Foundation quality detection method and device, electronic equipment and storage medium

Abstract

This application provides a method, apparatus, electronic device, and storage medium for foundation quality testing, belonging to the field of foundation testing technology. The method includes: obtaining standardized sensing data based on foundation dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data; extracting features from the standardized sensing data to obtain single-mode feature data; determining multi-mode feature fusion weights based on the single-mode feature data; fusing the single-mode feature data with the multi-mode feature fusion weights to obtain multi-mode fused feature data; obtaining target defect identification data based on the multi-mode fused feature data; mapping and updating the foundation digital twin basic model data based on the target defect identification data to obtain foundation digital twin model data with defect attributes; and determining the foundation quality testing results based on the foundation digital twin model data with defect attributes, superstructure load data, and stratum parameter data. This application can meet the high-precision requirements of foundation quality testing.

Description

Technical Field

[0001] This application belongs to the field of foundation testing technology, and more specifically, relates to a foundation quality testing method and device, electronic equipment, and storage medium. Background Technology

[0002] Foundation quality testing is a crucial aspect of geotechnical engineering and building structural safety, as its accuracy and reliability directly impact the overall stability and operational safety of buildings. Related technologies typically rely on foundation physical response information, which, after signal processing and defect identification, is combined with engineering parameters for quality assessment. This has become a standard technical approach for foundation construction acceptance and hazard identification. As building engineering progresses towards complex sites, deep foundations, and large structures, higher demands are placed on the spatialization, quantification, and comprehensiveness of testing results.

[0003] Existing foundation quality testing technologies mostly rely on single or discrete test data for analysis, identifying defects through experience-based interpretation or simple statistical methods. The defect identification results are mostly presented in the form of discrete data. When conducting quality evaluation based on this, the influence of defects on the foundation bearing capacity and deformation characteristics is not fully described, resulting in deviations between the quality evaluation results and the actual engineering conditions, making it difficult to meet the high-precision testing requirements of complex sites. Summary of the Invention

[0004] This application provides a foundation quality testing method, apparatus, electronic device, and storage medium to meet the high precision requirements of foundation quality testing.

[0005] According to one aspect of the embodiments of this application, a method for testing the quality of a foundation is provided, comprising: Acquire dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation, and obtain standardized sensing data based on the dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation; Feature extraction is performed on standardized sensing data to obtain multiple single-modal feature data. Multimodal feature fusion weights are determined based on the multiple single-modal feature data. The single-modal feature data and the multimodal feature fusion weights are then fused to obtain multimodal fused feature data. Target defect identification data is obtained based on multimodal fusion feature data; the foundation digital twin basic model data is mapped and updated based on the target defect identification data to obtain foundation digital twin model data with defect attributes; the foundation digital twin basic model data is constructed based on the foundation spatial structure and stratigraphic distribution data. The foundation quality inspection results are determined based on the foundation digital twin model data with defect attributes, the superstructure load data, and the stratum parameter data.

[0006] According to one aspect of the embodiments of this application, a foundation quality testing device is provided, comprising: The foundation data acquisition module is used to acquire foundation dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data, and to obtain standardized sensing data based on the foundation dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data; The multimodal feature fusion module is used to extract features from standardized perceptual data to obtain multiple single-modal feature data, determine multimodal feature fusion weights based on the multiple single-modal feature data, and fuse the single-modal feature data with the multimodal feature fusion weights to obtain multimodal fused feature data. The foundation digital twin model update module is used to obtain target defect identification data based on multimodal fusion feature data; to map and update the foundation digital twin basic model data based on the target defect identification data to obtain foundation digital twin model data with defect attributes; the foundation digital twin basic model data is constructed based on the foundation spatial structure and stratigraphic distribution data. The foundation quality analysis module is used to determine the foundation quality test results based on digital twin model data of the foundation with defect attributes, superstructure load data, and stratum parameter data.

[0007] According to one aspect of the embodiments of this application, an electronic device is provided, the electronic device including a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the above-described foundation quality detection method.

[0008] According to one aspect of the embodiments of this application, the computer program product includes a computer program stored in a computer-readable storage medium. A processor of an electronic device reads the computer program from the computer-readable storage medium and executes the computer program, causing the electronic device to perform the aforementioned foundation quality testing method.

[0009] The technical solutions provided in this application embodiment may have the following beneficial effects: This application's embodiments collect various sensing data, including dynamic strain, electromagnetic field response, and Rayleigh wave propagation, and perform standardization and multimodal fusion processing. This fully integrates detection information from different dimensions, avoiding the bias caused by single or discrete data, and making defect identification more comprehensive and stable. Based on this, this application's embodiments map the identified defect results to a digital twin model constructed based on the foundation's spatial structure and stratigraphic distribution, forming a foundation digital twin model with defect attributes. This transforms the originally discrete and abstract defect data into intuitive and structured information that matches the actual foundation's spatial location and stratigraphic distribution.

[0010] This application's embodiments rely on a digital twin model with defect attributes, combined with the superstructure load and stratum parameters for comprehensive analysis. This can fully and accurately reflect the real impact of defects on the foundation's bearing capacity and deformation characteristics, making the quality assessment results more consistent with the actual engineering conditions and reducing assessment deviations.

[0011] In summary, the embodiments of this application provide an intuitive, comprehensive, and accurate method for foundation quality testing in engineering scenarios such as complex sites and deep foundations. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic flowchart illustrating the foundation quality testing method provided in this application embodiment; Figure 2 A structural block diagram of the foundation quality testing device provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of a server provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a terminal provided in an embodiment of this application. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0015] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0016] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0017] It should be understood that although this application may use terms such as first, second, etc., to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, a first parameter may also be referred to as a second parameter, and similarly, a second parameter may also be referred to as a first parameter. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0018] It should be noted that this application may display prompt interfaces, pop-ups, or output voice prompts before and during the collection of user-related data. These prompt interfaces, pop-ups, or voice prompts are used to inform the user that their relevant data is being collected. This ensures that the application only begins the steps related to collecting user-related data after receiving confirmation from the user regarding the prompt interface or pop-up. Otherwise, if no confirmation is received from the user, the steps to collect user-related data end, meaning no user-related data is collected. In other words, all user data collected in this application is collected with the user's consent and authorization, and the collection, use, and processing of relevant user data must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0019] Figure 1 This is a flowchart of a foundation quality testing method provided in an embodiment of this application. The method is executed by an electronic device and may include: S101: Acquire dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation, and obtain standardized sensing data based on the dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation.

[0020] In this embodiment, acquiring foundation dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data includes: The excitation parameters are determined based on the foundation detection depth and geological condition data. The excitation parameters include the excitation frequency, excitation force, and excitation duration. A controllable excitation is applied to the foundation according to the excitation parameters, and dynamic strain data, electromagnetic field response data and Rayleigh wave propagation data of the foundation are acquired synchronously through a sensor array. The sensor array is used to synchronously collect dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation at different depths and locations; the sensor array is arranged according to the excitation parameters.

