Site grading engineering intelligent construction method and system

Abstract

The application discloses a site leveling engineering intelligent construction method and system, which comprises multi-source surveying and three-dimensional superposition modeling, construction quality zoning and initial optimization of earthwork allocation, hierarchical excavation and differential preprocessing according to processing levels, hierarchical filling and compaction process monitoring, stiffness field inversion and three-index joint determination, feedforward prediction and feedback correction, geological model rewriting and supplementary survey triggering, and cyclic filling and completion re-measurement; through the self-updating closed-loop control of zoning and grading, differential processing and acceptance, feedforward prediction and feedback correction, and geological model rewriting, the application makes the compaction quality data serve the current layer acceptance, the next layer correction, the upper layer prediction, and the geological model updating at the same time, so that the predictive control of the site leveling engineering construction quality and the dynamic optimization of the filling material resources are realized.

Description

Technical Field

[0001] This invention relates to the field of building construction technology, and in particular to an intelligent construction method and system for site leveling engineering. Background Technology

[0002] Site leveling is a fundamental process in construction projects, and its quality and efficiency directly affect the smooth progress of subsequent projects. Site leveling mainly includes excavation, filling, transportation, and compaction of the original terrain, involving a large amount of earthwork allocation.

[0003] With the development of measurement and sensing technologies, technologies such as UAV lidar topographic surveying, electronic earthwork volume calculation, and continuous compaction monitoring have been applied to site leveling construction. However, through engineering practice, the applicant has found that existing intelligent construction methods still have the following technical problems: (1) The measurement, allocation and monitoring systems are independent of each other, forming a “two-skinned measurement and control”: the data collected by the compaction monitoring system is only used for the post-acceptance judgment of the filling layer, and the acceptance data is not transmitted back or used in subsequent construction decisions; the earthwork allocation plan is determined once before the start of construction and is not adjusted, so it cannot respond to the actual geological and compaction quality conditions exposed during the construction process.

[0004] (2) The foundation treatment and compaction test acceptance are disconnected: the design of the soft foundation pretreatment and the compaction test acceptance standards are determined based on the experience of different links. In the same site, areas with different geological conditions often implement the same acceptance standard. The acceptance standard for areas with high treatment level is not raised accordingly, and the treatment effect cannot be verified through the acceptance link.

[0005] (3) Single compaction qualification indicator: Existing continuous compaction monitoring mostly uses a single indicator of compaction degree representative value or the degree of dispersion of continuous measurement to assist in the judgment. It cannot simultaneously reflect the overall compaction level, uniformity and local defects of the filling layer. The weak stiffness area is easily covered by the average indicator, forming hidden quality hazards.

[0006] (4) Construction control only has "post-judgment" and no "pre-prediction": The existing compaction monitoring is a post-control mode of "compacting and testing at the same time, and compacting again if it is not qualified". The number of compaction passes and the thickness of loose paving are initially determined by experience. It is impossible to predict the construction parameters required to meet the acceptance standards before each layer is constructed. Over-compaction waste and under-compaction rework are difficult to avoid. Moreover, the existing compaction data has not been settled into a reusable prediction model. Each layer starts from "zero experience".

[0007] (5) Ignoring the mutual influence of interlayer stiffness: The filling body is a layered structure that is stacked layer by layer. The compaction state of the lower layer directly affects the compaction efficiency and bearing capacity of the upper layer. The existing acceptance adopts independent judgment criteria for each layer, which may result in "interlayer stiffness inversion" where the upper layer stiffness is higher than the lower layer stiffness, resulting in differential settlement under load.

[0008] (6) Geological model is "determined for life after one exploration": Geological exploration is completed before construction begins. The spacing between exploration boreholes is limited. Unexplored weak bodies are only passively exposed after quality problems occur during filling. At this time, the cost of rework is high. A large amount of compaction response data generated during construction is not used to verify and correct the geological model, resulting in data waste. Summary of the Invention

[0009] The purpose of this invention is to overcome the above-mentioned shortcomings and provide an intelligent construction method and system for site leveling projects, so as to realize predictive control of construction quality and dynamic optimization of filler resources.

[0010] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A smart construction method for site leveling projects includes the following steps: S1. Multi-source survey and 3D overlay modeling: Establish a 3D overlay model that includes topography, soft strata and design surfaces; S2. Construction quality zoning and initial optimization of earthwork allocation: Based on the distribution of weak strata in the composite model, construction quality zones are divided and foundation treatment levels are determined. An earthwork allocation model is established with the goal of minimizing the sum of transportation costs and quality risk costs, and an initial earthwork allocation scheme is output. S3. Layered excavation and differentiated pretreatment according to treatment level: Layered excavation is carried out according to the initial earthwork allocation plan, and differentiated pretreatment is implemented according to the foundation treatment level of each construction quality zone. S4. Layered filling and compaction process monitoring: Layered filling and real-time acquisition of compaction process signals and rolling trajectory, calculation of compaction quality indicators characterizing the compaction state of the filling layer, and binding of compaction quality indicators with the spatial coordinates of the rolling trajectory; S5. Stiffness Field Inversion and Joint Judgment by Three Indicators: The compaction quality indicators bound to the compaction trajectory are inverted through gridded interpolation to generate a stiffness field distribution map of the fill layer; the corresponding differentiated acceptance threshold group is called according to the foundation treatment level of the construction quality zone, and the differentiated acceptance threshold group is dynamically corrected between layers based on the measured stiffness field of the next layer already filled; the three indicators of representative compaction quality indicator, stiffness field uniformity dispersion value, and area ratio of stiffness weak zone are used to jointly determine whether the fill layer is qualified; if it is determined to be unqualified, graded treatment is performed and retesting is carried out until it is determined to be qualified; S6. Feedforward Prediction and Feedback Correction: Based on the stiffness field distribution sequence of each filled layer and the corresponding construction parameters, a stiffness growth response model is established to characterize the changes in compaction quality indicators with the number of compaction passes and the loose paving thickness. Before the construction of each filling layer, the number of compaction passes and the loose paving thickness required for the filling layer to reach the differentiated acceptance threshold group are predicted in a feedforward manner to guide the construction of the layer. After the layer is deemed qualified, the model parameters of the stiffness growth response model are corrected by the deviation between the measured compaction quality indicators and the predicted values. At the same time, the stiffness field distribution map of the layer and the measured parameters of the fill material are fed back to the earthwork allocation model. The remaining earthwork allocation scheme is further optimized with the spatial location of the stiffness weak zone as a constraint, so that the fill material with the gradation and moisture content meeting the preset requirements is preferentially allocated to the construction quality zone where the stiffness weak zone is located. S7. Geological Model Backwriting and Supplementary Investigation Trigger: The compaction response distribution of each filling layer identified in step S5 is backwritten to the three-dimensional distribution model of weak strata in the composite model, and the spatial distribution of weak strata is dynamically corrected; when multiple consecutive filling layers at the same plane location are identified as weak stiffness areas, it is determined that there are unexplored weak bodies at that location, triggering supplementary investigation at that location, and updating the composite model, re-dividing the construction quality zones and their foundation treatment levels according to the supplementary investigation results, and reconstructing the earthwork allocation plan accordingly. S8. Repeat steps S4 to S7 until the filling is completed. The site is then finely leveled, and the site elevation is re-measured using GPS real-time dynamic measurement technology. After passing the inspection, the as-built data is generated.

