An intelligent false bedrock identification system and method for deep foundation pit support engineering of backfilling mine pit

Abstract

The present application relates to the field of urban rail transit engineering, foundation pit support construction monitoring and intelligent sensing while drilling technology, and provides a false bedrock intelligent identification system and method for backfilling mine pit deep foundation pit support engineering, the method comprising data integration before construction, generating a three-dimensional risk layer, and high-frequency sampling; real-time calculation and correction of drilling specific energy during drilling, and maintaining the sliding window mean value and its coefficient of variation of the specific energy; determining whether to trigger a suspected bedrock instruction; dynamically determining the penetration verification depth and continuing drilling, and determining whether the specific energy drops; measuring the longitudinal wave velocity of the medium at the bottom of the hole, outputting the true bedrock, weathered zone or abnormal hard rock determination result; writing the identification result back to the three-dimensional risk layer together with the coordinates and time stamp, triggering feedback update. The present scheme can accurately identify the false bedrock boulders and true bedrock in the backfilling mine pit in real time, effectively avoid the risk of mis-embedding the support structure, and ensure the safety of deep foundation pit construction and optimize the subsequent pile position construction parameters.

Description

Technical Field

[0001] This invention relates to the fields of urban rail transit engineering, foundation pit support construction monitoring and drilling intelligent sensing technology, specifically to a false bedrock intelligent identification system and method for deep foundation pit support engineering in backfilled mine pits. Background Technology

[0002] As urban rail transit networks extend into existing industrial and mining land, an increasing number of subway stations are being planned on top of backfilled open-pit mines, quarries, or abandoned mining sites. The engineering geological conditions of such sites differ significantly from conventional strata, mainly in the following three aspects:

[0003] First, the backfill is highly heterogeneous. The backfill material mainly includes blasted rock, original soil, construction waste, and large-diameter original rock blocks (i.e., boulders). These boulders can range in size from tens of centimeters to several meters, and their spatial distribution is random. Because they were originally bedrock blocks stripped by blasting, their lithology, mineral composition, hardness, and longitudinal wave velocity are almost identical to the underlying true bedrock, making them difficult to distinguish using conventional lithological comparison methods. Second, the top surface of the true bedrock is irregular. The bedrock residue from the original open-pit mining usually exhibits dramatic undulations, with localized steps, grooves, and mining pits. After backfilling, the interface between the bedrock surface and the backfill exhibits strong three-dimensional variability, which is difficult to accurately depict using traditional "one-hole-one-column diagrams." Third, the retaining structure is sensitive to the embedment depth. Deep foundation pits are generally 18-30m deep, and their retaining structures (diaphragm walls, bored piles, or interlocking piles) must be embedded in the true bedrock to a certain depth (usually not less than 2-3m) to provide sufficient passive resistance. If the pile bottom or wall toe is mistakenly placed on a boulder in the backfill, after the foundation pit is excavated and unloaded, the shear strength of the boulder and the surrounding weak backfill material will be far lower than the design assumption. The retaining structure may suffer kick-out failure, slide into the foundation pit as a whole, or even cause roof collapse, water inrush and settlement of surrounding buildings and structures. This is a high-level disaster mode in subway foundation pit accidents.

[0004] To address the above problems, existing technologies have the following limitations: (1) Traditional exploration density is insufficient. The "one pile, one hole" or quincunx drilling method arranged according to specifications is difficult to capture randomly distributed boulders, resulting in a high rate of missed detection. If the density is increased to cover all boulders, the exploration cost will increase exponentially and the construction period will be unacceptable. (2) Geophysical methods have limited reliability when used alone. Transient electromagnetic, ground-penetrating radar and cross-hole resistivity imaging can identify large high-resistivity anomalies, but their spatial resolution and depth inversion accuracy decrease significantly under strong interference from backfill. Furthermore, geophysical methods themselves cannot provide a definitive judgment on whether a boulder is a boulder. (3) Measurement while drilling (MWD) systems only record data and do not make decisions. Although existing MWD systems and construction monitoring platforms can collect parameters such as torque, footage, and feed pressure, their output is limited to post-construction analysis and report generation. They lack a closed-loop control logic that actively falsifies the key question of whether the current hard layer is a boulder or true bedrock during construction. (4) Experience judgment is highly subjective. On-site construction personnel rely on experience indicators such as “slowing down the drilling speed, drilling rig jumping, and rock cutting characteristics” to judge the rock entry. This is almost ineffective in the backfilling pit scenario where the lithology of the isolated rock and the true bedrock are completely consistent, and the human error rate is high.

[0005] In summary, existing technologies struggle to reliably determine the bedrock penetration status of deep foundation pit retaining piles in backfilled mining environments while balancing construction schedule and cost. Therefore, a real-time, automated, and falsifiable method for identifying false bedrock is urgently needed during construction. Summary of the Invention

[0006] To address the problems existing in the background technology, this invention proposes an intelligent identification system and method for false bedrock in deep foundation pit retaining engineering of backfilled mine pits. It can identify false bedrock boulders and real bedrock in backfilled mine pits in real time and accurately, effectively avoid the risk of mis-embedded retaining structures, ensure the safety of deep foundation pit construction, and optimize subsequent pile position construction parameters.

