Pile foundation reinforcing method and system based on karst area

Abstract

The invention relates to a pile foundation reinforcement method and system based on a karst area, and belongs to the technical field of foundation treatment and reinforcement. The pile foundation reinforcement method comprises the steps that a karst area pile foundation monitoring network is established, and karst cavity distribution characteristic parameters and underground water dynamic change parameters are obtained; constructing an adaptive evaluation model, and setting an initial real-time feedback correction parameter vector; based on a spatio-temporal evolution rule, generating a karst development comprehensive complexity index of dynamic evolution; a grading response mechanism is established, and pile foundation reinforcement early warning and response measures of corresponding grades are triggered; and constructing an intelligent decision-making system, and continuously optimizing an evaluation model and a reinforcement strategy through a feedback mechanism. According to the method, the karst development comprehensive complexity index is generated by introducing the underground water dynamic change parameters and combining the karst cavity distribution characteristic parameters, the evaluation of the geological environment is expanded from a static time point to a dynamic evolution process, and the comprehensiveness and accuracy of the evaluation are improved.

Description

Technical Field

[0001] The present invention relates to the field of foundation treatment and reinforcement, and in particular to a pile foundation reinforcement method and system based on karst areas. Background Art

[0002] Pile foundation construction in karst areas presents unique and severe technical challenges. The core issue stems from the inherent high complexity and dynamic uncertainty of the karst geological environment. Traditional pile foundation reinforcement methods rely primarily on geological surveys at the initial stages of the project. These surveys typically provide a static snapshot of geological data, reflecting static information such as the distribution of karst cavities and rock mass structure at the time of the survey.

[0003] These traditional methods have significant limitations. The data they obtain has limited coverage and timeliness, making it difficult to fully reflect the long-term dynamic evolution of the karst geological environment. The fluctuations in groundwater levels, changes in flow rate and pressure, and changes in water chemical composition in karst areas will continuously affect the stability of the rock mass. However, these dynamic factors are seriously ignored in traditional one-time assessments. Due to the lack of the ability to predict future changes in the geological environment, reinforcement schemes designed based on static data may not be able to adapt to future actual working conditions, thereby posing a hidden danger to the long-term safety and reliability of the project.

[0004] These limitations are primarily attributable to the limitations of data collection methods and assessment technologies. Fixed, one-time survey methods are unable to capture the dynamic process of geological environment changes over time, while traditional assessment models lack the ability to integrate multi-source dynamic information and make adaptive adjustments. When adverse changes occur in the geological environment where the pile foundation is located, managers are unable to obtain timely warnings and accurate assessment information, making it difficult to take effective preventive or remedial measures, thereby increasing engineering risks.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a pile foundation reinforcement method and system based on karst areas to solve the problems raised in the above background technology.

[0007] The technical solution of the present invention comprises the following methods: establishing a pile foundation monitoring network in karst areas, obtaining characteristic parameters of karst cavity distribution and groundwater dynamic change parameters by integrating multi-source data fusion technology of geological radar, acoustic wave detection and real-time sensors; Constructing an adaptive assessment model and setting an initial real-time feedback correction parameter vector, wherein the adaptive assessment model can automatically adjust its internal parameters according to changes in monitoring data to optimize the accuracy of the assessment; Based on the temporal and spatial evolution laws, the distribution characteristic parameters of karst cavities and the dynamic change parameters of groundwater are integrated and analyzed in multiple dimensions to generate a dynamic evolution comprehensive complexity index of karst development. Establish a graded response mechanism to trigger corresponding levels of pile foundation reinforcement warning and response measures based on the real-time trend of the comprehensive complexity index of karst development; Build an intelligent decision-making system to calculate the dynamic compensation strength, and use machine learning methods to establish a matching model between the historical case library and the current working conditions, and automatically generate the optimal pile foundation reinforcement compensation plan based on the strength; Implement closed-loop management, continuously monitor pile foundation response after implementing the reinforcement plan, and continuously optimize the evaluation model and reinforcement strategy through feedback mechanism.

[0008] Preferably, the step of obtaining characteristic parameters of karst cavity distribution specifically includes obtaining cavity volume ratio parameters, cavity distribution density parameters, cavity connectivity index parameters, cavity morphology complexity parameters, cavity burial depth influence parameters and cavity filling degree parameters; the step of obtaining groundwater dynamic change parameters specifically includes obtaining groundwater level change amplitude parameters, groundwater flow velocity change parameters, water pressure fluctuation parameters and water chemical erosion intensity parameters.

