Efficient grooving construction method for underground diaphragm wall

Abstract

The invention discloses an efficient grooving construction method for an underground diaphragm wall, and belongs to the technical field of underground diaphragm wall construction of geotechnical engineering. The method comprises the following steps: constructing a three-dimensional geologic model for representing stratigraphic distribution and a soft and hard abrupt change interface through gridding geological radar scanning and sparse inversion; dividing unit groove sections based on the model and planning an inter-jump type jump excavation sequence; establishing a construction coordinate reference system by taking the construction guide wall as a reference, and pre-burying a 5G signal transmitter at the boundary of the groove section; in the excavation process of the double-wheel slot milling machine according to the jump excavation sequence, equipment construction coordinates are calibrated in real time by recognizing 5G signals; meanwhile, model prediction control and model reference self-adaptive control are adopted to intelligently adjust tunneling parameters, and dynamic matching with stratum conditions is achieved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of underground continuous wall construction in geotechnical engineering, and particularly relates to an efficient trenching construction method for underground continuous walls. BACKGROUND

[0002] As a key supporting and anti-seepage structure in foundation pit engineering and underground structure, the quality and efficiency of trenching construction of underground continuous walls directly affect the safety and progress of the project. The traditional trenching construction method for underground continuous walls usually relies on limited pre-geological survey data, which is mostly derived from discrete borehole sampling and local geophysical prospecting results. Due to the significant spatial variability of underground stratum distribution, the data from discrete points cannot comprehensively and continuously reflect the stratum structure and geotechnical mechanical properties in the entire construction area, especially the identification accuracy of complex geological conditions such as sudden interfaces between soft and hard soil layers. This incompleteness of geological information leads to a lack of sufficient basis for trench segment division and excavation sequence planning, and risks such as trench wall collapse, equipment unbalanced loading, or sudden drilling efficiency reduction during construction.

[0003] In addition, during the trenching construction process, the positioning and attitude control of heavy equipment such as double-wheel trenchers mainly rely on external measurement means such as total stations for intermittent review. This method is greatly affected by the interference of the on-site environment, has a long measurement period, and cannot achieve continuous and real-time coordinate calibration during excavation. As the excavation depth increases, the measurement error gradually accumulates, which may cause the verticality deviation or axis offset of the trench, affecting the subsequent installation of the reinforcement cage and the integrity of the wall. SUMMARY

[0004] The application overcomes the shortcomings of the prior art and provides an efficient trenching construction method for underground continuous walls.

[0005] To achieve the above-mentioned purpose, the technical scheme adopted by the application is as follows: an efficient trenching construction method for underground continuous walls, comprising the following steps: S1: constructing a three-dimensional geological model; the three-dimensional geological model is used to represent stratum distribution and sudden interfaces between soft and hard layers; S2: based on the three-dimensional geological model, dividing the underground continuous wall into multiple standard unit trench segments and planning a skip excavation sequence; S3: constructing a guide wall, and establishing a construction coordinate reference system based on the guide wall; S4: based on the three-dimensional geological model, pre-reinforcing the stratum of the standard unit trench segment, and pre-burying a 5G signal transmitter at the segment boundary of the standard unit trench segment; S5: based on the skip excavation sequence, excavating the trench by a double-wheel trencher; during the trench excavation process, the construction coordinates of the double-wheel trencher in the construction coordinate reference system are calibrated in real time by identifying the 5G signal transmitter; S6: cleaning the trench bottom and pouring the wall to form an underground continuous wall.

[0006] Furthermore, the method for constructing the three-dimensional geological model in step S1 includes: S11: Obtain the original ground-penetrating radar dataset with spatial coordinate association through gridded ground-penetrating radar scanning; S12: Preprocess the raw ground-penetrating radar data to obtain preprocessed radar data; preprocessing includes filtering, frequency domain transformation, and time-frequency analysis. S13: Perform sparse inversion calculation on the preprocessed radar data to obtain the dielectric constant and conductivity of the three-dimensional spatial distribution. Combined with the local empirical correlation model established by the soil and rock sample test, obtain the elastic modulus and shear strength of the three-dimensional spatial distribution to form a three-dimensional geological parameter dataset. S14: Match the 3D geological parameter dataset to the voxel mesh to form a 3D geological model; S15: Based on the rate of change of elastic modulus and the rate of change of dielectric constant of adjacent voxel units, the interface between soft and hard abrupt changes is identified, and the spatial location data and tilt distribution map of the interface between soft and hard abrupt changes are obtained through surface fitting and geometric differential calculation.

[0007] Furthermore, the skip-dig sequence planned in step S2 is an intermittent skip-dig sequence, including: The standard unit slot segment is divided into end segment, main segment and transition segment; All standard unit trenches with odd numbers within the main excavation section; After the odd-numbered standard unit trenches are poured and their strength meets the standards, all even-numbered standard unit trenches in the main excavation section are excavated. After the main section is completed, the standard unit trench section of the transition section will be excavated. After the strength of the transition section meets the standard, the standard unit trench section of the end section is excavated.

[0008] Furthermore, in step S4, the strata of the trench wall are pre-reinforced by injecting nano-composite grout through mixing piles to form a bulge in the middle of the mixing piles.

[0009] Furthermore, the method for pre-embedding 5G signal transmitters in step S4 includes arranging 5G signal transmitters at intervals along the vertical direction on the boundary line of the standard unit slot, recording the coordinate data of each transmitter in the construction coordinate reference system, and forming a 5G signal transmitter array and corresponding coordinate dataset.

[0010] Furthermore, the method for real-time calibration of construction coordinates by identifying the 5G signal transmitter in step S5 includes: The 5G signal receiver on the dual-wheel grooving machine receives signals from multiple pre-embedded 5G signal transmitters, and the real-time coordinates of the dual-wheel grooving machine are calculated using a polygonal positioning algorithm. The attitude data of the twin-wheel grooving machine is acquired by the inertial measurement unit on the twin-wheel grooving machine, and the real-time coordinates are dynamically corrected. The corrected coordinates of the twin-wheel trenching machine are compared with the design excavation coordinates in the construction coordinate reference system. The position of the milling frame is adjusted through the control system of the twin-wheel trenching machine to complete the coordinate calibration.

[0011] Furthermore, in step S5, a model predictive control model is used to adjust the tunneling parameters of the twin-wheel trenching machine. The inputs of the model predictive control model include the stratum parameters of the current excavation area and the real-time sensing data of the twin-wheel trenching machine. The prediction layer predicts the stratum parameters in the future time domain based on the spatial distribution law of the stratum parameters. The output layer outputs the milling wheel cutting speed and output torque that need to be adjusted. The objective function is to minimize the rate of change of tunneling parameters and the estimated displacement of the trench wall.

[0012] Furthermore, when the twin-wheel milling machine advances to a distance of 1.5m-2.5m from the interface between soft and hard transitions, predictive adjustments are made based on the model prediction control model to make the milling wheel cutting speed decrease according to a cubic polynomial curve and the milling wheel output torque increase until it completely crosses the interface between soft and hard transitions.

[0013] Furthermore, in step S5, a model reference adaptive control model is used to adjust the working parameters of the milling wheel. The model reference adaptive control model adjusts the milling wheel drive voltage in real time through a proportional-integral adaptive law based on the error between the expected formation impedance obtained from the three-dimensional geological model and the actual output impedance of the equipment, thereby achieving dynamic matching between the output impedance of the dual-wheel trenching machine and the formation impedance.