[0007] To achieve the above objectives, the present invention adopts the following solution: A method for intelligent identification of false bedrock in deep foundation pit retaining engineering of backfilled mine pits includes the following steps: Step 1: Before construction, integrate the foundation pit retaining design documents with the site's 3D geophysical inversion data to generate a 3D risk layer with boulder risk attributes; when the distance between the drilling rig position and the center of any suspected boulder is less than a preset threshold, enter the high-frequency sampling state. Step 2: Calculate and correct the drilling specific energy in real time during the drilling process, and maintain the mean value of the sliding window of the specific energy and its coefficient of variation. Step 3: Based on the average specific energy and coefficient of variation within the current window, determine whether a suspected bedrock command has been triggered; if a suspected bedrock command has been triggered, proceed to step 4; otherwise, continue normal drilling. Step 4: Lock the current feed pressure and torque upper limit, dynamically determine the penetration verification depth and continue drilling; monitor whether the specific energy drops within the verification depth range. If it drops, it is determined to be a false bedrock boulder and proceeds to step 6; otherwise, proceed to step 5. Step 5: Conduct acoustic wave testing at the bottom of the borehole to measure the longitudinal wave velocity of the medium at the bottom of the borehole and compare it with the pre-calibrated regional bedrock longitudinal wave velocity benchmark value to output the result of determining whether it is true bedrock, weathered zone or abnormal hard rock. Step 6: Write the identification results from Steps 4 and 5, along with the coordinates and timestamps, back to the 3D risk layer and trigger feedback updates for subsequent drilling parameters.

[0008] Optionally, in step 2, the corrected drilling specific energy e Calculated based on the following formula: e =(1 / η)·[ F / A +(2π· n · T ) / ( A · v )] E loss ;; in, A η represents the effective cross-sectional area of ​​the drill bit; η∈(0,1] is the drill bit wear-efficiency correction coefficient. E loss This refers to the energy loss per unit volume caused by the circulation of the flushing fluid. T It is the rotational torque; v This refers to the instantaneous advance rate; n This represents the drill bit rotation speed.

[0009] Optionally, in step 3, the condition for triggering the suspected bedrock command is that the following conditions are met simultaneously: Firstly, in the recent continuous advance Δ h Within the 0.5m range, corrected drilling specific energy e Time average ē ≥ e th ,in, e th Based on the pre-calibrated strength threshold of moderately or slightly weathered bedrock according to the regional lithology database, and e th The uniaxial compressive strength of the rock meets the requirements e th = α · σ c ,in, α This is the conversion factor for the drilling specific energy threshold, and its value ranges from 0.8 to 1.2. σ c It represents the uniaxial compressive strength of the rock. Secondly, within the same range, correct the drilling specific energy. e coefficient of variation C V= σ e / ē ≤0.20.

[0010] Optionally, the penetration verification depth Ltest is dynamically determined by the following formula: L test =max{1.2· D est, h socket +Δs, L min}; Wherein, Dest is the equivalent particle size of the suspected boulder closest to the current pile location in the three-dimensional risk layer; when there are no prior boulder records around the current pile location, Dest is set to 0; h socket Δs represents the required embedment depth of the retaining structure in bedrock, a value obtained from the foundation pit retaining design documents; Δs is the safety margin, ranging from 0.5 to 1.0 m; L min The minimum mandatory verification depth is 2.0m.

[0011] Optionally, in step 4, the conditions for determining whether a rock is a false bedrock boulder are: Within any sub-interval of the penetration verification depth range, the corrected drilling specific energy e The drop amplitude satisfies Δ e = e peak e min ≥0.5 e th ; and accompanied by feed pressure F instantaneous drop or advance speed v The step increase; at this time, the top surface coordinates, thickness and equivalent particle size of the boulder are recorded and written back to the three-dimensional risk layer.

[0012] Optionally, in step 5, the specific method for comparing with the pre-calibrated regional bedrock P-wave velocity reference value is as follows: calculate the ratio of the measured P-wave velocity to the reference value. K = V p / V std When 0.9≤ K When the density is ≤1.15, it is considered true bedrock. KWhen K < 0.9, it is determined to be a weathered zone or a weak interlayer, and drilling continues. When K > 1.15, it indicates abnormally hard rock and triggers secondary penetration verification. The benchmark value... V std It is determined by the results of acoustic logging or indoor ultrasonic testing of rock samples during the site exploration phase.

[0013] Optionally, in step 6, the identification results specifically include: the spatial distribution of boulders and the bedrock surface elevation point cloud obtained after steps 4 and 5; the spatial distribution of boulders and the bedrock surface elevation point cloud, along with their coordinates and timestamps, are written back to the three-dimensional risk layer to form a dynamically updated three-dimensional geological-boulder joint digital model, which is used for subsequent pile locations within the same foundation pit. e th , Dest and L test Recalibrate.

[0014] Optionally, the method is applied to the construction of retaining structures for deep foundation pits built on backfilled mine pits, backfilled open-pit mines, or backfilled quarries, wherein the retaining structure is one or more of underground continuous walls, bored piles, or interlocking piles; in step 4, if the current pile position is located within the backfill body or bedrock transition zone marked by the three-dimensional risk layer, the safety margin Δs is taken as the upper limit of 1.0m.

