An Optimization Method and Device for Graded Backfilling of Prefabricated Channels Based on Non-uniform Settlement Control

Abstract

This invention relates to the field of tunnel filling technology, and specifically discloses a method and apparatus for optimizing graded backfilling of prefabricated tunnels based on uneven settlement control. The method includes: determining the target filler gradation range and initial layer backfill thickness, obtaining qualified filler material that meets the target filler gradation range, and recording the gradation test report for each batch of filler material; laying the qualified filler material according to the initial layer backfill thickness to form the first layer of backfill soil, and measuring the settlement values ​​of the top and bottom slabs of the prefabricated tunnel sidewalls and generating a layered backfill-settlement correspondence table by combining the compaction parameters of the intelligent vibration compaction unit; predicting the predicted settlement difference after the completion of the subsequent n layers of backfill construction, and adjusting any one of the parameters of the sieve vibration frequency of the gradation sieve unit, the initial layer backfill thickness, and the compaction parameters of the intelligent vibration compaction unit for the next round of layered backfill construction based on the predicted settlement difference, until all backfilling operations are completed; thereby improving the structural stability and safety of the tunnel.

Description

Technical Field

[0001] This invention relates to the field of tunnel filling technology, and in particular to a prefabricated tunnel graded backfilling optimization method and device based on uneven settlement control. Background Technology

[0002] In modern engineering construction, prefabricated access roads, as a highly efficient form of infrastructure, are widely used in highways, railways, municipal works, and many other fields. However, uneven settlement is a critical issue that seriously affects the stability and service life of prefabricated access roads during use. Uneven settlement caused by uneven backfill soil gradation and insufficient compaction can lead to structural deformation and cracking of the access road, thereby affecting its normal function and even endangering traffic safety. A graded backfill optimization method and device for prefabricated access roads based on uneven settlement control is of great significance for ensuring the engineering quality and long-term stability of prefabricated access roads. With the continuous expansion of infrastructure construction scale and increasingly stringent requirements for engineering quality, this technology has broad application prospects in the future engineering construction field and is expected to promote the development of prefabricated access road construction technology towards a more scientific and precise direction.

[0003] However, existing prefabricated tunnel backfilling technologies cannot comprehensively consider the foundation parameters of the construction area, the fill material gradation test report, and the layered backfill-settlement correspondence table to accurately predict the predicted settlement difference after subsequent backfilling construction. Consequently, they cannot adjust key parameters for the next round of layered backfilling based on the predicted settlement difference, such as the vibration frequency of the sieve holes in the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit. This leads to significant human error, low construction efficiency, and difficulty in effectively controlling uneven settlement in traditional backfilling processes, severely impacting the long-term stability and service life of prefabricated tunnels.

[0004] Therefore, this invention proposes an optimized method and apparatus for graded backfilling of prefabricated channels based on uneven settlement control. Summary of the Invention

[0005] This invention provides a method and apparatus for optimizing graded backfilling of prefabricated tunnels based on non-uniform settlement control. This method and apparatus, by real-time monitoring of backfill compaction and combined with numerical simulation to predict settlement trends, can dynamically adjust the backfill layer thickness and compaction parameters, ensuring that the fill particle size meets design requirements and achieving automated layered compaction. By closely linking settlement control standards with construction technology, forming a closed loop of "monitoring-prediction-adjustment," it not only improves construction efficiency but also effectively enhances the long-term stability of prefabricated tunnel structures.

[0006] This invention provides an optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control, comprising:

[0007] S1: Determine the target packing gradation range and initial layered backfill thickness, and put the packing to be backfilled into the gradation screening unit for gradation screening and crushing to obtain qualified packing that meets the target packing gradation range and transport it to the backfilling operation area. At the same time, record the gradation test report of each batch of packing.

[0008] S2: According to the initial layered backfill thickness, sieved qualified fill material is laid on both sides and top of the prefabricated passage to form the first layer of backfill soil. The settlement value of the top of the side wall and bottom plate of the prefabricated passage is measured. Combined with the excitation force and compaction pass of the intelligent vibration compaction unit, a layered backfill-settlement correspondence table is generated.

[0009] S3: Based on the foundation parameters of the prefabricated channel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table, predict the predicted settlement difference after the completion of the subsequent n layers of backfill construction. Based on the predicted settlement difference, adjust any one of the following parameters in the next round of layered backfill construction: the sieve vibration frequency of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit, until all backfill operations are completed.

[0010] Preferably, S1 includes:

[0011] S11: Obtain the soil type, natural moisture content and bearing capacity characteristics of the foundation soil in the prefabricated channel construction area, and determine the target fill material gradation range and initial layer backfill thickness in combination with the requirements for backfill soil in the pre-set roadbed construction technical specifications.

[0012] S12: The fill material to be backfilled is put into the gradation screening unit. The fill material is graded and screened by the three-layer vibrating screen in the gradation screening unit to obtain the screened fill material. The particle size monitoring camera in the gradation screening unit collects the particle image of the screened fill material in real time. The image recognition algorithm is used to analyze the particle image of the screened fill material and calculate the particle ratio of each particle size range.

[0013] S13: If the proportion of particles exceeding the standard in each particle size range is greater than the preset proportion threshold, the crushing component of the gradation screening unit is activated to crush the particles exceeding the standard until qualified filler that meets the target filler gradation range is obtained and transported to the backfilling operation area. At the same time, the gradation test report of each batch of filler is recorded.

[0014] Preferably, an image recognition algorithm is used to analyze the particle image of the packing material after sieving and calculate the proportion of particles in each particle size range, including:

[0015] The particle image of the screened filler was converted into a grayscale image using a weighted average method, and the particle edge contour was extracted from the grayscale image using the Canny operator.

[0016] The geometric features of each particle are calculated based on the particle edge contour, and a particle feature vector matrix is ​​constructed based on the geometric features of all particles. The particle feature vector matrix is ​​then standardized to obtain a standardized particle feature vector matrix.

[0017] The K-means clustering algorithm is used to cluster all the standardized particle feature vectors in the standardized particle feature vector matrix to obtain multiple feature vector clusters.

[0018] Calculate the average equivalent particle size of each feature vector cluster, and map each feature vector cluster to the corresponding target particle size interval according to the average equivalent particle size, thereby determining the particle size interval to which each feature vector cluster belongs.

[0019] The ratio of the total number of standardized particle feature vectors in each feature vector cluster to the total number of all particles is taken as the proportion of particles in the corresponding particle size range.

[0020] Preferably, S2 includes:

[0021] S21: According to the initial layered backfill thickness, sieved qualified fill material is laid on both sides and top of the prefabricated channel to form the first layer of backfill soil. At the same time, compaction degree sensing units are arranged in the first layer of backfill soil according to the preset size grid to collect the compaction degree data of the fill material in real time.

[0022] S22: Start the intelligent vibration compaction unit to compact the first layer of backfill soil, and control the ground radar monitoring unit to scan along the longitudinal direction of the channel at preset intervals during the compaction process to generate an image of the internal density distribution of the backfill soil.

[0023] S23: Calculate the average compaction degree of the fill material based on the compaction degree data. If the average compaction degree of the fill material is not less than the preset compaction degree threshold and there are no obvious voids or loose areas in the density distribution image inside the backfill soil, then stop the compaction operation of the intelligent vibration compaction unit on the first layer of backfill soil. Otherwise, adjust the excitation force and compaction pass of the intelligent vibration compaction unit.

[0024] S24: Repeat steps S21-S23 to carry out subsequent layered backfilling and compaction operations. After each preset layer of backfilling is completed, the settlement value of the top and bottom plate of the prefabricated passage sidewall is measured by the displacement monitoring unit. Combined with the recorded excitation force and number of compaction passes of the intelligent vibration compaction unit, a layered backfilling-settlement correspondence table is generated.

[0025] Preferably, the ground-penetrating radar monitoring unit scans along the longitudinal direction of the channel at preset intervals during the compaction process to generate an image of the internal density distribution of the backfill soil, including:

[0026] During the compaction process, the ground-penetrating radar monitoring unit scans along the longitudinal direction of the channel at preset intervals to obtain multiple scan lines. Each scan line contains a preset number of sampling points, and the propagation time of the reflected wave at each sampling point is collected.

[0027] Based on the propagation time of the reflected waves at all sampling points of all scan lines, a two-dimensional original signal propagation time matrix is ​​constructed. A background removal algorithm is then used to denoise the two-dimensional original signal propagation time matrix to obtain a corrected propagation time matrix.