[0021] In this embodiment, standardized sensing data is obtained based on foundation dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data, including: Wavelet denoising and outlier removal are performed on the dynamic strain data of the foundation to obtain denoised strain data; The electromagnetic field response data is normalized and interpolated in layers to obtain standard electromagnetic field data. Dispersion curve inversion and layered medium interpretation were performed on Rayleigh wave propagation data to obtain interpreted wave velocity data. The denoised strain data, standard electromagnetic field data, and interpreted wave velocity data are spatiotemporally aligned to obtain standardized sensing data.

[0022] In this embodiment, the dynamic strain data of the foundation is quantitative data characterizing the deformation response of the foundation after vibration, such as strain amplitude and strain change rate at different depths of the foundation. The electromagnetic field response data is feedback data on the electromagnetic field characteristics of the foundation after vibration, such as spatial distribution data of electric and magnetic field strength. The Rayleigh wave propagation data is characteristic data of Rayleigh waves propagating in the foundation medium, such as wave velocity, wave frequency, and propagation time. Standardized sensing data is unified and standardized data formed after processing multiple types of foundation detection data, such as strain and wave velocity fusion data after noise reduction, normalization, and spatiotemporal alignment. The vibration parameters are control indicators for applying vibration to the foundation, such as vibration frequency, vibration force, and vibration duration. The sensor array is a combined layout of multiple types of detection sensors, such as a layered array of strain sensors, electromagnetic probes, and Rayleigh wave sensors.

[0023] For example, in practical engineering applications, a ground survey can be conducted before implementing the method of this embodiment. The ground detection depth and geological condition data can be obtained through geological drilling and in-situ testing. The geological condition data includes information such as the distribution of the foundation soil layers, soil lithology, and soil density. Combining the detection depth requirements and geological condition data, appropriate excitation parameters are determined according to the geotechnical engineering testing specifications. The excitation frequency is adjusted according to the detection depth. High-frequency excitation parameters are selected for deep detection, and low-frequency excitation parameters are selected for shallow detection. The excitation force and excitation duration are reasonably set according to the soil hardness and density.

[0024] After the excitation parameters are determined, the sensor array is deployed on-site according to these parameters. The sensor array includes distributed strain sensors, electromagnetic probes, and Rayleigh wave sensors. It is deployed in a vertical layered and horizontal partitioned manner. The vertical layer spacing is set according to the foundation detection depth, and the horizontal partition spacing is adjusted according to the detection area to ensure that the sensors can cover different depths and positions of the detection target area, while ensuring that the acquisition ports of various sensors are on the same detection reference plane.

[0025] In this embodiment, a controllable excitation device is used to apply excitation to the foundation detection area according to the determined excitation parameters. During the excitation process, the output of the excitation device is kept stable to avoid additional vibration interference. At the same time, the sensor array is started to synchronously collect the dynamic strain data, electromagnetic field response data and Rayleigh wave propagation data of the foundation generated after the foundation is excited. During the acquisition process, the timestamps of various sensors are unified to ensure the spatiotemporal synchronization of signal acquisition. After the acquired signals are converted from analog to digital, the dynamic strain data, electromagnetic field response data and Rayleigh wave propagation data of the foundation are obtained.

[0026] This embodiment can preprocess three types of data separately. Specifically, for the dynamic strain data of the foundation, wavelet denoising is used to remove interference signals caused by environmental vibration and equipment noise. Then, outliers in the data are identified and removed through statistical analysis to obtain denoised strain data. For the electromagnetic field response data, the extreme value normalization method is used to map the data to a unified numerical range. Then, layered interpolation is performed according to the distribution of the foundation soil layers to supplement the missing data in the soil layer gaps to obtain standard electromagnetic field data. For the Rayleigh wave propagation data, the dispersion curve inversion method is used to restore the true distribution of wave velocity. Then, layered medium interpretation is performed in combination with the stratum distribution characteristics to clarify the Rayleigh wave propagation characteristics of different soil layers and obtain interpreted wave velocity data.

[0027] This embodiment can perform spatiotemporal alignment processing on the noise-reduced strain data, standard electromagnetic field data, and interpreted wave velocity data based on the deployment location and acquisition timestamp of various sensors. Using the spatial coordinates of the detection area as a reference, it matches the three types of data with the same spatial location and the same acquisition time to form a dataset with a unified format and spatiotemporal synchronization, thus obtaining standardized sensing data, which provides a data foundation for subsequent feature extraction and fusion processing.

[0028] This embodiment determines the excitation parameters by combining the foundation detection depth and geological conditions, ensuring the adaptability of the excitation to the detection requirements and improving the targeting of data acquisition. Relying on a sensor array to simultaneously acquire multiple types of data achieves spatiotemporal synchronization of the detection data, avoiding the one-sidedness of single data acquisition. Through categorized professional data processing and spatiotemporal alignment, data noise is effectively eliminated, data specifications are unified, and the resulting standardized sensing data possesses characteristics of uniform format and complete information, providing high-quality data support for subsequent multimodal feature fusion and enhancing the utilization value of foundation detection data.

[0029] S102: Extract features from standardized sensing data to obtain multiple single-modal feature data, determine multimodal feature fusion weights based on the multiple single-modal feature data, and fuse the single-modal feature data with the multimodal feature fusion weights to obtain multimodal fused feature data.

[0030] In this embodiment, the multiple single-modal feature data are obtained by independently extracting features from three different types of sensing data—dynamic strain, electromagnetic field response, and Rayleigh wave propagation—from the standardized sensing data. These three types of feature data are independent and each represents a single physical response characteristic of the foundation. The multimodal feature fusion weight is a fusion ratio coefficient assigned to each single-modal feature data, for example, a ratio determined based on the effectiveness of strain and wave velocity features. The multimodal fused feature data is the comprehensive feature data obtained by fusing the single-modal feature data according to the fusion weights, such as the comprehensive foundation response characteristics obtained by fusing strain, electromagnetic field, and wave velocity features according to their weights.

[0031] For example, in practical engineering applications, this embodiment can perform single-mode feature extraction operations on each type of standardized sensing data based on the data type classification. For the denoised strain data in the standardized sensing data, feature extraction methods are used to extract temporal and spatial features. Temporal feature extraction focuses on the strain variation trend over time, peak and trough characteristics, while spatial feature extraction focuses on the strain distribution differences and gradient characteristics at different spatial locations, forming strain-based single-mode feature data. For the standard electromagnetic field data, the spatial distribution characteristics, interlayer abrupt change characteristics, and regional correlation characteristics of the electric and magnetic field intensities are extracted, forming electromagnetic field-based single-mode feature data. For the interpreted wave velocity data, the wave velocity characteristics, dispersion characteristics, and propagation attenuation characteristics of Rayleigh waves in different soil layers are extracted, forming wave velocity-based single-mode feature data.

