Construction method of soft soil foundation CFG pile

Abstract

The present application relates to soft soil foundation CFG pile construction method, belongs to soft soil foundation treatment technical field. In view of the technical problems of difficult control of pile quality and low efficiency of pile forming in soft soil foundation construction, the technical scheme main points are: first, the site is leveled and compacted, and the pile position is marked out by measuring and lofting, and the ground pumping system including concrete truck pump and high pressure pipeline is established;Then move the long spiral drill to the pile position, drill at the initial speed, monitor the motor current to 150-180A, determine that it enters the hard soil layer, adjust the speed to 0.4-0.9m / min, drill to the designed pile depth;Then pump the CFG mixture, monitor the pumping pressure to 0.5-0.9MPa, then lift the drill rod at the initial speed, adjust the pumping displacement and lifting speed in real time according to the outlet pressure at the bottom of the drill rod;Finally, cover wet cloth and water curing after pouring to the top elevation of the pile. The method is mainly used for CFG pile construction of soft soil foundation, which can improve the pile quality and efficiency.

Description

Technical Field

[0001] This invention relates to the field of CFG pile construction technology. More specifically, this invention relates to a method for constructing CFG piles on soft soil foundations. Background Technology

[0002] In the field of soft soil foundation treatment, CFG piles are widely used in foundation reinforcement projects for roads, bridges, and buildings due to their advantages in effectively improving the bearing capacity of the foundation and reducing subsequent settlement. Soft soil foundations are characterized by high water content, large void ratio, high compressibility, and low shear strength. Before bearing the load of the superstructure, such foundations must be treated with methods such as CFG piles to improve their overall stability. During CFG pile construction, the mixture injection stage is the core process that determines the quality of the pile. This stage requires precise coordination between pumping pressure and drill rod lifting speed—only when the two are matched can the mixture be ensured to fill the pile hole evenly, avoiding quality defects such as pile voids, necking, or pile breakage, while also ensuring construction efficiency. Existing construction methods have not established a real-time feedback adjustment mechanism for pumping pressure and drill rod lifting speed, and cannot flexibly adapt construction parameters according to the dynamically changing resistance of the soft soil strata. Due to the strong heterogeneity of soft soil foundations, soil conditions may vary at different pile locations and at different depths of the same pile hole within the same construction area. Fixed parameters are difficult to adapt to all working conditions, resulting in poor stability of grouting quality and difficulty in guaranteeing the qualified rate of pile formation. This not only increases the cost of rework in the later stages but may also pose hidden dangers to the safety of the superstructure. Summary of the Invention

[0003] Another objective of this invention is to provide a method for constructing CFG piles on soft soil foundations.

[0004] To achieve these objectives and other advantages according to the present invention, a method for constructing CFG piles on soft soil foundations is provided, comprising the following steps:

[0005] Step 1: Level the soft soil foundation site and compact the ground with a road roller. Use a total station to measure and mark the center point of each CFG pile. At the same time, set up a ground pumping system, which includes a concrete truck pump, whose outlet is connected to the injection port at the top of the long spiral drill rod through a high-pressure pipeline.

[0006] Step 2: Move the long spiral drilling rig to the pile location, start the drilling rig and drive the drill rod downward to drill. The initial drilling speed is 1.0-1.8 m / min, and monitor the current value of the drilling rig motor in real time. When the current value reaches 150-180 A, it is determined that the drill bit has entered the hard soil layer. Adjust the drilling speed to 0.4-0.9 m / min and continue drilling to the designed pile depth.

[0007] Step 3: After drilling to the designed pile depth, CFG mixture is pumped into the ground pumping system for grouting. When the pumping pressure is monitored to rise to 0.5-0.9 MPa, the drill rod is lifted at an initial speed of 0.5-0.8 m / min. During the lifting process, the bottom outlet pressure of the drill rod is monitored in real time. When the bottom outlet pressure of the drill rod is lower than 0.5 MPa, the pumping capacity of the truck pump is increased simultaneously and the lifting speed of the drill rod is reduced to 0.3-0.4 m / min. When the bottom outlet pressure of the drill rod is higher than 0.9 MPa, the pumping capacity of the truck pump is reduced simultaneously and the lifting speed of the drill rod is increased to 0.9-1.0 m / min.

[0008] Step 4: After pouring to the top elevation of the pile, cover the top of the pile with wet burlap and water it regularly for maintenance.

[0009] Preferably, step two further includes establishing a formation-drilling parameter model, the method of which is as follows:

[0010] During the drilling process, the current value of the drilling rig motor and the corresponding drilling depth are recorded in real time at a preset sampling frequency to generate the current-depth profile data of the pile.

[0011] Based on the current-depth profile data, the depths of different soil layer interfaces penetrated during drilling are identified and recorded. For each soil layer segment, the ratio of the average current value to the thickness of the segment is defined as the average drilling resistance coefficient of that soil layer segment.

[0012] Before pumping and grouting begins in step three, an initial drill rod lifting speed control benchmark curve is generated for the current pile based on the average drilling resistance coefficient. In the drill rod lifting speed control benchmark curve, for soft soil sections where the average drilling resistance coefficient is less than the first threshold, the initial drill rod lifting speed is set to 0.7-0.8 m / min. For hard soil sections where the average drilling resistance coefficient is greater than or equal to the first threshold, the initial drill rod lifting speed is set to 0.4-0.6 m / min.

[0013] Preferably, when generating the drill pipe lifting speed control reference curve, the method further includes a step of compensating for pumping displacement in the soft soil section:

[0014] Based on the current-depth profile data and combined with the real-time collected drill rod torque data and drill rod angular velocity data, the soft soil disturbance effect factor when the drill bit drills in each soft soil section is calculated. The calculation formula is: W = ∫(torque T × angular velocity ω) dt, with the unit being kilojoules;

[0015] The calculated soft soil disturbance effect factor is compared with a first preset threshold. When the soft soil disturbance effect factor exceeds the first preset threshold, a dynamic compensation discharge is added to the initial pumping discharge in the soft soil section to compensate for the increased demand for the mixture caused by soil disturbance.

[0016] Preferably, the method further includes the following steps:

[0017] During the drilling and grouting process, an adaptive control database corresponding to the pile is established and updated in real time. The adaptive control database stores the elevation range of the grouting pile segment and its corresponding soft soil disturbance effect factor.

[0018] When the drill rod is raised to the starting elevation of a soft soil section to be grouted, the grouted section with the same elevation interval as the current soft soil section to be grouted is queried from the adaptive control database, and its soft soil disturbance effect factor is obtained.

[0019] The soft soil disturbance effect factor obtained from the query is compared with the second preset threshold.

[0020] Adjust the pumping discharge rate setting value of the soft soil section to be grouted based on the comparison results: if the soft soil disturbance effect factor is greater than the second preset threshold, increase the pumping discharge rate setting value; if the soft soil disturbance effect factor is less than or equal to the second preset threshold, decrease the pumping discharge rate setting value.

[0021] Preferably, in step three, the process of real-time monitoring of the drill pipe bottom outlet pressure and adjusting the pumping flow rate and drill pipe lifting speed accordingly further includes the following steps:

[0022] S1. Calculate the pressure change rate based on real-time monitored drill pipe bottom outlet pressure data;

[0023] S2. Determine whether the pressure change rate meets a preset risk condition. The determination of the risk condition includes:

[0024] Determination of blockage risk conditions: The pressure change rate is negative, and its absolute value is greater than the third preset threshold, and it continues for the first predetermined time.