[0028] The two-dimensional plane coordinates of all sampling points are determined based on the setting rules of all scan lines and all sampling points. The current electromagnetic wave velocity is calculated based on the dielectric constant of the backfill soil. The depth coordinates of each sampling point are calculated based on the current electromagnetic wave velocity and the corrected propagation time matrix. The three-dimensional physical coordinates of each sampling point are determined based on the two-dimensional plane coordinates and depth coordinates of each sampling point.

[0029] The density of each sampling point is calculated based on the maximum dry density of the filler, the standard electromagnetic wave velocity under compacted state, and the current electromagnetic wave velocity. A two-dimensional density matrix is ​​generated based on the density of all sampling points of all scan lines.

[0030] The two-dimensional density matrix is ​​interpolated into a uniform grid matrix using a bicubic interpolation algorithm. Based on the first derivative of each row element and the first derivative of each column element in the uniform grid matrix, all density abrupt change points are determined, and all density interval regions are determined based on all density abrupt change points.

[0031] The uniform grid matrix is ​​color-assigned according to all density intervals to obtain color assignment results, and an image of the density distribution inside the backfill soil is generated based on the color assignment results.

[0032] Preferably, all density interval regions are determined based on all density abrupt change points, including:

[0033] Cluster the three-dimensional physical coordinates of all density mutation points to obtain multiple clusters of neighboring mutation points;

[0034] The depth coordinates of all neighboring mutation points in each neighboring mutation point cluster are sorted to obtain a depth coordinate sequence. Based on the adjacent depth differences in the depth coordinate sequence, all region boundaries of each neighboring mutation point cluster are identified. Based on all the boundaries of each neighboring mutation point cluster, the corresponding depth coordinate sequence is divided to obtain multiple continuous depth sequences and corresponding mutation point sequences of the corresponding neighboring mutation point cluster.

[0035] Extract the density sequence corresponding to each mutation point sequence in the uniform grid matrix, and generate a depth-density two-dimensional data sequence by combining it with the corresponding continuous depth sequence. Use a quadratic polynomial to perform curve fitting on each depth-density two-dimensional data sequence to obtain a density-depth curve with depth as the independent variable and density as the dependent variable.

[0036] Calculate the first and second derivatives of each density depth curve, and based on the first and second derivatives of each density depth curve, determine the combination of adjacent density intervals for each density abrupt change point involved in each density depth curve;

[0037] All density interval regions are determined in a uniform grid matrix by combining adjacent density intervals at each density abrupt change point.

[0038] Preferably, based on the foundation parameters of the prefabricated tunnel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table, the predicted settlement difference after the completion of the subsequent n layers of backfill construction is predicted, including:

[0039] S31: Input the foundation parameters of the prefabricated passage construction area, the gradation test reports of all batches of fill material, and the corresponding table of layered backfill-settlement into the numerical simulation analysis unit to establish a coupled finite element model of prefabricated passage-backfill soil-foundation. The foundation parameters include the foundation soil type, natural moisture content, and characteristic value of foundation bearing capacity.

[0040] S32: Predict the settlement trend after the completion of the subsequent n layers of backfill construction using a coupled finite element model of prefabricated channel-backfill soil-foundation, and calculate the predicted settlement difference.

[0041] Preferably, based on the predicted settlement difference, any one of the following parameters is adjusted for the next round of layered backfilling operations: the sieve vibration frequency of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit, until all backfilling operations are completed, including:

[0042] S33: If the predicted settlement difference does not exceed the threshold of the uneven settlement control standard, then maintain the current layered backfill thickness and compaction parameters;

[0043] S34: If the predicted settlement difference exceeds the threshold of the uneven settlement control standard, analyze the reasons for the predicted settlement difference exceeding the standard.

[0044] S35: Adjust any one of the following parameters in the next round of layered backfilling construction operations based on the predicted cause of excessive settlement difference: sieve vibration frequency of the gradation screening unit, initial layered backfill thickness, and compaction parameters of the intelligent vibration compaction unit.

[0045] S36: Apply the adjusted sieve vibration frequency of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit to the next round of layered backfilling operations, and repeat steps S2-S3 until all backfilling operations are completed.

[0046] Preferably, S35 includes:

[0047] If the predicted cause of excessive settlement difference is fluctuation in the packing gradation, then adjust the vibration frequency of the sieve holes in the gradation screening unit.

[0048] If the predicted cause of excessive settlement difference is excessive layer thickness, then the layer backfill thickness shall be reduced by the proportion of the predicted excessive settlement difference.

[0049] If the predicted cause of excessive settlement difference is insufficient compaction, the excitation force of the intelligent vibration compaction unit is increased, and the compaction control threshold is increased according to the preset increase range.

[0050] This invention provides a prefabricated channel graded backfill optimization device based on uneven settlement control, comprising:

[0051] The gradation screening preparation module is used to determine the target filler gradation range and the initial layered backfill thickness, and to put the filler to be backfilled into the gradation screening unit for gradation screening and crushing to obtain qualified filler that meets the target filler gradation range and transport it to the backfilling operation area. At the same time, the gradation test report of each batch of filler is recorded.

[0052] The layered backfill construction module is used to lay sieved qualified fill material on both sides and top of the prefabricated passage according to the initial layered backfill thickness to form the first layer of backfill soil, and measure the settlement value of the top of the side wall and bottom plate of the prefabricated passage. Combined with the excitation force and number of compaction passes of the intelligent vibration compaction unit, a layered backfill-settlement correspondence table is generated.

[0053] The parameter dynamic adjustment module is used to predict the predicted settlement difference after the completion of the next n layers of backfilling based on the foundation parameters of the prefabricated channel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table. Based on the predicted settlement difference, it adjusts any one of the parameters of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit for the next round of layered backfilling operations until all backfilling operations are completed.

[0054] The beneficial effects of this invention compared to existing technologies are as follows: It determines the target filler gradation range and initial backfill thickness, and performs grading, screening, and crushing of the filler to ensure that the filler entering the backfilling area meets requirements, guaranteeing backfill quality from the source. Recording gradation test reports facilitates traceability. The first layer of backfill is then laid according to the initial thickness, and settlement values ​​are measured and combined with compaction parameters to generate a corresponding relationship table, providing data support for subsequent construction and helping to understand settlement patterns under different parameter combinations. Furthermore, based on foundation parameters, gradation reports, and the corresponding relationship table, the differential settlement of subsequent backfilling is predicted, and key parameters for the next round of construction are adjusted accordingly. This dynamically optimizes the backfilling process, effectively controls uneven settlement of prefabricated tunnels, improves the stability and safety of the tunnel structure, ensures the entire backfilling operation is carried out scientifically and efficiently, and enhances project quality.

[0055] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0056] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0057] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0058] Figure 1 This is a flowchart of the prefabricated channel graded backfill optimization method based on uneven settlement control in an embodiment of the present invention.

[0059] Figure 2 This is a flowchart illustrating the implementation of step S1 in an embodiment of the present invention;

[0060] Figure 3 This is a flowchart illustrating the implementation of step S2 in an embodiment of the present invention;

[0061] Figure 4 This is a flowchart illustrating the implementation of step S3 in an embodiment of the present invention. Detailed Implementation

[0062] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0063] like Figure 1 As shown, this invention provides an implementation method for a prefabricated channel graded backfill optimization method based on non-uniform settlement control, comprising:

[0064] S1: Determine the target packing gradation range and initial layered backfill thickness, and put the packing to be backfilled into the gradation screening unit for gradation screening and crushing to obtain qualified packing that meets the target packing gradation range and transport it to the backfilling operation area. At the same time, record the gradation test report of each batch of packing.

[0065] S2: According to the initial layered backfill thickness, sieved qualified fill material is laid on both sides and top of the prefabricated passage to form the first layer of backfill soil. The settlement value of the top of the side wall and bottom plate of the prefabricated passage is measured. Combined with the excitation force and compaction pass of the intelligent vibration compaction unit, a layered backfill-settlement correspondence table is generated.

[0066] S3: Based on the foundation parameters of the prefabricated channel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table, predict the predicted settlement difference after the completion of the subsequent n layers (n=5-10) of backfill construction. Based on the predicted settlement difference, adjust any one of the following parameters in the next round of layered backfill construction: the sieve vibration frequency of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit, until all backfill operations are completed.

[0067] In this embodiment, the target filler gradation range refers to the filler gradation standard determined based on the foundation soil type (cohesive soil / sandy soil / gravelly soil), natural moisture content, and foundation bearing capacity characteristics of the prefabricated tunnel construction area, combined with the requirements for backfill soil in the "Technical Specification for Highway Subgrade Construction" (JTG / T3610-2019). Specifically, it is a particle size ≤50mm with a particle ratio ≥90%, a uniformity coefficient between 5 and 10, and a curvature coefficient between 1 and 3.

