A multilayer composite foundation bearing capacity detection device and a detection method thereof

Abstract

The application relates to the technical field of composite foundation bearing capacity, and discloses a multilayer composite foundation bearing capacity detection device and a detection method thereof. The device comprises a sensor array module, a data processing module, a bearing capacity analysis module, a calibration output module, a communication module and a user interface module. Static force parameters of the foundation are collected through a static sounding sensor, dynamic parameters of the foundation are collected through a dynamic sounding sensor, and deformation parameters of the foundation are collected through a fiber grating sensor, so that integrated collection of multi-source static and dynamic parameter data is realized. Meanwhile, a wavelet transform algorithm is used for noise filtering, and a K nearest neighbor interpolation algorithm is used for data filling, so that standardized foundation parameter data is generated, the consistency and accuracy of data preprocessing are ensured, and the soil layer is intelligently identified and hierarchically divided through K-means clustering algorithm-based foundation hierarchical division processing of the standardized foundation parameter data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of composite foundation bearing capacity, in particular to a multi-layer composite foundation bearing capacity detection device and a detection method thereof. BACKGROUND

[0002] The composite foundation refers to the natural foundation in which part of the soil is enhanced or replaced during the foundation treatment, or the reinforced material is arranged in the natural foundation, the reinforced area is the artificial foundation composed of the base body and the reinforcing body, under the action of the load, the base body and the reinforcing body jointly bear the load, the foundation bearing capacity is the bearing potential of the unit area of the foundation soil with the increase of the load, and is a comprehensive term for evaluating the stability of the foundation; the foundation bearing capacity is a practical professional term for the foundation design to facilitate the evaluation of the strength and stability of the foundation, and is not a basic property index of the soil.

[0003] At present, in the multi-layer composite foundation bearing capacity detection process, due to the obvious stratification and anisotropy of the foundation soil, the traditional detection method usually relies on a single type of sensor to collect limited parameters, and it is difficult to synchronously obtain the complete mechanical response under static and dynamic load.

[0004] Therefore, the present application provides a multi-layer composite foundation bearing capacity detection device and a detection method thereof to solve the above problems. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a multi-layer composite foundation bearing capacity detection device and a detection method thereof, which solves the problem that the traditional detection method usually relies on a single type of sensor to collect limited parameters, and it is difficult to synchronously obtain the complete mechanical response under static and dynamic load.

[0006] To achieve the above purpose, the present application provides the following technical solutions: a multi-layer composite foundation bearing capacity detection device and a detection method thereof, the method comprising the following steps:

[0007] S1, collecting static and dynamic parameter data of the multi-layer composite foundation, including vertical displacement parameters, horizontal stress parameters and soil layer compression modulus parameters of the foundation;

[0008] S2, data preprocessing of the foundation parameters based on the static and dynamic parameter data, to generate standardized foundation parameter data;

[0009] S3, foundation stratification processing according to the standardized foundation parameter data, to generate foundation stratification data;

[0010] S4, preliminary calculation of the bearing capacity of each layer of foundation based on the foundation stratification data, to generate preliminary data of the bearing capacity of each layer of foundation;

[0011] S5, carrying capacity fusion analysis and processing is carried out according to the preliminary data of the bearing capacity of each layer of foundation, and the bearing capacity fusion data of the multi-layer composite foundation is generated;

[0012] S6, bearing capacity algorithm matching processing is carried out based on the bearing capacity fusion data of the multi-layer composite foundation and the standard foundation bearing capacity database, and the target foundation bearing capacity algorithm type data is generated;

[0013] S7, the foundation bearing capacity calibration summary data is constructed, and the final calibration processing of the foundation bearing capacity is carried out, and the final data of the bearing capacity of the multi-layer composite foundation is generated.

[0014] Preferably, the static and dynamic parameter data of the multi-layer composite foundation in S1 comprises the following steps:

[0015] S11, the static parameters of each layer of soil in the state of no load of the foundation are collected by the static sounding sensor, including the cone tip resistance parameter and the side friction resistance parameter, and the static parameter data of the foundation is generated;

[0016] S12, the dynamic parameters of each layer of soil in the state of dynamic load of the foundation are collected by the dynamic sounding sensor, including the number of blows parameter and the wave velocity parameter, and the dynamic parameter data of the foundation is generated;

[0017] S13, the deformation parameters of the foundation are collected by the optical fiber grating sensor, including the strain distribution and displacement change of each layer of soil, and the deformation parameter data of the foundation is generated;

[0018] S14, the static parameter data of the foundation, the dynamic parameter data of the foundation and the deformation parameter data of the foundation are integrated into the static and dynamic parameter data of the multi-layer composite foundation.