[0025] Determination of pressure stagnation risk conditions: The rate of pressure change is positive, and its value is greater than the fourth preset threshold, and it continues for the second predetermined time.

[0026] S3. If the blockage risk condition is met, a first intervention measure is implemented before the bottom outlet pressure of the drill pipe drops to 0.5 MPa. The first intervention measure includes increasing the pumping capacity of the truck pump and reducing the lifting speed of the drill pipe. If the pressure buildup risk condition is met, a second intervention measure is implemented before the bottom outlet pressure of the drill pipe rises to 0.9 MPa. The second intervention measure includes reducing the pumping capacity of the truck pump and increasing the lifting speed of the drill pipe.

[0027] The adjustment range of the pumping displacement and drill pipe lifting speed is positively correlated with the absolute value of the rate of change corresponding to the risk conditions met.

[0028] Preferably, in step S1, while calculating the pressure change rate, a pre-trained pressure trend prediction model is invoked to predict the pressure change trajectory and its corresponding prediction confidence level in real time within the next Δt time period.

[0029] In step S2, the determination of the risk condition further includes:

[0030] When the prediction confidence level is higher than the preset confidence threshold, and the predicted pressure change trajectory indicates that the bottom outlet pressure of the drill pipe will exceed the pressure range of 0.5-0.9 MPa within a time interval of Δt, the first or second intervention measure will be triggered in advance, and the adjustment range of the pumping displacement and the drill pipe lifting speed is positively correlated with the degree to which the predicted pressure deviates from the critical value.

[0031] When the prediction confidence level is lower than or equal to the preset confidence threshold, the system will switch to step S2 to make a risk judgment based on the rate of change of the bottom outlet pressure of the drill pipe.

[0032] Preferably, during the drill pipe lifting and grouting process in step three, the pressure trend prediction model is fine-tuned online using a dynamic learning rate;

[0033] The online fine-tuning steps include:

[0034] The actual monitoring value of the bottom outlet pressure of the drill pipe is obtained in real time and compared with the predicted value of the pressure trend prediction model at the previous moment to calculate the prediction error.

[0035] Based on the absolute value of the prediction error and the level of prediction confidence, the learning rate of the pressure trend prediction model parameters is dynamically adjusted. Using the updated pressure trend prediction model, a multi-step prediction process is executed: at each new sampling time, the pressure trend for a future period is predicted based on the latest data, thereby achieving continuous updates to the prediction results; and intervention measures are implemented based on these continuously updated prediction results.

[0036] Preferably, the method for dynamically adjusting the learning rate of the pressure trend prediction model is as follows:

[0037] When the absolute value of the prediction error is greater than a preset error threshold, and the prediction confidence is lower than a preset confidence threshold, the pressure trend prediction model is determined to be inaccurate, and the learning rate is increased for rapid correction.

[0038] When the absolute value of the prediction error is less than or equal to a preset error threshold, and the prediction confidence is higher than a preset confidence threshold, the pressure trend prediction model is determined to be in an accurate state, and the learning rate is reduced to maintain stability.

[0039] In other cases, the learning rate is kept at the default value.

[0040] Preferably, in step two, the adjustment range of the drilling speed is related to the current value:

[0041] When it is determined that the hard soil layer has been entered, the current value that triggers the speed adjustment is recorded;

[0042] Based on the preset mapping relationship between current value and drilling speed, a new drilling speed is determined in the range of 0.4-0.9 m / min. The mapping relationship satisfies the principle that the larger the current value, the smaller the drilling speed should be.

[0043] Preferably, in step four, an automatic maintenance system is used during watering maintenance. The automatic maintenance system includes a permeable and moisturizing pad covering the top of the pile, a humidity sensor installed on the moisturizing pad, and a water storage tank and a pump connected to the moisturizing pad through pipelines. Multiple humidity sensors are installed and distributed in different positions under the moisturizing pad.

[0044] The humidity sensor monitors the humidity data of the mixture at the top of the pile in real time and transmits it to the controller. The controller calculates the average value of all humidity sensor readings.

[0045] When the average value is lower than the fifth preset threshold, the controller instructs the pump to start and replenish the moisture to the moisturizing pad in a drip irrigation manner;

[0046] When the average value is higher than the sixth preset threshold, the controller commands the pump to shut down and stop water replenishment.

[0047] The present invention has at least the following beneficial effects:

[0048] This invention ensures drilling rig stability and accurate pile positioning through standardized site preparation and precise layout; dynamically adjusts drilling speed based on current value to adapt to differences in soft and hard soil layers; coordinates pump discharge and lifting speed in real time according to outlet pressure to ensure pile compaction; and strengthens maintenance control to enhance pile strength. The overall process is scientific and systematic, effectively improving pile quality and construction efficiency, and is applicable to CFG pile construction on various soft soil foundations. By establishing a stratum-drilling parameter model, using current-depth data to identify soil layer interfaces and calculate the average drilling resistance coefficient, a scientific benchmark is provided for the drill rod lifting speed. This eliminates reliance on experience in setting the lifting speed, enabling precise matching of different soil layer characteristics, avoiding parameter blindness, reducing pile defects caused by parameter mismatch with stratum, and improving the stability and specificity of grouting quality. By calculating the soft soil disturbance effect factor, the degree of soft soil disturbance is quantified, and the pump discharge is dynamically compensated accordingly, enabling targeted responses to changes in mixture requirements caused by soil disturbance. To avoid voids due to insufficient discharge in soft soil sections or waste caused by excessive discharge, ensure uniform pile filling, improve the quality of pile formation in soft soil sections, and reduce rework in later stages.

[0049] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Detailed Implementation

[0050] The present invention will be further described in detail below with reference to embodiments, so that those skilled in the art can implement it based on the description.

[0051] This invention provides a method for constructing CFG piles on soft soil foundations, comprising the following steps:

[0052] Step 1: Level the soft soil foundation site and compact the ground with a road roller. Use a total station to measure and mark the center point of each CFG pile. At the same time, set up a ground pumping system, which includes a concrete truck pump, whose outlet is connected to the injection port at the top of the long spiral drill rod through a high-pressure pipeline.

[0053] Step 2: Move the long spiral drilling rig to the pile location, start the drilling rig and drive the drill rod downward to drill. The initial drilling speed is 1.0-1.8 m / min, and monitor the current value of the drilling rig motor in real time. When the current value reaches 150-180 A, it is determined that the drill bit has entered the hard soil layer. Adjust the drilling speed to 0.4-0.9 m / min and continue drilling to the designed pile depth.

[0054] Step 3: After drilling to the designed pile depth, CFG mixture is pumped into the ground pumping system for grouting. When the pumping pressure is monitored to rise to 0.5-0.9 MPa, the drill rod is lifted at an initial speed of 0.5-0.8 m / min. During the lifting process, the bottom outlet pressure of the drill rod is monitored in real time. When the bottom outlet pressure of the drill rod is lower than 0.5 MPa, the pumping capacity of the truck pump is increased simultaneously and the lifting speed of the drill rod is reduced to 0.3-0.4 m / min. When the bottom outlet pressure of the drill rod is higher than 0.9 MPa, the pumping capacity of the truck pump is reduced simultaneously and the lifting speed of the drill rod is increased to 0.9-1.0 m / min.

[0055] Step 4: After pouring to the top elevation of the pile, cover the top of the pile with wet burlap and water it regularly for maintenance.