[0014] Furthermore, the method for establishing a construction coordinate reference system based on the guide wall in step S3 includes taking the intersection point on the inner side of the starting end of the guide wall as the origin of the construction coordinate reference system, taking the extension direction of the inner axis of the top of the guide wall as the X-axis, taking the direction perpendicular to the X-axis and pointing to the outside of the foundation pit as the Y-axis, and taking the geodetic elevation datum as the Z-axis to form a construction coordinate reference system.

[0015] This invention addresses the shortcomings of the prior art and has the following beneficial effects: A gridded ground-penetrating radar (GPR) scan was used to acquire a raw GPR dataset with spatial coordinate correlation. After filtering, frequency domain transformation, and time-frequency analysis, preprocessed GPR data was obtained. Sparse inversion calculations were then performed to obtain the dielectric constant and conductivity of the three-dimensional spatial distribution. Combined with a local empirical correlation model established through soil and rock sample tests, the elastic modulus and shear strength were derived, forming a complete three-dimensional geological parameter dataset, which was then matched to a voxel grid to construct a three-dimensional geological model. This process achieves accurate spatial characterization of stratigraphic distribution and abrupt transitions between soft and hard surfaces, comprehensively capturing the continuous variation patterns of underground geological conditions. Compared to traditional geological exploration methods relying on discrete borehole data, the three-dimensional geological model constructed in this invention provides more reliable and detailed geological information, enabling accurate identification of potential risk areas before construction, significantly improving the predictability and planning rationality of trenching construction, and reducing trench wall instability or equipment damage caused by stratigraphic uncertainties.

[0016] A construction coordinate reference system is established based on the guide wall, with the inner intersection point of the guide wall's starting end as the origin, the extension direction of the guide wall's top inner axis as the X-axis, the direction perpendicular to the X-axis pointing outwards from the pit as the Y-axis, and the geodetic datum as the Z-axis, forming a unified coordinate framework. A 5G signal transmitter is pre-embedded at the boundary of the standard unit trench section to record its coordinate data. During excavation, the 5G signal receiver on the twin-wheel trenching machine receives the signal, and a polygonal positioning algorithm is used to calculate the real-time coordinates. This is combined with attitude data obtained from the inertial measurement unit for dynamic correction, and the control system adjusts the position of the milling frame to complete coordinate calibration. This process ensures that the position of the twin-wheel trenching machine in the construction coordinate reference system remains consistent with the designed excavation coordinates, effectively overcoming the shortcomings of traditional manual measurement methods, such as low efficiency and easy accumulation of errors. It significantly improves the positional accuracy and construction efficiency of trench excavation, and avoids rework or quality defects caused by coordinate deviations.

[0017] A model predictive control model is used to adjust the tunneling parameters of a twin-wheel trench cutter. Inputs include the current geological parameters of the excavation area and real-time sensing data from the twin-wheel trench cutter. The prediction layer predicts future geological parameters based on the spatial distribution of these parameters, while the output layer outputs the milling wheel cutting speed and output torque to be adjusted. The objective function is to minimize the rate of change of tunneling parameters and the estimated trench wall displacement. Simultaneously, a model reference adaptive control model is employed. Based on the error between the expected geological impedance derived from the 3D geological model and the actual output impedance of the equipment, the milling wheel drive voltage is adjusted in real-time using a proportional-integral adaptive law, achieving dynamic matching between the output impedance and the geological impedance. This process allows for smooth and adaptive adjustment of tunneling parameters according to changes in geological conditions. Especially near the interface between soft and hard transitions, predictive adjustment causes the milling wheel cutting speed to decrease according to a cubic polynomial curve, while the output torque increases, effectively reducing the disturbance to the trench wall and equipment vibration caused by parameter abrupt changes. Compared to traditional control methods that rely on operator experience, this intelligent control method significantly improves the stability of the construction process and the equipment's adaptability to complex geological formations, extending the equipment's service life.

[0018] The stratum parameters provided by the 3D geological model serve as input data for the model predictive control model and the model reference adaptive control model. Real-time calibration under the construction coordinate reference system ensures that the control model performs parameter adjustments based on accurate spatial location. The optimized parameters output by the model predictive control model and the model reference adaptive control model further guarantee the excavation accuracy and stability of the twin-wheel trenching machine in complex strata. This enables the entire trenching construction process to achieve efficient, accurate, and adaptive operation under data-driven conditions, significantly improving the overall construction quality and engineering reliability of the diaphragm wall. It overcomes the shortcomings of traditional methods, such as incomplete geological information, coarse coordinate control, and lagging parameter adjustment. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 This is a flowchart of an efficient trenching construction method for diaphragm walls. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Unless otherwise specified below, all raw materials can be obtained from commercial purchases or prepared by conventional methods in the art.

[0021] like Figure 1 As shown, a method for efficient trenching construction of diaphragm walls includes the following steps: S1: Construct a three-dimensional geological model; the three-dimensional geological model is used to characterize stratigraphic distribution and the interface between soft and hard transitions. S2: Based on a three-dimensional geological model, the underground continuous wall is divided into multiple standard unit segments, and the skip excavation sequence is planned; S3: Construction guide wall, establishing a construction coordinate reference system based on the guide wall; S4: Based on the three-dimensional geological model, the strata of the trench wall of the standard unit trench section are pre-reinforced, and 5G signal transmitters are pre-embedded at the trench section boundary of the standard unit trench section. S5: Based on the skip-excavation sequence, trench excavation is carried out using a twin-wheel trenching machine; during the trench excavation process, the construction coordinates of the twin-wheel trenching machine in the construction coordinate reference system are calibrated in real time by identifying the 5G signal transmitter; S6: Clean the bottom of the trench and pour the wall to form a continuous underground wall.

[0022] Below, each step will be explained in detail.

[0023] Step S1 involves constructing a three-dimensional geological model including complete stratigraphic information and determining the spatial location data and dip distribution map of the abrupt transition between soft and hard surfaces. For example, step S1 includes the following implementation process.

[0024] S11: Obtain the original ground-penetrating radar dataset with spatial coordinate association through gridded ground-penetrating radar scanning.

[0025] The surface of the construction area is prepared by removing debris, pipelines, and other objects that may interfere with radar signals. The scanning surface is also leveled to ensure uniform contact between the radar probe and the ground. Based on the design axis and width of the diaphragm wall, the scanning range is defined, covering an area of ​​3 meters on each side of the designed diaphragm wall axis to ensure complete capture of geological information about the trench wall and any potentially influencing strata outside the trench.

[0026] A gridded ground-penetrating radar (GPR) scan was employed. During the scan, a basic grid size of 0.5m × 0.5m was set. In areas where abrupt changes in hardness were anticipated (such as rock outcrops identified in previous surveys), the grid size was increased to 0.3m × 0.3m to improve data resolution in localized areas. The GPR used a broadband electromagnetic pulse range of 200MHz-1500MHz. The low-frequency electromagnetic waves (200MHz-500MHz) were used for strong penetration to detect deep geological structures at least 5m below the design trench bottom, while the high-frequency electromagnetic waves (1000MHz-1500MHz) were used for high-resolution detailing of shallow stratigraphic interfaces 0-8m below the surface.

[0027] During the scanning process, the amplitude, phase, and time-domain waveform data of the radar signal are recorded simultaneously. After each grid cell is scanned, the real-time positioning plane coordinates of that grid cell are immediately associated to form a mapping between the coordinates and the radar signal, thus obtaining the corresponding original ground-penetrating radar dataset.

[0028] S12: Preprocess the raw ground-penetrating radar data to obtain preprocessed radar data; preprocessing includes filtering, frequency domain conversion, and time-frequency analysis.

[0029] The raw ground-penetrating radar data was filtered using Butterworth high-order digital filtering to remove interference while avoiding phase distortion of the reflected signals from the formation interfaces. Based on the analysis of the noise frequency bands in the raw ground-penetrating radar data, the low-pass cutoff frequency of the filter was set to 1.2 GHz and the high-pass cutoff frequency was set to 180 MHz to remove high-frequency noise and low-frequency interference signals.

