Pile-forming detection system and detection method for high-pressure jet grouting pile

Abstract

The invention discloses a pile-forming detection system and detection method for a high-pressure jet grouting pile, and relates to the technical field of pile-forming integrity detection.The pile-forming detection method comprises the steps that ultrasonic signals in the pile body range and apparent resistivity data of a pile-soil mixed medium are synchronously collected; performing imaging processing on the ultrasonic signal based on synthetic aperture focusing covered by dual rays of delay parameters, and constructing an acoustic image; based on the apparent resistivity data, establishing an apparent resistivity cloud picture reflecting the slurry diffusion range and the cement soil distribution uniformity; performing feature extraction and fusion according to the acoustic image and the apparent resistivity cloud picture to obtain fusion features after spatial consistency enhancement, and performing integrity evaluation on the formed pile according to the fusion features; correlation analysis is carried out according to the comprehensive integrity evaluation result and construction parameters, and the construction parameters are optimized for the recognized defects; according to the method, the problems that pile-forming integrity evaluation is inaccurate, defect identification is incomplete, and online optimization of construction parameters is difficult to realize are solved.

Description

Technical Field

[0001] This invention relates to the field of pile integrity testing technology, and more specifically, to a pile integrity testing system and method for high-pressure jet grouting piles. Background Technology

[0002] Cemented Soil Pressure Grouting (CSP) piles are widely used in soft soil foundation reinforcement, slope protection, and heavy building foundation engineering due to their advantages such as fast construction speed, high bearing capacity, and wide applicability. During CSP construction, cement grout is injected under high pressure and mixed with the soil to form the pile. The diffusion range of the grout and the uniformity of the pile-soil mixture directly affect the pile's bearing capacity and long-term stability. Therefore, real-time and accurate detection and evaluation of pile quality are crucial for ensuring project safety and optimizing construction techniques.

[0003] Existing technologies for processing and interpreting detection data largely rely on prior models or single-moment measurements, failing to fully utilize the continuously acquired acoustic and electrical information during construction. They also struggle to comprehensively analyze the temporal changes, propagation paths, and intensity of pile defects. For different types of defects such as necking, widening, segregation, or voids, existing methods lack quantitative, visual, and spatially relevant identification tools. Furthermore, they struggle to correlate detection results with construction parameters to achieve online optimization and control of the pile formation process. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a pile formation detection system and method for high-pressure jet grouting piles. By simultaneously acquiring acoustic and electrical information, and using a method based on synthetic aperture focusing imaging with dual-ray coverage and spatiotemporal fusion analysis, the system solves the problems of inaccurate pile integrity assessment, incomplete defect identification, and difficulty in achieving online optimization of construction parameters.

[0005] To achieve the above objectives, the present invention provides the following technical solution: Firstly, this application provides a method for pile formation inspection of high-pressure jet grouting piles. The method includes: simultaneously acquiring ultrasonic signals and apparent resistivity data of the pile-soil mixture within the pile body area; performing imaging processing on the ultrasonic signals using synthetic aperture focusing with dual-ray coverage based on delay parameters to construct an acoustic image reflecting the internal density and defects of the pile body; establishing an apparent resistivity cloud map reflecting the grout diffusion range and the uniformity of cement-soil distribution based on the apparent resistivity data; performing feature extraction and fusion on the acoustic image and the apparent resistivity cloud map respectively to obtain fused features with enhanced spatial consistency, and conducting a pile integrity assessment based on the fused features; and performing correlation analysis between the comprehensive integrity assessment results and construction parameters to optimize construction parameters for identified defects.

[0006] In one embodiment, synthetic aperture focusing based on delay parameters is used for imaging to construct an acoustic image reflecting the internal density and defects of the pile. This includes: constructing multiple propagation ray paths, obtaining the signal-to-noise ratio of each ray, and filtering effective rays to form an effective ray set; determining the dual-ray coverage area based on the effective ray set; performing cross-correlation analysis on the received signal and corresponding transmitted signal of each effective ray within the dual-ray coverage area to obtain delay parameters; performing delay compensation on signals from different receiving channels and different ray paths based on the delay parameters, and performing synthetic aperture focusing enhancement within the corresponding imaging spatial unit; after completing the synthetic aperture focusing enhancement, extracting the acoustic attribute parameters corresponding to each imaging spatial unit, and mapping them through spatial location to form an acoustic image.

[0007] In one embodiment, within the dual-ray coverage area, cross-correlation analysis is performed on the received signal and the corresponding transmitted signal of each effective ray to obtain delay parameters. This includes: preprocessing the received signal and the corresponding transmitted signal of each effective ray, and calculating the delay peak value of each ray through cross-correlation analysis; performing adjacent ray weighted averaging and consistency checks on the delay peak values ​​of each ray within the dual-ray coverage area to remove outliers and generate initial delay parameters corresponding to each imaging unit within the dual-ray coverage area; comparing the initial delay parameters of each ray, and using the minimum variance method to fuse them to obtain the final delay parameters with the smallest deviation.

[0008] In one embodiment, based on the apparent resistivity data, an apparent resistivity cloud map reflecting the slurry diffusion range and the uniformity of cement-soil distribution is established, including: collecting corresponding apparent resistivity data at multiple consecutive construction time nodes and adding a time stamp to each measurement data; constructing an apparent resistivity time series based on the apparent resistivity data located in the same imaging spatial unit according to the acquisition time sequence; preprocessing the apparent resistivity time series of each imaging spatial unit to obtain a stable apparent resistivity time series, and calculating the apparent resistivity change between adjacent time nodes for each imaging spatial unit to obtain a first-order time gradient; calculating the change between adjacent time nodes a second time based on the first-order time gradient to obtain a second-order time gradient, forming a time gradient feature set; determining the time nodes where significant changes occur and the change intensity parameters based on the time gradient feature set, and constructing the apparent resistivity cloud map.

