribution on each factor to form a risk assessment table; screening and matching the control data to obtain a prevention and control parameter set; and importing the parameter set into the BIM model for optimization to obtain control parameters. According to the invention, accurate identification and dynamic early warning of construction risks are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of risk identification, and particularly relates to a BIM-based metro engineering rotary pile construction risk source identification method and system. BACKGROUND

[0002] Rotary pile construction in metro engineering is a complex underground project, and its construction process involves multiple key links and risk points. Existing construction risk identification methods mainly rely on manual experience judgment and traditional monitoring means, and risk control is carried out through on-site patrol, settlement observation, and mud performance detection. With the development of BIM technology, some projects have begun to apply BIM to rotary pile construction management, realizing the visualization of the construction process and the management of basic data, and providing new technical means for risk identification.

[0003] However, the existing risk identification methods still have obvious deficiencies. First, the traditional risk identification method lacks deep fusion analysis of multi-source monitoring data, and it is difficult to fully reflect the risk evolution law in the construction process; second, the existing BIM application mainly stays at the visualization level, and fails to fully exert the potential of BIM technology in data analysis and risk early warning; third, the existing method lacks consideration of the pile-soil interaction mechanism, and it is difficult to accurately grasp the risk propagation path and influence mechanism. These problems lead to insufficient accuracy and timeliness of risk identification, affecting construction safety and quality control. SUMMARY

[0004] In view of the problems of insufficient multi-source data fusion and unclear risk propagation mechanism in the existing rotary pile construction risk identification method, the present application provides a BIM-based metro engineering rotary pile construction risk source identification method and system, which realizes accurate identification and dynamic early warning of construction risks by establishing a complete technical chain of data processing and risk analysis.

[0005] In a first aspect, the application provides a BIM-based metro engineering rotary pile construction risk source identification method, which comprises: obtaining drilling rig posture data, casing settlement data, pile forming quality data and surrounding environment data through a field acquisition device, digitally processing construction parameters and environmental parameters, and generating a rotary pile construction information dataset; according to the rotary pile construction information dataset, decoupling and reconstructing the pile forming process data and real-time monitoring data in the time-space dimension, and constructing a multi-dimensional risk database containing pile body integrity, casing stability and mud properties; according to the multi-dimensional risk database, dynamically coupling and interactively verifying the risk characteristic data such as drilling parameters, mud pressure data and pile bottom sediment, to obtain a risk identification dataset with a pile-soil interaction propagation chain; based on the risk identification dataset with the pile-soil interaction propagation chain, performing entropy dynamic iterative calculation and multi-level weight adaptive allocation on the pile forming quality factors, stratum mutation factors and surrounding environment factors, to form a risk assessment data table; for the risk assessment data table, screening and matching the casing stability control data, drilling parameter control data and mud performance control data to obtain a risk prevention and control parameter set; importing the risk prevention and control parameter set into a BIM model, performing numerical calculation and parameter optimization on the pile foundation construction process prevention and control effect, and obtaining construction control parameters.

[0006] In a second aspect, the application provides a BIM-based metro engineering rotary pile construction risk source identification system, which comprises:

[0007] An acquisition module is configured to obtain drilling rig posture data, casing settlement data, pile forming quality data and surrounding environment data through a field acquisition device, digitally process construction parameters and environmental parameters, and generate a rotary pile construction information dataset;

[0008] A reconstruction module is configured to decouple and reconstruct the pile forming process data and real-time monitoring data in the time-space dimension according to the rotary pile construction information dataset, and construct a multi-dimensional risk database containing pile body integrity, casing stability and mud properties;

[0009] A verification module is configured to dynamically couple and interactively verify the risk characteristic data such as drilling parameters, mud pressure data and pile bottom sediment according to the multi-dimensional risk database, to obtain a risk identification dataset with a pile-soil interaction propagation chain;

[0010] An iteration module is configured to perform entropy dynamic iterative calculation and multi-level weight adaptive allocation on the pile forming quality factors, stratum mutation factors and surrounding environment factors based on the risk identification dataset with the pile-soil interaction propagation chain, to form a risk assessment data table;

[0011] The screening module is configured to screen and match the protection wall stability control data, the drilling parameter control data and the mud performance control data against the risk assessment data table to obtain a risk prevention and control parameter set.

[0012] The optimization module is configured to import the risk prevention and control parameter set into the BIM model, perform numerical calculation and parameter optimization on the pile foundation construction process prevention and control effect, and obtain construction control parameters.

[0013] In the technical scheme provided in the present application, drilling rig posture data, pile casing settlement data, pile forming quality data and surrounding environment data are obtained by a field acquisition device, are digitally processed and are used to generate rotary drilling construction information data set, thereby providing a comprehensive data basis for subsequent risk analysis, and effective acquisition and integration of multi-source data at the construction site are achieved. Through spatio-temporal dimension decoupling and reconstruction processing of the pile forming process data and real-time monitoring data, a multi-dimensional risk database including pile body integrity, pile casing stability and mud properties is established, thereby solving the problem of single data dimension in the traditional method and improving the comprehensiveness of risk identification. According to the multi-dimensional risk database, dynamic coupling analysis and interactive verification are performed on risk characteristic data such as drilling parameters, mud pressure and pile bottom sediment, thereby obtaining a risk identification data set with a pile-soil interaction propagation chain, and the propagation mechanism and interaction relationship between risk factors are effectively revealed. Based on the risk identification data set with the pile-soil interaction propagation chain, an entropy value dynamic iteration calculation and multi-level weight adaptive distribution are performed, thereby forming a risk assessment data table, and the accuracy and scientificity of risk assessment are improved. According to the risk assessment data table, screening and matching are performed on the protection wall stability control data, the drilling parameter control data and the mud performance control data, thereby obtaining a risk prevention and control parameter set, and precise formulation of risk control measures is achieved. Finally, the risk prevention and control parameter set is imported into the BIM model, and construction control parameters are obtained through numerical calculation and parameter optimization, thereby achieving whole-process closed-loop management from risk identification to implementation of control measures, and significantly improving the risk management and control level and construction quality of rotary drilling pile construction. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical scheme of the embodiments of the present application, the drawings required in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0015] Figure 1 An embodiment schematic diagram of the BIM-based metro engineering rotary drilling pile construction risk source identification method in the embodiment of the present application;

[0016] Figure 2 An embodiment schematic diagram of the BIM-based metro engineering rotary drilling pile construction risk source identification system in the embodiment of the present application. Detailed Implementation

[0017] This application provides a BIM-based method and system for identifying risk sources in rotary drilling pile construction for subway engineering. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0018] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the BIM-based method for identifying risk sources in rotary pile construction for subway engineering in this application includes:

[0019] Step S101: Obtain drilling rig posture data, casing settlement data, pile quality data and surrounding environment data through on-site acquisition equipment, and digitally process the construction parameters and environmental parameters to generate a rotary drilling construction information dataset.

