A backfill compaction coefficient dynamic detection method and rapid determination system

Abstract

This invention discloses a dynamic detection method and rapid determination system for the compaction coefficient of backfill soil, belonging to the field of roadbed quality testing technology. The method includes: acquiring the compaction characteristic parameters and vibration response characteristic parameters of a road roller; analyzing wheel track direction differences and soil resonance effects based on the acquired parameters to determine the wheel track cumulative effect factor during the compaction process; constructing a backfill soil compaction prediction model based on the acquired parameters and the wheel track cumulative effect factor and outputting the predicted compaction coefficient value; acquiring groundwater distribution parameters in the backfill area and correcting the predicted compaction coefficient value based on the temporal fluctuation amplitude of the water level; and outputting the corrected predicted compaction coefficient value. The system includes a parameter acquisition module, an effect factor determination module, a model construction module, a prediction value correction module, and a prediction value output module. This invention effectively improves the accuracy and reliability of compaction coefficient prediction by quantifying the long-term impact of wheel track cumulative effects and groundwater level changes on compaction effects.

Description

Technical Field

[0001] This invention belongs to the field of roadbed quality testing technology, specifically relating to a dynamic testing method and rapid determination system for the compaction coefficient of backfill soil. Background Technology

[0002] Backfill compaction is a crucial step in roadbed construction, and its quality directly impacts the stability and long-term safety of the roadbed structure. The compaction coefficient, a core indicator for evaluating the density of backfill, is key to ensuring the stability and bearing capacity of the roadbed structure. Traditional testing methods primarily rely on on-site sampling combined with laboratory tests such as the ring sampler method and sand cone method. While these methods offer high accuracy, they suffer from low testing efficiency, limited coverage, and difficulty in reflecting compaction uniformity in real time.

[0003] In recent years, continuous compaction testing technology based on the response parameters of vibratory rollers has attracted attention. However, existing continuous compaction testing technologies still have the following shortcomings:

[0004] (1) Lack of quantitative analysis methods for the cumulative effect of vibratory roller tracks and soil resonance effect: In actual construction, the repeated superposition effect of roller tracks and the variability of rolling direction during the rolling process will have a significant impact on the compaction uniformity. However, existing technologies mostly ignore the dynamic control effect of rolling direction variability on compaction quality. At the same time, the resonance effect between the working frequency of the vibratory roller and the natural frequency of the backfill soil will enhance the compaction effect. However, existing technologies lack quantitative characterization of soil resonance mechanism.

[0005] (2) The dynamic changes in groundwater level do not take into account the time-varying correction requirements of compaction coefficient: fluctuations in groundwater level in the backfill area will affect long-term compaction stability by changing the soil moisture content. Existing technologies usually do not take into account the time-varying correction requirements of groundwater spatiotemporal distribution on compaction coefficient, resulting in deviations between prediction results and actual working conditions.

[0006] (3) The coverage is limited and cannot achieve dynamic and full coverage detection of large backfill areas.

[0007] In summary, existing technologies limit the accuracy, real-time performance, and engineering applicability of dynamic testing of backfill compaction coefficients. Therefore, there is an urgent need to develop a rapid method for determining the compaction coefficient of large-area backfill soil that can quantify the cumulative effect of wheel tracks, soil resonance effect, and the dynamic influence of groundwater. Summary of the Invention

[0008] To overcome the problems existing in the prior art, this invention provides a dynamic detection method and rapid determination system for the compaction coefficient of backfill soil. This method integrates the dynamic compaction parameters and vibration response characteristics of the road roller to construct a data-driven backfill soil compaction prediction model. It also introduces wheel track cumulative effect factor and soil resonance effect analysis, and combines the time-varying fluctuation of groundwater level to perform time-varying correction on the predicted compaction coefficient value, which effectively improves the accuracy and reliability of compaction coefficient prediction under different working conditions and long-term use conditions.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] In a first aspect, the present invention provides a method for dynamically detecting the compaction coefficient of backfill soil, comprising the following steps:

[0011] Obtain the compaction characteristic parameters of the road roller and the vibration response characteristic parameters of the road roller vibratory drum during the compaction process;

[0012] Based on the rolling characteristic parameters and vibration response characteristic parameters, the wheel track direction difference and soil resonance effect are analyzed to determine the wheel track cumulative effect factor during the rolling process;

[0013] Based on the rolling characteristic parameters and vibration response characteristic parameters, and combined with the wheel track cumulative effect factor, a backfill soil compaction prediction model is constructed and the predicted compaction coefficient value is output.

