Method for calculating bearing capacity of cast-in-situ bored pile

Abstract

The invention relates to the technical field of buildings, and discloses a cast-in-situ bored pile bearing capacity calculation method which comprises the following steps that pile body geometric parameter data, pile periphery soil property parameter data and pile top load historical data of a cast-in-situ bored pile are collected; pile body integrity detection processing is conducted on the basis of the pile body geometric parameter data, the pile periphery soil property parameter data and the pile top load historical data, and pile body integrity detection data are generated; when the bearing capacity of the cast-in-situ bored pile is calculated, a standard pile body defect type database is established, corresponding bearing capacity influence coefficients are matched according to different pile body defect types, the pertinence of bearing capacity calculation under different defect conditions is guaranteed, meanwhile, pile body defect characteristics detected in real time are intelligently matched with the standard database, and the bearing capacity of the cast-in-situ bored pile is calculated. The problem of defect influence coefficient value solidification in a traditional method is solved, scientificity and accuracy of a bearing capacity calculation result are guaranteed, and safety risks caused by improper defect evaluation are reduced.

Description

Technical Field

[0001] This invention relates to the field of building technology, specifically to a method for calculating the bearing capacity of bored piles. Background Technology

[0002] Drilled piles are a type of pile, referring to piles constructed by creating pile holes in the foundation soil on-site through mechanical drilling, steel pipe extrusion, or manual excavation, placing a reinforcing cage inside, and then pouring concrete. Pile foundations are widely used in bridges, high-rise buildings, and other constructions due to their advantages such as high bearing capacity, good stability, and small and uniform settlement. Piles are vertical and inclined foundation components set in the soil, and their function is to penetrate soft, highly compressible soil layers and water, transferring the load borne by the pile to a harder, denser, and less compressible bearing layer.

[0003] Currently, due to various uncertainties in pile defect identification and bearing capacity calculation during the construction of bored cast-in-place piles, the traditional detection system cannot dynamically match the pile defect type with the standard bearing capacity influence coefficient when performing real-time calculation of the bearing capacity of bored cast-in-place piles. When the defect identification error and the influence coefficient are not accurately matched, the deviation of the bearing capacity calculation result will accumulate significantly, and the accuracy of the safety assessment of pile foundation engineering cannot be guaranteed.

[0004] Therefore, a method for calculating the bearing capacity of bored piles is proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for calculating the bearing capacity of bored piles, which solves the problem mentioned in the background technology that the accumulated deviation of the bearing capacity calculation results is large and cannot guarantee the accuracy of the safety assessment of pile foundation projects.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for calculating the bearing capacity of bored piles, the method comprising the following steps: S1. Collect data on the geometric parameters of the bored pile, the soil parameters around the pile, and the historical data on the load on the pile top. S2. Based on the pile body geometric parameter data, pile surrounding soil parameter data and pile top load historical data, perform pile body integrity detection processing to generate pile body integrity detection data; S3. Based on the pile integrity detection data, perform pile defect type identification processing to generate pile defect type feature data; S4. Call the pre-stored standard pile defect type database, and perform pile bearing capacity influence coefficient matching processing based on the pile defect type feature data and the standard pile defect type database to generate pile bearing capacity influence coefficient data. S5. Based on the pile body geometric parameter data, pile surrounding soil parameter data, pile top load historical data and pile body bearing capacity influence coefficient data, perform initial bearing capacity calculation processing to generate initial bearing capacity data for bored piles. S6. Based on the initial bearing capacity data and the stress wave propagation characteristics of the pile body, the bearing capacity is corrected to generate corrected bearing capacity data for bored piles. S7. The modified bearing capacity data is optimized and verified using artificial intelligence algorithms to generate the final bearing capacity data of the bored pile. S8. Based on the final bearing capacity data, perform intelligent output and feedback control processing, including automatic generation and format optimization of bearing capacity reports, multi-platform synchronization and cloud storage of bearing capacity data, and real-time verification and abnormal warning push of calculation results.

