A pile hole diameter detection method for highway engineering construction

Abstract

The present application relates to the technical field of highway engineering pile foundation detection, in particular to a pile hole diameter detection method for highway engineering construction, comprising: a multi-degree-of-freedom detection holder is arranged at the pile hole opening, the holder is integrated with a structured light projector, a phased array ultrasonic probe and a temperature and humidity and pressure composite sensor, and a three-dimensional reference model of the pile hole is generated in advance according to design drawings and a geological report. The hole diameter is scanned in sections along the depth of the pile hole, the point cloud of the hole diameter, the ultrasonic echo signal of the hole wall and the sensing data of the environment in the hole are synchronously captured, the final hole diameter deviation field is obtained through point cloud model registration, ultrasonic signal correction and environmental disturbance compensation, the scanning parameters of the next depth are adaptively adjusted according to the deviation field, and the full-depth detection is completed in a loop. The method realizes synchronous acquisition of multi-source data and multi-level deviation correction closed-loop detection, and guarantees the accuracy of the pile hole diameter detection.

Description

Technical Field

[0001] This invention relates to the field of pile foundation testing technology in highway engineering, and in particular to a method for testing the diameter of pile holes used in highway engineering construction. Background Technology

[0002] Current methods for detecting the diameter of pile holes in highway engineering construction mostly employ single-method detection. Hole diameter profile, hole wall cracks, and internal environmental parameters are detected independently using different equipment. The detection equipment is not integrated, and the detection pan-tilt unit can only perform detection actions along a fixed path, failing to achieve synchronous and collaborative scanning of multiple sensor types. The existing detection process only performs a simple comparison between single-hole diameter data and the design model, without incorporating ultrasonic testing data of the hole wall and internal temperature, humidity, and pressure environmental data into the analysis and correction of hole diameter deviations. This prevents the formation of a multi-source data fusion system for calculating hole diameter deviations.

[0003] Existing discrete testing equipment suffers from inconsistent spatiotemporal references in its data collection. Precise matching of borehole diameter, crack, and environmental data at different depths is impossible. Structural deviations caused by internal borehole cracks and disturbances from microclimate changes within the borehole cannot be effectively eliminated, resulting in significant errors in borehole diameter deviation calculations. Furthermore, the scanning focus position and parameters of the testing gimbal cannot be dynamically adjusted based on real-time testing results, making it difficult to adapt to the morphological changes at different depths of the pile borehole. This invention integrates multiple types of sensors onto a single multi-degree-of-freedom testing gimbal to achieve segment-by-segment synchronous and collaborative scanning along the pile borehole depth direction. It constructs a multi-level data correction and environmental compensation borehole deviation field and adaptively adjusts scanning parameters based on the real-time deviation field to complete the full-depth testing of the pile borehole. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and propose a method for detecting the diameter of pile holes used in highway engineering construction.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for detecting the diameter of pile holes used in highway engineering construction, comprising:

[0006] A multi-degree-of-freedom detection platform is installed at the opening of the pile hole. The detection platform integrates a structured light projector for scanning the hole diameter profile, a phased array ultrasonic probe for measuring internal cracks in the hole wall, and a temperature, humidity and pressure composite sensor for sensing the microclimate inside the pile hole.

[0007] Based on the design drawings and geological exploration reports of the pile holes, a three-dimensional reference model of the pile holes is pre-generated in virtual space;

[0008] The detection pan-tilt unit is activated, and a collaborative scan is performed segment by segment along the depth direction of the pile hole, simultaneously capturing the hole diameter point cloud, the ultrasonic echo signal of the hole wall, and the sensor data of the hole environment at each depth.

[0009] The captured aperture point cloud is registered in real time with the corresponding cross section of the pile hole 3D reference model, the geometric deviation between the point cloud and the model cross section contour is calculated, and a preliminary aperture deviation field is generated.

[0010] The initial aperture deviation field is corrected by introducing the ultrasonic echo signal of the aperture wall captured by the phased array ultrasonic probe to generate a fused aperture deviation field.

[0011] Based on the environmental sensing data inside the hole sensed by the temperature, humidity and pressure composite sensor, environmental disturbance compensation is performed on the fused aperture deviation field to generate the final aperture deviation field.

[0012] Based on the final aperture deviation field, the driving detection gimbal automatically adjusts the focus position and scanning parameters of the next depth scanning area, and repeats the collaborative scanning and correction compensation process until the full depth of the pile hole is detected.

[0013] As a further aspect of the present invention, the method for generating the three-dimensional reference model of the pile hole is as follows:

[0014] Obtain the design parameters of the pile hole, including the design diameter, design depth, design verticality, and design hole wall inclination angle;

[0015] Obtain the geological survey report of the pile hole construction area and extract the type, density, internal friction angle, cohesion and moisture content range of soil layers at different depths;

[0016] Input the design parameters and the data from the geological survey report into the physical simulation engine;

[0017] The physical simulation engine simulates the stable borehole wall morphology of soil layers at different depths under the action of self-weight and lateral earth pressure in an undisturbed state, based on the input soil layer parameters.

[0018] By combining the designed borehole diameter with the simulated stable borehole wall morphology, a continuous three-dimensional surface model of the pile hole along the depth direction is generated;

[0019] The three-dimensional curved surface model is divided into a dense grid, and each grid cell is assigned a physical property label for the corresponding soil layer at its depth, thus forming the three-dimensional reference model of the pile hole.

[0020] As a further aspect of the present invention, the simultaneous acquisition of aperture point cloud, borehole wall ultrasonic echo signal, and borehole environment sensing data at each depth specifically includes:

[0021] Control the detection pan-tilt unit to descend along a preset trajectory to a specified depth and maintain stability;

[0022] The structured light projector projects an coded grating pattern onto the hole wall at the current depth, while the integrated camera simultaneously acquires the deformed grating image modulated by the hole wall.

[0023] The acquired deformed grating image is decoded and phase calculated to reconstruct the three-dimensional point coordinate set of the hole wall surface at the current depth, forming the aperture point cloud at the depth.

[0024] At the same time as acquiring the grating pattern, the phased array ultrasonic probe is driven to emit an array of ultrasonic beams of a specific frequency and to receive ultrasonic echo signals reflected from inside the aperture wall and the interface.

[0025] Simultaneously read the temperature, humidity and air pressure values ​​detected by the temperature, humidity and air pressure composite sensor at the current depth and time, and package them into an in-hole environment sensing data package;

[0026] The aperture point cloud, the ultrasonic echo signal, and the in-aperture environment sensing data packet at the same timestamp are associated and stored.

[0027] As a further aspect of the present invention, the calculation of the geometric deviation between the point cloud and the model cross-sectional profile to generate a preliminary aperture deviation field involves the following steps:

[0028] Extract the theoretical cross-sectional contour line that is exactly the same as the current detection depth from the three-dimensional reference model of the pile hole;

[0029] The reconstructed current depth aperture point cloud is projected onto the two-dimensional plane containing the theoretical cross-sectional profile.

[0030] In the two-dimensional plane, taking the points on the theoretical cross-sectional contour line as a reference, the nearest projected point cloud point is searched along the normal direction, and the distance in the normal direction is calculated.

[0031] By traversing all points on the theoretical cross-sectional profile, a set of normal distances corresponding one-to-one with the profile points is obtained. The set of normal distances characterizes the degree of convexity or inward contraction of the actual hole wall relative to the theoretical profile.

[0032] Using the theoretical cross-sectional profile as a reference frame, the set of normal distances is mapped onto the two-dimensional plane in the form of a field to form the preliminary aperture deviation field, where the value of each point in the field represents the actual deviation of the profile position.