[0068] In this embodiment, the initial backfill thickness is determined based on the different foundation soil types, with each layer of backfill soil being 30cm thick for sandy soil, 25cm thick for cohesive soil, and 35cm thick for gravelly soil.

[0069] In this embodiment, the backfill material is the material prepared for backfilling the prefabricated channel. In this embodiment, crushed stone from mountainous areas is used as an example.

[0070] In this embodiment, the gradation screening unit is a device unit for processing backfill material, comprising a three-layer vibrating screen (screen aperture sizes of 50mm, 20mm, and 5mm respectively) and a particle size monitoring camera (resolution 2048×1536, frame rate 15fps). Its function is to grade and screen the input backfill material. The particle size monitoring camera acquires real-time images of the particles in the screened material and calculates the proportion of particles in each size range using an image recognition algorithm. If the screened material does not meet the target gradation range, a crushing component (crushing pressure 15MPa) will be activated to crush the excessive particles, ensuring that the final screened material meets the requirements.

[0071] In this embodiment, the backfilling work area is the area where prefabricated channel backfilling construction is carried out, and qualified fillers that meet the target filler gradation range are transported here.

[0072] In this embodiment, the settlement values ​​of the top and bottom slabs of the prefabricated passage sidewalls are measured: During the backfilling construction of the prefabricated passage, after every three layers of backfilling are completed, the vertical movement distance of the top and bottom slabs of the prefabricated passage sidewalls relative to their initial positions is measured using a GNSS displacement monitoring unit (planar accuracy ±2mm, elevation accuracy ±5mm), and the settlement value s (mm) is obtained. These settlement values ​​are used to generate a layered backfilling-settlement correspondence table.

[0073] In this embodiment, the excitation force and number of compaction passes of the intelligent vibratory compaction unit are: the excitation force (300-400kN) generated by the intelligent vibratory compaction unit during operation and the number of compaction operations performed on the backfill soil.

[0074] In this embodiment, the layered backfill-settlement correspondence table records the settlement values ​​of the top of the sidewalls and the bottom slab after every three layers of backfilling during the layered backfilling process of the prefabricated passage, as well as the excitation force and number of compaction passes of the intelligent vibration compaction unit during each layer of backfilling. For example, the table may record information such as an excitation force of 320kN and 5 compaction passes during the 1st to 3rd layers of backfilling, with a corresponding settlement value of 5mm.

[0075] In this embodiment, the predicted settlement difference after the completion of the subsequent n layers of backfill construction is calculated by inputting foundation parameters, filler gradation data, compaction parameters and settlement values ​​into the numerical simulation analysis unit, and using the prefabricated channel-backfill soil-foundation coupled finite element model established by FLAC3D software to simulate the settlement after the completion of the subsequent n layers of backfill construction, and calculating the difference between the maximum and minimum longitudinal settlement of the channel, i.e., the predicted settlement difference.

[0076] In this embodiment, the sieve vibration frequency refers to the vibration frequency of the vibrating screen holes in the gradation screening unit. Adjusting this frequency can optimize the uniformity of the post-screening filler. For example, if the proportion of 20-50mm particles in the post-screening filler is <30% (below the lower limit of the target range), the vibration frequency of the 50mm sieve hole is increased by 2Hz to enhance the screening efficiency for larger particles, increasing the proportion of 20-50mm particles to 30-40%. If the proportion of <5mm particles in the post-screening filler is >20% (above the upper limit of the target range), the vibration frequency of the 5mm sieve hole is decreased by 2Hz to reduce the amount of fine particles passing through, reducing the proportion of <5mm particles to 10-20%. After adjustment, the gradation data of three batches of filler must be continuously monitored to ensure that the gradation meets the target range.

[0077] In this embodiment, the compaction parameters of the intelligent vibration compaction unit mainly refer to the excitation force and the number of compaction passes of the intelligent vibration compaction unit.

[0078] like Figure 2 As shown, to ensure that the packing material meets the gradation requirements and to record the test report, S1 is proposed, including...

[0079] S11: Obtain the soil type, natural moisture content and bearing capacity characteristics of the foundation soil in the prefabricated channel construction area, and determine the target fill material gradation range and initial layer backfill thickness in combination with the requirements for backfill soil in the pre-set roadbed construction technical specifications.

[0080] S12: The fill material to be backfilled is put into the gradation screening unit. The fill material is graded and screened by the three-layer vibrating screen in the gradation screening unit to obtain the screened fill material. The particle size monitoring camera in the gradation screening unit collects the particle image of the screened fill material in real time. The image recognition algorithm is used to analyze the particle image of the screened fill material and calculate the particle ratio of each particle size range.

[0081] S13: If the proportion of particles exceeding the standard in each particle size range is greater than the preset proportion threshold, the crushing component of the gradation screening unit is activated to crush the particles exceeding the standard until qualified filler that meets the target filler gradation range is obtained and transported to the backfilling operation area. At the same time, the gradation test report of each batch of filler is recorded.

[0082] In this embodiment, the prefabricated passage construction area refers to the specific site area where the prefabricated passage construction project is carried out.

[0083] In this embodiment, the foundation soil type is divided into categories such as cohesive soil, sandy soil, and gravelly soil.

[0084] In this embodiment, natural moisture content refers to the ratio of the mass of water contained in the foundation soil in its natural state to the mass of soil particles, expressed as a percentage (%).

[0085] In this embodiment, the characteristic value of foundation bearing capacity refers to the pressure value corresponding to the specified deformation within the linear deformation segment of the foundation soil pressure-deformation curve determined by load tests, and its unit is kilopascal (kPa). It represents the ability of the foundation soil to withstand the load of the superstructure and is a key indicator for measuring the stability and bearing capacity of the foundation.

[0086] In this embodiment, the requirements for backfill soil in the pre-defined roadbed construction technical specifications are based on the various standards for backfill soil stipulated in relevant industry standards such as the "Technical Specifications for Highway Roadbed Construction" (JTG / T3610-2019). For example, the target fill material gradation range requires that the proportion of particles with a particle size ≤50mm be ≥90%, the non-uniformity coefficient be between 5 and 10, and the curvature coefficient be between 1 and 3.

[0087] In this embodiment, the soil type, natural moisture content, and bearing capacity characteristics of the prefabricated tunnel construction area are obtained. Combined with the requirements for backfill soil in the pre-defined roadbed construction technical specifications, the target fill material gradation range and initial layer backfill thickness are determined. Basic data such as soil type, natural moisture content, and bearing capacity characteristics of the prefabricated tunnel construction area are obtained through geological surveys. Then, based on the pre-defined requirements for backfill soil in documents such as the "Highway Roadbed Construction Technical Specifications," these factors are comprehensively considered to determine the appropriate target fill material gradation range and initial layer backfill thickness.

[0088] In this embodiment, the particle proportion of each particle size range refers to the percentage of particles in different size ranges relative to the total number of particles in the backfill material after it has passed through the vibrating screen of the grading screening unit. For example, if the particle size is divided into ranges such as >50mm, 20-50mm, 5-20mm, and <5mm, images are captured by a particle size monitoring camera, and an image recognition algorithm is used to calculate the particle proportion of each range. For instance, the calculation shows that the particle proportion in the >50mm size range is 3%, and the particle proportion in the 20-50mm size range is 35%, etc.

[0089] In this embodiment, the preset percentage threshold is a set standard ratio value used to determine whether the percentage of particles in a certain size range in the packing material after sieving exceeds the standard.

[0090] In this embodiment, the percentage of particles exceeding the standard is the percentage of particles in the filler material after sieving whose particle size exceeds a preset threshold (e.g., 50 mm).

[0091] In this embodiment, the crushing component of the gradation screening unit is activated to crush particles exceeding the standard size: when the proportion of particles in a certain size range in the backfill exceeds a preset threshold, the crushing component within the gradation screening unit is activated to ensure the backfill gradation meets the target range. This component applies a crushing pressure of 15 MPa to crush the particles exceeding the standard size. For example, when the proportion of particles >50 mm exceeds the standard, the crushing component crushes the large particles, reducing their size. After re-screening, the proportion of particles in each size range conforms to the target gradation range, thereby ensuring the quality and performance of the backfill soil and reducing the risk of uneven settlement.