[0056] In this technical solution, in step one, a road roller with a self-weight of 15t-20t can be used to compact the ground, with the number of compaction cycles controlled at 3-5 times to ensure that the ground compaction degree reaches more than 85%. For surveying and setting out, a total station with an accuracy of ±2mm+2ppm can be selected. After marking the center point of the pile position, a 5cm-8cm diameter steel bar can be inserted into the ground as a marker, with the steel bar protruding 10cm-15cm above the ground. In the ground pumping system, a concrete truck pump with a theoretical conveying capacity of 30m³ / h-60m³ / h can be selected. Wear-resistant alloy steel pipes with an inner diameter of 125mm-150mm can be used for the high-pressure pipeline. Sealing gaskets must be used at the pipeline connections to prevent leakage. The connection between the high-pressure pipeline and the top injection port of the long spiral drill rod can be achieved through flanges, with rubber gaskets installed between the flanges to enhance sealing. In step two, a long spiral drill rig with a drill rod diameter of 500mm-800mm can be selected. The initial drilling speed can be selected from 1.0m / min, 1.2m / min, 1.5m / min, and 1.8m / min depending on the density of the soft soil. A current sensor with an accuracy of ±1A can be used to monitor the drill rig's motor current. When the current value reaches 150A, 160A, 170A, or 180A, it is determined that a hard soil layer has been entered. At this time, the drilling speed can be adjusted to 0.4m / min, 0.6m / min, 0.8m / min, or 0.9m / min. During the drilling process, the verticality deviation of the drill rod must be kept below 1%, which can be monitored in real time using the drill rig's built-in verticality monitor. In step three, the CFG mixture is prepared by mixing cement, crushed stone, sand, and water in a specific ratio. P.O42.5 grade ordinary Portland cement can be selected, the crushed stone particle size can be selected from 5mm to 20mm, and the sand can be medium sand with a mud content not exceeding 3%. Pump pressure monitoring can use a range of 0-2MPa with an accuracy of ±0.05MPa. The pressure sensor can be selected for the initial drill pipe lifting speed of 0.5m / min, 0.6m / min, 0.7m / min, or 0.8m / min. The pressure sensor at the bottom outlet of the drill pipe, with a range of 0-2MPa and an accuracy of ±0.05MPa, is also used for monitoring. When the outlet pressure is below 0.5MPa, the pumping capacity of the truck-mounted pump can be increased by 5m³ / h-10m³ / h, and the lifting speed reduced to 0.3m / min or 0.4m / min. When the outlet pressure is above 0.9MPa, the pumping capacity is reduced by 5m³ / h-10m³ / h, and the lifting speed is increased to 0.9m / min or 1.0m / min. In step four, the wet burlap covering the pile top can be made of 2mm-3mm thick cotton burlap. Watering should be done 3-4 times daily, ensuring the burlap is completely wet each time, and the curing time should be no less than 7 days.

[0057] During the work, the site is first leveled and compacted, then the pile position is located and marked using a total station. Finally, the concrete truck pump and high-pressure pipeline are assembled, ensuring a tight connection between the pipeline and the drill rod injection port. The drilling rig is moved to the pile position and aligned with the mark. The drilling rig is started and drilled at the initial speed. At the same time, the current is monitored by a current sensor. After the set current value is reached, the speed is adjusted and drilling continues to the designed pile depth. During this period, the verticality of the drill rod is monitored in real time. After drilling to the designed pile depth, the truck pump is started to pump the mixture. After monitoring that the pumping pressure reaches the set range, the drill rod is lifted. The outlet pressure is monitored in real time, and the pumping flow rate and lifting speed are adjusted according to the pressure changes. After the concrete is poured to the top elevation of the pile, it is immediately covered with wet burlap. Water is then sprayed at the set frequency to keep the burlap moist until the curing is completed.

[0058] This invention employs a scientific and systematic approach, ensuring drilling rig stability and accurate pile positioning through standardized site preparation and precise layout; dynamically adjusting drilling speed based on current values ​​to adapt to differences in soil layers (soft vs. hard); coordinating pump discharge and lifting speed in real-time according to outlet pressure to guarantee pile compaction; and strengthening maintenance control to enhance pile strength. The overall process is scientific and systematic, effectively improving pile quality and construction efficiency, and is suitable for CFG pile construction on various soft soil foundations.

[0059] In another technical solution, step two further includes establishing a formation-drilling parameter model, the method of which is as follows:

[0060] During the drilling process, the current value of the drilling rig motor and the corresponding drilling depth are recorded in real time at a preset sampling frequency to generate the current-depth profile data of the pile.

[0061] Based on the current-depth profile data, the depths of different soil layer interfaces penetrated during drilling are identified and recorded. For each soil layer segment, the ratio of the average current value to the thickness of the segment is defined as the average drilling resistance coefficient of that soil layer segment.

[0062] Before pumping and grouting begins in step three, an initial drill rod lifting speed control benchmark curve is generated for the current pile based on the average drilling resistance coefficient. In the drill rod lifting speed control benchmark curve, for soft soil sections where the average drilling resistance coefficient is less than the first threshold, the initial drill rod lifting speed is set to 0.7-0.8 m / min. For hard soil sections where the average drilling resistance coefficient is greater than or equal to the first threshold, the initial drill rod lifting speed is set to 0.4-0.6 m / min. The sampling frequency during drilling can be selected as 1, 2, or 3 times / second. The current sensor used to record the current value has an accuracy of ±1A. The drilling depth can be recorded using the drilling rig's built-in depth encoder, with a depth measurement accuracy of ±1cm. The generated current-depth profile data can be stored in the drilling rig's data acquisition system in Excel or CSV format for easy subsequent processing. Identifying soil interface depths can be achieved by analyzing abrupt changes in the current-depth profile data. When the current value change between adjacent sampling points exceeds 20A, it is determined to be a soil interface. When calculating the average drilling resistance coefficient, the soil layer is first divided into sections, each with a thickness of not less than 0.5m. Then, the average value of all current values ​​within that section is calculated. The average drilling resistance coefficient is obtained by dividing the average value by the section thickness. The calculation process can be completed using an Excel spreadsheet or professional data processing software. The first threshold is obtained through extensive experimentation and can be set to 50A / m or 60A / m. Or 70A / m. When the average drilling resistance coefficient is less than 50A / m (or 60A / m, 70A / m), it is determined to be a soft soil section, and the initial drill rod lifting speed in this section can be set to 0.7m / min-0.8m / min; when the average drilling resistance coefficient is greater than or equal to 50A / m (or 60A / m, 70A / m), it is determined to be a hard soil section, and the initial drill rod lifting speed can be set to 0.4m / min-0.6m / min. The drill rod lifting speed control benchmark curve can be generated using CAD or professional drawing software. The horizontal axis of the curve is the drilling depth, and the vertical axis is the lifting speed. The first threshold is obtained through a large amount of experimental data. During the operation, the current value and drilling depth are recorded synchronously at the set sampling frequency. The data is transmitted to the data acquisition system in real time and stored in a specified format to form the current-depth profile data of a single pile; for the current-depth profile... The depth profile data is analyzed to find the current change point and determine the soil layer interface. After dividing the soil layer into sections, the average current value and thickness of each section are calculated to obtain the average drilling resistance coefficient. A first threshold is set, and the soil layer type is determined based on the average drilling resistance coefficient of each soil layer section. An initial lifting speed is set for different types of sections, and a complete lifting speed control baseline curve is generated by drawing software.By adopting this technical solution, the present invention establishes a formation-drilling parameter model, uses current-depth data to identify soil interfaces, and calculates the average drilling resistance coefficient, providing a scientific benchmark for drill rod lifting speed. This eliminates reliance on experience in setting the lifting speed, enabling precise matching of different soil characteristics, avoiding parameter blindness, reducing pile defects caused by parameter mismatch with formation, and improving the stability and specificity of grouting quality.