[0030] A Fast Fourier Transform (FFT) is performed on the filtered radar data to convert the signal from a time-domain representation to a frequency-domain representation. Different strata exhibit different absorption and scattering characteristics of electromagnetic wave energy. Soft soil has a strong absorption capacity for high-frequency electromagnetic waves, and its frequency-domain energy is concentrated in the low-frequency band; hard rock has a weak absorption capacity for high-frequency electromagnetic waves, and its frequency-domain energy is more evenly distributed in the mid-to-high frequency band. By analyzing this difference in frequency-domain energy distribution, the strata types are preliminarily classified.

[0031] Continuous wavelet transform is performed on the radar data after frequency domain conversion for time-frequency analysis. By dynamically adjusting the size of the time-frequency window, the continuous wavelet transform maintains a high time resolution in the low-frequency band (corresponding to deep strata) and a high frequency resolution in the high-frequency band (corresponding to shallow strata), capturing and locating reflection signals from non-stationary strata interfaces at different depths, especially strong reflection waves generated by abrupt transitions between soft and hard surfaces.

[0032] Through the above processing, preprocessed radar data is obtained that removes major interferences and highlights the time-frequency domain characteristics of the strata.

[0033] S13: Perform sparse inversion calculations on the preprocessed radar data to obtain the dielectric constant and conductivity of the three-dimensional spatial distribution. Combined with the local empirical correlation model established by the soil and rock sample test, obtain the elastic modulus and shear strength of the three-dimensional spatial distribution to form a three-dimensional geological parameter dataset.

[0034] The preprocessed radar data was subjected to sparse inversion calculations based on the least squares criterion. This calculation was based on the assumption that the electromagnetic parameters of the geological medium are spatially sparsity, meaning that abrupt changes in the electromagnetic parameters only occur at the interfaces between different strata, while the electromagnetic parameters within the same stratum are relatively uniform. Through iterative solutions, a stratigraphic parameter model was obtained that minimizes the difference between the theoretically synthesized radar data and the preprocessed radar data observed in actual observations. The dielectric constant and conductivity distributed in three-dimensional space were calculated. Specifically, the dielectric constant of soft soil was 8-15 F / m, and the conductivity was 10-30 mS / m; the dielectric constant of hard rock was 3-8 F / m, and the conductivity was 1-5 mS / m. The dielectric constant reflects the soil density and water content, while the conductivity reflects the soil composition and pore fluid properties.

[0035] To obtain mechanical parameters that are instructive for construction, soil and rock samples were collected from 3-5 boreholes within the construction area. Indoor physical and mechanical tests were conducted on these samples, including uniaxial compressive strength tests and direct shear tests. Local empirical correlation models for dielectric constant and elastic modulus, electrical conductivity, and shear strength were established using the test data. The model coefficients were determined through regression analysis of the test data. Substituting the three-dimensional dielectric constant and three-dimensional electrical conductivity data obtained in step S13 into these models, the three-dimensional spatially distributed elastic modulus and shear strength were obtained, forming a three-dimensional geological parameter dataset including dielectric constant, electrical conductivity, elastic modulus, and shear strength.

[0036] S14: Match the 3D geological parameter dataset to the voxel mesh to form a 3D geological model.

[0037] The underground space of the scanning area is divided into voxel grids, with the grid extending from the surface to 5m below the bottom of the designed trench. The size of the divided voxel unit is 0.1m×0.1m×0.05m, which balances model accuracy and computational efficiency. The horizontal resolution of 0.1m reflects the parameter changes of the strata near the trench wall, while the vertical resolution of 0.05m captures the thinner soft and hard interlayer strata.

[0038] The dielectric constant, electrical conductivity, elastic modulus, and shear strength in the three-dimensional geological parameter dataset obtained in step S13 are matched one by one to the corresponding voxel units according to spatial coordinates, so that each voxel unit carries the complete four parameter attributes, forming a digital geological body that can characterize the underground geological conditions.

[0039] The digital geological body was visualized and rendered using 3D modeling software (such as Surfer and FLAC3D) to obtain a 3D geological model. Optionally, voxel elements were colored according to their elastic modulus values, with soft soil areas (elastic modulus less than 500 MPa) rendered in red and hard rock areas (elastic modulus greater than 2000 MPa) rendered in blue, visually representing the stratigraphic distribution. Simultaneously, two borehole data points not involved in the empirical correlation model calibration were selected, and the actual stratigraphic parameters (dielectric constant, electrical conductivity, elastic modulus, and shear strength) corresponding to the borehole locations were extracted and compared with the parameters at the corresponding spatial locations in the 3D geological model. The error between the two was ensured to be less than 10%, thereby verifying the reliability of the 3D geological model.

[0040] S15: Based on the rate of change of elastic modulus and the rate of change of dielectric constant of adjacent voxel units, the interface between soft and hard abrupt changes is identified, and the spatial location data and tilt distribution map of the interface between soft and hard abrupt changes are obtained through surface fitting and geometric differential calculation.

[0041] A dual-parameter joint physical criterion is employed to identify abrupt transitions between soft and hard formations. Based on the parameter changes of adjacent voxel elements, a transition between soft and hard formations is determined when the rate of change of the elastic modulus of adjacent voxel elements exceeds 150% and the rate of change of the dielectric constant exceeds 80%. The rate of change of the elastic modulus directly reflects abrupt changes in formation stiffness, while the rate of change of the dielectric constant helps verify abrupt changes in formation composition and density. Combining these two methods can improve the accuracy of interface identification to over 95%.

[0042] The spatial coordinates (X, Y, and Z) of all voxel elements identified as abrupt transition interfaces between soft and hard surfaces are extracted. A quadratic polynomial surface fitting algorithm based on the least squares criterion is used to process these discrete coordinates. By minimizing the sum of squared errors, a continuous surface that optimally approximates the discrete points is constructed, forming a 3D surface model of the abrupt transition interface. This model can directly output the interface depth (Z) corresponding to any planar coordinate (X, Y), obtaining complete spatial location data of the interface.

[0043] Geometric differential calculations are performed on the three-dimensional surface model of the abrupt change in softness / hardness interface to obtain dip angle data. A tangent plane is taken at any point on the surface, and the angle between the tangent plane and the horizontal plane is calculated (the angle range is 0°-90°). In key areas such as the middle and ends of the preset trench section, calculation points are selected at 5m intervals, and the dip angle values ​​of each calculation point are statistically analyzed. The average dip angle and its variation range are further processed to obtain the dip angle distribution map of the abrupt change in softness / hardness interface.

[0044] Step S2, based on the stratigraphic parameters in the three-dimensional geological model, completes the division of the underground continuous wall into unit segments, forming an intermittent skip-excavation sequence. For example, step S2 includes the following implementation process.

[0045] S21: Determine the division parameters of the standard unit slot segment.

[0046] Extract the stratigraphic parameters of the area coinciding with the axis of the diaphragm wall in the three-dimensional geological model, analyze the spatial distribution characteristics of elastic modulus and shear strength in this area, and count the distribution length of different strata (soft soil, hard rock, soft-hard transition layer) in the direction of the diaphragm wall axis. These parameters directly determine the risk areas to be avoided when dividing the unit trench (such as continuous hard rock sections with a length of more than 5.0m).