[0092] To accurately analyze the particle size distribution of the post-screening filler, an image recognition algorithm is proposed to analyze the particle images of the post-screening filler and calculate the proportion of particles in each size range, including:

[0093] The particle image of the screened filler was converted into a grayscale image using a weighted average method, and the particle edge contour was extracted from the grayscale image using the Canny operator.

[0094] The geometric features of each particle are calculated based on the particle edge contour, and a particle feature vector matrix is ​​constructed based on the geometric features of all particles. The particle feature vector matrix is ​​then standardized to obtain a standardized particle feature vector matrix.

[0095] The K-means clustering algorithm is used to cluster all the standardized particle feature vectors in the standardized particle feature vector matrix to obtain multiple feature vector clusters.

[0096] Calculate the average equivalent particle size of each feature vector cluster, and map each feature vector cluster to the corresponding target particle size interval according to the average equivalent particle size, thereby determining the particle size interval to which each feature vector cluster belongs.

[0097] The ratio of the total number of standardized particle feature vectors in each feature vector cluster to the total number of all particles is taken as the proportion of particles in the corresponding particle size range.

[0098] In this embodiment, the particle image of the screened packing material is converted into a grayscale image using a weighted average method, i.e., a weighted average formula is employed. The channel values ​​of each pixel in the particle image are calculated for the color particle image of the packing material after sieving, and then converted into a grayscale image.

[0099] In this embodiment, the edge contours of particles are extracted from the grayscale image using the Canny operator. The Canny operator is used to set a threshold range of 50-150 to identify and extract the edge contour information of particles from the converted grayscale image.

[0100] In this embodiment, the geometric features of each particle are calculated based on the particle edge contour. Based on the extracted particle edge contour, the equivalent particle size (calculated according to the diameter of the particle's circumscribed circle), perimeter, area, and other geometric feature data of each particle are calculated.

[0101] In this embodiment, a particle feature vector matrix is ​​constructed based on the geometric features of all particles. The geometric features of each particle, such as the equivalent particle size, perimeter, and area, are combined into a feature vector. An m×3 matrix is ​​constructed using the feature vectors of all particles (m is the number of particles, and each row corresponds to the [equivalent particle size, perimeter, and area] of a single particle).

[0102] In this embodiment, the particle feature vector matrix is ​​standardized to obtain a standardized particle feature vector matrix. The ratios of the equivalent particle diameter, perimeter, and area of ​​the corresponding particle to the largest equivalent particle diameter, largest perimeter, and largest area among all particles are taken as the corresponding standardized values. The standardized calculation is performed on each element in the particle feature vector matrix to obtain the standardized particle feature vector matrix.

[0103] In this embodiment, the K-means clustering algorithm is used to cluster all standardized particle feature vectors in the standardized particle feature vector matrix, obtaining multiple feature vector clusters. K=4 is set (corresponding to four particle size ranges: >50mm, 20-50mm, 5-20mm, <5mm). Four particle feature vectors are randomly selected as initial cluster centers. The Euclidean distance between each particle feature vector and each cluster center is calculated, and the particles are assigned to the cluster with the smallest distance. The centers of each cluster are recalculated, and this process is repeated until the change in cluster centers is less than 1. Thus, four feature vector clusters are obtained.

[0104] In this embodiment, the average equivalent particle size of each feature vector cluster is calculated, the equivalent particle sizes of all particles in each feature vector cluster are added together, and then divided by the number of particles in the cluster to obtain the average equivalent particle size.

[0105] In this embodiment, each feature vector cluster is assigned to a corresponding target particle size interval according to the average equivalent particle size, and the particle size interval to which each feature vector cluster belongs is determined. The calculated average equivalent particle size is compared with four target particle size intervals (>50mm, 20-50mm, 5-20mm, <5mm). If the average equivalent particle size is >50mm, it corresponds to the >50mm interval.

[0106] like Figure 3 As shown, in order to accurately obtain the correlation data between backfilling and settlement, S2 is proposed, including:

[0107] S21: According to the initial layered backfill thickness, sieved qualified fill material is laid on both sides and top of the prefabricated channel to form the first layer of backfill soil. At the same time, compaction degree sensing units are arranged in the first layer of backfill soil according to the preset size grid to collect the compaction degree data of the fill material in real time.

[0108] S22: Start the intelligent vibration compaction unit to compact the first layer of backfill soil, and control the ground radar monitoring unit to scan along the longitudinal direction of the channel at preset intervals during the compaction process to generate an image of the internal density distribution of the backfill soil.

[0109] S23: Calculate the average compaction degree of the fill material based on the compaction degree data. If the average compaction degree of the fill material is not less than the preset compaction degree threshold and there are no obvious voids or loose areas in the density distribution image inside the backfill soil, then stop the compaction operation of the intelligent vibration compaction unit on the first layer of backfill soil. Otherwise, adjust the excitation force and compaction pass of the intelligent vibration compaction unit.

[0110] S24: Repeat steps S21-S23 to carry out subsequent layered backfilling and compaction operations. After each preset layer of backfilling is completed, the settlement value of the top and bottom plate of the prefabricated passage sidewall is measured by the displacement monitoring unit. Combined with the recorded excitation force and number of compaction passes of the intelligent vibration compaction unit, a layered backfilling-settlement correspondence table is generated.

[0111] In this embodiment, the preset size grid refers to the 5m×5m grid specification used when arranging the compaction sensing unit in the first layer of backfill soil. Arranging the sensors according to this rule can collect compaction data of the fill material at different locations more evenly.

[0112] In this embodiment, the compaction degree sensing unit collects the compaction degree data of the fill material. The electromagnetic compaction degree sensor (measurement range 0-100%, accuracy ±1%) collects the compaction degree value of the fill material in real time at each point of the preset size grid. These data can intuitively reflect the compaction effect of the backfill soil.

[0113] In this embodiment, the ground-penetrating radar monitoring unit is a device with a center frequency of 1.5 GHz and a detection depth in the range of 0-1.5 m. During the backfill compaction process, it detects the internal structure of the backfill soil by emitting and receiving electromagnetic waves and obtains information related to the compaction degree.

[0114] In this embodiment, scanning along the longitudinal direction of the channel at preset intervals means that the ground-penetrating radar monitoring unit performs a scanning operation every 2m along the longitudinal direction of the prefabricated channel. By scanning at fixed intervals, information about the backfill soil at different longitudinal positions of the channel can be systematically obtained.

[0115] In this embodiment, the density distribution image inside the backfill soil is generated by the ground-penetrating radar monitoring unit after scanning the backfill soil, followed by signal processing and a series of calculations. The image uses different colors (e.g., red for loose, yellow for acceptable, and green for dense) to show the density of different areas inside the backfill soil, allowing people to intuitively understand the internal quality distribution.

[0116] In this embodiment, the average compaction degree of the fill material is calculated based on the compaction degree data. This is achieved by adding up all the compaction degree data collected by the compaction degree sensing unit at each point of a preset size grid, and then dividing by the number of data collection points. The resulting value is the average compaction degree of the backfill soil layer, which represents the overall compaction degree of this backfill soil layer.

[0117] In this embodiment, the preset compaction threshold is set at 93% according to the compaction requirements of highway subgrade. This is the standard value for judging whether the compaction of the backfill soil layer is qualified. When the average compaction degree reaches or exceeds this threshold, it indicates that the compaction degree of the backfill soil layer initially meets the requirements.

[0118] In this embodiment, the density distribution image of the backfill soil does not contain obvious voids or loose areas. This means that by analyzing and identifying the generated density distribution image, no void areas with a volume greater than 0.01 m³ or loose areas with a compaction degree less than 90% were found. This indicates that the internal structure of the backfill soil is relatively uniform and the density is good. Specifically, to identify void areas with a volume greater than 0.01 m³ based on the two-dimensional density distribution image of the backfill soil, it is necessary to first find the area representing the looseness in the image, obtain its coordinates, calculate the horizontal area S based on the correspondence between the image and the actual physical coordinates through pixel statistics, determine the depth h according to the depth mapping rule, calculate the volume using V=S×h, and then compare it with 0.01 m³. If it is greater than 0.01 m³, it is determined that there is an excessive void area; otherwise, it does not exist.

[0119] In this embodiment, the excitation force and number of compaction passes of the intelligent vibration compaction unit are adjusted. If the average compaction degree does not reach the preset compaction degree threshold or obvious voids or loose areas are found in the density distribution image, the working parameters of the intelligent vibration compaction unit need to be adjusted. Specifically, the excitation force is increased by 5% at a time, and one compaction operation is added each time. Then the compaction operation is repeated until the compaction requirements are met.