[0019] Preferably, the data preprocessing of the foundation parameters in S2 comprises the following steps:

[0020] S21, the static and dynamic parameter data is obtained, and the wavelet transform algorithm is used for noise filtering processing to remove high-frequency interference signals, and the filtered foundation parameter data is generated;

[0021] S22, the filtered foundation parameter data is normalized, and the parameter values of different sensors are scaled to a unified dimension, and the normalized foundation parameter data is generated;

[0022] S23, the missing values in the normalized foundation parameter data are detected, and the K nearest neighbor interpolation algorithm is used for data filling, and the standardized foundation parameter data is generated.

[0023] Preferably, the foundation level division processing in S3 comprises the following steps:

[0024] S31, based on the standardized foundation parameter data, extracting the depth characteristics and soil characteristics of each soil layer, including soil layer thickness, soil density and water content;

[0025] S32, using K-means clustering algorithm to analyze the depth characteristics and soil characteristics, dividing the foundation into multiple levels, and generating foundation level division data;

[0026] S33, giving each level a unique identifier and storing level attribute information.

[0027] Preferably, the S4 preliminary calculation of the bearing capacity of each layer of foundation includes the following steps:

[0028] S41, according to the foundation level division data, selecting the corresponding bearing capacity calculation model for each level;

[0029] S42, for shallow soil, using the Terzaghi bearing capacity formula for calculation;

[0030] S43, for deep soil, using the modified lateral pressure meter model for calculation;

[0031] S44, integrating the calculation results of each level to generate preliminary data of the bearing capacity of each layer of foundation.

[0032] Preferably, the S5 foundation bearing capacity fusion analysis processing includes the following steps:

[0033] S51, obtaining the preliminary data of the bearing capacity of each layer of foundation, and using entropy weight method to determine the weight coefficient of each level;

[0034] S52, based on the weight coefficient, generating preliminary fusion data through linear weighted fusion algorithm;

[0035] S53, inputting the preliminary fusion data into the pre-trained BP neural network model for nonlinear correction to generate multi-layer composite foundation bearing capacity fusion data.

[0036] Preferably, the S6 bearing capacity algorithm matching processing includes the following steps:

[0037] S61, establishing a standard foundation bearing capacity database containing standard parameter combinations corresponding to various bearing capacity algorithms, including Prandtl algorithm, Hansen algorithm and Vesic algorithm;

[0038] S62, matching the multi-layer composite foundation bearing capacity fusion data with the standard parameter combination using cosine similarity algorithm to calculate the matching degree;

[0039] S63, based on the matching degree result, selecting the optimal algorithm type to generate target foundation bearing capacity algorithm type data.

[0040] Preferably, the final calibration process of the foundation bearing capacity in S7 comprises the following steps:

[0041] S71, combining the multi-layer composite foundation bearing capacity fusion data and the target foundation bearing capacity algorithm type data into foundation bearing capacity calibration summary data;

[0042] S72, calling the corresponding calibration program according to the target algorithm type to perform final calibration on the foundation bearing capacity data, including deviation correction and confidence interval calculation;

[0043] S73, outputting the final data of the multi-layer composite foundation bearing capacity and generating a detection report.

[0044] Preferably, the device comprises a sensor array module, a data processing module, a bearing capacity analysis module, and a calibration output module;

[0045] The sensor array module uses a static touch unit to collect foundation static parameters, acquires foundation dynamic response data through a dynamic touch unit, and outputs foundation deformation parameters from a deformation monitoring unit;

[0046] The data processing module receives the foundation static parameters, foundation dynamic response data, and foundation deformation parameters, performs noise filtering and normalization processing through a data preprocessing unit, performs intelligent identification of foundation levels using a hierarchical division unit, and outputs standardized foundation parameter data and foundation level division data;

[0047] The bearing capacity analysis module receives the standardized foundation parameter data and foundation level division data, performs independent calculation of the bearing capacity of each layer through a preliminary calculation unit, performs weighted and nonlinear correction of multi-source data using a fusion analysis unit, and outputs multi-layer composite foundation bearing capacity fusion data;

[0048] The calibration output module receives the multi-layer composite foundation bearing capacity fusion data, calls the optimal calculation model from the standard database through an algorithm matching unit, performs deviation correction and confidence evaluation using a calibration execution unit, and generates and outputs the final data of the multi-layer composite foundation bearing capacity and a detection report.