[0009] In one embodiment, the process involves determining the time nodes and intensity parameters of significant changes based on a time gradient feature set, and constructing an apparent resistivity cloud map. This includes: recording the time node of the first occurrence of a significant electrical change for each imaging spatial unit identified as having undergone a significant electrical change, based on the time gradient feature set; sorting the imaging spatial units according to the first occurrence time of the significant change to determine the spatial propagation order of the electrical effects of the slurry or cement-soil; obtaining the intensity parameters of the change within the significant electrical change for each imaging spatial unit based on the time gradient feature set; weighting and fusing the apparent resistivity data of the imaging spatial unit and its spatially adjacent imaging spatial units, combining the propagation order and the intensity parameters, to generate apparent resistivity values ​​for imaging spatial units with clear spatial orientation; and mapping the apparent resistivity values ​​of each imaging spatial unit according to its spatial location to generate an apparent resistivity cloud map.

[0010] In one embodiment, feature extraction and fusion are performed on the acoustic image and the apparent resistivity cloud map respectively to obtain fused features with enhanced spatial consistency. This includes: dividing the three-dimensional imaging space corresponding to the acoustic image and the apparent resistivity cloud map into several imaging units; extracting acoustic feature vectors and electrical feature vectors for each imaging unit and forming a preliminary fused feature set; determining the set of spatially neighboring units for each imaging unit and assigning initial weights to each neighboring unit; using a convolutional kernel smoothing method to perform a weighted average of the preliminary fused feature set of the unit itself and the preliminary fused feature sets of neighboring units according to the initial weights to obtain a smoothed feature vector; wherein, the convolutional kernel smoothing method further includes adjusting the weight distribution and radius of the convolutional kernel according to the spatial consistency requirements; and using the smoothed feature vector of each imaging unit as the fused feature with enhanced spatial consistency.

[0011] In one embodiment, the integrity assessment of the pile based on the fusion features includes: using the imaging units of the fusion features as nodes of a graph structure, with node features being smoothed feature vectors, and establishing edges of the graph structure to obtain a spatial graph structure; performing graph Laplacian smoothing on the spatial graph structure, and obtaining the deviation index between the node features and neighboring features for the smoothed node features; if the deviation index exceeds a preset threshold, marking the node as an abnormal node; determining defect areas based on the topological relationship of the abnormal nodes in the spatial graph; extracting geometric features for each defect area; combining the defect areas with normal node areas, and performing a comprehensive integrity assessment based on the pile design parameters, wherein the comprehensive integrity assessment includes the pile diameter, necking / expansion position, and internal segregation or void defects.

[0012] In one embodiment, a correlation analysis is performed between the comprehensive integrity assessment results and construction parameters. For identified defects, the construction parameters are optimized, including: parameterizing the comprehensive integrity assessment results to obtain an integrity assessment parameter sequence; acquiring the construction parameters corresponding to the integrity assessment parameter sequence; aligning the integrity assessment parameter sequence and construction parameters according to pile depth markings, so that the integrity assessment results for each depth segment correspond to a set of actual construction parameters; statistically analyzing the correspondence between different combinations of construction parameters and defect types and degrees on the aligned integrity assessment parameters and construction parameters to obtain abnormal construction parameter combinations; determining the adjustment direction and magnitude of the construction parameters based on the defect type and its corresponding abnormal construction parameter combinations to obtain a construction parameter adjustment scheme; and sending the determined construction parameter adjustment scheme to the construction control system for real-time adjustment of the construction parameters during subsequent construction sections or when re-spraying is performed at the axial position corresponding to the defect.

[0013] Secondly, this application provides a pile formation inspection system for high-pressure jet grouting piles. The system includes: an acquisition module for simultaneously acquiring ultrasonic signals and apparent resistivity data of the pile-soil mixture within the pile body area; an acoustic image construction module for imaging the ultrasonic signals using synthetic aperture focusing with dual-ray coverage based on delay parameters to construct an acoustic image reflecting the internal density and defects of the pile body; a resistivity cloud map construction module for establishing an apparent resistivity cloud map reflecting the grout diffusion range and cement-soil distribution uniformity based on the apparent resistivity data; an integrity assessment module for performing feature extraction and fusion based on the acoustic image and the apparent resistivity cloud map to obtain spatially consistent fused features, and performing a pile integrity assessment based on the fused features; and a correlation adjustment module for performing correlation analysis based on the comprehensive integrity assessment results and construction parameters, and optimizing construction parameters for identified defects.

[0014] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: By simultaneously acquiring dual-mode data of ultrasound and apparent resistivity under the same construction conditions, and constructing high-precision acoustic imaging based on dual-ray coverage and apparent resistivity cloud maps based on time gradients, a spatially consistent acoustic-electric feature fusion mechanism is introduced to effectively solve the inherent shortcomings of single detection methods in terms of resolution, directivity, and noise resistance. At the same time, by combining convolutional kernel smoothing with graph structure neighborhood consistency analysis, stable identification and precise positioning of internal defects in the pile body are achieved. Furthermore, the identification results are quantitatively correlated with construction parameters to form a closed-loop feedback control process of defect-cause-parameter regulation. Thus, technically, the entire process of multi-source sensing, fine imaging, reliable assessment and online optimization of construction parameters is achieved, which has innovative and practical value in significantly improving the accuracy of pile quality assessment, enhancing the ability to identify complex defects, and realizing active regulation of the construction process. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a pile formation testing method for high-pressure jet grouting piles provided in an embodiment of this application.