[0015] A false bedrock intelligent identification system for deep foundation pit retaining engineering in backfilled mine pits, the system comprising a drilling multi-parameter sensing subsystem, a geological prior data integration module, an edge computing decision unit, a penetration verification execution unit, an acoustic final judgment unit, and a feedback update module, all connected in sequence; wherein: The multi-parameter sensing subsystem for drilling includes torque sensors, speed encoders, displacement sensors, feed pressure sensors, and flushing fluid inlet / outlet pressure or flow sensors deployed on the drill rig's power head, drill pipe, and drill string, respectively. These sensors are used to synchronously acquire the rotational torque T and drill bit speed during the drilling process at a sampling frequency of not less than 50Hz. n Instantaneous advance rate v Pressure input F and flushing hydraulic differential Δ P f Time series data; The geological prior data integration module stores three-dimensional geophysical data of the construction area obtained by combining one or more methods, such as transient electromagnetic method, ground-penetrating radar or trans-hole resistivity imaging, and constructs a three-dimensional risk layer with the center coordinates of suspected boulders, estimated equivalent grain size and confidence level of high resistivity anomaly as attributes. The edge computing decision unit is connected to the drilling multi-parameter sensing subsystem via a bus, and has a built-in drilling specific energy calculation engine and formation attribute discriminator, used to convert in-situ mechanical characteristics into formation anti-fracture strength indicators. e The system combines the three-dimensional risk layer to output three types of instructions: suspected bedrock, suspected boulder, or continue drilling. The penetration verification execution unit is used to automatically calculate and issue a penetration verification depth to the drilling rig when the edge computing decision unit outputs a suspected bedrock command. L test Control the drilling rig to continue drilling downwards under the current power parameters. L test And the energy ratio within this range e The magnitude of the drop Δ e Conduct closed-loop monitoring; The acoustic wave final determination unit includes an acoustic transducer installed at the bottom of the drill pipe or lowered through the inner cavity of the drill pipe. After the penetration verification execution unit determines that the specific energy is stable, it uses the pulse transmission method to measure the longitudinal wave velocity of the medium at the bottom of the borehole. V p and compared with the pre-calibrated regional bedrock longitudinal wave velocity reference value. V std The comparison is performed, and the final result of determining whether the bedrock is real or fake is output. The feedback update module is used to write back the spatial distribution of boulders and the bedrock surface elevation point cloud obtained after penetration verification and acoustic final judgment to the geological prior data integration module, forming a dynamically updated three-dimensional geological-boulder joint digital model as construction progresses, which is used for subsequent pile locations within the same foundation pit. e th , Dest and L test Recalibrate.

[0016] Optionally, after determining that it is a false bedrock boulder, the edge calculation decision unit further adjusts the torque limiter and feed pressure limit of the drilling rig in real time based on the recorded boulder thickness and position. When the drill bit is passing through the interface between the boulder and the backfill, the torque limit is reduced to 70% to 85% of the rated value to suppress hole deviation and drill jamming caused by lateral impact.

[0017] The beneficial effects of this invention are as follows: First, this solution proposes a four-level serial closed-loop identification architecture of geophysical prior, drilling energy perception, forced penetration falsification, and acoustic wave velocity final judgment, which significantly reduces the rock entry misjudgment rate. Moreover, each misjudgment can be actively discovered and corrected by subsequent penetration verification, thus significantly improving the identification reliability.

[0018] Furthermore, this solution will verify depth L test Explicitly bound to the geometric valuation of the boulderD est, rock embedment depth of retaining structure design h socket With a minimum safety margin Δs, the actual rock embedment depth of each retaining pile or each diaphragm wall section is not less than the design value, thus eliminating the risk of "suspended piles" and the resulting damage to the foundation pit excavation kick-off from the source.

[0019] In addition, compared with the conservative approach of "deeply penetrating all suspected hard layers", this solution only triggers penetration verification at the location where the discrimination condition is met, which can reduce invalid penetration footage by an average of 30%-50%, significantly reduce drill bit wear and hole-forming time, and achieve a balance between construction efficiency and cost.

[0020] Furthermore, this scheme not only performs identification at individual pile locations, but also writes the identification results of each pile location back to the site's three-dimensional risk layer, enabling subsequent pile locations in the same foundation pit to be identified. e th Dest and L test The calibration can be continuously refined as construction progresses, significantly reducing the probability of misjudging pile positions in later stages. At the same time, the resulting three-dimensional geological-isolated boulder joint model can directly support the exploration and design optimization of subsequent sections of the same project and surrounding subway stations and tunnel sections. Attached Figure Description

[0021] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the overall architecture and sensor distribution of the system in an embodiment of the present invention; Figure 3 This is a schematic diagram comparing the vertical spatial characteristics of isolated boulders and true bedrock in a backfilled mine pit scenario according to an embodiment of the present invention, wherein (a) is... Figure 3 (a) is a bedrock-isolated rock condition; (b) is Figure 3 Zhongzhen bedrock working conditions; Figure 4 Correcting drilling specific energy in embodiments of the present invention e With depth h Typical characteristic curves of change; Figure 5 This is a sub-flowchart of the forced penetration verification step in an embodiment of the present invention; Figure 6 This is a sub-flowchart of the data write-back and dynamic update steps in an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the construction progress and model accuracy evolution in an embodiment of the present invention; Figure 8 This is a schematic diagram showing the superposition of the three-dimensional geophysical risk layer and the pile location planar layout in an embodiment of the present invention. Detailed Implementation

[0022] To make the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the given embodiments are merely one implementation method and do not represent all embodiments.

[0023] Example 1 Combination Figure 1 The flowchart shown illustrates an embodiment of the present invention that provides an intelligent identification method for false bedrock in deep foundation pit retaining engineering for backfilled mine pits. This method aims to solve the problems of misjudgment of the bearing layer and risk of retaining structure failure caused by the consistency of lithology between isolated boulders and true bedrock, severe undulation of the bedrock surface, and the lack of active falsification capability in existing monitoring systems. The steps include the following: Step 1, Multi-source Data Integration and Early Warning Zoning: Before construction, the foundation pit retaining design documents are integrated with the site's 3D geophysical inversion data to generate a 3D risk layer with boulder risk attributes. When the distance between the drilling rig location and the center of any suspected boulder is less than a preset threshold, high-frequency sampling is initiated. In this example, the preset threshold is 0.5m.