[0049] Preferably, the device further comprises a communication module and a user interface module;

[0050] The communication module sends the final bearing capacity data and the detection report to a remote monitoring platform through a wireless transmission unit;

[0051] The user interface module receives the final bearing capacity data and the detection report, generates a bearing capacity change curve through a data visualization unit, prompts an abnormal state using an alarm information generation unit, and receives user instructions through a human-computer interaction unit.

[0052] Compared with the prior art, the multi-layer composite foundation bearing capacity detection device and the detection method thereof have the following beneficial effects:

[0053] 1. In the present application, the static parameters of the foundation are collected by the static sounding sensor, the dynamic parameters of the foundation are collected by the dynamic sounding sensor, and the deformation parameters of the foundation are collected by the fiber grating sensor, realizing integrated collection of multi-source static and dynamic parameter data, and at the same time, the wavelet transform algorithm is used for noise filtering and the K nearest neighbor interpolation algorithm is used for data filling, to generate standardized foundation parameter data, ensuring the consistency and accuracy of data preprocessing.

[0054] 2. In the present application, the K-means clustering algorithm is used to divide the standardized foundation parameter data into foundation levels, realizing intelligent identification and hierarchical division of soil layers, and at the same time, the shallow soil body is calculated by using the Terzaghi bearing capacity formula and the deep soil body is calculated by using the modified lateral pressure meter model, to ensure independent and accurate evaluation of the bearing capacity of each layer, and to adapt to the complex soil characteristics of the multi-layer composite foundation.

[0055] 3. In the present application, the entropy weight method is used to determine the weight coefficients of each level of bearing capacity and combine the BP neural network model for nonlinear correction, to realize weighted fusion and fusion analysis of multi-source data, and at the same time, the cosine similarity algorithm is used to match the fusion data with the standard foundation bearing capacity database, to generate target foundation bearing capacity algorithm type data, ensuring the comprehensive accuracy and adaptability of the bearing capacity calculation. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 The flowchart of the multi-layer composite foundation bearing capacity detection method of the present application;

[0057] Figure 2 The framework diagram of the multi-layer composite foundation bearing capacity detection device of the present application. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0059] Specific embodiments: a multi-layer composite foundation bearing capacity detection device and a detection method thereof, the method comprising the following steps:

[0060] S1, collecting static and dynamic parameter data of the multi-layer composite foundation, including vertical displacement parameters, horizontal stress parameters and soil layer compression modulus parameters of the foundation;

[0061] S2, data preprocessing of foundation parameters based on static and dynamic parameter data, generating standardized foundation parameter data;

[0062] S3, foundation level division processing according to standardized foundation parameter data, generating foundation level division data;

[0063] S4, preliminary calculation of bearing capacity of each layer of foundation based on foundation level division data, generating preliminary data of bearing capacity of each layer of foundation;

[0064] S5, bearing capacity fusion analysis processing of foundation according to preliminary data of bearing capacity of each layer of foundation, generating multi-layer composite foundation bearing capacity fusion data;

[0065] S6, bearing capacity algorithm matching processing based on multi-layer composite foundation bearing capacity fusion data and standard foundation bearing capacity database, generating target foundation bearing capacity algorithm type data;

[0066] S7, building foundation bearing capacity calibration summary data, carrying out final calibration of foundation bearing capacity, generating multi-layer composite foundation bearing capacity final data.

[0067] S1 in the collection of static and dynamic parameter data of multi-layer composite foundation, including the following steps:

[0068] S11, collecting static parameters of each layer of soil in the state of foundation without load by static cone penetration sensor, including cone tip resistance parameter and side friction resistance parameter, and generating foundation static parameter data;

[0069] S12, collecting dynamic parameters of each layer of soil in the state of foundation under dynamic load by dynamic penetration sensor, including blow count parameter and wave velocity parameter, and generating foundation dynamic parameter data;

[0070] S13, collecting deformation parameters of foundation by fiber grating sensor, including strain distribution and displacement change of each layer of soil, and generating foundation deformation parameter data;

[0071] S14, integrating foundation static parameter data, foundation dynamic parameter data and foundation deformation parameter data into static and dynamic parameter data of multi-layer composite foundation.

