A real-time intelligent detection method for three parameters of mud in pile foundation construction.

Abstract

This invention belongs to the field of pile foundation construction technology, specifically a real-time intelligent detection method for three parameters of pile foundation construction mud. The method includes the following steps: Step 1, installing an online detection mechanism on the mud circulation pipeline; the online detection mechanism in Step 1 includes a connecting pipe, the surface of which is fixedly connected to a sampling pipe. This real-time intelligent detection method for three parameters of pile foundation construction mud, by setting up an online detection mechanism and periodically calibrating the dynamic monitoring loop online through a static high-precision loop, fundamentally solves the problems of long-term drift and interference from operating conditions of online sensors, ensuring long-term accuracy and reliability of data. Combined with software compensation algorithms for flow velocity and temperature, it effectively isolates the impact of drastic fluctuations in the main pipeline flow on the measurement; the static detection loop completely eliminates interference from dynamic factors, thus solving the problems of poor real-time performance and low efficiency of existing manual detection methods.

Description

Technical Field

[0001] This invention relates to the field of pile foundation construction technology, and in particular to a real-time intelligent detection method for three parameters of pile foundation construction mud. Background Technology

[0002] Drilled piles are a widely used type of pile in foundation engineering for bridges, high-rise buildings, and other structures. During the drilling process, circulating drilling mud plays a crucial role in wall protection, spoil removal, cooling, and lubrication. The properties of the drilling mud, especially its three core parameters—specific gravity, viscosity, and sand content—directly affect the stability of the borehole wall, drilling efficiency, and the final quality of the pile. When constructing ultra-long piles (pile length > 50 meters) in deep, easily collapsible soil layers (such as silty fine sand layers and silty soft soil), the control requirements for drilling mud properties are even more stringent.

[0003] Currently, the detection of the three parameters of mud at construction sites generally adopts traditional manual methods: measuring specific gravity with a mud hydrometer, measuring viscosity with a funnel viscometer, and measuring sand content with a sieve analysis sand content meter. These methods not only suffer from the problems of intermittent sampling inspections, which cannot reflect the continuous changes in mud performance during construction and make it difficult to detect mud deterioration in a timely manner, resulting in poor real-time performance, but also have the drawbacks of cumbersome and time-consuming testing procedures, which cannot meet the needs of modern rapid construction and thus lead to low efficiency. Therefore, there is a need for a real-time intelligent detection method for the three parameters of mud in pile foundation construction. Summary of the Invention

[0004] To address the technical problems of poor real-time performance and low efficiency in existing manual detection methods, this invention proposes a real-time intelligent detection method for three parameters of mud in pile foundation construction.

[0005] This invention proposes a real-time intelligent detection method for three parameters of mud slurry in pile foundation construction, comprising the following steps: Step 1: Install an online detection mechanism on the mud circulation pipeline; The online detection mechanism in step one includes a connecting pipe, a sampling pipe is fixedly connected to the surface of the connecting pipe, a circular diversion box is fixedly connected to the top of the sampling pipe, a sealing cover is threadedly connected to the upper surface of the circular diversion box, and a PLC controller and a wireless communication module are fixedly installed on the upper surface of the sealing cover respectively. The wireless communication module is connected to the PLC controller via a cable. The wireless communication module is connected to a cloud control platform via a data connection. Step 2: Isokinetic sampling and import. The sampling drive motor is automatically controlled by the PLC controller to drive the sampling shaft and sampling spiral blade to rotate. Step 3: Parallel detection with dual loops. The mud entering the circular diversion box is divided into two independent detection loops: a continuous dynamic monitoring loop and a static high-precision re-inspection loop. Step 4: Data Acquisition and Upload. The PLC controller synchronously acquires raw data from five sensors on two loops and performs preliminary digital filtering and temperature compensation. Step 5: Cloud-based intelligent data processing and calibration; Step Six: On-site Execution and Display. The on-site touchscreen displays all information sent from the cloud in real time. When a control suggestion is received, the corresponding operation can be performed manually or automatically through the dosing system to complete the closed loop from detection and analysis to control.

[0006] Preferably, a sampling drive motor is fixedly installed on the upper surface of the sealing cover. The sampling drive motor is electrically connected to the PLC controller via a cable. The output shaft of the sampling drive motor is fixedly connected to a sampling shaft via a coupling. One end of the sampling shaft passes through and extends to the inner wall of the sampling tube. A sampling spiral blade is fixedly sleeved on the surface of the sampling shaft. A spiral baffle is fixedly connected to the surface of the sampling spiral blade. Both the surface of the sampling spiral blade and the surface of the spiral baffle are slidably connected to the inner wall of the sampling tube. One end of the sampling tube extends to the axis of the connecting tube. A sampling groove is formed on the surface of the sampling tube inside the connecting tube. Two sampling grooves are symmetrically distributed with the axis of the sampling tube as the center. The inner wall of the sampling groove is connected to the inner wall of the sampling tube.

[0007] Preferably, two symmetrically distributed arc-shaped diverter plates are fixedly connected to the surface of the sampling shaft, and the surfaces of the arc-shaped diverter plates are slidably connected to the inner wall of the circular diverter box and the lower surface of the sealing cover, respectively. The inner bottom wall of the circular diversion box is fixedly connected to an online detection tube and a static pressure detection tube, respectively. The online detection tube and the static pressure detection tube form two independent detection loops: a continuous dynamic monitoring loop and a static high-precision re-inspection loop. The surface of the online detection tube is U-shaped, and one end of the online detection tube penetrates and extends to the inner wall of the connecting tube. An online circulating pump and a static pressure circulating pump are fixedly installed on the surface of the online detection tube and the surface of the static pressure detection tube, respectively. Both the online circulating pump and the static pressure circulating pump are electrically connected to the PLC controller via cables.