[0024] Step 2, Real-time calculation of drilling specific energy: Real-time calculation and correction of drilling specific energy during drilling. e And maintain the sliding window mean of this specific energy. ē and its coefficient of variation C V. Wherein, the modified drilling specific energy e Calculated based on the following formula: e =(1 / η)·[ F / A +(2π· n · T ) / ( A · v )] E loss ; In the formula, A η represents the effective cross-sectional area of ​​the drill bit; η∈(0,1] is the drill bit wear-efficiency correction coefficient. E loss This refers to the energy loss per unit volume caused by the circulation of the flushing fluid. T It is the rotational torque; v This refers to the instantaneous advance rate; n The drill bit rotation speed is given by the following formula: E loss = k f ·Δ P f · Q / ( A · v ); In the formula, Q This refers to the flushing fluid flow rate. k f The dissipation coefficient is related to the geometry of the annulus inside the hole; e , F / A and E loss All are expressed in megapascals (MPa) or equivalent megajoules per cubic meter (MJ / m³). 3 (in units of )

[0025] In this embodiment, the drill bit wear-efficiency coefficient η and the flushing fluid energy dissipation term E are introduced. loss This is to eliminate the interference of drill string deterioration and mud circulation on formation strength indicators, thereby improving the corrected drilling specific energy. e The values ​​can remain comparable during long-term construction, thus allowing for the setting of stable bedrock identification thresholds. e th 。

[0026] Step 3, Preliminary assessment of suspected bedrock: Based on the average specific energy and coefficient of variation within the current window, determine whether a suspected bedrock command has been triggered; if a suspected bedrock command has been triggered, proceed to Step 4; if no suspected bedrock command has been triggered, maintain normal drilling operations. The conditions for triggering a suspected bedrock command must simultaneously meet the following two conditions, combined with... Figure 4 As shown, it specifically includes: Firstly, in the recent continuous advance Δ h Within the 0.5m range, corrected drilling specific energy e Time average ē ≥ e th ,in, e th Based on the pre-calibrated strength threshold of moderately or slightly weathered bedrock according to the regional lithology database, and e th The uniaxial compressive strength of the rock meets the requirements e th = α · σ c, where α This is the conversion factor for the drilling specific energy threshold, and its value ranges from 0.8 to 1.2. σ c It represents the uniaxial compressive strength of the rock.

[0027] Secondly, within the same range, correct the drilling specific energy. e coefficient of variation C V= σ e / ē ≤0.20.

[0028] Step 4, forced penetration verification: Lock the current feed pressure and torque upper limit, dynamically determine the penetration verification depth and continue drilling; Monitor whether the specific energy drops within the verification depth range. If it drops, it is determined as a false bedrock boulder and go to Step 6, otherwise go to Step 5. The process is as Figure 5 shown.

[0029] Among them, the penetration verification depth L test is dynamically determined by the following formula: L test = max{1.2· D est, h socket + Δ s , L min}; In the formula, Dest is the equivalent particle size of the suspected boulder closest to the current pile position in the three-dimensional risk layer. When there is no prior boulder record around the current pile position, Dest takes the value of 0; h socket is the depth of the retaining structure embedded in the bedrock required by the design, and its value is obtained by reading the foundation pit retaining design document; Δ s is the safety margin, and its value range is 0.5 - 1.0 m ；L min is the minimum forced verification depth, and its value is not less than 2.0 m. By explicitly associating L test with the designed bedrock embedding depth h socket , it can ensure that the actual rock embedding depth of each retaining pile or each diaphragm wall trench section is not less than the design value, and eliminate the risk of "hanging piles" and the resulting kick - out failure during foundation pit excavation from the source.

[0030] [[ID=The condition for determining a false bedrock boulder is that in any sub - interval within the penetration verification depth range, the drop amplitude of the corrected drilling specific energy e satisfies Δ e = e peak e min ≥ 0.5· e th ; Among them, e peak is the peak value of the corrected drilling specific energy, e min is the minimum value of the corrected drilling specific energy; and accompanied by an instantaneous drop in the feed pressure F or the penetration ratev Step - up; at this time, record the top - surface coordinates, thickness, and equivalent particle size of the boulder and write them back to the three - dimensional risk layer.

[0031] Different from the penetration judgment of fixed values or empirical values in the prior art, in this embodiment, the penetration verification depth L test is explicitly bound to the boulder geometric estimation Dest , the designed rock - socketed depth of the retaining structure h socket and the minimum safety margin Δ s . As Figure 3 shown, the boulders in the backfilled pit have natural truncation characteristics in the vertical dimension, and their lower part must be backfill or void medium; while the real bedrock has continuity characteristics in the vertical dimension. Based on the above geological characteristic differences, after the forced penetration L test depth, monitor whether there is a significant drop in the corrected drilling specific energy to achieve the discrimination of boulders. The distinction between boulders and bedrock is converted from the judgment of lithological attributes to the active falsification of spatial continuity, realizing the essential improvement of the existing empirical discrimination logic.

[0032] Step 5, acoustic wave final judgment: Conduct acoustic wave testing at the bottom of the hole, measure the longitudinal wave velocity of the bottom - hole medium, and compare it with the pre - calibrated longitudinal wave velocity reference value of the regional bedrock to output the judgment results of true bedrock, weathered zone, or abnormally hard rock.