[0063] In another technical solution, the step of compensating for pumping displacement in the soft soil section is also included when generating the drill pipe lifting speed control reference curve.

[0064] Based on the current-depth profile data and combined with the real-time collected drill rod torque data and drill rod angular velocity data, the soft soil disturbance effect factor when the drill bit drills in each soft soil section is calculated. The calculation formula is: W = ∫(torque T × angular velocity ω) dt, with the unit being kilojoules;

[0065] The calculated soft soil disturbance effect factor is compared with a first preset threshold. When the soft soil disturbance effect factor exceeds the first preset threshold, a dynamic compensation discharge is added to the initial pumping discharge in the soft soil section to compensate for the increased demand for the mixture caused by soil disturbance. Torque data of the drill rod can be acquired by a torque sensor installed at the drive end of the drill rod. The torque sensor can be selected with a range of 0-50 kN·m and an accuracy of ±0.1 kN·m. Angular velocity data can be calculated from the drill rig motor speed. Motor speed monitoring uses a speed sensor with an accuracy of ±1 r / min. The angular velocity is calculated by multiplying the speed (r / min) by 2π and then dividing by 60, converting the unit to rad / s. The soft soil disturbance effect factor W can be calculated using MATLAB or Python software, with the integration time step consistent with the sampling frequency, ranging from 1 s to 3 s. The first preset threshold is obtained from experimental data and can be set to 500 kJ, 600 kJ, or 700 kJ. The initial pumping discharge rate can be determined according to the pile diameter. For pile diameters of 500 mm-600 mm, the initial discharge rate is 30 m³ / h-40 m³ / h; for pile diameters of 600 mm-800 mm... The initial determination of the dynamic compensation displacement is achieved through a preset control strategy, which can be manifested in one or more of the following ways. The determination of the first preset threshold needs to match the set drilling speed. For example, when the set drilling speed for the soft soil section is 1.5 m / min and the thickness of the section is 1.0 m, the drilling time is approximately 40 seconds. For a drilling rig with a power of 90 kW, the energy consumed in this time at a load rate of 60% is approximately: W = 90 kW × 60% × 40 s = 2160 kJ. Based on this calculation and considering the safety factor for soil variability, the first preset threshold is set to 2500 kJ. Conditional Judgment Method: The system has at least one preset displacement compensation level. For example, when the soft soil disturbance effect factor W exceeds the seventh preset threshold (W0) but does not reach the higher eighth preset threshold, the first compensation displacement △Q1 is activated; when W exceeds the eighth threshold, the second compensation displacement △Q2 is activated, and the second compensation displacement is greater than the first compensation displacement. Lookup Table Method: A "disturbance factor - compensation displacement" lookup table is pre-stored in the control system. The system directly locks the corresponding optimal compensation displacement △Q by looking up the table based on the calculated real-time W value, and superimposes it with the initial pumping displacement. Example Adjustment Method: The compensation displacement △Q is calculated by multiplying a benchmark displacement (which can be the initial pumping displacement or its proportion) by a compensation coefficient k that is proportional to (W - W0), that is, following the basic principle of "the greater the disturbance, the more compensation".Displacement adjustment can be achieved by operating the control panel of the concrete pump truck. After adjustment, the outlet pressure needs to be monitored by the pressure sensor to ensure that the pressure is stable within a reasonable range. When generating the reference curve for drill rod lifting speed control, the pumping displacement compensation parameters of the soft soil section and the lifting speed parameters need to be integrated. In addition to marking the lifting speed, the pumping displacement of the corresponding section also needs to be marked in the curve for easy viewing by operators. Parameter integration can be completed using an Excel spreadsheet. First, list the depth, lifting speed, initial discharge rate, and compensation discharge rate for each soft soil section. Then, import the data into plotting software to generate a baseline curve containing discharge rate information. During operation: When drilling into the soft soil section, a torque sensor collects torque T in real time, and a speed sensor monitors the motor speed and calculates the angular velocity ω. Integrate T×ω over a set time step to obtain the soft soil disturbance effect factor W. Set a first preset threshold and an initial pumping discharge rate. Compare the calculated W with the threshold. If it exceeds the threshold, add compensation discharge rate to the initial discharge rate. After adjustment, monitor the outlet pressure to ensure normal grouting. Based on the generated lifting speed baseline curve, supplement the pumping discharge rate data (including compensation discharge rate) for each soft soil section to form a complete speed-discharge joint control curve, providing clear parameter guidance for subsequent grouting. Using this technical solution, this invention quantifies the degree of soft soil disturbance by calculating the soft soil disturbance effect factor and dynamically compensates the pumping discharge rate accordingly, enabling targeted responses to changes in mixture demand caused by soil disturbance. To avoid voids due to insufficient discharge in soft soil sections or waste caused by excessive discharge, ensure uniform pile filling, improve the quality of pile formation in soft soil sections, and reduce rework in later stages.

[0066] Another technical solution also includes the following steps:

[0067] During the drilling and grouting process, an adaptive control database corresponding to the pile is established and updated in real time. The adaptive control database stores the elevation range of the grouting pile segment and its corresponding soft soil disturbance effect factor.

[0068] When the drill rod is raised to the starting elevation of a soft soil section to be grouted, the grouted section with the same elevation interval as the current soft soil section to be grouted is queried from the adaptive control database, and its soft soil disturbance effect factor is obtained.

[0069] The soft soil disturbance effect factor obtained from the query is compared with the second preset threshold.

[0070] Adjust the pumping displacement setting value of the soft soil section to be grouted according to the comparison result: if the soft soil disturbance effect factor is greater than the second preset threshold, increase the pumping displacement setting value; if the soft soil disturbance effect factor is less than or equal to the second preset threshold, decrease the pumping displacement setting value; the database can use a MySQL or SQLServer database management system. The elevation range of the already grouted pile sections stored needs to be accurate to 0.1 m, such as "1.0 m - 1.5 m", "1.5 m - 2.0 m". The soft soil disturbance effect factor needs to be reserved to one decimal place. Data entry can be manually input through the database client or automatically imported through the data acquisition system to ensure real-time data update. The database needs to be installed in the control computer supporting the drill rig; when the drill pipe is lifted to the starting elevation of the soft soil section to be grouted, determine the starting elevation through the depth encoder of the drill rig with an accuracy of ±1 cm, and then retrieve the records of the same elevation range in the database through an SQL query statement, such as "SELECT soft soil disturbance effect factor FROM database table WHERE elevation range = '1.0 m - 1.5 m'". The query result needs to be fed back within 1 s to ensure that the construction progress is not affected. If multiple records of the same elevation range are retrieved, take the average value of the soft soil disturbance effect factors as the reference value; if no records of the same range are retrieved, use the initial pumping displacement setting value; the second preset threshold can be set to 550 kJ, 650 kJ or 750 kJ. If the retrieved soft soil disturbance effect factor is greater than the threshold, the pumping displacement setting value can be increased by 5 m 3 / h - 8 m 3 / h; if it is less than or equal to the threshold, the pumping displacement setting value can be decreased by 3 m 3 / h - 5 m 3 / h. The displacement adjustment is achieved through the electric control system of the concrete truck pump. After adjustment, the actual displacement needs to be monitored through a flow sensor. The flow sensor can be selected with a measurement range of 0 - 100 m 3 / h and an accuracy of ±1 m 3The model is specified as / h, ensuring the actual discharge rate matches the set value; the second preset threshold should be higher than the first preset threshold, and can be set to 3500 kJ. The working process is as follows: A database management system is installed on the control computer. During the grouting process, the elevation intervals of the already grouted pile segments and the corresponding soft soil disturbance effect factors are collected in real time. The data is entered into the database according to the specified format, enabling real-time data updates and storage. After the drill rod is raised to the starting elevation of the section to be grouted, the elevation value is obtained, and the disturbance effect factor of the corresponding interval is retrieved from the database. A reference factor is determined based on the query results; if no results are found, the initial discharge rate is used. A second preset threshold is set, and the reference disturbance effect factor is compared with the threshold. The pump discharge rate setting is adjusted according to the results, and the actual discharge rate is monitored by a flow sensor to ensure it matches the set value. Then, grouting of that section begins. Using this technical solution, this invention establishes an adaptive control database and uses data from the already grouted pile segments to guide parameter adjustments in the section to be grouted, achieving self-optimization of the construction process. This allows the pumping displacement setting to better match actual working conditions, improves parameter adaptability, and is especially suitable for soft soil foundations with complex soil layers. It reduces quality fluctuations caused by parameter solidification and improves the pile qualification rate.