[0020] Step S102: Based on the rotary drilling construction information dataset, the spatiotemporal dimensions of the pile formation process data and real-time monitoring data are decoupled and reconstructed to build a multi-dimensional risk database that includes pile integrity, casing stability, and mud properties.

[0021] Step S103: Based on the multi-dimensional risk database, perform dynamic coupling analysis and interactive verification on risk characteristic data such as drilling parameters, mud pressure data, and pile bottom sediment to obtain a risk identification dataset with pile-soil action propagation chain.

[0022] Step S104: Based on the risk identification dataset with pile-soil action propagation chain, perform dynamic iterative calculation of entropy values ​​and multi-level weight adaptive allocation for pile quality factors, stratum change factors and surrounding environmental factors to form a risk assessment data table.

[0023] Step S105: For the risk assessment data table, screen and match the wall stability control data, drilling parameter control data and mud performance control data to obtain the risk prevention and control parameter set;

[0024] In step S106, the risk control parameter set is imported into the BIM model, numerical calculation and parameter optimization are performed on the pile foundation construction process control effect, and construction control parameters are obtained.

[0025] It can be understood that the execution subject of the present application can be a BIM-based subway engineering rotary drilling pile construction risk source identification system, and can also be a terminal or a server, which is not limited here. The server is taken as an example for description in the embodiments of the present application.

[0026] Specifically, the data information of the construction site is collected by a field collection device. The drilling machine posture data is obtained by an inclination sensor and a gyroscope sensor installed on the drilling machine to obtain the real-time inclination angle and rotation speed of the drilling machine in the construction process. The casing settlement data is collected by displacement sensors arranged around the casing to obtain the settlement displacement change of the casing. The pile quality data is obtained by scanning and detecting the pile body by a sound wave detector and an ultrasonic detection device. The surrounding environment data is collected by monitoring points arranged in the surrounding buildings and the ground to obtain displacement deformation information. The original data is digitally processed, including data standardization, abnormal value elimination and data smoothing processing, to form a standard format rotary drilling construction information data set. After obtaining the rotary drilling construction information data set, the pile forming process data and real-time monitoring data are decoupled and reconstructed in time and space dimensions. Time-space dimension decoupling refers to segmenting the data according to the time stamp, separating the spatial dimension and the time dimension, and facilitating the analysis of time sequence characteristics and spatial distribution characteristics. The time sequence characteristics are extracted by time sequence decomposition of the separated data, and the spatial dimension characteristics are extracted by spatial correlation analysis. The extracted characteristics are reconstructed to establish a multi-dimensional risk database containing pile body integrity, casing stability and mud properties.

[0027] The data in the multi-dimensional risk database is subjected to dynamic coupling analysis and interactive verification. Numerical range analysis and abnormal value screening are performed on the drilling parameters to generate a sequence reflecting the characteristics of the drilling process. Mud pressure data is subjected to pressure gradient calculation and volatility analysis to form a pressure change law. The pile bottom sediment data is subjected to settlement rate analysis and distribution characteristic extraction. Through the correlation calculation of these data, a risk coupling matrix is formed, and the node connectivity of the risk propagation path is analyzed and the propagation intensity is quantified to construct a pile-soil interaction propagation chain and form a risk identification data set. Based on the risk identification data set, the entropy value dynamic iteration calculation and multi-level weight distribution are performed on each risk factor. First, the pile quality factors are subjected to numerical discretization and data standardization processing to form a quality evaluation index sequence. The stratum mutation factors are subjected to stratum interface identification and mutation feature extraction to generate a stratum layer data set. The surrounding environmental factors are subjected to spatial correlation analysis and influence domain division to obtain an environmental influence parameter table. An entropy value dynamic matrix is generated by information entropy calculation and data iteration processing, and the matrix is used for weight distribution to form a risk assessment data table.

[0028] The control data in the risk assessment data table is screened and matched. Stability threshold judgment and parameter interval division are performed on the pile casing stability control data; drilling parameter control data are analyzed for drilling rate and rotational speed range; and rheological property analysis and density comparison are performed on the mud performance control data. Through cross-validation and correlation calculation, a parameter matching matrix is generated, mutual feedback effect analysis and constraint condition screening are performed, a control parameter combination table is formed, and a risk prevention and control parameter set is obtained. After importing the risk prevention and control parameter set into the BIM model, the construction process data is time-decomposed and process-split, and a construction process control node sequence is generated. The sequence is used to reproduce the scene and simulate the process of the pile foundation construction condition, forming a construction state data stream. Through quantitative analysis and numerical operation of the response characteristics and effect of the prevention and control measures, a prevention and control response matrix is obtained. According to the matrix, the sensitivity of the key control parameters is analyzed and the influence is quantified, a parameter sensitivity table is formed, and constraint boundary adjustment and objective function construction are performed, through multi-objective balance calculation and feedback correction processing, the construction control parameters are obtained.

[0029] For example, the drilling rig attitude data obtained by the field acquisition device shows that the drilling rig inclination angle changes periodically during the construction process. Through smoothing processing of the original data, abnormal values caused by equipment jitter are removed, and the change trend of the drilling rig inclination angle is obtained. After standardization processing of the pile casing settlement data, it is found that the settlement around the pile casing is uneven, with a maximum difference of 15 mm. The pile quality data obtained by ultrasonic detection reflects that there are micro-cracks in the pile body after signal processing. After decoupling of the time and space dimensions, it is found that there is a significant correlation between the drilling rig attitude change and the pile casing settlement, and the key risk propagation nodes are determined through time sequence feature extraction. Analysis of the mud pressure data shows that the mud pressure fluctuates when passing through the silt layer, which forms a coupling effect with the stratum mutation factor. Based on these data characteristics, the weights of each risk factor are determined through entropy calculation, and a risk prevention and control parameter set including pile casing stability control, drilling parameter control and mud performance control is generated.

[0030] In the embodiments of the present application, drilling rig posture data, casing settlement data, pile quality data and surrounding environment data are obtained by field acquisition equipment, digitized and generated into rotary drilling construction information data set, providing a comprehensive data basis for subsequent risk analysis, and realizing effective collection and integration of multi-source data in the construction site. Through spatio-temporal decoupling and reconstruction processing of the pile construction process data and real-time monitoring data, a multi-dimensional risk database containing pile integrity, casing stability and mud properties is established, solving the problem of single data dimension in traditional methods and improving the comprehensiveness of risk identification. According to the multi-dimensional risk database, dynamic coupling analysis and interactive verification are performed on the risk characteristic data such as drilling parameters, mud pressure and pile bottom sediment, and a risk identification data set with pile-soil interaction propagation chain is obtained, effectively revealing the propagation mechanism and interaction relationship between risk factors. Based on the risk identification data set with pile-soil interaction propagation chain, through entropy dynamic iteration calculation and multi-level weight adaptive allocation, a risk assessment data table is formed, improving the accuracy and scientificity of risk assessment. According to the risk assessment data table, risk control parameter set is obtained by screening and matching casing stability control data, drilling parameter control data and mud performance control data, realizing accurate formulation of risk control measures. Finally, the risk control parameter set is imported into the BIM model, and the construction control parameters are obtained through numerical calculation and parameter optimization, realizing the whole process closed-loop management from risk identification to control measure implementation, and significantly improving the risk control level and construction quality of rotary drilling pile construction.