[0033] This step includes: calculating the ratio of the measured longitudinal wave velocity to the reference value K = V p / V std ; when 0.9 ≤ K ≤ 1.15, it is judged as true bedrock; when K < 0.9, it is judged as weathered zone or soft interlayer and continue drilling is carried out; when K > 1.15, it is prompted as abnormally hard rock and secondary penetration verification is triggered; the reference value V std is determined by the acoustic wave logging in the field exploration stage or the ultrasonic test results of indoor rock samples.

[0034] Step 6, data writing back and dynamic update: Write the boulder spatial distribution and bedrock surface elevation point cloud obtained through the penetration verification in step 4 and the acoustic wave final judgment in step 5, together with coordinates and timestamps, back to the three - dimensional risk layer, forming a three - dimensional geological - boulder joint digital model that is dynamically updated with the progress of construction, and used for e th , Dest and L test of the subsequent pile positions in the same foundation pit to be recalibrated. As Figure 6This step not only performs identification at a single pile position, but also writes back the identification results of each pile position to the three-dimensional risk layer of the construction area, enabling the e th 、 Dest 与 L test calibration to be continuously refined as the construction progresses, significantly reducing the probability of misjudgment of subsequent pile positions, as shown in the Figure 7 schematic diagram of the construction progress and model accuracy evolution shown.

[0035] Example Two Combined with the Figure 2 system framework diagram shown, this example provides an intelligent identification system for false bedrock in deep foundation pit retaining engineering for backfilled mine pits, used to implement the method of Example One. The system includes a downhole multi-parameter perception subsystem, a geological prior data integration module, an edge computing decision unit, a penetration verification execution unit, an acoustic wave final judgment unit, and a feedback update module that are sequentially communicatively connected.

[0036] Specifically, the downhole multi-parameter perception subsystem includes torque sensors, rotational speed encoders, displacement sensors, feed pressure sensors, and flushing fluid inlet and outlet pressure or flow sensors respectively deployed on the drill rig power head, drill pipe, and drill tool, used to synchronously obtain the rotational torque T, drill bit rotational speed n 、instantaneous penetration rate v 、feed pressure F 及冲洗液压差Δ P f time series data at a sampling frequency not lower than 50Hz; The geological prior data integration module stores three-dimensional geophysical exploration data obtained by joint inversion of any one or more of transient electromagnetic method, ground penetrating radar, or cross-hole resistivity imaging in the construction area, and constructs a three-dimensional risk layer with the suspected boulder center coordinates ( x i ,y i ,z i ), estimated equivalent particle size Dest,i 及高阻异常置信度 p i as attributes.

[0037] The edge computing decision unit is connected to the downhole multi-parameter perception subsystem through a bus, and is built-in with a drilling specific energy calculation engine and a formation property discriminator, used to convert in-situ mechanical characteristic quantities into formation anti-fracture strength indexes eand output three types of instructions, namely suspected bedrock, suspected boulder or continue drilling, in combination with the three-dimensional risk layer. After determining false bedrock boulders, the edge computing decision unit further adjusts the torque limiter and the upper limit of the feed pressure of the drill rig in real time based on the recorded boulder thickness and position. When the drill bit is passing through the interface between the boulder and the backfill, the upper torque limit is reduced to 70% - 85% of the rated value to inhibit hole deviation and sticking caused by lateral impact.

[0038] The penetration verification execution unit is used to automatically calculate and issue the penetration verification depth to the drill rig when the edge computing decision unit outputs a suspected bedrock instruction L test to control the drill rig to continue drilling downward under the current power parameters L test and perform closed-loop monitoring on the drop amplitude Δ e of the specific energy e in this interval.

[0039] The acoustic wave final judgment unit includes an acoustic wave transducer installed at the bottom of the drill pipe or lowered through the inner cavity of the drill pipe, which is used to actually measure the longitudinal wave velocity of the bottom hole medium by the pulse transmission method after the penetration verification execution unit determines that the specific energy is stable V p and compare it with the pre-calibrated regional bedrock longitudinal wave velocity reference value V std to output the final true or false bedrock judgment result.

[0040] The feedback update module is used to write back the boulder spatial distribution and bedrock surface elevation point cloud obtained after penetration verification and acoustic wave final judgment to the geological prior data integration module, forming a three-dimensional geological-boulder joint digital model that is dynamically updated with the progress of construction, and is used for e th , Dest and L test of subsequent pile positions in the same foundation pit for recalibration.

[0041] It should be noted that the acquisition method of the three-dimensional geophysical exploration data is not limited to transient electromagnetic method, ground penetrating radar or cross-hole resistivity imaging. Any geophysical exploration or comprehensive geophysical exploration means that can invert the spatial distribution of high-resistance anomalies can be used as the prior data source of the present invention; the acoustic wave transducer can adopt a single-shot single-receive, single-shot double-receive or array structure; the edge computing decision unit can be implemented in the form of an industrial control computer, an embedded controller or an industrial gateway with local computing capabilities. The above alternative methods all fall within the protection scope of the present invention.

[0042] [[ID=**39**]]Simulation example: To verify the effectiveness and accuracy of an intelligent identification method and system for false bedrock in deep foundation pit enclosure engineering for backfilled mine pits proposed in this invention, this embodiment combines an example of a deep foundation pit enclosure project for a city built on a backfilled open-pit mine, and details the specific implementation process and verification results of this method.

[0043] This project is a planned subway station, which is located above the backfill area of a certain abandoned open-pit quarry. The main body of the station is 220m long, 22m wide, and the foundation pit is 24m deep. A retaining structure of 1000mm thick bored cast-in-place piles plus an internal support system is adopted. The design requirement is that each retaining pile should be embedded in moderately and slightly weathered granite for no less than 2.5m. The field investigation data reveals that the thickness of the backfill body ranges from 10 to 22m, there are a large number of boulders in the backfill body, and the undulation height difference of the bedrock surface of the original quarry reaches 6m. The standard value of the uniaxial compressive strength of granite σ c ≈120MPa, and the indoor ultrasonic measurement gives V std = 5200m / s.