[0007] Preferably, the step S1, which involves collecting data on the geometric parameters of the bored pile, the soil parameters around the pile, and the historical load data at the pile top, includes the following steps: S11. Multi-point acoustic velocity measurement is performed along the length of the pile using an ultrasonic testing instrument to generate pile geometric parameter data, including pile diameter, pile length, and pile cross-sectional change rate. S12. Soil samples are collected at different depths around the pile using a soil sampler, and laboratory geotechnical tests are conducted to generate soil parameter data around the pile. The soil parameter data around the pile includes the soil's internal friction angle, cohesion, and compression modulus. S13. Apply a phased load to the top of the pile using a load sensor and record the settlement to generate historical load data at the top of the pile. The historical load data at the top of the pile includes the load-settlement curve and the maximum test load.

[0008] Preferably, the generation of pile integrity detection data in step S2 includes the following steps: S21. Import the generated pile body geometric parameter data, pile surrounding soil parameter data, and pile top load historical data into the pile foundation monitoring platform; S22. The fast Fourier transform algorithm is used to perform frequency analysis on the acoustic velocity measurement data, extract the characteristic frequency of pile integrity, and generate pile integrity detection data. S23. Remove environmental noise interference through filtering algorithms to optimize the data signal-to-noise ratio.

[0009] Preferably, the generation of pile defect type feature data in S3 includes the following steps: S31. Obtain the pile integrity detection data; S32. Input the data into a pre-trained convolutional neural network model to perform pile body defect image recognition. Generate pile body defect type feature data based on the recognition results. When a crack defect is identified, output the defect type as crack. When a void defect is identified, output the defect type as void. When a segregation defect is identified, output the defect type as segregation.

[0010] Preferably, the generation of pile bearing capacity influence coefficient data in S4 includes the following steps: S41. Establish a standard pile defect type database, which contains bearing capacity influence coefficients corresponding to various defect types; S42. The pile defect type feature data is matched with the database for similarity, and the matching degree is calculated using the Euclidean distance algorithm to generate the pile bearing capacity influence coefficient data.

[0011] Preferably, the process of generating the initial bearing capacity data of the bored pile in step S5 includes the following steps: S51. Obtain the pile body geometric parameter data, pile surrounding soil parameter data, pile top load historical data, and pile body bearing capacity influence coefficient data; S52. Calculate the pile side surface area and pile bottom area based on the pile body geometric parameter data, and determine the soil bearing capacity parameters by combining the soil around the pile parameter data. S53. Analyze the load transfer law based on the historical data of pile top load, and calculate the components of pile side friction and pile end resistance. S54. Introduce pile bearing capacity influence coefficient data to dynamically adjust calculation parameters and comprehensively calculate initial bearing capacity; S55. The initial bearing capacity calculation results are verified by iterative optimization algorithm to generate the initial bearing capacity data of bored piles.

[0012] Preferably, the step S6 of generating the corrected bearing capacity data for bored piles includes the following steps: S61. Obtain the initial bearing capacity data; S62. Apply an impact load to the top of the pile using a stress wave sensor, measure the propagation time and attenuation of the stress wave in the pile body, and generate stress wave propagation data. S63. Based on stress wave propagation data, the initial bearing capacity is corrected using a wave equation correction model. The correction process is based on wave velocity variation and attenuation coefficient to generate corrected bearing capacity data for bored piles.

[0013] Preferably, the step of generating the final bearing capacity data of the bored pile in S7 includes the following steps: S71. Construct a neural network model, with the input being the corrected bearing capacity data, pile geometric parameter data, and pile surrounding soil parameter data, and the output being the optimized bearing capacity; S72. The neural network is trained using historical carrying capacity data as labels, and the prediction error is minimized through the backpropagation algorithm. S73. Input the data into the trained model to generate the final bearing capacity data of the bored pile.

[0014] Preferably, the intelligent output and feedback control processing based on the final bearing capacity data in step S8 includes the following steps: S81. Import the final bearing capacity data into the report generation system; S82. Automatically generate a bearing capacity calculation report, including a data summary, calculation process, result analysis, and safety recommendations; S83. Display the bearing capacity settlement curve and defect distribution map through a graphical interface.