[0047] Based on the technical parameters of the trenching equipment and the stability requirements of mud wall protection, the constraints for dividing the unit trench sections were determined. Specifically, the effective working range of the trenching equipment is 6.0-8.0m, determined by the milling wheel cutting radius and the machine's movement accuracy. The stability requirements for mud wall protection dictate that the length of each unit trench section should not exceed 8.0m; exceeding this length will increase the exposed area of ​​the trench wall, making it difficult to maintain uniform mud pressure and potentially causing trench wall deformation. Considering the formation parameters, equipment range, and wall protection conditions, the reasonable range for the length of each unit trench section is determined to be 6.0-7.0m.

[0048] S22: Based on the parameters determined in step S21, the diaphragm wall is segmented to form standard unit trench segments.

[0049] Along the design axis of the diaphragm wall, the entire diaphragm wall structure is divided into multiple standard unit segments connected end to end, with the length of each standard unit segment controlled between 6.0 and 7.0 meters. A length less than 6.0 meters will increase the number of interfaces between unit segments, leading to an increase in the workload of subsequent wall joint treatment; a length greater than 7.0 meters will exceed the upper limit of the stability conditions of the slurry wall protection, increasing the risk of wall instability.

[0050] During the division process, each standard unit trench segment is bound to the corresponding spatial region in the three-dimensional geological model. The stratigraphic composition details within each standard unit trench segment are recorded, including the proportion of soft soil area (30%-70%), hard rock area (10%-50%), soft-hard transition layer area (10%-30%) (the specific proportion is adjusted according to the actual geological conditions), as well as whether the standard unit trench segment contains a soft-hard abrupt transition interface, and the distribution depth of the interface within the trench segment (e.g., 12.0-15.0m underground), forming a stratigraphic parameter list for each standard unit trench segment.

[0051] S23: Based on the spatial location of standard unit trench segments and the corresponding list of geological parameters, classify all standard unit trench segments, determine the excavation sequence and time interval, and plan the inter-slope skip excavation sequence.

[0052] First, based on the location of the standard unit trench segments within the overall space of the foundation pit, all standard unit trench segments are divided into end segments located at both ends of the foundation pit, main segments located in the middle area of ​​the foundation pit, and transition segments between the end segments and the main segments. The total length of the end segments accounts for 10%-15% of the total length of the foundation pit, the total length of the main segments accounts for 70%-80% of the total length, and the total length of the transition segments accounts for 5%-10% of the total length. During classification, the stratigraphic parameter list for each standard unit trench segment is referenced simultaneously to ensure that standard unit trench segments within the same category have similar stratigraphic complexity (e.g., standard unit trench segments within the main segment do not contain continuous hard rock layers exceeding 3.0m in length).

[0053] Subsequently, the specific execution order of the intermittent skip-excavation sequence was determined. First, all odd-numbered standard unit trenches within the main section were excavated. During excavation, it was ensured that at least one unexcavated even-numbered standard unit trench was maintained between any two odd-numbered standard unit trenches under excavation, forming an intermittent excavation pattern to provide space for stress release in the strata. Second, after all odd-numbered standard unit trenches had completed concrete pouring and the concrete strength reached at least 70% of the design strength, all even-numbered standard unit trenches within the main section were excavated. The hardened walls of the odd-numbered standard unit trenches could provide lateral support to the trench walls of adjacent even-numbered standard unit trenches. Third, after all standard unit trenches in the main section had completed wall construction, the standard unit trenches in the transition section were excavated. The walls of the main section could reduce the disturbance of the strata in the core area of ​​the foundation pit caused by the excavation of the transition section. Fourth, after the walls of the standard unit trenches in the transition section reached at least 70% of the design strength, the standard unit trenches in the end sections were excavated.

[0054] A time interval of 72-96 hours is set between each excavation step to ensure that the strata have sufficient time to redistribute stress and that the strength of the constructed wall has fully increased. The final intermittent skip excavation sequence includes the classification results of all standard unit trenches, the order of excavation, the time interval requirements between each step, and the strata parameters that need to be carefully considered when excavating each standard unit trench (such as whether there is a sudden change in the interface between soft and hard rock, and the proportion of hard rock area).

[0055] Step S3 establishes a construction coordinate reference system based on the guide wall. For example, step S3 includes the following implementation process.

[0056] S31: Based on the design requirements of the diaphragm wall and the strata parameters in the three-dimensional geological model, determine the design parameters of the guide wall, such as cross-sectional dimensions, material strength, and burial depth.

[0057] The geological parameters within a range of 1.0m-1.5m on both sides of the diaphragm wall axis in the 3D geological model were extracted. The shear strength and elastic modulus of the soil in this area were analyzed in detail to determine the supporting capacity of the strata for the guide wall. Soft soil strata (shear strength 15kPa-30kPa, elastic modulus 200MPa-500MPa) provide weaker lateral constraints on the guide wall, while hard rock strata (shear strength 80kPa-120kPa, elastic modulus 2000MPa-3000MPa) provide stronger lateral constraints. The difference in strata directly affects the setting of the guide wall burial depth parameters.

[0058] Based on the working load of the trenching equipment (maximum static load 50kN-80kN, dynamic load 30kN-50kN) and the design width of the diaphragm wall (600mm-1200mm), the cross-sectional dimensions of the guide wall are determined. The top width of the guide wall is set at 600mm-800mm, the bottom width at 500mm-700mm, and the cross-sectional height at 1.0m-1.5m. These dimensions ensure that the guide wall can withstand the load of the trenching equipment and provide guiding space for the subsequent lowering of the reinforcing cage.

[0059] The material strength and burial depth parameters of the guide wall were determined. The concrete strength grade of the guide wall was selected as C25-C30, which can meet the requirements of the guide wall for bearing capacity and crack resistance. The burial depth of the guide wall was set at 1.5m-2.0m in soft soil and 1.0m-1.5m in hard rock. The difference in burial depth can ensure that the guide wall has sufficient stability under different strata conditions and avoid tilting or displacement of the guide wall during the trenching process.

[0060] S32: Cast the guide wall according to the design parameters determined in S31.

[0061] The surface of the guide wall construction area is treated by removing surface debris, loose soil and stones. A light roller is used to compact the base, and the compaction degree is controlled at 90%-95% to ensure that the bearing capacity of the soil base of the guide wall meets the design requirements (not less than 150kPa).

[0062] The reinforcing mesh for the guide wall is tied, consisting of main bars and stirrups. The main bars are HRB400 grade steel bars with a diameter of 12mm-16mm, with a longitudinal spacing of 150mm-200mm and a transverse spacing of 180mm-220mm. The stirrups are HPB300 grade steel bars with a diameter of 8mm-10mm, spaced at 250mm-300mm intervals. During the tying process, ensure that the intersections of the main bars and stirrups are tied in a staggered pattern, that the tying firmness meets the specifications, and that the concrete cover thickness is controlled to 30mm-40mm.

[0063] Install guide wall formwork, using steel formwork with a thickness of 5mm-8mm. Seal the joints of the formwork with sealing strips to prevent grout leakage during concrete pouring. Use a total station to check the verticality and axial position of the formwork. After checking, use steel pipe scaffolding to fix the formwork to ensure that it does not shift during the pouring process.

[0064] For pouring C25-C30 concrete, a concrete pump is used to deliver the concrete into the formwork. During the pouring process, an immersion vibrator (vibration frequency 2800r / min-3200r / min) is used to vibrate in layers. The thickness of each layer is controlled at 300mm-500mm, and the vibration time is controlled at 20s-30s until there are no obvious air bubbles overflowing from the concrete surface, ensuring the density of the concrete.

[0065] After the concrete is poured, cover it with geotextile and water it for curing. The curing time is set at 7-10 days, and the geotextile must be kept moist at all times during the curing period. After curing, the concrete strength of the guide wall is tested using the rebound method to ensure that the test results reach more than 70% of the design strength, thus forming a guide wall structure with load-bearing and guiding functions.