[0011] Preferably, step S1 specifically involves: using a drone equipped with a lidar scanner to perform a full-coverage scan of the construction site, acquiring site cloud data, establishing a three-dimensional terrain model, interpolating the geological survey data to build a three-dimensional distribution model of weak strata, and overlaying it with the design surface model established according to the design elevation to form an overlay model.

[0012] Preferably, step S2 specifically involves: dividing the construction site into several construction quality zones based on the burial depth and thickness of the weak strata in the superimposed model, and determining the foundation treatment level of each construction quality zone; calculating the excavation and filling volumes of each construction quality zone based on the superimposed model, establishing an earthwork allocation model with the objective function of minimizing the sum of transportation costs and quality risk costs, and solving and outputting the initial earthwork allocation scheme.

[0013] Preferably, in step S2, the foundation treatment level is divided into three levels according to the burial depth and thickness of the weak strata: Level I is where there are no weak strata distributed within the construction quality zone; Level II is where there are weak strata with a burial depth greater than the preset burial depth value and a thickness less than the preset thickness value; and Level III is where the burial depth of the weak strata is not greater than the preset burial depth value or the thickness is not less than the preset thickness value.

[0014] Preferably, in step S2, the constraints of the earthwork allocation model include: the moisture content of each material source filler must be within a preset moisture content range; the daily allocation amount of each construction quality zone does not exceed the daily production capacity of the compaction equipment; and the working face timing constraint, that is, before the construction quality zone has completed differentiated pretreatment and passed acceptance, the filling working face corresponding to that zone will not arrange for filler to be transferred in.

[0015] Preferably, step S3 specifically involves: excavating the excavation area layer by layer from top to bottom according to the initial earthwork allocation plan; implementing differentiated pretreatment for the filling area according to the foundation treatment level of each construction quality zone. The higher the foundation treatment level, the greater the dynamic compaction energy level or the greater the replacement thickness used in the pretreatment. Only construction quality zones that have completed pretreatment and passed acceptance can proceed to the filling process.

[0016] Preferably, step S4 specifically involves: filling soil in layers within the qualified construction quality zone; using an intelligent compaction monitoring device on the compaction machinery to collect vibration acceleration signals and rolling trajectories during the compaction process in real time; calculating compaction quality indicators that characterize the compaction state of the filling layer from the vibration acceleration signals; and binding and uploading the compaction quality indicators with the spatial coordinates of the rolling trajectories.

[0017] Preferably, in step S5, the method for generating the stiffness field distribution map is as follows: taking the sampling points on the compaction trajectory as discrete points, the compaction quality index bound to each sampling point is mapped to a regular grid using inverse distance weighted interpolation or Kriging interpolation, and the grid size is no greater than half the width of the compaction machine roller, to obtain the stiffness field distribution map.

[0018] Preferably, in step S5, the rule for the joint determination of the three indicators is as follows: the representative value of the compaction quality index is not lower than the lower limit of the representative value in the differentiated acceptance threshold group, the discrete value of the stiffness field uniformity is not greater than the upper limit of the discrete value in the differentiated acceptance threshold group, and the area ratio of the stiffness weak zone is not greater than the upper limit of the ratio in the differentiated acceptance threshold group. When all three indicators are met, the filling layer is determined to be qualified; wherein the stiffness weak zone is a continuous grid area in the stiffness field distribution map where the compaction quality index is lower than the lower limit of the representative value.

[0019] Preferably, the rule for determining the value of the differentiated acceptance threshold group is as follows: the higher the foundation treatment level, the larger the lower limit of the corresponding compaction quality index representative value, and the smaller the upper limit of the uniformity dispersion value of the stiffness field and the upper limit of the area ratio of the stiffness weak zone.

[0020] Preferably, in step S5, the graded treatment is carried out in the following order: First level, the area with weak stiffness is compacted and then retested; Second level, if the retest after compaction is still unqualified, the fill material in the area with weak stiffness is dried or improved by admixtures and then compacted and retested; Third level, if the retest after the second level of treatment is still unqualified, the area with weak stiffness is partially replaced and retested, and the replacement location and replacement material information are backfilled into the earthwork mixing model.

[0021] Preferably, in step S5, the rule for interlayer dynamic correction is as follows: when the measured compaction quality index of the grid position corresponding to the next layer of filled material is lower than the lower limit of the representative value of the differentiated acceptance threshold group of the corresponding zone, the lower limit of the representative value of the grid position corresponding to the current filling layer is increased by a preset amount to prevent the formation of interlayer stiffness inversion where the upper layer stiffness is higher than the lower layer stiffness; when the measured compaction quality index of the grid position corresponding to the next layer of filled material is higher than the lower limit of the representative value to a preset margin, the target number of compaction passes of the current filling layer is reduced by a preset amount.

[0022] Preferably, in step S6, the method for establishing and correcting the stiffness growth response model is as follows: using the sampling sequence of compaction quality index at the same grid location of each filled layer as the number of compaction passes increases as training data, a response curve characterizing the mapping relationship between compaction quality index and number of compaction passes and loose paving thickness is fitted; when constructing the next filled layer, the target number of compaction passes and target loose paving thickness required to reach the differentiated acceptance threshold group are extrapolated from the response curve; after the construction of this layer is completed, the model parameters of the response curve are updated with the deviation between the measured compaction quality index sequence and the predicted value of the response curve.

[0023] Preferably, in step S7, the triggering condition for supplementary investigation is: the same plane location is identified as a weak stiffness zone in the stiffness field distribution map of N consecutive filling layers, where N is a preset integer not less than 2; after the supplementary investigation is completed, the spatial range of the weak body revealed by the investigation is written into the three-dimensional distribution model of the weak strata, and the construction quality zoning, foundation treatment level determination and earthwork allocation scheme solution in step S2 are re-executed.