[0015] Preferably, the method further includes a step S9 of real-time monitoring of load-bearing capacity changes: S91. Deploy long-term monitoring sensors on the pile body to continuously collect load and deformation data; S92. Data is transmitted to the cloud analysis system in real time through the Internet of Things platform; S93. Regularly update the bearing capacity model to achieve dynamic calibration; S94. When the monitoring data is abnormal, the early warning mechanism is automatically triggered.

[0016] Compared with the prior art, the present invention provides a method for calculating the bearing capacity of bored piles, which has the following beneficial effects: 1. In this invention, when calculating the bearing capacity of bored piles, a standard pile defect type database is established, and corresponding bearing capacity influence coefficients are matched for different pile defect types. This ensures the pertinence of bearing capacity calculations under different defect conditions. At the same time, the real-time detected pile defect characteristics are intelligently matched with the standard database, which can dynamically determine the actual impact of pile defects on bearing capacity. This solves the problem of fixed defect influence coefficient values ​​in traditional methods, ensures the scientificity and accuracy of bearing capacity calculation results, and reduces safety risks caused by improper defect assessment.

[0017] 2. In this invention, when calculating the bearing capacity of bored piles, the initial bearing capacity is corrected in real time by analyzing the propagation characteristics of stress waves. The propagation state of stress waves in the pile body is dynamically judged to determine whether it is abnormal. The potential impact of wave velocity changes on the bearing capacity result can be identified in a timely manner. When an abnormal wave velocity is detected, the bearing capacity value can be adjusted in real time through the wave equation model. This ensures that the bearing capacity calculation process can be dynamically optimized according to the actual response of the pile body, thereby improving the reliability of the calculation results.

[0018] 3. In this invention, when calculating the bearing capacity of bored piles, artificial intelligence algorithms are used to fuse and optimize multi-source calculation data, verify the output results of different calculation models in real time, and automatically allocate weights and calibrate results based on the deviation values ​​of different models. This enables intelligent decision-making through multi-model collaboration, avoids the limitations of single calculation methods, reduces bearing capacity calculation errors caused by improper model selection and data errors, and comprehensively improves the accuracy and applicability of pile foundation engineering safety assessment. Attached Figure Description

[0019] Figure 1 This is a flowchart of the borehole pile bearing capacity calculation method of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] For specific implementation examples, please refer to: Figure 1 A method for calculating the bearing capacity of bored piles, characterized by comprising the following steps: S1. Collect data on the geometric parameters of the bored pile, the soil parameters around the pile, and the historical data on the load on the pile top. S2. Based on the pile body geometric parameter data, pile surrounding soil parameter data and historical data of pile top load, perform pile body integrity detection processing to generate pile body integrity detection data; S3. Based on the pile integrity detection data, identify the type of pile defect and generate pile defect type feature data. S4. Call the pre-stored standard pile defect type database, and perform pile bearing capacity influence coefficient matching processing based on the pile defect type feature data and the standard pile defect type database to generate pile bearing capacity influence coefficient data. S5. Based on the pile body geometric parameter data, pile surrounding soil parameter data, pile top load historical data, and pile body bearing capacity influence coefficient data, perform initial bearing capacity calculation and generate initial bearing capacity data for bored piles. S6. Based on the initial bearing capacity data and the stress wave propagation characteristics of the pile body, the bearing capacity is corrected to generate corrected bearing capacity data for bored piles. S7. The modified bearing capacity data is optimized and verified using artificial intelligence algorithms to generate the final bearing capacity data of the bored pile. S8. Intelligent output and feedback control processing based on the final bearing capacity data, including automatic generation and format optimization of bearing capacity reports, multi-platform synchronization and cloud storage of bearing capacity data, real-time verification of calculation results and abnormal early warning push.