[0044] In the simulation, the typical pile position, pile No. A23, is selected as the verification object. Before construction, the three-dimensional geophysical exploration data jointly inverted by transient electromagnetic method and ground penetrating radar in the field area and the deep foundation pit enclosure design documents (including the design pile bottom elevation and rock-socketed depth requirements for each pile h socket =2.5m) are imported into the edge computing decision unit to generate a three-dimensional risk layer with boulder risk attributes. The layer marks that there are several high-resistance anomalies in the depth range of 12 - 22m. The equivalent diameter of the suspected boulder closest to the pile position of pile No. A23 D est≈1.2m, and the confidence level p = 0.7.

[0045] According to the process of the method described in this invention, the following steps are implemented in sequence: Step B1, data integration in the pre-construction stage: The system completes data import and three-dimensional risk layer construction before construction. When the drill rig is located at the pile position of pile No. A23 and the horizontal distance between its three-dimensional coordinates and the center of the suspected boulder is less than 0.5m, the system automatically enters the high-frequency sampling state.

[0046] Step B2, real-time calculation of drilling specific energy: The rotary drilling rig starts the hole-forming operation of pile No. A23. When drilling to a depth of 20.5m, the following parameter jumps are monitored by the multi-parameter perception subsystem during drilling: torque T jumps from 80kN·m to 220kN·m, the instantaneous penetration rate v drops from 1.5m / h to 0.2m / h, and the feed pressure F rises from 600kN to 1100kN.

[0047] Calculate the specific energy at the current moment according to the modified penetration specific energy formula: Take the effective cross-sectional area of the drill bit A = 0.785 m 2 (drill bit with a diameter of 1.0 m), the bit wear - efficiency correction factor η = 0.85 (the cumulative footage of this drill bit is 1800 m, check the life curve), and the energy loss of the flushing fluid circulation E loss = 3 MJ / m 3 , and calculate the modified penetration specific energy at the current moment e ≈ 182 MJ / m 3 .

[0048] Step S3, preliminary judgment of suspected bedrock: The threshold of medium / weakly weathered bedrock strength pre-calibrated in this project e th = α · σ c = 1.0 × 120 = 120 MJ / m³; The current specific energy is 182 MJ / m 3 far exceeds this threshold. In the interval of the most recent continuous footage Δ h = 0.5 m, the time average value of the modified penetration specific energy e is ē = 178 MJ / m³, C V = 0.14 ≤ 0.20. Both conditions meet the trigger conditions. The system outputs a suspected bedrock command and enters the forced penetration verification stage.

[0049] Step B4, forced penetration verification: Calculate the penetration verification depth according to the formula: L test = max{1.2 · D est, h socket + Δs, L min}; That is, 1.2 · D est = 1.44 m; h socket + Δs = 2.5 + 1.0 = 3.5 m (this pile position is in the transition zone between the backfill and the bedrock, and Δs takes the upper limit); L min = 2.0 m; Therefore, the maximum value of the three, L test = 3.5 m. The system locks the current power parameters and commands the drill rig to continue drilling down 3.5 m.

[0050] When it is Case A (false bedrock - boulder): That is, if in any sub-interval within the verification depth range, the specific energy e drops (that is, Δe = e p eak e min ≥0.5· e th ) and accompanied by the instantaneous unloading of the feeding pressure or the stepwise increase of the penetration rate, it is determined that the current interface is a false bedrock boulder. In this embodiment, when the drill rig penetrates to 21.8 m, the special energy e suddenly drops from 185 MJ / m 3 to 24 MJ / m 3 , Δ e = 161 MJ / m 3 ≥ 0.5· e th = 60 MJ / m 3 , and at the same time, the feeding pressure F shows instantaneous unloading, and the penetration rate v stepwise increases to 2.1 m / h, meeting the above criteria for false bedrock boulders. The system records the top surface coordinates, thickness, and equivalent particle size of the boulder and writes them back to the three-dimensional risk layer. And it instructs the drill rig to continue to search downward for the true bedrock bearing layer, and at the same time temporarily reduces the torque upper limit to 80% of the rated value to prevent deviation or sticking of the drill bit caused by sudden unloading when passing through the bottom interface of the boulder.

[0051] When it is case B (true bedrock): That is, if the drill rig completes L test = 3.5 m of penetration (the depth reaches 24.0 m), and the special energy e always remains stable within the range of 175 ± 12 MJ / m 3 during the entire verification depth range without dropping, then the system determines that the current interface has the characteristics of true bedrock and transfers to the acoustic wave final judgment.

[0052] Step B5, acoustic wave final judgment: The edge computing decision unit instructs the drill rig to stop advancing, lowers the acoustic wave transducer to the bottom of the hole through the inner cavity of the drill pipe, and measures the actual V p = v = 4850 m / s by the pulse transmission method. Calculate K = 4850 / 5200 = 0.933, which satisfies 0.9 ≤ K ≤ 1.15, and it is determined as the true bedrock bearing layer. The system confirms that the actual rock socket depth of this pile ≥ 2.5 m, meeting the design requirements, and allows subsequent processes such as hole cleaning, placing the steel reinforcement cage, and pouring concrete.