[0033] As a further aspect of the present invention, the preliminary aperture deviation field is corrected to generate a fused aperture deviation field, and the specific steps are as follows:

[0034] The ultrasonic echo signal captured by the phased array ultrasonic probe is processed to extract the amplitude attenuation characteristics and travel time characteristics of the signal;

[0035] Based on the amplitude attenuation characteristics and travel time characteristics, the average elastic modulus distribution and crack density distribution within a set depth range below the surface of the borehole wall are calculated by inversion.

[0036] Based on the average elastic modulus distribution obtained by inversion, an empirical mapping relationship between the elastic modulus and the actual yield strength of the pore wall material is established, and then the local yield strength distribution of the pore wall is estimated.

[0037] Spatially align the local yield strength distribution with the preliminary aperture deviation field;

[0038] Based on the spatially aligned data, each deviation value in the preliminary aperture deviation field is weighted and adjusted, where the weight coefficient is determined by the local yield strength at the corresponding position. The lower the yield strength of the region, the higher the confidence weight is given to the apparent geometric deviation in the weighting adjustment.

[0039] After weighted adjustment, a new deviation field is obtained, which is the fused aperture deviation field, and it simultaneously reflects the weakening information of geometric size deviation and the mechanical properties of the aperture wall material.

[0040] As a further aspect of the present invention, environmental disturbance compensation is performed on the fused aperture deviation field to generate the final aperture deviation field. The specific steps are as follows:

[0041] Read the environmental sensor data packet inside the hole associated with the current scanning depth and parse out the temperature, humidity and air pressure values;

[0042] Based on the temperature value, consult the material thermal expansion coefficient table to calculate the theoretical change in the borehole diameter caused by the thermal expansion and cooling effect of the rock and soil in the borehole wall at the current temperature.

[0043] Based on the humidity value, consult the model of the relationship between the water content and volume change of the soil and rock mass, and calculate the theoretical change in the borehole diameter caused by the moisture absorption or drying effect of the soil and rock mass on the borehole wall under the current humidity.

[0044] Based on the air pressure value and the porosity parameters of the rock and soil in the borehole wall, the small elastic deformation caused by the pressure difference between the inside and outside of the borehole wall is calculated.

[0045] The theoretical change in aperture caused by the thermal expansion and cooling effect, the theoretical change in aperture caused by the moisture absorption or drying effect, and the small elastic deformation are vector-synthesized to obtain the total environmental disturbance compensation vector field acting on aperture deviation under the current environmental conditions.

[0046] The final aperture deviation field is obtained by subtracting the total environmental disturbance compensation vector field from the fused aperture deviation field to eliminate the influence of environmental factors on the aperture measurement results.

[0047] As a further aspect of the present invention, the driving detection gimbal automatically adjusts the focus position and scanning parameters of the next depth scanning area, specifically including:

[0048] Analyze the spatial distribution characteristics of the final aperture deviation field to identify local abnormal regions where the deviation value exceeds the warning threshold;

[0049] Based on the area, deviation gradient, and spatial location of the local anomaly region, calculate the pitch angle, yaw angle, and focus distance that the detection gimbal needs to adjust for the next depth scan.

[0050] Based on the overall uniformity of the final aperture deviation field, the projection pattern density of the structured light projector and the scanning frequency of the phased array ultrasonic probe are dynamically adjusted.

[0051] The calculated pitch angle, yaw angle, focus distance, and the adjusted projection pattern density and scanning frequency are packaged together into the next scanning instruction set;

[0052] The next scan instruction set is sent to the motion controller and sensor controller of the detection pan-tilt unit, driving the detection pan-tilt unit and integrated sensor to perform adjustments.

[0053] As a further aspect of the present invention, after completing the full-depth detection of the pile hole, the method further includes:

[0054] All the final aperture deviation fields continuously acquired along the depth direction are stacked in the order of their corresponding detection depths to construct a complete body model of the measured aperture deviation of the pile hole in three-dimensional space.

[0055] The measured diameter deviation model of the pile hole is compared with the three-dimensional reference model of the pile hole in a unified coordinate system.

[0056] The difference between each voxel in the measured borehole diameter deviation volumetric model and the design value at the corresponding position in the three-dimensional reference model of the borehole is calculated to generate a three-dimensional deviation map of the entire borehole.

[0057] As a further aspect of the present invention, based on the full-hole three-dimensional deviation map, the following processing is performed:

[0058] In the full-hole three-dimensional deviation map, a series of horizontal sections are extracted along the depth direction, and the average deviation, maximum deviation and deviation distribution variance of each section are calculated.

[0059] The average deviation, maximum deviation, and variance of deviation distribution of each section are correlated with the design depth of the section and the corresponding soil layer type to form a section deviation characteristic sequence;

[0060] Based on the cross-sectional deviation characteristic sequence of all cross sections, the trend decomposition method is used to separate the characteristics of the deviation trend component caused by systematic construction factors and the deviation residual component caused by random disturbances.

[0061] The characteristics of the deviation trend components are fed back into the generation process of the three-dimensional reference model of the pile hole to optimize the simulation settings of relevant soil layer parameters in the physical simulation engine.

[0062] As a further aspect of the present invention, the simulation settings of relevant soil layer parameters in the optimized physical simulation engine are specifically as follows:

[0063] From the cross-sectional deviation feature sequence, depth ranges in which the deviation trend component features show regular changes are identified;

[0064] Match the depth range with the soil layers in the geological survey report to identify the specific soil layers for which parameters need to be optimized;

[0065] Based on the characteristics of the deviation trend components in the corresponding soil layers, the simulated values ​​of soil density, internal friction angle, or cohesion in the physical simulation engine are adjusted in reverse.

[0066] Using the adjusted soil parameters, the physical simulation engine was rerun to generate an updated 3D reference model of the pile hole.

[0067] The updated 3D reference model for pile holes will serve as a new benchmark model for real-time registration and deviation calculation when conducting borehole diameter testing on subsequent pile holes of the same type or subsequent pile holes in the same project.

[0068] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0069] The multi-degree-of-freedom detection gimbal integrates a structured light projector, a phased array ultrasonic probe, and a temperature, humidity, and pressure composite sensor. It performs collaborative scanning operations segment by segment along the depth direction of the pile hole, and can simultaneously acquire hole diameter point cloud data, hole wall ultrasonic echo signals, and pile hole internal environment sensing data at the same cross-sectional location and during the same acquisition period. This ensures that hole diameter contour information, hole wall internal crack information, and hole temperature, humidity, and pressure parameters are under a unified spatiotemporal reference, eliminating the spatiotemporal misalignment problem caused by step-by-step acquisition by different devices. The three types of detection data can directly participate in the calculation and correction process of the hole diameter deviation field, reducing the analysis bias caused by data mismatch and ensuring the spatiotemporal consistency and correspondence of the detection data.

[0070] A preliminary aperture deviation field is obtained by real-time registration of the aperture point cloud with the corresponding cross-section of the pile hole 3D reference model. Ultrasonic echo signals from the borehole wall are then introduced to correct this preliminary aperture deviation field, forming a fused aperture deviation field. Environmental disturbance compensation is then implemented using environmental sensing data collected by a temperature, humidity, and pressure composite sensor to obtain the final aperture deviation field. This process eliminates data interference caused by internal borehole wall cracks and microclimate changes within the borehole, making the aperture deviation calculation results more closely match the actual structural morphology of the pile hole. Based on the final aperture deviation field, the detection pan-tilt unit automatically adjusts the focusing position and scanning parameters for the next depth scanning area. This allows the scanning focus state and data acquisition parameters to adapt to the structural changes at different depths of the pile hole, ensuring that the scanning parameters match the actual working conditions of the pile hole. This avoids focusing offset and data acquisition gaps caused by fixed scanning parameters, achieving real-time linkage between detection parameters and detection results, and guaranteeing the authenticity and reliability of the full-depth aperture detection results for the pile hole. Attached Figure Description

[0071] Figure 1 This is a flowchart of a method for detecting the diameter of pile holes used in highway engineering construction, as described in this invention.