[0066] S33: Using the completed guide wall as a reference, and combining external measurement control points, a construction coordinate reference system is established through total station measurement and coordinate calibration.

[0067] Establish 3-5 horizontal control points and 2-3 vertical control points at stable locations outside the construction area. Horizontal control points will use forced-centering observation piers, while vertical control points will use concrete leveling points. A total station will be used to observe the horizontal control points, and a leveling instrument will be used to observe the vertical control points, obtaining the precise coordinates (horizontal coordinates X, Y, and vertical coordinates Z) of each control point.

[0068] The origin and coordinate axes of the construction coordinate reference system are set with the guide wall as the reference. The intersection point on the inner side of the starting end of the guide wall is selected as the origin of the construction coordinate reference system. The extension direction of the inner axis of the top of the guide wall is taken as the X-axis (parallel to the design axis of the diaphragm wall). The direction perpendicular to the X-axis and pointing to the outside of the pit is taken as the Y-axis. The geodetic elevation datum is taken as the Z-axis (the elevation starting point is consistent with the external elevation control point).

[0069] A total station was used to measure multiple feature points on the inner side axis of the top of the guide wall (one feature point was selected at intervals of 5m-10m). The coordinates (X, Y, and Z) of these feature points in the construction coordinate reference system were obtained and correlated with the coordinates of the external plane control points and elevation control points to ensure that the construction coordinate reference system is consistent with the external measurement benchmark.

[0070] The construction coordinate reference system is calibrated by repeatedly measuring the coordinates of feature points 3-5 times and calculating the deviation value of each measurement result. After the calibration is passed, the origin position, coordinate axis direction, accuracy index, and correlation parameters with external control points of the construction coordinate reference system are recorded to form the construction coordinate reference system.

[0071] Step S4 involves reinforcing the soil strata of the trench wall before excavation of the standard unit trench segment, and simultaneously deploying 5G signal transmitters. For example, step S4 includes the following implementation process.

[0072] S41: Based on the stratigraphic parameters in the three-dimensional geological model and the list of stratigraphic parameters for the standard unit trench, determine the layout range, size parameters, and nanocomposite grout ratio of the mixing piles.

[0073] The stratigraphic parameters of the corresponding area of ​​the standard unit trench segment in the 3D geological model were extracted, and the elastic modulus and shear strength of the soil within a range of 1.0m-1.5m outside the trench wall were analyzed. For soft soil strata (elastic modulus 200MPa-500MPa, shear strength 15kPa-30kPa), the pre-reinforcement range needs to be expanded to 1.5m outside the trench wall; for hard rock strata (elastic modulus 2000MPa-3000MPa, shear strength 80kPa-120kPa), the pre-reinforcement range can be reduced to 1.0m outside the trench wall, thus matching the stability requirements of different strata.

[0074] The diameter of the mixing piles is set at 0.8m-1.2m, the center-to-center spacing is set at 0.6m-0.9m, and the horizontal overlap rate is controlled at 25%-35% to ensure that adjacent mixing piles form a continuous solidified body. The mixing pile depth needs to penetrate the interface between soft and hard transitions and enter the stable stratum 2.0m-3.0m below the interface. The pier is set in the middle of the mixing pile (corresponding to a depth range of 8.0m-12.0m underground), and the diameter of the pier is 0.3m-0.5m larger than the diameter of the mixing pile to enhance the lateral support force on the trench wall.

[0075] The nanocomposite grout uses high-strength Portland cement as the main cementing material, accounting for 60%-70% by mass; ultrafine silica powder (average particle size 3μm-5μm) accounts for 15%-20% by mass, used to fill the micropores in the soil; nano-silica (average particle size 20nm-50nm) accounts for 5%-10% by mass, optimizing the microstructure of the grout; and high molecular polymers (such as polyacrylamide) account for 3%-5% by mass, improving the toughness of the solidified body. The water-cement ratio of the grout is set at 0.8-1.2, adjusted according to the formation permeability. A water-cement ratio of 0.8-1.0 is used for formations with high permeability, and a water-cement ratio of 1.0-1.2 is used for formations with low permeability.

[0076] S42: Construction mixing pile, injecting nano-composite grout to form a thickened pier.

[0077] A high-pressure jet grouting drill was used to mark and position the pre-reinforcement area. During drilling, the drilling speed was controlled at 0.5 m / min-1.0 m / min, and the borehole diameter was consistent with the diameter of the mixing pile (0.8 m-1.2 m) until the designed depth was reached (penetrating 2.0 m-3.0 m below the soft-hard interface). During drilling, strata parameters (such as resistance and grout return volume) were collected simultaneously in the borehole and compared with the corresponding location parameters in the three-dimensional geological model to ensure that the borehole depth matched the strata.

[0078] Start the mixing device and inject the nano-composite grout into the hole through the internal channel of the drill rod. Control the grouting pressure at 15MPa-20MPa and the grouting flow rate at 50L / min-80L / min. Control the mixing blade speed at 30r / min-50r / min. Use a two-way mixing process of grouting during drilling and re-mixing during drilling. Grout to the designed depth during drilling and increase the speed to 50r / min-60r / min during drilling to ensure that the grout is fully mixed with the soil.

[0079] In the middle section of the mixing pile (8.0m-12.0m underground), a roughing operation is performed. By increasing the grouting pressure to 20MPa-25MPa and reducing the drilling speed to 0.2m / min-0.3m / min, the grout is allowed to fully diffuse in this area, forming a roughing body with a diameter of 1.1m-1.7m. After the construction of each mixing pile is completed, an ultrasonic testing instrument is used to test the integrity of the pile body to ensure that there are no broken piles or voids, and that the uniaxial compressive strength of the solidified body reaches 50MPa-70MPa after 28 days, and the flexural strength reaches 10MPa-15MPa.

[0080] S43: A 5G signal transmitter is pre-buried at the boundary of the slot section.

[0081] At the boundary of each standard unit trench section, a 5G signal transmitter is deployed at vertical intervals of 3.0m-5.0m, with 3-5 transmitters deployed on each boundary line (covering from the ground surface to the designed trench bottom depth). The transmitter placement should avoid the thickened mixing pile piers to ensure unobstructed signal transmission.

[0082] The 5G signal transmitter is encapsulated using a stainless steel sheath with a diameter of 50mm-60mm and a length of 150mm-200mm. The inner wall of the sheath is lined with an insulating layer (2mm-3mm thick) to prevent corrosion and interference from groundwater and electrolytes. The transmitter operates at a frequency of 3.5GHz-5.0GHz, and its built-in storage module records its serial number and initial coordinate data.

[0083] A small drilling machine was used to drill holes at the trench section boundaries. The encapsulated 5G signal transmitters were placed into the holes, and the gaps were filled with micro-expansion cement mortar to fix the transmitters. The spatial coordinates of each transmitter were measured using a total station. Based on the construction coordinate reference system established by S3, the coordinate data (X, Y, and Z) of each transmitter were recorded and written to the storage module to ensure that the transmitter coordinates were consistent with the construction coordinate reference system, thus forming a 5G signal transmitter array and its corresponding coordinate dataset.

[0084] Step S5 involves using a twin-wheel trenching machine to excavate the trench, while simultaneously calibrating the construction coordinates via a 5G signal. For example, step S5 includes the following implementation process.

[0085] In this step, following the skip-excavation sequence determined in step S23, the standard unit trenches with odd numbers within the main section are excavated first. The initial excavation speed of the twin-wheel trenching machine is set to 0.8 m / h-1.2 m / h, and the speed is adjusted according to the real-time feedback of the stratum resistance (it can be increased to 1.2 m / h-1.5 m / h for soft soil strata and decreased to 0.5 m / h-0.8 m / h for hard rock strata).