[0053] Step B6, feedback update: Regardless of whether it is Case A or Case B, the system will write back the identification results of Pile A23 (the spatial distribution of boulders and the measured elevation of the true bedrock surface) to the 3D risk layer through the feedback update module, forming a 3D geological-boulder joint digital model that is dynamically updated as the construction progresses. As Figure 8 shown, for adjacent Pile Nos. A24 and A25, the system recalculates D est and L test , making the penetration verification of subsequent pile positions more targeted.

[0054] In this embodiment, through the simulated drilling data of Pile A23, the functions of the method and system of the present invention are fully verified: (1) The recognition reliability is significantly improved: In the engineering scenario of a typical backfilled mine pit subway station, through four-level closed-loop identification of geophysical prior experience, real-time specific energy during drilling, forced penetration falsification, and final judgment of acoustic wave velocity, the misjudgment rate of rock penetration can be reduced from the level of 15%-25% in manual experience judgment to less than 5%, and each misjudgment can be actively discovered and corrected by subsequent penetration verification.

[0055] (2) The safety margin of the retaining structure is controllable: By explicitly correlating L test with the designed rock socket depth h socket , it can be ensured that the actual rock socket depth of each retaining pile or each diaphragm wall trench section is not less than the designed value, eliminating the risk of "hanging piles" and the resulting kick-out failure during foundation pit excavation at the source.

[0056] (3) Construction efficiency and cost are balanced: Compared with the conservative approach of "penetrating all suspected hard layers to a certain depth", the present invention only triggers penetration verification at positions where the discrimination conditions are met, which can reduce the ineffective penetration footage by 30%-50% on average, significantly reducing bit wear and hole formation time.

[0057] (4) Data assets can be reused: The 3D geological-boulder joint model formed by the feedback update module can directly support the optimization of the exploration and design of subsequent sections of the same project and surrounding subway stations and tunnel sections.

[0058] The specific embodiments of the present invention have been described in detail above in conjunction with the drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, without departing from the principles and spirit of the present invention, various changes, modifications, substitutions, and variations to these embodiments still fall within the protection scope of the present invention.

Claims

1. A method for intelligent identification of false bedrock in deep foundation pit retaining engineering for backfilled mine pits, characterized in that, Includes the following steps: Step 1: Before construction, integrate the foundation pit retaining design documents with the site's 3D geophysical inversion data to generate a 3D risk layer with boulder risk attributes; when the distance between the drilling rig position and the center of any suspected boulder is less than a preset threshold, enter the high-frequency sampling state. Step 2: Calculate and correct the drilling specific energy in real time during the drilling process, and maintain the sliding window mean and its coefficient of variation of the specific energy; Step 3: Based on the average specific energy and coefficient of variation within the current window, determine whether a suspected bedrock command has been triggered; if a suspected bedrock command has been triggered, proceed to step 4; otherwise, continue normal drilling. Step 4: Lock the current feed pressure and torque upper limit, dynamically determine the penetration verification depth and continue drilling; monitor whether the specific energy drops within the verification depth range. If it drops, it is determined to be a false bedrock boulder and proceeds to step 6; otherwise, proceed to step 5. Step 5: Conduct acoustic wave testing at the bottom of the borehole to measure the longitudinal wave velocity of the medium at the bottom of the borehole and compare it with the pre-calibrated regional bedrock longitudinal wave velocity benchmark value to output the result of determining whether it is true bedrock, weathered zone or abnormal hard rock. Step 6: Write the identification results from Steps 4 and 5, along with the coordinates and timestamps, back to the 3D risk layer and trigger feedback updates for subsequent drilling parameters.

2. The method for intelligent identification of false bedrock in deep foundation pit retaining engineering for backfilled mine pits according to claim 1, characterized in that: In step 2, the corrected drilling specific energy e Calculated based on the following formula: e =(1 / n)·[ F / A +(2π· n · T ) / ( A · v )] E loss ; in, A η represents the effective cross-sectional area of ​​the drill bit; η∈(0,1] is the drill bit wear-efficiency correction coefficient. E loss This refers to the energy loss per unit volume caused by the circulation of the flushing fluid. T It is the rotational torque; v This refers to the instantaneous advance rate; n This represents the drill bit rotation speed.

3. The method for intelligent identification of false bedrock in deep foundation pit retaining engineering for backfilled mine pits according to claim 2, characterized in that: In step 3, the conditions for triggering a suspected bedrock command are that the following conditions are met simultaneously: Firstly, in the recent continuous advance Δ h Within the 0.5m range, corrected drilling specific energy e Time average ē ≥ e th ,in, e th Based on the pre-calibrated strength threshold of moderately or slightly weathered bedrock according to the regional lithology database, and e th The uniaxial compressive strength of the rock meets e th =α·σ c ,in, α This is the conversion factor for the drilling specific energy threshold, and its value ranges from 0.8 to 1.

2. σ c It represents the uniaxial compressive strength of the rock. Secondly, within the same range, correct the drilling specific energy. e coefficient of variation C V= σ e / ē ≤0.

20.

4. The method for intelligent identification of false bedrock in deep foundation pit retaining engineering for backfilled mine pits according to claim 1, characterized in that: In step 4, the penetration verification depth is determined. L test Determined dynamically according to the following formula: L test =max{1.2· D East, h socket +Δ s , L min }; in, Dest The equivalent particle size of the suspected boulder closest to the current pile location in the three-dimensional risk layer is used when there are no prior boulder records around the current pile location. Dest The value is 0; h socket The depth of the retaining structure embedded in the bedrock as required by the design is obtained from the foundation pit retaining design documents; Δ s For safety margin, its value ranges from 0.5 to 1.0 m; L min The minimum mandatory verification depth is 2.0m.