[0022] The steps involved in collecting pile geometry parameters, surrounding soil parameters, and historical load data at the pile top in S1 are as follows: S11. Multi-point acoustic velocity measurement is performed along the length of the pile using an ultrasonic testing instrument to generate pile geometric parameter data, including pile diameter, pile length, and pile cross-sectional change rate. S12. Soil samples are collected at different depths around the pile using a soil sampler, and laboratory geotechnical tests are conducted to generate soil parameter data around the pile, including the soil's internal friction angle, cohesion, and compression modulus. S13. Apply a phased load to the top of the pile using a load sensor and record the settlement to generate historical load data for the top of the pile. The historical load data for the top of the pile includes the load-settlement curve and the maximum test load.

[0023] The steps involved in generating pile integrity detection data in S2 are as follows: S21. Import the generated pile geometry parameter data, pile surrounding soil parameter data, and pile top load historical data into the pile foundation monitoring platform; S22. The fast Fourier transform algorithm is used to perform frequency analysis on the acoustic velocity measurement data, extract the characteristic frequencies of pile integrity, and generate pile integrity detection data. This includes the following steps: S221. Sample the sound wave velocity measurement data at equal intervals according to the time series to generate a discrete signal sequence; S222. Apply the Hanning window function to the discrete signal sequence to reduce spectral leakage. This includes the following steps: The formula for the Hanning window function is as follows: ; The butterfly algorithm calculation formula is as follows: ; The formula for the Fast Fourier Transform is as follows: ; S224. Extract the frequency component with the largest amplitude in the frequency domain signal as the characteristic frequency of pile integrity. S23. Remove environmental noise interference and optimize the data signal-to-noise ratio using a filtering algorithm, specifically including the following steps: S231. Set the frequency range of the bandpass filter to the bandwidth of the dominant frequency of the acoustic signal; S232. An infinite impulse response filter is used to perform convolution operations on the acoustic velocity measurement data to suppress high-frequency noise and low-frequency drift. The transfer function formula for the infinite impulse response filter is: ; S233. Adjust the phase of the filtered signal using zero-phase filtering technology to reduce signal distortion; S234. Calculate the signal-to-noise ratio of the filtered data. If it is lower than the threshold, readjust the filter parameters. The signal-to-noise ratio (SNR) is calculated using the following formula: .

[0024] The steps involved in generating pile defect type feature data in S3 are as follows: S31. Obtain pile integrity detection data; S32. Input the data into the pre-trained convolutional neural network model to perform pile defect image recognition. Generate pile defect type feature data based on the recognition results. When a crack defect is identified, the output defect type is crack; when a void defect is identified, the output defect type is void; when a segregation defect is identified, the output defect type is segregation. Specifically, this includes the following steps: S321. Construct a convolutional neural network structure, with the input layer receiving pile integrity detection data as a grayscale image; The formula for the convolution operation is as follows: ; S322. Extract local features from the image through convolutional layers and enhance nonlinearity using the ReLU activation function; The ReLU activation function formula is as follows: ; S323. Pooling layers downsample the feature maps, preserving key features and reducing the number of parameters. S324, the fully connected layer maps features to the defect type classification results, and outputs the probability distribution of cracks, voids and segregation; The formula for the Softmax function is: ; The steps involved in generating pile bearing capacity influence coefficient data in S4 are as follows: S41. Establish a standard pile defect type database, which contains bearing capacity influence coefficients corresponding to various defect types; S42. Perform similarity matching between the pile defect type feature data and the database, calculate the matching degree using the Euclidean distance algorithm, and generate pile bearing capacity influence coefficient data. This includes the following steps: S421. Represent the pile defect type feature data as a multi-dimensional vector; S422. Calculate the Euclidean distance between this vector and the vector corresponding to each defect type in the standard database; The Euclidean distance formula is as follows: ; S423. Select the standard defect type with the smallest Euclidean distance as the matching result; S424. Output the corresponding bearing capacity influence coefficient based on the matching results.

