Static pressure pipe pile body stress real-time monitoring method and system

Abstract

The application provides a static pressure pipe pile body stress real-time monitoring method and system, which realizes high-quality data efficient transmission and processing through systematic arrangement of sensing units based on geological layering and combination of in-situ calibration output standardized digital signal flow; through signal flow execution of abnormal elimination, baseline correction and self-adaptive packaging scheduling, realizes high-quality data efficient transmission and processing; through multi-sensor time synchronization and continuous internal force curve reconstruction, discrete observation is converted into full pile body continuous internal force distribution and abnormal section is accurately positioned; the damage degree of the quantitative joint and the body is quantified, and the local integrity index is combined to inversely calculate the side friction resistance, the separation end resistance and the bearing capacity time correction, finally generate the comprehensive safety coefficient and the construction disposal instruction, form a complete real-time closed loop from perception, analysis to decision execution, so that the monitoring result can timely reflect the real stress state and bearing performance evolution of the pile body, and the timeliness and accuracy of the static pressure pipe pile construction quality control are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of pipe pile monitoring technology, specifically to a method and system for real-time monitoring of static pressure pipe pile stress. Background Technology

[0002] In the construction of static pressure pipe piles, real-time monitoring of pile stress is a core technical means to evaluate the quality of pile driving and the bearing capacity of the pile. Existing technologies typically collect stress data by pre-embedding strain sensors inside the pipe pile and transmitting the signals to ground-based receiving equipment for processing and analysis via wired or wireless transmission. While this approach can obtain pile stress information within a certain depth range, sensor deployment often relies on engineering experience and lacks a systematic correlation with geological stratification. Furthermore, the raw signals acquired are susceptible to interference from various factors such as construction vibration and temperature drift, making it difficult to guarantee the continuity and accuracy of the monitoring data.

[0003] Existing monitoring methods mostly focus on displaying stress values ​​at a single cross-section or performing post-hoc offline analysis, and have not yet established a complete real-time processing chain from data cleaning and internal force reconstruction to damage identification and bearing capacity extrapolation. Especially under complex geological conditions, there is a lack of effective interpolation reconstruction methods for the internal force distribution between discrete observation nodes, insufficient accuracy in locating abnormal sections, and an inability to dynamically and quantitatively assess the time-dependent changes in pile integrity damage types and bearing capacity by combining pile driving process parameters. Ultimately, this results in monitoring results lagging behind the actual construction progress, making it difficult to provide timely guidance for on-site construction and disposal decisions. Summary of the Invention

[0004] The present invention aims to provide a method and system for real-time monitoring of stress in static pressure pipe piles, which can reflect the actual stress state and bearing capacity evolution of the pile body in real time, and significantly improve the timeliness and accuracy of static pressure pipe pile construction quality control.

[0005] To achieve the above objectives, the technical solution adopted by this invention is: a method for real-time monitoring of static pressure pipe pile stress, comprising: The encapsulated sensing unit is implanted inside the pipe pile. The number of monitoring sections is determined based on geological stratification, and the sensing units are evenly arranged along the circumference of each section. The embedded sensing unit is subjected to in-situ signal conditioning and calibration, and a standardized digital signal stream is output. The system receives the standardized digital signal stream, performs outlier removal and baseline drift correction to obtain the corrected axial force value, extracts feature vectors and statistical summaries based on the corrected axial force value, performs adaptive data encapsulation and priority scheduling, and outputs a structured feature data packet. Receive the structured feature data packet, perform multi-sensor time synchronization correction, reconstruct the continuous pile internal force distribution curve based on the time-aligned data, locate abnormal sections based on the continuous pile internal force distribution curve, and complete three-dimensional visualization mapping and structured storage. Based on the continuous pile internal force distribution curve and the abnormal section, the joint damage index is calculated and the pile body damage type is identified. The degree of damage is quantified to obtain the overall pile integrity score and the local integrity index of each section. Based on the local integrity index of each section and the internal force distribution curve of the continuous pile body, the unit side friction resistance after correction is back-calculated and the effective end resistance is separated. The bearing capacity is then corrected by the soil consolidation parameters to obtain the equivalent static bearing capacity. The comprehensive safety factor and structural reliability reduction factor are then output. Based on the comprehensive safety factor, the structural reliability reduction factor, and the comprehensive score of the whole pile integrity, a construction disposal instruction is generated and pushed to the construction site terminal, an electronic file is automatically generated, and a group quality trend analysis is performed.