[0008] Preferably, a tuning fork density sensor, an online viscometer, and a sand content sensor are fixedly mounted on the surface of the online detection tube, and the tuning fork density sensor, the online viscometer, and the sand content sensor are all electrically connected to the PLC controller via cables.

[0009] Preferably, a one-way valve is fixedly installed on the surface of the static pressure detection tube near the output end of the static pressure circulating pump, and one end of the static pressure detection tube penetrates and extends to the inner bottom wall of the circular diverter box. The surface of the static pressure detection tube is respectively provided with a static pressure density detection area and an anti-backflow pressure holding area. A differential pressure density sensor and a sand content sensor are installed in the static pressure density detection area of ​​the static pressure detection tube.

[0010] Preferably, in step two, the spiral blades draw the slurry at the center of the main pipeline through symmetrical sampling slots on the sampling tube at a rate approximately equal to the flow velocity at the center of the main pipeline, and transport it to the circular diversion box at the top; the arc-shaped diversion blades guide the slurry evenly to avoid swirling.

[0011] Preferably, in step three, the continuous dynamic monitoring loop is started by an online circulation pump to pump the mud into the U-shaped online detection tube; the mud circulates in the loop at a flow rate similar to that of the main pipeline; the dynamic density of the mud is measured in real time by a tuning fork density sensor, the apparent viscosity of the mud is measured in real time by an online viscometer, and the sand content of the mud is measured in real time by a sand content sensor. After the measurement is completed, the mud flows back to the main pipeline.

[0012] Preferably, in step three, the static high-precision re-inspection circuit is intermittently started by a static pressure circulation pump to pump a small portion of mud into the static pressure detection tube until the static pressure density detection zone is filled; subsequently, the static pressure circulation pump stops, a one-way valve prevents backflow, and the backflow prevention pressure holding zone ensures that the mud in the detection zone is in a relatively static state; under this static environment, a differential pressure density sensor is used to perform high-precision density measurement under static conditions, which has higher accuracy than the tuning fork sensor, and a sand content sensor is used to measure the sand content under static conditions as a comparison benchmark for sensor one.

[0013] Preferably, during cloud-based intelligent data processing and calibration in step five, the cloud control platform receives the data and executes the following algorithm flow: S1. Data validity verification: Check signal strength and data packet integrity, and remove invalid data. S2. Continuous dynamic monitoring loop data dynamic compensation: The readings of the online viscometer are significantly affected by flow rate and temperature. The compensation formula is as follows:

[0014] Where, η comp (t) represents the compensated viscosity value (mPa·s) at time t; η raw (t) represents the raw reading of the viscosity sensor at time t (mPa·s); v(t) represents the mud flow velocity at time t (m / s), measured or estimated by a flow meter; T(t) represents the mud temperature at time t (°C); f v (v(t)) is the flow velocity compensation function at time t, obtained through experimental calibration; f t (T(t)) is the temperature compensation coefficient at time t; α calThe periodic calibration factor; the initial value of the density / sand content of the continuous dynamic monitoring loop is denoted as p. dy and S dy1 ; Flow rate compensation function:

[0015] When the flow velocity v is within the effective measurement range [vmin, vmax]: Where v is the real-time measured mud flow velocity (unit: m / s); a0, a1, and a2 are compensation coefficients, determined through experimental calibration; f v (v) is the compensation coefficient, used to correct the original measurement value; Temperature compensation coefficient: , where f t (T) is the temperature compensation coefficient, used to correct the original measurement value; T is the current mud temperature; T ref β is the reference temperature; β is the material constant, the mud temperature sensitivity coefficient, which is related to the physicochemical properties of the mud; 273.15 is the conversion factor from Celsius to absolute temperature. S3. Static high-precision re-inspection loop triggering and high-precision benchmark acquisition, calibration triggering strategy: A combination of "timed triggering" and "event triggering" is used to start the static pressure circulation pump for re-inspection; the static high-precision density value p is acquired. st and sand content value S st ; S4. Dynamic calibration and data fusion, including density fusion calibration, sand content fusion calibration and comprehensive confidence assessment; Among them, density fusion calibration involves calculating the dynamic offset of the tuning fork density sensor, and the calibration offset calculation uses... , , where p st ,S st This is a static high-precision measurement value; p dy (t c ),S dy1 (t c ) represents the dynamic measurement value at the calibration trigger time; t c To calibrate the trigger time; The decay weight function is Where k(t) is the decay weight value at time t, ranging from [0,1]; t is the current data acquisition time; t c The time point of the most recent calibration trigger, i.e., the moment of the static high-precision measurement; T decay The decay period is the time required for the weight to decay linearly from 1 to 0; max(0,·) is the maximum value function to ensure that the weight is non-negative. The formula for calculating the density report value is as follows: The formula for the sand content report value is: , where p report (t), S report (t) represents the final reported value at time t; p dy (t), S dy1 (t) represents the real-time measurement value of the dynamic sensor at time t; Δp and ΔS are the calibration offsets, i.e., the difference between the static high-precision value and the dynamic measurement value; k(t) is the decay weighting function, which decays linearly from 1 to 0 over time; Before the next calibration, the real-time density report value of the continuous dynamic monitoring loop will use the calibrated value: , where k(t) is the confidence coefficient that decays over time; Among them, the sand content fusion calibration is the calibration of sand content sensor one: The comprehensive confidence assessment assigns a comprehensive confidence score C (0-1) to the current reported value based on the consistency of the main and static high-precision re-inspection loop data and the sensor's own status. The comprehensive confidence assessment model is as follows: C consisitency For data consistency confidence, the smaller the deviation of the main static high-precision re-inspection loop, the closer the value is to 1, and the parameter change rate is... C sensor The sensor health status is rated (0-1), i.e. The five sensors are a tuning fork density sensor, an online viscometer, a sand content sensor one, a differential pressure density sensor, and a sand content sensor two. If all are functioning normally, the result is 1; C trend For the confidence level of trend stability, The slower the change, the closer the value is to 1; w1, w2, and w3 are weighting coefficients, individual weights, and w1 + w2 + w3 = 1. S5. Intelligent diagnosis and early warning: Establish a mud condition assessment model and integrate diagnostic results and real-time parameters (p) report η comp S report ), confidence level C, and control recommendations are sent to the field PLC controller and remote monitoring terminal; Among them, the mud condition assessment model when "S" report Continuing to rise and η comp When the mud continues to decline, the diagnosis is "insufficient mud carrying capacity, requiring the addition of binder," meaning... When "p report When the level drops abnormally and the confidence level C is high, a warning is issued stating that "the mud may be diluted by groundwater." , where θ S θ η θ p This is the threshold parameter.