[0031] In a specific embodiment, the process of step S101 can specifically include the following steps:

[0032] (1) Collect the drilling rig inclination angle, rotation speed and drilling depth in the drilling process through the inertial sensor, and obtain the drilling rig posture data according to the time sequence characteristic processing;

[0033] (2) Continuously sample the displacement changes around the casing by using the displacement sensor, and generate the casing settlement data through data smoothing processing;

[0034] (3) Scan the pile integrity according to the acoustic detector, and form the pile quality data in combination with the frequency spectrum analysis of the ultrasonic detection signal;

[0035] (4) Collect the displacement deformation information of surrounding buildings and ground surface from the ground settlement monitoring points, and obtain the surrounding environment data through spatial interpolation processing;

[0036] (5) Based on the drilling rig posture data, casing settlement data, pile quality data and surrounding environment data, establish a parameter comparison table through data standardization processing, and obtain the construction parameters;

[0037] (6) According to the construction parameters, combined with the geological survey information, the data correlation analysis is carried out to generate the rotary drilling construction information data set.

[0038] Specifically, drilling process data is collected by an inertial sensor. The inertial sensor includes a gyroscope and an accelerometer, wherein the gyroscope measures the angular velocity of the drilling machine, and the accelerometer measures the linear acceleration. By fusing the data of the two sensors, the attitude information of the drilling machine is obtained. The time sequence feature processing adopts a sliding window method to process the original data in segments. The data in each time window is processed by denoising and smoothing to extract the change trend of the drilling machine inclination angle, rotation speed and drilling depth, forming a continuous drilling machine attitude data stream. The pile casing settlement monitoring adopts a high-precision displacement sensor, which is uniformly arranged around the pile casing. The displacement sensor continuously collects data at a fixed sampling frequency. The collected original data is processed by data smoothing to eliminate random noise. The data smoothing processing adopts a weighted moving average method to perform weighted averaging on adjacent data points in the time sequence, and the weight decreases with the increase of the distance from the center point, thereby retaining the true change trend of the data and filtering out short-term fluctuations.

[0039] The pile body integrity detection adopts acoustic wave detection technology. The acoustic wave detector transmits acoustic wave signals to judge the integrity of the pile body by receiving reflected waveforms. The acoustic wave signals are analyzed by frequency spectrum analysis, and the time domain signals are converted into frequency domain signals by fast Fourier transform, and the energy distribution characteristics of different frequency bands are analyzed. Ultrasonic detection transmits ultrasonic waves to measure the propagation time and attenuation characteristics of the sound waves in the concrete, and combines the sound wave test results to comprehensively evaluate the pile body quality, forming pile quality data containing crack, necking, mud inclusion and other defect information. The surrounding environment monitoring is realized by a network of monitoring points arranged on the building and the ground. The monitoring points collect vertical and horizontal displacement data, and the discrete monitoring point data is processed by a spatial interpolation method. The spatial interpolation adopts an inverse distance weighting method to determine the weight according to the distance between the to-be-solved point and the known monitoring points. The closer the monitoring points, the greater the influence on the interpolation result, so as to obtain the continuous deformation field distribution of the entire monitoring area.

[0040] Data standardization is to eliminate the differences in dimension and numerical range between different types of data. The drilling rig attitude data, casing settlement data, pile quality data and surrounding environment data are standardized respectively, and all kinds of data are converted into a unified numerical interval. Through the establishment of parameter comparison table, the corresponding relationship and conversion rule between various data are clarified, and a unified construction parameter system is formed. Data correlation analysis associates the standardized construction parameters with the geological survey information. The geological survey information includes soil layer distribution, groundwater level, soil physical and mechanical properties and other contents. Through correlation analysis, the mapping relationship between construction parameters and geological conditions is established, the key geological factors affecting construction risk are found out, and the rotary drilling construction information data set containing construction process information and geological environment information is generated.

[0041] Taking the rotary drilling pile construction of a subway station with a depth of 35 meters as an example, the original data collected by the inertial sensor shows that the attitude of the drilling rig fluctuates when it passes through the soft soil layer. Through sliding window analysis, a 5-second time window is set to smooth the original data and obtain the continuous change curve of the inclination angle of the drilling rig. At the same time, the displacement sensors arranged around the casing collect data at a frequency of 2 times per second, and after weighted moving average processing, it is found that the settlement rate of the east side of the casing is significantly faster than that of other directions. The acoustic wave detection is scanned every 3 meters along the pile body, and after frequency spectrum analysis of the collected acoustic wave signals, abnormal reflection waveforms are detected at a depth of 15-20 meters, combined with the ultrasonic detection results, it is judged that there is a lack of concrete density at this position. The surrounding environment monitoring sets up 20 monitoring points, and uses inverse distance weighted interpolation to calculate the surface subsidence distribution of the entire construction influence area.

[0042] In a specific embodiment, the process of performing step S102 can specifically include the following steps:

[0043] (1) According to the time series data in the rotary drilling construction information data set, the pile forming process data is segmented according to the time stamp, the spatial dimension and the time dimension are separated to obtain dimension separation data;

[0044] (2) Using the dimension separation data, the pile forming process data is decomposed in time sequence by signal processing method to generate time sequence feature matrix;

[0045] (3) Based on the time sequence feature matrix, the real-time monitoring data is deconstructed and data verified in layers to form a monitoring data verification table;

[0046] (4) According to the monitoring data verification table, the monitoring features of pile body integrity and casing stability are reorganized to obtain pile foundation risk feature data;

[0047] (5) For the pile foundation risk feature data, the mud property parameters are extracted by numerical fitting method to obtain mud property data;

[0048] (6) The pile foundation risk feature data and the mud property data are reconstructed and associated in time and space dimensions to construct a multi-dimensional risk database containing pile body integrity, casing stability and mud properties.