[0009] Preferably, the step of generating the comprehensive complexity index of karst development further comprises: establishing a spatial feature recognition model, and analyzing the spatial distribution pattern and aggregation characteristics of karst cavities by pattern recognition technology based on image processing and data mining; Construct a time evolution prediction model to predict the impact of groundwater dynamics on karst development based on historical monitoring data and geological evolution mechanisms; Develop a coupling analysis engine to identify the nonlinear coupling relationship between spatial distribution and temporal variation and discover key influencing factors; A multi-objective optimization method is used to comprehensively consider project safety, economy and construction feasibility to generate comprehensive evaluation indicators.

[0010] Preferably, the pile foundation design parameters are adjusted in real time, and the allowable bearing capacity and safety factor of the pile foundation are dynamically updated according to the dynamic bearing capacity correction coefficient; graded reinforcement measures are triggered, and when the dynamic bearing capacity correction coefficient is lower than the preset threshold, the corresponding level of reinforcement plan is automatically started; the construction plan is optimized, and the construction sequence and reinforcement key areas are dynamically adjusted according to the changing trend of the dynamic bearing capacity correction coefficient; long-term maintenance strategies are guided, and a pile foundation health assessment system and maintenance plan based on the dynamic bearing capacity correction coefficient are established.

[0011] Preferably, the step of calculating the dynamic compensation strength also includes: setting an evaluation factor weight for each evaluation factor in the multi-parameter coupling evaluation matrix, and defining an evaluation factor response function corresponding to the evaluation factor; processing the multi-parameter coupling evaluation matrix through the evaluation factor weight and the evaluation factor response function to calculate the dynamic compensation strength.

[0012] Preferably, the step of generating a corresponding pile foundation reinforcement compensation scheme based on the value of the dynamic compensation strength also includes: comparing the dynamic compensation strength with a preset compensation level threshold to determine the compensation level; based on the compensation level, matching and selecting the corresponding pile foundation reinforcement compensation scheme from a preset scheme library.

[0013] Preferably, the step of updating the real-time feedback correction parameter vector includes: updating the real-time feedback correction parameter vector of the previous time node based on the reinforcement feedback data and the preset learning rate parameter to obtain an updated real-time feedback correction parameter vector.

[0014] Preferably, the step of obtaining characteristic parameters of karst cavity distribution is achieved by adopting high-precision geological radar and acoustic wave detection technology; the step of obtaining dynamic change parameters of groundwater is achieved by real-time monitoring of groundwater level, flow rate, pressure and chemical composition.

[0015] A pile foundation reinforcement system based on karst areas, comprising: A data acquisition module is used to obtain karst cavity distribution characteristic parameters representing the spatial distribution state of karst cavities, and groundwater dynamic change parameters representing the dynamic change of groundwater; An initial parameter setting module is used to set the initial real-time feedback correction parameter vector; The complexity index generation module is used to generate a comprehensive karst development complexity index that comprehensively reflects the complexity of karst development based on the karst cavity distribution characteristic parameters and groundwater dynamic change parameters, and set the karst development comprehensive complexity index corresponding to the initial exploration time node as the benchmark complexity index; A correction coefficient generation module is used to generate a dynamic bearing capacity correction coefficient based on the relationship between the real-time updated karst development comprehensive complexity index and a preset maximum value; Evaluation matrix construction module, used to construct a multi-parameter coupling evaluation matrix integrating karst cavity distribution characteristic parameters, groundwater dynamic change parameters and real-time feedback correction parameter vectors; The compensation strength calculation module is used to set weights and response functions for each evaluation factor in the multi-parameter coupling evaluation matrix, and perform comprehensive processing on the matrix to calculate the dynamic compensation strength; A reinforcement scheme generation module is used to generate a corresponding pile foundation reinforcement compensation scheme according to the value of the dynamic compensation strength; The feedback and update module is used to monitor the status of the pile foundation to obtain reinforcement feedback data after executing the pile foundation reinforcement compensation plan, and to update the real-time feedback correction parameter vector based on the reinforcement feedback data.

[0016] The present invention provides a pile foundation reinforcement method and system for karst areas through improvements, which have the following improvements and advantages compared with the prior art: 1. By introducing groundwater dynamic change parameters and combining them with karst cavity distribution characteristic parameters, a comprehensive karst development complexity index is generated. This expands the assessment of the geological environment from a static time point to a dynamic evolution process, greatly improving the comprehensiveness and accuracy of the assessment. 2. By calculating dynamic compensation strength, multi-dimensional assessment information is converted into a single, quantifiable reinforcement strength index. Based on this, corresponding pile foundation reinforcement compensation schemes are generated from the scheme library, achieving differentiated and precise reinforcement, effectively avoiding waste caused by over-design or safety risks caused by under-design; 3. By generating a dynamic bearing capacity correction coefficient, the system can dynamically adjust the bearing capacity assessment based on real-time monitored changes in environmental complexity, providing early warning of risks. This dynamic adjustment capability significantly improves the long-term safety and reliability of pile foundation projects in complex karst geological environments. 4. The present invention introduces a feedback and update module, which iteratively updates the real-time feedback correction parameter vector through monitoring data after reinforcement. The closed-loop adaptive optimization path enables the system to have the ability of continuous learning and self-improvement, significantly improving the intelligent management level, engineering safety, efficiency and economy of the entire process of pile foundation reinforcement. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further explained below in conjunction with the accompanying drawings and Examples: Figure 1 It is a flow chart of a pile foundation reinforcement method based on karst areas according to the present invention. DETAILED DESCRIPTION