[0014] Obtain the groundwater distribution parameters of the backfill area, and correct the predicted compaction coefficient based on the temporal fluctuation amplitude of the groundwater level;

[0015] Output the corrected compaction coefficient prediction value to achieve dynamic detection of the compaction state of backfill soil.

[0016] Furthermore, the vibration response characteristic parameters include vibration frequency, amplitude, and wheel load pressure.

[0017] Furthermore, determining the wheel track cumulative effect factor during the compaction process includes:

[0018] The backfill area is divided into grid cells, and the compaction feature parameters corresponding to each grid cell are extracted based on the compaction feature parameters; the compaction feature parameters of each grid cell include the cumulative number of compactions and the compaction direction angle corresponding to each compaction.

[0019] Based on the compaction characteristic parameters, the differences between the current compaction direction and the historical compaction direction are analyzed, and the wheel track direction variation factor corresponding to the grid cell is calculated;

[0020] Based on vibration response characteristic parameters, the frequency deviation between the vibration frequency of the vibratory roller drum during the compaction process and the natural frequency of the soil is determined, and the soil resonance effect factor corresponding to the grid cell is calculated.

[0021] Based on the wheel track direction variation factor and the soil resonance effect factor, the wheel track effect factor of each grid cell is determined, and the average value of the wheel track effect factors of all grid cells is taken as the wheel track cumulative effect factor; the wheel track cumulative effect factor is used to characterize the compaction uniformity and density of the backfill area under vibratory rolling.

[0022] Furthermore, the formula for calculating the wheel track direction variation factor is as follows:

[0023] ,

[0024] In the formula, This represents the variation factor of the wheel track direction corresponding to the mesh cell. Indicates the current rolling direction angle. Indicates the first The direction and angle of this historical crushing. Indicates the direction sensitivity coefficient. This indicates the cumulative number of times the grid cell has been compacted.

[0025] Further, the soil resonance effect factor is calculated, including:

[0026] Based on vibration response characteristic parameters, the vibration frequency of the vibratory roller drum during the compaction process is extracted;

[0027] The natural frequency of the soil in the backfill area is obtained, and the frequency deviation between the vibration frequency and the natural frequency of the soil is determined; the frequency deviation is used to characterize the degree of difference between the vibration frequency of the vibrating wheel and the natural frequency of the soil.

[0028] Based on the aforementioned frequency deviation, the soil resonance effect factor is calculated using the following formula:

[0029] ,

[0030] In the formula, Indicates soil resonance effect factor, This indicates the vibration frequency of the vibratory drum of the road roller during the compaction process. This indicates the inherent frequency of the soil.

[0031] Furthermore, the construction of the backfill soil compaction prediction model and the output of the predicted compaction coefficient include:

[0032] Based on compaction characteristic parameters, vibration response characteristic parameters, and wheel track cumulative effect factor, an input feature vector is constructed, and feature engineering processing is performed on the input feature vector; the feature engineering processing includes missing value imputation, filtering and denoising, and standardization;

[0033] A backfill compaction prediction model is constructed based on at least one data-driven modeling method, such as support vector machine, random forest, or deep learning network, and the model is trained and validated. The optimal model parameters are determined by combining cross-validation and hyperparameter search to obtain the trained backfill compaction prediction model.

[0034] The backfill compaction prediction model, after training, infers from the input feature vector and outputs the predicted compaction coefficient value for the corresponding construction location.