[0006] Preferably, the step of embedding the encapsulated sensing unit inside the pipe pile includes: A multi-level thermal protection packaging structure is constructed. A polyimide flexible circuit board is selected as the signal acquisition substrate. The strain-sensitive grid wire and the temperature compensation resistor are soldered to the signal acquisition substrate in a full-bridge manner. An organosilicon-modified epoxy resin adhesive layer is coated on the circuit surface as primary insulation protection. A groove-shaped shell is formed by stamping titanium alloy sheets. The coated polyimide flexible circuit board is placed inside the groove-shaped shell. After the alumina ceramic filler is injected and cured, a sealed cavity is formed, resulting in a sensing unit whose size is controlled within a preset range.

[0007] Preferably, the step of determining the number of monitoring sections based on geological stratification and uniformly arranging the sensing units along the circumferential direction in each section includes: The number of monitoring sections is determined based on the number of soil layers penetrated by the pile as revealed in the geological survey report, so that each soil layer segment has no less than two effective sensing sections. Four sensing units are evenly arranged along the circumference of each monitoring section. The sensing units are fastened to the inside of the main reinforcement using stainless steel clamps. The lead wire is laid longitudinally along the inside of the main reinforcement to the reserved wiring cavity at the pile end and fixed. Quick-drying cement slurry is applied to the outer surface of the sensing unit to ensure deformation coordination with the pile body material.

[0008] Preferably, the in-situ signal conditioning and calibration of the embedded sensing unit includes: Record the initial output voltage value of each channel under no external load conditions and the ambient temperature of the pile body. After applying the standard preload, record the change in output voltage, calculate the actual sensitivity coefficient of each sensing channel and write it into the non-volatile memory. The on-chip analog-to-digital converter is enabled to set the oversampling rate. A digital low-pass filter is used to filter out high-frequency noise. The conditioned digital signal is packaged according to a predefined frame format to generate the standardized digital signal stream. Each frame contains a channel number, timestamp, original code value, temperature value, and checksum field.

[0009] Preferably, the step of performing outlier removal and baseline drift correction to obtain the corrected axial force value includes: The latest sampled values ​​are stored in a ring buffer in real time. The absolute deviation between each new data point and the median of the historical data in the window is calculated. When the absolute deviation exceeds a preset threshold, it is judged as an outlier and replaced with a linear interpolation result. When multiple consecutive data points are judged as outliers, the sensor fault flag is triggered. During the interval between two pile driving actions, the average data value is used as the dynamic zero point. During the dynamic loading segment of continuous pile driving, the temperature compensation is superimposed based on the dynamic zero point, and the corrected axial force value is calculated in combination with the actual sensitivity coefficient.

[0010] Preferably, the step of extracting feature vectors and statistical summaries based on the corrected axial force values ​​includes: Three types of triggering events are defined: extreme axial force events, rate change events, and periodic integrity events. Each time an event is triggered, a feature vector containing the event type code, occurrence time, peak axial force, valley axial force, average axial force, maximum axial force change rate, and duration is extracted. During non-event periods, a statistical summary containing the mean, variance, sample count, and a globally monotonically increasing sequence number is reported at a preset frequency. The globally monotonically increasing sequence number is uniformly allocated by the edge nodes to eliminate ambiguity in multi-channel synchronization.

[0011] Preferably, the reconstructing of the internal force distribution curve of the continuous pile body based on time-aligned data includes: Two-way message exchange and linear regression fitting are used to correct the timestamps of the data reported by each edge node, so that the data of each sensing unit can achieve millisecond-level time consistency. The pile body is divided into soil layers according to the geological survey report. Within each segment, the corrected axial force values ​​of two adjacent sensing sections are used as interpolation endpoints. Piecewise linear spatial interpolation is performed at any depth position to obtain a continuous pile body internal force distribution curve.