[0018] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0019] Example 1 See also Figure 1 The present invention provides a pile foundation reinforcement method and system technical solution based on karst areas, including: establishing a pile foundation monitoring network in karst areas, and obtaining karst cavity distribution characteristic parameters and groundwater dynamic change parameters by integrating multi-source data fusion technology of geological radar, acoustic wave detection and real-time sensors; Construct an adaptive assessment model and set the initial real-time feedback correction parameter vector. The adaptive assessment model can automatically adjust its internal parameters according to changes in monitoring data to optimize the accuracy of the assessment. Based on the temporal and spatial evolution laws, the distribution characteristic parameters of karst cavities and the dynamic change parameters of groundwater are integrated and analyzed in multiple dimensions to generate a dynamic evolution comprehensive complexity index of karst development. Establish a graded response mechanism to trigger corresponding levels of pile foundation reinforcement warning and response measures based on the real-time trend of the comprehensive complexity index of karst development; Build an intelligent decision-making system to calculate the dynamic compensation strength, and use machine learning methods to establish a matching model between the historical case library and the current working conditions, and automatically generate the optimal pile foundation reinforcement compensation plan based on the strength; Implement closed-loop management, continuously monitor pile foundation response after implementing the reinforcement plan, and continuously optimize the evaluation model and reinforcement strategy through feedback mechanism.

[0020] The present embodiment provides a pile foundation reinforcement method based on karst areas, which obtains karst cavity distribution characteristic parameters that characterize the spatial distribution state of karst cavities and groundwater dynamic change parameters that characterize the dynamic change of groundwater. The karst cavity distribution characteristic parameters and groundwater dynamic change parameters provide a data basis for the subsequent generation of a comprehensive complexity index of karst development. The setting of the initial real-time feedback correction parameter vector provides initial conditions for the dynamic adaptive adjustment of the compensation evaluation model. The generation of the comprehensive complexity index of karst development can quantitatively reflect the comprehensive complexity of the karst geological environment. The comprehensive complexity index of karst development corresponding to the initial survey time node is set as the benchmark complexity index, which establishes a reference benchmark for the subsequent complexity change evaluation, and the dynamic bearing capacity correction. The generation of positive coefficients can dynamically adjust the bearing capacity of traditional pile foundations based on the real-time changes in the comprehensive complexity index of karst development. The construction of a multi-parameter coupling evaluation matrix integrates geological static characteristics, hydrological dynamic characteristics, and feedback information after reinforcement. The calculation of dynamic compensation strength converts multi-dimensional evaluation information into a single, quantifiable reinforcement strength indicator. The generation of pile foundation reinforcement compensation schemes matches the optimal engineering measures based on the dynamic compensation strength to achieve differentiated and precise reinforcement. After reinforcement is executed, the reinforcement feedback data obtained by monitoring the pile foundation status is used to update the real-time feedback correction parameter vector, forming a closed-loop adaptive optimization path. This method significantly improves the safety and long-term reliability of pile foundation projects in complex karst geological environments.

[0021] The machine learning methods used by the intelligent decision-making system specifically include: building a historical engineering case database to extract key features including geological conditions, reinforcement schemes, and implementation effects; using supervised learning algorithms, such as support vector machines or random forest algorithms, to establish a mapping model between geological conditions and reinforcement schemes; optimizing model parameters through cross-validation to improve the accuracy of scheme matching; and in actual applications, inputting current working condition characteristics into the trained model to output recommendations for the optimal reinforcement scheme.

[0022] The specific implementation of closed-loop management includes: establishing a real-time monitoring system to continuously collect pile foundation displacement, stress, tilt and other state parameters; setting a performance evaluation index system to quantitatively evaluate the reinforcement effect; generating optimization suggestions by comparing and analyzing the deviation between actual effects and expected goals; feeding the optimization suggestions back to the evaluation model to update model parameters and decision-making rules; forming a complete closed loop of "monitoring-evaluation-decision-making-implementation-feedback" to achieve continuous optimization of the system.