[0120] In this embodiment, the settlement values ​​of the top and bottom slabs of the prefabricated passage sidewalls are measured by a displacement monitoring unit. Combined with the recorded excitation force and compaction pass number of the intelligent vibration compaction unit, a layered backfill-settlement correspondence table is generated. After every 3 layers of backfilling are completed, the settlement values ​​of the top and bottom slabs of the prefabricated passage sidewalls are measured by a GNSS displacement monitoring unit (planar accuracy ±2mm, elevation accuracy ±5mm). These settlement values ​​are then correlated with the recorded excitation force and compaction pass number of the intelligent vibration compaction unit for the corresponding layer to form a table.

[0121] To accurately represent the internal density of the backfill soil, it is proposed that the ground-penetrating radar monitoring unit scan along the longitudinal direction of the channel at preset intervals during the compaction process, generating an image of the internal density distribution of the backfill soil, including:

[0122] During the compaction process, the ground-penetrating radar monitoring unit scans along the longitudinal direction of the channel at preset intervals to obtain multiple scan lines. Each scan line contains a preset number of sampling points, and the propagation time of the reflected wave at each sampling point is collected.

[0123] Based on the propagation time of the reflected waves at all sampling points of all scan lines, a two-dimensional original signal propagation time matrix is ​​constructed. A background removal algorithm is then used to denoise the two-dimensional original signal propagation time matrix to obtain a corrected propagation time matrix.

[0124] The two-dimensional plane coordinates of all sampling points are determined based on the setting rules of all scan lines and all sampling points. The current electromagnetic wave velocity is calculated based on the dielectric constant of the backfill soil. The depth coordinates of each sampling point are calculated based on the current electromagnetic wave velocity and the corrected propagation time matrix. The three-dimensional physical coordinates of each sampling point are determined based on the two-dimensional plane coordinates and depth coordinates of each sampling point.

[0125] The density of each sampling point is calculated based on the maximum dry density of the filler, the standard electromagnetic wave velocity under compacted state, and the current electromagnetic wave velocity. A two-dimensional density matrix is ​​generated based on the density of all sampling points of all scan lines.

[0126] The two-dimensional density matrix is ​​interpolated into a uniform grid matrix using a bicubic interpolation algorithm. Based on the first derivative of each row element and the first derivative of each column element in the uniform grid matrix, all density abrupt change points are determined, and all density interval regions are determined based on all density abrupt change points.

[0127] The uniform grid matrix is ​​color-assigned according to all density intervals to obtain color assignment results, and an image of the density distribution inside the backfill soil is generated based on the color assignment results.

[0128] In this embodiment, the ground-penetrating radar monitoring unit is controlled to scan along the longitudinal direction of the channel at preset intervals during the compaction process to obtain multiple scan lines. Each scan line contains a preset number of sampling points. That is, during the compaction of the backfill soil, the ground-penetrating radar monitoring unit is operated to scan along the longitudinal direction of the channel every 2m. Each scan yields one scan line, and each scan line contains a fixed 500 sampling points, thereby obtaining a large number of data points reflecting the internal information of the backfill soil.

[0129] In this embodiment, the propagation time of the reflected wave at each sampling point is collected. When the ground-penetrating radar monitoring unit is working, it records the time taken for the reflected wave to travel from transmission to reception for each of the 500 sampling points along each scan line.

[0130] In this embodiment, a two-dimensional original signal propagation time matrix is ​​constructed based on the propagation time of reflected waves from all sampling points of all scan lines. The propagation time data of reflected waves collected from each sampling point of each scan line are organized into a two-dimensional matrix. The rows of the matrix correspond to different scan lines (length of the longitudinal construction section of the common channel / 2m scan lines), and the columns correspond to 500 sampling points on each scan line. The matrix elements are the propagation time of reflected waves at each sampling point (in ns), thus visually presenting all the collected propagation time data.

[0131] In this embodiment, a background removal algorithm is used to denoise the two-dimensional original signal propagation time matrix to obtain a corrected propagation time matrix. By calculating the mean of all elements in the two-dimensional original signal propagation time matrix, the mean is subtracted from each element in the matrix to remove system noise. Then, a 3×3 sliding window filter (the mean of elements in the window is taken) is used to further smooth the data and eliminate single-point time anomalies caused by electromagnetic interference in the construction environment, thereby obtaining a corrected propagation time matrix that more accurately reflects the true situation of the backfill soil.

[0132] In this embodiment, the setting rules for all scan lines and all sampling points are as follows: one scan line is set every 2m along the longitudinal direction of the channel (the number of scan lines is determined according to the length of the longitudinal construction section of the channel, which is the length of the longitudinal construction section of the channel / 2m, and is taken as a positive integer); the 500 sampling points on each scan line are evenly distributed in the horizontal direction at intervals of 0.01m (from 0 to 5m), covering a horizontal detection range of 2.5m on both sides of the channel.

[0133] In this embodiment, the two-dimensional plane coordinate values ​​of all sampling points are determined based on the setting rules for all scan lines and all sampling points. According to the setting rules, the X-coordinate of the i-th scan line is set as follows: This indicates the position of the scan line along the longitudinal direction of the channel; each scan line has 500 sampling points arranged according to... Determine its horizontal coordinates to obtain the position coordinates of each sampling point on the two-dimensional plane.

[0134] In this embodiment, the current electromagnetic wave velocity is calculated based on the dielectric constant of the backfill soil, and the depth coordinates of each sampling point are calculated based on the current electromagnetic wave velocity and the corrected propagation time matrix. The depth coordinates of each sampling point are then determined according to the dielectric constant of the backfill soil. (Indoor tests on crushed stone filler in mountainous areas determined ε=3-5), using the formula ( Speed ​​of light Calculate the propagation speed of electromagnetic waves in backfill soil. Then, based on the modified propagation time matrix, according to... ( To correct the propagation time matrix, the propagation time unit is... Convert to Calculate the depth coordinates of each sampling point. Determine its position in the vertical direction.

[0135] In this embodiment, the three-dimensional physical coordinates of each sampling point are determined based on the two-dimensional planar coordinates and depth coordinates of each sampling point, and the previously obtained two-dimensional planar coordinates of each sampling point are used to... With the calculated depth coordinates By combining these, the physical coordinates of each sampling point in three-dimensional space can be obtained. This allows for precise location of each sampling point within the backfill soil.

[0136] In this embodiment, the maximum dry density of the fill material refers to the unit volume mass of the backfill soil in its densest state, as determined by indoor compaction tests. In this embodiment, the maximum dry density of the crushed stone fill material in the mountainous area is... It is an important parameter for calculating the compaction of backfill soil.

[0137] In this embodiment, the standard electromagnetic wave velocity under compacted conditions is the propagation speed of electromagnetic waves in compacted backfill soil, calibrated through experiments. It is used to calculate the compaction of backfill soil by comparing it with the current electromagnetic wave velocity.

[0138] In this embodiment, the density of each sampling point is calculated based on the maximum dry density of the filler, the standard electromagnetic wave velocity under compacted conditions, and the current electromagnetic wave velocity, according to the formula. The current electromagnetic wave velocity corresponding to each sampling point Maximum dry density of filler and the standard electromagnetic wave velocity in a dense state Substitute the values ​​and calculate the density of each sampling point. (Unit: kg / m³), reflecting the compaction density of the backfill soil at that point.

[0139] In this embodiment, a two-dimensional density matrix is ​​generated based on the density of all sampling points of all scan lines. The density data calculated for each sampling point of all scan lines is organized into a two-dimensional matrix. The rows of the matrix correspond to the scan lines, the columns correspond to the sampling points, and the elements are the density of each sampling point, thereby comprehensively displaying the density distribution of each location inside the backfill soil.

[0140] In this embodiment, a bicubic interpolation algorithm is used to interpolate the two-dimensional compaction matrix into a uniform grid matrix. The bicubic interpolation algorithm is then used to process the two-dimensional compaction matrix, converting it into a uniform grid matrix with a grid size of 0.1m × 0.1m. This improves the resolution of the data, making the subsequent analysis of the compaction of the backfill soil and the generation of images more accurate, and enabling the clear identification of voids such as those with a volume > 0.01m³.

[0141] In this embodiment, all density abrupt change points are determined based on the first derivatives of each row and column element in the uniform grid matrix. The first derivative is calculated for each row of data in the uniform grid matrix along the column direction (depth direction) and for each column of data along the row direction (the ratio of the density difference between adjacent elements to the corresponding coordinate difference). When the first derivative at a point along either the row or column direction is < -50 kg / (m³·m), that point is determined to be a density abrupt change point. These points may correspond to defects such as backfill voids or loose areas.