[0016] Figure 2 This is a schematic diagram of a high-pressure jet grouting pile pile testing system provided in an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] Reference Figure 1 As shown in the diagram, a pile formation testing method for high-pressure jet grouting piles provided by the present invention includes the following steps: Step S1: Synchronously collect ultrasonic signals and apparent resistivity data of the pile-soil mixture within the pile body area.

[0019] During the high-pressure jet grouting construction, a central measuring pipe is pre-embedded at the center of the construction pile location, and several peripheral monitoring pipes are arranged in a regular polygon around it; a shear wave ultrasonic transducer and a high-precision electrode are integrated and installed inside the central measuring pipe; and an ultrasonic receiving transducer array and a potential measuring electrode are integrated and installed in each peripheral monitoring pipe. First, the shear wave ultrasonic transducer and high-precision electrode in the central measuring tube, as well as the ultrasonic receiving transducer array and potential measuring electrode in each peripheral monitoring tube, are uniformly initialized and clock-synchronized. After establishing a unified time reference, the shear wave ultrasonic transducer in the central measuring tube periodically emits shear wave ultrasonic signals according to a preset frequency and energy parameters. The ultrasonic receiving transducer array in the peripheral monitoring tubes synchronously receives the transmitted and reflected wave signals in the pile body and pile-soil mixture within the same time window and records complete waveform data to obtain ultrasonic signals. Simultaneously, a stable excitation current is injected into the pile-soil mixture by the high-precision electrode in the central measuring tube. The potential measuring electrodes in each peripheral monitoring tube synchronously collect the potential values ​​of each measuring point under the same time stamp conditions as the ultrasonic acquisition. By recording the injected current and potential difference, the apparent resistivity data at the corresponding time is obtained. All ultrasonic signals and apparent resistivity data are appended with a unified time stamp and stored in association with construction parameters such as the current injection pressure, lifting speed, and rotation speed. This achieves synchronous and corresponding acquisition of acoustic and electrical information within the pile body under the same construction state, providing a consistent and reliable data foundation for subsequent dual-mode imaging and integrity evaluation.

[0020] It should be noted that ultrasonic signals refer to time-domain or frequency-domain acoustic response signals that reflect the internal density, structural continuity, and defect characteristics of the medium; apparent resistivity data refers to equivalent resistivity parameters used to characterize the grout diffusion range and the electrical distribution characteristics of the pile-soil mixture.

[0021] Step S2: The ultrasonic signal is imaged using synthetic aperture focusing with dual-ray coverage based on delay parameters to construct an acoustic image reflecting the internal density and defects of the pile.

[0022] In this embodiment, the ultrasonic signal is imaged using synthetic aperture focusing with dual-ray coverage based on delay parameters to construct an acoustic image reflecting the internal density and defects of the pile, including: Based on the spatial relative positions of the transmitting transducer and each receiving transducer, multiple propagation ray paths of ultrasonic waves are constructed inside the pile. The signal-to-noise ratio of each ray is obtained, and rays with a signal-to-noise ratio greater than a set threshold are selected to form an effective ray set for imaging. The threshold is determined by pre-experimental statistics based on engineering experience of similar pile foundation detection and combined with the background noise level at the site, and is used to eliminate invalid rays that are greatly affected by noise. Based on the effective ray set, the spatial region within any imaging spatial unit is obtained that is traversed by at least two rays propagating in different directions, and this region is defined as the dual-ray coverage region to ensure the redundancy and reliability of acoustic information within the imaging unit. The imaging spatial unit is formed by discretizing the space of the pile body and the pile-soil mixture to be detected into multiple imaging units along the radial and depth directions.

[0023] Within the dual-ray coverage area, cross-correlation analysis is performed on the received signal and the corresponding transmitted signal of each effective ray to obtain the delay parameters; Based on the delay parameter, delay compensation is performed on ultrasonic signals from different receiving channels and different X-ray paths, and signal synthesis is performed in the corresponding imaging spatial unit using a coherent superposition method to achieve synthetic aperture focusing enhancement. Understandably, for each imaging unit, the ultrasonic signals from different receiving channels and different X-ray paths are first time-compensated according to the delay parameter to align the arrival times of each signal in the imaging unit; then, the compensated signals are superimposed within the spatial unit, and a coherent superposition method can be used to preserve the signal phase information to enhance the focusing resolution, thereby completing the synthetic aperture focusing enhancement.

[0024] After completing the synthetic aperture focusing enhancement, the acoustic property parameters corresponding to each imaging spatial unit are extracted. The acoustic property parameters include sound wave propagation speed, energy intensity and attenuation characteristics. Based on acoustic property parameters, the parameter values ​​of each imaging spatial unit are mapped according to its spatial location to form an acoustic image.

[0025] In the acoustic image, areas where the sound wave propagation speed is below a preset sound velocity threshold, where the energy abnormally attenuates to below a preset energy threshold, or where the distribution is discontinuous are identified and marked to highlight potential structural defects within the pile, such as necking, voids, segregation, or insufficient density. Dense areas of the pile-soil mixture typically have high mechanical stiffness, resulting in relatively fast ultrasonic wave propagation speeds; conversely, the propagation speed decreases significantly in voids or loose areas. Therefore, the sound velocity distribution directly reflects the density within the pile. Dense areas have good interface continuity, resulting in low ultrasonic wave transmission loss and strong received signal energy; while in areas of low density, wave energy attenuation is significant. Therefore, the energy intensity distribution can be used to determine density. The sound velocity threshold is determined based on the theoretical sound velocity range of the designed pile material and the dense pile-soil mixture, combined with on-site measured calibration data; the energy threshold is obtained by statistical analysis of the received signal energy in a normal dense area; the threshold for determining discontinuous distribution is set based on the variation amplitude of acoustic attribute parameters of adjacent imaging spatial units. When a sudden change in sound velocity or energy parameters occurs in space exceeding the preset variation amplitude, it is identified as a structural discontinuity or potential defect area.