[0072] Figure 2 A flowchart for the synchronous capture and associated storage of collaborative scan data;

[0073] Figure 3 This is a sequence diagram of the full-hole depth deviation characteristics;

[0074] Figure 4 A diagram illustrating the environmental disturbance compensation process for integrating the aperture deviation field;

[0075] Figure 5 This is a three-dimensional deviation map of the entire borehole. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0077] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0078] See Figure 1 This invention provides a method for detecting the diameter of pile holes used in highway engineering construction, the specific method including:

[0079] A multi-degree-of-freedom (DOF) detection platform is erected at the borehole opening. This platform integrates a structured light projector, a phased array ultrasonic probe, and a temperature, humidity, and pressure composite sensor. The structured light projector is used to scan and reconstruct the borehole profile, the phased array ultrasonic probe is used to measure the internal structure of the borehole wall, and the temperature, humidity, and pressure composite sensor is used to sense the temperature, humidity, and air pressure microclimate data within the borehole. Before the detection, a three-dimensional reference model of the borehole needs to be pre-generated in virtual space based on the borehole design drawings and geological exploration report. During the detection, the detection platform is activated and descends segment by segment along the depth direction of the borehole, performing a collaborative scan. At each stopping depth, the borehole point cloud data, borehole wall ultrasonic echo signal, and borehole environment sensing data are simultaneously captured. The real-time captured borehole point cloud is registered with the theoretical cross-section of the three-dimensional reference model of the borehole at that depth, and the geometric deviation of their profiles is calculated to generate a preliminary borehole deviation field. The borehole wall material information reflected by the ultrasonic echo signal captured by the phased array ultrasonic probe is introduced to correct the preliminary borehole deviation field, generating a fused borehole deviation field that incorporates the material's mechanical properties. Environmental sensor data captured by a temperature, humidity, and pressure composite sensor is used to compensate for environmental disturbances in the fused aperture deviation field, eliminating measurement errors caused by temperature, humidity, and pressure variations, thus obtaining the final aperture deviation field. This final aperture deviation field is not only used to evaluate the construction quality of the current depth section, but also to drive the detection pan-tilt unit to automatically adjust the focus position and scanning parameters of the next depth scanning area, such as adjusting the sensor's pitch angle or scanning frequency. The detection pan-tilt unit moves to the next depth, repeating the entire process from collaborative scanning to generating the final aperture deviation field, until adaptive closed-loop detection of the entire pile hole depth is completed.

[0080] In one embodiment of the present invention, the generation of a three-dimensional reference model for the pile borehole requires obtaining the design parameters of the pile borehole and a geological survey report of the construction area. The design parameters include the design borehole diameter, design depth, design verticality, and design borehole wall inclination angle. The geological survey report needs to provide the type, density, internal friction angle, cohesion, and water content range of soil layers at different depths. The design parameters and the data from the geological survey report are input into a physical simulation engine. Based on the input soil layer physical and mechanical parameters, the physical simulation engine simulates the stable borehole wall morphology that soil layers at different depths can maintain under their own weight and lateral earth pressure without construction disturbance. Combining the design borehole diameter and the simulated stable borehole wall morphology, a three-dimensional curved surface model of the pile borehole that continuously varies along the depth direction is generated. This three-dimensional curved surface model is then meshed, and each mesh cell is assigned a physical attribute label corresponding to the soil layer at its corresponding depth, ultimately forming a three-dimensional reference model of the pile borehole with multiple physical attribute information.

[0081] The specific steps for synchronously capturing data during the detection process are as follows: (See attached document) Figure 2 The system controls the detection gimbal to descend along a preset trajectory to a specified depth and maintain stable hovering. A structured light projector projects an coded grating pattern onto the borehole wall at the current depth. A camera integrated on the gimbal simultaneously acquires the deformed grating image modulated by the borehole wall surface. The acquired deformed grating image is decoded and its phase is calculated to reconstruct the three-dimensional point coordinate set of the borehole wall surface at the current depth, forming the borehole point cloud at that depth. Simultaneously with the acquisition of the grating pattern, a phased array ultrasonic probe emits an array of ultrasonic beams at a specific frequency and receives ultrasonic echo signals reflected from inside the borehole wall and from interfaces between different materials. The temperature, humidity, and pressure values ​​detected by the temperature, humidity, and pressure composite sensor at the current depth and time are simultaneously read and packaged into a borehole environment sensing data package. The borehole point cloud, ultrasonic echo signals, and environment sensing data package, all with the same timestamp, are associated and stored.

[0082] In practical implementation, the generation of the 3D reference model for the pile borehole relies on accurate design input and geological data. The design parameters for the pile borehole are obtained, including the design diameter, design depth, design verticality, and design borehole wall inclination angle. A detailed geological survey report of the construction area is obtained, extracting information on soil type, density, internal friction angle, cohesion, and moisture content range at different depths. The complete design parameters and soil layer data from the geological survey report are input into a pre-defined physical simulation engine software module. Based on the input soil physical and mechanical parameters, the physical simulation engine simulates the mechanical equilibrium of soil layers at different depths under self-weight stress and lateral earth pressure in the absence of external construction disturbance, thereby calculating the stable borehole wall profile that each soil layer can maintain. Combining the designed borehole diameter and the stable borehole wall profiles simulated by the physical simulation engine, a smooth 3D surface model of the pile borehole that continuously varies along the depth direction is generated using a 3D surface modeling algorithm. The generated 3D surface model is meshed, dividing the surface into a series of dense triangular or quadrilateral mesh elements. Each mesh element is assigned a physical attribute label corresponding to the soil layer at its depth. These physical attribute labels include at least soil type, density, internal friction angle, and cohesion, ultimately forming a 3D reference model of the pile hole with multiple physical attribute labels. The assignment of physical attribute labels in the 3D reference model of the pile hole follows a mapping relationship, expressed by the formula:

[0083]

[0084] in: Indicates the depth The A set of physical attribute labels for each grid cell, function Represents the mapping rules. Indicates depth in the geological survey report The soil layer type defined at the location, This represents a set of physical and mechanical parameters associated with this soil type.

[0085] In practice, the simultaneous acquisition of borehole point cloud, borehole wall ultrasonic echo signals, and borehole environment sensing data is a process with strict time alignment. The multi-degree-of-freedom (DOF) detection platform is controlled to descend along a preset vertical trajectory along the central axis of the pile borehole to a specified depth defined by the detection plan. After reaching the specified depth, the multi-DOF detection platform is locked by a braking mechanism to maintain spatial pose stability. A structured light projector integrated on the detection platform projects an coded grating pattern onto the circumferential region of the borehole wall at the current depth. A coaxially integrated industrial camera simultaneously acquires the grating image, which is deformed after modulation by the borehole wall surface morphology. The single or multiple deformed grating images acquired by the industrial camera are decoded and phase-calculated. A 3D reconstruction algorithm is used to calculate the 3D spatial coordinate set of the sampling points on the borehole wall surface at the current depth, forming the borehole point cloud data for the current depth. At the same moment the structured light projector projects the pattern and the industrial camera acquires the image, the phased array ultrasonic probe is triggered to emit an array of ultrasonic beams with a specific center frequency and receive ultrasonic echo signals reflected from the borehole wall internal medium, defect interfaces, and the borehole wall-soil interface. Simultaneously, the real-time output of the temperature, humidity, and pressure composite sensor is read. The sensor output includes the temperature, humidity, and air pressure values ​​at the current depth. These three values ​​are packaged into a structured borehole environment sensing data package. The aperture point cloud data, ultrasonic echo signal data stream, and borehole environment sensing data package, all with the same millisecond-level timestamp, generated in the above process, are transmitted to the processing unit via a data bus. The processing unit associates and stores them in a single data frame based on their timestamps.