[0142] In this embodiment, all density range regions include: loose range (low density that decreases rapidly with depth), acceptable range (medium density that changes gradually), dense range (high density that increases rapidly with depth), and other different density range regions.

[0143] In this embodiment, the uniform grid matrix is ​​color-assigned according to all density intervals to obtain color assignment results. Based on the color assignment results, an image of the density distribution inside the backfill soil is generated. Then, according to the color standard corresponding to different density intervals, each element in the uniform grid matrix is ​​color-assigned. For example, density... (For areas with a compaction degree <90%, loose areas) are assigned a red value; (For areas with a compaction degree of 90%-93%, considered acceptable) the value is assigned as yellow; (Corresponding to compaction degree > 93%, dense area) is assigned a green value. A color image is generated based on the color assignment results. The horizontal axis of the image represents the longitudinal distance of the channel, and the vertical axis represents the horizontal detection distance. This visually displays the internal density distribution of the backfill soil and the location of abnormal areas, making it easier for construction personnel to understand the quality status of the backfill soil.

[0144] To accurately determine the compaction range of backfill soil, a method is proposed to identify all compaction range regions based on all compaction abrupt change points, including:

[0145] Cluster the three-dimensional physical coordinates of all density mutation points to obtain multiple clusters of neighboring mutation points;

[0146] The depth coordinates of all neighboring mutation points in each neighboring mutation point cluster are sorted to obtain a depth coordinate sequence. Based on the adjacent depth differences in the depth coordinate sequence, all region boundaries of each neighboring mutation point cluster are identified. Based on all the boundaries of each neighboring mutation point cluster, the corresponding depth coordinate sequence is divided to obtain multiple continuous depth sequences and corresponding mutation point sequences of the corresponding neighboring mutation point cluster.

[0147] Extract the density sequence corresponding to each mutation point sequence in the uniform grid matrix, and generate a depth-density two-dimensional data sequence by combining it with the corresponding continuous depth sequence. Use a quadratic polynomial to perform curve fitting on each depth-density two-dimensional data sequence to obtain a density-depth curve with depth as the independent variable and density as the dependent variable.

[0148] Calculate the first and second derivatives of each density depth curve, and based on the first and second derivatives of each density depth curve, determine the combination of adjacent density intervals for each density abrupt change point involved in each density depth curve;

[0149] All density interval regions are determined in a uniform grid matrix by combining adjacent density intervals at each density abrupt change point.

[0150] In this embodiment, clustering the three-dimensional physical coordinates of all density mutation points to obtain multiple neighboring mutation point clusters means collecting the three-dimensional coordinates (including channel longitudinal distance, horizontal detection distance and depth) of all previously determined density mutation points, and using a clustering algorithm (such as the DBSCAN clustering algorithm) to group points with similar spatial locations together based on the distance relationship between points, thereby obtaining multiple neighboring mutation point clusters.

[0151] In this embodiment, a neighboring mutation point refers to a mutation point within the same neighboring mutation point cluster.

[0152] In this embodiment, all region boundaries of each cluster of neighboring mutation points are identified based on the depth differences between adjacent depths in the depth coordinate sequence. For each cluster of neighboring mutation points, the depth coordinates of all mutation points are extracted and arranged in ascending order to form a depth coordinate sequence. Then, the difference between adjacent depth values ​​in the sequence is calculated. When the difference between adjacent depths is greater than 0.1m, the corresponding point is identified as a region boundary. These region boundaries mark the boundaries of regions with different density variations within the cluster.

[0153] In this embodiment, the corresponding depth coordinate sequence is divided based on all the boundary points of each neighboring mutation point cluster, resulting in multiple continuous depth sequences and corresponding mutation point sequences for each neighboring mutation point cluster. Using the identified region boundary points as limits, the depth coordinate sequence of each neighboring mutation point cluster is divided into multiple small segments, each of which is a continuous depth sequence. Simultaneously, mutation points belonging to each continuous depth sequence are grouped together to form a corresponding mutation point sequence.

[0154] In this embodiment, the density sequence corresponding to each mutation point sequence is extracted from the uniform grid matrix, and combined with the corresponding continuous depth sequence to generate a depth-density two-dimensional data sequence. Based on the position of each mutation point sequence in the uniform grid matrix, the density values ​​corresponding to these points are extracted to form a density sequence. Then, this density sequence is combined with the corresponding continuous depth sequence to construct a depth-density two-dimensional data sequence.

[0155] In this embodiment, a quadratic polynomial is used to perform curve fitting on each depth-density two-dimensional data sequence to obtain a density-depth curve with depth as the independent variable and density as the dependent variable. Mathematical methods are employed to perform quadratic polynomial fitting on each depth-density two-dimensional data sequence. That is, a quadratic polynomial function is found that describes the relationship between depth and density as accurately as possible; the curve corresponding to this function is the density-depth curve.

[0156] In this embodiment, based on the first and second derivatives of each density depth curve, the combination of adjacent density intervals for each density abrupt change point involved in each density depth curve is determined. The first and second derivatives of the density depth curves are then calculated. For example, at a density abrupt change point, if the value of the first derivative is large and negative, it indicates that the density decreases rapidly with depth; simultaneously, if the second derivative is negative, it means that the rate of density decrease is accelerating. Combined with pre-defined density interval standards, this situation aligns with the setting that low density and a rapid rate of change correspond to a loose interval.

[0157] If the density to the left of the mutation point was originally in the acceptable range (e.g., density between 850-930 kg / m³), while the right side is determined to be in the loose range due to the aforementioned derivative characteristics (e.g., density below 850 kg / m³ and a rapid rate of decrease), then the combination of density ranges corresponding to both sides of this density mutation point is from the acceptable range to the loose range.

[0158] For example, if at another abrupt change point the first derivative is positive and has a large value, and the second derivative is also positive, it indicates that the density increases with depth and the rate of increase is accelerating. If the density to the left of the abrupt change point is within the acceptable range, and the density to the right meets the standard of "high density and rapid increase with depth corresponding to a dense range (e.g., density higher than 930 kg / m³ and a rapid rate of increase)" due to this derivative characteristic, then the combination of density ranges on both sides of this abrupt change point represents the transition from the acceptable range to the dense range.

[0159] In this embodiment, all density interval regions are determined in the uniform grid matrix based on the adjacent density interval combinations of each density mutation point. According to the adjacent density interval combination information of each density mutation point, the corresponding position range is found in the uniform grid matrix, and regions with the same or similar density interval combination characteristics are defined as a density interval region.

[0160] like Figure 4 As shown, to provide data support for construction adjustments, a method is proposed to predict the differential settlement after the completion of subsequent n layers of backfill, based on the foundation parameters of the prefabricated tunnel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table. This includes:

[0161] S31: Input the foundation parameters of the prefabricated passage construction area, the gradation test reports of all batches of fill material, and the corresponding table of layered backfill-settlement into the numerical simulation analysis unit to establish a coupled finite element model of prefabricated passage-backfill soil-foundation. The foundation parameters include the foundation soil type, natural moisture content, and characteristic value of foundation bearing capacity.

[0162] S32: Predict the settlement trend after the completion of the subsequent n layers of backfill construction using a coupled finite element model of prefabricated channel-backfill soil-foundation, and calculate the predicted settlement difference.

[0163] In this embodiment, the numerical simulation analysis unit is a tool that uses professional software (such as FLAC3D software) to simulate and analyze the construction process of prefabricated passageways. It can integrate various types of construction data and simulate actual construction conditions through specific algorithms and models, providing a basis for adjusting construction parameters and predicting results.

[0164] In this embodiment, the foundation parameters of the prefabricated tunnel construction area (such as foundation soil type, natural moisture content, and characteristic values ​​of foundation bearing capacity), the gradation test reports of all batches of fill material (including the proportion of each particle size range, screening time, etc.), and the layered backfill-settlement correspondence table (recording the compaction parameters and corresponding settlement values ​​of each layer of backfill) are input into the numerical simulation analysis unit to establish a coupled finite element model of the prefabricated tunnel-backfill soil-foundation. This means that by using the numerical simulation analysis unit, these key data from actual construction are used as input conditions, and based on finite element theory, the prefabricated tunnel, backfill soil, and foundation are modeled as an interacting whole system. In the model, the backfill soil adopts the Mohr-Coulomb constitutive model (its cohesion c and internal friction angle φ are determined based on the indoor test of the screened fill material, c=15-25kPa, φ=25-35°), and the prefabricated tunnel adopts the elastic constitutive model (elastic modulus 34.5GPa, Poisson's ratio 0.2) to simulate their mechanical response and interaction relationship in actual construction.