[0012] Preferably, locating abnormal sections based on the continuous pile internal force distribution curve includes: Calculate the first derivative of the internal force distribution curve of the continuous pile body. When the absolute value of the first derivative deviates from the expected lateral resistance range of the soil layer by more than a preset proportion, it is marked as a candidate anomaly point. The fluctuation energy of the internal force curve is calculated within a preset length window centered on the candidate anomaly point. When the ratio of the fluctuation energy to the average energy of the entire pile exceeds a preset multiple, it is confirmed as an effective anomaly section, and the start and end depths and severity index are recorded.

[0013] Preferably, the calculation of the joint damage index and identification of the pile body damage type includes: The measured transfer coefficient is calculated by extracting the axial force of the upstream section and the axial force of the downstream section of the joint in the abnormal section. The transfer coefficient is then established in conjunction with the historical maximum clamping force. The joint damage index is obtained based on the ratio of the measured transfer coefficient to the associated benchmark. Check whether the abnormal section is located at the soil layer interface. If not, extract the local gradient deviation and energy attenuation rate. Based on the combination relationship of the local gradient deviation, energy attenuation rate and joint damage index, determine the damage type of transverse crack, necking, segregation or longitudinal microcrack.

[0014] On the other hand, the present invention proposes a real-time monitoring system for static pressure pipe pile stress, comprising: The sensor integration module is used to embed the packaged sensor unit inside the pipe pile, determine the number of monitoring sections based on geological stratification, and evenly arrange the sensor unit along the circumferential direction in each section. The embedded sensor unit is subjected to in-situ signal conditioning and calibration, and a standardized digital signal stream is output. The edge processing module is used to receive the standardized digital signal stream output by the sensing integration module, perform outlier removal and baseline drift correction to obtain the corrected axial force value, extract feature vectors and statistical summaries based on the corrected axial force value, perform adaptive data encapsulation and priority scheduling, and output structured feature data packets. The cloud reconstruction module is used to receive the structured feature data packets output by the edge processing module, perform multi-sensor time synchronization correction, reconstruct the internal force distribution curve of the continuous pile body based on the time-aligned data, locate abnormal sections according to the internal force distribution curve of the continuous pile body, and complete three-dimensional visualization mapping and structured storage. The integrity discrimination module is used to calculate the joint damage index and identify the damage type of the pile body based on the continuous pile internal force distribution curve and the abnormal section output by the cloud reconstruction module, and to quantify the degree of damage to obtain the overall integrity score of the whole pile and the local integrity index of each section. The bearing capacity derivation module is used to back-calculate the corrected unit side friction and separate the effective end resistance based on the local integrity index of each section output by the integrity discrimination module and the internal force distribution curve of the continuous pile body output by the cloud reconstruction module. It then combines the soil consolidation parameters to perform bearing capacity time correction to obtain the equivalent static bearing capacity and outputs the comprehensive safety factor and structural reliability reduction factor. The decision output module is used to generate construction disposal instructions and push them to the construction site terminal based on the comprehensive safety factor output by the bearing capacity deduction module, the structural reliability reduction factor, and the comprehensive integrity score of the whole pile output by the integrity discrimination module. It also automatically generates electronic files and performs group quality trend analysis.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention ensures the representativeness and accuracy of data acquisition by systematically deploying sensing units based on geological stratification and combining in-situ calibration to output standardized digital signal streams. Through anomaly removal, baseline correction, and adaptive encapsulation scheduling of the signal stream, it achieves efficient transmission and processing of high-quality data. By multi-sensor time synchronization and continuous internal force curve reconstruction, discrete observations are transformed into a continuous internal force distribution across the entire pile, accurately locating abnormal sections. Furthermore, it quantifies the degree of joint and body damage, and combines local integrity index to back-calculate side friction and separation end resistance, and performs bearing capacity time-based correction, ultimately generating a comprehensive safety factor and construction handling instructions. This forms a complete real-time closed loop from perception and analysis to decision execution, enabling monitoring results to instantly reflect the true stress state and bearing capacity evolution of the pile body, significantly improving the timeliness and accuracy of static pressure pipe pile construction quality control. Attached Figure Description

[0016] Figure 1 This is a flowchart of the method for real-time monitoring of static pressure pipe pile stress according to the present invention; Figure 2 This is a block diagram of the static pressure pipe pile real-time monitoring system of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be noted that if the embodiments of the present invention involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a specific posture. If the specific posture changes, the directional indicators will also change accordingly.