[0086] In some embodiments, the physical simulation engine operates based on the finite element method, treating the soil layer as an elasto-plastic material. The engine iteratively calculates the displacement field of the borehole wall under gravitational and initial geostress conditions, outputting the final stable borehole wall morphology when the displacement field stabilizes. In some embodiments, the structured light projector employs digital grating projection technology, where the coded grating pattern is a sinusoidally modulated grayscale grating or a binary coded grating. After an industrial camera acquires a single frame of deformed grating image, a 3D point cloud is reconstructed using a phase-shifting algorithm or Fourier transform contouring. Alternatively, after the industrial camera acquires multiple frames of grating images with a fixed phase difference, a multi-step phase-shifting algorithm is used to reconstruct the 3D point cloud. It can be understood that the focusing depth and scanning sector angle of the ultrasonic beam array emitted by the phased array ultrasonic probe are adjustable, and the scanning plane of the ultrasonic beam array coincides with or is parallel to the projection plane of the structured light projector in space. It can be understood that the temperature, humidity and pressure composite sensor includes a temperature sensor, a humidity sensor and a pressure sensor. The temperature sensor, humidity sensor and pressure sensor are packaged in the same probe. The temperature, humidity and pressure composite sensor transmits the sensing data in real time through wired or wireless means.

[0087] In one embodiment of the present invention, the geometric deviation between the point cloud and the model cross-sectional profile is calculated to generate a preliminary aperture deviation field. The steps include: extracting a theoretical cross-sectional profile that is identical to the current detection depth from a pre-generated three-dimensional reference model of the pile hole; projecting the aperture point cloud at the current depth obtained from on-site reconstruction onto a two-dimensional plane containing the theoretical cross-sectional profile; using each point on the theoretical cross-sectional profile as a reference point, searching for the nearest projected point cloud point along the normal direction of the profile, and calculating the distance between the two points in the normal direction, where a positive distance value indicates outward convexity of the actual hole wall, and a negative value indicates inward contraction; traversing all points on the theoretical cross-sectional profile yields a set of normal distances corresponding one-to-one with each profile point, representing the overall deviation of the actual hole wall profile from the theoretical profile; using the theoretical cross-sectional profile as a reference frame, mapping the set of normal distances onto the plane in the form of a two-dimensional field, where the value of each point in the field represents the actual geometric deviation at the profile position, thereby forming the preliminary aperture deviation field.

[0088] In practice, generating the initial aperture deviation field begins with extracting the theoretical profile from the 3D reference model of the pile hole. From the pre-generated 3D reference model, a horizontal theoretical cross-sectional profile is extracted based on the current depth value of the detection platform. The theoretical cross-sectional profile is a closed spatial curve formed by the intersection of the 3D reference model of the pile hole at that depth and the horizontal plane, defined by a series of ordered discrete point coordinate sequences. The aperture point cloud at the current depth, reconstructed on-site using structured light scanning, is projected onto the 2D plane containing the theoretical cross-sectional profile through coordinate transformation. This 2D plane is typically a plane with the pile hole design center axis as its normal. Within this 2D plane, using the coordinates of each discrete point on the theoretical cross-sectional profile as a reference point, the unit normal vector of the theoretical cross-sectional profile at that point is calculated. The nearest projected point cloud point is searched along both the positive and negative directions of this unit normal vector, and the distance between the reference point and the found nearest projected point cloud point along the direction of the unit normal vector is calculated. By traversing all discrete points on the theoretical cross-sectional profile, and repeatedly performing the nearest point search and normal distance calculation along the unit normal vector direction for each point, a set of normal distances corresponding one-to-one with the theoretical cross-sectional profile points is obtained. Each value in the set of normal distances represents the degree of outward convexity or inward contraction of the actual hole wall relative to the theoretical profile at the corresponding profile point position; positive numbers indicate that the actual hole wall convexes outward, and negative numbers indicate that the actual hole wall contracts inward.

[0089] In practical implementation, the set of normal distances is mapped to a two-dimensional preliminary aperture deviation field. Using the theoretical cross-sectional profile as a spatial reference frame, the calculated set of normal distances is mapped onto the plane containing the theoretical cross-sectional profile as a two-dimensional scalar field. The mapping relationship is that the coordinates of each discrete point of the theoretical profile on the plane correspond to a grid node in the deviation field, and the value stored in this grid node is the normal distance corresponding to that point. For the region between the theoretical cross-sectional profiles, the deviation value of the grid nodes is calculated using an interpolation algorithm, including bilinear interpolation or nearest neighbor interpolation. Finally, a continuous scalar field is formed on the two-dimensional plane; this scalar field is the preliminary aperture deviation field, and the value of each point in the preliminary aperture deviation field represents the actual geometric deviation of that profile position. The formula for calculating the deviation is expressed as:

[0090]

[0091] in: Indicates the first Normal deviation distance at each theoretical profile point Indicates the first The coordinate vectors of the theoretical contour points This indicates the distance found in the direction of the normal. The coordinate vectors of the most recently measured point cloud points, Indicates in The unit outward normal vector of the point-processing contour is represented by the symbol "·", which indicates the dot product operation of the vectors.

[0092] In some embodiments, the process of searching for the nearest projected point cloud is accelerated using a spatial index structure, including a KD tree or an octree. Calculating the unit normal vector requires parametric fitting of the theoretical cross-sectional profile. This parametric fitting uses cubic spline curve fitting, and the tangent direction at each point is obtained by differentiating the fitted spline curve function, thus calculating the normal direction. In some embodiments, the theoretical cross-sectional profile is pre-discretely stored in the 3D reference model of the pile hole, and the sequence of discrete point coordinates of the theoretical cross-sectional profile is consistent with the vertex data of the model mesh. It can be understood that the two-dimensional plane is defined as perpendicular to the pile hole design axis. When the pile hole has a design inclination, the two-dimensional plane is a plane perpendicular to the tangent of the pile hole design centerline at that depth. It can also be understood that the mesh resolution of the preliminary aperture deviation field is pre-set, and the mesh resolution is higher than the density of the discrete points of the theoretical profile to ensure the continuity of the deviation field representation.

[0093] In one embodiment of the present invention, the preliminary aperture deviation field is corrected to generate a fused aperture deviation field, which is specifically implemented as follows: The ultrasonic echo signal captured by the phased array ultrasonic probe is processed to extract the amplitude attenuation characteristics and ultrasonic travel time characteristics from the signal. Based on the extracted amplitude attenuation characteristics and travel time characteristics, the average elastic modulus distribution and crack density distribution within a set depth range below the surface of the aperture wall are calculated by inversion. Based on the average elastic modulus distribution obtained by inversion, the local yield strength distribution of the aperture wall is estimated through an empirical mapping relationship. This local yield strength distribution is spatially aligned and matched with the preliminary aperture deviation field. Based on the spatially aligned data, each deviation value in the preliminary aperture deviation field is weighted and adjusted, with the weight coefficient determined by the local yield strength corresponding to that location. In regions with lower yield strength, it indicates that the mechanical properties of the aperture wall material are weakened, and its geometric deformation may be more significant or more unstable. Therefore, a higher confidence weight is assigned to the geometric deviation at that location in the weighted adjustment. After this weighted adjustment process, a new deviation field is obtained, which is the fused aperture deviation field. It simultaneously reflects the geometric deviation of the aperture and the weakening information of the mechanical properties of the aperture wall material.