[0035] Furthermore, the correction of the predicted compaction coefficient based on the temporal fluctuation amplitude of groundwater level includes:

[0036] Based on the groundwater distribution parameters of the backfill area, the groundwater level fluctuation amplitude within a preset time window is extracted;

[0037] Based on the time-varying attenuation of the long-term stability of the predicted compaction coefficient value according to the temporal fluctuation amplitude of groundwater level, the time-varying attenuation factor is calculated:

[0038] ,

[0039] In the formula, Represents the time-varying decay factor. Indicates the attenuation sensitivity coefficient. Indicates the preset time window The amplitude of water level fluctuations within the area, Indicates the reference water level height;

[0040] The predicted compaction coefficient is corrected by using a time-varying decay factor to obtain the corrected predicted compaction coefficient.

[0041] Secondly, the present invention provides a rapid system for determining the compaction coefficient of backfill soil, comprising:

[0042] The parameter acquisition module is used to acquire the compaction characteristic parameters of the road roller and the vibration response characteristic parameters of the vibratory drum during the compaction process; the vibration response characteristic parameters include vibration frequency, amplitude and drum load pressure;

[0043] The effect factor determination module is used to analyze wheel track direction differences and soil resonance effects based on rolling characteristic parameters and vibration response characteristic parameters, and to determine the wheel track cumulative effect factor during the rolling process.

[0044] The model building module is used to construct a backfill soil compaction prediction model based on rolling characteristic parameters, vibration response characteristic parameters, and wheel track cumulative effect factor, and output the predicted compaction coefficient value.

[0045] The prediction correction module is used to obtain the groundwater distribution parameters of the backfill area and correct the predicted compaction coefficient based on the temporal fluctuation amplitude of the groundwater level.

[0046] The prediction output module is used to output the corrected predicted compaction coefficient to achieve dynamic detection of the compaction state of backfill soil.

[0047] Thirdly, the present invention provides an electronic device, characterized in that it includes: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the dynamic detection method for the compaction coefficient of backfill soil by calling the computer program stored in the memory.

[0048] Fourthly, the present invention provides a computer-readable storage medium having stored computer program instructions thereon, which, when executed on a computer, cause the computer to perform the dynamic detection method for the compaction coefficient of backfill soil.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] This invention analyzes the differences in wheel track direction and soil resonance effects to determine the wheel track cumulative effect factor, thereby constructing a backfill compaction prediction model that integrates the dynamic compaction parameters and vibration response characteristics of the road roller. Furthermore, it incorporates time-varying corrections to the predicted compaction coefficient based on groundwater level fluctuations, effectively improving the accuracy and reliability of compaction state assessment under different working conditions and long-term use. Compared with traditional static or point measurement methods, this invention can quantify the impact of wheel track direction differences and soil vibration characteristics on compaction uniformity and density, while also quantifying the long-term impact of groundwater level changes on compaction effects. This significantly improves the accuracy and reliability of compaction coefficient prediction, achieving full-cycle dynamic monitoring from construction process control to long-term performance prediction. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating the dynamic detection method for the compaction coefficient of backfill soil provided by the present invention.

[0052] Figure 2 This is a schematic diagram of the contact between the road roller and the soil provided by the present invention;

[0053] Figure 3 This is a structural schematic diagram of the rapid determination system for the compaction coefficient of backfill soil provided by the present invention;

[0054] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention;

[0055] The labels in the diagram are as follows: 1-cab, 2-vibration wheel, 3-drive wheel. Detailed Implementation

[0056] To make the technical solution of the present invention clearer, the technical solution of the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

[0057] like Figure 1 As shown, the present invention provides a method for dynamically detecting the compaction coefficient of backfill soil, comprising the following steps:

[0058] S100. Obtain the compaction characteristic parameters of the road roller and the vibration response characteristic parameters of the road roller vibratory drum during the compaction process; the vibration response characteristic parameters include vibration frequency, amplitude and wheel load pressure;

[0059] Among them, the compaction characteristic parameters are obtained through the GNSS / INS integrated navigation system, encoder and control unit integrated on the road roller; the GNSS / INS system calculates the plane coordinates and heading angle of the roller drum center in real time; the vibration response characteristic parameters are directly measured by the triaxial acceleration sensor, frequency sensor and hydraulic system pressure sensor installed on the vibratory wheel bearing seat or frame, so as to synchronously obtain the vibration frequency, amplitude and wheel load pressure reflecting the dynamic compaction state of the soil.