[0032] After extracting single-modal feature data, this embodiment determines the multimodal feature fusion weights based on the effectiveness and relevance of each single-modal feature data. First, feature effectiveness analysis is used to evaluate the contribution of each single-modal feature data to foundation defect identification, eliminating invalid or low-contribution features. Then, feature correlation analysis is used to calculate the degree of association between each single-modal feature data, avoiding feature redundancy. This embodiment combines the effectiveness and correlation analysis results and uses an objective weighting method to allocate corresponding fusion weights based on the contribution value of each single-modal feature data. The higher the contribution, the larger the proportion of the fusion weight. Simultaneously, the sum of the fusion weights of all single-modal feature data is ensured to be a unified benchmark value, thus completing the determination of the multimodal feature fusion weights.

[0033] This embodiment can perform a fusion operation between single-modal feature data and multi-modal feature fusion weights to form multi-modal fused feature data. During the fusion process, each single-modal feature data is first normalized to ensure that single-modal feature data of different types and dimensions are within the same numerical range, eliminating fusion bias caused by differences in dimensions. This embodiment can then perform a weighted superposition process on the normalized single-modal feature data according to the determined multi-modal feature fusion weights. High-contribution single-modal feature data is assigned a higher weight to strengthen its role in the comprehensive feature set, while low-contribution single-modal feature data is assigned a lower weight to reduce its interference.

[0034] After weighted superposition, this embodiment can integrate and optimize the fused feature data, eliminating redundant and abnormal features generated during the fusion process, and retaining comprehensive feature information that can fully and accurately reflect the physical characteristics of the foundation. Simultaneously, this embodiment can perform format unification processing on the integrated feature data to ensure consistency in data structure and dimensions, ultimately forming multimodal fused feature data. This data integrates the core information of three single-mode features: strain, electromagnetic field, and Rayleigh wave propagation, and can reflect the actual state of the foundation from multiple dimensions, providing comprehensive and accurate feature support for subsequent foundation defect identification.

[0035] This embodiment extracts single-modal feature data in a targeted manner to accurately mine the core information of various types of sensing data. Based on feature validity and relevance, fusion weights are determined to achieve scientific weight allocation and avoid feature redundancy and invalid interference. The multimodal fusion feature data formed by weighted fusion integrates multi-dimensional ground information, improving the comprehensiveness and accuracy of the features and laying a high-quality feature foundation for subsequent defect identification.

[0036] S103: Target defect identification data is obtained based on multimodal fusion feature data; the foundation digital twin basic model data is mapped and updated based on the target defect identification data to obtain foundation digital twin model data with defect attributes; the foundation digital twin basic model data is constructed based on the foundation spatial structure and stratum distribution data.

[0037] In this embodiment, target defect identification data is obtained based on multimodal fusion feature data, including: The multimodal fusion feature data is compared and matched with the defect features in the foundation testing knowledge graph data to obtain initial matching data; the foundation testing knowledge graph data is obtained based on foundation testing specifications, historical engineering cases, and defect judgment rules. The initial matching data is constrained and filtered according to the knowledge graph judgment rules to obtain the target matching data according to the knowledge graph judgment rules. The target defect type, target defect location, and target defect level are obtained by reasoning from the target matching data, and the target defect type, target defect location, and target defect level are used as target defect identification data.

[0038] In this embodiment, the target defect identification data includes the target defect type, the target defect location, and the target defect level; Based on the target defect identification data, the foundation digital twin model data is mapped and updated to obtain foundation digital twin model data with defect attributes, including: Spatial coordinate matching is performed between the target defect location and the digital twin foundation model data to obtain the coordinate matching result; Based on the coordinate matching results, the target defect type and target defect level are assigned to the corresponding spatial location in the foundation digital twin basic model data, so that the target defect type, target defect location and target defect level are associated and fused with the foundation spatial structure and stratum distribution data to obtain foundation digital twin model data with defect attributes.

[0039] In this embodiment, target defect identification data is judgment data characterizing the specific features of defects within the foundation, reflecting the core attributes of the defects, such as combined data of fissure-type defects in a certain foundation area, defects at 3 meters underground, and level 3 defects. Foundation digital twin basic model data is digital model data constructed based on the actual space and stratum information of the foundation, such as a three-dimensional spatial grid model matching the actual dimensions of the foundation, or digital model data superimposed with soil layer distribution information. Foundation inspection knowledge graph data is feature matching data formed by integrating relevant foundation inspection standards and cases, such as structured graph data containing various defect features and judgment criteria. Initial matching data is the preliminary result data after comparing multimodal fusion feature data with knowledge graph defect features, such as multiple sets of defect feature matching data with feature similarity reaching a threshold. Target matching data is valid matching data filtered by knowledge graph judgment rules, such as unique defect feature matching data that meets the standard requirements. Coordinate matching results are the corresponding data after matching the defect location with the spatial coordinates of the digital twin model, such as the three-dimensional coordinate data of the model corresponding to the actual location of the defect. Digital twin model data of a foundation with defect attributes is digital twin model data with superimposed defect features, such as digital model data in which the defect type and level are labeled at the corresponding coordinates of a 3D model.

[0040] For example, in this embodiment, foundation testing knowledge graph data can be constructed in advance, and the current foundation testing industry standards, defect detection cases of similar projects over the years, and industry-standard defect judgment rules data can be sorted out. The characteristics, judgment standards, and related effects of various defects can be structured and sorted out to form foundation testing knowledge graph data containing defect feature library and judgment rule library, thus completing the construction of basic matching basis.

[0041] In this embodiment, the multimodal fusion feature data obtained in the early stage can be input into the foundation detection knowledge graph system to start the feature comparison and matching process. Based on the defect feature library in the knowledge graph, the various feature information in the multimodal fusion feature data is matched one by one, the similarity between features is calculated, and the matching results with similarity meeting the preset threshold are selected to obtain the initial matching data. The initial matching data can contain multiple sets of different defect feature matching information, providing a basis for subsequent accurate judgment.

[0042] This embodiment can constrain and filter the initial matching data based on the content in the knowledge graph judgment rule base. According to the requirements of foundation testing specifications and the judgment criteria of similar engineering cases, matching results that do not meet the similarity standard, have incomplete feature matching, or do not conform to the judgment rules are eliminated, and unique matching results that meet all judgment requirements are retained to obtain the target matching data under the knowledge graph judgment rules.