[0086] During the excavation process, mud slurry was used for wall protection. Bentonite mud slurry was used, with a viscosity controlled at 18s-22s and a density controlled at 1.05g / cm³. 3 -1.20g / cm 3 The filtration loss is controlled below 10 mL / 30 min, and the pH value is maintained between 9.0 and 10.5. The mud is continuously injected into the trench through the mud circulation system to ensure that the mud level is always 1.5 m to 2.0 m above the groundwater level, so as to balance the lateral earth pressure on the trench wall and prevent the trench wall from collapsing.

[0087] After each odd-numbered standard unit trench segment is excavated to the design depth (0.5m-1.0m below the design trench bottom), the verticality of the trench wall is checked using an ultrasonic wall gauge. Once all odd-numbered standard unit trench segments have been excavated and the concrete has been poured and cured to more than 70% of the design strength, the even-numbered standard unit trench segments in the main section are excavated according to the same process. Subsequently, the standard unit trench segments of the transition section and the end section are excavated in sequence.

[0088] Preferably, 5G signal receivers are installed at the top and middle of the milling frame of the twin-wheel milling machine. The operating frequency of the receivers is matched with the 5G signal transmitter pre-embedded in step S4, and the receivers receive signals from 3-5 surrounding transmitters in real time. At the same time, an inertial measurement unit, including a three-axis accelerometer and a three-axis gyroscope, is installed in the middle of the milling frame to collect the attitude data (pitch angle, roll angle, yaw angle) of the twin-wheel milling machine.

[0089] The signal arrival times of multiple pre-embedded transmitters are obtained by the 5G signal receiver on the dual-wheel grooving machine. The distance between the receiver and each transmitter is calculated. Combined with the known coordinates of the transmitters (from the S43 coordinate dataset), a polygonal positioning algorithm is used to solve the real-time coordinates of the dual-wheel grooving machine. Simultaneously, the attitude change of the twin-wheel slotting machine is acquired through an inertial measurement unit, and the calculated coordinates are dynamically corrected to obtain the corrected coordinates of the twin-wheel slotting machine. .

[0090] Compare the corrected coordinates of the twin-wheel trenching machine with the design excavation coordinates in the S3 construction coordinate reference system. Compare and calculate the coordinate deviation value. The control system of the twin-wheel trenching machine is used to adjust the position of the milling frame to complete the construction coordinate calibration. The calibration process is repeated every 1.0m-2.0m of excavation depth to ensure that the positional accuracy of the twin-wheel trenching machine meets the requirements throughout the excavation process, and a coordinate calibration record is generated (including the deviation value, adjustment amount, and calibration time for each calibration).

[0091] Preferably, a model predictive control (MMCC) model is used to achieve smooth adjustment of tunneling parameters (cutting speed, milling wheel torque), avoiding trench wall disturbances caused by sudden parameter changes. The MMCC model has rolling optimization and multivariate constraint control capabilities, and can optimize tunneling parameters based on geological prediction data over a future period, adapting to the requirements for smooth parameter adjustment in strata with abrupt changes in hardness.

[0092] For example, the configuration of the model predictive control model is as follows: The input layer includes the geological parameters (elastic modulus) of the current excavation area in the 3D geological model. Shear strength ) and real-time sensing data (current depth) of the twin-wheel grooving machine Real-time cutting force The input dimension is set to 4 dimensions. , , , ).

[0093] The prediction layer includes setting the prediction time domain to 5s-10s (covering the time it takes for the twin-wheel trencher to tunnel to a depth of 0.1m-0.2m), and based on the spatial distribution of formation parameters, predicting the elastic modulus of the formation that the twin-wheel trencher will encounter within the next 5s-10s. Formation shear strength .

[0094] The output layer includes the output of the tunneling parameters that need to be adjusted, namely the real-time cutting speed of the milling wheel. Milling wheel output torque The output dimension is set to 2-dimensional.

[0095] The objective function is set to minimize the rate of change of tunneling parameters and the estimated displacement of the trench wall, i.e. ,in, The objective function value is dimensionless. The smaller the value, the smoother the parameter adjustment and the smaller the tank wall displacement. , and The weighting coefficients are 0.4, 0.4, and 0.2 respectively, with the weighting prioritizing parameter smoothness. The number of time-domain steps for prediction (values ​​range from 5 to 10); Let k be the milling wheel cutting speed (m / h) predicted at step k. The actual milling wheel cutting speed (m / h) in step k-1. The predicted milling wheel output torque (kN·m) is given at step k. The actual milling wheel output torque (kN·m) in step k-1. The estimated trench wall displacement (mm) at step k is derived from the formation elastic modulus. Formation shear strength With cutting force The calculation yielded the result.

[0096] The dataset used for model training consists of parameter samples (1000-2000 sets) of different strata in the 3D geological model and historical tunneling parameter adjustment records (500-800 sets) of the same strata. During the training process, the gradient descent method is used to minimize the objective function to ensure that the adjustment error of the model in strata with abrupt changes in hardness is ≤5%.

[0097] For example, when the twin-wheel trenching machine excavates to a distance of 1.5m-2.5m from the interface between soft and hard transitions (based on the interface spatial location data of S15), predictive adjustments are made according to the model predictive control model.

[0098] First, input the current formation parameters (elastic modulus 300MPa-2500MPa, shear strength 20kPa-100kPa) and the real-time data of the twin-wheel milling machine (current depth 10m-20m, real-time cutting force 50kN-200kN) to predict the formation parameters in the next 5s-10s, with an elastic modulus of 500MPa-3000MPa and a shear strength of 30kPa-120kPa.

[0099] Secondly, based on the predicted formation parameters and objective function, the adjustment trajectory of the milling wheel cutting speed is calculated: from the steady-state speed of 1.2m / h-1.5m / h in soft soil formations, it is smoothly reduced according to a cubic polynomial curve to the suitable speed of 0.5m / h-0.8m / h in hard rock formations, and the speed change rate during the adjustment process does not exceed 0.2m / h·s.

[0100] Simultaneously, the adjustment trajectory of the milling wheel output torque is calculated: from the steady-state torque of 30kN·m-50kN·m in soft soil strata, it is smoothly increased to the suitable torque of 80kN·m-120kN·m in hard rock strata, with the torque change rate not exceeding 5kN·m·s, to avoid sudden torque changes that could cause milling wheel vibration.

[0101] During the adjustment process, real-time parameters of the twin-wheel grooving machine (milling wheel cutting speed, milling wheel output torque, and real-time cutting force) are collected every 0.5 seconds and fed back to the model predictive control model. The objective function is recalculated and the adjustment trajectory is corrected to ensure that the deviation between the actual parameters and the predicted trajectory does not exceed 0.1 m / h (speed) and 3 kN·m (torque) until the twin-wheel grooving machine completely crosses the interface between soft and hard transitions (the excavation exceeds 2.0m-3.0m below the interface).

[0102] Preferably, based on the real-time vibration feedback and formation impedance calculation of the twin-wheel trenching machine, a model reference adaptive control model is adopted to adjust the working parameters of the milling wheels, thereby achieving dynamic matching between the output impedance of the twin-wheel trenching machine and the formation impedance, reducing equipment vibration and formation disturbance. The model reference adaptive control model has real-time adaptive capability, which can adjust the control parameters according to the dynamic changes in formation impedance, adapting to scenarios with rapid impedance changes in formations with abrupt changes in hardness.

[0103] For example, the configuration of the model predictive control model is as follows: The reference model is set as the response model of a two-wheeled trenching machine under ideal impedance matching conditions, and the input is the desired formation impedance. (Based on the elastic modulus in the three-dimensional geological model) With density Calculations show that The output is the ideal milling wheel vibration frequency. and amplitude .