[0094] In practical implementation, the processing of ultrasonic echo signals begins with feature extraction. The raw ultrasonic echo signals captured by the phased array ultrasonic probe are filtered and gain-compensated. Two key features are extracted from the processed signal: amplitude attenuation and travel time. Amplitude attenuation refers to the rate at which the signal amplitude decreases as the ultrasonic wave propagates through the material, while travel time refers to the time difference between the ultrasonic wave emission and reception at a specific interface. Based on the extracted amplitude attenuation and travel time features, and combined with the ultrasonic propagation model in porous media, the average elastic modulus distribution and fracture density distribution within a predetermined depth range below the surface of the pore wall are calculated. The average elastic modulus distribution represents the stiffness characteristics of the material in different regions of the pore wall, while the fracture density distribution represents the degree of fracture development within the material. Based on the inverted average elastic modulus distribution, an empirical mapping relationship between the average elastic modulus and the actual yield strength of the pore wall material is established by querying a pre-established mapping table, thereby estimating the local yield strength distribution at various locations on the pore wall. The local yield strength distribution is represented in two-dimensional matrix form, where the value of each element in the matrix represents the estimated yield strength of the material at the corresponding spatial coordinate point.

[0095] In practice, the mechanical property field and the geometric deviation field are fused. The local yield strength distribution is aligned with the preliminary aperture deviation field obtained from the same depth section in spatial coordinates. This alignment ensures that each element of the local yield strength distribution matrix and the corresponding element of the preliminary aperture deviation field matrix represent the same position of the aperture wall. Based on the spatially aligned data, each deviation value in the preliminary aperture deviation field is weighted and adjusted. The weighting coefficients used for this adjustment are determined by the local yield strength value corresponding to that position. The rule for weighting is that in regions with lower local yield strength, the apparent geometric deviation is assigned a higher confidence weight in the weighting. After weighted adjustment, a new two-dimensional deviation field is obtained. This new two-dimensional deviation field is the fused aperture deviation field, which simultaneously contains information on the geometric size deviation of the aperture and the weakening information on the mechanical properties of the aperture wall material. The formula for weighted adjustment is expressed as:

[0096]

[0097] in: This represents the deviation value after merging at coordinates (x, y). This represents the initial deviation value of the preliminary aperture deviation field at coordinates (x, y). This represents the estimated local yield strength value at coordinates (x, y). It is a weighting function, and its output is a weighting coefficient calculated based on the local yield strength value.

[0098] In some embodiments, ultrasonic signal inversion calculations employ tomographic imaging algorithms, dividing the borehole wall scanning area into a grid. The average elastic modulus and fracture density of each grid are reconstructed by solving a series of acoustic equations. Empirical mapping relationships are stored in tabular form, recording the relationship between the average elastic modulus range and the corresponding yield strength range for different geological formations. Weighting function. The specific form is a piecewise linear function, whose function value varies with the local yield strength. The weighting factor increases as the yield strength decreases. In some embodiments, it can be understood that in regions where the local yield strength is higher than a set threshold, the weighting factor can be set to zero, meaning no correction is performed; in regions where the local yield strength is lower than the set threshold, the weighting factor is positive. It can be understood that the spatial alignment of the local yield strength distribution with the preliminary aperture deviation field depends on the unified coordinate calibration of the detection pan-tilt unit to ensure that the coordinate systems of structured light scanning and ultrasonic scanning are consistent. See Table 1.

[0099] Table 1: Example of mapping relationship between local yield strength range and weighting coefficient

[0100]

[0101] See Figure 3 This is a sequence diagram of the overall borehole depth deviation, used to assess the variation of pile hole construction quality along the depth direction. The average deviation gradually increases with depth, starting from approximately 4.2 mm at the borehole opening (0 m), reaching approximately 10 mm at a depth of 20 m, showing a trend of slow increase with local fluctuations. This reflects the overall expansion of the actual borehole diameter relative to the design value as depth increases. The maximum deviation rises rapidly from approximately 11.7 mm at the borehole opening to approximately 28 mm at a depth of 20 m, with a growth rate much higher than the average deviation. This indicates the existence of severe local over-diameter / under-diameter deviations in the deeper parts of the borehole, representing a key risk point for construction quality control. The deviation variance rises slowly from approximately 1.8 mm at the borehole opening to approximately 4.2 mm at a depth of 20 m, reflecting a slight increase in the dispersion of borehole deviation with depth, indicating that the uniformity of the borehole wall is slightly worse in deeper areas than in shallower areas. The maximum deviation curve can be used to quickly locate abnormal areas in the deeper parts of the borehole, guiding subsequent reinforcement construction.

[0102] In one embodiment of the present invention, environmental disturbance compensation is performed on the fused aperture deviation field. The process is as follows: The borehole environment sensor data packet associated with the current scanning depth is read, and the temperature, humidity, and air pressure values ​​are parsed. Based on the temperature value, a pre-stored material thermal expansion coefficient table is consulted to calculate the theoretical change in aperture caused by thermal expansion and contraction of the borehole wall soil at the current temperature. Based on the humidity value, a model relating soil moisture content to volume change is consulted to calculate the theoretical change in aperture caused by hygroscopic or drying effects of the borehole wall soil at the current humidity. Based on the air pressure value, combined with the porosity parameter of the borehole wall soil, the small elastic deformation caused by the pressure difference between the inside and outside of the pile hole is calculated. The calculated theoretical changes in thermal expansion and contraction, hygroscopic drying, and elastic deformation caused by the pressure difference are vector-synthesized to obtain the total environmental disturbance compensation vector field acting on aperture measurement under the current environmental conditions. This total environmental disturbance compensation vector field is subtracted from the fused aperture deviation field to eliminate the influence of environmental factors on the measured aperture result; the resulting deviation field is the final aperture deviation field.

[0103] The detection gimbal automatically adjusts its focus position and scanning parameters for the next depth scan area by analyzing the spatial distribution characteristics of the final aperture deviation field and identifying local anomaly areas where the deviation value exceeds a preset warning threshold. Based on the size of the identified local anomaly areas, the trend of deviation gradient changes, and their spatial location, the gimbal calculates the pitch angle, yaw angle, and focus distance that need to be adjusted for the next depth scan. Simultaneously, based on the overall uniformity of the final aperture deviation field, the projection pattern density of the structured light projector and the scanning frequency of the phased array ultrasonic probe are dynamically adjusted. The calculated pitch angle, yaw angle, focus distance, and the adjusted projection pattern density and scanning frequency parameters are packaged into a next scan command set. This command set is sent to the motion controller and sensor controllers of the detection gimbal, driving the detection gimbal and its integrated sensors to perform corresponding pose and parameter adjustments.

[0104] In practical implementation, environmental disturbance compensation for the fused aperture deviation field begins with the analysis of sensor data. The borehole environment sensor data package associated with the current scanning depth is read, and temperature, humidity, and air pressure values ​​are extracted from it. Based on the extracted temperature value, a pre-stored table of material thermal expansion coefficients is consulted. This table records the linear expansion coefficients of different soil and rock types within different temperature ranges. The theoretical change in aperture due to thermal expansion and contraction of the soil and rock at the current temperature is calculated based on the current borehole wall soil and rock type and temperature value. Based on the extracted humidity value, a pre-established model of the relationship between soil and rock moisture content and volume change is consulted. This model describes the quantitative relationship of volume expansion or contraction of a specific soil and rock during moisture absorption or drying. Based on this model, the theoretical change in aperture due to moisture absorption or drying of the soil and rock at the current humidity level is calculated. Based on the analyzed air pressure values ​​and the porosity parameters of the borehole wall soil obtained from the geological survey report, the air pressure difference between the internal air pressure and the external atmospheric pressure is calculated. Based on the theory of elasticity, the small elastic deformation caused by this air pressure difference to the borehole wall of the porous medium is calculated.