[0043] This embodiment can perform defect reasoning based on target matching data. Combining the pre-set reasoning logic in the knowledge graph, it extracts defect-related feature information from the target matching data and sequentially derives the corresponding target defect type, target defect location, and target defect level. The defect type is classified according to the defect category in the knowledge graph corresponding to the feature matching results. The defect location is accurately located by combining the spatial distribution information of the features. The defect level is classified according to the level classification standard in the feature severity matching judgment rule. The derived three types of data are integrated to form target defect identification data.

[0044] After acquiring the target defect identification data, this embodiment can first rely on the spatial structure and stratum distribution data of the foundation obtained from the on-site foundation survey to construct the foundation digital twin foundation model data. The actual spatial size and three-dimensional shape of the foundation are restored through three-dimensional modeling. At the same time, information such as soil layer type, soil layer thickness, and distribution location in the stratum distribution data are superimposed on the three-dimensional model to form a digital foundation model that corresponds one-to-one with the actual foundation.

[0045] Based on the target defect location in the target defect identification data, spatial coordinate matching is performed with the ground digital twin basic model data. The actual spatial coordinates of the target defect location are converted into the three-dimensional coordinate system of the digital twin model. Through coordinate calibration, position correspondence and other means, the specific three-dimensional coordinates of the target defect location in the digital twin basic model are determined, and the coordinate matching results are obtained to ensure that the defect location corresponds accurately in the model.

[0046] Based on the coordinate matching results, this embodiment can locate the corresponding spatial coordinate position in the foundation digital twin basic model data, perform defect attribute assignment operations, and accurately mark the target defect type and target defect level at the coordinate position. During the assignment process, it is ensured that the defect attributes are associated with the model spatial position and stratum distribution data, so that the target defect type, target defect position and target defect level are deeply associated and integrated with the foundation spatial structure and stratum distribution data, and the mapping and update of the foundation digital twin basic model data are completed. Finally, foundation digital twin model data with defect attributes is obtained. This model can intuitively reflect the specific distribution of defects in the foundation space and their relationship with the strata.

[0047] This embodiment relies on knowledge graphs to achieve accurate comparison and judgment of defect features, improving the accuracy and standardization of target defect identification data and avoiding biases from experience-based interpretation. This embodiment maps defect attributes to a digital twin base model through spatial coordinate matching, realizing the association and fusion of defect information with foundation spatial structure and stratigraphic distribution data. This allows defect features to be presented in an intuitive digital form, and the resulting foundation digital twin model data with defect attributes provides an accurate and intuitive model foundation for subsequent comprehensive foundation quality evaluation.

[0048] S104: Determine the foundation quality inspection results based on digital twin model data of the foundation with defect attributes, superstructure load data, and stratum parameter data.

[0049] In this embodiment, the foundation quality inspection results are determined based on digital twin model data of the foundation with defect attributes, superstructure load data, and stratum parameter data, including: Mechanical coupling analysis was performed on the digital twin model data of the foundation with defect attributes, the superstructure load data, and the stratum parameter data to obtain the foundation bearing capacity data; Based on the digital twin model data of the foundation with defect attributes, the superstructure load data and the stratum parameter data, the foundation deformation simulation analysis is carried out to obtain settlement prediction data. Foundation quality is assessed based on foundation bearing capacity data and settlement prediction data to determine the foundation quality test results.

[0050] In this embodiment, the superstructure load data refers to the quantitative data related to the load transferred from the building structure to the foundation, such as the additional pressure at the base, the load distribution range, and the long-term load amplitude. The soil parameter data characterizes the mechanical and physical properties of each soil layer, such as the soil compression modulus, internal friction angle, cohesion, and unit weight. Mechanical coupling analysis is a comprehensive analysis method of the foundation's mechanical properties combining multi-source engineering data. The foundation bearing capacity data is the quantitative data characterizing the foundation's load-bearing capacity, such as the ultimate bearing capacity and allowable bearing capacity. Settlement prediction data is the quantitative data predicting the deformation of the foundation after loading, such as the final settlement and stratified settlement. The foundation quality inspection results are a comprehensive judgment of the overall quality of the foundation, such as a qualified, requiring reinforcement, or unqualified result.

[0051] For example, in practical engineering applications, this embodiment can first organize and preprocess the basic data. The foundation digital twin model data with defect attributes can directly use the results of the previous mapping and update. The superstructure load data is obtained through the calculation of the building structure design documents. Combined with the structural form, self-weight, and live load of the building, the actual load of the foundation is calculated to clarify the magnitude, distribution form and transmission method of the load. The stratum parameter data is obtained through in-situ testing and indoor geotechnical testing. In-situ testing adopts static load testing, vane shear testing and other methods. Indoor testing performs compression and shear tests on soil samples to obtain the mechanical and physical parameters of each soil layer. At the same time, the spatial reference of all data is unified with the digital twin model to ensure the consistency of analysis.

[0052] This embodiment can perform mechanical coupling analysis to obtain foundation bearing capacity data. It uses the foundation digital twin model data with defect attributes as the basis for analysis, superimposes the upper load data and stratum parameter data, and, based on the principles of geotechnical engineering mechanics analysis, considers the weakening effect of foundation defects on the continuity and mechanical properties of the soil. For soil layers corresponding to different defect locations and levels in the model, the mechanical parameter values ​​are adjusted, and the overall foundation is subjected to stress analysis. The bearing capacity contribution of each soil layer is calculated layer by layer, and the overall bearing capacity of the foundation and the bearing capacity of each region are comprehensively determined to obtain foundation bearing capacity data including allowable bearing capacity and ultimate bearing capacity.

[0053] After completing the bearing capacity analysis, this embodiment can perform foundation deformation simulation analysis to obtain settlement prediction data. Based on the digital twin model of the foundation with defect attributes, combined with the application method of the superstructure load and the stratum parameters, the layered summation method is used to calculate the deformation. The focus is on the change of soil compression characteristics in the defect area. Differential deformation analysis is performed on the defect area and the normal stratum. At the same time, combined with the consolidation characteristics of the foundation, the settlement of the foundation under instantaneous load and long-term load is predicted, including instantaneous settlement, consolidation settlement and secondary consolidation settlement. Settlement prediction data in different time dimensions and different areas are obtained, and the settlement distribution law and maximum settlement value are clarified.

[0054] This embodiment can conduct foundation quality assessment based on foundation bearing capacity data and settlement prediction data. First, it sorts out the current geotechnical engineering quality acceptance specifications, clarifies the passing judgment criteria for foundation bearing capacity and settlement amount, compares the calculated foundation bearing capacity data with the allowable bearing capacity required by the specifications to determine whether the foundation bearing capacity meets the engineering requirements. At the same time, it compares the settlement prediction data with the settlement limit and differential settlement limit specified by the specifications to analyze whether the foundation deformation is within the allowable range. This embodiment can comprehensively evaluate based on the results of the two analyses, combined with the influence degree of defects on the foundation performance. If both the bearing capacity and settlement amount meet the specification requirements, it is determined that the foundation quality is qualified. If one or both do not meet the requirements, judgment conclusions such as to-be-reinforced and unqualified are given according to the deviation degree, and finally a complete foundation quality inspection result is formed.