[0104] The controlled object is the milling wheel drive system of a two-wheel milling machine, and the input is the milling wheel drive voltage. The output is the actual milling wheel vibration frequency. ,amplitude With the output impedance of the device ( Milling wheel torque Rotation speed Calculations show that ).

[0105] The adaptive law adopts a proportional-integral adaptive law, based on the error between the output of the reference model and the output of the controlled object. , Adjust the drive voltage U in real time to reduce the error. ≤1Hz ≤0.05mm, the adaptive law formula is: ,in Let V be the milling wheel drive voltage at step k; The milling wheel drive voltage (V) is for the (k-1)th step. This is a proportionality coefficient (with a value of 0.5-1.0). This is the integral coefficient (values ​​range from 0.1 to 0.3). Let be the vibration frequency error (Hz) at step k. To adjust the time (s); Let be the integral variable (s).

[0106] Desired formation impedance in the reference model The elastic modulus and density at the current excavation location in the 3D geological model need to be updated in real time, with the update frequency synchronized with the tunneling speed of the twin-wheel trenching machine (0.5s / time); the proportional coefficient of the adaptive law... Integral coefficient In soft soil layers (elastic modulus < 500 MPa), the smaller value should be taken. =0.5、 =0.1), take the larger value when in hard rock formations (elastic modulus > 2000 MPa). =1.0、 =0.3), which improves the impedance matching response speed of hard rock formations.

[0107] For example, during the excavation process of a twin-wheel trenching machine, the vibration signal of the milling wheel (via the triaxial accelerometer built into the twin-wheel trenching machine) and the driving parameters (milling wheel torque) are collected in real time. Rotation speed ).

[0108] First, perform a Fast Fourier Transform on the vibration signal to extract the actual vibration frequency. (25Hz-60Hz) and amplitude (0.2mm-0.8mm); simultaneously calculate the equipment output impedance. (40kN·s / rad - 150kN·s / rad), calculate the current formation impedance (Based on vibration frequency) ,amplitude With cutting force Calculations show that ).

[0109] Secondly, the calculated formation impedance The reference model of the input model predictive control model is used to generate the ideal vibration frequency. (20Hz-50Hz) and amplitude (0.1mm-0.3mm), calculation error , .

[0110] when >1Hz or When the diameter is greater than 0.05mm, the milling wheel drive voltage is adjusted according to the adaptive law. :like > (The formation impedance is greater than the equipment output impedance), increase To increase the milling wheel speed With torque ,make Approaching ;like < (The formation impedance is less than the equipment output impedance), reduce To reduce and To avoid overloading equipment or excessive disturbance to the formation.

[0111] During the adjustment process, the error is updated every 0.1 seconds. , With driving voltage Ensure the output impedance of the device With formation impedance With a matching error of ≤10%, the dynamic impedance matching between the twin-wheel trenching machine and the stratum is ultimately achieved, reducing equipment wear and trench wall disturbance during the tunneling process.

[0112] Step S6 removes sediment from the bottom of the trench, completes the installation of the reinforcing cage and the pouring of concrete, forming a complete unit wall segment. For example, step S6 includes the following implementation process.

[0113] S61: Remove sediment from the bottom of the trench.

[0114] A reverse circulation airlift system is used for cleaning the bottom of the trench. A combination of high-pressure air hoses and suction hoses is lowered to a depth of 200-300mm from the bottom of the trench, and high-pressure air at a pressure of 0.6MPa-0.75MPa is introduced. The high-pressure air mixes with the sludge inside the suction hoses to form a low-density gas-liquid mixture. The buoyancy generated by the density difference lifts the mixture of sediment and sludge from the bottom of the trench to the surface. After separation by a sludge separator, the sediment is transported off-site for disposal, while the qualified sludge is returned to the trench.

[0115] During the cleanup process, fresh bentonite slurry was continuously added to the trench to maintain the slurry level 1.5m-2.0m above the groundwater level, while simultaneously controlling the slurry performance parameters to remain stable within the specified range: slurry viscosity 18s-22s, density 1.05g / cm³. 3 -1.20g / cm 3Filtration loss ≤10mL / 30min, pH value 9.0-10.5, to avoid deterioration of mud properties leading to tank wall instability or sediment suspension.

[0116] An ultrasonic sediment detector with a frequency of 200kHz was used to monitor the sediment thickness at the bottom of the tank in real time. The detector probe was placed at intervals of 1.5m-2.0m along the length of the trench, and each measurement was repeated three times. The average value was taken as the sediment thickness at that point. Tank bottom cleaning operations were stopped when the sediment thickness at all detection points was less than 30mm. The sediment thickness data for each detection point was recorded, and a tank bottom sediment detection report was generated.

[0117] S62: Place the precast steel cage.

[0118] The fabrication of precast reinforcing cages must match the design dimensions of the diaphragm wall. The length of the reinforcing cage should be 500mm-800mm greater than the trench depth (to allow for the top removal section), and the width should be 60mm-80mm less than the trench width (to allow for the concrete cover). The main reinforcement bars of the reinforcing cage should be HRB400 grade steel bars with a diameter of 20mm-25mm, spaced 150mm-200mm longitudinally; the stirrups should be HPB300 grade steel bars with a diameter of 8mm-10mm, spaced 250mm-300mm transversely. A reinforcing hoop (16mm-18mm in diameter) should be installed every 2.0m-3.0m on the inner side of the reinforcing cage to enhance its overall rigidity. Concrete cover spacers (strength grade ≥ C30) should be tied to the outer side of the reinforcing cage, spaced 1.5m-2.0m apart, ensuring a concrete cover thickness of 30mm-40mm between the reinforcing cage and the trench wall.

[0119] A crawler crane with a lifting capacity of 50t-80t is used to lift the rebar cage. The main hook of the crane is connected to the lifting point at the top of the rebar cage, and the auxiliary hook assists in adjusting the posture of the rebar cage to avoid deformation during the lifting process. After the rebar cage is lifted above the trench, it is slowly lowered at a controlled speed of 1.0m / min-1.5m / min. During the lowering process, the 5G signal receiver on the dual-wheel trenching machine receives the signal from the 5G signal transmitter pre-embedded in S4, and combined with the construction coordinate reference system established in S3, the planar position and verticality of the rebar cage are monitored in real time.

[0120] When the reinforcing cage is lowered to the design elevation (top 300mm-500mm above the top surface of the guide wall), it is fixed to the guide wall using structural steel to prevent it from floating or shifting. After fixing, the planar position and verticality of the reinforcing cage are checked again. Once the deviation is confirmed to meet the requirements, the final positioning data of the reinforcing cage is recorded, and a reinforcing cage positioning report is generated.

[0121] S63: Underwater concrete pouring is carried out to form a diaphragm wall.

[0122] The concrete used for pouring is self-compacting underwater concrete, with a strength grade of C30-C40, a slump controlled at 200mm-230mm, a slump spread controlled at 650mm-750mm, and an initial setting time greater than 8 hours to ensure good fluidity and anti-segregation ability in the underwater environment. The raw materials for the concrete include continuously graded crushed stone with a particle size of 5mm-25mm as coarse aggregate, medium sand (fineness modulus 2.3-3.0) as fine aggregate, and a cementitious material dosage of 380kg / m³. 3 -420kg / m 3 The water-to-binder ratio is 0.40-0.45.