[0049] Specifically, the time series data in the rotary drilling construction information dataset are processed. The time series data are continuous data streams recorded according to time stamps, containing time variation information of construction parameters such as drilling speed, drilling depth and casing settlement. The time stamp is a time marker recording the time of data collection. The continuous data stream is segmented according to fixed time intervals through the time stamp. The segmented data contain time dimension information (time point of data collection) and space dimension information (position, depth and other spatial attributes of data collection), which are separated by data dimension reduction technology to obtain dimension-separated data. The dimension-separated data generate a time series feature matrix after signal processing. The signal processing methods include wavelet transform and empirical mode decomposition. The wavelet transform is used to extract the local features of the signal, and the time series is decomposed at different scales to obtain multi-scale feature information. The empirical mode decomposition decomposes the complex time series into a finite number of intrinsic mode functions, each of which represents an oscillation component of a specific frequency. Through the combination of the two methods, the pile forming process data are decomposed into multiple characteristic components to form a time series feature matrix containing time-frequency features.

[0050] The real-time monitoring data are hierarchically deconstructed and data-verified based on the time series feature matrix. The hierarchical deconstruction is to classify and organize the monitoring data according to the monitoring objects (pile body, casing, mud, etc.) and monitoring parameters (displacement, pressure, density, etc.). The data verification includes numerical range verification, variation trend verification and correlation verification. By setting reasonable threshold ranges, abnormal data are eliminated to ensure the effectiveness and reliability of the data. The verified data are recorded in the monitoring data verification table, which contains the effectiveness mark and verification results of the data. According to the monitoring data verification table, the monitoring features of the pile body integrity and the casing stability are reorganized. The pile body integrity contains acoustic detection data, ultrasonic detection data and other indicators reflecting the pile body quality, and the casing stability contains displacement, inclination, stress and other indicators reflecting the casing state. During the data reorganization process, the monitoring data from different sources are integrated according to the spatial position and time sequence to establish the corresponding relationship between the monitoring parameters and generate the pile foundation risk feature data.

[0051] For the pile foundation risk feature data, the numerical fitting method is used to extract the characteristics of the mud property parameters. The mud property parameters include density, viscosity, sand content and other indicators reflecting the physical properties of mud. Numerical fitting uses polynomial regression and neural network methods to establish the mapping relationship between mud parameters and construction conditions, extract the change rule of mud properties, and form mud property data. Finally, the pile foundation risk feature data and the mud property data are reconstructed and associated in time and space dimensions. Time and space dimension reconstruction is to recombine the time and space dimension information separated in the early stage to restore the time and space correlation of the data. Correlation analysis and cluster analysis are used to identify the correlation between different monitoring parameters, and a multi-dimensional risk database including pile integrity, casing stability and mud properties is constructed.

[0052] For example, the rotary drilling construction information dataset records 8 hours of continuous construction process data. The data is segmented at 5-minute intervals by timestamp, separating the spatial data in the depth direction and the time series change data. After wavelet transform and empirical mode decomposition of the separated data, the time series features of key construction parameters such as drilling speed change and casing settlement are extracted. After hierarchical decomposition and verification of the real-time monitoring data, it is found that the pile body acoustic detection data at 25 meters in depth is abnormal, and the casing inclination monitoring data shows that the inclination angle increases. Through data recombination, it is found that there is a temporal correlation between the two. Further analysis of the mud property parameters shows that the mud density fluctuates before the anomaly occurs, and the corresponding relationship between the mud density change and the stratum condition is determined through numerical fitting. After time and space dimension reconstruction, the spatial distribution and time series evolution rule of pile quality anomaly, casing inclination and mud property change are determined.

[0053] In a specific embodiment, the process of performing step S103 can specifically include the following steps:

[0054] (1) Based on the multi-dimensional risk database, numerical range analysis and outlier screening are performed on the drilling parameters to generate a drilling feature sequence;

[0055] (2) According to the drilling feature sequence, pressure gradient calculation and fluctuation analysis are performed on the mud pressure data to form a pressure change data table;

[0056] (3) Using the pressure change data table, settlement rate analysis and distribution feature extraction are performed on the pile bottom sediment data to obtain sediment distribution data;

[0057] (4) Through the drilling feature sequence, the pressure change data table and the sediment distribution data, correlation calculation is performed to obtain a risk coupling matrix;

[0058] (5) According to the risk coupling matrix, node connectivity analysis and propagation intensity quantification are performed on the risk propagation path to form a pile-soil interaction propagation chain;

[0059] (6) The pile-soil interaction propagation chain is dynamically associated with the risk characteristic data to obtain a risk identification data set with the pile-soil interaction propagation chain.

[0060] Specifically, numerical range analysis and abnormal value screening are performed on the drilling parameters. The drilling parameters include key parameters such as drilling speed, drilling pressure and rotation speed. The original data is screened by setting engineering experience values and statistical thresholds to eliminate abnormal values beyond the reasonable range. The numerical range analysis adopts the 3σ criterion for preliminary screening, and the data points beyond the range of ±3 times the standard deviation of the mean value are marked. Further abnormal value identification is performed through box plot analysis, and the data points beyond 1.5 times the interquartile range of the upper and lower quartiles are determined as abnormal points. The data screened by the abnormal value is rearranged according to the time sequence to form a drilling feature sequence reflecting the characteristics of the drilling process. Based on the drilling feature sequence, the mud pressure data is analyzed and processed. The mud pressure data is obtained by collecting the pressure at different depths through a pressure sensor. The pressure gradient calculation is to analyze the pressure change rate between adjacent measuring points. By calculating the pressure change amount per unit depth, the vertical distribution characteristics of the mud pressure are obtained. The fluctuation analysis focuses on the time sequence change characteristics of the pressure value. Variance analysis and frequency spectrum analysis methods are used to identify the periodicity and mutation characteristics of the pressure fluctuation. These characteristics are recorded in the pressure change data table.

[0061] Using the information in the pressure change data table, further analysis is performed on the pile bottom sediment data. The pile bottom sediment data reflects the accumulation condition of the drilling sludge at the pile bottom, and the thickness and distribution of the sediment are recorded by the monitoring equipment. The settlement rate analysis calculates the change amount of the sediment thickness per unit time, and the spatial distribution data are combined to analyze the distribution characteristics of the sediment at the pile bottom, including whether the sediment is uniformly deposited, whether there is local accumulation, etc. These information is integrated to form the sediment distribution data.

[0062] The calculation of the risk coupling matrix uses the following formula:

[0063]

[0064] Wherein: R ij represents the coupling risk intensity; W k represents the weight coefficient of each factor; X i represents the normalized value of the drilling feature sequence; Y j represents the normalized value of the pressure change data; Z k represents the normalized value of the sediment distribution data. represents the coupling operator.

[0065] Based on the risk coupling matrix, the risk propagation path is analyzed, and the propagation intensity is quantified using the following formula:

[0066]

[0067] wherein: P mn represents the propagation strength from node m to node n; a represents the propagation attenuation coefficient; b represents the space influence coefficient; D mn represents the distance between nodes; F q represents the influence factor; s represents the number of influence factors.