[0105] In practical implementation, the theoretical changes caused by various environmental factors are synthesized. The calculated theoretical changes due to thermal expansion and contraction, moisture absorption or drying, and the small elastic deformation caused by pressure difference are treated as vectors. Each vector has magnitude and direction, with the direction perpendicular to the orifice wall surface and pointing inwards or outwards. In a two-dimensional deviation field coordinate system, these three vectors are vector synthesized at each spatial grid point. The vector synthesis yields the total environmental disturbance compensation vector field acting on the aperture measurement under the current environmental conditions. The total environmental disturbance compensation vector field is subtracted from the fused aperture deviation field. The vector value of the total environmental disturbance compensation vector field at each grid point represents the measurement deviation caused by environmental factors. The subtraction operation eliminates the influence of environmental factors on the aperture measurement results. The resulting two-dimensional scalar field after the subtraction operation is the final aperture deviation field. The formula for calculating the total environmental disturbance compensation vector field is expressed as:

[0106]

[0107] in: This represents the total environmental disturbance compensation vector at coordinates (x, y). Indicated by temperature The determined thermal expansion and contraction compensation vector, Indicates humidity The determined moisture absorption and drying effect compensation vector, Indicated by air pressure and porosity The determined pressure deformation effect compensation vector.

[0108] In some embodiments, the driver pan-tilt unit automatically adjusts the focus position and scanning parameters of the next depth scanning area based on real-time analysis of the final aperture deviation field. The spatial distribution characteristics of the final aperture deviation field are analyzed to identify local anomaly regions where the deviation value exceeds a preset warning threshold. A local anomaly region refers to an area where the deviation value of a continuous grid point is greater than the positive threshold or less than the negative threshold. Based on the size of the identified local anomaly region, the deviation gradient change at the region's edge, and the region's spatial coordinates in the two-dimensional field, the pitch angle, yaw angle, and focus distance that the pan-tilt unit needs to adjust for the next depth scan are calculated. The projection pattern density of the structured light projector and the scanning frequency of the phased array ultrasonic probe are dynamically adjusted based on the overall uniformity of the final aperture deviation field. The overall uniformity of the final aperture deviation field can be quantified by calculating the variance or standard deviation of the entire two-dimensional field. A larger variance indicates poorer uniformity, requiring a higher density projection pattern and a higher ultrasonic scanning frequency to obtain more detailed data. The calculated pitch angle, yaw angle, focus distance, and adjusted projection pattern density parameters, along with the ultrasonic scanning frequency parameters, are packaged together into a next scan instruction set. The next set of scanning instructions is sent to the motion controller and sensor controller of the detection pan-tilt unit through the communication interface. The motion controller drives the pan-tilt unit's robotic arm to adjust its posture according to the instructions, and the sensor controller adjusts the internal operating parameters of the structured light projector and the phased array ultrasonic probe according to the instructions.

[0109] Optionally, the material thermal expansion coefficient table can be refined for different soil layers, defining different thermal expansion coefficient values ​​for each soil layer. Optionally, the relationship between soil and rock water content and volume change can be described using a piecewise linear model or an exponential model. It can be understood that the directions of each component vector in the total environmental disturbance compensation vector field are defined as follows: when the environmental effect causes the aperture measurement value to be too large, the compensation vector direction points inward (negative direction); when the environmental effect causes the aperture measurement value to be too small, the compensation vector direction points outward (positive direction). It can be understood that when calculating the focusing distance, it is necessary to consider the extension trend of local abnormal areas in the depth direction, and combine the scanning data of the previous depth to predict the possible target area distance at the next depth. See Table 2.

[0110] Table 2: Correspondence between Environmental Parameter Changes and Environmental Disturbance Compensation Coefficients

[0111]

[0112] See Figure 4This is a diagram illustrating the environmental disturbance compensation process for the fused borehole diameter deviation field, used to quantify the correction effect of environmental factors on borehole diameter measurement results. The deviations in the 0-15m and 30-45m depth ranges are positive, indicating over-diameter boreholes, with a maximum over-diameter of approximately 2.2mm. The deviations in the 15-30m and 45-50m depth ranges are negative, indicating narrowing boreholes, with a maximum narrowing of approximately -2.2mm. The deviations across the entire depth exhibit periodic fluctuations, consistent with the actual engineering patterns of soil stratification and construction disturbances. The blue final deviation almost completely overlaps with the green fused deviation, with only a slight difference. This is because the environmental disturbance compensation is a constant small correction (approximately -0.2mm), used to eliminate the systematic influence of temperature, humidity, and pressure on soil deformation, ensuring the accuracy of the final deviation. A continuous and significant narrowing (maximum -2.2mm) occurs in the 20-30m depth range, representing a weak point in the borehole's bearing capacity, requiring a thorough investigation of soil creep and borehole collapse risks. Eliminating environmental interference provides accurate and reliable quantitative data for borehole construction acceptance.

[0113] In one embodiment of the present invention, after completing the full-depth detection of the pile hole, all the final hole diameter deviation fields continuously acquired along the depth direction are stacked according to their corresponding detection depth order to construct a complete pile hole measured hole diameter deviation volume model in three-dimensional space. This pile hole measured hole diameter deviation volume model is then compared and aligned with the pile hole three-dimensional reference model in a unified three-dimensional coordinate system. The difference between each voxel in the pile hole measured hole diameter deviation volume model and its corresponding design value in the pile hole three-dimensional reference model is calculated to generate a full-hole three-dimensional deviation map, which can three-dimensionally display the deviation distribution throughout the entire pile hole space.

[0114] Based on the generated full-hole 3D deviation map, further processing is performed. A series of horizontal sections are extracted along the depth direction from the full-hole 3D deviation map, and the average deviation, maximum deviation, and deviation distribution variance of each section are calculated. The average deviation, maximum deviation, and deviation distribution variance of each section are correlated with the design depth and corresponding soil layer type information of that section to form a section deviation feature sequence. Based on the section deviation feature sequences of all sections, a trend decomposition method is used to separate the characteristics of the deviation trend component caused by systematic construction factors and the deviation residual component caused by random disturbances.

[0115] The analyzed deviation trend component characteristics are fed back into the initial generation process of the pile borehole 3D reference model to optimize the simulation settings of relevant soil layer parameters in the physical simulation engine. The specific optimization process is as follows: From the cross-sectional deviation characteristic sequence, depth intervals where the deviation trend component characteristics exhibit regular changes are identified. These depth intervals are matched with the detailed soil layer stratification in the geological survey report to determine the specific soil layers requiring simulation parameter optimization. Based on the performance of the deviation trend component characteristics in the corresponding soil layer, the simulated density, internal friction angle, or cohesion values ​​of that soil layer in the physical simulation engine are adjusted in reverse. Using the adjusted soil layer parameters, the physical simulation engine is rerun to generate an updated pile borehole 3D reference model. This updated pile borehole 3D reference model will serve as a new benchmark model for real-time registration and deviation calculation when subsequent pile boreholes of the same type or in the same project undergo borehole diameter testing.

[0116] In practice, after completing the full-depth detection of the pile hole, all final borehole diameter deviation fields acquired along the depth direction are processed. These fields are then stacked according to their corresponding detection depth coordinates, from the borehole opening to the bottom. The stacking operation is performed in three-dimensional space, with each final borehole diameter deviation field serving as a two-dimensional layer. The spacing between layers is determined by the scanning depth interval. Data is filled between layers using a three-dimensional interpolation algorithm, thus constructing a complete volumetric model of the measured borehole diameter deviation. This model is a three-dimensional volumetric data set, where each voxel stores the deviation between the actual borehole diameter and the theoretical value at that spatial location. The volumetric model and the three-dimensional reference model of the pile hole (the theoretical model) are then imported into a unified three-dimensional coordinate system, and the two models are spatially overlaid and compared. The algebraic difference between the value of each voxel in the measured borehole diameter deviation volumetric model and the design value of the corresponding voxel in the three-dimensional reference model of the borehole is calculated. The design value is zero or the theoretical borehole diameter value. The calculated difference value is assigned to the corresponding position of a new three-dimensional volumetric data. This new three-dimensional volumetric data containing the difference value is the full-hole three-dimensional deviation map. The full-hole three-dimensional deviation map intuitively displays the construction deviation of each point in the three-dimensional space of the entire borehole.