[0072] S2 in the data preprocessing of foundation parameters, including the following steps:

[0073] S21, obtaining static and dynamic parameter data, using wavelet transform algorithm for noise filtering processing, removing high frequency interference signal, generating filtered foundation parameter data;

[0074] In the implementation of the wavelet transform algorithm, firstly, a suitable wavelet basis function Daubechies 5 wavelet is selected, and the original signal containing noise is decomposed into different frequency subbands; then, a threshold processing function is applied to the wavelet coefficients obtained by decomposition, the coefficients lower than the set threshold are regarded as noise and are set to zero or attenuated, and the coefficients higher than the threshold representing effective signal characteristics are retained; finally, the processed wavelet coefficients are used for signal reconstruction, thereby generating filtered foundation parameter data; this process can eliminate high-frequency random interference while retaining the mutation characteristics of the signal, thereby providing a high-quality data basis for subsequent analysis;

[0075] S22, normalizing the filtered foundation parameter data to scale the parameter values of different sensors to a unified dimension, and generating normalized foundation parameter data;

[0076] In the implementation, the minimum-maximum normalization method is used to linearly transform each parameter data to the interval [0, 1], and the calculation formula is as follows:

[0077] ;

[0078] wherein, represents the original value of a certain parameter, and respectively represent the maximum value and the minimum value of the parameter in the data set, is the normalized result; this step ensures that all parameters are in the same scale, facilitating subsequent fusion calculation and comparative analysis;

[0079] S23, detecting missing values in the normalized foundation parameter data, using the K-nearest neighbor interpolation algorithm to fill in the data, and generating standardized foundation parameter data;

[0080] In the implementation, firstly, the sample in which the missing data is located is determined, then the Euclidean distance between the sample and other samples in the complete sample space is calculated, the K nearest complete samples are selected, and K is usually 5; finally, the mean value of the values of the K nearest neighbor samples in the missing attribute is taken as the estimated value of the missing value; this method reasonably estimates according to the similarity between data samples, and maintains the inherent distribution characteristics of the data set.

[0081] The foundation level division process in S3 includes the following steps:

[0082] S31, based on the standardized foundation parameter data, extracting the depth characteristics and soil characteristics of each soil layer, including soil layer thickness, soil density and water content;

[0083] S32, using the K-means clustering algorithm to perform clustering analysis on the depth characteristics and soil characteristics, dividing the foundation into multiple levels, and generating foundation level division data;

[0084] In implementation, firstly, the number of clusters K is determined according to the prior knowledge of the engineering geological survey report; then, the algorithm randomly initializes K cluster centers, and then iteratively executes the following two steps until the center point is stable:

[0085] Assignment step: assign the soil parameters at different depths represented by each data point to the nearest cluster center;

[0086] Update step: recalculate the mean of all data points in each cluster as the new cluster center;

[0087] After iterative convergence, the continuous depth section with the same class label is identified as an independent soil layer, thereby generating the stratum hierarchy division data;

[0088] S33, assign a unique identifier to each hierarchy and store the hierarchy attribute information;

[0089] In implementation, the identifier can use a structured coding method, "L<layer number><depth range><main soil type>", where "layer number" is numbered in the order from top to bottom, "depth range" records the burial depth of the top plate and bottom plate of the layer, and "main soil type" is determined according to the parameter characteristics of the cluster center; this unique identifier facilitates fast retrieval, positioning and management in the database, and provides clear indexing for subsequent layered bearing capacity calculation.

[0090] The preliminary calculation and processing of the bearing capacity of each layer in S4 include the following steps:

[0091] S41, according to the stratum hierarchy division data, select the corresponding bearing capacity calculation model for each hierarchy;

[0092] In implementation, a model rule library is established: for shallow soil layers, the Terzaghi bearing capacity formula is preferred; for deep soil layers and pile tip soil layers, the modified side pressure meter model considering depth effect and soil compressibility is selected; according to the depth information and soil type in the soil layer identifier, the corresponding calculation model is automatically matched and called to realize the adaptation of the calculation method;

[0093] S42, for shallow soil, the Terzaghi bearing capacity formula is used for calculation:

[0094] ;

[0095] Wherein is the ultimate bearing capacity of the foundation, is the cohesion of the soil, is the effective pressure of the overburden at the base, is the effective unit weight of the soil, is the width of the foundation, , , The coefficient of Terzaghi bearing capacity;

[0096] S43, for deep soil, the modified lateral pressure test model is used for calculation, the core idea is to convert the ultimate pressure obtained by lateral pressure test into foundation bearing capacity, and introduce depth correction coefficient, the formula can be expressed as:

[0097] ;

[0098] Wherein is the ultimate bearing capacity of foundation, is the ultimate pressure derived from the lateral pressure test data, is the depth correction coefficient;

[0099] S44, the calculation results of each level are integrated to generate the preliminary data of foundation bearing capacity of each layer.