[0068] For example, during the construction process, the abnormal value screening of drilling parameters shows that the drilling speed in the soft soil layer section appears abnormal fluctuations. Through pressure gradient analysis, it is found that the mud pressure gradient deviates from the normal value at this depth section, and the fluctuation analysis shows that the pressure fluctuation frequency increases. At the same time, the pile bottom sediment monitoring data shows that the sediment thickness is unevenly distributed, and the local accumulation is obvious. Correlation calculation is performed on these three groups of data, and the risk coupling matrix generated reflects that there is a significant correlation between drilling speed anomaly, mud pressure fluctuation and sediment accumulation. Through node connectivity analysis and propagation strength quantification, the risk propagation path is determined: drilling speed anomaly leads to mud pressure fluctuation, which in turn causes uneven distribution of pile bottom sediment, affecting the pile foundation construction quality. This propagation chain is integrated into the risk identification data set to guide the dynamic adjustment of construction parameters.

[0069] In a specific embodiment, the process of performing step S104 can specifically include the following steps:

[0070] (1) According to the risk identification data set with the pile-soil interaction propagation chain, the pile quality factors are subjected to numerical discretization and data standardization processing to obtain a quality evaluation index sequence;

[0071] (2) According to the quality evaluation index sequence, the stratum mutation factors are subjected to stratum interface identification and mutation feature extraction to form a stratum layering data set;

[0072] (3) Based on the stratum layering data set, the surrounding environmental factors are subjected to spatial correlation analysis and influence domain division to obtain an environmental influence parameter table;

[0073] (4) The quality evaluation index sequence, the stratum layering data set and the environmental influence parameter table are subjected to information entropy calculation, and the entropy value dynamic matrix is generated through data iteration processing;

[0074] (5) The entropy value dynamic matrix is used for weight initialization, the different hierarchical risk factors are subjected to hierarchical decomposition and weight distribution, and a multi-level weight distribution table is constructed;

[0075] (6) According to the multi-level weight distribution table, the risk assessment indicators are subjected to comprehensive calculation and hierarchical processing to form a risk assessment data table.

[0076] Specifically, in the process of identifying risk sources of rotary drilling pile construction in BIM-based subway engineering, the risk identification dataset with the pile-soil interaction propagation chain is first processed. According to the dataset, the pile quality factors are numerically discretized and data standardized. Numerical discretization is to convert continuous quality evaluation indicators into discrete numerical intervals, including pile integrity index, concrete strength, pile diameter deviation, etc. Data standardization uses the maximum and minimum value standardization method to convert indicators of different dimensions into a unified numerical interval, forming a comparable quality evaluation index sequence.

[0077] According to the quality evaluation index sequence, the stratum mutation factors are analyzed. The stratum interface identification determines the boundary position between different strata by analyzing the change characteristics of the physical and mechanical properties of the soil layer. The mutation feature extraction focuses on the areas with sharp changes in stratum properties, such as the interface between soft and hard soil layers, and areas with sudden changes in water content. These feature information is organized and recorded in the stratum layering dataset, providing basic data support for subsequent analysis. Based on the stratum layering dataset, the spatial correlation analysis of the surrounding environmental factors is carried out. The spatial correlation analysis includes the correlation analysis between environmental impact factors such as building settlement and ground deformation and the construction location. The influence domain division divides the construction influence range into strong, medium and weak influence areas based on the spatial decay law, forming an environmental impact parameter table.

[0078] The information entropy calculation uses the following formula:

[0079]

[0080] where H represents the comprehensive information entropy; γ ijk represents the normalized evaluation index value; ψ ijk represents the index weight adjustment factor; n represents the number of quality indicators; m represents the number of stratum layers; and l represents the number of environmental impact factors.

[0081] By calculating the information entropy of the quality evaluation index sequence, the stratum layering dataset, and the environmental impact parameter table, and combining data iteration processing, an entropy value dynamic matrix reflecting the uncertainty degree of each factor is generated. Using this matrix, weight initialization is performed, and the analytic hierarchy process is used to hierarchically decompose different level risk factors. In the hierarchical decomposition process, the risk factors are subdivided layer by layer according to the hierarchical relationship, from the top-level comprehensive risk index to the bottom-level specific evaluation index, and the weight distribution of each level index is allocated to construct a multi-level weight distribution table. Finally, according to the multi-level weight distribution table, the risk assessment indicators are comprehensively calculated and classified. The comprehensive calculation considers the weight and mutual correlation of each indicator, and the classification processing divides the risk level into different grades according to the calculation results, forming a risk assessment data table.

[0082] For example, the pile integrity data and concrete strength detection data obtained by acoustic detection are discretized to obtain segmented quality evaluation indexes. Geological survey data show that there are multiple layers of soft and hard soil interbedded structures at the pile location, and the key horizon change points are determined through stratum interface identification. The surrounding environment monitoring data show that there is a building within the construction influence range, and the correlation between the building settlement and the construction progress is determined through spatial correlation analysis. After information entropy calculation, it is reflected that the uncertainty of the pile quality at the soft and hard soil layer interface is high, and the correlation between the position and the building settlement is also strong. Through hierarchical decomposition, the risk factors are divided into three levels of site conditions, construction conditions and environmental impact, and the weight distribution of each level index is based on the entropy value result. The risk assessment result shows that the construction risk level of the pile at the soft and hard soil layer interface is high, and the construction parameters and the monitoring of the surrounding environment response need to be controlled.

[0083] In a specific embodiment, the process of performing step S105 can specifically include the following steps:

[0084] (1) According to the risk assessment data table, the stability threshold value judgment and parameter interval division of the retaining wall stability control data are performed to obtain the retaining wall control index set;

[0085] (2) Using the retaining wall control index set, the drilling rate analysis and rotation speed range determination of the drilling parameter control data are performed to form a drilling control data table;

[0086] (3) Through the drilling control data table, the rheological property analysis and density comparison of the mud performance control data are performed to obtain a mud parameter sequence;

[0087] (4) According to the retaining wall control index set, the drilling control data table and the mud parameter sequence, cross-validation and correlation calculation are performed to generate a parameter matching matrix;

[0088] (5) Based on the parameter matching matrix, mutual feedback effect analysis and constraint condition screening are performed on the control parameters to form a control parameter combination table;

[0089] (6) The parameters in the control parameter combination table are subjected to comprehensive balance processing and optimization calculation to obtain a risk prevention and control parameter set.

[0090] Specifically, in the process of risk control of rotary drilling pile construction in subway engineering, firstly, the stability control data of the retaining wall are analyzed and processed according to the risk assessment data table. The stability threshold determines the critical state of retaining wall instability based on engineering experience and geological conditions, including key indicators such as the inclination angle of the retaining wall, the displacement of the surrounding soil, and ground settlement. Parameter interval division divides the effective value range of each control indicator into multiple sub-intervals, such as dividing the inclination angle of the retaining wall into a safe interval, a warning interval, and a dangerous interval, forming a retaining wall control indicator set. The retaining wall control indicator set is used to guide the control of drilling parameters. Drilling rate analysis considers the optimal drilling speed under different soil conditions, combined with the stability requirements of the retaining wall, to determine the safe drilling speed range. The speed range is defined based on soil properties and drilling tool performance, setting reasonable upper and lower limits to avoid excessive disturbance that leads to instability of the retaining wall. These parameters are recorded in the drilling control data table as the basis for construction process control.