[0019] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the use of "and / or" or "and / or" throughout the text includes three parallel solutions. For example, "A and / or B" includes solution A, solution B, or a solution where both A and B are satisfied simultaneously. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0020] Please see Figure 1 This invention provides a method for real-time monitoring of stress in static pressure pipe piles, comprising: The encapsulated sensing unit is implanted inside the pipe pile. The number of monitoring sections is determined based on geological stratification, and sensing units are evenly arranged along the circumference of each section. The embedded sensing unit is subjected to in-situ signal conditioning and calibration, and a standardized digital signal stream is output. The process of embedding the encapsulated sensing unit inside the pipe pile includes: constructing a multi-layer thermal protection encapsulation structure; selecting a polyimide flexible circuit board as the signal acquisition substrate; welding the strain-sensitive grid wire and temperature compensation resistor to the signal acquisition substrate in a full-bridge configuration; coating the circuit surface with an organosilicon-modified epoxy resin adhesive layer as primary insulation protection; stamping a slotted shell using a titanium alloy sheet; placing the coated polyimide flexible circuit board into the slotted shell; and filling it with alumina ceramic filler adhesive, which, after curing, forms a sealed cavity, thus obtaining a sensing unit whose size is controlled within a preset range.

[0021] Furthermore, the number of monitoring sections is determined based on geological stratification, and sensing units are evenly arranged along the circumference of each section. This includes: determining the number of monitoring sections based on the number of soil layers the pile penetrates as revealed in the geological survey report, so that each soil layer segment has no less than two effective sensing sections; evenly arranging four sensing units along the circumference of each monitoring section, using stainless steel clamps to fasten the sensing units to the inside of the main reinforcement, laying the lead wire longitudinally along the inside of the main reinforcement to the reserved wiring cavity at the pile end and fixing it, and applying quick-drying cement slurry to the outer surface of the sensing unit to ensure deformation coordination with the pile material.

[0022] In-situ signal conditioning and calibration of the embedded sensing units include: recording the initial output voltage value of each channel under no external load conditions and the ambient temperature of the pile body; recording the output voltage change after applying standard preload; calculating the actual sensitivity coefficient of each sensing channel and writing it into non-volatile memory; enabling the on-chip analog-to-digital converter to set the oversampling rate; using a digital low-pass filter to filter out high-frequency noise; and packaging the conditioned digital signal into a standardized digital signal stream according to a predefined frame format, with each frame containing the channel number, timestamp, original code value, temperature value, and checksum field.

[0023] The multi-layer thermal protection encapsulation structure ensures the long-term stability and sealing reliability of the sensing unit under high temperature and high pressure pile driving environment. The cross-section layout strategy based on geological stratification and the circumferential uniform arrangement method ensure the spatial representativeness and deformation coordination of the internal force acquisition of each soil layer segment. In-situ calibration and in-chip oversampling filtering are used to eliminate individual differences and high frequency noise interference.

[0024] The system receives a standardized digital signal stream, performs outlier removal and baseline drift correction to obtain the corrected axial force value, extracts feature vectors and statistical summaries based on the corrected axial force value, performs adaptive data encapsulation and priority scheduling, and outputs a structured feature data packet. The process of removing outliers and correcting baseline drift to obtain the corrected axial force value includes: maintaining a ring buffer to store the latest sampled values ​​in real time; calculating the absolute deviation between each new data point and the median of historical data within the window; when the absolute deviation exceeds a preset threshold, it is judged as an outlier and replaced with a linear interpolation result; when multiple consecutive data points are judged as outliers, the sensor fault flag is triggered; during the interval between two pile driving actions, the average data is used as the dynamic zero point; during the dynamic loading segment of continuous pile driving, the temperature compensation is superimposed based on the dynamic zero point, and the corrected axial force value is calculated in combination with the actual sensitivity coefficient.

[0025] Furthermore, feature vectors and statistical summaries are extracted based on the corrected axial force values, including: defining three types of triggering events: axial force extreme value events, rate change events, and periodic integrity events; extracting feature vectors containing event type codes, occurrence times, peak axial force, valley axial force, average axial force, maximum axial force change rate, and duration for each event trigger; and reporting statistical summaries containing mean, variance, sample count, and globally monotonically increasing sequence number at a preset frequency during non-event periods. The globally monotonically increasing sequence number is uniformly assigned by edge nodes to eliminate ambiguity in multi-channel synchronization.