[0026] Further, in step S2, within the dual-ray coverage area, a cross-correlation analysis is performed on the received signal and the corresponding transmitted signal of each effective ray to obtain delay parameters, including: The received signal and the corresponding transmitted signal of each effective ray are preprocessed, including DC removal, bandpass filtering and normalization. The peak delay of each ray is calculated by cross-correlation analysis based on the preprocessed received and transmitted signals. Specifically, the cross-correlation function is calculated based on the preprocessed received and transmitted signals, and the peak delay time is determined by searching for the global maximum peak value of the cross-correlation function. The cross-correlation function is a mathematical tool used to quantify how the similarity between two signal sequences changes with time delay. The peak position of the cross-correlation function corresponds to the time delay of optimal alignment between the received and transmitted signals, which is the propagation time of the ultrasound wave from transmission to reception. By searching for the maximum peak of the cross-correlation function, the delay time of the ray path can be determined, thus providing a basis for delay compensation of the imaging unit.

[0027] The delay peak values ​​of each ray within the dual-ray coverage area are weighted and checked for consistency with adjacent rays to remove outliers and generate the initial delay parameters for each imaging unit within the dual-ray coverage area. In this process, a weighted average is taken of the delay peak of each delay peak and the delay peak of its spatial neighboring rays, while outliers that deviate from the neighboring values ​​by more than a set threshold are removed (i.e., consistency check). This process yields the initial delay parameters of each imaging unit within the dual-ray coverage area.

[0028] For each imaging unit traversed by multiple effective rays, the initial delay parameters of each ray are compared, and the minimum variance method is used for fusion to obtain the final delay parameters with the smallest deviation and full utilization of ray redundancy information, so as to ensure the focusing accuracy of the dual-ray coverage area.

[0029] For example, the initial delay parameters obtained from each ray are first statistically analyzed as a set of observations. The mean and variance of each initial delay parameter are calculated, and the deviation of each initial delay parameter from the mean is used as a consistency evaluation index. Based on the consistency evaluation index, a weight is assigned to the initial delay parameter of each ray, and the weight is inversely proportional to the variance corresponding to the initial delay parameter of that ray. On this basis, a weighted minimum variance fusion method is used to sum the initial delay parameters of multiple rays in a weighted manner to obtain the final delay parameter that minimizes the overall variance of the weighted result. Among them, the delay parameters of rays with small deviations and low variances are arranged from largest to smallest, and the weights of the delay parameters of rays with large deviations are reduced during the fusion process. This allows for the full utilization of ray redundancy information while suppressing abnormal delays, obtaining stable and reliable final delay parameters, and ensuring the accuracy and consistency of subsequent delay compensation and synthetic aperture focusing.

[0030] It should be noted that by constructing multiple ray paths inside the pile and selecting high signal-to-noise ratio effective rays, cross-correlation analysis of the received and transmitted signals is performed within the dual-ray coverage area to obtain delay parameters. Then, the final delay parameters are obtained through weighted averaging, consistency checks, and minimum variance fusion. This achieves high-precision delay compensation and synthetic aperture focusing enhancement of ultrasonic signals. Combined with the extraction and spatial mapping of acoustic property parameters, high-resolution and reliable acoustic images of the pile interior can be generated. Potential defects can be identified through sound velocity, energy, and distribution anomalies, thus balancing the accuracy and stability of density assessment and defect location. This significantly improves the imaging accuracy, noise resistance, and detection capability of defects in complex mixed media structures within the pile.

[0031] Step S3: Based on the apparent resistivity data, establish an apparent resistivity cloud map that reflects the slurry diffusion range and the uniformity of cement-soil distribution.

[0032] In this embodiment, based on the apparent resistivity data, an apparent resistivity cloud map reflecting the slurry diffusion range and the uniformity of cement-soil distribution is established, including: Collect corresponding apparent resistivity data at multiple consecutive construction time points, and add a time stamp to each measurement data; Based on apparent resistivity data, the apparent resistivity data located in the same imaging spatial unit are organized in the order of acquisition time to construct the apparent resistivity time series corresponding to each imaging spatial unit. The apparent resistivity time series of each imaging spatial unit is preprocessed, the preprocessing including: Outliers are removed from measurements that deviate significantly from the statistical mean of the time series and exceed a preset multiple of the standard deviation; time series of different imaging spatial units are time-aligned; and the time series are smoothed using moving average or median filtering to reduce the impact of construction disturbances and instantaneous noise on time variation analysis, thereby obtaining a stable apparent resistivity time series. Based on a stable apparent resistivity time series, the apparent resistivity change between adjacent time nodes is calculated for each imaging spatial unit. That is, the first-order time gradient is obtained by the ratio of the apparent resistivity data difference between adjacent construction time nodes to the corresponding time interval. The change between adjacent time nodes is calculated twice for the first-order time gradient to obtain the second-order time gradient, thereby forming a time gradient feature set that characterizes the rate and trend of electrical change of the imaging spatial unit. Based on the time gradient feature set, the time nodes where significant changes occur and the parameters of change intensity are determined, and an apparent resistivity cloud map is constructed.