[0024] Preferably, it also includes environmental protection measures during construction, including: setting up dust nets or water spraying devices at the boundary of excavation and filling construction; setting up temporary drainage ditches around the construction area; and ensuring that the exhaust emissions of construction machinery meet the requirements of the national standard GB 20891-2014 "Emission Limits and Measurement Methods for Exhaust Pollutants from Diesel Engines for Non-Road Mobile Machinery".

[0025] In addition, this invention also discloses an intelligent construction system for site leveling engineering, used to implement the above-mentioned intelligent construction method for site leveling engineering, comprising: The data acquisition module includes a drone equipped with a lidar scanner and an inertial measurement unit, used to collect three-dimensional point cloud data of the construction site; The 3D modeling and analysis module communicates with the data acquisition module to establish an overlay model by combining point cloud data, geological survey data, and design elevations; divide construction quality zones and determine foundation treatment levels; establish and solve the earthwork allocation model; and update the overlay model based on the write-back data. The intelligent compaction monitoring module, installed on compaction machinery, includes an acceleration sensor, a vibration frequency acquisition device, and a wireless data transmission unit, used to collect and upload vibration acceleration signals and compaction trajectory in real time. The construction control center communicates with the 3D modeling and analysis module and the intelligent compaction monitoring module, and has built-in stiffness field inversion unit, joint judgment unit, feedforward prediction unit, feedback correction unit, and model update unit: The stiffness field inversion unit is used to generate a stiffness field distribution map by interpolating the compaction quality indicators into a grid; the joint judgment unit is used to call differentiated acceptance threshold groups according to the foundation treatment level of the construction quality zone and perform inter-layer dynamic correction, and perform joint judgment of three indicators; the feedforward prediction unit is used to establish a stiffness growth response model and predict the number of compaction passes and loose paving thickness of the next filling layer; the feedback correction unit is used to correct the model parameters of the stiffness growth response model based on the deviation between the measured values ​​and the predicted values, and to send the judgment results and stiffness field distribution map back to the 3D modeling and analysis module to trigger the re-optimization of the earthwork allocation plan; the model update unit is used to write the compaction response distribution back to the 3D distribution model of the weak strata, and to trigger supplementary investigation when multiple consecutive filling layers are judged to be weak stiffness zones at the same plane location; GPS real-time dynamic measurement unit is used for final elevation re-measurement.

[0026] Furthermore, the construction control center is equipped with a BIM integration platform, which integrates and displays the composite model, construction quality zoning, stiffness field distribution map of each filling layer, stiffness growth response curve, joint judgment results and construction progress, so as to realize the visualization management of the entire construction process.

[0027] Beneficial effects of this invention: 1. This invention enables the compaction quality data to simultaneously serve the acceptance of the current layer, the correction of the lower layer, the prediction of the upper layer, and the updating of the geological model through a self-updating closed-loop control of zoning and grading, differentiated processing and acceptance, feedforward prediction and feedback correction, and geological model writing, thereby achieving predictive control of the construction quality of site leveling projects and dynamic optimization of filler resources.

[0028] 2. This invention upgrades post-construction control to predictive control. Its stiffness growth response model uses data from each filled layer to fit the mapping relationship between compaction quality indicators and the number of compaction passes and loose paving thickness. Before construction of each layer, the target number of compaction passes and loose paving thickness are predicted using feedforward. After construction, the model parameters are corrected using measured deviations, forming a double loop of feedforward prediction and feedback correction. The compaction parameters are transformed from being initially determined based on experience and requiring rework if unqualified, to being predicted by the model and evolving with construction. At the same time, it suppresses waste from over-compaction and rework from under-compaction (see...). Figure 5 ).

[0029] (2) The present invention eliminates the inversion of interlayer stiffness. Its acceptance threshold group is dynamically corrected by the measured stiffness field of the next layer. The lower layer under-compression position corresponds to the increase of the upper layer acceptance standard, and the lower layer with sufficient margin corresponds to the decrease of the upper layer target compaction pass, so as to ensure the coordinated distribution of the stiffness of the filling body along the layer and avoid differential settlement between layers.

[0030] (3) The compaction data of the present invention feeds back into the geological model, realizing the compaction response distribution of each layer back to the three-dimensional distribution model of the weak strata; when multiple layers in the same plane are identified as weak, supplementary exploration is triggered, and the exploration results update the superimposed model and re-zone and reconstruct the allocation plan. The unexplored weak bodies are transformed from passive exposure after a quality accident to continuous weak automatic early warning and active supplementary exploration, and each supplementary exploration makes the model more accurate. The superimposed model continues to evolve throughout the entire construction cycle.

[0031] (4) This invention forms a triple nested closed loop: inner loop (this layer: monitoring, inversion, judgment, and graded handling), middle loop (interlayer: feedforward prediction, construction, feedback correction, and threshold inheritance correction), and outer loop (the entire field: geological model rewriting, supplementary exploration, rezoning, and reconfiguration). The three loops share the same stiffness field data stream, and the data is collected once and reused at four levels (acceptance, lower-level correction, upper-level prediction, and geological update). The functions of each link are mutually restrictive and mutually supportive, achieving a synergistic effect that cannot be achieved by superimposing a single technology.

[0032] (5) The beneficial effects of the present invention, which links the treatment and acceptance levels, jointly judges the three indicators, and adjusts the multiple constraints, are superimposed with the aforementioned mechanisms, so that the treatment effect is closed-loop through differentiated acceptance verification, targeted reinforcement of weak areas, and configuration of high-quality fillers as needed. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the overall process of the intelligent construction method for site leveling engineering of the present invention, in which the dashed arrows represent the closed-loop feedback and model write-back paths.

[0034] Figure 2 This is a schematic diagram of the architecture of the intelligent construction system of the present invention.

[0035] Figure 3This is a schematic diagram illustrating the working principle of compaction monitoring, stiffness field inversion and feedforward prediction, and feedback correction in steps S4 to S6.

[0036] Figure 4 This is a schematic diagram illustrating the linkage between construction quality zoning, foundation treatment level, and differentiated acceptance threshold groups.

[0037] Figure 5 This is a schematic diagram illustrating the principles of feedforward prediction-feedback correction dual-loop control, interlayer stiffness transfer correction, and geological model write-back triggering.