[0016] Preferably, the amount of additive added to the dosing system in step six is ​​calculated as follows: Where Qadditive(t) is the rate at which the additive is added at time t; k p k is the proportionality coefficient. i k is the integral coefficient; d These are the differential coefficients; Density deviation; The integral of the density deviation; This represents the rate of change of density deviation.

[0017] The beneficial effects of this invention are as follows: By setting up an online detection mechanism and periodically calibrating the dynamic monitoring loop using a static high-precision loop, the long-term drift and operational interference problems of online sensors are fundamentally solved, ensuring long-term accuracy and reliability of data. Combined with software compensation algorithms for flow velocity and temperature, the impact of drastic fluctuations in the main pipeline flow pattern on the measurement is effectively isolated. The static detection loop completely eliminates interference from dynamic factors, thus solving the problems of poor real-time performance and low efficiency in existing manual detection. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of a real-time intelligent detection method for three parameters of mud in pile foundation construction proposed in this invention. Figure 2 The left view of the connecting pipe structure is a method for real-time intelligent detection of three parameters of mud in pile foundation construction proposed in this invention. Figure 3 The top view of the circular diversion box structure is a method for real-time intelligent detection of three parameters of mud in pile foundation construction proposed in this invention. Figure 4 A three-dimensional view of the sampling drive motor structure for a real-time intelligent detection method for three parameters of mud in pile foundation construction proposed in this invention. Figure 5 This is a flowchart of the cloud-based intelligent data processing and calibration algorithm for a real-time intelligent detection method of three parameters of mud in pile foundation construction proposed in this invention. Figure 6 This is a diagram illustrating the overall architecture and data interaction of a real-time intelligent detection method for three parameters of mud in pile foundation construction proposed in this invention.

[0019] In the diagram: 1. Connecting pipe; 2. Sampling pipe; 3. Circular diverter box; 4. Sealing cover; 5. PLC controller; 6. Wireless communication module; 7. Sampling drive motor; 8. Sampling shaft; 9. Sampling spiral blade; 10. Spiral baffle; 11. Sampling groove; 12. Arc-shaped diverter; 13. Online detection tube; 14. Static pressure detection tube; 15. Online circulation pump; 16. Static pressure circulation pump; 17. Tuning fork density sensor; 18. Online viscometer; 19. Sand content sensor one; 20. Check valve; 21. Differential pressure density sensor; 22. Sand content sensor two. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0021] Reference Figures 1-6 A real-time intelligent detection method for three parameters of mud slurry in pile foundation construction includes the following steps: Step 1: Install an online detection device on the mud circulation pipeline.

[0022] Furthermore, in step one, the online detection mechanism includes a connecting pipe 1, a sampling pipe 2 is fixedly connected to the surface of the connecting pipe 1, a circular diversion box 3 is fixedly connected to the top of the sampling pipe 2, a sealing cover 4 is threadedly connected to the upper surface of the circular diversion box 3, and a PLC controller 5 and a wireless communication module 6 are fixedly installed on the upper surface of the sealing cover 4 respectively. The wireless communication module 6 is connected to the PLC controller 5 via a cable.

[0023] The wireless communication module 6 is connected to a cloud control platform via data connection.

[0024] A sampling drive motor 7 is fixedly installed on the upper surface of the sealing cover 4. The sampling drive motor 7 is electrically connected to the PLC controller 5 via a cable. The output shaft of the sampling drive motor 7 is fixedly connected to a sampling shaft 8 via a coupling. One end of the sampling shaft 8 passes through and extends to the inner wall of the sampling tube 2. A sampling spiral blade 9 is fixedly sleeved on the surface of the sampling shaft 8.