[0165] In this embodiment, a coupled finite element model of the prefabricated tunnel-backfill soil-foundation is used to predict the settlement trend after the completion of the subsequent n layers of backfill construction, and the predicted settlement difference is calculated. This involves using the established coupled finite element model to simulate the construction process of laying and compacting the subsequent n layers of backfill soil, analyzing the settlement changes of the prefabricated tunnel during this process, and thus predicting the settlement trend, i.e., the direction of settlement change over time or construction progress. Simultaneously, the difference between the maximum and minimum longitudinal settlement of the tunnel is calculated, which is the predicted settlement difference.

[0166] like Figure 4 As shown, in order to effectively control uneven settlement, it is proposed to adjust any one of the following parameters in the next round of layered backfilling construction operations: the vibration frequency of the sieve aperture of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit, based on the predicted settlement difference, until all backfilling operations are completed, including:

[0167] S33: If the predicted settlement difference does not exceed the threshold of the uneven settlement control standard, then maintain the current layered backfill thickness and compaction parameters;

[0168] S34: If the predicted settlement difference exceeds the threshold of the uneven settlement control standard, analyze the reasons for the predicted settlement difference exceeding the standard.

[0169] S35: Adjust any one of the following parameters in the next round of layered backfilling construction operations based on the predicted cause of excessive settlement difference: sieve vibration frequency of the gradation screening unit, initial layered backfill thickness, and compaction parameters of the intelligent vibration compaction unit.

[0170] S36: Apply the adjusted sieve vibration frequency of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit to the next round of layered backfilling operations, and repeat steps S2-S3 until all backfilling operations are completed.

[0171] In this embodiment, the non-uniform settlement control standard threshold refers to the specific numerical limit used to determine whether the prefabricated passage has excessive non-uniform settlement, which is 10mm / 10m in this embodiment.

[0172] In this embodiment, the reasons for the predicted settlement difference exceeding the standard are analyzed. When the predicted settlement difference exceeds the non-uniform settlement control standard threshold of 10mm / 10m, the actual proportion of each particle size range in the test reports of each batch of filler gradation is compared with the target range by collecting the test reports of each batch of filler gradation. If the proportion of a certain particle size range deviates from the target range by ±3%, it may be due to the fluctuation of filler gradation causing the settlement difference to exceed the standard. The layer thickness data in the construction record is checked and compared with the initial layer backfill thickness. If the actual layer thickness is significantly too large, it may be due to the excessive layer thickness. The data collected by the compaction degree sensing unit is reviewed and the average compaction degree is calculated. If the average compaction degree does not reach 93%, it may be due to insufficient compaction causing the settlement difference to exceed the standard. The reasons for exceeding the standard are specifically analyzed in this way so as to adjust the construction parameters in a targeted manner to control the non-uniform settlement.

[0173] To precisely adjust construction parameters to control uneven settlement, S35 is proposed, including:

[0174] If the predicted cause of excessive settlement difference is fluctuation in the packing gradation, then adjust the vibration frequency of the sieve holes in the gradation screening unit.

[0175] If the predicted cause of excessive settlement difference is excessive layer thickness, then the layer backfill thickness shall be reduced by the proportion of the predicted excessive settlement difference.

[0176] If the predicted cause of excessive settlement difference is insufficient compaction, the excitation force of the intelligent vibration compaction unit is increased, and the compaction control threshold is increased according to the preset increase range.

[0177] In this embodiment, adjusting the vibration frequency of the sieve apertures in the gradation screening unit refers to the measures taken when the predicted cause of excessive settling difference is fluctuation in the packing gradation. For example, if the proportion of 20-50mm particles in the post-screening packing is lower than the lower limit of the target range (<30%), the vibration frequency of the 50mm sieve apertures is increased by 2Hz. This enhances the screening efficiency for large particles, increasing the proportion of 20-50mm particles to 30-40%. If the proportion of <5mm particles in the post-screening packing is higher than the upper limit of the target range (>20%), the vibration frequency of the 5mm sieve apertures is decreased by 2Hz to reduce the amount of fine particles passing through the sieve, reducing the proportion of <5mm particles to 10-20%.

[0178] In this embodiment, reducing the layered backfill thickness according to the proportion of predicted settlement difference exceeding the standard is a response method when the predicted cause of excessive settlement difference is excessive layer thickness. The specific calculation method is based on... Reduce the thickness of layered backfill (in (For predicting differential settlement). For example, when the initial layered backfill thickness... If the differential settlement is predicted The value was 15mm / 10m, exceeding the standard by a certain percentage. The adjusted layered backfill thickness for (Actual value is 27-30cm).

[0179] In this embodiment, the excitation force of the intelligent vibration compaction unit is increased, and the compaction degree control threshold is increased according to a preset increase range. The excitation force is increased in a 5% gradient. For example, if the current excitation force is 300kN, it will be increased to... At the same time, the compaction control threshold was increased from K≥93% to K≥94%.

[0180] This invention provides an implementation method for a prefabricated channel graded backfill optimization device based on uneven settlement control, comprising:

[0181] The gradation screening preparation module is used to determine the target filler gradation range and the initial layered backfill thickness, and to put the filler to be backfilled into the gradation screening unit for gradation screening and crushing to obtain qualified filler that meets the target filler gradation range and transport it to the backfilling operation area. At the same time, the gradation test report of each batch of filler is recorded.

[0182] The layered backfill construction module is used to lay sieved qualified fill material on both sides and top of the prefabricated passage according to the initial layered backfill thickness to form the first layer of backfill soil, and measure the settlement value of the top of the side wall and bottom plate of the prefabricated passage. Combined with the excitation force and number of compaction passes of the intelligent vibration compaction unit, a layered backfill-settlement correspondence table is generated.

[0183] The parameter dynamic adjustment module is used to predict the predicted settlement difference after the completion of the next n layers of backfilling based on the foundation parameters of the prefabricated channel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table. Based on the predicted settlement difference, it adjusts any one of the parameters of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit for the next round of layered backfilling operations until all backfilling operations are completed.

[0184] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. An optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control, characterized in that, include: S1: Determine the target packing gradation range and initial layered backfill thickness, and put the packing to be backfilled into the gradation screening unit for gradation screening and crushing to obtain qualified packing that meets the target packing gradation range and transport it to the backfilling operation area. At the same time, record the gradation test report of each batch of packing. S2: According to the initial layered backfill thickness, sieved qualified fill material is laid on both sides and top of the prefabricated passage to form the first layer of backfill soil. The settlement value of the top of the side wall and bottom plate of the prefabricated passage is measured. Combined with the excitation force and compaction pass of the intelligent vibration compaction unit, a layered backfill-settlement correspondence table is generated. S3: Based on the foundation parameters of the prefabricated channel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table, predict the predicted settlement difference after the completion of the subsequent n layers of backfill construction. Based on the predicted settlement difference, adjust any one of the parameters of the sieve vibration frequency of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit for the next round of layered backfill construction until all backfill operations are completed. Wherein, S2 includes: S21: According to the initial layered backfill thickness, sieved qualified fill material is laid on both sides and top of the prefabricated channel to form the first layer of backfill soil. At the same time, compaction degree sensing units are arranged in the first layer of backfill soil according to the preset size grid to collect the compaction degree data of the fill material in real time. S22: Start the intelligent vibration compaction unit to compact the first layer of backfill soil, and control the ground radar monitoring unit to scan along the longitudinal direction of the channel at preset intervals during the compaction process to generate an image of the internal density distribution of the backfill soil. S23: Calculate the average compaction degree of the fill material based on the compaction degree data. If the average compaction degree of the fill material is not less than the preset compaction degree threshold and there are no obvious voids or loose areas in the density distribution image inside the backfill soil, then stop the compaction operation of the intelligent vibration compaction unit on the first layer of backfill soil. Otherwise, adjust the excitation force and compaction pass of the intelligent vibration compaction unit. S24: Repeat steps S21-S23 to carry out subsequent layered backfilling and compaction operations. After each preset layer of backfilling is completed, the settlement value of the top and bottom plate of the prefabricated passage sidewall is measured by the displacement monitoring unit. Combined with the recorded excitation force and number of compaction passes of the intelligent vibration compaction unit, a layered backfilling-settlement correspondence table is generated.