[0060] S200. Based on the rolling characteristic parameters and vibration response characteristic parameters, the wheel track direction difference and soil resonance effect are analyzed to determine the wheel track cumulative effect factor during the rolling process.

[0061] The wheel track cumulative effect refers to the superposition of compaction energy and effect applied by a road roller to backfill soil during vibratory compaction. It is a dynamic concept that integrates the spatial distribution of compaction and the efficiency of compaction energy transfer. The spatial distribution of compaction is characterized by the variation of wheel track direction, while the efficiency of compaction energy transfer is characterized by the soil resonance effect. The core meaning of the wheel track cumulative effect is that the final compaction state of the backfill soil is not a simple linear superposition of each compaction action, but is jointly affected by the directional change of the compaction path and whether the vibration energy is efficiently absorbed by the soil. Therefore, the wheel track cumulative effect factor is a comprehensive indicator used to quantitatively evaluate the real and effective compaction effect obtained in a specific area under complex construction history.

[0062] The determination of the wheel track cumulative effect factor during the compaction process specifically includes:

[0063] Grid division and extraction of compaction feature parameters: The backfill area is divided into grid cells, and based on the center plane coordinates and heading angle of the roller at each sampling time, the number of times each grid cell is compacted is counted as the cumulative number of compactions, and the compaction direction angle corresponding to each compaction is recorded; the cumulative number of compactions and the compaction direction angle constitute the compaction feature parameters of each grid cell.

[0064] Wheel track direction variation factor calculation: Based on the rolling characteristic parameters, analyze the difference between the current rolling direction and the historical rolling direction, and calculate the wheel track direction variation factor corresponding to the grid cell;

[0065] Specifically, wheel track direction variation is crucial for ensuring compaction uniformity. If the roller always rolls in the same direction, fixed wheel track grooves are easily formed, leading to stress concentration and uneven compaction. By using a wheel track direction variation factor to encourage cross-rolling or changing the rolling angle, significant variations in the rolling direction can generate shear and compressive forces in different directions on the soil particles, breaking down the original particle arrangement and causing them to realign to a denser and more uniform state. The formula for calculating the wheel track direction variation factor is:

[0066] ,

[0067] In the formula, This represents the variation factor of the wheel track direction corresponding to the mesh cell. Indicates the current rolling direction angle. Indicates the first The direction and angle of the second historical crushing; This represents the angle difference, used to describe the relationship between the current compaction direction and the historical [number missing]. The geometric differences between compaction directions; since the compaction direction itself is periodic, for example, 0° and 180° can be considered as opposite directions on the same axis, direct subtraction cannot accurately reflect the actual physical differences; therefore, the following method is adopted. The function performs the processing. A metric representing the degree of directional difference, used to quantify the inconsistency in the directions of two compaction operations. The function can eliminate the influence of directional periodicity: when two compaction directions are the same or opposite, they are still on the same path axis, and their effect on disrupting the directional arrangement of soil particles is limited. Therefore, the difference is rated as 0, i.e. , ; This means that the difference between the current compaction direction and all previous historical compaction directions is accumulated, i.e., the sum of the differences in historical directions, and the influence of the number of compaction times on the sum of the differences in historical directions is eliminated by taking the average value. Indicates the direction sensitivity coefficient. This represents the cumulative number of compaction cycles per grid cell. Initial directional changes, such as from unidirectional to intersecting, have a significant effect. However, once the directions have diversified, the marginal benefit of increasing vertical compaction diminishes. An exponential function effectively simulates this non-linear relationship of diminishing returns. Therefore, an exponential decay function is used... The average directional difference is mapped to the wheel track directional variation factor;

[0068] Soil resonance effect factor calculation: Based on vibration response characteristic parameters, determine the frequency deviation between the vibration frequency of the road roller vibratory wheel during the compaction process and the natural frequency of the soil, and calculate the soil resonance effect factor corresponding to the grid cell;