[0025] A spiral baffle 10 is fixedly connected to the surface of the sampling spiral blade 9. Both the surface of the sampling spiral blade 9 and the surface of the spiral baffle 10 are slidably connected to the inner wall of the sampling tube 2. One end of the sampling tube 2 extends to the axis of the connecting tube 1. A sampling groove 11 is opened on the surface of the sampling tube 2 inside the connecting tube 1. The two sampling grooves 11 are symmetrically distributed with the axis of the sampling tube 2 as the center. The inner wall of the sampling groove 11 is connected to the inner wall of the sampling tube 2.

[0026] During use, the sampling drive motor 7 is automatically controlled by the PLC controller 5. The output shaft of the sampling drive motor 7 drives the sampling shaft 8 to rotate through the coupling. The sampling shaft 8 drives the sampling spiral blade 9 to rotate, which rotates the mud inside the connecting pipe 1 into the sampling pipe 2 and then into the circular diversion box 3.

[0027] Two symmetrically distributed arc-shaped diverter plates 12 are fixedly connected to the surface of the sampling shaft 8. The surfaces of the arc-shaped diverter plates 12 are slidably connected to the inner wall of the circular diverter box 3 and the lower surface of the sealing cover 4, respectively.

[0028] During use, the mud enters the circular diversion box 3 through the sampling tube 2 and is then diverted by the arc-shaped diversion plate 12.

[0029] The inner bottom wall of the circular diversion box 3 is fixedly connected to the online detection tube 13 and the static pressure detection tube 14. The online detection tube 13 and the static pressure detection tube 14 form two independent detection circuits: a continuous dynamic monitoring circuit and a static high-precision re-inspection circuit, respectively. The surface of the online detection tube 13 is U-shaped, and one end of the online detection tube 13 penetrates and extends to the inner wall of the connecting tube 1.

[0030] An online circulating pump 15 and a static pressure circulating pump 16 are fixedly installed on the surface of the online detection tube 13 and the static pressure detection tube 14, respectively. Both the online circulating pump 15 and the static pressure circulating pump 16 are electrically connected to the PLC controller 5 via cables.

[0031] In use, the mud inside the circular diversion box 3 is pumped into the online detection pipe 13 by the online circulation pump 15 for detection, and then returned to the connecting pipe 1. The mud inside the circular diversion box 3 is pumped into the static pressure detection pipe 14 by the static pressure circulation pump 16 for static pressure detection.

[0032] A tuning fork density sensor 17, an online viscometer 18, and a sand content sensor 19 are fixedly installed on the surface of the online detection tube 13. The tuning fork density sensor 17, the online viscometer 18, and the sand content sensor 19 are all electrically connected to the PLC controller 5 via cables.

[0033] Furthermore, the tuning fork density sensor is a physical instrument that measures the density of liquids online based on the principle of vibration.

[0034] The online viscometer 18 is an industrial instrument for real-time monitoring of fluid viscosity, supporting a measurement range of 0.1-10,000,000 mPa·s and an applicable temperature range of -40 to 450℃. The device employs either a vibration (amplitude decay method) or rotation (rotor resistance method) principle. The torque micro-oscillation design calculates viscosity by measuring the amplitude decay of spherical resonance, and features low shear rate measurement characteristics.

[0035] The sand content sensor-19 employs an ultrasonic attenuation sensor. When ultrasonic waves pass through mud, their energy is absorbed and scattered by solid particles (especially sand particles), causing them to attenuate. By measuring the degree of attenuation of the ultrasonic waves at the receiving end, the sand content can be calculated, thus enabling online measurement.

[0036] A one-way valve 20 is fixedly installed on the surface of the static pressure detection tube 14 near the output end of the static pressure circulating pump 16. One end of the static pressure detection tube 14 penetrates and extends to the inner bottom wall of the circular diversion box 3.

[0037] The surface of the static pressure detection tube 14 is provided with a static pressure density detection area and an anti-backflow pressure holding area. A differential pressure density sensor 21 and a sand content sensor 22 are installed in the static pressure density detection area of ​​the static pressure detection tube 14.

[0038] Furthermore, a static pressure circulation pump 16 pumps the mud from inside the circular distribution box 3 into the static pressure detection tube 14, and a one-way valve 20 prevents the mud from flowing back into the static pressure detection tube 14. A backflow prevention pressure-maintaining zone is also provided to prevent the mud from the circular distribution box 3 from entering the static pressure density detection zone through the other end of the static pressure detection tube 14. After the mud is pumped into the static pressure detection tube 14 and flows back into the circular distribution box 3 through one end of the static pressure detection tube 14, the static pressure circulation pump 16 stops. Then, the density of the mud is measured by a differential pressure density sensor 21. This is achieved by periodically using a differential pressure density sensor 21 with higher accuracy than the tuning fork density sensor 17 to re-check and calibrate the mud density. Additionally, a sand content sensor 22 is used to re-check the data detected by the sand content sensor 19.

[0039] Step 2: Isokinetic sampling and import. The sampling drive motor 7 is automatically controlled by the PLC controller 5 to drive the sampling shaft 8 and sampling spiral blade 9 to rotate.

[0040] Specifically, in step two, the spiral blades draw the slurry at the center of the main pipeline through the symmetrical sampling grooves 11 on the sampling pipe 2 at a rate approximately equal to the flow velocity at the center of the main pipeline, and then transport it to the circular diversion box 3 at the top. The arc-shaped diversion plate 12 guides the slurry evenly, preventing swirling.