[0023] Example 2 The steps of obtaining characteristic parameters of karst cavity distribution include obtaining cavity volume ratio parameters, cavity distribution density parameters, cavity connectivity index parameters, cavity morphology complexity parameters, cavity burial depth influence parameters and cavity filling degree parameters; the steps of obtaining groundwater dynamic change parameters include obtaining groundwater level change amplitude parameters, groundwater flow velocity change parameters, water pressure fluctuation parameters and water chemical erosion intensity parameters. The steps to generate the comprehensive complexity index of karst development include: Establish a spatial feature recognition model and analyze the spatial distribution and aggregation characteristics of karst cavities through pattern recognition technology based on image processing and data mining; Construct a time evolution prediction model to predict the impact of groundwater dynamics on karst development based on historical monitoring data and geological evolution mechanisms; Develop a coupling analysis engine to identify the nonlinear coupling relationship between spatial distribution and temporal variation and discover key influencing factors; A multi-objective optimization method is used to comprehensively consider project safety, economy and construction feasibility to generate comprehensive evaluation indicators.

[0024] The step of obtaining the characteristic parameters of karst cavity distribution is to construct a karst cavity distribution characteristic parameter matrix by collecting specific values ​​of multiple dimensions such as cavity volume ratio parameter, cavity distribution density parameter, and cavity connectivity index parameter. The step of obtaining groundwater dynamic change parameters is to establish a groundwater dynamic change parameter vector by monitoring time-varying factors such as groundwater level change amplitude parameter and groundwater flow velocity change parameter, thereby achieving comprehensive quantification of the static structure and dynamic changes of the geological environment. In the step of generating the comprehensive complexity index of karst development, these parameters are used for calculation. The calculation of the spatial distribution complexity function converts the multiple static geological characteristics contained in the karst cavity distribution characteristic parameter matrix into a scalar for evaluating the complexity of the spatial structure. The calculation of the temporal change complexity function integrates the hydrological dynamic information in the groundwater dynamic change parameter vector into an indicator reflecting the intensity of the temporal evolution of the geological environment. The calculation of the space-time coupling complexity function reveals the interactive influence between the static distribution of karst cavities and the dynamic changes of groundwater. Finally, the comprehensive complexity index of karst development is generated by weighted summation of the results of these three functions.

[0025] The steps of obtaining characteristic parameters of karst cavity distribution specifically include obtaining cavity volume ratio parameters, cavity distribution density parameters, cavity connectivity index parameters, cavity morphology complexity parameters, cavity burial depth influence parameters and cavity filling degree parameters; the steps of obtaining groundwater dynamic change parameters specifically include obtaining groundwater level change amplitude parameters, groundwater flow velocity change parameters, water pressure fluctuation parameters and water chemical erosion intensity parameters.

[0026] The process of generating the comprehensive complexity index of karst development adopts the following technical logic: First, a spatial distribution complexity assessment is conducted. The system uses three-dimensional feature recognition technology to conduct a comprehensive analysis of the acquired characteristic parameters of karst cavity distribution. Normalization processing technology is used to eliminate the dimensional differences between different parameters to ensure the objectivity of the assessment. Based on the engineering experience database, weights are dynamically assigned to each parameter, and the determination of the weights fully considers the mutual influence and synergistic effect between the parameters. Through the fuzzy comprehensive evaluation method, multiple parameters are intelligently integrated to generate a spatial distribution complexity evaluation result.

[0027] Secondly, the complexity of temporal changes is evaluated; a sliding time window is established in the system to continuously analyze the evolution trend of groundwater dynamic change parameters; through time series analysis technology, the periodic change patterns, mutation events and change acceleration of parameters are identified; the evaluation process pays special attention to the synchronization and lag characteristics of parameter changes, establishes a time evolution prediction model, and generates evaluation indicators reflecting the complexity of dynamic changes.

[0028] Secondly, conduct a space-time coupling complexity assessment; systematically and in-depth analyze the response mechanism of spatial defects under dynamic hydrological conditions, and identify the most dangerous "spatial weakness-temporal threat" combination; by establishing a correlation matrix, evaluate the mutual amplification effect of spatial and temporal factors, and identify key coupling points that may lead to chain reactions.

[0029] Finally, a comprehensive complexity index is generated. The system uses adaptive integration technology to intelligently integrate the evaluation results of the three dimensions of space, time and coupling. During the integration process, the contribution weights of each dimension are dynamically adjusted according to the actual working conditions and risk characteristics, and finally a comprehensive index is generated that can comprehensively and accurately reflect the complexity of karst development.

[0030] In specific implementation, the construction process of the spatial feature recognition model is as follows: first, the three-dimensional scanning data obtained by high-precision geological radar is used to identify the boundaries of the cavity through image segmentation algorithm; second, the cluster analysis method is used to identify the spatial distribution pattern of the cavity, including discrete, clustered, chain, etc.; third, by calculating the shortest distance between the cavities, connectivity probability and other indicators, the overall danger of the cavity group is evaluated; finally, combined with geomechanical analysis, the key areas where instability is most likely to occur are identified.