[0069] Specifically, soil is not only the object of the road roller's operation, but also the object on which the road roller bears its load; for example... Figure 2 As shown, the vibratory roller consists of a vibratory drum 2, a drive wheel 3, and a cab 1. During construction, the vibratory roller transmits periodic vibrations to the soil, causing it to change from a static to a vibrating state. Due to the change in the soil's physical and mechanical properties, relative motion occurs between soil clods and between soil particles, resulting in rearrangement and achieving soil compaction. Soil resonance is key to ensuring compaction efficiency and depth. When the frequency of the roller's vibratory drum approaches the natural frequency of the soil, soil resonance occurs. At this point, vibrational energy can be transmitted to the deep soil layers with minimal loss, causing intense relative motion of soil particles and thus achieving optimal compaction. Therefore, the efficiency of this energy transmission is identified and quantified through soil resonance effect factors.

[0070] The calculation process for the soil resonance effect factor specifically includes:

[0071] Based on vibration response characteristic parameters, the vibration frequency of the vibratory roller drum during the compaction process is extracted;

[0072] The natural frequency of the soil in the backfill area is obtained, and the frequency deviation between the vibration frequency and the natural frequency of the soil is determined; the frequency deviation is used to characterize the degree of difference between the vibration frequency of the vibrating wheel and the natural frequency of the soil.

[0073] Based on the aforementioned frequency deviation, the soil resonance effect factor is calculated using a cosine similarity function:

[0074] ,

[0075] In the formula, Indicates soil resonance effect factor, This indicates the vibration frequency of the vibratory drum of the road roller during the compaction process. Indicates the inherent frequency of the soil;

[0076] Determination of wheel track cumulative effect factor: based on wheel track direction variation factor Resonance effect factor with soil Determine the wheel track effect factor for each grid cell. As shown in the following formula:

[0077] ;

[0078] The average value of the wheel track effect factor of all grid cells is used as the wheel track cumulative effect factor to characterize the compaction uniformity and density of the backfill area under vibratory rolling.

[0079] S300, based on rolling characteristic parameters and vibration response characteristic parameters, and combined with wheel track cumulative effect factor, constructs a backfill soil compaction prediction model and outputs the predicted compaction coefficient value;

[0080] By introducing the wheel track cumulative effect factor, the backfill soil compaction prediction model not only includes rolling characteristic parameter information, but also further couples geometric path information and energy transfer efficiency information. This makes the prediction model take a big step from simple statistical correlation to physical mechanism driven, and significantly improves the accuracy and reliability of the compaction coefficient prediction value.

[0081] The construction of the backfill soil compaction prediction model and the output of the predicted compaction coefficient specifically include:

[0082] Based on compaction characteristic parameters, vibration response characteristic parameters, and wheel track cumulative effect factor, an input feature vector is constructed, and feature engineering processing is performed on the input feature vector; the feature engineering processing includes missing value imputation, filtering and denoising, and standardization;

[0083] Based on at least one data-driven modeling method among support vector machines, random forests, or deep learning networks, a backfill soil compaction prediction model structure is constructed, and the model is trained and validated. The optimal model parameters are determined by combining cross-validation and hyperparameter search to obtain the trained backfill soil compaction prediction model.

[0084] The backfill compaction prediction model, after training, infers from the input feature vector and outputs the predicted compaction coefficient value for the corresponding construction location.

[0085] S400. Obtain the groundwater distribution parameters of the backfill area and correct the predicted compaction coefficient based on the temporal fluctuation amplitude of the groundwater level.

[0086] Specifically, during backfill construction, changes in groundwater level significantly affect the soil's water content, pore structure, and compaction performance. Soil is prone to local liquefaction or rebound under high water content, leading to a decrease in the compaction coefficient. Simply relying on rolling characteristics and vibration response parameters to predict the compaction coefficient cannot fully reflect the long-term impact of dynamic changes in groundwater on the soil compaction effect.