[0091] The drilling control data table is used to guide the control of mud performance. Rheological property analysis focuses on rheological parameters such as mud viscosity and yield value to ensure that the mud has good carrying performance and retaining wall performance. Density ratio comparison determines the optimal mud density range by calculating the balance between mud density and formation pore water pressure. These performance indicators form a mud parameter sequence to guide mud configuration and adjustment. Cross-validation and correlation degree calculation are performed based on the above three groups of data. Cross-validation is used to verify the rationality of different parameter combinations and analyze the coordination between parameters. Correlation degree calculation uses the following formula:

[0092] C ijkl = λ1·σ ij + λ2·τ jk + λ3·η kl + λ4·ω li

[0093] Where: C ijkl represents the correlation degree of parameters; σ ij represents the correlation coefficient of retaining wall indicators and drilling parameters; τ jk represents the correlation coefficient of drilling parameters and mud performance; η kl represents the correlation coefficient of mud performance and retaining wall stability; ω li represents the interaction coefficient between parameters; λ1, λ2, λ3, λ4 represent weight coefficients.

[0094] The calculation results form a parameter matching matrix, reflecting the coupling relationship between the parameters. Based on the matrix, mutual feedback effect analysis is carried out to study the mutual influence mechanism between different parameters, and parameter combinations meeting the constraint conditions are screened to form a control parameter combination table. Finally, the parameters in the control parameter combination table are balanced and optimized, considering construction efficiency, safety and economy, etc., to select the optimal parameter combination and form the risk prevention and control parameter set.

[0095] For example, the retaining wall stability control data shows that the retaining wall is at high risk of instability in the silt layer section, and the threshold value of the inclination angle of the casing in this section is determined to be 2° through threshold determination. Based on the retaining wall control requirements, the drilling control data table stipulates that the upper limit of the drilling rate in this soil layer is 0.5 m / min, and the rotation speed range is 8-12 r / min. In terms of mud performance control, through rheological property analysis, it is determined that the mud viscosity needs to be maintained at 28-32 s, the density is controlled at 1.15-1.20 g / cm 3 . These parameters are cross-validated, which shows that when the drilling rate approaches the upper limit, the mud viscosity needs to be increased accordingly to maintain the stability of the retaining wall. Through mutual feedback effect analysis and constraint condition screening, the optimal parameter combination for this soil layer section is determined: drilling rate 0.4 m / min, rotation speed 10 r / min, mud viscosity 30 s, and density 1.18 g / cm 3 , forming the prevention and control parameter set.

[0096] In a specific embodiment, the process of performing step S106 can specifically include the following steps:

[0097] (1) According to the risk prevention and control parameter set, the construction process data is time-decomposed and process-split to generate a construction process control node sequence;

[0098] (2) Using the construction process control node sequence, the pile foundation construction condition is scene-recreated and process-simulated to form a construction state data stream;

[0099] (3) Through the construction state data stream, the response characteristics and effect of the prevention and control measures are quantitatively analyzed and numerically calculated to obtain a prevention and control response matrix;

[0100] (4) According to the prevention and control response matrix, the sensitivity analysis and influence quantification of the key control parameters are carried out to form a parameter sensitivity table;

[0101] (5) Based on the parameter sensitivity table, the constraint boundary adjustment and objective function construction of the control parameters are carried out to obtain a parameter optimization sequence;

[0102] (6) The parameter optimization sequence is processed through multi-objective balance calculation and feedback correction to obtain the construction control parameters.

[0103] Specifically, the construction process data is processed according to the risk prevention and control parameter set. Time sequence decomposition is to divide the continuous construction process into multiple construction stages according to the time sequence, including pile casing embedding, drilling into hole, hole cleaning, reinforcement cage installation and concrete pouring and other main processes. Process decomposition is to subdivide each process into specific operation steps, such as drill bit positioning, starting drilling, adjusting parameters, adding mud and other operation steps in the drilling into hole stage. These decomposed processes and steps form a construction process control node sequence. The control node sequence is used for scene reproduction and process simulation. Scene reproduction is to reproduce the construction conditions in the BIM environment, including stratum conditions, drilling state, mud circulation and other construction scenes. Process simulation is to dynamically simulate the construction process, analyze the connection relationship between each process and the change law of the construction parameters, and generate a construction state data stream reflecting the whole construction process.

[0104] The construction state data stream is continuously recorded working condition data. Through analysis of these data, the response characteristics and effect of the prevention and control measures are evaluated. The response characteristics include the response time, response amplitude and stability after parameter adjustment, and the effect reflects the control degree of the prevention and control measures on the construction risk. These analysis results form a prevention and control response matrix.

[0105] According to the response matrix, sensitivity analysis is performed on the key control parameters to quantify the influence degree of each parameter on the construction risk. The sensitivity analysis adopts the single factor change method, observes the change of the system response by changing a single parameter, calculates the sensitivity coefficient, and forms a parameter sensitivity table. Based on the sensitivity analysis results, the objective function is constructed, and the following multi-objective optimization model is adopted:

[0106]

[0107] Wherein: Q(x) represents the comprehensive objective function value; Θ v (x) represents the current value of the vth safety index (including the retaining wall stability index, the pile body integrity index, etc.); represents the minimum allowable value of the vth safety index; represents the maximum allowable value of the vth safety index; Π u (x) represents the current value of the uth efficiency index (including drilling rate, mud circulation efficiency, etc.); represents the minimum allowable value of the uth efficiency index; represents the maximum allowable value of the uth efficiency index; μ v represents the weight coefficient of the vth safety index, reflecting the importance of the index; ν uweight coefficient of the u-th efficiency index, reflecting the importance of the index; p represents the total number of safety indexes; q represents the total number of efficiency indexes; x represents a control parameter vector, including drilling rate, rotating speed, and mud performance parameters.

[0108] By constraint boundary adjustment and objective function optimization, a parameter optimization sequence is formed. The multi-objective balance calculation considers the balance between construction safety and efficiency, and continuously optimizes the parameter combination through feedback correction to obtain the construction control parameters.