2. The optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control according to claim 1, characterized in that, S1 includes: S11: Obtain the soil type, natural moisture content and bearing capacity characteristics of the foundation soil in the prefabricated channel construction area, and determine the target fill material gradation range and initial layer backfill thickness in combination with the requirements for backfill soil in the pre-set roadbed construction technical specifications. S12: The fill material to be backfilled is put into the gradation screening unit. The fill material is graded and screened by the three-layer vibrating screen in the gradation screening unit to obtain the screened fill material. The particle size monitoring camera in the gradation screening unit collects the particle image of the screened fill material in real time. The image recognition algorithm is used to analyze the particle image of the screened fill material and calculate the particle ratio of each particle size range. S13: If the proportion of particles exceeding the standard in each particle size range is greater than the preset proportion threshold, the crushing component of the gradation screening unit is activated to crush the particles exceeding the standard until qualified filler that meets the target filler gradation range is obtained and transported to the backfilling operation area. At the same time, the gradation test report of each batch of filler is recorded.

3. The optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control according to claim 2, characterized in that, Image recognition algorithms were used to analyze the particle images of the packing material after sieving and to calculate the proportion of particles in each size range, including: The particle image of the screened filler was converted into a grayscale image using a weighted average method, and the particle edge contour was extracted from the grayscale image using the Canny operator. The geometric features of each particle are calculated based on the particle edge contour, and a particle feature vector matrix is ​​constructed based on the geometric features of all particles. The particle feature vector matrix is ​​then standardized to obtain a standardized particle feature vector matrix. The K-means clustering algorithm is used to cluster all the standardized particle feature vectors in the standardized particle feature vector matrix to obtain multiple feature vector clusters. Calculate the average equivalent particle size of each feature vector cluster, and map each feature vector cluster to the corresponding target particle size interval according to the average equivalent particle size, thereby determining the particle size interval to which each feature vector cluster belongs. The ratio of the total number of standardized particle feature vectors in each feature vector cluster to the total number of all particles is taken as the proportion of particles in the corresponding particle size range.

4. The optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control according to claim 1, characterized in that, The ground-penetrating radar monitoring unit scans along the longitudinal direction of the channel at preset intervals during the compaction process, generating images of the internal density distribution of the backfill soil, including: During the compaction process, the ground-penetrating radar monitoring unit scans along the longitudinal direction of the channel at preset intervals to obtain multiple scan lines. Each scan line contains a preset number of sampling points, and the propagation time of the reflected wave at each sampling point is collected. Based on the propagation time of the reflected waves at all sampling points of all scan lines, a two-dimensional original signal propagation time matrix is ​​constructed. A background removal algorithm is then used to denoise the two-dimensional original signal propagation time matrix to obtain a corrected propagation time matrix. The two-dimensional plane coordinates of all sampling points are determined based on the setting rules of all scan lines and all sampling points. The current electromagnetic wave velocity is calculated based on the dielectric constant of the backfill soil. The depth coordinates of each sampling point are calculated based on the current electromagnetic wave velocity and the corrected propagation time matrix. The three-dimensional physical coordinates of each sampling point are determined based on the two-dimensional plane coordinates and depth coordinates of each sampling point. The density of each sampling point is calculated based on the maximum dry density of the filler, the standard electromagnetic wave velocity under compacted state, and the current electromagnetic wave velocity. A two-dimensional density matrix is ​​generated based on the density of all sampling points of all scan lines. The two-dimensional density matrix is ​​interpolated into a uniform grid matrix using a bicubic interpolation algorithm. Based on the first derivative of each row element and the first derivative of each column element in the uniform grid matrix, all density abrupt change points are determined, and all density interval regions are determined based on all density abrupt change points. The uniform grid matrix is ​​color-assigned according to all density intervals to obtain color assignment results, and an image of the density distribution inside the backfill soil is generated based on the color assignment results.

5. The optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control according to claim 4, characterized in that, All density intervals were determined based on all density abrupt change points, including: Cluster the three-dimensional physical coordinates of all density mutation points to obtain multiple clusters of neighboring mutation points; The depth coordinates of all neighboring mutation points in each neighboring mutation point cluster are sorted to obtain a depth coordinate sequence. Based on the adjacent depth differences in the depth coordinate sequence, all region boundaries of each neighboring mutation point cluster are identified. Based on all the boundaries of each neighboring mutation point cluster, the corresponding depth coordinate sequence is divided to obtain multiple continuous depth sequences and corresponding mutation point sequences of the corresponding neighboring mutation point cluster. Extract the density sequence corresponding to each mutation point sequence in the uniform grid matrix, and generate a depth-density two-dimensional data sequence by combining it with the corresponding continuous depth sequence. Use a quadratic polynomial to perform curve fitting on each depth-density two-dimensional data sequence to obtain a density-depth curve with depth as the independent variable and density as the dependent variable. Calculate the first and second derivatives of each density depth curve, and based on the first and second derivatives of each density depth curve, determine the combination of adjacent density intervals for each density abrupt change point involved in each density depth curve; All density interval regions are determined in a uniform grid matrix by combining adjacent density intervals at each density abrupt change point.

6. The optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control according to claim 1, characterized in that, Based on the foundation parameters of the prefabricated tunnel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table, the predicted differential settlement after the completion of the subsequent n layers of backfill construction is predicted, including: S31: Input the foundation parameters of the prefabricated passage construction area, the gradation test reports of all batches of fill material, and the corresponding table of layered backfill-settlement into the numerical simulation analysis unit to establish a coupled finite element model of prefabricated passage-backfill soil-foundation. The foundation parameters include the foundation soil type, natural moisture content, and characteristic value of foundation bearing capacity. S32: Predict the settlement trend after the completion of the subsequent n layers of backfill construction using a coupled finite element model of prefabricated channel-backfill soil-foundation, and calculate the predicted settlement difference.

7. The optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control according to claim 1, characterized in that, Based on the predicted settlement difference, adjust any one of the following parameters for the next round of layered backfilling: the sieve vibration frequency of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit, until all backfilling operations are completed, including: S33: If the predicted settlement difference does not exceed the threshold of the uneven settlement control standard, then maintain the current layered backfill thickness and compaction parameters; S34: If the predicted settlement difference exceeds the threshold of the uneven settlement control standard, analyze the reasons for the predicted settlement difference exceeding the standard. S35: Adjust any one of the following parameters in the next round of layered backfilling construction operations based on the predicted cause of excessive settlement difference: sieve vibration frequency of the gradation screening unit, initial layered backfill thickness, and compaction parameters of the intelligent vibration compaction unit. S36: Apply the adjusted sieve vibration frequency of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit to the next round of layered backfilling operations, and repeat steps S2-S3 until all backfilling operations are completed.

8. The optimized method for graded backfilling of prefabricated passageways based on non-uniform settlement control according to claim 7, characterized in that, The S35 includes: If the predicted cause of excessive settlement difference is fluctuation in the packing gradation, then adjust the vibration frequency of the sieve holes in the gradation screening unit. If the predicted cause of excessive settlement difference is excessive layer thickness, then the layer backfill thickness shall be reduced by the proportion of the predicted excessive settlement difference. If the predicted cause of excessive settlement difference is insufficient compaction, the excitation force of the intelligent vibration compaction unit is increased, and the compaction control threshold is increased according to the preset increase range.

9. A prefabricated channel graded backfilling optimization device based on uneven settlement control, characterized in that, The prefabricated channel graded backfill optimization method based on non-uniform settlement control, as described in any one of claims 1 to 8, comprises: The gradation screening preparation module is used to determine the target filler gradation range and the initial layered backfill thickness, and to put the filler to be backfilled into the gradation screening unit for gradation screening and crushing to obtain qualified filler that meets the target filler gradation range and transport it to the backfilling operation area. At the same time, the gradation test report of each batch of filler is recorded. The layered backfill construction module is used to lay sieved qualified fill material on both sides and top of the prefabricated passage according to the initial layered backfill thickness to form the first layer of backfill soil, and measure the settlement value of the top of the side wall and bottom plate of the prefabricated passage. Combined with the excitation force and number of compaction passes of the intelligent vibration compaction unit, a layered backfill-settlement correspondence table is generated. The parameter dynamic adjustment module is used to predict the predicted settlement difference after the completion of the next n layers of backfilling based on the foundation parameters of the prefabricated channel construction area, the gradation test reports of all batches of fill material, and the layered backfill-settlement correspondence table. Based on the predicted settlement difference, it adjusts any one of the parameters of the gradation screening unit, the initial layered backfill thickness, and the compaction parameters of the intelligent vibration compaction unit for the next round of layered backfilling operations until all backfilling operations are completed.

Images