[0100] The fusion analysis and processing of foundation bearing capacity in S5 include the following steps:

[0101] S51, the preliminary data of foundation bearing capacity of each layer is obtained, and the weight coefficient of each level bearing capacity is determined by using entropy weight method:

[0102] Index proportion calculation:

[0103] ;

[0104] Wherein is the proportion of the ith sample under the jth index, is the original value of the jth index of the ith sample, is the total number of samples;

[0105] Information entropy calculation:

[0106] ;

[0107] Wherein is the information entropy of the jth index and the value range is [0, 1], is the total number of samples, is the adjustment coefficient, k = 1 / ln(m);

[0108] Weight calculation:

[0109] ;

[0110] ;

[0111] Wherein is the maximum weight of the jth index, is the difference coefficient of the jth index, is the total number of indexes;

[0112] The smaller the entropy value, the greater the difference coefficient, indicating that the information provided by the index is greater, and the weight is also higher;

[0113] S52, generating preliminary fusion data based on the weight coefficient through a linear weighted fusion algorithm;

[0114] For each calculation point i, the preliminary fusion bearing capacity value of the multi-layer composite foundation is obtained by linearly weighting the bearing capacity value of each soil layer at the point and its weight, and the calculation formula is:

[0115] ;

[0116] Wherein is the preliminary fusion bearing capacity value of the i-th calculation point, is the preliminary calculation value of the bearing capacity of the i-th calculation point on the j-th layer of soil, is the total number of characteristic parameters;

[0117] S53, inputting the preliminary fusion data into the pre-trained BP neural network model for nonlinear correction to generate multi-layer composite foundation bearing capacity fusion data;

[0118] In specific implementation, the network is a three-layer structure: the input layer is the preliminary fusion bearing capacity value and related influence parameters, the hidden layer contains a number of neurons and uses the Sigmoid activation function, and the output layer is the final bearing capacity value after correction; The network is trained through historical data, with the goal of minimizing the prediction error, learning the complex nonlinear mapping relationship from the preliminary fusion value to the true bearing capacity, thereby generating multi-layer composite foundation bearing capacity fusion data.

[0119] The bearing capacity algorithm matching process in S6 includes the following steps:

[0120] S61, establishing a standard foundation bearing capacity database containing a variety of bearing capacity algorithm corresponding standard parameter combination, the algorithm has Prandtl algorithm, Hansen algorithm and Vesic algorithm;

[0121] The database is a relational database, and the core data table fields include but are not limited to:

[0122] algorithm ID, algorithm name, applicable soil layer type, cohesion c reference value range, internal friction angle φ reference value range, foundation width B influence coefficient, overload q influence coefficient, algorithm formula;

[0123] Parameter combination example: "Plattner algorithm" applied to "saturated soft clay", a typical record in the database may contain: algorithm name = "Plattner algorithm", applicable soil type = "clay", cohesion c reference value range = "10-25 kPa", internal friction angle φ reference value range = "0-5°", and indicates that under this condition, the bearing capacity calculation mainly depends on the cohesion c item;

[0124] This database provides a benchmark data set for subsequent cosine similarity matching by pre-computing and storing the results of standard algorithms under different geological conditions;

[0125] S62, similarity matching of multi-layer composite foundation bearing capacity fusion data and standard parameter combination, using cosine similarity algorithm to calculate the matching degree:

[0126] ;

[0127] Where is the cosine similarity value and ranges between [-1, 1], is the feature vector of the foundation data to be tested, is the feature vector of a case in the standard case database, is the dimension of the feature vector, is the summation index, , is the component value of vector and vector in the i-th dimension;

[0128] The closer the similarity value is to 1, the more similar the two features are;

[0129] S63, based on the matching degree result, select the optimal algorithm type, and generate target foundation bearing capacity algorithm type data.

[0130] The final calibration process of the foundation bearing capacity in S7 includes the following steps:

[0131] S71, combine the multi-layer composite foundation bearing capacity fusion data and the target foundation bearing capacity algorithm type data into foundation bearing capacity calibration summary data;

[0132] S72, according to the target algorithm type, call the corresponding calibration program to perform final calibration on the foundation bearing capacity data, including bias correction and confidence interval calculation;

[0133] In specific implementation, according to the target algorithm type, the error model of its historical application data is called to perform systematic bias compensation on the calculation result; then, the confidence interval is calculated; based on the statistical distribution characteristics of the error, which is usually assumed to be normally distributed, the confidence interval of the bearing capacity is calculated under a given confidence level:

[0134] ;

[0135] ;

[0136] wherein is a confidence interval, is a bearing capacity calibration value after bias correction, is a quantile of standard normal distribution, is an estimated standard deviation, is a sample size, is a standard error;

[0137] S73, output the final data of the multi-layer composite foundation bearing capacity, and generate a detection report.