[0109] For example, key control nodes are identified after the construction process is split, including parameter adjustment points in the soft soil layer drilling section, mud replacement points, and hole cleaning points. Through BIM scene reproduction, simulation analysis shows that there is a significant correlation between drilling rate and wall protection stability in the soft soil layer section. Construction state data shows that when the drilling rate increases from 0.4 m / min to 0.5 m / min, the response time of the casing inclination angle is about 5 minutes, and the displacement increases significantly. Sensitivity analysis shows that the drilling rate is the most sensitive parameter affecting the wall protection stability, followed by the mud density. Based on these analysis results, a multi-objective optimization function including construction efficiency and safety is set, and the control parameters are determined through iterative optimization: the drilling rate is controlled in the range of 0.35-0.45 m / min in the soft soil layer section, and the mud density is appropriately increased to 1.20 g / cm 3 , to ensure the stability of the wall protection.

[0110] The above describes the BIM-based metro engineering rotary drilling pile construction risk source identification method in the embodiments of the present application, and the BIM-based metro engineering rotary drilling pile construction risk source identification system in the embodiments of the present application is described below. Please refer to Figure 2 , an embodiment of the BIM-based metro engineering rotary drilling pile construction risk source identification system in the embodiments of the present application includes:

[0111] The acquisition module is used to obtain drilling posture data, casing settlement data, pile quality data, and surrounding environment data through field acquisition equipment, and to digitally process construction parameters and environmental parameters to generate rotary drilling construction information data sets;

[0112] The reconstruction module is used to perform space-time dimension decoupling and reconstruction processing on the pile construction process data and real-time monitoring data according to the rotary drilling construction information data sets, and to construct a multi-dimensional risk database including pile integrity, casing stability, and mud properties;

[0113] The verification module is used to perform dynamic coupling analysis and interactive verification on risk characteristic data such as drilling parameters, mud pressure data, and pile bottom sediment, according to the multi-dimensional risk database, to obtain a risk identification data set with a pile-soil interaction propagation chain;

[0114] an iteration module configured to perform entropy dynamic iteration calculation and multi-level weight adaptive allocation on the pile-forming quality factor, the stratum mutation factor and the surrounding environment factor based on the risk identification data set with the pile-soil interaction propagation chain to form a risk assessment data table;

[0115] a screening module configured to screen and match the pile protection stability control data, the drilling parameter control data and the mud performance control data against the risk assessment data table to obtain a risk prevention and control parameter set;

[0116] an optimization module configured to import the risk prevention and control parameter set into the BIM model, perform numerical calculation and parameter optimization on the pile foundation construction process prevention and control effect, and obtain the construction control parameter.

[0117] Through the cooperation of the above components, the drilling machine posture data, the casing settlement data, the pile-forming quality data and the surrounding environment data are obtained by the on-site acquisition equipment, are digitally processed and are generated into the rotary drilling construction information data set, providing a comprehensive data basis for subsequent risk analysis, and realizing effective collection and integration of multi-source data on the construction site. Through spatio-temporal decoupling and reconstruction processing of the pile-forming process data and the real-time monitoring data, a multi-dimensional risk database containing the pile integrity, the casing stability and the mud properties is established, solving the problem of single data dimension in the traditional method and improving the comprehensiveness of risk identification. According to the multi-dimensional risk database, dynamic coupling analysis and interactive verification are performed on the risk characteristic data such as the drilling parameter, the mud pressure and the pile bottom sediment, and the risk identification data set with the pile-soil interaction propagation chain is obtained, effectively revealing the propagation mechanism and the interaction relationship between the risk factors. Based on the risk identification data set with the pile-soil interaction propagation chain, the risk assessment data table is formed through entropy dynamic iteration calculation and multi-level weight adaptive allocation, improving the accuracy and the scientificity of risk assessment. The risk prevention and control parameter set is obtained by screening and matching the pile protection stability control data, the drilling parameter control data and the mud performance control data against the risk assessment data table, realizing accurate formulation of risk control measures. Finally, the risk prevention and control parameter set is imported into the BIM model, and the construction control parameter is obtained through numerical calculation and parameter optimization, realizing whole-process closed-loop management from risk identification to control measure implementation, and significantly improving the risk management and control level and the construction quality of the rotary drilling pile construction.

[0118] The above and the above embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying a risk source of a rotary drilling pile construction of a BIM-based subway project, characterized in that, The BIM-based subway engineering rotary drilling pile construction risk source identification method comprises: Obtain drilling rig posture data, casing settlement data, pile forming quality data and surrounding environment data through field acquisition equipment, and digitally process construction parameters and environmental parameters to generate rotary drilling construction information data set; According to the rotary drilling construction information data set, decouple and reconstruct the time and space dimensions of the pile forming process data and real-time monitoring data, and construct a multi-dimensional risk database containing pile body integrity, casing stability and mud properties; According to the multi-dimensional risk database, dynamically couple and interactively verify the risk characteristic data of drilling parameters, mud pressure data and pile bottom sediment, and obtain a risk identification data set with pile-soil interaction propagation chain; Based on the risk identification data set with pile-soil interaction propagation chain, perform entropy dynamic iteration calculation and multi-level weight adaptive allocation on the pile forming quality factors, stratum mutation factors and surrounding environment factors to form a risk assessment data table; According to the risk assessment data table, screen and match the casing stability control data, drilling parameter control data and mud performance control data to obtain a risk prevention and control parameter set; Import the risk prevention and control parameter set into the BIM model, perform numerical calculation and parameter optimization on the pile foundation construction process prevention and control effect, and obtain the construction control parameters.

2. The BIM-based subway engineering rotary digging pile construction risk source identification method according to claim 1, characterized in that, The method for obtaining drilling rig posture data, casing settlement data, pile forming quality data and surrounding environment data through field acquisition equipment, and digitally processing construction parameters and environmental parameters to generate rotary drilling construction information data set comprises: Collect the drilling rig inclination angle, rotation speed and drilling depth during drilling through an inertial sensor, and obtain the drilling rig posture data according to time sequence characteristic processing; Use a displacement sensor to continuously sample the displacement changes around the casing, and generate casing settlement data through data smoothing processing; Scan the pile body integrity according to the acoustic detector, and form the pile forming quality data by combining the frequency spectrum analysis of the ultrasonic detection signal; Collect the displacement deformation information of surrounding buildings and the ground surface from the ground settlement monitoring points, and obtain the surrounding environment data through spatial interpolation processing; Based on the drilling rig posture data, casing settlement data, pile forming quality data and surrounding environment data, establish a parameter comparison table through data standardization processing to obtain construction parameters; According to the construction parameters, perform data correlation analysis in combination with geological survey information to generate the rotary drilling construction information data set.