[0025] The steps involved in generating initial bearing capacity data for bored piles in S5 are as follows: S51. Obtain pile geometric parameters, pile surrounding soil parameters, historical pile top load data, and pile bearing capacity influence coefficient data. S52. Calculate the pile side surface area and pile bottom area based on the pile body geometric parameter data, and determine the soil bearing capacity parameters by combining the soil around the pile parameter data. S53. Analyze the load transfer law based on the historical data of pile top load, and calculate the components of pile side friction and pile end resistance. S54. Introduce pile bearing capacity influence coefficient data to dynamically adjust calculation parameters and comprehensively calculate initial bearing capacity; S55. Verify the initial bearing capacity calculation results through iterative optimization algorithms to generate initial bearing capacity data for bored piles, specifically including the following steps: S551. Set the error threshold and maximum number of iterations for the initial bearing capacity calculation results; S552. Calculate the predicted value of pile top settlement corresponding to the current bearing capacity result, and compare it with the measured settlement value. S553. When the error exceeds the threshold, adjust the pile side friction coefficient and pile end resistance coefficient, and recalculate the bearing capacity. The error calculation formula is as follows: ; S554. Repeat the iteration until the error is lower than the threshold and the maximum number of iterations is reached, and output the final verified bearing capacity data. The iterative update formula is as follows: .

[0026] Generating corrected bearing capacity data for bored piles in S6 includes the following steps: S61. Obtain initial bearing capacity data; S62. Apply an impact load to the top of the pile using a stress wave sensor, measure the propagation time and attenuation of the stress wave in the pile body, and generate stress wave propagation data. S63. Based on stress wave propagation data, the initial bearing capacity is corrected using a wave equation correction model. The correction process is based on wave velocity variation and attenuation coefficient to generate corrected bearing capacity data for bored piles. Specifically, this includes the following steps: S631. Establish a one-dimensional stress wave propagation equation and define the density and elastic modulus parameters of the pile material. The one-dimensional stress wave equation is as follows: ; S632. Substitute the stress wave propagation data into the equation to calculate the wave velocity change and attenuation coefficient. S633. Solve the wave equation analytically to obtain the bearing capacity correction factor; The formula for the bearing capacity correction factor is as follows: ; S634. Multiply the correction factor by the initial bearing capacity data to generate the corrected bearing capacity result.

[0027] The steps involved in generating the final bearing capacity data for bored piles in S7 are as follows: S71. Construct a neural network model, with the inputs being corrected bearing capacity data, pile geometric parameter data, and pile surrounding soil parameter data, and the output being optimized bearing capacity. This includes the following steps: S711. Design a feedforward neural network architecture, with the input layer containing corrected bearing capacity data, pile geometry parameters, and soil parameters. S712, The hidden layer uses the hyperbolic tangent activation function for nonlinear transformation; The hyperbolic tangent activation function formula is as follows: ; S713, The output layer generates optimized load-bearing capacity values ​​through a linear activation function; The formula for the linear activation function is: ; S714. Initialize network weights using the stochastic gradient descent algorithm; The formula for the stochastic gradient descent algorithm is as follows: ; S72. Training the neural network uses historical carrying capacity data as labels and minimizes the prediction error through backpropagation algorithm, specifically including the following steps: S721, Forward propagation calculates the error between the output value of the neural network and the historical load capacity label; S722. Calculate the gradient of the error function with respect to each weight, and backpropagate the error using the chain rule. The error function formula is as follows: ; S723. Update the weight matrix according to the gradient descent rule, and set the learning rate to adaptive adjustment mode; The weight update formula is as follows: ; S724. Repeat the training until the error function converges to the preset tolerance range; S73. Input the data into the trained model to generate the final bearing capacity data of the bored pile.

[0028] The intelligent output and feedback control processing based on the final bearing capacity data in S8 includes the following steps: S81. Import the final load-bearing capacity data into the report generation system, which includes the following steps: S811, Design report template, defining the chapter structure of data summary, calculation process, results analysis and safety recommendations; S812. Automatically fill the final bearing capacity data into the corresponding fields of the template, and format the numerical precision and unit; The data formatting formula is as follows: ; The formula for generating the chart is: ; S814. Export the report as a PDF and synchronize it to the cloud storage platform; S82. Automatically generate a bearing capacity calculation report, including a data summary, calculation process, result analysis, and safety recommendations; S83. Display the bearing capacity settlement curve and defect distribution map through a graphical interface.