[0031] Acquisition of void volume ratio parameters: The total void volume is calculated using the voxelization method using the three-dimensional data obtained through geological radar scanning and divided by the total volume of the survey area; Acquisition of void connectivity index parameters: The degree of hydraulic connectivity between voids is determined using tracer tests or geophysical detection methods, and connectivity evaluation indicators are established; Acquisition of hydrochemical erosion intensity parameters: Groundwater samples are collected regularly to measure indicators such as pH value, dissolved oxygen, and calcium and magnesium ion concentrations, calculate the calcium carbonate saturation index, and evaluate erosion intensity.

[0032] Example 3 Adjust pile foundation design parameters in real time, and dynamically update the allowable bearing capacity and safety factor of the pile foundation based on the dynamic bearing capacity correction coefficient; trigger graded reinforcement measures, and automatically start the corresponding level of reinforcement plan when the dynamic bearing capacity correction coefficient is lower than the preset threshold; optimize the construction plan, and dynamically adjust the construction sequence and reinforcement key areas according to the changing trend of the dynamic bearing capacity correction coefficient; guide long-term maintenance strategies, and establish a pile foundation health assessment system and maintenance plan based on the dynamic bearing capacity correction coefficient.

[0033] The step of calculating the dynamic compensation strength also includes: setting an evaluation factor weight for each evaluation factor in the multi-parameter coupling evaluation matrix, and defining an evaluation factor response function corresponding to the evaluation factor; processing the multi-parameter coupling evaluation matrix using the evaluation factor weight and the evaluation factor response function to calculate the dynamic compensation strength; The steps for generating the dynamic bearing capacity correction coefficient are based on a nonlinear calculation of the comprehensive complexity index of karst development. Preset correction strength adjustment parameters are used to adjust the shape of the correction curve to adapt to the engineering experience of different geological regions. This calculation ensures that the range of the correction coefficient is reasonable through the ratio of the comprehensive complexity index of karst development to a preset maximum value.

[0034] The steps for generating the dynamic bearing capacity correction factor adopt the following technical solution: The system first determines the theoretical bearing capacity baseline for the pile foundation based on initial survey data and relevant design specifications. Taking into account the unique geological conditions of karst areas, it sets a corresponding safety reserve factor. Subsequently, an intelligent assessment algorithm determines the current geological risk level based on a real-time, updated karst development complexity index.

[0035] The correction coefficient is generated using a segmented mapping mechanism: when the comprehensive complexity index is in the low-risk range, the correction coefficient is close to 1, indicating that the geological conditions have little impact on the bearing capacity; as the complexity index increases, the correction coefficient gradually decreases, reflecting the weakening effect of geological risks on the bearing capacity; when the complexity index approaches the preset maximum value, the correction coefficient reaches the minimum value, triggering the highest level of safety warning.

[0036] The system introduces a correction strength adjustment parameter, which is personalized according to the project's importance level, service life requirements and risk tolerance, so that the change curve of the correction coefficient can adapt to the specific needs of different projects.

[0037] The process of calculating the dynamic compensation strength is as follows: The first step is to construct a multi-parameter coupled assessment matrix. The system integrates the characteristic parameters of karst cavity distribution, groundwater dynamics, and real-time feedback correction parameters into a structured, three-dimensional assessment matrix. The rows of the matrix represent different monitoring locations, the columns represent different types of assessment parameters, and the layers represent time series, achieving comprehensive coverage across the three dimensions of space, parameters, and time.

[0038] The second step is to design the response mechanism for the assessment factors. The system designs unique response characteristics for each assessment factor to accurately reflect its impact on pile foundation safety. For parameters with linear effects, a proportional response mode is used; for parameters with threshold effects, a step response mode is used; and for parameters sensitive within a specific range, an S-shaped response mode is used. This differentiated response design ensures the accuracy and pertinence of the assessment.

[0039] The third step is generating dynamic compensation strength. Using intelligent processing algorithms, the system extracts key risk characteristics from the assessment matrix, comprehensively considers the weights and response characteristics of each factor, and calculates a dynamic compensation strength value. This value quantitatively reflects the degree of pile foundation reinforcement required under current geological conditions, providing clear guidance for subsequent engineering decisions.