[0033] Furthermore, based on the time gradient feature set, the time nodes where significant changes occur and the parameters of change intensity are determined, and an apparent resistivity cloud map is constructed, including: Based on the set of time gradient features, a threshold judgment is made on the time gradient of each imaging spatial unit. When the absolute value of its time gradient exceeds the preset change threshold, it is determined that the imaging spatial unit has undergone a significant electrical change at the corresponding time node, and the time node at which the significant change first occurs is recorded. Based on the time of first occurrence of significant changes in each imaging spatial unit, the imaging spatial units are sorted to determine the spatial propagation order of the electrical effects of slurry or cement-soil. For example, the time of the first occurrence of a significant electrical change corresponding to each imaging spatial unit is used as the time identifier parameter of that imaging spatial unit, and all imaging spatial units are sorted in ascending order according to the time identifier parameter; thus forming a spatial propagation sequence that reflects the electrical influence during slurry diffusion or cement-soil formation.

[0034] Based on the time gradient feature set, the apparent resistivity change amplitude of each imaging spatial unit within a significant electrical change is statistically analyzed to obtain the corresponding change intensity parameter, which is used to characterize the degree of influence of slurry diffusion or cement-soil formation on each imaging spatial unit. Based on the apparent resistivity data of the imaging spatial unit and its spatially adjacent imaging spatial units, and combined with the propagation sequence and change intensity parameters, normalized weighted fusion is performed to generate an apparent resistivity value of the imaging spatial unit with clear spatial orientation. The weighting is related to the start time and intensity parameter of the change in the imaging spatial unit, so that the imaging spatial units with early start time and large intensity parameter are given high weights during the fusion process.

[0035] By mapping the apparent resistivity values ​​of each imaging spatial unit according to their spatial location, an apparent resistivity cloud map is generated. Based on the spatial continuity of apparent resistivity changes, the distribution of change start time, and the distribution of change intensity, the slurry diffusion range, the uniformly distributed cement-soil area, and the area that may have insufficient diffusion or abnormal distribution are identified and marked.

[0036] It should be noted that the constructed apparent resistivity cloud map is based on temporal apparent resistivity data during continuous construction. It introduces temporal gradient features to characterize electrical changes, transforming the traditional apparent resistivity imaging method based on single-moment spatial interpolation into a spatiotemporal joint expression method that takes into account "change rate, change sequence, and change intensity". This effectively reduces the spatial orientation ambiguity problem caused by the volume average effect of apparent resistivity. At the same time, by jointly constraining the start time and change intensity of significant changes, the imaging spatial units that are truly affected by grout diffusion are highlighted in the data-driven weighted fusion process. This makes the generated apparent resistivity cloud map not only reflect the electrical distribution results, but also implicitly contain the temporal path and intensity of grout diffusion, which can more accurately and stably characterize the grout diffusion range and cement-soil distribution uniformity in the high-pressure jet grouting pile.

[0037] Step S4: Perform feature extraction and fusion based on the acoustic image and apparent resistivity cloud map respectively to obtain fused features with enhanced spatial consistency, and evaluate the integrity of the pile based on the fused features.

[0038] In this embodiment, feature extraction and fusion are performed on the acoustic image and the apparent resistivity cloud map respectively to obtain fused features with enhanced spatial consistency, including: The three-dimensional imaging space corresponding to the acoustic image and the apparent resistivity cloud map is divided into several imaging units, and each imaging unit corresponds to a spatial node or voxel. Acoustic and electrical feature vectors are extracted for each imaging unit, and a preliminary fusion feature set is formed. The acoustic and electrical feature vectors are normalized to ensure dimensional uniformity in subsequent weighted fusion. The acoustic feature vectors include sound wave propagation speed, signal energy intensity, and attenuation characteristics; the electrical feature vectors include resistivity time gradient, uniformity index, and electrical change intensity parameters.

[0039] For each imaging unit, its set of spatially neighboring units is determined. Neighboring units can be selected by grid nodes within a sphere with the center of the unit as the center and a radius equal to the size of the imaging unit. An initial weight is assigned to each neighboring unit. The initial weight is calculated based on the spatial distance, which can be the reciprocal of the Euclidean distance from the center of the neighboring unit to the center of the target unit. The initial weight is then normalized by dividing the initial weight of each neighboring imaging spatial unit by the sum of the corresponding weights, so that the sum of the normalized weights is 1. This ensures that the feature magnitude is not changed during the weighted fusion process and improves the stability of spatial consistency fusion. The convolution kernel smoothing method is adopted to perform a weighted average of the initial fusion feature set of itself and the initial fusion feature set of neighboring units according to the initial weights to obtain a smooth feature vector. During the weighting process, the acoustic feature vector and the electrical feature vector are weighted separately to ensure that the influence of each feature on the result in the fusion is relatively balanced, to avoid a certain type of feature from dominating the fusion result due to its large numerical magnitude, and to maintain the consistency of the dimensions. The smoothing process can employ single-pass convolution or multiple-pass iterative convolution, with the number of iterations ranging from 1 to 3. The process can be adjusted according to spatial consistency requirements and local feature preservation needs to suppress mismatches caused by local outliers or resolution differences, while preserving key feature signals in areas of significant change. In the process of smoothing each imaging unit using convolutional kernels, the mean of each feature vector in the preliminary fusion feature set of each unit in its neighborhood is obtained. The difference between each neighboring unit and the mean is calculated as a difference vector, and the difference vector is squared and averaged along the feature dimension to obtain a spatial consistency index. When the index is lower than a preset threshold (e.g., the variance is less than 10% of the overall feature variance), the initial weights are used for smoothing. When the index is higher than the threshold, it indicates that there is a significant feature change in the unit. In this case, the radius of action of the convolutional kernel is reduced or the weights of neighboring units are reduced according to a preset ratio to reduce the impact of smoothing on the features of the unit, thereby suppressing mismatches caused by local outliers or resolution differences, while preserving the true key feature signals of the unit.

[0040] The smoothed feature vector of each imaging unit is used as the fusion feature after spatial consistency enhancement.