[0087] To this end, by acquiring the groundwater distribution parameters of the backfill area and extracting the fluctuation amplitude of the groundwater level within a preset time window, dynamic correction of the predicted compaction coefficient value is achieved. Specifically, by introducing a time-varying attenuation factor to quantify the influence of water level fluctuation amplitude on the long-term stability of the compaction coefficient, the effect of hydrological environment changes on soil density can be scientifically reflected. The time-varying attenuation factor fully considers the temporal variation characteristics of groundwater in the construction area, enhances the adaptability of the backfill compaction coefficient prediction model to actual working conditions, and can reasonably simulate the attenuation effect of groundwater fluctuations on soil compaction performance, making the predicted value more consistent with the actual soil stress and compaction state variation law.

[0088] The correction of the predicted compaction coefficient based on the temporal fluctuation amplitude of groundwater level specifically includes:

[0089] Based on the groundwater distribution parameters of the backfill area, the groundwater level fluctuation amplitude within a preset time window is extracted;

[0090] The long-term stability of the predicted compaction coefficient is assessed based on the temporal fluctuation amplitude of groundwater level, and the time-varying attenuation factor is calculated.

[0091] ,

[0092] In the formula, Represents the time-varying decay factor. Indicates the attenuation sensitivity coefficient. Indicates the preset time window The amplitude of water level fluctuations within the area, Indicates the reference water level height;

[0093] The predicted compaction coefficient is corrected by a time-varying decay factor to obtain the corrected predicted compaction coefficient; wherein, the corrected predicted compaction coefficient... In the formula This represents the predicted compaction coefficient. This represents the time-varying decay factor.

[0094] S500 outputs the corrected predicted compaction coefficient value to achieve dynamic detection of the compaction state of backfill soil.

[0095] like Figure 3 As shown, the present invention provides a rapid system for determining the compaction coefficient of backfill soil, comprising:

[0096] The parameter acquisition module 210 is used to acquire the compaction characteristic parameters of the road roller and the vibration response characteristic parameters of the road roller vibratory drum during the compaction process; the vibration response characteristic parameters include vibration frequency, amplitude and drum load pressure;

[0097] The effect factor determination module 220 is used to analyze wheel track direction differences and soil resonance effects based on compaction characteristic parameters and vibration response characteristic parameters, and to determine the wheel track cumulative effect factor during the compaction process. Specifically, it includes: dividing the backfill area into grid cells, extracting the cumulative number of compactions and compaction direction angle for each grid cell; calculating the wheel track direction variation factor based on the difference between the current compaction direction and the historical compaction direction; calculating the soil resonance effect factor based on the frequency deviation between the vibration frequency and the natural frequency of the soil; combining the wheel track direction variation factor and the soil resonance effect factor to determine the wheel track effect factor for each grid cell, and taking the average wheel track effect factor of all grid cells as the wheel track cumulative effect factor to characterize the compaction uniformity and density of the backfill area.

[0098] The model building module 230 is used to construct a backfill compaction prediction model and output a predicted compaction coefficient value based on rolling feature parameters, vibration response feature parameters, and wheel track cumulative effect factors. Specifically, it includes: constructing an input feature vector and performing missing value imputation, filtering, denoising, and standardization; constructing a backfill compaction prediction model based on at least one data-driven modeling method among support vector machines, random forests, or deep learning networks, and performing model training and validation; determining the optimal model parameters by combining cross-validation and hyperparameter search to obtain the trained backfill compaction prediction model; and using the trained backfill compaction prediction model to infer the input feature vector and output the predicted compaction coefficient value corresponding to the construction location.

[0099] The prediction correction module 240 is used to obtain the groundwater distribution parameters of the backfill area and correct the predicted compaction coefficient based on the temporal fluctuation amplitude of the groundwater level. Specifically, it includes: extracting the water level fluctuation amplitude of the groundwater level within a preset time window; calculating the time-varying attenuation factor based on the temporal fluctuation amplitude of the groundwater level; and performing time-varying attenuation correction on the predicted compaction coefficient through the time-varying attenuation factor to obtain the corrected predicted compaction coefficient.

[0100] The prediction output module 250 is used to output the corrected compaction coefficient prediction value to achieve dynamic detection of the compaction state of the backfill soil.