[0038] Figure labeling: 1. Unmanned Aerial Vehicle (UAV); 2. LiDAR Scanner; 10. Data Acquisition Module; 20. 3D Modeling and Analysis Module; 30. Intelligent Compaction Monitoring Module; 31. Accelerometer; 32. Vibration Frequency Acquisition Device; 33. Wireless Data Transmission Unit; 40. Construction Control Center; 41. BIM Integration Platform; 42. Feedforward Prediction Unit; 43. Model Update Unit; 50. GPS Real-time Dynamic Measurement Unit; 60. Compaction Machinery; 61. Roller Wheel; 70. Fill Layer; 70a. Filled Layer; 70b. Current Fill Layer; 71. Stiffness Field Distribution Grid; 72. Stiffness Weak Zone; 74. Stiffness Growth Response Curve; 75. Predicted Trajectory; 76. Measured Sampling Point; 80. Natural Foundation; 81. Soft Strata; 82. Replacement Zone; 83. Supplementary Exploration Borehole. Detailed Implementation

[0039] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0040] Example 1: An intelligent construction method for site leveling engineering, including the following steps: S1 Multi-source survey and 3D overlay modeling; S2 Initial optimization of construction quality zoning and earthwork allocation; S3 Layered excavation and differentiated preprocessing according to treatment level; S4 Layered filling and compaction process monitoring; S5 Stiffness field inversion and joint judgment of three indicators; S6 Feedforward prediction and feedback correction; S7 Geological model writing back and supplementary survey triggering; S8 Cyclic filling and as-built re-survey. The specific contents of each step are as follows: S1. Multi-source survey and 3D overlay modeling: Use a drone equipped with a lidar scanner to conduct a full-coverage scan of the construction site, obtain site cloud data, establish a 3D terrain model, and interpolate the geological survey data to build a 3D distribution model of weak strata, which is then overlaid with the design surface model established according to the design elevation to form an overlay model. S2. Initial Optimization of Construction Quality Zoning and Earthwork Allocation: Based on the depth and thickness of the soft strata in the superimposed model, the construction site is divided into several construction quality zones, and the foundation treatment level of each construction quality zone is determined; the excavation and filling volumes of each construction quality zone are calculated based on the superimposed model, and an earthwork allocation model is established with the objective function of minimizing the sum of transportation costs and quality risk costs, and the initial earthwork allocation scheme is output. S3. Layered excavation and differentiated pretreatment according to treatment level: The excavation area is excavated in layers from top to bottom according to the initial earthwork allocation plan; the filling area is pretreated in a differentiated manner according to the foundation treatment level of each construction quality zone. The higher the foundation treatment level, the greater the dynamic compaction energy level or the greater the replacement thickness used in the pretreatment. Only construction quality zones that have completed pretreatment and passed acceptance can enter the filling process. S4. Layered filling and compaction process monitoring: In the construction quality zone that has passed the acceptance inspection, the soil material is filled in layers. The intelligent compaction monitoring device set on the compaction machinery collects the vibration acceleration signal and rolling trajectory in real time during the compaction process. The compaction quality index that characterizes the compaction state of the filling layer is calculated from the vibration acceleration signal, and the compaction quality index is bound to the spatial coordinates of the rolling trajectory and uploaded. S5. Stiffness Field Inversion and Joint Judgment by Three Indicators: The compaction quality indicators bound to the compaction trajectory are inverted through gridded interpolation to generate a stiffness field distribution map of the fill layer; the corresponding differentiated acceptance threshold group is called according to the foundation treatment level of the construction quality zone, and the differentiated acceptance threshold group is dynamically corrected between layers based on the measured stiffness field of the next layer already filled; the fill layer is jointly judged as qualified based on three indicators: the representative value of the compaction quality indicator, the discrete value of the stiffness field uniformity, and the area ratio of the stiffness weak zone; if the judgment is unqualified, graded treatment is performed and retesting is carried out until the judgment is qualified; S6. Feedforward Prediction and Feedback Correction: Based on the stiffness field distribution sequence of each filled layer and the corresponding construction parameters, a stiffness growth response model is established to characterize the changes in compaction quality indicators with the number of compaction passes and the loose paving thickness. Before the construction of each filling layer, the number of compaction passes and the loose paving thickness required for the filling layer to reach the differentiated acceptance threshold group are predicted in a feedforward manner to guide the construction of the layer. After the layer is deemed qualified, the model parameters of the stiffness growth response model are corrected by the deviation between the measured compaction quality indicators and the predicted values. At the same time, the stiffness field distribution map of the layer and the measured parameters of the fill material are fed back to the earthwork allocation model. The remaining earthwork allocation scheme is further optimized with the spatial location of the stiffness weak zone as a constraint, so that the fill material with the gradation and moisture content meeting the preset requirements is preferentially allocated to the construction quality zone where the stiffness weak zone is located. S7. Geological Model Backwriting and Supplementary Investigation Trigger: The compaction response distribution of each filling layer identified in step S5 is backwritten to the three-dimensional distribution model of the weak strata in the composite model, and the spatial distribution of the weak strata is dynamically corrected; when multiple consecutive filling layers at the same plane location are identified as weak stiffness areas, it is determined that there are unexplored weak bodies at that location, triggering supplementary investigation at that location, and updating the composite model, re-dividing the construction quality zones and their foundation treatment levels according to the supplementary investigation results, and reconstructing the earthwork allocation plan accordingly; S8. Repeat steps S4 to S7 until the filling is completed. The site is then finely leveled, and the site elevation is re-measured using GPS real-time dynamic measurement technology. After passing the inspection, the as-built data is generated.

[0041] In step S2, the foundation treatment level is divided into three levels according to the burial depth and thickness of the weak strata: Level I is where there are no weak strata distributed within the construction quality zone; Level II is where there are weak strata with a burial depth greater than the preset burial depth value and a thickness less than the preset thickness value; and Level III is where the burial depth of the weak strata is not greater than the preset burial depth value or the thickness is not less than the preset thickness value.

[0042] In step S5, the stiffness field distribution map is generated as follows: using the sampling points on the compaction trajectory as discrete points, the compaction quality index bound to each sampling point is mapped to a regular grid using inverse distance weighted interpolation or Kriging interpolation. The grid size is no larger than half the width of the compaction machine roller, thus obtaining the stiffness field distribution map.

[0043] In step S5, the rule for the joint judgment of the three indicators is as follows: the representative value of the compaction quality index is not lower than the lower limit of the representative value in the differentiated acceptance threshold group, the discrete value of the stiffness field uniformity is not greater than the upper limit of the discrete value in the differentiated acceptance threshold group, and the area ratio of the stiffness weak zone is not greater than the upper limit of the ratio in the differentiated acceptance threshold group. When all three indicators are met, the filling layer is judged to be qualified. Among them, the stiffness weak zone is the continuous grid area in the stiffness field distribution map where the compaction quality index is lower than the lower limit of the representative value.