[0041] It should be noted that during feature fusion, acoustic and electrical feature vectors may have issues such as resolution differences, uneven sampling points, and local outliers or noise interference. Direct fusion may lead to mismatches, local discontinuities, or amplification of abnormal information. By using convolutional kernel smoothing, the feature vector of each imaging unit can be weighted and averaged using spatial neighborhood information, effectively suppressing errors caused by local anomalies and uneven sampling, while maintaining the feature signals in areas of significant change. This enhances the spatial consistency of acoustic and electrical features, improves the reliability, continuity, and sensitivity to real defects of the fused features, and provides a more accurate and robust foundation for subsequent integrity assessment.

[0042] Furthermore, based on the fusion characteristics, an integrity assessment of the piles is performed, including: The imaging units with the fused features are used as nodes of the graph structure. The node features are smooth feature vectors. The edges of the graph structure are established according to the proximity relationship in three-dimensional space to obtain the spatial graph structure. For each imaging unit, its center coordinates in three-dimensional space are recorded. The neighborhood radius R (e.g., 2 to 3 times the size of the imaging unit) is set according to the Euclidean distance to determine the set of neighboring units. The edges of the graph structure are established between neighboring units. The spatial graph structure is subjected to graph Laplacian smoothing, which enhances the local consistency of node features by minimizing the weighted difference between node features and the features of neighboring nodes. For the node features after Laplacian smoothing of the graph, obtain the deviation index between the node features and the neighbor features, such as the z-score standardized difference; If the deviation index exceeds the preset threshold, the node is marked as an abnormal node, indicating that there may be defects such as necking, voids, segregation or insufficient density. Based on the topological relationship of abnormal nodes in the spatial graph, the connected subgraph analysis method is used to classify spatially adjacent and similar abnormal nodes into the same defect region. Geometric features are extracted for each defect area, including the location of the spatial center of gravity, volume, maximum extension direction (principal inertial direction), and offset distance relative to the central axis of the pile, providing basic data for defect location; By combining the defective areas with normal node areas, and based on the pile design parameters, the overall integrity of the pile diameter, necking / expansion location, and internal segregation or void defects is evaluated.

[0043] The process involves combining the defective areas with normal node areas and evaluating the comprehensive integrity of the pile diameter, necking / enlargement locations, and internal segregation or void defects based on pile design parameters. This can be understood as follows: First, using the pile design parameters (including design pile diameter, design axis position, and allowable deviation range) as a benchmark, the identified defective areas are aligned radially and axially with the pile design geometric model. By statistically analyzing the spatial distribution range of normal nodes on each depth section, the effective diameter of the actual pile at different depths is inferred and compared with the design pile diameter. When the effective diameter is less than the design pile diameter and the difference exceeds a preset allowable deviation threshold, it is determined that there is a necking phenomenon at that depth layer. When the effective diameter is greater than the design pile diameter and exceeds the allowable deviation threshold, it is determined that there is an enlargement phenomenon. Areas with the same deviation in multiple consecutive adjacent depth layers are identified as the specific locations of necking or enlargement. Meanwhile, for defect areas located inside the pile body and discontinuously distributed, based on their size, spatial connectivity, and differences in acoustic and electrical characteristics with surrounding normal nodes, material segregation anomalies and void anomalies are distinguished. When the defect area volume is less than a preset threshold, is discretely distributed, or has local connectivity with surrounding normal nodes, and its acoustic characteristics show a gradual decrease in sound velocity and energy, and its electrical characteristics show a moderate change in apparent resistivity with a transitional distribution with the surrounding area, the defect is judged to be more consistent with the characteristics of material segregation anomalies. Conversely, when the defect area volume is greater than a preset threshold, has strong spatial connectivity, or forms an obvious continuous cavity structure, and its acoustic characteristics show a significant attenuation of sound energy and a wave velocity significantly lower than the average value of the surrounding area, and its electrical characteristics show a sudden change in apparent resistivity, a steep gradient, and obvious discontinuity with surrounding normal nodes, the defect is judged to be more consistent with the characteristics of void anomalies.

[0044] Step S5: Based on the comprehensive integrity assessment results and construction parameters, conduct correlation analysis, optimize construction parameters for identified defects, and achieve online control of pile quality.