[0101] like Figure 4 As shown, the present invention also provides an electronic device 300, including a memory 310, a processor 320 and a communication bus 330; the memory 310 and the processor 320 are connected through the communication bus 330; the memory 310 stores a computer program, and when the computer program is loaded and executed by the processor 320, it realizes the dynamic detection method for the compaction coefficient of the backfill soil.

[0102] Specifically, the memory 310 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 310 includes a program storage area and a data storage area, wherein the program storage area stores instructions for implementing the dynamic detection method for the compaction coefficient of the backfill soil; and the data storage area stores data involved in the dynamic detection method for the compaction coefficient of the backfill soil.

[0103] Specifically, processor 320 includes one or more processing cores. Processor 320 executes instructions, programs, code sets, or instruction sets stored in memory 310, and calls data stored in memory 310 to perform the dynamic detection method for the compaction coefficient of backfill soil. Processor 320 is at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), controller, microcontroller, and microprocessor. Understandably, for different devices, the electronic devices used to implement the functions of processor 320 can be other types, and no specific limitations are made here.

[0104] Specifically, the communication bus 330 may include a path for transferring information between the memory 310 and the processor 320. The communication bus 330 is a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or other standard buses. The communication bus 330 is divided into an address bus, a data bus, and a control bus. For ease of representation, Figure 4 The symbol is represented by a single double arrow, but this does not mean that there is only one bus or one type of bus.

[0105] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is loaded and executed by a processor, the method for dynamically detecting the compaction coefficient of backfill soil is implemented.

[0106] Specifically, a computer-readable storage medium is a tangible device for holding and storing instructions used by an instruction execution device. Computer-readable storage media include, but are not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any combination thereof. Exemplarily, computer-readable storage media include portable computer disks, hard disks, USB flash drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, optical disks, magnetic disks, mechanical encoding devices, and any combination thereof.

[0107] The term "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only the listed elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0108] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made by those skilled in the art within the scope of the technology disclosed in this invention, based on the technical solution and inventive concept of the present invention, should be covered within the protection scope of this invention. Therefore, the protection scope of this invention should be determined by the scope of the claims.

Claims

1. A method for dynamically detecting a compaction coefficient of backfill soil, characterized in that, include: Obtain the compaction characteristic parameters of the road roller and the vibration response characteristic parameters of the road roller vibratory drum during the compaction process; Based on the rolling characteristic parameters and vibration response characteristic parameters, the wheel track direction difference and soil resonance effect are analyzed to determine the wheel track cumulative effect factor during the rolling process; Based on the rolling characteristic parameters and vibration response characteristic parameters, and combined with the wheel track cumulative effect factor, a backfill soil compaction prediction model is constructed and the predicted compaction coefficient value is output. Obtain the groundwater distribution parameters of the backfill area, and correct the predicted compaction coefficient based on the temporal fluctuation amplitude of the groundwater level; Output the corrected compaction coefficient prediction value to achieve dynamic detection of the compaction state of backfill soil.

2. The method according to claim 1, wherein, The vibration response characteristic parameters include vibration frequency, amplitude, and wheel load pressure.

3. The method according to claim 2, wherein, The determination of the wheel track cumulative effect factor during the compaction process includes: The backfill area is divided into grid cells, and the compaction feature parameters corresponding to each grid cell are extracted based on the compaction feature parameters; the compaction feature parameters of each grid cell include the cumulative number of compactions and the compaction direction angle corresponding to each compaction. Based on the compaction characteristic parameters, the differences between the current compaction direction and the historical compaction direction are analyzed, and the wheel track direction variation factor corresponding to the grid cell is calculated; Based on vibration response characteristic parameters, the frequency deviation between the vibration frequency of the vibratory roller drum during the compaction process and the natural frequency of the soil is determined, and the soil resonance effect factor corresponding to the grid cell is calculated. Based on the wheel track direction variation factor and the soil resonance effect factor, the wheel track effect factor of each grid cell is determined, and the average value of the wheel track effect factors of all grid cells is taken as the wheel track cumulative effect factor; the wheel track cumulative effect factor is used to characterize the compaction uniformity and density of the backfill area under vibratory rolling.