[0044] In step S5, the graded treatment is carried out in the following order: Level 1, after supplementary compaction of the weak stiffness area, retesting is performed; Level 2, if the retesting after supplementary compaction is still unqualified, the fill material in the weak stiffness area is turned over or improved by admixtures and then re-compacted and retested; Level 3, if the retesting after Level 2 treatment is still unqualified, the weak stiffness area is partially replaced and retested, and at the same time, the replacement location and replacement material information are backfilled into the earthwork mixing model.

[0045] In step S2, the constraints of the earthwork allocation model include: the moisture content of each material source must be within the preset moisture content range; the daily allocation amount of each construction quality zone does not exceed the daily production capacity of the compaction equipment; and the working face time constraint, that is, before the construction quality zone has completed differentiated pretreatment and passed acceptance, the filling working face corresponding to the zone will not be arranged to transfer filler.

[0046] The rules for determining the value of the differentiated acceptance threshold group are as follows: the higher the foundation treatment level, the larger the lower limit of the corresponding compaction quality index representative value, and the smaller the upper limit of the uniformity dispersion value of the stiffness field and the upper limit of the area ratio of the stiffness weak zone.

[0047] In step S6, the method for establishing and correcting the stiffness growth response model is as follows: using the sampling sequence of compaction quality index at the same grid location of each filled layer as the number of compaction passes increases as training data, a response curve characterizing the mapping relationship between compaction quality index and number of compaction passes and loose paving thickness is fitted; when constructing the next filled layer, the target number of compaction passes and target loose paving thickness required to reach the differentiated acceptance threshold group are extrapolated from the response curve; after the construction of this layer is completed, the model parameters of the response curve are updated with the deviation between the measured compaction quality index sequence and the predicted value of the response curve.

[0048] In step S5, the rules for interlayer dynamic correction are as follows: when the measured compaction quality index of the grid position corresponding to the next layer of filled material is lower than the lower limit of the representative value of the differential acceptance threshold group of the corresponding zone, the lower limit of the representative value of the grid position corresponding to the current filling layer is increased by a preset amount to prevent the formation of interlayer stiffness inversion where the upper layer stiffness is higher than the lower layer stiffness; when the measured compaction quality index of the grid position corresponding to the next layer of filled material is higher than the lower limit of the representative value to a preset margin, the target number of compaction passes of the current filling layer is reduced by a preset amount.

[0049] In step S7, the triggering condition for supplementary exploration is: the same plane location is identified as a weak stiffness zone in the stiffness field distribution map of N consecutive filling layers, where N is a preset integer not less than 2; after the supplementary exploration is completed, the spatial range of the weak body revealed by the exploration is written into the three-dimensional distribution model of the weak strata, and the construction quality zoning, foundation treatment level determination and earthwork allocation scheme solution in step S2 are re-executed.

[0050] Environmental protection measures during construction include: setting up dust nets or water spraying devices at the boundaries of excavation and filling construction; setting up temporary drainage ditches around the construction area; and ensuring that the exhaust emissions of construction machinery meet the requirements of the national standard GB 20891-2014 "Emission Limits and Measurement Methods for Exhaust Pollutants from Diesel Engines for Non-Road Mobile Machinery (China Stage III and IV)".

[0051] Example 2: An intelligent construction system for implementing the above method includes a data acquisition module, a 3D modeling and analysis module, an intelligent compaction monitoring module, a construction control center, and a GPS real-time dynamic measurement unit. The construction control center incorporates a stiffness field inversion unit, a joint judgment unit, a feedforward prediction unit, a feedback correction unit, and a model update unit. (See attached text) Figure 2 The specific contents of each module are as follows: The data acquisition module 10 includes a drone 1 equipped with a lidar scanner 2 and an inertial measurement unit, used to acquire three-dimensional point cloud data of the construction site; The 3D modeling and analysis module 20 is connected to the data acquisition module 10. It is used to establish an overlay model by combining point cloud data, geological survey data, and design elevation, divide construction quality zones and determine the foundation treatment level, establish and solve the earthwork allocation model, and update the overlay model based on the write-back data. The intelligent compaction monitoring module 30 is installed on the compaction machine 60 and includes an acceleration sensor 31, a vibration frequency acquisition device 32 and a wireless data transmission unit 33, which is used to collect vibration acceleration signals and compaction trajectory in real time and upload them. The construction control center 40 is communicatively connected to the 3D modeling and analysis module 20 and the intelligent compaction monitoring module 30. It includes a stiffness field inversion unit, a joint judgment unit, a feedforward prediction unit, a feedback correction unit, and a model update unit. The stiffness field inversion unit generates a stiffness field distribution map by interpolating the compaction quality indicators into a grid. The joint judgment unit calls differentiated acceptance threshold groups according to the foundation treatment level of the construction quality zone and performs inter-layer dynamic correction, conducting joint judgment of the three indicators. The feedforward prediction unit establishes a stiffness growth response model and predicts the number of compaction passes and the loose paving thickness of the next filling layer. The feedback correction unit corrects the model parameters of the stiffness growth response model based on the deviation between measured and predicted values, and sends the judgment results and stiffness field distribution map back to the 3D modeling and analysis module 20, triggering further optimization of the earthwork allocation scheme. The model update unit writes the compaction response distribution back to the 3D distribution model of the weak strata and triggers supplementary investigation when multiple consecutive filling layers are all judged as stiffness-weak areas at the same plane location. GPS real-time dynamic measurement unit 50 is used for final elevation re-measurement.

[0052] The construction control center 40 is equipped with a BIM integration platform 41, which integrates and displays the composite model, construction quality zoning, stiffness field distribution map of each filling layer, stiffness growth response curve, joint judgment results and construction progress, so as to realize the visualization management of the entire construction process.

[0053] Example 3: As Figure 1 As shown, the intelligent construction method for site leveling engineering in this embodiment is implemented according to the following process: Step S1: Use UAV 1 equipped with LiDAR scanner 2 to perform a full-coverage scan of the construction site at a flight altitude of 80-120m and a scanning point density of no less than 50 points / m² to acquire site ground cloud data and establish a three-dimensional terrain model with an accuracy of no less than 5cm; use three-dimensional interpolation to establish a three-dimensional distribution model of weak strata based on geological survey data such as borehole and static penetration tests; establish a design surface model according to the design elevation; and overlay the three models in a unified coordinate system to form an overlay model.