[0071] In another technical solution, step three, which involves real-time monitoring of the drill pipe bottom outlet pressure and adjusting the pumping volume and drill pipe lifting speed accordingly, further includes the following steps:

[0072] S1. Calculate the pressure change rate based on real-time monitored drill pipe bottom outlet pressure data;

[0073] S2. Determine whether the pressure change rate meets a preset risk condition. The determination of the risk condition includes:

[0074] Determination of blockage risk conditions: The pressure change rate is negative, and its absolute value is greater than the third preset threshold, and it continues for the first predetermined time.

[0075] Determination of pressure stagnation risk conditions: The rate of pressure change is positive, and its value is greater than the fourth preset threshold, and it continues for the second predetermined time.

[0076] S3. If the blockage risk condition is met, a first intervention measure is implemented before the bottom outlet pressure of the drill pipe drops to 0.5 MPa. The first intervention measure includes increasing the pumping capacity of the truck pump and reducing the lifting speed of the drill pipe. If the pressure buildup risk condition is met, a second intervention measure is implemented before the bottom outlet pressure of the drill pipe rises to 0.9 MPa. The second intervention measure includes reducing the pumping capacity of the truck pump and increasing the lifting speed of the drill pipe.

[0077] The adjustment range of the pumping displacement and drill pipe lifting speed is positively correlated with the absolute value of the rate of change corresponding to the risk conditions met. The drill pipe bottom outlet pressure data is collected by a pressure sensor. The pressure sensor can be selected with a range of 0-2 MPa and an accuracy of ±0.05 MPa. The data sampling frequency is 1-2 times / second. The pressure change rate is calculated as (current pressure value - previous pressure value) / time interval, with the time interval consistent with the sampling frequency. The calculation result is rounded to three decimal places and the unit is MPa / s. The calculation process can be implemented through the PLC module in the drilling rig control system. The PLC can be a transistor output type with a response time of no more than 0.1s to ensure real-time calculation. The third preset threshold can be set to 0.1MPa / s, 0.15MPa / s, or 0.2MPa / s, and the first preset time can be set to 2s, 3s, or 4s. The fourth preset threshold can be set to 0.1MPa / s, 0.15MPa / s, or 0.2MPa / s, and the second preset time can be set to 2s, 3s, or 4s. If a blockage risk is identified, the pumping capacity can be increased by 5m³ / h-10m³ / h and the drill pipe lifting speed can be reduced by 0.1m / min-0.2m / min before the pressure drops to 0.5MPa. If a pressure buildup risk is identified, the pumping capacity can be reduced by 5m³ / h-10m³ / h and the lifting speed can be increased by 0.1m / min-0.2m / min before the pressure rises to 0.9MPa. The adjustment range is positively correlated with the absolute value of the rate of change. For example, the adjustment range is at its maximum when the absolute value of the rate of change is 0.2 MPa / s, and at its minimum when the absolute value is 0.1 MPa / s. Adjustment is achieved by controlling the actuators of the drilling rig and the pump truck. After adjustment, pressure changes must be continuously monitored to ensure the pressure returns to the normal range. The working process is as follows: the pressure sensor collects outlet pressure data at a set frequency and transmits the data to the PLC module. The PLC calculates the pressure change rate according to the formula and outputs the calculation result in real time. A preset threshold and a predetermined time are set in the PLC. The PLC compares the pressure change rate with the threshold in real time, monitors the duration, and determines the risk type when the conditions are met, displaying a risk warning on the screen. After determining the risk, the adjustment range is determined based on the absolute value of the pressure change rate, and the actuator is controlled to adjust the displacement and lifting speed, continuously monitoring the pressure until it stabilizes within the range of 0.5 MPa-0.9 MPa. Using this technical solution, this invention achieves proactive intervention by monitoring the pressure change rate and predicting risks in advance, avoiding the lag problem of adjusting only after the pressure exceeds the standard. It can effectively prevent the risk of blockage or pressure buildup, reduce equipment wear and construction interruption, ensure continuous and stable grouting process, improve construction safety and efficiency, and ensure uniform pile quality.

[0078] In another technical solution, in step S1, while calculating the pressure change rate, a pre-trained pressure trend prediction model is invoked to predict the pressure change trajectory and its corresponding prediction confidence level in real time within the future time interval Δt.

[0079] In step S2, the determination of the risk condition further includes:

[0080] When the prediction confidence level is higher than the preset confidence threshold, and the predicted pressure change trajectory indicates that the bottom outlet pressure of the drill pipe will exceed the pressure range of 0.5-0.9 MPa within a time interval of Δt, the first or second intervention measure will be triggered in advance, and the adjustment range of the pumping displacement and the drill pipe lifting speed is positively correlated with the degree to which the predicted pressure deviates from the critical value.