[0045] In this embodiment, a correlation analysis is performed between the comprehensive integrity assessment results and construction parameters. For identified defects, the construction parameters are optimized to achieve online control of pile formation quality, including: The integrity assessment results are parameterized to obtain an integrity assessment parameter sequence. The integrity assessment parameter sequence includes the actual pile diameter at each depth location, the axial position and range of the necking or expanding neck, and the defect degree index corresponding to the defect type (necking / expanding neck position and internal segregation or void defects). The defect degree index is determined by the weighted result of the abnormal amplitude of acoustic characteristics and the intensity of changes in electrical characteristics. Obtain the construction parameters corresponding to the integrity assessment parameter sequence, including injection pressure, lifting speed, and rotary jet rotation speed; Based on the pile depth markings, the integrity assessment parameter sequence is aligned with the construction parameters so that the integrity assessment results for each depth segment correspond to a set of actual construction parameters. Statistical analysis was conducted on the correspondence between different combinations of construction parameters and defect types and degrees, and abnormal combinations of construction parameters were obtained by comparing the aligned integrity assessment parameters and construction parameters. When a defect is identified within a certain axial depth segment and its defect severity index exceeds the preset defect judgment threshold, the construction parameter group corresponding to that depth segment is recorded as an abnormal construction parameter combination, and its corresponding defect type is marked. Based on the defect type and its corresponding abnormal construction parameter combination, the adjustment direction and adjustment range of the construction parameters are determined, resulting in a construction parameter adjustment plan, which includes: When the defect type is necking, it is determined that the injection pressure needs to be increased, and the adjustment range is determined according to the deviation ratio between the defect severity index and the design threshold. When the defect type is necking defect, it is determined that it is caused by excessive concentration of injection energy in the local axial or radial range, and it is determined that the injection pressure needs to be reduced and / or the lifting speed increased in order to reduce the grouting energy input per unit length. The reduction in injection pressure or the increase in lifting speed are determined based on the deviation ratio between the actual equivalent pile diameter of the enlarged neck section and the designed pile diameter, and the adjusted injection pressure and lifting speed are limited to the range of process parameters allowed by the design. When the defect type is a void defect, it is determined that the grout inside the space is not fully filled or the grout continuity is insufficient. It is determined that supplementary grouting or re-spraying should be carried out at the corresponding axial position. During the re-spraying process, the spraying pressure should be appropriately increased and / or the lifting speed should be reduced to enhance the filling capacity of the void area of ​​the grout. The volume estimate of the cavity defect (obtaining the spatial distribution of the cavity defect area, using the three-dimensional mesh accumulation method to mark the continuous abnormal imaging units for connectivity, and calculating their total volume to obtain the defect volume estimate), the spatial connectivity index (the number of voxels of the largest connected component among the imaging units in the defect area as the spatial connectivity index), and the acoustic and electrical characteristic differences between the cavity defect and the surrounding normal area are normalized and then weighted and fused according to preset weights to form a comprehensive severity index. Based on the correspondence between the comprehensive severity index and the preset graded threshold interval, the corresponding construction parameter adjustment range is determined based on the standard construction parameters in each interval and limited to the preset safety threshold range. When the defect type is segregation, it is determined that the jet energy distribution needs to be adjusted to make the jet energy within a unit length tend to be uniform. The determined construction parameter adjustment plan is sent to the construction control system, and the spraying pressure, lifting speed and rotary spraying speed are adjusted in real time during subsequent construction sections or when re-spraying is carried out at the axial position corresponding to the defect.

[0046] Reference Figure 2 As shown in the diagram, a structural schematic of a high-pressure jet grouting pile pile testing system provided by the present invention includes an acquisition module, an acoustic image construction module, a resistivity cloud map construction module, an integrity assessment module, and a correlation adjustment module. The modules are interconnected. The acquisition module is used to simultaneously acquire ultrasonic signals and apparent resistivity data of the pile-soil mixture within the pile body area; An acoustic image construction module is used to perform imaging processing on the ultrasonic signal based on the delay parameter and the synthetic aperture focusing of dual-ray coverage to construct an acoustic image reflecting the internal density and defects of the pile. The resistivity cloud map construction module is used to establish an apparent resistivity cloud map reflecting the slurry diffusion range and the uniformity of cement-soil distribution based on the apparent resistivity data. The integrity assessment module is used to extract and fuse features based on acoustic images and apparent resistivity cloud maps respectively, to obtain fused features with enhanced spatial consistency, and to assess the integrity of the pile based on the fused features. The correlation adjustment module is used to perform correlation analysis between the comprehensive integrity assessment results and construction parameters, and to optimize the construction parameters for identified defects.

[0047] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0048] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0049] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0050] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0051] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for pile completion testing of high-pressure jet grouting piles, characterized in that, Includes the following steps: Simultaneously collect ultrasonic signals and apparent resistivity data of the pile-soil mixture within the pile body area; The ultrasonic signal is imaged using synthetic aperture focusing with dual-ray coverage based on delay parameters to construct an acoustic image reflecting the internal density and defects of the pile. Based on the apparent resistivity data, an apparent resistivity cloud map reflecting the slurry diffusion range and the uniformity of cement-soil distribution is established. Feature extraction and fusion are performed on acoustic images and apparent resistivity cloud maps respectively to obtain fused features with enhanced spatial consistency, and the integrity of piles is evaluated based on the fused features. Based on the correlation analysis between the comprehensive integrity assessment results and the construction parameters, the construction parameters are optimized for the identified defects.

2. The pile formation testing method for high-pressure jet grouting piles according to claim 1, characterized in that, The ultrasonic signal is imaged using synthetic aperture focusing with dual-ray coverage based on delay parameters to construct an acoustic image reflecting the internal density and defects of the pile, including: Construct multiple propagation ray paths, obtain the signal-to-noise ratio of each ray, and filter out effective rays to form an effective ray set; Based on the effective ray set, the dual-ray coverage area is determined; Within the dual-ray coverage area, cross-correlation analysis is performed on the received signal and the corresponding transmitted signal of each effective ray to obtain the delay parameters; Based on the delay parameter, delay compensation is performed on signals from different receiving channels and different ray paths, and synthetic aperture focusing enhancement is performed in the corresponding imaging spatial unit; After completing the synthetic aperture focusing enhancement, the acoustic attribute parameters corresponding to each imaging spatial unit are extracted and mapped through spatial position to form an acoustic image.

3. The pile formation testing method for high-pressure jet grouting piles according to claim 2, characterized in that, Within the dual-ray coverage area, cross-correlation analysis is performed on the received signal and the corresponding transmitted signal of each effective ray to obtain delay parameters, including: The received signal and the corresponding transmitted signal of each effective ray are preprocessed, and the delay peak value of each ray is calculated by cross-correlation analysis; The delay peak values ​​of each ray within the dual-ray coverage area are weighted and checked for consistency with adjacent rays to remove outliers and generate the initial delay parameters for each imaging unit within the dual-ray coverage area. By comparing the initial delay parameters of each ray, the minimum variance method is used to fuse them, and the final delay parameters with the smallest deviation are obtained.