[0054] Step S2: Based on the depth and thickness of the weak stratum 81 in the composite model, the site is divided into several construction quality zones and graded: Zones without weak strata are classified as Grade I; zones with weak strata depth greater than the preset depth value (2m in this embodiment) and thickness less than the preset thickness value (2m in this embodiment) are classified as Grade II; zones with a depth not greater than 2m or a thickness not less than 2m are classified as Grade III. Based on the composite model, the grid method is used to calculate the excavation and filling volume of each zone, and an allocation model is established: the objective function is to minimize the sum of transportation cost and quality risk cost, where the quality risk cost is weighted by the foundation treatment level of the receiving zone based on the transportation distance; constraints include the fill material moisture content being within the range of optimum moisture content ±2%, the daily allocation volume of each zone not exceeding the daily capacity of the compaction equipment, and working face timing constraints (fill material is not arranged to be transferred in before the zone has passed acceptance). The initial earthwork allocation scheme is output from the solution.

[0055] Step S3: Excavate the excavation area in layers from top to bottom according to the allocation plan. The thickness of each excavation layer is 200-300mm. The excavator is equipped with a Beidou / GPS positioning system to record the excavation trajectory in real time. For the fill area, pre-treatment is differentiated according to the treatment level: Level I areas are directly cleared and leveled; Level II areas are treated with dynamic compaction, with a point compaction energy of 2000-4000 kN·m, a point compaction spacing of 4-6m, and a full compaction energy of 1000-1500 kN·m; Level III areas are treated with dynamic compaction replacement or backfilling, using graded crushed stone or gravel as the backfill material, with a backfill thickness of not less than 1.5m and a point compaction energy of 4000-6000 kN·m. Only areas that have completed pre-treatment and passed acceptance can proceed to the filling process.

[0056] Step S4: Fill the zoning and layering in the qualified areas. The thickness of the first layer is 250-350mm (the subsequent layers are determined by the feedforward prediction in step S6). The acceleration sensor 31 and vibration frequency collector 32 on the compaction machine 60 collect the vibration acceleration signal in real time during the rolling process and calculate the compaction quality index (the compaction value CMV is used in this embodiment, but the resonance value RMV or the subgrade reaction coefficient based on the second integral of acceleration can also be used). The wireless data transmission unit 33 binds the compaction quality index with the rolling trajectory coordinates and uploads it to the construction control center 40. The moisture content of the fill material is controlled within ±2% of the optimum moisture content. Soil materials that do not meet the requirements are used after being sun-dried or watered.

[0057] Step S5: The stiffness field inversion unit uses the compaction trajectory sampling points as discrete points and employs inverse distance weighted interpolation to map the compaction quality index to a regular grid (grid size not greater than half the width of the compaction wheel 61), generating a stiffness field distribution map of the fill layer (e.g., Figure 3 The stiffness field distribution grid 71 in the middle); the joint judgment unit first calls the differentiated acceptance threshold group according to the zoning processing level (the higher the processing level, the larger the lower limit of the representative value, and the smaller the upper limit of the discrete value and the upper limit of the proportion). Then, the threshold group is dynamically corrected between layers by the measured stiffness field of the next layer of filled layer: if the measured compaction quality index of the corresponding grid position of the next layer is lower than the lower limit of the representative value, the lower limit of the representative value of the corresponding grid position of the current layer is increased by a preset range (3% to 5% in this embodiment) to prevent the formation of interlayer stiffness inversion with hard upper and soft lower layers; if the measured compaction quality index of the next layer is higher than the lower limit of the representative value to reach a preset margin (8% in this embodiment), the target compaction pass of the current layer is reduced by a preset range. Then, the three indicators are jointly judged: ① the representative value of the compaction quality index is not lower than the corrected lower limit of the representative value; ② the discrete value of the stiffness field uniformity (the coefficient of variation is used in this embodiment) is not greater than the upper limit of the discrete value; ③ the area proportion of the stiffness weak zone 72 is not greater than the upper limit of the proportion. If all three conditions are met, the judgment is qualified.

[0058] If the test result is unqualified, the following graded measures shall be taken: Grade 1: Compact the weak area 72 with 2 to 4 additional passes and retest; Grade 2: If the test result is still unqualified, the filler in the weak area shall be sun-dried or improved with admixtures (lime, graded crushed stone) and then compacted and retested; Grade 3: If the test result is still unqualified, the weak area shall be partially replaced (forming a replacement area 82) and retested. At the same time, the replacement location and replacement material information shall be backfilled into the mixing model.

[0059] Step S6: The feedforward prediction unit 42 uses the sampling sequence of compaction quality index at the same grid location of each filled layer as the number of compaction passes as training data to fit the stiffness growth response curve 74 (in this embodiment, an asymptotic function is used for fitting; piecewise linear or machine learning regression models can also be used) to characterize the mapping relationship between compaction quality index and the number of compaction passes and loose paving thickness. Before constructing the next filling layer, the target number of compaction passes and target loose paving thickness required to reach the differentiated acceptance threshold group are extrapolated from the response curve and sent to the compaction machinery to guide construction. After the layer is constructed and judged to be qualified, the feedback correction unit updates the model parameters of the response curve with the deviation between the measured compaction quality index sequence (measured sampling point 76) and the predicted trajectory 75 of the response curve, so that the model evolves layer by layer with the construction process (see...). Figure 5 Simultaneously, the feedback correction unit transmits the stiffness field distribution map of this layer and the measured parameters of the fill material back to the three-dimensional modeling and analysis module 20. Using the spatial location of the stiffness-weak zone 72 as a constraint, the remaining earthwork allocation scheme is further optimized, increasing the quality risk cost weight of the zone where the weak zone is located, so that fill material with good gradation and suitable moisture content is preferentially allocated to this zone.

[0060] Step S7: The model update unit 43 writes back the compaction response distribution of each filling layer identified in step S5 to the three-dimensional distribution model of the weak stratum, and dynamically corrects the spatial distribution of the weak stratum 81; when the same plane location is identified as a weak stiffness area in the stiffness field distribution map of N consecutive filling layers (N is 3 in this embodiment), it is determined that there is an unexplored weak body at this location, and a supplementary exploration command is automatically issued to arrange supplementary exploration boreholes 83 to reveal the range of the weak body; after the supplementary exploration is completed, the exploration results are written into the three-dimensional distribution model of the weak stratum, the superimposed model is updated, and the zoning, grade determination and allocation scheme solution of step S2 are re-executed to achieve "exploration by compaction" (see Figure 5 ).

[0061] Step S8: Repeat steps S4 to S7 until filling is completed. The site is then finely leveled, and the site elevation is re-measured using a GPS real-time dynamic measurement unit 50. The elevation deviation is controlled within ±30mm. After passing inspection, as-built documentation is generated. This documentation includes stiffness field distribution maps for each filling layer, joint judgment records, stiffness growth response curve evolution records, and supplementary investigation trigger records, achieving full traceability of quality throughout the entire process.