[0040] Example 4 The step of generating a corresponding pile foundation reinforcement compensation scheme based on the value of the dynamic compensation strength also includes: comparing the dynamic compensation strength with a preset compensation level threshold to determine the compensation level; and selecting a corresponding pile foundation reinforcement compensation scheme from a preset scheme library based on the compensation level; The step of obtaining characteristic parameters of karst cavity distribution is achieved by using high-precision geological radar and acoustic wave detection technology; the step of obtaining groundwater dynamic change parameters is achieved by real-time monitoring of groundwater level, flow rate, pressure and chemical composition; the step of updating the real-time feedback correction parameter vector includes: based on reinforcement feedback data and a preset learning rate parameter, updating the real-time feedback correction parameter vector of the previous time node to obtain an updated real-time feedback correction parameter vector; The generation of the pile foundation reinforcement compensation scheme depends on the calculation results of the dynamic compensation strength. The dynamic compensation strength value is used to compare with multiple preset compensation level thresholds. The result of this comparison directly determines the compensation level of the reinforcement measure, such as light, moderate or heavy compensation. Based on the determined compensation level, the system matches and selects the pile foundation reinforcement compensation scheme that best suits the current level from a preset scheme library containing a variety of standardized engineering measures. The characteristic parameters of karst cavity distribution are obtained through non-destructive exploration technologies such as high-precision geological radar and acoustic wave detection, ensuring the accuracy of the data. The dynamic change parameters of groundwater are obtained by long-term real-time monitoring of groundwater level, flow rate, etc. through on-site sensors, ensuring the timeliness of the data. After the reinforcement plan is implemented, the update of the real-time feedback correction parameter vector is the core of achieving system self-optimization. This update is based on the monitored reinforcement feedback data and a preset learning rate parameter, and the parameter vector of the previous time node is iteratively corrected. The update process is driven by the following formula: ; in: represents the updated real-time feedback correction parameter vector, represents the real-time feedback correction parameter vector of the previous time node, Indicates the correction amount calculated based on reinforcement feedback data. The calculation is based on the comparison of the monitoring results of key performance indicators after pile foundation reinforcement with the expected target values. For example, one or more performance indicators can be defined, such as the pile top settlement rate. , bearing capacity changes etc., the correction amount can be defined as the actual value and target value The weighted error between , can be calculated as: ; in: and is the dimensionless weight coefficient, the correction amount Essentially, it is the error signal in feedback control theory, used to guide the iterative optimization direction of parameters; Represents the preset learning rate parameter. This closed-loop feedback and update mechanism enables the entire reinforcement assessment system to learn and adapt, and can continuously improve the accuracy of assessment and decision-making; In the implementation of the present invention, multiple preset parameters and thresholds are the basis for implementing the present method and system, and their settings can be based on the following principles: Weight coefficient ( ), these weights can be scientifically determined by organizing experts in the field to score the degree of influence of various factors on pile foundation stability under different geological conditions through the hierarchical analysis method; the maximum complexity index Based on the statistical analysis of historical survey data of the project area or similar geological areas, the maximum possible karst development complexity or an upper limit with sufficient safety margin can be taken; the intensity adjustment parameters can be modified. As a regulating parameter, it is calibrated by fitting the historical data of existing projects to make the change trend of the correction coefficient most consistent with actual engineering experience; the learning rate is a key parameter for iterative optimization, and its value is usually a small positive number, such as arrive The trial and error method or adaptive learning rate algorithm can be used to determine the compensation level threshold: the dynamic compensation intensity The thresholds for different compensation levels can be combined with risk level assessment, for example: mild compensation (0 < ≤0.3): low risk, the corresponding measure is to increase the frequency of monitoring; moderate compensation (0.3< ≤0.7): medium risk, the corresponding measures are local or selective grouting reinforcement; heavy compensation ( >0.7): High risk. Corresponding measures include full grouting, additional pile foundations, or structural design adjustments.