[0041] Step 3: Parallel detection with dual loops. The mud entering the circular diversion box 3 is divided into two independent detection loops: a continuous dynamic monitoring loop and a static high-precision re-inspection loop.

[0042] Specifically, in step three, the continuous dynamic monitoring loop is started by the online circulation pump 15, which pumps the mud into the U-shaped online detection pipe 13. The mud circulates in this loop at a flow rate similar to that of the main pipeline. The dynamic density of the mud is measured in real time by a tuning fork density sensor 17, and the apparent viscosity of the mud is measured in real time by an online viscometer 18. The sand content of the mud is measured in real time by a sand content sensor 19. After the measurements are completed, the mud is returned to the main pipeline.

[0043] In step three, the static high-precision re-inspection circuit is intermittently started by the static pressure circulation pump 16, pumping a small portion of the mud into the static pressure detection tube 14 until the static pressure density detection area is filled. Subsequently, the static pressure circulation pump 16 stops, the one-way valve 20 prevents backflow, and the backflow prevention pressure-holding zone ensures the mud in the detection area remains in a relatively static state. Under this static environment, a differential pressure density sensor 21 performs high-precision density measurement under static conditions, with higher accuracy than the tuning fork sensor. Sand content sensor 22 is used to measure the sand content under static conditions, serving as a comparison benchmark for sensor 1.

[0044] Step 4: Data Acquisition and Upload. PLC controller 5 synchronously acquires raw data from five sensors on two loops and performs preliminary digital filtering and temperature compensation.

[0045] The processed data packet in step four is uploaded to the cloud control platform via wireless communication module 6. The processed data packet contains information such as timestamp, sensor readings, and device status. Wireless communication module 6 uses a 5G communication module.

[0046] Step 5, Cloud-based Intelligent Data Processing and Calibration: During step 5, the cloud control platform receives the data and executes the following algorithm: S1. Data validity verification: Check signal strength and data packet integrity, and remove invalid data.

[0047] S2. Dynamic compensation of continuous dynamic monitoring loop data: The reading of the online viscometer 18 is significantly affected by flow rate and temperature. The compensation formula is as follows: Where, η comp (t) represents the compensated viscosity value (mPa·s) at time t; η raw (t) represents the raw reading of the viscosity sensor at time t (mPa·s); v(t) represents the mud flow velocity at time t (m / s), measured or estimated by a flow meter; T(t) represents the mud temperature at time t (°C); f v (v(t)) is the flow velocity compensation function at time t, obtained through experimental calibration; f t (T(t)) is the temperature compensation coefficient at time t; α cal The periodic calibration factor; the initial value of the density / sand content of the continuous dynamic monitoring loop is denoted as p. dyand S dy1 .

[0048] Flow rate compensation function:

[0049] When the flow velocity v is within the effective measurement range [vmin, vmax]: Where v is the real-time measured mud flow velocity (unit: m / s); a0, a1, and a2 are compensation coefficients, determined through experimental calibration; f v (v) is the compensation coefficient, used to correct the original measurement value.

[0050] Furthermore, a0, a1, and a2 are compensation coefficients, determined through experimental calibration. The experimental steps are as follows: Y1. Prepare standard mud samples and measure their true viscosity η. true .

[0051] Y2, at different flow velocities v i Below, record the sensor's raw reading η. raw (v) i ).

[0052] Y3. Calculate the compensation coefficient for each flow velocity. .

[0053] Y4, for data points (v) i ,f v,i Perform a quadratic polynomial fitting to obtain the coefficients a0, a1, a2.

[0054] a0 is a constant term, representing the theoretical compensation coefficient when the flow velocity is zero; a1 is a linear coefficient, describing the linear rate of change of the compensation coefficient with the flow velocity; a2 is a quadratic coefficient, describing the nonlinear change of the compensation coefficient.

[0055] When the flow rate exceeds the effective range, a conservative boundary value is used, and the flow rate is cut off at a low velocity (v). <v min ), take f v (v) min High-speed truncation (v>v) max ), take f v (v) max ).

[0056] As described above, when in use, the system can prevent the mud from settling and becoming unstable when the mud flow rate is too low, and prevent the generation of bubbles and severe turbulence when the flow rate is too high, thus avoiding unreasonable extreme values ​​generated by polynomial extrapolation. This achieves the effect of preventing control misjudgments caused by abnormal compensation coefficients by using conservative boundary values.

[0057] Temperature compensation coefficient: , where f t(T) is the temperature compensation coefficient, used to correct the original measurement value; T is the current mud temperature; T ref The reference temperature is usually taken as 20℃; β is the material constant, the mud temperature sensitivity coefficient, which is related to the physicochemical properties of the mud; 273.15 is the conversion factor from Celsius temperature to absolute temperature (Kelvin).

[0058] Based on the Arrhenius equation, the effect of temperature on mud viscosity / density is quantified. When T>T, f T If (T) < 1, it can correct the problem of "inflated viscosity due to increased temperature"; conversely, it can correct the inflated viscosity and ensure that viscosity data at different temperatures are comparable.

[0059] S3. Static high-precision re-inspection loop triggering and high-precision benchmark acquisition, calibration triggering strategy: A combination of "timed triggering" (every 30 minutes) and "event triggering" (continuous dynamic monitoring loop data mutation exceeding a threshold) is used to start the static pressure circulation pump 16 for re-inspection. The static high-precision density value p is acquired. st and sand content value S st .

[0060] S4. Dynamic calibration and data fusion, including density fusion calibration, sand content fusion calibration and comprehensive confidence assessment.