[0026] Real-time identification and robust replacement of outliers are achieved through a circular buffer and median absolute deviation detection. Combined with dynamic zero-point updates and temperature compensation during intermittent periods, the effects of baseline drift and temperature drift are eliminated, ensuring the long-term accuracy of axial force data. At the same time, key feature vectors are adaptively extracted based on an event-triggered mechanism and statistical summaries are compressed and reported during non-event periods. In conjunction with a globally monotonically increasing sequence number, multi-channel synchronization ambiguity is eliminated. While ensuring the complete preservation of key mechanical information, the data transmission load is significantly reduced and the timing consistency is improved.

[0027] Receive structured feature data packets, perform multi-sensor time synchronization correction, reconstruct the internal force distribution curve of the continuous pile body based on time-aligned data, locate abnormal sections based on the internal force distribution curve of the continuous pile body, and complete three-dimensional visualization mapping and structured storage. The process of reconstructing the internal force distribution curve of the continuous pile body based on time-aligned data includes: correcting the timestamps of the data reported by each edge node by using bidirectional message exchange and linear regression fitting to achieve millisecond-level time consistency of the data of each sensing unit; dividing the pile body into soil layers according to the geological survey report, and using the corrected axial force values ​​of two adjacent sensing sections as interpolation endpoints within each segment, performing piecewise linear spatial interpolation at any depth position to obtain the internal force distribution curve of the continuous pile body.

[0028] The abnormal section is located based on the internal force distribution curve of the continuous pile body, including: calculating the first derivative of the internal force distribution curve of the continuous pile body, and marking it as a candidate abnormal point when the absolute value of the first derivative deviates from the expected lateral resistance range of the soil layer by more than a preset proportion; calculating the fluctuation energy of the internal force curve within a preset length window centered on the candidate abnormal point, and confirming it as a valid abnormal section when the ratio of the fluctuation energy to the average energy of the entire pile exceeds a preset multiple, and recording the start and end depths and severity index.

[0029] By eliminating data time-series deviations through multi-sensor millisecond-level time synchronization correction, and combining piecewise linear spatial interpolation based on geological stratification to reconstruct the axial force of discrete sections into a continuous pile internal force distribution curve, the problem of missing internal force information between observation points in traditional monitoring is overcome. Furthermore, by using the dual criteria of first-order derivative deviation degree and fluctuation energy ratio, abnormal sections are accurately identified and the severity is quantified, achieving high-fidelity restoration and anomaly location from discrete data to the continuous mechanical state of the entire pile.

[0030] Based on the continuous pile internal force distribution curve and abnormal sections, the joint damage index is calculated and the pile body damage type is identified. The degree of damage is quantified to obtain the overall pile integrity score and the local integrity index of each section. Specifically, the joint damage index is calculated and the type of pile body damage is identified, including: extracting the axial force of the upstream and downstream sections of the joint in the abnormal section to calculate the measured transfer coefficient, establishing a transfer coefficient correlation benchmark in combination with the historical maximum clamping force, and obtaining the joint damage index based on the ratio of the measured transfer coefficient to the correlation benchmark; checking whether the abnormal section is located at the soil layer interface, if not, extracting the local gradient deviation and energy attenuation rate, and judging the damage type of transverse cracks, necking, segregation or longitudinal microcracks based on the combination relationship of local gradient deviation, energy attenuation rate and joint damage index.

[0031] By establishing a benchmark for the joint transfer coefficient, the degree of joint damage is quantitatively characterized. Combined with multi-dimensional features such as soil layer location, local gradient deviation, and energy attenuation rate, the type of pile body damage is determined, overcoming the limitations of traditional methods that can only make qualitative judgments or evaluate with a single index. On this basis, a comprehensive score for the integrity of the entire pile and a local integrity index for each section are generated, transforming discrete anomaly information into quantifiable and comparable structural health status indicators, providing a refined damage basis for subsequent bearing capacity back calculation and construction decision-making.