[0055] This embodiment conducts mechanical coupling analysis by integrating the digital twin model with defect attributes, upper loads, and formation parameters, fully considering the actual impact of defects on the foundation bearing capacity, making the bearing capacity data more in line with the engineering reality. Combining the deformation simulation analysis of multi-source data, it accurately predicts the foundation settlement characteristics and realizes the quantitative evaluation of foundation deformation. The comprehensive quality assessment based on the specification standards makes the inspection results have a rigorous engineering basis, improves the accuracy and scientificity of foundation quality judgment, and provides reliable decision-making support for the subsequent construction and use of the project.

[0056] In one embodiment of the present application, target defect recognition data is obtained based on multi-modal fusion feature data, including: Obtain the target geological scenario of the foundation and determine the target defect recognition model corresponding to the target geological scenario; the target geological scenario is miscellaneous fill, karst development area, reclamation foundation or frozen soil foundation; the target defect recognition model is obtained by training the initial defect recognition model based on the complex geological scenario sample data under the target geological scenario; the complex geological scenario sample data includes historical multi-modal fusion feature data and sample labels; the sample labels include defect type, defect location, and defect grade; the initial defect recognition model is a neural network model combining a convolutional neural network and an attention mechanism, including a convolutional layer, a pooling layer, an attention layer, a fully connected layer, and an output layer; Based on the multi-modal fusion feature data, obtain the target defect recognition data through the target defect recognition model; the target defect recognition data includes the target defect type, the target defect location, and the target defect grade.

[0057] In this embodiment, the target geological scene refers to the specific complex geological category of the foundation, serving as the basis for targeted defect identification, such as miscellaneous fill, karst development areas, reclaimed land foundations, or permafrost foundations. The target defect identification model refers to a defect identification model adapted to a specific target geological scene, such as a dedicated defect identification model trained for karst development areas, which can output defect judgment results based on multimodal fusion feature data. Complex geological scene sample data refers to the labeled dataset used for model training under the target geological scene, such as processed multimodal fusion feature data and corresponding annotation information from historical projects under the permafrost foundation scene. The initial defect identification model refers to the basic identification model that has not undergone scene-specific training; it is a neural network model combining convolutional neural networks and attention mechanisms, including convolutional layers, pooling layers, attention layers, fully connected layers, and an output layer. The convolutional layer is used to extract deep features from the data; the pooling layer is used for dimensionality reduction and filtering of the extracted features; the attention layer is used to enhance the target feature response; the fully connected layer is used for feature fusion and dimensionality mapping; and the output layer is used to output the final judgment result of the model. Sample labels refer to defect-related attribute information labeled for sample data of complex geological scenes, which are used for model supervised training, including defect type, defect location and defect level.

[0058] For example, in practical engineering applications, it is necessary to first obtain the target geological scene of the foundation. Through on-site geological surveys, drilling and sampling, in-situ testing, and sorting out regional geological data, the soil composition, geological structure, hydrological conditions, and special geological phenomena of the foundation are analyzed to determine the target geological scene to which the foundation belongs, and to clarify that it is one of miscellaneous fill, karst development area, reclaimed land foundation, or frozen soil foundation, so as to provide accurate basis for subsequent model matching and sample data selection.

[0059] This embodiment can determine the target defect recognition model corresponding to the target geological scene. This model is obtained by training an initial defect recognition model under scene-based supervision. The initial defect recognition model is a neural network model combining convolutional neural networks and attention mechanisms. The core includes convolutional layers, pooling layers, attention layers, fully connected layers, and an output layer. The convolutional layers are configured with multiple convolutional kernels of different scales to extract deep features of different dimensions from the input data. The pooling layers use downsampling to reduce the dimensionality of the features output by the convolutional layers, reducing computation while retaining core feature information. The attention layer strengthens the relevant defect features of the geological scene and suppresses irrelevant interference features through dynamic weight allocation. The fully connected layer realizes the fusion and linear mapping of multi-dimensional features. The output layer corresponds to three judgment dimensions: defect type, defect location, and defect level, and outputs the corresponding judgment results.

[0060] Before model training, a complex geological scene sample dataset is constructed under the target geological scenario. In this embodiment, historical multimodal fusion feature data matching the target geological scenario can be collected through engineering databases, historical inspection reports, or industry shared data. These data are comprehensive feature data obtained after standardized perception data feature extraction and weight fusion under the corresponding scenario. In this embodiment, the collected historical multimodal fusion feature data can be screened to remove invalid data with missing data, too many outliers, or inconsistencies between the detection environment and the actual project, and retain valid data with complete data and clear features. Finally, sample labels are labeled on the screened valid historical multimodal fusion feature data. Based on the actual inspection conclusions and defect verification reports of the corresponding project, the defect type, defect location, and defect level corresponding to each data point are accurately labeled to complete the construction of the sample dataset. The dataset is then divided into training set, validation set, and test set according to the proportions to provide data support for model training.

[0061] This embodiment can train the initial defect recognition model in a scenario-based manner based on a constructed complex geological scene sample dataset. The historical multimodal fusion feature data of the training set is input into the initial defect recognition model, and multiple convolution operations are performed through convolutional layers to extract deep local features that represent the physical state of the foundation and defect characteristics from the data. The pooling layer performs dimensionality reduction and filtering on the feature map output by the convolutional layer, retaining key feature information and reducing the risk of model overfitting. The attention layer assigns higher weights to the features corresponding to typical defects in the target geological scene according to the feature patterns of the scene, strengthening the defect feature response, while weakening irrelevant features such as environmental interference and equipment interference in the scene. The fully connected layer fully fuses the high-dimensional features processed by the attention layer, completing the mapping transformation of features from high-dimensional space to low-dimensional judgment space, providing a data foundation for accurate defect judgment. The output layer outputs the prediction results of defect type, defect location, and defect level according to the output results of the fully connected layer.

[0062] This embodiment compares the model's prediction results with the sample labels, calculates the error values ​​of various judgment results, iteratively optimizes the parameters of each layer of the model through the backpropagation algorithm, monitors the training effect of the model in real time using the validation set, and dynamically adjusts training hyperparameters such as learning rate and number of iterations. Once the model's defect recognition accuracy, location determination accuracy, and grade classification conformity rate on the test set all reach the preset standards, the training is completed, and the initial defect recognition model completes the scenario adaptation and is transformed into a target defect recognition model for the target geological scene.

[0063] After determining the target defect identification model, this embodiment can preprocess the multimodal fusion feature data obtained from the current engineering preprocessing, including data normalization, dimension adjustment, format conversion, etc., so that the numerical range, dimension size, and data format of the data are fully matched with the input requirements of the target defect identification model, ensuring that the data can be successfully input into the model for feature processing and defect judgment.