[0081] When the predicted confidence level is lower than or equal to the preset confidence threshold, the system will switch to step S2 to make a risk judgment based on the rate of change of the bottom outlet pressure of the drill pipe. The pressure trend prediction model can adopt a 3-layer structure, with 5-10 neurons in the input layer corresponding to the pressure data of the previous 5-10 time moments, 10-20 neurons in the hidden layer, and 1 neuron in the output layer corresponding to the pressure value in the future time interval Δt; Δt can be set to 2s, 3s, or 4s, the prediction confidence level is determined by the probability value output by the model, and the confidence threshold can be set to 0.8, 0.85, or 0.9. The dataset used for model training consists of historical pressure data from construction projects, with a minimum of 1000 datasets. The Adam algorithm can be selected for training, with 1000-2000 training iterations to ensure the model's prediction error is less than 0.05 MPa. When the prediction confidence level exceeds a preset confidence threshold, and the predicted pressure exceeds the 0.5-0.9 MPa range, the adjustment magnitude of the intervention measures is positively correlated with the degree to which the predicted pressure deviates from the critical value. For example, when the predicted pressure is 0.4 MPa (deviating from 0.1 MPa), the discharge rate is increased by 5 m³ / h; when the predicted pressure is 1.0 MPa (deviating from 0.1 MPa), the discharge rate is decreased by 5 m³ / h. Adjustments are made through the electrical control systems of the pump truck and drilling rig. After adjustment, the model continues to predict pressure changes until the pressure stabilizes. When the confidence level is lower than or equal to the preset confidence threshold, the system automatically switches to a control mode based on the real-time instantaneous pressure value and rate of change. During model operation, the predicted results need to be compared with the actual pressure data in real time. If the deviation between the predicted and actual values ​​exceeds 0.05 MPa, the latest data needs to be imported to retrain the model and update the model parameters to ensure the model's prediction accuracy. The model parameters are updated once a day, or once after the completion of 10 piles. The update process is carried out during construction breaks and does not affect normal construction. During operation, a three-layer pressure trend prediction model is constructed, and the model is trained using historical pressure data. Δt and confidence thresholds are set. The model receives the current pressure data in real time and outputs the pressure trajectory and confidence level for the future Δt time. The predicted confidence level is compared with the preset confidence threshold. When the confidence level is high and the pressure is about to exceed the limit, the parameters are adjusted according to the degree of deviation. When the confidence level is low, the system switches to real-time control mode, continuously monitors the pressure and adjusts the parameters. During construction, the predicted and actual pressure values ​​are compared in real time. When the deviation exceeds the limit, the model is retrained with the latest data during construction breaks, the parameters are updated, and the model continues to be used for prediction to ensure model reliability. This invention, employing this technical solution, introduces a pressure trend prediction model to achieve advance prediction of pressure change trends and dynamically adjusts the control mode based on confidence levels. This allows for early intervention before pressure exceeds limits, improving the timeliness and accuracy of risk avoidance. Confidence level assessment ensures the adaptability of the control mode, reduces misjudgments, further guarantees the stability of grouting parameters, and improves pile quality.

[0082] In another technical solution, during the drill pipe lifting and grouting process in step three, the pressure trend prediction model is fine-tuned online using a dynamic learning rate;

[0083] The online fine-tuning steps include:

[0084] The actual monitoring value of the bottom outlet pressure of the drill pipe is obtained in real time and compared with the predicted value of the pressure trend prediction model at the previous moment to calculate the prediction error.

[0085] Based on the absolute value of the prediction error and the level of prediction confidence, the learning rate of the pressure trend prediction model parameters is dynamically adjusted. Using the updated pressure trend prediction model, a multi-step prediction process is executed: at each new sampling time, the pressure trend for a future period is predicted based on the latest data, thus continuously updating the prediction results; and intervention measures are implemented based on these continuously updated prediction results. The prediction error is the difference between the actual monitored value of the drill pipe bottom outlet pressure and the predicted value at the previous moment, and the calculation frequency is consistent with the sampling frequency. The initial learning rate of the pressure trend prediction model can be set to 0.001, 0.005, or 0.01; in the dynamic adjustment of the learning rate and multi-step prediction, the preset error threshold can be set to 0.05MPa, 0.08MPa, or 0.1MPa. When the absolute value of the prediction error is greater than the preset error threshold and the prediction confidence is lower than the preset confidence threshold, the learning rate can be increased to 1.5-2 times the original value; when the absolute value of the prediction error is less than or equal to the preset error threshold and the prediction confidence is higher than the preset confidence threshold, the learning rate can be decreased to 0.5-0.8 times the original value. At each new sampling moment, the updated model is used to predict the pressure trend for a future period, and intervention measures are implemented based on the prediction results. During operation, the actual monitoring value of the outlet pressure at the bottom of the drill pipe is obtained in real time, compared with the predicted value of the pressure trend prediction model at the previous moment, the prediction error is calculated, and the learning rate of the model is dynamically adjusted according to the prediction error and prediction confidence. After updating the model parameters, multi-step prediction is performed, the prediction results are continuously updated, and intervention measures are implemented accordingly. By adopting this technical solution, the pressure trend prediction model can self-correct with changes in working conditions by calculating the prediction error in real time and dynamically adjusting the learning rate. This avoids the model becoming inaccurate after long-term use, ensures the continuity of prediction accuracy, ensures that the intervention measures based on the prediction results are reliable and effective, and improves the adaptive control capability of the grouting process.

[0086] In another technical solution, the method for dynamically adjusting the learning rate of the pressure trend prediction model is as follows:

[0087] When the absolute value of the prediction error is greater than a preset error threshold, and the prediction confidence is lower than a preset confidence threshold, the pressure trend prediction model is determined to be inaccurate, and the learning rate is increased for rapid correction.

[0088] When the absolute value of the prediction error is less than or equal to a preset error threshold, and the prediction confidence is higher than a preset confidence threshold, the pressure trend prediction model is determined to be in an accurate state, and the learning rate is reduced to maintain stability.

[0089] In other cases, the learning rate is kept at the default value; the preset error threshold can be set to 0.04MPa, 0.05MPa or 0.06MPa, and the preset confidence threshold can be set to 0.75, 0.8 or 0.85. The model state determination is realized by the logic judgment program in the control computer. Based on the absolute value of the input prediction error and the confidence level, the program outputs the inaccurate state, accurate state or default state according to the set conditions. The judgment result must be output within 0.1 seconds to ensure that the learning rate adjustment time is not affected. When the judgment is inaccurate, the learning rate can be increased by 2-3 times the initial value, for example, if the initial learning rate is 0.001, increase it to 0.002-0.003. When the judgment is accurate, the learning rate can be decreased by 0.5-0.8 times the initial value, for example, if the initial learning rate is 0.001, decrease it to 0.0005-0.0008. In the default state, the initial learning rate remains unchanged, and the initial learning rate can be selected as 0.001, 0.005, or 0.01. According to the model status judgment result, the learning rate is adjusted through the parameter control interface. When inaccurate, the learning rate is increased to quickly correct the model; when accurate, the learning rate is decreased to maintain stability; in the default state, it remains unchanged. After adjustment, it is synchronized to the model training stage. The adjusted learning rate needs to be verified by the model prediction effect. The verification method is to compare the adjusted learning rate with the model prediction effect for 5-10 seconds. If the absolute value of the error between the predicted and actual pressure values ​​is less than the preset error threshold, the adjustment is considered effective. If the error still exceeds the limit, the state judgment and learning rate adjustment need to be triggered again until the error meets the requirements. The verification process is carried out automatically in the background without affecting normal construction. The verification results can be stored in the model log for subsequent analysis. During operation, error and confidence thresholds are preset in the logic judgment program. The program receives the real-time calculated absolute value of error and confidence level, judges the current state of the model according to the conditions, and quickly outputs the judgment result. After the learning rate is adjusted, the background automatically compares the errors of multiple sets of predicted and actual values ​​to judge the adjustment effect. If the effect is not good, it is readjusted to ensure that the model always runs with an appropriate learning rate. This technical solution scientifically adjusts the learning rate based on prediction error and confidence level, enabling the model to quickly correct when it is inaccurate and run stably when it is accurate, avoiding insufficient or excessive correction caused by a fixed learning rate. This ensures that the model always maintains high prediction accuracy, provides reliable support for pressure trend prediction, and ensures the effectiveness and timeliness of intervention measures.