4. The method for dynamically detecting the compaction coefficient of backfill soil according to claim 3, characterized in that, The formula for calculating the wheel track direction variation factor is as follows: ， In the formula, This represents the variation factor of the wheel track direction corresponding to the mesh cell. Indicates the current rolling direction angle. Indicates the first The direction and angle of this historical crushing. Indicates the direction sensitivity coefficient. This indicates the cumulative number of times the grid cell has been compacted.

5. The method for dynamically detecting the compaction coefficient of backfill soil according to claim 3, characterized in that, The calculation of the soil resonance effect factor includes: Based on vibration response characteristic parameters, the vibration frequency of the vibratory roller drum during the compaction process is extracted; The natural frequency of the soil in the backfill area is obtained, and the frequency deviation between the vibration frequency and the natural frequency of the soil is determined; the frequency deviation is used to characterize the degree of difference between the vibration frequency of the vibrating wheel and the natural frequency of the soil. Based on the aforementioned frequency deviation, the soil resonance effect factor is calculated using the following formula: ， In the formula, Indicates soil resonance effect factor, This indicates the vibration frequency of the vibratory drum of the road roller during the compaction process. This indicates the inherent frequency of the soil.

6. The method for dynamically detecting the compaction coefficient of backfill soil according to claim 1, characterized in that, The construction of the backfill soil compaction prediction model and the output of the predicted compaction coefficient include: Based on compaction characteristic parameters, vibration response characteristic parameters, and wheel track cumulative effect factor, an input feature vector is constructed, and feature engineering processing is performed on the input feature vector; the feature engineering processing includes missing value imputation, filtering and denoising, and standardization; A backfill compaction prediction model is constructed based on at least one data-driven modeling method, such as support vector machine, random forest, or deep learning network, and the model is trained and validated. The optimal model parameters are determined by combining cross-validation and hyperparameter search to obtain the trained backfill compaction prediction model. The backfill compaction prediction model, after training, infers from the input feature vector and outputs the predicted compaction coefficient value for the corresponding construction location.

7. The method for dynamically detecting the compaction coefficient of backfill soil according to claim 1, characterized in that, The correction of the predicted compaction coefficient based on the temporal fluctuation amplitude of groundwater level includes: Based on the groundwater distribution parameters of the backfill area, the groundwater level fluctuation amplitude within a preset time window is extracted; Based on the time-varying attenuation of the long-term stability of the predicted compaction coefficient value according to the temporal fluctuation amplitude of groundwater level, the time-varying attenuation factor is calculated: ， In the formula, Represents the time-varying decay factor. Indicates the attenuation sensitivity coefficient. Indicates the preset time window The amplitude of water level fluctuations within the area, Indicates the reference water level height; The predicted compaction coefficient is corrected by using a time-varying decay factor to obtain the corrected predicted compaction coefficient.

8. A rapid system for determining the compaction coefficient of backfill soil, characterized in that, A method for dynamically detecting the compaction coefficient of backfill soil according to any one of claims 1-7 includes: The parameter acquisition module is used to acquire the compaction characteristic parameters of the road roller and the vibration response characteristic parameters of the road roller vibratory drum during the compaction process; The effect factor determination module is used to analyze wheel track direction differences and soil resonance effects based on rolling characteristic parameters and vibration response characteristic parameters, and to determine the wheel track cumulative effect factor during the rolling process. The model building module is used to construct a backfill soil compaction prediction model based on rolling characteristic parameters, vibration response characteristic parameters, and wheel track cumulative effect factor, and output the predicted compaction coefficient value. The prediction correction module is used to obtain the groundwater distribution parameters of the backfill area and correct the predicted compaction coefficient based on the temporal fluctuation amplitude of the groundwater level. The prediction output module is used to output the corrected predicted compaction coefficient to achieve dynamic detection of the compaction state of backfill soil.

9. An electronic device, characterized in that, include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes a dynamic detection method for the compaction coefficient of backfill soil as described in any one of claims 1-7 by calling the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that, It stores computer program instructions, which, when executed on a computer, cause the computer to perform a dynamic detection method for the compaction coefficient of backfill soil as described in any one of claims 1-7.

Images