[0064] This embodiment can input the adapted multimodal fusion feature data into the target defect recognition model. The model internally completes feature processing and defect judgment in sequence according to a preset process: the convolutional layer performs deep feature extraction on the input multimodal fusion feature data to capture the subtle features of hidden foundation defects in the data; the pooling layer performs dimensionality reduction and filtering on the convolution results to retain the core defect features; the attention layer combines the feature patterns of the target geological scene to further enhance the defect feature response in this scene and weaken the interference features generated in the field detection; the fully connected layer fuses and maps the processed features, and the output layer finally outputs the corresponding target defect type, target defect location and target defect level. By integrating the three types of data, the final target defect recognition data is obtained, completing the acquisition of target defect recognition data based on multimodal fusion feature data.

[0065] This embodiment employs a dedicated target defect recognition model for different complex geological scenarios, achieving precise scenario-based adaptation of defect recognition and avoiding recognition biases caused by differences in the characteristics of various geological scenarios. The initial defect recognition model combines convolutional neural networks and attention mechanisms, with each network layer having a clear division of labor, enabling accurate extraction and enhancement of defect features. After training with sample data from complex geological scenarios, the model's recognition accuracy and anti-interference capability are significantly improved. The target defect recognition data obtained based on this model can accurately reflect the actual defect state of the foundation, providing high-quality data support for subsequent digital twin model mapping updates.

[0066] Corresponding to the foundation quality testing method in the above embodiment, Figure 2 This is a structural block diagram of a foundation quality testing device provided according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The foundation quality testing device 20 includes: a foundation data acquisition module 21, a multimodal feature fusion module 22, a foundation digital twin model update module 23, and a foundation quality analysis module.

[0067] Among them, the foundation data acquisition module 21 is used to acquire foundation dynamic strain data, electromagnetic field response data and Rayleigh wave propagation data, and obtain standardized sensing data based on the foundation dynamic strain data, electromagnetic field response data and Rayleigh wave propagation data; The multimodal feature fusion module 22 is used to extract features from standardized perceptual data to obtain multiple single-modal feature data, determine multimodal feature fusion weights based on the multiple single-modal feature data, and fuse the single-modal feature data with the multimodal feature fusion weights to obtain multimodal fused feature data. The foundation digital twin model update module 23 is used to obtain target defect identification data based on multimodal fusion feature data; to map and update the foundation digital twin basic model data based on the target defect identification data to obtain foundation digital twin model data with defect attributes; the foundation digital twin basic model data is constructed based on the foundation spatial structure and stratigraphic distribution data of the foundation. The foundation quality analysis module 24 is used to determine the foundation quality test results based on the foundation digital twin model data with defect attributes, the superstructure load data, and the stratum parameter data.

[0068] In one embodiment of this application, the foundation data acquisition module 21 is specifically used for: The excitation parameters are determined based on the foundation detection depth and geological condition data. The excitation parameters include the excitation frequency, excitation force, and excitation duration. A controllable excitation is applied to the foundation according to the excitation parameters, and dynamic strain data, electromagnetic field response data and Rayleigh wave propagation data of the foundation are acquired synchronously through a sensor array. The sensor array is used to synchronously collect dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation at different depths and locations; the sensor array is arranged according to the excitation parameters.

[0069] In one embodiment of this application, the foundation data acquisition module 21 is further used for: Wavelet denoising and outlier removal are performed on the dynamic strain data of the foundation to obtain denoised strain data; The electromagnetic field response data is normalized and interpolated in layers to obtain standard electromagnetic field data. Dispersion curve inversion and layered medium interpretation were performed on Rayleigh wave propagation data to obtain interpreted wave velocity data. The denoised strain data, standard electromagnetic field data, and interpreted wave velocity data are spatiotemporally aligned to obtain standardized sensing data.

[0070] In one embodiment of this application, the foundation digital twin model update module 23 is specifically used for: The multimodal fusion feature data is compared and matched with the defect features in the foundation testing knowledge graph data to obtain initial matching data; the foundation testing knowledge graph data is obtained based on foundation testing specifications, historical engineering cases, and defect judgment rules. The initial matching data is constrained and filtered according to the knowledge graph judgment rules to obtain the target matching data according to the knowledge graph judgment rules. The target defect type, target defect location, and target defect level are obtained by reasoning from the target matching data, and the target defect type, target defect location, and target defect level are used as target defect identification data.

[0071] In one embodiment of this application, the foundation digital twin model update module 23 is specifically used for: The target geological scene of the foundation is obtained, and the target defect recognition model corresponding to the target geological scene is determined. The target geological scene is a mixed fill soil, a karst development area, a reclaimed land foundation, or a permafrost foundation. The target defect recognition model is obtained by training an initial defect recognition model based on sample data of complex geological scenes under the target geological scene. The sample data of complex geological scenes includes historical multimodal fusion feature data and sample labels. The sample labels include defect type, defect location, and defect level. The initial defect recognition model is a neural network model that combines convolutional neural network and attention mechanism, including convolutional layer, pooling layer, attention layer, fully connected layer, and output layer. Based on multimodal fusion feature data, target defect identification data is obtained through a target defect identification model; the target defect identification data includes target defect type, target defect location, and target defect level.

[0072] In one embodiment of this application, the foundation digital twin model update module 23 is further used for: Spatial coordinate matching is performed between the target defect location and the digital twin foundation model data to obtain the coordinate matching result; Based on the coordinate matching results, the target defect type and target defect level are assigned to the corresponding spatial location in the foundation digital twin basic model data, so that the target defect type, target defect location and target defect level are associated and fused with the foundation spatial structure and stratum distribution data to obtain foundation digital twin model data with defect attributes.

[0073] In one embodiment of this application, the foundation quality analysis module 24 is specifically used for: Mechanical coupling analysis was performed on the digital twin model data of the foundation with defect attributes, the superstructure load data, and the stratum parameter data to obtain the foundation bearing capacity data; Based on the digital twin model data of the foundation with defect attributes, the superstructure load data and the stratum parameter data, the foundation deformation simulation analysis is carried out to obtain settlement prediction data. Foundation quality is assessed based on foundation bearing capacity data and settlement prediction data to determine the foundation quality test results.

[0074] It should be noted that the specific limitations of the above-described foundation quality testing device 20 can be found in the limitations of the foundation quality testing method described above, and will not be repeated here. Each module of the above device can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in the processor of the electronic device in hardware form or independent of the processor, or stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of each module.

[0075] This application also provides an electronic device, which includes: a processor and a memory, wherein the memory stores a computer program; the processor is used to execute the computer program in the memory to implement the foundation quality detection method provided in the above method embodiments.