[0090] In another technical solution, in step two, the adjustment range of the drilling speed is related to the current value:

[0091] When it is determined that the hard soil layer has been entered, the current value that triggers the speed adjustment is recorded;

[0092] Based on the preset mapping relationship between current value and drilling speed, a new drilling speed is determined within the range of 0.4-0.9 m / min. This mapping relationship satisfies the principle that the larger the current value, the smaller the drilling speed should be. The current value range for entering hard soil layers is determined to be 150A-180A. This current value can be recorded by the data logger in the drilling rig control system. The data logger can be selected with a sampling frequency of 1 time / second and a storage capacity of 16GB or more. The recorded content includes the trigger time, current value and corresponding drilling depth, and the data format is TXT or CSV for easy subsequent querying. After recording, the data must be automatically transmitted to the parameter matching module with a transmission delay of no more than 1 second. The mapping relationship between current value and drilling speed can be preset in the parameter matching module in tabular form. For example, a current value of 150A-160A corresponds to a drilling speed of 0.8m / min-0.9m / min, 160A-170A corresponds to 0.6m / min-0.8m / min, and 170A-180A corresponds to 0.4m / min-0.6m / min. The mapping relationship can be fine-tuned according to the actual geological conditions of the project, but the principle of lower speed for higher current values ​​must be followed. The parameter matching module can be a PLC-based control module with a response time of no more than 0.5 seconds to ensure rapid determination of the new drilling speed. After determining the new drilling speed, the drill rod drilling speed needs to be adjusted through the drill rig's speed control unit. The speed control unit can be a frequency converter with a speed adjustment accuracy of ±0.05m / min. During the adjustment process, the drill rod rotation speed and drilling depth need to be monitored in real time to ensure that the actual speed is consistent with the set speed. Simultaneously, the motor current value needs to be continuously monitored. If the current value still exceeds 180A, the drilling speed needs to be adjusted down again according to the mapping relationship until the current stabilizes within a reasonable range. During operation, when the current sensor detects a current value of 150A-180A, the system determines that it has entered a hard soil layer. The data logger immediately records the current value, time, and depth at this time and transmits the data to the parameter matching module. The parameter matching module receives the trigger current value, queries the preset mapping relationship table, determines the corresponding drilling speed range based on the current value range, and then fine-tunes it according to the actual geological conditions, outputting the final new drilling speed. The speed control unit receives the new drilling speed command, adjusts the drill rod speed through the frequency converter, monitors the actual speed and current value in real time, corrects the speed deviation when it exceeds the limit, and adjusts the speed down again when the current is too high to ensure stable drilling in hard soil layers. By adopting this technical solution, this invention establishes a mapping relationship between current value and drilling speed, making the drilling speed setting in hard soil layers more precise. The higher the current value, the lower the speed, which can avoid equipment damage caused by excessive drilling load, while ensuring drilling efficiency and borehole stability. Reduce construction delays or quality problems caused by unreasonable speed settings, and improve the safety and economy of drilling in hard soil layers.

[0093] In another technical solution, in step four, during watering maintenance, an automatic maintenance system is used. The automatic maintenance system includes a permeable and moisturizing pad covering the top of the pile, a humidity sensor installed on the moisturizing pad, and a water storage tank and a pump connected to the moisturizing pad through pipelines. Multiple humidity sensors are installed and distributed in different positions under the moisturizing pad.

[0094] The humidity sensor monitors the humidity data of the mixture at the top of the pile in real time and transmits it to the controller. The controller calculates the average value of all humidity sensor readings.

[0095] When the average value is lower than the fifth preset threshold, the controller instructs the pump to start and replenish the moisture to the moisturizing pad in a drip irrigation manner;

[0096] When the average value exceeds the sixth preset threshold, the controller instructs the pump to shut down and stop water replenishment. The permeable moisture-retaining pad can be a 3mm-5mm thick polyester fiber pad, covering the top of the pile and a 50cm perimeter to ensure complete moisture retention. The humidity sensor can be a soil moisture sensor with a range of 0%-100% RH and an accuracy of ±3% RH. Three to five sensors should be evenly distributed on the pile top surface below the moisture-retaining pad, ensuring close contact between the sensors and the pile top surface for accurate monitoring data. The water storage tank can be a polyethylene tank with a capacity of 50L-100L, the pump can be a miniature diaphragm pump with a flow rate of 1L / min-3L / min, and the pipeline can be a PVC pipe with an inner diameter of 10mm-15mm. The humidity sensor collects the humidity data of the mixture at the top of the pile in real time, with a collection frequency of once every 5min to once every 10min. The data is transmitted to the controller via a wireless transmission module, which can be a Bluetooth or LoRa module with a transmission distance of no more than 100m. The controller can be a single-chip microcomputer controller to calculate the average value of all sensor readings with a calculation accuracy of ±1% RH. The fifth preset threshold can be set to 60% RH-70% RH, and the sixth preset threshold can be set to 85% RH-90%. These thresholds can be adjusted according to the curing requirements of the mixture. When the average humidity is below the fifth preset threshold, the controller instructs the micro diaphragm pump to start, replenishing water to the moisture-retaining pad via drip irrigation through PVC pipes. The replenishment time is determined based on the humidity difference; for example, 5-10 minutes of replenishment for a 10% RH difference. When the average humidity is above the sixth preset threshold, the controller instructs the pump to shut down, stopping the replenishment. Simultaneously, the controller needs to record the replenishment time, frequency, and humidity change data, storing them in local memory for subsequent curing effect analysis. During operation, a polyester fiber moisturizing pad is placed on top of the pile. Three to five humidity sensors are evenly distributed under the pad and close to the pile top, connected to a water storage tank, a miniature diaphragm pump, and PVC piping, ensuring the piping is connected to the pad. The humidity sensors collect data at a set frequency and transmit it wirelessly to a microcontroller. The controller calculates the average humidity and compares it with preset third and sixth thresholds to determine if water replenishment is needed. Based on the humidity comparison results, the controller sends start / stop pump commands. When the pump starts, it replenishes water via drip irrigation; when the high threshold is reached, it stops, recording maintenance data to ensure the pile top humidity remains stable within a suitable range. This automated maintenance system, through multi-point humidity monitoring and intelligent water replenishment, achieves precise control of pile top humidity, avoiding the unevenness and untimely nature of traditional manual maintenance. It ensures the pile is maintained in a suitable humidity environment, reducing surface cracks, promoting stable strength growth, improving pile durability and load-bearing capacity, and reducing maintenance costs.