4. The pile formation testing method for high-pressure jet grouting piles according to claim 1, characterized in that, The step of establishing an apparent resistivity cloud map reflecting the slurry diffusion range and the uniformity of cement-soil distribution based on the apparent resistivity data includes: Collect corresponding apparent resistivity data at multiple consecutive construction time points, and add a time stamp to each measurement data; Based on apparent resistivity data, apparent resistivity data located in the same imaging spatial unit are constructed into an apparent resistivity time series according to the acquisition time order; The apparent resistivity time series of each imaging spatial unit is preprocessed to obtain a stable apparent resistivity time series, and the apparent resistivity change between adjacent time nodes is calculated for each imaging spatial unit to obtain the first-order time gradient. The change between adjacent time nodes is calculated twice for the first-order time gradient to obtain the second-order time gradient, forming a time gradient feature set. Based on the time gradient feature set, the time nodes where significant changes occur and the change intensity parameters are determined, and an apparent resistivity cloud map is constructed.

5. The pile formation testing method for high-pressure jet grouting piles according to claim 4, characterized in that, The process of determining the time nodes and intensity parameters of significant changes based on the time gradient feature set, and constructing an apparent resistivity cloud map, includes: Based on the time gradient feature set, for imaging spatial units that are determined to have undergone significant electrical changes, the time node at which the significant change first occurs is recorded; Based on the time of first occurrence of significant changes in each imaging spatial unit, the imaging spatial units are sorted to determine the spatial propagation order of the electrical effects of slurry or cement-soil. Based on the time gradient feature set, the change intensity parameters of each imaging spatial unit within a significant electrical change are obtained; Based on the apparent resistivity data of the imaging spatial unit and its spatially adjacent imaging spatial units, and combined with the propagation sequence and change intensity parameters, a weighted fusion is performed to generate the apparent resistivity value of the imaging spatial unit with a clear spatial orientation. By mapping the apparent resistivity values ​​of each imaging spatial unit according to its spatial location, an apparent resistivity cloud map is generated.

6. The pile formation testing method for high-pressure jet grouting piles according to claim 1, characterized in that, The step of extracting and fusing features from acoustic images and apparent resistivity cloud maps to obtain spatially consistent fused features includes: The three-dimensional imaging space corresponding to the acoustic image and the apparent resistivity cloud map is divided into several imaging units; Acoustic and electrical feature vectors are extracted for each imaging unit, and a preliminary fusion feature set is formed. For each imaging unit, determine its set of spatially neighboring units and assign initial weights to each neighboring unit; The convolutional kernel smoothing method is used to perform a weighted average of the initial fused feature set of itself and the initial fused feature set of neighboring units according to the initial weights to obtain a smooth feature vector. Among them, the convolution kernel smoothing method also includes adjusting the weight distribution and radius size of the convolution kernel according to spatial consistency requirements; The smoothed feature vector of each imaging unit is used as the fusion feature after spatial consistency enhancement.

7. The pile formation testing method for high-pressure jet grouting piles according to claim 1, characterized in that, The integrity assessment of the pile foundation based on the fusion characteristics includes: The imaging units with fused features are used as nodes of the graph structure, the node features are smooth feature vectors, and the edges of the graph structure are established to obtain the spatial graph structure. The spatial graph structure is smoothed using graph Laplacian smoothing, and the deviation index between the smoothed node features and the neighbor features is obtained. If the deviation index exceeds a preset threshold, the node is marked as an abnormal node; Defect areas are determined based on the topological relationships of abnormal nodes in the spatial graph; Extract geometric features for each defect region; By combining the defective areas with the normal node areas, a comprehensive integrity assessment is conducted based on the pile design parameters. The comprehensive integrity assessment includes the pile diameter, necking / expansion location, and internal segregation or void defects.

8. The pile formation testing method for high-pressure jet grouting piles according to claim 1, characterized in that, The process involves correlation analysis between the comprehensive integrity assessment results and construction parameters, and optimization of construction parameters for identified defects, including: The integrity assessment parameters are parameterized based on the comprehensive integrity assessment results to obtain the integrity assessment parameter sequence. Obtain the construction parameters corresponding to the integrity assessment parameter sequence; Based on the pile depth markings, the integrity assessment parameter sequence is aligned with the construction parameters so that the integrity assessment results for each depth segment correspond to a set of actual construction parameters. Statistical analysis was conducted on the correspondence between different combinations of construction parameters and defect types and degrees, and abnormal combinations of construction parameters were obtained by comparing the aligned integrity assessment parameters and construction parameters. Based on the defect type and its corresponding abnormal construction parameter combination, determine the adjustment direction and adjustment range of the construction parameters to obtain the construction parameter adjustment plan; The determined construction parameter adjustment plan is sent to the construction control system, and the construction parameters are adjusted in real time during subsequent construction sections or when re-spraying is carried out at the axial position corresponding to the defect.

9. A system for pile testing of high-pressure jet grouting piles as described in any one of claims 1-8, characterized in that, include: The acquisition module is used to simultaneously acquire ultrasonic signals and apparent resistivity data of the pile-soil mixture within the pile body area; An acoustic image construction module is used to perform imaging processing on the ultrasonic signal based on the delay parameter and the synthetic aperture focusing of dual-ray coverage to construct an acoustic image reflecting the internal density and defects of the pile. The resistivity cloud map construction module is used to establish an apparent resistivity cloud map reflecting the slurry diffusion range and the uniformity of cement-soil distribution based on the apparent resistivity data. The integrity assessment module is used to extract and fuse features based on acoustic images and apparent resistivity cloud maps respectively, to obtain fused features with enhanced spatial consistency, and to assess the integrity of the pile based on the fused features. The correlation adjustment module is used to perform correlation analysis between the comprehensive integrity assessment results and construction parameters, and to optimize the construction parameters for identified defects.

Images