[0061] Among them, density fusion calibration involves calculating the dynamic offset of the tuning fork density sensor, and the calibration offset calculation uses... , , where p st ,S st This is a static high-precision measurement value; p dy (t c ),S dy1 (t c ) represents the dynamic measurement value at the calibration trigger time; t c To calibrate the trigger time.

[0062] The decay weight function is Where k(t) is the decay weight value at time t, ranging from [0,1]; t is the current data acquisition time; t c The time point of the most recent calibration trigger, i.e., the moment of the static high-precision measurement; T decay The decay period is the time required for the weight to decay linearly from 1 to 0; max(0,·) is the maximum value function to ensure that the weight is non-negative.

[0063] When used, this function causes the "static calibration correction" to decay linearly over time, i.e., t=t c When k=1, t=t c +T decayWhen k=0, it achieves the effect of balancing real-time performance and accuracy by utilizing the high precision of static calibration and avoiding errors caused by expired calibration.

[0064] The formula for calculating the density report value is as follows: The formula for the sand content report value is: , where p report (t), S report (t) represents the final reported value at time t; p dy (t), S dy1 (t) represents the real-time measurement value of the dynamic sensor at time t; Δp and ΔS are the calibration offsets, i.e., the difference between the static high-precision value and the dynamic measurement value; k(t) is the decay weighting function, which decays linearly from 1 to 0 over time.

[0065] Specifically, p report (t) represents the final reported density of the mud at time t, S report (t) represents the reported final sand content of the mud at time t; p dy (t) represents the measurement value of the tuning fork density sensor 17 in the continuous dynamic monitoring loop at time t; S dy1 (t) represents the measured value of the sand content sensor-19 in the continuous dynamic monitoring loop at time t; Δp is the density calibration offset; ΔS is the sand content calibration offset. Before the next calibration, the real-time density report value of the continuous dynamic monitoring loop will use the calibrated value: , where k(t) is the confidence coefficient that decays over time (linearly decays to 0).

[0066] Among them, the sand content fusion calibration is the calibration of the sand content sensor-19: .

[0067] The comprehensive confidence assessment assigns a comprehensive confidence score C (0-1) to the current reported value based on the consistency of the main and static high-precision re-inspection loop data and the sensor's own status.

[0068] The comprehensive confidence assessment model is C consisitency For data consistency confidence, the smaller the deviation of the main static high-precision re-inspection loop, the closer the value is to 1, and the parameter change rate is... C sensor The sensor health status is rated (0-1), i.e. The five sensors are a tuning fork density sensor 17, an online viscometer 18, a sand content sensor 19, a differential pressure density sensor 21, and a sand content sensor 22. If all are functioning normally, the result is 1; C trend For the confidence level of trend stability, The slower the change, the closer the value is to 1; w1, w2, w3 are weighting coefficients, the weights of each item, w1+w2+w3=1.

[0069] S5. Intelligent diagnosis and early warning: Establish a mud condition assessment model and integrate diagnostic results and real-time parameters (p) report η comp S report The confidence level C and control recommendations are sent to the field PLC controller 5 and the remote monitoring terminal.

[0070] Among them, the mud condition assessment model when "S" report Continuing to rise and η comp When the mud continues to decline, the diagnosis is "insufficient mud carrying capacity, requiring the addition of binder," meaning... When "p report When the level drops abnormally and the confidence level C is high, a warning is issued stating that "the mud may be diluted by groundwater." , where θ S θ η θ p For threshold parameters; θ C The confidence threshold is set to 0.8; θ S Threshold for sand content increase; θ η θ is the viscosity decrease threshold. p This represents the density decrease threshold. This represents the rate of change of the reported sand content value. To compensate for the rate of change in viscosity; This represents the rate of change of the reported density values.

[0071] Step Six: On-site Execution and Display. The on-site touchscreen displays all information sent from the cloud in real time. Upon receiving control recommendations, the corresponding operations can be executed manually or automatically through the dosing system, completing the closed loop from detection and analysis to control.

[0072] Specifically, the amount of additive to be added in the dosing system in step six is ​​calculated as follows: Where Qadditive(t) is the rate at which the additive is added at time t; k p k is the proportionality coefficient. i k is the integral coefficient; d These are the differential coefficients; Density deviation; The integral of the density deviation; This represents the rate of change of density deviation.

[0073] When using it, the proportional term (k) p The larger the deviation of ΔP, the faster the addition rate, and the faster the response to deviation; the integral term ( If a small deviation persists for a long period, the cumulative integral will gradually increase the addition rate until the deviation is eliminated; the differential term (k d If the deviation decreases rapidly, the differential term will reduce the addition rate to avoid "over-addition leading to density overshoot" and ensure stable adjustment.

[0074] By setting up an online detection mechanism and periodically calibrating the dynamic monitoring loop using a static high-precision loop, the long-term drift and operational interference problems of online sensors are fundamentally solved, ensuring long-term accuracy and reliability of data. Combined with software compensation algorithms for flow velocity and temperature, the impact of drastic fluctuations in the main pipeline flow pattern on the measurement is effectively isolated. The static detection loop completely eliminates interference from dynamic factors, thus solving the problems of poor real-time performance and low efficiency in existing manual detection.