[0097] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. A method for constructing CFG piles on soft soil foundations, characterized in that, Includes the following steps: Step 1: Level the soft soil foundation site and compact the ground with a road roller. Use a total station to measure and mark the center point of each CFG pile. At the same time, set up a ground pumping system, which includes a concrete truck pump, whose outlet is connected to the injection port at the top of the long spiral drill rod through a high-pressure pipeline. Step 2: Move the long spiral drilling rig to the pile location, start the drilling rig and drive the drill rod downward to drill. The initial drilling speed is 1.0-1.8 m / min, and monitor the current value of the drilling rig motor in real time. When the current value reaches 150-180 A, it is determined that the drill bit has entered the hard soil layer. Adjust the drilling speed to 0.4-0.9 m / min and continue drilling to the designed pile depth. Step 3: After drilling to the designed pile depth, CFG mixture is pumped into the ground pumping system for grouting. When the pumping pressure is monitored to rise to 0.5-0.9 MPa, the drill rod is lifted at an initial speed of 0.5-0.8 m / min. During the lifting process, the bottom outlet pressure of the drill rod is monitored in real time. When the bottom outlet pressure of the drill rod is lower than 0.5 MPa, the pumping capacity of the truck pump is increased simultaneously and the lifting speed of the drill rod is reduced to 0.3-0.4 m / min. When the bottom outlet pressure of the drill rod is higher than 0.9 MPa, the pumping capacity of the truck pump is reduced simultaneously and the lifting speed of the drill rod is increased to 0.9-1.0 m / min. Step 4: After pouring to the top elevation of the pile, cover the top of the pile with wet burlap and water it regularly for curing. Step two also includes establishing a formation-drilling parameter model, which is established as follows: During the drilling process, the current value of the drilling rig motor and the corresponding drilling depth are recorded in real time at a preset sampling frequency to generate the current-depth profile data of the pile. Based on the current-depth profile data, the depths of different soil layer interfaces penetrated during drilling are identified and recorded. For each soil layer segment, the ratio of the average current value to the thickness of the segment is defined as the average drilling resistance coefficient of that soil layer segment. Before pumping and grouting begins in step three, an initial drill rod lifting speed control benchmark curve is generated for the current pile based on the average drilling resistance coefficient. In the drill rod lifting speed control benchmark curve, for soft soil sections where the average drilling resistance coefficient is less than the first threshold, the initial drill rod lifting speed is set to 0.7-0.8 m / min. For hard soil sections where the average drilling resistance coefficient is greater than or equal to the first threshold, the initial drill rod lifting speed is set to 0.4-0.6 m / min. The process of generating the drill pipe lifting speed control baseline curve also includes a step of compensating for pumping displacement in soft soil sections. Based on the current-depth profile data and combined with the real-time collected drill rod torque data and drill rod angular velocity data, the soft soil disturbance effect factor when the drill bit drills in each soft soil section is calculated. The calculation formula is: W = ∫(torque T × angular velocity ω) dt, with the unit being kilojoules; The calculated soft soil disturbance effect factor is compared with a first preset threshold. When the soft soil disturbance effect factor exceeds the first preset threshold, a dynamic compensation discharge is added to the initial pumping discharge in the soft soil section to compensate for the increased demand for the mixture caused by soil disturbance.

2. The method for constructing CFG piles on soft soil foundations as described in claim 1, characterized in that, It also includes the following steps: During the drilling and grouting process, an adaptive control database corresponding to the pile is established and updated in real time. The adaptive control database stores the elevation range of the grouting pile segment and its corresponding soft soil disturbance effect factor. When the drill rod is raised to the starting elevation of a soft soil section to be grouted, the grouted section with the same elevation interval as the current soft soil section to be grouted is queried from the adaptive control database, and its soft soil disturbance effect factor is obtained. The soft soil disturbance effect factor obtained from the query is compared with the second preset threshold. Adjust the pumping discharge rate setting value of the soft soil section to be grouted based on the comparison results: if the soft soil disturbance effect factor is greater than the second preset threshold, increase the pumping discharge rate setting value; if the soft soil disturbance effect factor is less than or equal to the second preset threshold, decrease the pumping discharge rate setting value.

3. The method for constructing CFG piles on soft soil foundations as described in claim 1, characterized in that, Step three, which involves real-time monitoring of the drill pipe bottom outlet pressure and adjusting the pumping rate and drill pipe lifting speed accordingly, further includes the following steps: S1. Calculate the pressure change rate based on real-time monitored drill pipe bottom outlet pressure data; S2. Determine whether the pressure change rate meets a preset risk condition. The determination of the risk condition includes: Determination of blockage risk conditions: The pressure change rate is negative, and its absolute value is greater than the third preset threshold, and it continues for the first predetermined time. Determination of pressure stagnation risk conditions: The rate of pressure change is positive, and its value is greater than the fourth preset threshold, and it continues for the second predetermined time. S3. If the blockage risk condition is met, a first intervention measure is implemented before the bottom outlet pressure of the drill pipe drops to 0.5 MPa. The first intervention measure includes increasing the pumping capacity of the truck pump and reducing the lifting speed of the drill pipe. If the pressure buildup risk condition is met, a second intervention measure is implemented before the bottom outlet pressure of the drill pipe rises to 0.9 MPa. The second intervention measure includes reducing the pumping capacity of the truck pump and increasing the lifting speed of the drill pipe. The adjustment range of the pumping displacement and drill pipe lifting speed is positively correlated with the absolute value of the rate of change corresponding to the risk conditions met.

4. The method for constructing CFG piles on soft soil foundations as described in claim 3, characterized in that, In step S1, while calculating the pressure change rate, a pre-trained pressure trend prediction model is invoked to predict the pressure change trajectory and its corresponding prediction confidence level in real time within the next Δt time period. In step S2, the determination of the risk condition further includes: When the prediction confidence level is higher than the preset confidence threshold, and the predicted pressure change trajectory indicates that the bottom outlet pressure of the drill pipe will exceed the pressure range of 0.5-0.9 MPa within a time interval of Δt, the first or second intervention measure will be triggered in advance, and the adjustment range of the pumping displacement and the drill pipe lifting speed is positively correlated with the degree to which the predicted pressure deviates from the critical value. When the prediction confidence level is lower than or equal to the preset confidence threshold, the system will switch to step S2 to make a risk judgment based on the rate of change of the bottom outlet pressure of the drill pipe.

5. The method for constructing CFG piles on soft soil foundations as described in claim 4, characterized in that, During the drill pipe lifting and grouting process in step three, the pressure trend prediction model is fine-tuned online using a dynamic learning rate. The online fine-tuning steps include: The actual monitoring value of the bottom outlet pressure of the drill pipe is obtained in real time and compared with the predicted value of the pressure trend prediction model at the previous moment to calculate the prediction error. Based on the absolute value of the prediction error and the level of prediction confidence, the learning rate of the pressure trend prediction model parameters is dynamically adjusted. Using the updated pressure trend prediction model, a multi-step prediction process is executed: at each new sampling time, the pressure trend for a future period is predicted based on the latest data, thereby achieving continuous updates to the prediction results; and intervention measures are implemented based on these continuously updated prediction results.

6. The method for constructing CFG piles on soft soil foundations as described in claim 5, characterized in that, The method for dynamically adjusting the learning rate of the stress trend prediction model is as follows: When the absolute value of the prediction error is greater than a preset error threshold, and the prediction confidence is lower than a preset confidence threshold, the pressure trend prediction model is determined to be inaccurate, and the learning rate is increased for rapid correction. When the absolute value of the prediction error is less than or equal to a preset error threshold, and the prediction confidence is higher than a preset confidence threshold, the pressure trend prediction model is determined to be in an accurate state, and the learning rate is reduced to maintain stability. In other cases, the learning rate is kept at the default value.

7. The method for constructing CFG piles on soft soil foundations as described in claim 1, characterized in that, In step two, the adjustment range of the drilling speed is related to the current value: When it is determined that the hard soil layer has been entered, the current value that triggers the speed adjustment is recorded; Based on the preset mapping relationship between current value and drilling speed, a new drilling speed is determined in the range of 0.4-0.9 m / min. The mapping relationship satisfies the principle that the larger the current value, the smaller the drilling speed should be.

8. The method for constructing CFG piles on soft soil foundations as described in claim 1, characterized in that, In step four, during watering and maintenance, an automatic maintenance system is used. The automatic maintenance system includes a permeable and moisturizing pad covering the top of the pile, a humidity sensor installed on the moisturizing pad, and a water storage tank and a pump connected to the moisturizing pad through pipelines. Multiple humidity sensors are installed and distributed in different positions under the moisturizing pad. The humidity sensor monitors the humidity data of the mixture at the top of the pile in real time and transmits it to the controller. The controller calculates the average value of all humidity sensor readings. When the average value is lower than the fifth preset threshold, the controller instructs the pump to start and replenish the moisture to the moisturizing pad in a drip irrigation manner; When the average value is higher than the sixth preset threshold, the controller commands the pump to shut down and stop water replenishment.