[0138] The device comprises a sensor array module, a data processing module, a bearing capacity analysis module, and a calibration output module;

[0139] The sensor array module collects static force parameters of the foundation by using a static force touch unit, acquires dynamic response data of the foundation by using a dynamic force touch unit, and outputs deformation parameters of the foundation by using a deformation monitoring unit;

[0140] The data processing module receives the static force parameters of the foundation, the dynamic response data of the foundation, and the deformation parameters of the foundation, performs noise filtering and normalization processing by using a data preprocessing unit, performs intelligent identification of foundation levels by using a hierarchical division unit, and outputs standardized foundation parameter data and foundation level division data;

[0141] The bearing capacity analysis module receives the standardized foundation parameter data and the foundation level division data, performs independent operation of bearing capacity of each layer by using a preliminary calculation unit, performs weighted and nonlinear correction of multi-source data by using a fusion analysis unit, and outputs fusion data of the multi-layer composite foundation bearing capacity;

[0142] The calibration output module receives the fusion data of the multi-layer composite foundation bearing capacity, calls an optimal calculation model from a standard database by using an algorithm matching unit, performs bias correction and confidence evaluation by using a calibration execution unit, and generates and outputs final data of the multi-layer composite foundation bearing capacity and a detection report.

[0143] The device further comprises a communication module and a user interface module;

[0144] The communication module sends the final data of the bearing capacity and the detection report to a remote monitoring platform by using a wireless transmission unit;

[0145] The user interface module receives the final data of the bearing capacity and the detection report, generates a bearing capacity change curve by using a data visualization unit, prompts an abnormal state by using an alarm information generation unit, and receives user instructions by using a human-computer interaction unit.

[0146] The operating steps of this device and method are as follows:

[0147] The detection method begins with the acquisition of static and dynamic parameter data. Static parameters of each soil layer under unloaded conditions are collected using static cone penetration sensors, including cone tip resistance and side skin friction parameters, generating static foundation parameter data. Simultaneously, dynamic cone penetration sensors collect dynamic parameters of each soil layer under dynamic load conditions, generating dynamic foundation parameter data. Fiber optic sensors collect foundation deformation parameters, including strain distribution and displacement changes in each soil layer, generating foundation deformation parameter data. These data are integrated into static and dynamic parameter data for the multi-layered composite foundation, providing a foundation for subsequent processing.

[0148] Next, the foundation parameter data is preprocessed to generate standardized foundation parameter data. First, wavelet transform algorithm is used to filter noise from the static and dynamic parameter data, removing high-frequency interference signals to generate filtered foundation parameter data. Then, the filtered data is normalized to scale the parameter values ​​from different sensors to a uniform dimension, generating normalized foundation parameter data. Finally, missing values ​​in the normalized data are detected and imputed using the K-nearest neighbor interpolation algorithm to ensure data integrity and consistency.

[0149] Based on standardized foundation parameter data, foundation layer division processing is performed to generate foundation layer division data. Depth and soil characteristics, including soil layer thickness, density, and water content, are extracted from the standardized data for each soil layer. K-means clustering is then used to perform cluster analysis on these characteristics, dividing the foundation into multiple layers. Each layer is assigned a unique identifier, and layer attribute information is stored to achieve intelligent identification of soil layers.

[0150] Based on the foundation layer division data, preliminary calculations of the bearing capacity of each layer are performed to generate preliminary bearing capacity data for each layer. A corresponding bearing capacity calculation model is selected for each layer: for shallow soil, the Terzaghi bearing capacity formula is used; for deep soil, a modified pressuremeter model is used. Finally, the calculation results from each layer are integrated to ensure independent assessment of the bearing capacity of each layer.

[0151] Based on the preliminary bearing capacity data of each foundation layer, a fusion analysis of the bearing capacity is performed to generate fusion data of the bearing capacity of multi-layer composite foundations. The entropy weight method is used to determine the weight coefficients of the bearing capacity at each layer; based on these weight coefficients, a linear weighted fusion algorithm is used to generate preliminary fusion data; finally, the preliminary fusion data is input into a pre-trained BP neural network model for nonlinear correction to improve the fusion accuracy.