[0032] Based on the local integrity index of each section and the internal force distribution curve of the continuous pile body, the unit side friction resistance after correction is back-calculated and the effective end resistance is separated. The bearing capacity is then corrected by the soil consolidation parameters to obtain the equivalent static bearing capacity. The comprehensive safety factor and structural reliability reduction factor are then output. By incorporating the local integrity index of each section into the back calculation process of side friction, the overestimation of bearing capacity caused by pile damage is effectively corrected. Combined with the soil consolidation parameters, the end resistance is corrected for timeliness, thus realizing the accurate conversion from dynamic test data to equivalent static bearing capacity. On this basis, a comprehensive safety factor and structural reliability reduction factor are output, and the pile integrity damage and soil time effect are coupled into the bearing capacity assessment system.

[0033] Based on the comprehensive safety factor, structural reliability reduction factor, and overall pile integrity score, construction handling instructions are generated and pushed to the construction site terminal. Electronic files are automatically generated, and a group-wide quality trend analysis is performed. By integrating the safety factor, reliability reduction factor, and integrity score into a multi-dimensional system to generate precise construction handling instructions, a closed-loop feedback and real-time response from data assessment to on-site decision-making are achieved. Simultaneously, the automatic generation of electronic files and the implementation of group-wide quality trend analysis ensure the traceability and compliance of the basis for individual pile handling.

[0034] On the other hand, this invention proposes a real-time monitoring system for static pressure pipe pile stress, such as... Figure 2 As shown, it includes: The sensor integration module is used to embed the packaged sensor unit inside the pipe pile, determine the number of monitoring sections based on geological stratification, and evenly arrange the sensor unit along the circumferential direction in each section. The embedded sensor unit is subjected to in-situ signal conditioning and calibration, and a standardized digital signal stream is output. The edge processing module is used to receive the standardized digital signal stream output by the sensing integration module, perform outlier removal and baseline drift correction to obtain the corrected axial force value, extract feature vectors and statistical summaries based on the corrected axial force value, perform adaptive data encapsulation and priority scheduling, and output structured feature data packets. The cloud reconstruction module is used to receive the structured feature data packets output by the edge processing module, perform multi-sensor time synchronization correction, reconstruct the internal force distribution curve of the continuous pile body based on the time-aligned data, locate abnormal sections according to the internal force distribution curve of the continuous pile body, and complete three-dimensional visualization mapping and structured storage. The integrity discrimination module is used to calculate the joint damage index and identify the damage type of the pile body based on the continuous pile internal force distribution curve and the abnormal section output by the cloud reconstruction module, and to quantify the degree of damage to obtain the overall integrity score of the whole pile and the local integrity index of each section. The bearing capacity derivation module is used to back-calculate the corrected unit side friction and separate the effective end resistance based on the local integrity index of each section output by the integrity discrimination module and the internal force distribution curve of the continuous pile body output by the cloud reconstruction module. It then combines the soil consolidation parameters to perform bearing capacity time correction to obtain the equivalent static bearing capacity and outputs the comprehensive safety factor and structural reliability reduction factor. The decision output module is used to generate construction disposal instructions and push them to the construction site terminal based on the comprehensive safety factor output by the bearing capacity deduction module, the structural reliability reduction factor, and the comprehensive integrity score of the whole pile output by the integrity discrimination module. It also automatically generates electronic files and performs group quality trend analysis.

[0035] Furthermore, the above modules, during execution, are also used to implement other steps of the aforementioned method for real-time monitoring of static pressure pipe pile stress, as follows: Step 1: Integration and adaptive deployment of embedded sensing units into the pile body; During the preparation stage of static pressure pipe pile driving construction, a sensing unit with data acquisition and transmission functions is embedded inside the pipe pile. This embedding process, through integrated structural design, makes the sensing unit an integral part of the pile's load-bearing system. The core challenge lies in addressing the survival of the sensor under high-temperature, high-pressure steam curing conditions and the effective coupling of the sensing signal within the concrete medium. Specifically, this includes: 1.1: To address the destructive impact of the high-temperature and high-pressure steam curing process on electronic components during the production of static pressure pipe piles, a multi-level thermal protection encapsulation structure is constructed before the sensor unit is implanted.

[0036] A polyimide flexible circuit board with a temperature resistance rating of over 200℃ was selected as the signal acquisition substrate. The strain-sensitive grid wire and the temperature compensation resistor were soldered onto the substrate in a full-bridge configuration. Subsequently, a silicone-modified epoxy resin adhesive layer with a thickness of 0.3mm to 0.5mm was coated on the circuit surface as primary insulation protection.