[0041] Example 5 A pile foundation reinforcement system based on karst areas includes: a data acquisition module for acquiring karst cavity distribution characteristic parameters representing the spatial distribution state of karst cavities and groundwater dynamic change parameters representing the dynamic change of groundwater; an initial parameter setting module for setting an initial real-time feedback correction parameter vector; a complexity index generation module for generating a karst development comprehensive complexity index that comprehensively reflects the complexity of karst development based on the karst cavity distribution characteristic parameters and the groundwater dynamic change parameters, and setting the karst development comprehensive complexity index corresponding to the initial exploration time node as the benchmark complexity index; a correction coefficient generation module for generating a karst development comprehensive complexity index based on the real-time updated karst development comprehensive complexity index and the preset maximum complexity index. The dynamic bearing capacity correction coefficient is generated based on the relationship between the values; the evaluation matrix construction module is used to construct a multi-parameter coupling evaluation matrix that integrates the karst cavity distribution characteristic parameters, groundwater dynamic change parameters and real-time feedback correction parameter vectors; the compensation strength calculation module is used to set weights and response functions for each evaluation factor in the multi-parameter coupling evaluation matrix, and perform comprehensive processing on the matrix to calculate the dynamic compensation strength; the reinforcement scheme generation module is used to generate the corresponding pile foundation reinforcement compensation scheme according to the value of the dynamic compensation strength; the feedback and update module is used to monitor the status of the pile foundation after executing the pile foundation reinforcement compensation scheme to obtain reinforcement feedback data, and update the real-time feedback correction parameter vector based on the reinforcement feedback data; This embodiment provides a pile foundation reinforcement system based on karst areas. The data acquisition module in the system is responsible for performing exploration and monitoring tasks, obtaining karst cavity distribution characteristic parameters and groundwater dynamic change parameters, the initial parameter setting module sets the initial real-time feedback correction parameter vector for the system's dynamic evaluation model, the complexity index generation module receives data from the data acquisition module, calculates the karst development comprehensive complexity index, and solidifies the benchmark complexity index, the correction coefficient generation module generates a dynamic bearing capacity correction coefficient according to the real-time change of the complexity index, the evaluation matrix construction module integrates all relevant parameters, and constructs a multi-parameter coupling evaluation matrix as the basis for comprehensive evaluation, the compensation strength calculation module processes the matrix, and outputs a quantified dynamic compensation strength, and the reinforcement formula The solution generation module automatically matches and generates pile foundation reinforcement compensation solutions from a solution library based on dynamic compensation intensity. The solution library is a pre-established database of standardized engineering measures that maps compensation levels to specific, executable reinforcement solutions. An exemplary solution library may include: a mild compensation solution: increasing monitoring frequency to daily and adding groundwater level and microseismic monitoring points; a moderate compensation solution: injecting ordinary cement grout into identified void clusters and grouting sleeve valve pipes around key pile foundations; and a severe compensation solution: performing full-scale pressure grouting using cement-water glass dual-liquid grout, inserting micro-steel pipe piles between existing pile foundations, and rechecking the design to consider reducing superstructure loads. This solution library can be continuously enriched and improved based on engineering practice. After reinforcement is completed, the feedback and update module iteratively updates the real-time feedback correction parameter vector through monitoring data. The system integrates data collection, complexity assessment, dynamic correction, solution decision-making and closed-loop feedback update into an organic whole, realizing the full-process intelligent management of pile foundation reinforcement in karst areas. Compared with the traditional step-by-step working mode that relies on manual judgment, it significantly improves the safety, efficiency and economy of the project.

[0042] During project implementation, this method improved the reinforcement effect through the following innovations: establishing a three-dimensional visualization monitoring platform to display the distribution and evolution process of karst cavities in real time; adopting a zoning and grading management strategy to formulate differentiated reinforcement plans for areas with different risk levels; introducing an intelligent early warning system to identify potential risks in advance and take proactive preventive measures; and establishing an expert knowledge base to transform engineering experience into reusable decision-making rules.

[0043] The technical advantages of this invention are: achieving a shift from passive reinforcement to active prevention; establishing a complete "monitoring-assessment-decision-making-implementation-feedback" closed-loop system; providing traceable decision-making basis and quantitative evaluation indicators; and significantly improving the safety and reliability of pile foundation projects in karst areas.

[0044] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A pile foundation reinforcement method based on karst areas, characterized in that: include: Establish a pile foundation monitoring network in karst areas, and obtain karst cavity distribution characteristic parameters and groundwater dynamic change parameters by integrating multi-source data fusion technology of geological radar, acoustic wave detection and real-time sensors; Constructing an adaptive assessment model and setting an initial real-time feedback correction parameter vector, wherein the adaptive assessment model can automatically adjust its internal parameters according to changes in monitoring data to optimize the accuracy of the assessment; Based on the temporal and spatial evolution laws, the distribution characteristic parameters of karst cavities and the dynamic change parameters of groundwater are integrated and analyzed in multiple dimensions to generate a dynamic evolution comprehensive complexity index of karst development. Establish a graded response mechanism to trigger corresponding levels of pile foundation reinforcement warning and response measures based on the real-time trend of the comprehensive complexity index of karst development; Build an intelligent decision-making system to calculate the dynamic compensation strength, and use machine learning methods to establish a matching model between the historical case library and the current working conditions, and automatically generate the optimal pile foundation reinforcement compensation plan based on the strength; Implement closed-loop management, continuously monitor pile foundation response after implementing the reinforcement plan, and continuously optimize the evaluation model and reinforcement strategy through feedback mechanism.