[0029] The method also includes step S9, which involves real-time monitoring of changes in bearing capacity. S91. Deploy long-term monitoring sensors on the pile body to continuously collect load and deformation data; S92. Transmit data to the cloud analysis system in real time through the Internet of Things (IoT) platform, specifically including the following steps: S921: Configure the communication protocol for sensor nodes, supporting 4G, 5G and LoRa wireless transmission; S922. Design the data packet structure, including timestamp, sensor ID, load value, and deformation fields; S923, Upload real-time data to the cloud database via the MQTT protocol; S924: Set up a data verification mechanism in the cloud to discard abnormal and duplicate data packets; The data packet verification formula is as follows: ; The formula for detecting abnormal data is as follows: ; S93. Regularly update the bearing capacity model to achieve dynamic calibration; S94. When the monitoring data is abnormal, the early warning mechanism is automatically triggered.

[0030] The operational steps of this method for calculating the bearing capacity of bored piles are as follows: Step 1: Mechanism for Fusion and Collaborative Processing of Multi-Source Heterogeneous Data The starting point of the method is to establish a unified data fusion framework. Its principle is not to simply collect geometric, soil and load data in parallel, but to enable parameters of different dimensions to produce a synergistic effect on a unified analysis platform through data alignment and feature extraction. The soil parameter data around the pile is combined with the historical load data of the pile top to invert the real response of the soil under load. At the same time, the geometric parameter data of the pile body provides three-dimensional spatial constraints for this response. This fusion mechanism ensures that the physical meaning of subsequent analysis is clear and the data foundation is solid, reducing the model distortion caused by data silos.

[0031] Step 2: Intelligent diagnosis of pile defects based on frequency domain analysis and deep learning This method upgrades pile integrity detection from traditional time-domain waveform interpretation to a dual diagnostic mode of "frequency domain feature extraction + AI image recognition". The principle is as follows: First, the acoustic signal is converted from the time domain to the frequency domain using Fast Fourier Transform. Because different types of defects, such as cracks and cavities, will produce unique frequency response "fingerprints" for stress waves, that is, specific characteristic frequencies. Then, these frequency domain features and the original data are constructed into an "image" that can be analyzed and input into a pre-trained convolutional neural network. Through its multi-layer convolution and pooling structure, the convolutional neural network can automatically learn and identify complex defect patterns that are difficult for the human eye to detect, thereby generating high-precision pile defect type feature data. This principle greatly reduces the defect misjudgment rate.

[0032] Step 3: Dynamic matching and adaptive correction of bearing capacity influence coefficient The key innovation of this method lies in solving the inherent problem of the "one-size-fits-all" approach to the defect impact coefficient. Its principle is to construct a dynamic retrieval and matching engine. This engine compares the characteristic data of the pile defect type identified in the previous intelligent diagnosis with a massive number of cases in a pre-stored standard pile defect type database in real time. The matching process does not look for "completely identical" defects, but rather for defect cases with "equivalent mechanical effects". This results in the output of a dynamic pile bearing capacity impact coefficient that is highly adapted to the current actual condition of the pile. This allows the bearing capacity calculation to truly reflect the quantitative impact of specific defects.

[0033] Step 4: Real-time online correction of bearing capacity based on wave theory After obtaining the initial bearing capacity data, the method introduces a dynamic verification and correction mechanism based on wave theory. The principle is as follows: by applying an instantaneous impact load to the pile top and analyzing the propagation speed and energy attenuation of the stress wave in the pile body, the integrity of the pile body and the dynamic stiffness of the pile-soil system can be deduced. By solving the one-dimensional wave equation, these wave parameters can be quantified into a bearing capacity correction factor. This factor acts as a "real-time calibrator" to correct the initial static calculation results and generate corrected bearing capacity data for bored piles. This principle is equivalent to adding a "dynamic check-up" step to the bearing capacity calculation, enabling it to capture changes in pile body impedance that cannot be reflected by static calculation.