[0076] This application also provides an electronic device, which includes a processor and a memory, wherein at least one computer program is stored in the memory. The at least one computer program is loaded and executed by one or more processors to enable the electronic device to implement any of the aforementioned foundation quality testing methods. The electronic device can be a server or a terminal; the structures of servers and terminals will be described below.

[0077] Figure 3 This is a schematic diagram of a server structure provided in an embodiment of this application. The server can vary significantly due to differences in configuration or performance. It may include one or more Central Processing Units (CPUs) 31 and one or more memories 32. The one or more memories 32 store at least one computer program, which is loaded and executed by the one or more processors 31 to enable the server to implement the foundation quality detection method provided in the various method embodiments described above. Of course, the server may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The server may also include other components for implementing device functions, which will not be elaborated upon here.

[0078] Figure 4 This is a schematic diagram of the structure of a terminal provided in an embodiment of this application. The terminal may also be referred to as user equipment, portable terminal, laptop terminal, desktop terminal, or other names.

[0079] Typically, a terminal includes a processor 41 and a memory 42.

[0080] Processor 41 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 41 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 41 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 41 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 41 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0081] The memory 42 may include one or more computer-readable storage media, which may be non-transitory. The memory 42 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 42 are used to store at least one instruction, which is executed by the processor 41 to enable the terminal to implement the foundation quality detection method provided in the method embodiments of this application.

[0082] In some embodiments, the terminal may also optionally include: a peripheral device interface 43 and at least one peripheral device. The processor 41, memory 42, and peripheral device interface 43 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 43 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of: a radio frequency circuit 44, a display screen 45, a camera assembly 46, an audio circuit 47, and a power supply 48.

[0083] Peripheral interface 43 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 41 and memory 42. In some embodiments, processor 41, memory 42 and peripheral interface 43 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 41, memory 42 and peripheral interface 43 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0084] The radio frequency (RF) circuit 44 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 44 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 44 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF circuit 44 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 44 can communicate with other terminals via at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: metropolitan area networks (MANs), various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks (WLANs), and / or Wi-Fi (Wireless Fidelity) networks. In some embodiments, the RF circuit 44 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0085] The display screen 45 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When the display screen 45 is a touch display, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to the processor 41 for processing. In this case, the display screen 45 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 45, located on the front panel of the terminal; in other embodiments, there may be at least two display screens, respectively located on different surfaces of the terminal or in a folded design; in still other embodiments, the display screen 45 may be a flexible display screen, located on a curved or folded surface of the terminal. Furthermore, the display screen 45 may be configured as a non-rectangular, irregular shape, i.e., a non-rectangular screen. The display screen 45 may be made of materials such as LCD (Liquid Crystal Display) or OLED (Organic Light-Emitting Diode).

[0086] The camera assembly 46 is used to acquire images or videos. Optionally, the camera assembly 46 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the terminal, and the rear-facing camera is located on the back of the terminal. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly 46 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash is a combination of a warm-light flash and a cool-light flash, which can be used for light compensation at different color temperatures.

[0087] The audio circuit 47 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting them into electrical signals that are input to the processor 41 for processing, or to the radio frequency circuit 44 for voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, each positioned at a different location on the terminal. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert electrical signals from the processor 41 or the radio frequency circuit 44 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuit 47 may also include a headphone jack.

[0088] The power source 48 is used to power the various components in the terminal. The power source 48 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When the power source 48 includes a rechargeable battery, the rechargeable battery can support wired or wireless charging. The rechargeable battery can also be used to support fast charging technology.

[0089] In some embodiments, the terminal further includes one or more sensors 49. The one or more sensors 49 include, but are not limited to: an accelerometer 410, a gyroscope 411, a pressure sensor 412, an optical sensor 413, and a proximity sensor 414.

[0090] Accelerometer 410 can detect the magnitude of acceleration along the three coordinate axes of a coordinate system established by the terminal. For example, accelerometer 410 can be used to detect the components of gravitational acceleration along the three coordinate axes. Processor 41 can control display screen 45 to display the user interface in either a landscape or portrait view based on the gravitational acceleration signal acquired by accelerometer 410. Accelerometer 410 can also be used for collecting motion data from games or users.

[0091] The gyroscope sensor 411 can detect the terminal's orientation and rotation angle. The gyroscope sensor 411 can work in conjunction with the accelerometer sensor 410 to collect the user's 3D movements on the terminal. Based on the data collected by the gyroscope sensor 411, the processor 41 can perform the following functions: motion sensing (e.g., changing the UI based on the user's tilt), image stabilization during shooting, game control, and inertial navigation.

[0092] The pressure sensor 412 can be disposed on the side bezel of the terminal and / or the lower layer of the display screen 45. When the pressure sensor 412 is disposed on the side bezel of the terminal, it can detect the user's grip signal on the terminal, and the processor 41 can perform left / right hand recognition or quick operation based on the grip signal collected by the pressure sensor 412. When the pressure sensor 412 is disposed on the lower layer of the display screen 45, the processor 41 can control the operable controls on the UI interface based on the user's pressure operation on the display screen 45. The operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0093] Optical sensor 413 is used to collect ambient light intensity. In one embodiment, processor 41 can control the display brightness of display screen 45 based on the ambient light intensity collected by optical sensor 413. Specifically, when the ambient light intensity is high, the display brightness of display screen 45 is increased; when the ambient light intensity is low, the display brightness of display screen 45 is decreased. In another embodiment, processor 41 can also dynamically adjust the shooting parameters of camera assembly 46 based on the ambient light intensity collected by optical sensor 1613.

[0094] The proximity sensor 414, also known as a distance sensor, is typically installed on the front panel of the terminal. The proximity sensor 414 is used to detect the distance between the user and the front of the terminal. In one embodiment, when the proximity sensor 414 detects that the distance between the user and the front of the terminal is gradually decreasing, the processor 41 controls the display screen 45 to switch from a screen-on state to a screen-off state; when the proximity sensor 414 detects that the distance between the user and the front of the terminal is gradually increasing, the processor 41 controls the display screen 45 to switch from a screen-off state to a screen-on state.

[0095] Those skilled in the art will understand that Figure 4The structure shown does not constitute a limitation on the terminal and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0096] In an exemplary embodiment, a computer-readable storage medium is also provided, which stores at least one computer program that is loaded and executed by a processor of an electronic device to enable the computer to implement any of the above-described foundation quality testing methods.

[0097] In one possible implementation, the aforementioned computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a solid-state drive (SSD), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc. The random access memory can include resistive random access memory (ReRAM) and dynamic random access memory (DRAM).

[0098] In an exemplary embodiment, a computer program or computer program product is also provided, which includes computer instructions loaded and executed by a processor to enable the computer to implement any of the above-described foundation quality testing methods.