[0062] Example 4: Figure 2 As shown, the intelligent construction system in this embodiment includes a data acquisition module 10, a 3D modeling and analysis module 20, an intelligent compaction monitoring module 30, a construction control center 40, and a GPS real-time dynamic measurement unit 50. The modules are connected via wireless network communication. The construction control center 40 has a built-in stiffness field inversion unit, a joint judgment unit, a feedforward prediction unit 42, a feedback correction unit, and a model update unit 43. It is also equipped with a BIM integration platform 41, which integrates and displays the superimposed model, construction quality zoning, stiffness field distribution map of each layer, stiffness growth response curve, judgment results, and construction progress.

[0063] The core of this invention lies in a triple-nested closed loop: the inner loop handles monitoring, inversion, judgment, and tiered processing within the current layer; the middle loop handles inter-layer feedforward prediction, construction, feedback correction, and threshold inheritance adjustment; and the outer loop handles geological model rewriting, supplementary exploration, rezoning, and reconfiguration for the entire field. All three loops share the same stiffness field data stream, enabling compaction data to be collected once and reused at four levels (current layer acceptance, lower layer correction, upper layer prediction, and geological update). Surveying and modeling, quality zoning, allocation optimization, and compaction monitoring are truly integrated into a self-updating whole, achieving synergistic effects that cannot be achieved by implementing each stage independently.

[0064] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A smart construction method for site leveling projects, characterized in that, Includes the following steps: S1. Multi-source survey and 3D overlay modeling: Establish a 3D overlay model that includes topography, soft strata and design surfaces; S2. Construction quality zoning and initial optimization of earthwork allocation: Based on the distribution of weak strata in the composite model, construction quality zones are divided and foundation treatment levels are determined. An earthwork allocation model is established with the goal of minimizing the sum of transportation costs and quality risk costs, and an initial earthwork allocation scheme is output. S3. Layered excavation and differentiated pretreatment according to treatment level: Layered excavation is carried out according to the initial earthwork allocation plan, and differentiated pretreatment is implemented according to the foundation treatment level of each construction quality zone. S4. Layered filling and compaction process monitoring: Layered filling and real-time acquisition of compaction process signals and rolling trajectory, calculation of compaction quality indicators characterizing the compaction state of the filling layer, and binding of compaction quality indicators with the spatial coordinates of the rolling trajectory; S5. Stiffness Field Inversion and Joint Judgment by Three Indicators: The compaction quality indicators bound to the compaction trajectory are inverted through gridded interpolation to generate a stiffness field distribution map of the fill layer; the corresponding differentiated acceptance threshold group is called according to the foundation treatment level of the construction quality zone, and the differentiated acceptance threshold group is dynamically corrected between layers based on the measured stiffness field of the next layer already filled; the fill layer is jointly judged as qualified based on three indicators: the representative value of the compaction quality indicator, the discrete value of the stiffness field uniformity, and the area ratio of the stiffness weak zone; if the judgment is unqualified, graded treatment is performed and retesting is carried out until the judgment is qualified; S6. Feedforward Prediction and Feedback Correction: Based on the stiffness field distribution sequence of each filled layer and the corresponding construction parameters, a stiffness growth response model is established to characterize the changes in compaction quality indicators with the number of compaction passes and the loose paving thickness. Before the construction of each filling layer, the number of compaction passes and the loose paving thickness required for the filling layer to reach the differentiated acceptance threshold group are predicted in a feedforward manner to guide the construction of the layer. After the layer is deemed qualified, the model parameters of the stiffness growth response model are corrected by the deviation between the measured compaction quality indicators and the predicted values. At the same time, the stiffness field distribution map of the layer and the measured parameters of the fill material are fed back to the earthwork allocation model. The remaining earthwork allocation scheme is further optimized with the spatial location of the stiffness weak zone as a constraint, so that the fill material with the gradation and moisture content meeting the preset requirements is preferentially allocated to the construction quality zone where the stiffness weak zone is located. S7. Geological Model Backwriting and Supplementary Investigation Trigger: The compaction response distribution of each filling layer identified in step S5 is backwritten to the three-dimensional distribution model of weak strata in the composite model, and the spatial distribution of weak strata is dynamically corrected; when multiple consecutive filling layers at the same plane location are identified as weak stiffness areas, it is determined that there are unexplored weak bodies at that location, triggering supplementary investigation at that location, and updating the composite model, re-dividing the construction quality zones and their foundation treatment levels according to the supplementary investigation results, and reconstructing the earthwork allocation plan accordingly. S8. Repeat steps S4 to S7 until the filling is completed. The site is then finely leveled, and the site elevation is re-measured using GPS real-time dynamic measurement technology. After passing the inspection, the as-built data is generated.

2. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: Step S1 specifically involves: using a drone equipped with a lidar scanner to perform a full-coverage scan of the construction site, acquiring site cloud data, establishing a three-dimensional terrain model, interpolating the geological survey data to create a three-dimensional distribution model of the soft strata, and overlaying it with the design surface model established according to the design elevation to form an overlay model.

3. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: Step S2 specifically involves: dividing the construction site into several construction quality zones based on the depth and thickness of the weak strata in the superimposed model, and determining the foundation treatment level for each construction quality zone; calculating the excavation and filling volumes for each construction quality zone based on the superimposed model, establishing an earthwork allocation model with the objective function of minimizing the sum of transportation costs and quality risk costs, and solving and outputting the initial earthwork allocation scheme.

4. The intelligent construction method for site leveling engineering according to claim 3, characterized in that: In step S2, the foundation treatment level is divided into three levels according to the burial depth and thickness of the weak strata: Level I is where there are no weak strata distributed within the construction quality zone; Level II is where there are weak strata with a burial depth greater than the preset burial depth value and a thickness less than the preset thickness value; and Level III is where the burial depth of the weak strata is not greater than the preset burial depth value or the thickness is not less than the preset thickness value.

5. The intelligent construction method for site leveling engineering according to claim 3, characterized in that: In step S2, the constraints of the earthwork allocation model include: the moisture content of each material source must be within the preset moisture content range; the daily allocation amount of each construction quality zone does not exceed the daily production capacity of the compaction equipment; and the working face time constraint, that is, before the construction quality zone has completed differentiated pretreatment and passed acceptance, the filling working face corresponding to the zone will not be arranged to transfer filler.

6. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: Step S3 specifically involves: excavating the excavation area layer by layer from top to bottom according to the initial earthwork allocation plan; implementing differentiated pretreatment for the filling area according to the foundation treatment level of each construction quality zone. The higher the foundation treatment level, the greater the dynamic compaction energy level or the greater the replacement thickness used in the pretreatment. Only construction quality zones that have completed pretreatment and passed acceptance can proceed to the filling process.

7. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: Step S4 specifically involves: filling soil in layers within the qualified construction quality zone; using an intelligent compaction monitoring device on the compaction machinery to collect vibration acceleration signals and rolling trajectories in real time during the compaction process; calculating compaction quality indicators that characterize the compaction state of the filling layer from the vibration acceleration signals; and binding and uploading the compaction quality indicators with the spatial coordinates of the rolling trajectories.

8. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: In step S5, the stiffness field distribution map is generated as follows: using the sampling points on the compaction trajectory as discrete points, the compaction quality index bound to each sampling point is mapped to a regular grid using inverse distance weighted interpolation or Kriging interpolation. The grid size is no larger than half the width of the compaction machine roller, thus obtaining the stiffness field distribution map.

9. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: In step S5, the rule for the joint judgment of the three indicators is as follows: the representative value of the compaction quality index is not lower than the lower limit of the representative value in the differentiated acceptance threshold group, the discrete value of the stiffness field uniformity is not greater than the upper limit of the discrete value in the differentiated acceptance threshold group, and the area ratio of the stiffness weak zone is not greater than the upper limit of the ratio in the differentiated acceptance threshold group. When all three indicators are met, the filling layer is judged to be qualified. Among them, the stiffness weak zone is the continuous grid area in the stiffness field distribution map where the compaction quality index is lower than the lower limit of the representative value.

10. The intelligent construction method for site leveling engineering according to claim 9, characterized in that: The rules for determining the value of the differentiated acceptance threshold group are as follows: the higher the foundation treatment level, the larger the lower limit of the corresponding compaction quality index representative value, and the smaller the upper limit of the uniformity dispersion value of the stiffness field and the upper limit of the area ratio of the stiffness weak zone.

11. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: In step S5, the graded treatment is carried out in the following order: Level 1, after supplementary compaction of the weak stiffness area, retesting is performed; Level 2, if the retesting after supplementary compaction is still unqualified, the fill material in the weak stiffness area is turned over or improved by admixtures and then re-compacted and retested; Level 3, if the retesting after Level 2 treatment is still unqualified, the weak stiffness area is partially replaced and retested, and at the same time, the replacement location and replacement material information are backfilled into the earthwork mixing model.

12. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: In step S5, the rules for interlayer dynamic correction are as follows: when the measured compaction quality index of the grid position corresponding to the next layer of filled material is lower than the lower limit of the representative value of the differential acceptance threshold group of the corresponding zone, the lower limit of the representative value of the grid position corresponding to the current filling layer is increased by a preset amount to prevent the formation of interlayer stiffness inversion where the upper layer stiffness is higher than the lower layer stiffness; when the measured compaction quality index of the grid position corresponding to the next layer of filled material is higher than the lower limit of the representative value to a preset margin, the target number of compaction passes of the current filling layer is reduced by a preset amount.

13. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: In step S6, the method for establishing and correcting the stiffness growth response model is as follows: using the sampling sequence of compaction quality index at the same grid location of each filled layer as the number of compaction passes increases as training data, a response curve characterizing the mapping relationship between compaction quality index and number of compaction passes and loose paving thickness is fitted; when constructing the next filled layer, the target number of compaction passes and target loose paving thickness required to reach the differentiated acceptance threshold group are extrapolated from the response curve; after the construction of this layer is completed, the model parameters of the response curve are updated with the deviation between the measured compaction quality index sequence and the predicted value of the response curve.

14. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: In step S7, the triggering condition for supplementary exploration is: the same plane location is identified as a weak stiffness zone in the stiffness field distribution map of N consecutive filling layers, where N is a preset integer not less than 2; after the supplementary exploration is completed, the spatial range of the weak body revealed by the exploration is written into the three-dimensional distribution model of the weak strata, and the construction quality zoning, foundation treatment level determination and earthwork allocation scheme solution in step S2 are re-executed.

15. The intelligent construction method for site leveling engineering according to claim 1, characterized in that: It also includes environmental protection measures during construction, including: setting up dust nets or water spraying devices at the boundaries of excavation and filling construction; setting up temporary drainage ditches around the construction area; and ensuring that the exhaust emissions of construction machinery meet the requirements of the national standard GB 20891-2014 "Emission Limits and Measurement Methods for Exhaust Pollutants from Diesel Engines for Non-Road Mobile Machinery (China Stage III and IV)".

16. An intelligent construction system for site leveling engineering, used to implement the intelligent construction method for site leveling engineering as described in any one of claims 1 to 15, characterized in that: include: The data acquisition module (10) includes a drone (1) equipped with a lidar scanner (2) and an inertial measurement unit, used to acquire three-dimensional point cloud data of the construction site; The 3D modeling and analysis module (20) is connected to the data acquisition module (10) to establish a superimposed model of point cloud data, geological survey data, and design elevation, divide the construction quality zones and determine the foundation treatment level, establish and solve the earthwork allocation model, and update the superimposed model according to the write-back data. The intelligent compaction monitoring module (30) is installed on the compaction machinery (60) and includes an acceleration sensor (31), a vibration frequency collector (32) and a wireless data transmission unit (33) for real-time acquisition of vibration acceleration signals and compaction trajectory and uploading. The construction control center (40) is connected to the three-dimensional modeling and analysis module (20) and the intelligent compaction monitoring module (30) respectively. It has built-in stiffness field inversion unit, joint judgment unit, feedforward prediction unit, feedback correction unit and model update unit: the stiffness field inversion unit is used to generate stiffness field distribution map by grid interpolation of compaction quality index; the joint judgment unit is used to call the differentiated acceptance threshold group according to the foundation treatment level of construction quality zone and perform inter-layer dynamic correction, and perform joint judgment of three indicators; the feedforward prediction unit is used to establish stiffness growth response model and predict the number of compaction passes and loose paving thickness of the next filling layer; the feedback correction unit is used to correct the model parameters of stiffness growth response model with the deviation between measured value and predicted value, and send the judgment result and stiffness field distribution map back to the three-dimensional modeling and analysis module (20) to trigger the earthwork allocation scheme to be re-optimized; the model update unit is used to write the compaction response distribution back to the three-dimensional distribution model of weak strata, and trigger supplementary investigation when multiple consecutive filling layers are judged as stiffness weak areas in the same plane position. GPS real-time dynamic measurement unit (50) is used for the re-measurement of the final elevation.

17. The intelligent construction system for site leveling engineering according to claim 16, characterized in that: The construction control center (40) is equipped with a BIM integration platform (41), which integrates and displays the composite model, construction quality zoning, stiffness field distribution map of each filling layer, stiffness growth response curve, joint judgment results and construction progress, so as to realize the visualization management of the entire construction process.

Images