[0034] Step 5: AI-driven multi-model decision fusion and optimization To enhance the robustness of the final result, the method employs an artificial intelligence algorithm as the decision-making center. The principle is to construct a neural network optimizer, taking the outputs of the preceding steps, including corrected bearing capacity data, pile geometry parameters, and surrounding soil parameters, as input features. This network is trained with a large amount of historical successful case data, learning how to weigh the outputs of different calculation models and automatically correcting system biases. It does not simply accept the result of the previous step, but performs multi-model decision fusion, ultimately outputting a globally optimized and more reliable final bearing capacity data for bored piles. This essentially embeds the engineer's expert experience into the system in a data-driven manner.

[0035] Step Six: Full Lifecycle Data Management and Intelligent Feedback Control Principles This method views load-bearing capacity calculation as a continuous process throughout the project's lifecycle. Its principle is to establish a cloud-based data closed loop, where the final load-bearing capacity data and all process data are synchronized to the cloud platform. This not only automatically generates structured reports but, more importantly, provides an initial benchmark for subsequent long-term monitoring. By deploying long-term sensors, load and deformation data are continuously collected and transmitted back. By periodically comparing real-time data with the prediction model, dynamic calibration of the load-bearing capacity is achieved. When data anomalies occur, an early warning is automatically triggered, forming an intelligent feedback control loop of "calculation-monitoring-calibration-early warning," thereby achieving true preventative safety maintenance.

[0036] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0037] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for calculating the bearing capacity of a cast-in-place pile, characterized by: The method comprises the following steps: S1, collecting pile body geometric parameter data, pile soil parameter data and pile top load history data of the bored pile; S2, performing pile body integrity detection processing based on the pile body geometric parameter data, pile soil parameter data and pile top load history data, and generating pile body integrity detection data; S3, performing pile body defect type identification processing according to the pile body integrity detection data, and generating pile body defect type feature data; S4, calling a pre-stored standard pile body defect type database, performing pile body bearing capacity influence coefficient matching processing based on the pile body defect type feature data and the standard pile body defect type database, and generating pile body bearing capacity influence coefficient data; S5, performing initial bearing capacity calculation processing according to the pile body geometric parameter data, pile soil parameter data, pile top load history data and pile body bearing capacity influence coefficient data, and generating bored pile initial bearing capacity data; S6, performing bearing capacity correction processing based on the initial bearing capacity data in combination with pile body stress wave propagation characteristics, and generating bored pile corrected bearing capacity data; S7, performing optimization verification processing on the corrected bearing capacity data through an artificial intelligence algorithm, and generating bored pile final bearing capacity data; S8, performing intelligent output and feedback control processing based on the final bearing capacity data, including automatic generation and format optimization of the bearing capacity report, multi-platform synchronization and cloud storage of the bearing capacity data, real-time verification and abnormal early warning push of the calculation result.

2. The bored pile bearing capacity calculation method according to claim 1, characterized in that: The S1 of collecting the pile body geometric parameter data, the pile soil parameter data and the pile top load history data of the bored pile comprises the following steps: S11, performing multi-point sound wave velocity measurement along the length direction of the pile body through an ultrasonic detector, generating pile body geometric parameter data, and the pile body geometric parameter data comprising pile diameter, pile length and pile body cross section change rate; S12, collecting soil samples at different depths around the pile through a soil sampler, and performing laboratory soil test, generating pile soil parameter data, and the pile soil parameter data comprising soil internal friction angle, cohesion and compression modulus; S13, applying a phased load on the pile top through a load sensor and recording the settlement, generating pile top load history data, and the pile top load history data comprising load settlement curve and maximum test load.

3. The bored pile bearing capacity calculation method according to claim 2, characterized in that: The S2 of generating the pile body integrity detection data comprises the following steps: S21, importing the generated pile body geometric parameter data, pile soil parameter data and pile top load history data into a pile foundation supervision platform; S22, performing frequency analysis on the sound wave velocity measurement data through a fast Fourier transform algorithm, extracting pile body integrity feature frequency, and generating pile body integrity detection data; S23, removing environmental noise interference through a filtering algorithm, and optimizing data signal-to-noise ratio.