[0152] Based on the multi-layer composite foundation bearing capacity fusion data, the bearing capacity algorithm matching processing is performed with the standard foundation bearing capacity database to generate target foundation bearing capacity algorithm type data. The standard database contains multiple algorithms: Prandtl algorithm, Hansen algorithm and Vesic algorithm corresponding standard parameter combination; cosine similarity algorithm is used to calculate the matching degree of fusion data and standard parameters; the optimal algorithm type is selected according to the matching degree result.

[0153] Finally, the foundation bearing capacity calibration summary data is constructed, and the final calibration processing of the foundation bearing capacity is performed to generate the final data of the multi-layer composite foundation bearing capacity. The fusion data and the target algorithm type data are combined; the corresponding calibration program is called according to the target algorithm type to perform bias correction and confidence interval calculation; the final data is output and a detection report is generated.

[0154] The detection device is based on modular design, including sensor array module, data processing module, bearing capacity analysis module and calibration output module. The sensor array module uses static touch unit to collect foundation static parameters, dynamic touch unit to obtain foundation dynamic response data, and deformation monitoring unit to output foundation deformation parameters. The data processing module receives these parameters, performs noise filtering and normalization processing through the data preprocessing unit, performs intelligent identification of foundation levels by using the hierarchical division unit, and outputs standardized foundation parameter data and foundation level division data. The bearing capacity analysis module receives the above data, performs independent operation of each layer bearing capacity through the preliminary calculation unit, performs multi-source data weighting and nonlinear correction by using the fusion analysis unit, and outputs the multi-layer composite foundation bearing capacity fusion data. The calibration output module receives the fusion data, calls the optimal calculation model from the standard database through the algorithm matching unit, performs bias correction and confidence evaluation by using the calibration execution unit, and generates and outputs the final data and the detection report. In addition, the device also contains communication module and user interface module: the communication module sends data to the remote monitoring platform through the wireless transmission unit; the user interface module generates bearing capacity change curve through the data visualization unit, prompts abnormal state through the alarm information generation unit, and receives user instructions through the man-machine interaction unit. The whole device works cooperatively to realize efficient and accurate foundation bearing capacity detection.

[0155] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting; it is not intended to exclude myriad other embodiments of the present application that other inventors can develop based on the same general inventive concepts embodied by the described embodiments. That is, although the present application is described in terms of particular embodiments and implementations, it is to be understood that the terminology used is for the purpose of descriptive clarity and that it should be taken in a descriptive sense and not a limiting sense.

[0156] While the embodiments of the application have been shown and described herein, it is to be understood that the application is not limited to these embodiments. Rather, many modifications, changes, substitutions, and alterations can be made to the embodiments of the application without departing from the spirit and scope of the application as defined by the appended claims and their equivalents.

Claims

1. A method for detecting the bearing capacity of a multi-layer composite foundation, characterized in that: The method comprises the following steps: S1, collecting static and dynamic parameter data of the multi-layer composite foundation, including vertical displacement parameters, horizontal stress parameters and soil layer compression modulus parameters of the foundation; S2, data preprocessing of foundation parameters based on the static and dynamic parameter data, to generate standardized foundation parameter data; S3, foundation level division processing according to the standardized foundation parameter data, including extracting depth characteristics and soil characteristics of each soil layer, and using K-means clustering algorithm to perform clustering analysis on the depth characteristics and soil characteristics, dividing the foundation into multiple levels to generate foundation level division data; S4, preliminary calculation processing of the bearing capacity of each layer of foundation based on the foundation level division data, including selecting a corresponding bearing capacity calculation model for each level, using the Terzaghi bearing capacity formula for calculation for shallow soil, and using the modified lateral pressure instrument model for calculation for deep soil, and integrating the calculation results of each level to generate preliminary bearing capacity data of each layer of foundation; S5, fusion analysis processing of the bearing capacity of the foundation according to the preliminary bearing capacity data of each layer of foundation, including determining the weight coefficient of each level of bearing capacity by using the entropy weight method, generating preliminary fusion data based on the weight coefficient by using the linear weighted fusion algorithm, and inputting the preliminary fusion data into the pre-trained BP neural network model for nonlinear correction to generate multi-layer composite foundation bearing capacity fusion data; S6, bearing capacity algorithm matching processing based on the multi-layer composite foundation bearing capacity fusion data and the standard foundation bearing capacity database to generate target foundation bearing capacity algorithm type data; S7, constructing foundation bearing capacity calibration summary data to perform final calibration processing of the foundation bearing capacity to generate final data of the bearing capacity of the multi-layer composite foundation.