[0123] Before pouring concrete, install the concrete pouring guide pipe. The guide pipe should be a seamless steel pipe with an inner diameter of 300mm-350mm, with each section being 2.0m-3.0m long. Rubber sealing rings should be used at the joints to prevent leakage. A water tightness test (test pressure 0.6MPa-0.8MPa) should be conducted before lowering the guide pipe to ensure there is no leakage. When lowering the guide pipe into the trench, the distance between the bottom of the guide pipe and the bottom of the trench should be controlled at 400mm-600mm, and the top of the guide pipe should extend 1.5m-2.0m above the top surface of the guide wall to facilitate concrete pouring.

[0124] Start the concrete pump for pouring. The initial pour volume should be sufficient to embed the bottom of the guide pipe 1.0m-1.5m below the concrete surface. The initial pouring time should be controlled within 10-15 minutes. During subsequent pours, maintain continuous concrete delivery, controlling the pouring speed at 3m / s². 3 / h-5m 3 / h, while monitoring the concrete liquid level in real time with a measuring rope (with a weight), and adjusting the lifting speed of the guide pipe (1.0m / min-2.5m / min) to ensure that the bottom of the guide pipe is always buried 2.5m-3.5m below the concrete liquid level, so as to avoid the guide pipe being pulled out of the concrete liquid level and causing the pile to break.

[0125] When the concrete is poured to 500mm-1000mm above the design elevation, pouring should be stopped (leaving a section for removing the top laitance). After pouring is completed, wait for the concrete to initially set (4-6 hours after pouring), then remove the pouring guide pipe and the steel section fixing the reinforcing cage. After the concrete has cured for 7-10 days, use an ultrasonic flaw detector to test the integrity of the unit wall section to ensure that the wall is free of defects such as holes and delamination, and that the 28-day compressive strength of the concrete reaches more than 100% of the design strength. A diaphragm wall pouring record and quality inspection report should then be generated.

[0126] Based on the preferred embodiments of the present invention described above, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A method for efficient trenching construction of diaphragm walls, characterized in that, Includes the following steps: S1: Construct a three-dimensional geological model; the three-dimensional geological model is used to characterize the stratigraphic distribution and the interface between soft and hard transitions. S2: Based on the aforementioned three-dimensional geological model, the underground continuous wall is divided into multiple standard unit segments, and the skip excavation sequence is planned; S3: Construction guide wall, establishing a construction coordinate reference system based on the guide wall; S4: Based on the three-dimensional geological model, the strata of the trench wall of the standard unit trench segment are pre-reinforced, and a 5G signal transmitter is pre-embedded at the trench segment boundary of the standard unit trench segment. S5: Based on the skip-dig sequence, trench excavation is carried out using a twin-wheel trenching machine; during the trench excavation process, the construction coordinates of the twin-wheel trenching machine in the construction coordinate reference system are calibrated in real time by identifying the 5G signal transmitter; S6: Clean the bottom of the trench and pour the wall to form a continuous underground wall.

2. The efficient trenching construction method for diaphragm walls according to claim 1, characterized in that, The methods for constructing the three-dimensional geological model in step S1 include: S11: Obtain the original ground-penetrating radar dataset with spatial coordinate association through gridded ground-penetrating radar scanning; S12: Preprocess the raw ground-penetrating radar data to obtain preprocessed radar data; the preprocessing includes filtering, frequency domain conversion, and time-frequency analysis. S13: Perform sparse inversion calculation on the preprocessed radar data to obtain the dielectric constant and conductivity of the three-dimensional spatial distribution, and combine them with the local empirical correlation model established by the soil and rock sample test to obtain the elastic modulus and shear strength of the three-dimensional spatial distribution, forming a three-dimensional geological parameter dataset. S14: Match the three-dimensional geological parameter dataset to a voxel grid to form a three-dimensional geological model; S15: Based on the rate of change of elastic modulus and the rate of change of dielectric constant of adjacent voxel units, the interface between soft and hard abrupt changes is identified, and the spatial location data and tilt distribution map of the interface between soft and hard abrupt changes are obtained through surface fitting and geometric differential calculation.

3. The efficient trenching construction method for diaphragm walls according to claim 1, characterized in that, The skip-excavation sequence planned in step S2 is an intermittent skip-excavation sequence, including: The standard unit slot segment is divided into end segment, main segment and transition segment; Excavate all standard unit trenches with odd numbers within the main section at intervals; After the odd-numbered standard unit trenches are poured and reach the required strength, all even-numbered standard unit trenches in the main section are excavated. After the main section is completed, the standard unit trench section of the transition section is excavated. After the strength of the transition section meets the standard, the standard unit trench section of the end section is excavated.

4. The efficient trenching construction method for diaphragm walls according to claim 1, characterized in that, In step S4, the strata of the trench wall are pre-reinforced by injecting nano-composite grout through mixing piles to form a bulge in the middle of the mixing piles.

5. The efficient trenching construction method for diaphragm walls according to claim 1, characterized in that, The method for pre-embedding 5G signal transmitters in step S4 includes arranging 5G signal transmitters at intervals along the vertical direction on the boundary line of the standard unit slot, recording the coordinate data of each transmitter in the construction coordinate reference system, and forming a 5G signal transmitter array and corresponding coordinate dataset.

6. The efficient trenching construction method for diaphragm walls according to claim 1, characterized in that, The method for real-time calibration of construction coordinates by identifying the 5G signal transmitter in step S5 includes: The real-time coordinates of the dual-wheel groove milling machine are calculated by receiving signals from multiple pre-embedded 5G signal transmitters via a 5G signal receiver on the machine. The attitude data of the twin-wheel grooving machine is acquired by the inertial measurement unit on the twin-wheel grooving machine, and the real-time coordinates are dynamically corrected. The corrected coordinates of the twin-wheel trenching machine are compared with the design excavation coordinates in the construction coordinate reference system. The position of the milling frame is adjusted through the control system of the twin-wheel trenching machine to complete the coordinate calibration.

7. The efficient trenching construction method for diaphragm walls according to claim 1, characterized in that, In step S5, a model predictive control model is used to adjust the tunneling parameters of the twin-wheel trenching machine. The input of the model predictive control model includes the stratum parameters of the current excavation area and the real-time sensing data of the twin-wheel trenching machine. The prediction layer predicts the stratum parameters in the future time domain based on the spatial distribution law of the stratum parameters. The output layer outputs the milling wheel cutting speed and output torque that need to be adjusted. The objective function is to minimize the rate of change of tunneling parameters and the estimated displacement of the trench wall.

8. The efficient trenching construction method for diaphragm walls according to claim 7, characterized in that, When the twin-wheel milling machine advances to a distance of 1.5m-2.5m from the interface between soft and hard transitions, it makes predictive adjustments based on the model prediction control model, so that the milling wheel cutting speed decreases according to a cubic polynomial curve and the milling wheel output torque increases until it completely crosses the interface between soft and hard transitions.

9. The efficient trenching construction method for diaphragm walls according to claim 1, characterized in that, In step S5, a model reference adaptive control model is also used to adjust the working parameters of the milling wheel. The model reference adaptive control model adjusts the milling wheel drive voltage in real time through a proportional-integral adaptive law based on the error between the expected formation impedance obtained from the three-dimensional geological model and the actual output impedance of the equipment, so as to achieve dynamic matching between the output impedance of the dual-wheel trenching machine and the formation impedance.

10. The efficient trenching construction method for diaphragm walls according to claim 1, characterized in that, The method for establishing a construction coordinate reference system based on the guide wall in step S3 includes taking the intersection point on the inner side of the starting end of the guide wall as the origin of the construction coordinate reference system, taking the extension direction of the inner axis of the top of the guide wall as the X-axis, taking the direction perpendicular to the X-axis and pointing to the outside of the foundation pit as the Y-axis, and taking the geodetic elevation datum as the Z-axis, thus forming the construction coordinate reference system.

Images