[0099] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, data stored, data displayed, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0100] In other words, the data collection and processing in this application should strictly comply with the requirements of relevant national laws and regulations, obtain the informed consent or separate consent of the personal information subject, and conduct subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0101] It should be further noted that the terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The implementation methods described in the above exemplary embodiments do not represent all implementation methods consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application.

[0102] It should be understood that "multiple" as used in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0103] Furthermore, the step numbers described herein are merely illustrative of one possible execution order between steps. In some other embodiments, the steps may not be executed in the order of their numbers, such as two steps with different numbers being executed simultaneously, or two steps with different numbers being executed in the reverse order of the illustration. This application does not limit this.

[0104] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. Optionally, the program is stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0105] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for testing the quality of foundation soil, characterized in that, include: Acquire dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation, and obtain standardized sensing data based on the dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation; Feature extraction is performed on the standardized sensing data to obtain multiple single-modal feature data, and multimodal feature fusion weights are determined based on the multiple single-modal feature data; The single-modal feature data and the multimodal feature fusion weights are fused to obtain multimodal fused feature data; Target defect identification data is obtained based on multimodal fusion feature data; Based on the target defect identification data, the foundation digital twin basic model data is mapped and updated to obtain foundation digital twin model data with defect attributes; the foundation digital twin basic model data is constructed based on the foundation spatial structure and stratigraphic distribution data; The foundation quality inspection results are determined based on the foundation digital twin model data with defect attributes, the superstructure load data, and the stratum parameter data.

2. The foundation quality testing method as described in claim 1, characterized in that, The acquisition of foundation dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data includes: The excitation parameters are determined based on the foundation detection depth and geological condition data. The excitation parameters include the excitation frequency, excitation force, and excitation duration. A controllable vibration excitation is applied to the foundation according to the excitation parameters, and the dynamic strain data of the foundation, the electromagnetic field response data, and the Rayleigh wave propagation data are synchronously acquired through a sensor array. The sensor array is used to synchronously collect dynamic strain data, electromagnetic field response data, and Rayleigh wave propagation data of the foundation at different depths and locations; the sensor array is arranged according to the excitation parameters.

3. The foundation quality testing method as described in claim 2, characterized in that, The standardized sensing data obtained based on the foundation dynamic strain data, the electromagnetic field response data, and the Rayleigh wave propagation data includes: Wavelet denoising and outlier removal are performed on the dynamic strain data of the foundation to obtain denoised strain data; The electromagnetic field response data is normalized and hierarchically interpolated to obtain standard electromagnetic field data; The Rayleigh wave propagation data are subjected to dispersion curve inversion and layered medium interpretation to obtain the interpreted wave velocity data; The standardized sensing data is obtained by spatiotemporally aligning the noise-reduced strain data, the standard electromagnetic field data, and the interpreted wave velocity data.

4. The foundation quality testing method as described in claim 1, characterized in that, The target defect identification data obtained based on multimodal fusion feature data includes: The multimodal fusion feature data is compared and matched with the defect features in the foundation testing knowledge graph data to obtain initial matching data; the foundation testing knowledge graph data is obtained based on foundation testing specifications, historical engineering cases, and defect judgment rules. The initial matching data is constrained and filtered according to the knowledge graph determination rules to obtain the target matching data according to the knowledge graph determination rules. The target defect type, target defect location, and target defect level are obtained by reasoning from the target matching data, and the target defect type, target defect location, and target defect level are used as the target defect identification data.

5. The foundation quality testing method as described in claim 1, characterized in that, The target defect identification data obtained based on multimodal fusion feature data includes: The target geological scene of the foundation is obtained, and the target defect recognition model corresponding to the target geological scene is determined. The target geological scene is a mixed fill soil, a karst development area, a reclaimed land foundation, or a permafrost foundation. The target defect recognition model is obtained by training an initial defect recognition model based on sample data of complex geological scenes under the target geological scene. The sample data of complex geological scenes includes historical multimodal fusion feature data and sample labels. The sample labels include defect type, defect location, and defect level. The initial defect recognition model is a neural network model combining convolutional neural network and attention mechanism, including convolutional layer, pooling layer, attention layer, fully connected layer, and output layer. Based on the multimodal fusion feature data, the target defect identification data is obtained through the target defect identification model; the target defect identification data includes the target defect type, the target defect location, and the target defect level.

6. The foundation quality testing method as described in claim 1, characterized in that, The target defect identification data includes the target defect type, target defect location, and target defect level; The process of mapping and updating the foundation digital twin model data based on the target defect identification data to obtain foundation digital twin model data with defect attributes includes: Spatial coordinate matching is performed between the target defect location and the foundation digital twin model data to obtain the coordinate matching result; Based on the coordinate matching results, the target defect type and the target defect level are assigned to the spatial location corresponding to the foundation digital twin basic model data, so that the target defect type, the target defect location, and the target defect level are associated and fused with the foundation spatial structure and the stratum distribution data to obtain the foundation digital twin model data with defect attributes.

7. The foundation quality testing method as described in claim 1, characterized in that, The determination of foundation quality inspection results based on the foundation digital twin model data with defect attributes, superstructure load data, and stratum parameter data includes: Mechanical coupling analysis is performed on the digital twin model data of the foundation with defective attributes, the superstructure load data, and the stratum parameter data to obtain the foundation bearing capacity data; Based on the foundation digital twin model data with defective attributes, the superstructure load data, and the stratum parameter data, foundation deformation simulation analysis is performed to obtain settlement prediction data. Based on the foundation bearing capacity data and the settlement prediction data, the foundation quality is assessed to obtain the foundation quality test results.

8. A foundation quality testing device, characterized in that, include: The foundation data acquisition module is used to acquire foundation dynamic strain data, electromagnetic field response data and Rayleigh wave propagation data, and obtain standardized sensing data based on the foundation dynamic strain data, electromagnetic field response data and Rayleigh wave propagation data; A multimodal feature fusion module is used to extract features from the standardized perception data to obtain multiple single-modal feature data, and to determine the multimodal feature fusion weights based on the multiple single-modal feature data; The single-modal feature data and the multimodal feature fusion weights are fused to obtain multimodal fused feature data; The ground-based digital twin model update module is used to obtain target defect identification data based on multimodal fusion feature data; Based on the target defect identification data, the foundation digital twin basic model data is mapped and updated to obtain foundation digital twin model data with defect attributes; the foundation digital twin basic model data is constructed based on the foundation spatial structure and stratigraphic distribution data; The foundation quality analysis module is used to determine the foundation quality test results based on the foundation digital twin model data with defect attributes, superstructure load data, and stratum parameter data.

9. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory storing a computer program, which is loaded and executed by the processor to implement the foundation quality testing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is loaded and executed by a processor to implement the foundation quality testing method as described in any one of claims 1 to 7.

Images