3. The BIM-based subway engineering rotary digging pile construction risk source identification method according to claim 1, characterized in that, The method for decoupling and reconstructing the time and space dimensions of the pile forming process data and real-time monitoring data according to the rotary drilling construction information data set, and constructing a multi-dimensional risk database containing pile body integrity, casing stability and mud properties comprises: According to the time sequence data in the rotary drilling construction information data set, segment the pile forming process data according to time stamp, separate the spatial dimension and the time dimension to obtain dimension separation data; Use the dimension separation data to perform time sequence decomposition on the pile forming process data through a signal processing method to generate a time sequence characteristic matrix; Based on the time sequence characteristic matrix, perform hierarchical deconstruction and data verification on the real-time monitoring data to form a monitoring data verification table; The pile body integrity, the monitoring characteristic of the casing stability is reorganized according to the monitoring data check table, and the pile foundation risk characteristic data is obtained; According to the pile foundation risk characteristic data, the mud property parameter is extracted by a numerical fitting method, and mud characteristic data is obtained; The pile foundation risk characteristic data and the mud characteristic data are reconstructed in time and space dimensions and are associated, and a multi-dimensional risk database including pile body integrity, casing stability and mud property is constructed.

4. The BIM-based subway engineering rotary digging pile construction risk source identification method according to claim 1, characterized in that, According to the multi-dimensional risk database, the drilling parameters, mud pressure data and pile bottom sediment data are dynamically coupled and interactively verified, and a risk identification data set with a pile-soil interaction propagation chain is obtained, including: Based on the multi-dimensional risk database, numerical range analysis and abnormal value screening are performed on the drilling parameters to generate a drilling feature sequence; According to the drilling feature sequence, pressure gradient calculation and fluctuation analysis are performed on the mud pressure data to form a pressure change data table; Using the pressure change data table, the settlement rate analysis and distribution characteristic extraction are performed on the pile bottom sediment data to obtain sediment distribution data; Through the drilling feature sequence, the pressure change data table and the sediment distribution data, the correlation calculation is performed to obtain a risk coupling matrix; According to the risk coupling matrix, the node connectivity analysis and propagation intensity quantification are performed on the risk propagation path to form a pile-soil interaction propagation chain. The pile-soil interaction propagation chain and the risk characteristic data are dynamically associated to obtain a risk identification data set with a pile-soil interaction propagation chain.

5. The BIM-based subway engineering rotary digging pile construction risk source identification method according to claim 1, characterized in that, Based on the risk identification data set with the pile-soil interaction propagation chain, the entropy dynamic iterative calculation and multi-level weight adaptive allocation are performed on the pile forming quality factors, the stratum mutation factors and the surrounding environmental factors to form a risk assessment data table, including: According to the risk identification data set with the pile-soil interaction propagation chain, the numerical discretization and data standardization processing are performed on the pile forming quality factors to obtain a quality evaluation index sequence; According to the quality evaluation index sequence, the stratum interface identification and mutation characteristic extraction are performed on the stratum mutation factors to form a stratum layering data set; Based on the stratum layering data set, the spatial correlation analysis and influence domain division are performed on the surrounding environmental factors to obtain an environmental influence parameter table; The quality evaluation index sequence, the stratum layering data set and the environmental influence parameter table are subjected to information entropy calculation, and an entropy dynamic matrix is generated through data iteration processing; Using the entropy dynamic matrix, the weight initialization is performed, the hierarchical decomposition and weight allocation are performed on different level risk factors, and a multi-level weight allocation table is constructed; According to the multi-level weight allocation table, the risk assessment index is comprehensively calculated and processed to form a risk assessment data table.

6. The BIM-based subway engineering rotary digging pile construction risk source identification method according to claim 1, characterized in that, According to the risk assessment data table, the casing stability control data, the drilling parameter control data and the mud performance control data are screened and matched to obtain a risk prevention and control parameter set, including: According to the risk assessment data table, the stability threshold determination and parameter interval division are performed on the casing stability control data to obtain a casing control index set; The drilling rate of the drilling parameter control data is analyzed and the rotation speed range is defined by using the wall protection control index set, and a drilling control data table is formed; The rheological property analysis and density comparison of the mud performance control data are performed by using the drilling control data table, and a mud parameter sequence is obtained; Cross-validation and correlation calculation are performed according to the wall protection control index set, the drilling control data table and the mud parameter sequence, and a parameter matching matrix is generated; Based on the parameter matching matrix, mutual feedback effect analysis and constraint condition screening of the control parameters are performed, and a control parameter combination table is formed; The parameters in the control parameter combination table are comprehensively balanced and optimized, and a risk prevention and control parameter set is obtained.

7. The BIM-based subway engineering rotary digging pile construction risk source identification method according to claim 1, characterized in that, The risk prevention and control parameter set is imported into the BIM model, numerical calculation and parameter optimization of the pile foundation construction process prevention and control effect are performed, and construction control parameters are obtained, including: According to the risk prevention and control parameter set, the time sequence decomposition and process splitting of the construction process data are performed, and a construction process control node sequence is generated; The construction process control node sequence is used to reproduce the scene and simulate the process of the pile foundation construction condition, and a construction state data stream is formed; The response characteristics and effect of the prevention and control measures are quantitatively analyzed and numerically calculated through the construction state data stream, and a prevention and control response matrix is obtained; According to the prevention and control response matrix, the sensitivity analysis and influence quantification of the key control parameters are performed, and a parameter sensitivity table is formed; Based on the parameter sensitivity table, the constraint boundary adjustment and objective function construction of the control parameters are performed, and a parameter optimization sequence is obtained; The parameter optimization sequence is processed through multi-objective balance calculation and feedback correction, and construction control parameters are obtained.

8. A BIM-based underground engineering rotary digging pile construction risk source identification system for implementing the BIM-based underground engineering rotary digging pile construction risk source identification method according to any one of claims 1-7, characterized in that, The BIM-based metro engineering rotary drilling pile construction risk source identification system includes: The acquisition module is used to obtain drilling posture data, casing settlement data, pile forming quality data and surrounding environment data through field acquisition equipment, to digitally process construction parameters and environmental parameters, and to generate a rotary drilling construction information data set; The reconstruction module is used to decouple and reconstruct the time and space dimensions of the pile forming process data and real-time monitoring data according to the rotary drilling construction information data set, and to construct a multi-dimensional risk database including pile integrity, casing stability and mud properties; The verification module is used to perform dynamic coupling analysis and interactive verification of drilling parameters, mud pressure data, pile bottom sediment and other risk characteristic data according to the multi-dimensional risk database, and to obtain a risk identification data set with a pile-soil interaction propagation chain; The iteration module is used to perform entropy dynamic iteration calculation and multi-level weight adaptive allocation of pile forming quality factors, stratum mutation factors and surrounding environment factors based on the risk identification data set with the pile-soil interaction propagation chain, and to form a risk assessment data table; The screening module is used to screen and match the wall protection stability control data, drilling parameter control data and mud performance control data according to the risk assessment data table, and to obtain a risk prevention and control parameter set; The optimization module is used to import the risk prevention and control parameter set into the BIM model, to perform numerical calculation and parameter optimization of the pile foundation construction process prevention and control effect, and to obtain construction control parameters.

Images