2. The method for detecting the bearing capacity of a multi-layer composite foundation according to claim 1, characterized in that: The static and dynamic parameter data of the multi-layer composite foundation in S1 comprises the following steps: S11, collecting static parameters of each layer of soil in the unloaded state of the foundation by a static cone penetration sensor, including cone tip resistance parameters and side friction resistance parameters, and generating foundation static parameter data; S12, collecting dynamic parameters of each layer of soil in the dynamic load state of the foundation by a dynamic cone penetration sensor, including blow count parameters and wave velocity parameters, and generating foundation dynamic parameter data; S13, collecting deformation parameters of the foundation by a fiber grating sensor, including strain distribution and displacement change of each layer of soil, and generating foundation deformation parameter data; S14, integrating the foundation static parameter data, the foundation dynamic parameter data and the foundation deformation parameter data into the static and dynamic parameter data of the multi-layer composite foundation.

3. The method of claim 1, wherein: The data preprocessing of the foundation parameters in S2 comprises the following steps: S21, obtaining static and dynamic parameter data, and performing noise filtering processing by using a wavelet transform algorithm to remove high-frequency interference signals to generate filtered foundation parameter data; S22, performing normalization processing on the filtered foundation parameter data to scale the parameter values of different sensors to a unified dimension to generate normalized foundation parameter data; S23, detecting missing values in the normalized foundation parameter data, and filling the data by using a K-nearest neighbor interpolation algorithm to generate standardized foundation parameter data.

4. The method for detecting the bearing capacity of a multi-layer composite foundation according to claim 1, characterized in that: The bearing capacity algorithm matching processing in S6 comprises the following steps: S61, a standard foundation bearing capacity database is established, including a plurality of bearing capacity algorithm corresponding standard parameter combination, the algorithm has prandtl algorithm, hansen algorithm and visic algorithm; S62, the multi-layer composite foundation bearing capacity fusion data is matched with the standard parameter combination, and the matching degree is calculated by using a cosine similarity algorithm; S63, based on the matching degree result, the optimal algorithm type is selected, and target foundation bearing capacity algorithm type data is generated.

5. The method for detecting the bearing capacity of a multi-layer composite foundation according to claim 1, characterized in that: The final calibration processing of the foundation bearing capacity in the S7 includes the following steps: S71, the multi-layer composite foundation bearing capacity fusion data and the target foundation bearing capacity algorithm type data are combined into foundation bearing capacity calibration summary data; S72, according to the target algorithm type, the corresponding calibration program is called, and the final calibration of the foundation bearing capacity data is carried out, including deviation correction and confidence interval calculation; S73, the final data of the multi-layer composite foundation bearing capacity is output, and a detection report is generated.

6. A device for detecting the bearing capacity of a multi-layer composite foundation, for implementing the method for detecting the bearing capacity of a multi-layer composite foundation according to any one of claims 1-5, characterized in that: The device includes a sensor array module, a data processing module, a bearing capacity analysis module and a calibration output module; The sensor array module collects the foundation static parameters by using a static touch element, obtains the foundation dynamic response data through a dynamic touch element, and outputs the foundation deformation parameters by a deformation monitoring element; The data processing module receives the foundation static parameters, the foundation dynamic response data and the foundation deformation parameters, carries out noise filtering and normalization processing through a data preprocessing unit, carries out intelligent identification of foundation levels by using a hierarchical division unit, and outputs standardized foundation parameter data and foundation level division data; The bearing capacity analysis module receives the standardized foundation parameter data and the foundation level division data, carries out independent operation of each layer bearing capacity through a preliminary calculation unit, carries out weighted and nonlinear correction of multi-source data by using a fusion analysis unit, and outputs multi-layer composite foundation bearing capacity fusion data; The calibration output module receives the multi-layer composite foundation bearing capacity fusion data, calls the optimal calculation model from the standard database through an algorithm matching unit, carries out deviation correction and confidence evaluation by using a calibration execution unit, and generates and outputs the final data of the multi-layer composite foundation bearing capacity and the detection report.

7. The device for detecting the bearing capacity of a multi-layer composite foundation according to claim 6, characterized in that: The device further includes a communication module and a user interface module; The communication module sends the final data of the bearing capacity and the detection report to a remote monitoring platform through a wireless transmission unit; The user interface module receives the final data of the bearing capacity and the detection report, generates a bearing capacity change curve by using a data visualization unit, prompts an abnormal state by using an alarm information generation unit, and receives user instructions by using a man-machine interaction unit.

Images