[0075] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for real-time intelligent detection of three parameters of mud slurry in pile foundation construction, characterized in that, Includes the following steps: Step 1: Install an online detection mechanism on the mud circulation pipeline; The online detection mechanism in step one includes a connecting pipe (1), a sampling pipe (2) is fixedly connected to the surface of the connecting pipe (1), a circular diversion box (3) is fixedly connected to the top of the sampling pipe (2), a sealing cover (4) is threadedly connected to the upper surface of the circular diversion box (3), and a PLC controller (5) and a wireless communication module (6) are fixedly installed on the upper surface of the sealing cover (4). The wireless communication module (6) is connected to the PLC controller (5) via a cable. The wireless communication module (6) is connected to a cloud control platform via a data connection; Step 2, constant velocity sampling and import: The sampling drive motor (7) is automatically controlled by the PLC controller (5) to drive the sampling shaft (8) and sampling spiral blade (9) to rotate. Step 3: Parallel detection with dual loops. The mud entering the circular diversion box (3) is divided into two independent detection loops: a continuous dynamic monitoring loop and a static high-precision re-inspection loop. Step 4: Data Acquisition and Upload. The PLC controller (5) synchronously acquires the raw data from five sensors on two loops and performs preliminary digital filtering and temperature compensation. Step 5: Cloud-based intelligent data processing and calibration; Step Six: On-site Execution and Display. The on-site touchscreen displays all information sent from the cloud in real time. When a control suggestion is received, the corresponding operation can be performed manually or automatically through the dosing system to complete the closed loop from detection and analysis to control.

2. The method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 1, characterized in that: A sampling drive motor (7) is fixedly installed on the upper surface of the sealing cover (4). The sampling drive motor (7) is electrically connected to the PLC controller (5) via a cable. The output shaft of the sampling drive motor (7) is fixedly connected to a sampling shaft (8) via a coupling. One end of the sampling shaft (8) passes through and extends to the inner wall of the sampling tube (2). A sampling spiral blade (9) is fixedly sleeved on the surface of the sampling shaft (8). A spiral baffle (10) is fixedly connected to the surface of the sampling spiral blade (9). The surfaces of the sampling spiral blade (9) and the spiral baffle (10) are slidably connected to the inner wall of the sampling tube (2). One end of the sampling tube (2) extends to the axis of the connecting tube (1). A sampling groove (11) is opened on the surface of the sampling tube (2) inside the connecting tube (1). The two sampling grooves (11) are symmetrically distributed with the axis of the sampling tube (2) as the center. The inner wall of the sampling groove (11) is connected to the inner wall of the sampling tube (2).

3. The method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 2, characterized in that: Two symmetrically distributed arc-shaped diverter plates (12) are fixedly connected to the surface of the sampling shaft (8). The surfaces of the arc-shaped diverter plates (12) are slidably connected to the inner wall of the circular diverter box (3) and the lower surface of the sealing cover (4), respectively. The inner bottom wall of the circular diversion box (3) is fixedly connected to an online detection tube (13) and a static pressure detection tube (14). The online detection tube (13) and the static pressure detection tube (14) form two independent detection circuits: a continuous dynamic monitoring circuit and a static high-precision re-inspection circuit, respectively. The surface of the online detection tube (13) is U-shaped, and one end of the online detection tube (13) extends through and to the inner wall of the connecting tube (1). An online circulating pump (15) and a static pressure circulating pump (16) are fixedly installed on the surface of the online detection tube (13) and the surface of the static pressure detection tube (14), respectively. The online circulating pump (15) and the static pressure circulating pump (16) are electrically connected to the PLC controller (5) through cables.

4. The method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 3, characterized in that: The online detection tube (13) is fixedly mounted with a tuning fork density sensor (17), an online viscometer (18), and a sand content sensor (19). The tuning fork density sensor (17), the online viscometer (18), and the sand content sensor (19) are all electrically connected to the PLC controller (5) via cables.

5. The method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 4, characterized in that: A one-way valve (20) is fixedly installed on the surface of the static pressure detection tube (14) near the output end of the static pressure circulating pump (16), and one end of the static pressure detection tube (14) penetrates and extends to the inner bottom wall of the circular diversion box (3). The surface of the static pressure detection tube (14) is provided with a static pressure density detection area and an anti-backflow pressure holding area. A differential pressure density sensor (21) and a sand content sensor (22) are installed in the static pressure density detection area of ​​the static pressure detection tube (14).

6. The method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 5, characterized in that: In step two, the spiral blades draw out the mud at the center of the main pipeline through the symmetrical sampling grooves (11) on the sampling tube (2) at a rate close to the flow velocity at the center of the main pipeline, and transport it to the circular diversion box (3) at the top; the arc-shaped diversion plate (12) guides the mud evenly to avoid swirling.

7. The method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 6, characterized in that: In step three, the continuous dynamic monitoring loop is started by an online circulation pump (15) to pump the mud into the U-shaped online detection tube (13). The mud is kept circulating in the loop at a flow rate similar to that of the main pipeline. The dynamic density of the mud is measured in real time by a tuning fork density sensor (17), the apparent viscosity of the mud is measured in real time by an online viscometer (18), and the sand content of the mud is measured in real time by a sand content sensor (19). After the measurement is completed, the mud is returned to the main pipeline.