5. The method for intelligent identification of false bedrock in deep foundation pit retaining engineering for backfilled mine pits according to claim 4, characterized in that: In step 4, the conditions for determining whether a rock is a false bedrock are as follows: Within any sub-interval of the penetration verification depth range, the corrected drilling specific energy e The drop amplitude satisfies Δ e = e peak e min ≥0.5 e th ,in, e peak To correct the peak value of drilling specific energy, e min Correct the minimum drilling specific energy; and accompany the feed pressure. F instantaneous drop or advance speed v The step increase; at this time, the top surface coordinates, thickness and equivalent particle size of the boulder are recorded and written back to the three-dimensional risk layer.

6. The method for intelligent identification of false bedrock in deep foundation pit retaining engineering for backfilled mine pits according to claim 1, characterized in that: In step 5, the specific method for comparing the measured P-wave velocity with the pre-calibrated regional bedrock P-wave velocity benchmark value is as follows: calculate the ratio of the measured P-wave velocity to the benchmark value. K = V p / V std When 0.9≤ K When the density is ≤1.15, it is determined to be true bedrock. K When K < 0.9, it is determined to be a weathered zone or a weak interlayer, and drilling continues. When K > 1.15, it indicates abnormally hard rock and triggers secondary penetration verification. The benchmark value... V std It is determined by the results of acoustic logging or indoor ultrasonic testing of rock samples during the site exploration phase.

7. The method for intelligent identification of false bedrock in deep foundation pit retaining engineering for backfilled mine pits according to claim 1, characterized in that: In step 6, the identification results specifically include: the spatial distribution of boulders and the bedrock surface elevation point cloud obtained after steps 4 and 5; the spatial distribution of boulders and the bedrock surface elevation point cloud, along with their coordinates and timestamps, are written back to the three-dimensional risk layer to form a dynamically updated three-dimensional geological-boulder joint digital model, which is used to identify subsequent pile locations within the same foundation pit. e th , Dest and L test Recalibrate.

8. The method for intelligent identification of false bedrock in deep foundation pit retaining engineering for backfilled mine pits according to claim 1, characterized in that: The method is applied to the construction of retaining structures for deep foundation pits built on backfilled mine pits, backfilled open-pit mines, or backfilled quarries. The retaining structure is one or more of the following: diaphragm wall, bored piles, or interlocking piles. In step 4, if the current pile position is located within the backfill body or bedrock transition zone marked by the three-dimensional risk layer, the safety margin Δs is taken as the upper limit of 1.0m.

9. A false bedrock intelligent identification system for deep foundation pit retaining engineering in backfilled mine pits, used to implement the method described in any one of claims 1-8, characterized in that: The system comprises, in sequence, a drilling multi-parameter sensing subsystem, a geological prior data integration module, an edge computing decision unit, a penetration verification execution unit, an acoustic final judgment unit, and a feedback update module; wherein: The multi-parameter sensing subsystem for drilling includes torque sensors, speed encoders, displacement sensors, feed pressure sensors, and flushing fluid inlet / outlet pressure or flow sensors deployed on the drill rig's power head, drill pipe, and drill string, respectively. These sensors are used to synchronously acquire the rotational torque T and drill bit speed during the drilling process at a sampling frequency of not less than 50Hz. n Instantaneous advance rate v Pressure input F and flushing hydraulic differential Δ P f Time series data; The geological prior data integration module stores three-dimensional geophysical data of the construction area obtained by combining one or more methods, such as transient electromagnetic method, ground-penetrating radar or trans-hole resistivity imaging, and constructs a three-dimensional risk layer with the center coordinates of suspected boulders, estimated equivalent grain size and confidence level of high resistivity anomaly as attributes. The edge computing decision unit is connected to the drilling multi-parameter sensing subsystem via a bus, and has a built-in drilling specific energy calculation engine and formation attribute discriminator, used to convert in-situ mechanical characteristics into formation anti-fracture strength indicators. e The system combines the three-dimensional risk layer to output three types of instructions: suspected bedrock, suspected boulder, or continue drilling. The penetration verification execution unit is used to automatically calculate and issue a penetration verification depth to the drilling rig when the edge computing decision unit outputs a suspected bedrock command. L test Control the drilling rig to continue drilling downwards under the current power parameters. L test And the energy ratio within this range e The magnitude of the drop Δ e Conduct closed-loop monitoring; The acoustic wave final determination unit includes an acoustic transducer installed at the bottom of the drill pipe or lowered through the inner cavity of the drill pipe. After the penetration verification execution unit determines that the specific energy is stable, it uses the pulse transmission method to measure the longitudinal wave velocity of the medium at the bottom of the borehole. V p and compared with the pre-calibrated regional bedrock longitudinal wave velocity reference value. V std The comparison is performed, and the final result of determining whether the bedrock is real or fake is output. The feedback update module is used to write back the spatial distribution of boulders and the bedrock surface elevation point cloud obtained after penetration verification and acoustic final judgment to the geological prior data integration module, forming a dynamically updated three-dimensional geological-boulder joint digital model as construction progresses, which is used for subsequent pile locations within the same foundation pit. e th , Dest and L test Recalibrate.

10. The intelligent identification system for false bedrock in deep foundation pit retaining engineering of backfilled mine pits according to claim 9, characterized in that: After determining that it is a false bedrock boulder, the edge calculation decision unit further adjusts the torque limiter and feed pressure limit of the drilling rig in real time based on the recorded boulder thickness and position. When the drill bit is passing through the interface between the boulder and the backfill, the torque limit is reduced to 70% to 85% of the rated value to suppress hole deviation and drill jamming caused by lateral impact.

Images