4. The bored pile bearing capacity calculation method according to claim 3, characterized in that: The S3 of generating the pile body defect type feature data comprises the following steps: S31, obtaining the pile body integrity detection data; S32, input the data into a pre-trained convolutional neural network model, perform pile body defect image recognition, generate pile body defect type feature data according to the recognition result, when a crack defect is recognized, output the defect type as a crack, when a hollow defect is recognized, output the defect type as a hollow, and when an exudation defect is recognized, output the defect type as an exudation.

5. The bored pile bearing capacity calculation method according to claim 4, characterized in that: The step S4 of generating the pile body bearing capacity influence coefficient data includes the following steps: S41, establish a standard pile body defect type database, wherein the database contains bearing capacity influence coefficients corresponding to various defect types; S42, perform similarity matching of the pile body defect type feature data with the database, calculate the matching degree by using the Euclidean distance algorithm, and generate the pile body bearing capacity influence coefficient data.

6. The bored pile bearing capacity calculation method according to claim 5, characterized in that: The step S5 of generating the initial bearing capacity data of the cast-in-place bored pile includes the following steps: S51, obtain the pile body geometric parameter data, the soil parameter data around the pile, the pile top load history data, and the pile body bearing capacity influence coefficient data; S52, calculate the pile side surface area and the pile bottom area based on the pile body geometric parameter data, and determine the soil bearing capacity parameter in combination with the soil parameter data around the pile; S53, analyze the load transfer law according to the pile top load history data, and calculate the pile side friction resistance component and the pile end resistance component; S54, introduce the pile body bearing capacity influence coefficient data to dynamically adjust the calculation parameters, and comprehensively calculate the initial bearing capacity; S55, verify the initial bearing capacity calculation result by using an iterative optimization algorithm, and generate the initial bearing capacity data of the cast-in-place bored pile.

7. The bored pile bearing capacity calculation method according to claim 6, characterized in that: The step S6 of generating the corrected bearing capacity data of the cast-in-place bored pile includes the following steps: S61, obtain the initial bearing capacity data; S62, apply an impact load on the pile top by using a stress wave sensor, measure the propagation time and the attenuation degree of the stress wave in the pile body, and generate stress wave propagation data; S63, based on the stress wave propagation data, use a wave equation correction model to correct the initial bearing capacity, the correction process is based on the wave velocity change and the attenuation coefficient, and generate the corrected bearing capacity data of the cast-in-place bored pile.

8. The bored pile bearing capacity calculation method according to claim 7, characterized in that: The step S7 of generating the final bearing capacity data of the cast-in-place bored pile includes the following steps: S71, construct a neural network model, the input is the corrected bearing capacity data, the pile body geometric parameter data, and the soil parameter data around the pile, and the output is the optimized bearing capacity; S72, train the neural network using the historical bearing capacity data as labels, and minimize the prediction error by using a back propagation algorithm; S73, input the data into the trained model, and generate the final bearing capacity data of the cast-in-place bored pile.

9. The bored pile bearing capacity calculation method according to claim 8, characterized in that: The step S8 of performing intelligent output and feedback control processing based on the final bearing capacity data includes the following steps: S81, import the final bearing capacity data into a report generation system; S82, automatically generate a bearing capacity calculation report, including data summary, calculation process, result analysis, and safety suggestions; S83, display the bearing capacity settlement curve and the defect distribution graph through a graphical interface.

10. The bored pile bearing capacity calculation method according to claim 1, characterized in that: The method further includes a step S9 of performing real-time monitoring of the bearing capacity change: S91, deploy a long-term monitoring sensor on the pile body, and continuously collect load and deformation data; S92, transmit the data to a cloud analysis system in real time through an Internet of Things platform; S93, periodically update the bearing capacity model to realize dynamic calibration; S94, when the monitoring data is abnormal, automatically trigger the early warning mechanism.

Images