8. The method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 7, characterized in that: In step three, the static high-precision re-inspection circuit is started intermittently by the static pressure circulation pump (16) to pump a small portion of mud into the static pressure detection tube (14) until the static pressure density detection area is filled. Then, the static pressure circulation pump (16) stops, the check valve (20) prevents backflow, and the backflow prevention pressure holding area ensures that the mud in the detection area is in a relatively static state. Under this static environment, the differential pressure density sensor (21) is used to perform high-precision density measurement under static conditions. Its accuracy is higher than that of the tuning fork sensor. The sand content sensor (22) is used to perform sand content measurement under static conditions as a comparison benchmark for sensor one.

9. The method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 8, characterized in that: In step five, during cloud-based intelligent data processing and calibration, the cloud control platform receives the data and executes the following algorithm: S1. Data validity verification: Check signal strength and data packet integrity, and remove invalid data. S2. Dynamic compensation of continuous dynamic monitoring loop data. The reading of the online viscometer (18) is significantly affected by flow rate and temperature. The compensation formula is as follows: Where, η comp (t) represents the compensated viscosity value (mPa·s) at time t; η raw (t) represents the raw reading of the viscosity sensor at time t (mPa·s); v(t) represents the mud flow velocity at time t (m / s), measured or estimated by a flow meter; T(t) represents the mud temperature at time t (°C); f v (v(t)) is the flow velocity compensation function at time t, obtained through experimental calibration; f t (T(t)) is the temperature compensation coefficient at time t; α cal The periodic calibration factor; the initial value of the density / sand content of the continuous dynamic monitoring loop is denoted as p. dy and S dy1 ; Flow rate compensation function: When the flow velocity v is within the effective measurement range [vmin, vmax]: Where v is the real-time measured mud flow velocity (unit: m / s); a0, a1, and a2 are compensation coefficients, determined through experimental calibration; f v (v) is the compensation coefficient, used to correct the original measurement value; Temperature compensation coefficient: , where f t (T) is the temperature compensation coefficient, used to correct the original measurement value; T is the current mud temperature; T ref β is the reference temperature; β is the material constant, the mud temperature sensitivity coefficient, which is related to the physicochemical properties of the mud; 273.15 is the conversion factor from Celsius to absolute temperature. S3. Static high-precision re-inspection loop triggering and high-precision benchmark acquisition, calibration triggering strategy: adopt a combination of "timed triggering" and "event triggering" to start the static pressure circulation pump (16) for re-inspection; acquire the static high-precision density value p st and sand content value S st ; S4. Dynamic calibration and data fusion, including density fusion calibration, sand content fusion calibration and comprehensive confidence assessment; Among them, density fusion calibration involves calculating the dynamic offset of the tuning fork density sensor, and the calibration offset calculation uses... , , where p st ,S st This is a static high-precision measurement value; p dy (t c ),S dy1 (t c ) represents the dynamic measurement value at the calibration trigger time; t c To calibrate the trigger time; The decay weighting function is Where k(t) is the decay weight value at time t, ranging from [0,1]; t is the current data acquisition time; t c The time point of the most recent calibration trigger, i.e., the moment of the static high-precision measurement; T decay The decay period is the time required for the weight to decay linearly from 1 to 0; max(0,·) is the maximum value function to ensure that the weight is non-negative. The formula for calculating the density report value is as follows: The formula for the sand content report value is: , where p report (t), S report (t) represents the final reported value at time t; p dy (t), S dy1 (t) represents the real-time measurement value of the dynamic sensor at time t; Δp and ΔS are the calibration offsets, i.e., the difference between the static high-precision value and the dynamic measurement value; k(t) is the decay weighting function, which decays linearly from 1 to 0 over time; Before the next calibration, the real-time density report value of the continuous dynamic monitoring loop will use the calibrated value: , where k(t) is the confidence coefficient that decays over time; Among them, the sand content fusion calibration is to calibrate the sand content sensor (19): ; The comprehensive confidence assessment assigns a comprehensive confidence score C (0-1) to the current reported value based on the consistency of the main and static high-precision re-inspection loop data and the sensor's own status. The comprehensive confidence assessment model is as follows: C consisitency For data consistency confidence, the smaller the deviation of the main static high-precision re-inspection loop, the closer the value is to 1, and the parameter change rate is... C sensor The sensor health status is rated (0-1), i.e. The five sensors are a tuning fork density sensor (17), an online viscometer (18), a sand content sensor one (19), a differential pressure density sensor (21), and a sand content sensor two (22). If all are normal, the value is 1; C trend For the confidence level of trend stability, The slower the change, the closer the value is to 1; w1, w2, and w3 are weighting coefficients, individual weights, and w1 + w2 + w3 = 1. S5. Intelligent diagnosis and early warning: Establish a mud condition assessment model and integrate diagnostic results and real-time parameters (p) report η comp S report ), confidence level C and control recommendations, and send them to the field PLC controller (5) and remote monitoring terminal; Among them, the mud state assessment model when "S report Continuing to rise and η comp When the mud continues to decline, the diagnosis is "insufficient mud carrying capacity, requiring the addition of binder," meaning... When "p report When the level is abnormally low and confidence level C is high, an early warning is issued stating that "the mud may be diluted by groundwater." , where θ S θ η θ p This is the threshold parameter.

10. A method for real-time intelligent detection of three parameters of mud in pile foundation construction according to claim 9, characterized in that: The calculation of the additive amount in step six of the dosing system is as follows: Where Qadditive(t) is the rate at which the additive is added at time t; k p k is the proportionality coefficient. i k is the integral coefficient; d These are the differential coefficients; Density deviation; The integral of the density deviation; This represents the rate of change of density deviation.

Images