[0117] In practical implementation, cross-sectional feature extraction and trend analysis are performed based on the full-hole 3D deviation map. A series of horizontal cross-sections are extracted at fixed intervals along the depth direction from the full-hole 3D deviation map; these cross-sections are planes perpendicular to the depth coordinates. The average deviation, maximum deviation, and variance of the deviation distribution for each extracted horizontal cross-section are calculated. These values ​​are then correlated with the design depth value of that cross-section and the soil layer type corresponding to that depth obtained from the geological survey report, forming a depth-ordered sequence of cross-sectional deviation features. Based on the cross-sectional deviation feature sequences of all cross-sections, a trend decomposition method is used to separate the deviation trend component caused by systematic construction factors and the deviation residual component caused by random disturbances. Trend decomposition can employ moving average, polynomial fitting, or filtering methods. The deviation trend component typically exhibits a slow curve that changes with depth, while the deviation residual component shows random fluctuations around zero.

[0118] In practice, the analyzed deviation trend component characteristics are fed back to the physical simulation engine to optimize simulation parameters. From the cross-sectional deviation characteristic sequence, depth intervals exhibiting regular monotonically increasing, decreasing, or periodic changes in the deviation trend component characteristics are identified. These depth intervals are matched with detailed soil layer data in the geological survey report to determine which soil layer(s) require optimization of simulation parameters. Based on the performance of the deviation trend component characteristics in the corresponding soil layer, the simulated density, internal friction angle, or cohesion values ​​of the soil layer input into the physical simulation engine are adjusted in reverse. Using the adjusted simulated density, internal friction angle, or cohesion values, the physical simulation engine is rerun. Based on the new parameters, the physical simulation engine recalculates the stable borehole wall morphology, generating an updated 3D reference model for the pile borehole. This updated 3D reference model will serve as a new benchmark model for real-time registration and deviation calculation when subsequent similar pile boreholes or subsequent pile boreholes in the same project undergo borehole diameter testing. The variance calculation formula for the cross-sectional deviation characteristic sequence is expressed as:

[0119]

[0120] in: Indicates the first The variance of the deviation distribution of each horizontal cross section Indicates the first The total number of effective voxels on each cross section Indicates the first On the first cross section The deviation value of individual elements, Indicates the first The average value of all voxel deviations on each cross section.

[0121] In some embodiments, the 3D interpolation algorithm employs trilinear interpolation or nearest-neighbor interpolation to ensure the continuity of the measured borehole diameter deviation volumetric model. Before the alignment comparison, the measured borehole diameter deviation volumetric model and the 3D reference model of the borehole need to be coordinate registered to ensure that the spatial origin and axis of the two models are aligned. In some embodiments, the cross-sectional deviation feature sequence is stored in a table or linked list data structure, where each element contains fields such as depth, soil type, average deviation, maximum deviation, and variance. It can be understood that the deviation trend component characteristics are expressed as a function related to soil properties or drilling rig parameters. For example, in clay layers, the deviation trend component may be positive and increase with depth, while in sandy soil layers, the deviation trend component may fluctuate around zero. It can be understood that adjusting soil parameters is an iterative process, comparing the updated 3D reference model of the borehole with the actual detection data, calculating new deviation trend components, until the deviation trend component characteristics are weakened to an acceptable threshold.

[0122] See Figure 5 This is a three-dimensional deviation map of the entire borehole, visually presenting the distribution of construction deviations throughout the borehole. From 0-10m, the map is predominantly blue / light blue, with deviations mostly negative (-4~0mm), indicating localized diameter reduction. The borehole opening area is significantly affected by construction disturbances. From 10-50m, the map is predominantly red / light red, with deviations mostly positive (0~10mm), showing an overall over-diameter state. The over-diameter increases slightly with depth, consistent with the engineering principles of increased soil self-weight stress and borehole wall creep. The deviations are randomly distributed circumferentially, without significant systematic circumferential deviations, indicating good verticality of the drilling rig and no skewness. Localized high-value over-diameter deviations (>8mm) appear near 20-30m depth and 350°, representing a risk point for borehole wall collapse and requiring close investigation. Continuous diameter reduction in the 0-5m borehole opening area may be due to borehole wall rebound after drilling and disturbance caused by drilling rig lifting; the load-bearing stability of the pile top needs to be monitored. It presents the deviation distribution of the pile hole at all depths and circumferences in one go, with no blind spots in the detection, providing a complete basis for construction acceptance.

[0123] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for detecting the diameter of pile holes used in highway engineering construction, characterized in that, include: A multi-degree-of-freedom detection platform is installed at the opening of the pile hole. The detection platform integrates a structured light projector for scanning the hole diameter profile, a phased array ultrasonic probe for measuring internal cracks in the hole wall, and a temperature, humidity and pressure composite sensor for sensing the microclimate inside the pile hole. Based on the design drawings and geological exploration reports of the pile holes, a three-dimensional reference model of the pile holes is pre-generated in virtual space; The detection pan-tilt unit is activated, and a collaborative scan is performed segment by segment along the depth direction of the pile hole, simultaneously capturing the hole diameter point cloud, the ultrasonic echo signal of the hole wall, and the sensor data of the hole environment at each depth. The captured aperture point cloud is registered in real time with the corresponding cross section of the pile hole 3D reference model, the geometric deviation between the point cloud and the model cross section contour is calculated, and a preliminary aperture deviation field is generated. The initial aperture deviation field is corrected by introducing the ultrasonic echo signal of the aperture wall captured by the phased array ultrasonic probe to generate a fused aperture deviation field. Based on the environmental sensing data inside the hole sensed by the temperature, humidity and pressure composite sensor, environmental disturbance compensation is performed on the fused aperture deviation field to generate the final aperture deviation field. Based on the final aperture deviation field, the driving detection gimbal automatically adjusts the focus position and scanning parameters of the next depth scanning area, and repeats the collaborative scanning and correction compensation process until the full depth of the pile hole is detected.

2. The method for detecting the diameter of pile holes used in highway engineering construction according to claim 1, characterized in that, The method for generating the three-dimensional reference model of the pile hole is as follows: Obtain the design parameters of the pile hole, including the design diameter, design depth, design verticality, and design hole wall inclination angle; Obtain the geological survey report of the pile hole construction area and extract the type, density, internal friction angle, cohesion and moisture content range of soil layers at different depths; Input the design parameters and the data from the geological survey report into the physical simulation engine; The physical simulation engine simulates the stable borehole wall morphology of soil layers at different depths under the action of self-weight and lateral earth pressure in an undisturbed state, based on the input soil layer parameters. By combining the designed borehole diameter with the simulated stable borehole wall morphology, a continuous three-dimensional surface model of the pile hole along the depth direction is generated; The three-dimensional curved surface model is divided into a dense grid, and each grid cell is assigned a physical property label for the corresponding soil layer at its depth, thus forming the three-dimensional reference model of the pile hole.