2. The pile foundation reinforcement method in karst areas according to claim 1, characterized in that: The steps of obtaining characteristic parameters of karst cavity distribution specifically include obtaining cavity volume ratio parameters, cavity distribution density parameters, cavity connectivity index parameters, cavity morphology complexity parameters, cavity burial depth influence parameters and cavity filling degree parameters; the steps of obtaining groundwater dynamic change parameters specifically include obtaining groundwater level change amplitude parameters, groundwater flow velocity change parameters, water pressure fluctuation parameters and water chemical erosion intensity parameters.

3. The pile foundation reinforcement method in karst areas according to claim 1, characterized in that: The steps to generate the comprehensive complexity index of karst development include: Establish a spatial feature recognition model and analyze the spatial distribution and aggregation characteristics of karst cavities through pattern recognition technology based on image processing and data mining; Construct a time evolution prediction model to predict the impact of groundwater dynamics on karst development based on historical monitoring data and geological evolution mechanisms; Develop a coupling analysis engine to identify the nonlinear coupling relationship between spatial distribution and temporal variation and discover key influencing factors; A multi-objective optimization method is used to comprehensively consider project safety, economy and construction feasibility to generate comprehensive evaluation indicators.

4. The pile foundation reinforcement method in karst areas according to claim 1, characterized in that: Adjust pile foundation design parameters in real time, and dynamically update the allowable bearing capacity and safety factor of the pile foundation based on the dynamic bearing capacity correction coefficient; trigger graded reinforcement measures, and automatically start the corresponding level of reinforcement plan when the dynamic bearing capacity correction coefficient is lower than the preset threshold; optimize the construction plan, and dynamically adjust the construction sequence and reinforcement key areas according to the changing trend of the dynamic bearing capacity correction coefficient; guide long-term maintenance strategies, and establish a pile foundation health assessment system and maintenance plan based on the dynamic bearing capacity correction coefficient.

5. The pile foundation reinforcement method in karst areas according to claim 1, characterized in that: The step of calculating the dynamic compensation strength also includes: setting an evaluation factor weight for each evaluation factor in the multi-parameter coupling evaluation matrix, and defining an evaluation factor response function corresponding to the evaluation factor; processing the multi-parameter coupling evaluation matrix through the evaluation factor weight and the evaluation factor response function to calculate the dynamic compensation strength.

6. The pile foundation reinforcement method in karst areas according to claim 1, characterized in that: The step of generating a corresponding pile foundation reinforcement compensation scheme based on the value of the dynamic compensation strength also includes: comparing the dynamic compensation strength with a preset compensation level threshold to determine the compensation level; and based on the compensation level, matching and selecting a corresponding pile foundation reinforcement compensation scheme from a preset scheme library.

7. The pile foundation reinforcement method in karst areas according to claim 1, characterized in that: The step of updating the real-time feedback correction parameter vector includes: updating the real-time feedback correction parameter vector of the previous time node based on the reinforcement feedback data and the preset learning rate parameter to obtain an updated real-time feedback correction parameter vector.

8. The pile foundation reinforcement method in karst areas according to claim 1, characterized in that: The step of obtaining characteristic parameters of karst cavity distribution is achieved by adopting high-precision geological radar and acoustic wave detection technology; the step of obtaining dynamic change parameters of groundwater is achieved by real-time monitoring of groundwater level, flow rate, pressure and chemical composition.

9. A pile foundation reinforcement system based on karst areas, applied to a pile foundation reinforcement method based on karst areas according to any one of claims 1 to 8, characterized in that: include: A data acquisition module is used to obtain karst cavity distribution characteristic parameters representing the spatial distribution state of karst cavities, and groundwater dynamic change parameters representing the dynamic change of groundwater; An initial parameter setting module is used to set the initial real-time feedback correction parameter vector; The complexity index generation module is used to generate a comprehensive karst development complexity index that comprehensively reflects the complexity of karst development based on the karst cavity distribution characteristic parameters and groundwater dynamic change parameters, and set the karst development comprehensive complexity index corresponding to the initial exploration time node as the benchmark complexity index; A correction coefficient generation module is used to generate a dynamic bearing capacity correction coefficient based on the relationship between the real-time updated karst development comprehensive complexity index and a preset maximum value; Evaluation matrix construction module, used to construct a multi-parameter coupling evaluation matrix integrating karst cavity distribution characteristic parameters, groundwater dynamic change parameters and real-time feedback correction parameter vectors; The compensation strength calculation module is used to set weights and response functions for each evaluation factor in the multi-parameter coupling evaluation matrix, and perform comprehensive processing on the matrix to calculate the dynamic compensation strength; A reinforcement scheme generation module is used to generate a corresponding pile foundation reinforcement compensation scheme according to the value of the dynamic compensation strength; The feedback and update module is used to monitor the status of the pile foundation to obtain reinforcement feedback data after executing the pile foundation reinforcement compensation plan, and to update the real-time feedback correction parameter vector based on the reinforcement feedback data.

Images