3. The method for detecting the diameter of pile holes used in highway engineering construction according to claim 2, characterized in that, The simultaneous acquisition of aperture point cloud, borehole wall ultrasonic echo signal, and borehole environment sensing data at each depth specifically includes: Control the detection pan-tilt unit to descend along a preset trajectory to a specified depth and maintain stability; The structured light projector projects an coded grating pattern onto the hole wall at the current depth, while the integrated camera simultaneously acquires the deformed grating image modulated by the hole wall. The acquired deformed grating image is decoded and phase calculated to reconstruct the three-dimensional point coordinate set of the hole wall surface at the current depth, forming the aperture point cloud at the depth. At the same time as acquiring the grating pattern, the phased array ultrasonic probe is driven to emit an array of ultrasonic beams of a specific frequency and to receive ultrasonic echo signals reflected from inside the aperture wall and the interface. Simultaneously read the temperature, humidity and air pressure values ​​detected by the temperature, humidity and air pressure composite sensor at the current depth and time, and package them into an in-hole environment sensing data package; The aperture point cloud, the ultrasonic echo signal, and the in-aperture environment sensing data packet at the same timestamp are associated and stored.

4. The method for detecting the diameter of pile holes used in highway engineering construction according to claim 3, characterized in that, The geometric deviation between the calculated point cloud and the model cross-sectional profile is used to generate a preliminary aperture deviation field. The specific steps are as follows: Extract the theoretical cross-sectional contour line that is exactly the same as the current detection depth from the three-dimensional reference model of the pile hole; The reconstructed current depth aperture point cloud is projected onto the two-dimensional plane containing the theoretical cross-sectional profile. In the two-dimensional plane, taking the points on the theoretical cross-sectional contour line as a reference, the nearest projected point cloud point is searched along the normal direction, and the distance in the normal direction is calculated. By traversing all points on the theoretical cross-sectional profile, a set of normal distances corresponding one-to-one with the profile points is obtained. The set of normal distances characterizes the degree of convexity or inward contraction of the actual hole wall relative to the theoretical profile. Using the theoretical cross-sectional profile as a reference frame, the set of normal distances is mapped onto the two-dimensional plane in the form of a field to form the preliminary aperture deviation field, where the value of each point in the field represents the actual deviation of the profile position.

5. The method for detecting the diameter of pile holes used in highway engineering construction according to claim 4, characterized in that, The preliminary aperture deviation field is corrected to generate a fused aperture deviation field. The specific steps are as follows: The ultrasonic echo signal captured by the phased array ultrasonic probe is processed to extract the amplitude attenuation characteristics and travel time characteristics of the signal; Based on the amplitude attenuation characteristics and travel time characteristics, the average elastic modulus distribution and crack density distribution within a set depth range below the surface of the borehole wall are calculated by inversion. Based on the average elastic modulus distribution obtained by inversion, an empirical mapping relationship between the elastic modulus and the actual yield strength of the hole wall material is established, and then the local yield strength distribution of the hole wall is estimated. Spatially align the local yield strength distribution with the preliminary aperture deviation field; Based on the spatially aligned data, each deviation value in the preliminary aperture deviation field is weighted and adjusted, where the weight coefficient is determined by the local yield strength at the corresponding position. The lower the yield strength of the region, the higher the confidence weight is given to the apparent geometric deviation in the weighting adjustment. After weighted adjustment, a new deviation field is obtained, which is the fused aperture deviation field, and it simultaneously reflects the weakening information of geometric size deviation and the mechanical properties of the aperture wall material.

6. The method for detecting the diameter of pile holes used in highway engineering construction according to claim 5, characterized in that, Environmental disturbance compensation is applied to the fused aperture deviation field to generate the final aperture deviation field. The specific steps are as follows: Read the environmental sensor data packet inside the hole associated with the current scanning depth and parse out the temperature, humidity and air pressure values; Based on the temperature value, consult the material thermal expansion coefficient table to calculate the theoretical change in the borehole diameter caused by the thermal expansion and cooling effect of the rock and soil in the borehole wall at the current temperature. Based on the humidity value, consult the model of the relationship between the water content and volume change of the soil and rock mass, and calculate the theoretical change in the borehole diameter caused by the moisture absorption or drying effect of the soil and rock mass on the borehole wall under the current humidity. Based on the air pressure value and the porosity parameters of the rock and soil in the borehole wall, the small elastic deformation caused by the pressure difference between the inside and outside of the borehole wall is calculated. The theoretical change in aperture caused by the thermal expansion and cooling effect, the theoretical change in aperture caused by the moisture absorption or drying effect, and the small elastic deformation are vector-synthesized to obtain the total environmental disturbance compensation vector field acting on aperture deviation under the current environmental conditions. The final aperture deviation field is obtained by subtracting the total environmental disturbance compensation vector field from the fused aperture deviation field to eliminate the influence of environmental factors on the aperture measurement results.

7. The method for detecting the diameter of pile holes used in highway engineering construction according to claim 6, characterized in that, The drive detection gimbal automatically adjusts the focus position and scanning parameters of the next depth scanning area, specifically including: Analyze the spatial distribution characteristics of the final aperture deviation field to identify local abnormal regions where the deviation value exceeds the warning threshold; Based on the area, deviation gradient, and spatial location of the local anomaly region, calculate the pitch angle, yaw angle, and focus distance that the detection gimbal needs to adjust for the next depth scan. Based on the overall uniformity of the final aperture deviation field, the projection pattern density of the structured light projector and the scanning frequency of the phased array ultrasonic probe are dynamically adjusted. The calculated pitch angle, yaw angle, focus distance, and the adjusted projection pattern density and scanning frequency are packaged together into the next scanning instruction set; The next scan instruction set is sent to the motion controller and sensor controller of the detection pan-tilt unit, driving the detection pan-tilt unit and integrated sensor to perform adjustments.

8. The method for detecting the diameter of pile holes used in highway engineering construction according to claim 7, characterized in that, After completing the full-depth inspection of the pile hole, the following further steps are included: All the final aperture deviation fields continuously acquired along the depth direction are stacked in the order of their corresponding detection depths to construct a complete body model of the measured aperture deviation of the pile hole in three-dimensional space. The measured diameter deviation model of the pile hole is compared with the three-dimensional reference model of the pile hole in a unified coordinate system. The difference between each voxel in the measured borehole diameter deviation volumetric model and the design value at the corresponding position in the three-dimensional reference model of the borehole is calculated to generate a three-dimensional deviation map of the entire borehole.

9. A method for detecting the diameter of pile holes used in highway engineering construction according to claim 8, characterized in that, Based on the full-hole three-dimensional deviation map, the following processing is performed: In the full-hole three-dimensional deviation map, a series of horizontal sections are extracted along the depth direction, and the average deviation, maximum deviation and deviation distribution variance of each section are calculated. The average deviation, maximum deviation, and variance of deviation distribution of each section are correlated with the design depth of the section and the corresponding soil layer type to form a section deviation characteristic sequence; Based on the cross-sectional deviation characteristic sequence of all cross sections, the trend decomposition method is used to separate the characteristics of the deviation trend component caused by systematic construction factors and the deviation residual component caused by random disturbances. The characteristics of the deviation trend components are fed back into the generation process of the three-dimensional reference model of the pile hole to optimize the simulation settings of relevant soil layer parameters in the physical simulation engine.

10. A method for detecting the diameter of pile holes used in highway engineering construction according to claim 9, characterized in that, The simulation settings for relevant soil layer parameters in the optimized physical simulation engine are as follows: From the cross-sectional deviation feature sequence, depth ranges in which the deviation trend component features show regular changes are identified; Match the depth range with the soil layers in the geological survey report to identify the specific soil layers for which parameters need to be optimized; Based on the characteristics of the deviation trend components in the corresponding soil layers, the simulated values ​​of soil density, internal friction angle, or cohesion in the physical simulation engine are adjusted in reverse. Using the adjusted soil parameters, the physical simulation engine was rerun to generate an updated 3D reference model of the pile hole. The updated 3D reference model for pile holes will serve as a new benchmark model for real-time registration and deviation calculation when conducting borehole diameter testing on subsequent pile holes of the same type or subsequent pile holes in the same project.

Images