[0037] Based on this, a groove-shaped outer shell is formed by stamping titanium alloy sheets. The coated flexible circuit board is placed inside the groove, and then alumina ceramic filler with a thermal conductivity greater than 1.5 W / (m·K) is poured in. After curing, a sealed cavity is formed. The overall dimensions of the sensing unit after the above encapsulation process are controlled within 40 mm in length, 25 mm in width, and 12 mm in height to ensure that it does not significantly weaken the cross-sectional strength of the pipe pile when embedded in the pile wall, while ensuring that the acoustic impedance matching between the encapsulation shell and the concrete interface meets the stress wave transmission requirements.

[0038] 1.2: The completed sensing unit needs to be positioned at the stress characteristic section of the pile body during the stage of binding the pipe pile reinforcement cage.

[0039] Number of monitoring sections Must meet ,in, The number of soil layers penetrated by the pile body as revealed in the geological survey report.

[0040] For example, when the pile passes through 5 soil layers, at least 11 monitoring sections need to be set up to ensure that there are no less than 2 effective sensing sections in each soil layer segment, so as to provide sufficient data support for subsequent segmented linear interpolation.

[0041] Four sensing units are evenly arranged along the circumference of each cross-section, with an included angle of 90° between adjacent units. During installation, stainless steel clamps are used to secure the sensing units to the inside of the main reinforcement, with a 5mm thick rubber buffer pad placed between the clamps and the main reinforcement. The lead wires of the sensing units are made of 0.8mm diameter PTFE-sheathed wires, laid longitudinally along the inside of the main reinforcement to the pre-reserved wiring cavity at the pile end, and secured with nylon cable ties every 300mm, with the tightening torque of the cable ties controlled between 0.5N·m and 0.8N·m.

[0042] After the wiring is completed, a layer of quick-drying cement slurry with a thickness of about 2mm is applied to the outer surface of the sensing unit. When the pile body is centrifuged, it will be integrated with the fresh concrete to ensure the deformation coordination between the sensing unit and the pile body material.

[0043] 1.3: Based on the deformation coordination state achieved in step 1.2, after the pipe pile demolding and curing are completed, the embedded sensing unit is subjected to in-situ signal conditioning and calibration.

[0044] During calibration, a portable high-precision data acquisition instrument is used to connect the lead wires in the reserved wiring cavity at the pile end to record the initial output voltage value of each channel under no external load conditions. and pile body ambient temperature Subsequently, a standard preload of 5% of the estimated ultimate bearing capacity was applied to the top of the pile. Record the change in output voltage after the load stabilizes. Calculate the actual sensitivity coefficient of each sensing channel. (Unit: mV / kN).

[0045] This coefficient is written into the sensor unit's internal non-volatile memory, replacing the factory default value. Simultaneously, the on-chip 24-bit analog-to-digital converter is enabled, with the oversampling rate set to 64 times the base sampling rate. A digital low-pass filter is used to set the cutoff frequency to 20Hz to filter out high-frequency noise caused by hydraulic system pulsation and mechanical friction. The conditioned digital signal is packaged according to a predefined frame format. Each frame contains five fields: channel number, timestamp, original code value, temperature value, and checksum, with a total length of 16 bytes, providing a standardized data stream for subsequent edge processing.

[0046] Step 2: Edge end data cleaning and feature extraction triggered by pile driving; After the integrated sensing unit from step one enters the pile driving face along with the pipe pile, the system transitions to the edge data processing stage. This stage is executed on edge computing nodes located near the sensing unit or within the pile driver control cabinet. Its core task is to perform real-time cleaning and feature compression of the rapidly incoming raw stress data, completing quality verification and information condensation before the data leaves the pile body. This ensures that subsequent wireless transmissions only contain valid information reflecting the true mechanical state of the pile body, specifically including: 2.1: Based on the standardized digital signal stream output by the sensing unit, the edge computing node first performs an outlier removal operation based on a sliding window.

[0047] Edge nodes maintain a circular buffer of length 32, storing the latest sampled value x[n] in real time. For each new data point, the absolute deviation d[n] between it and the median of the historical data within the window is calculated.

[0048] If d[n] exceeds the preset threshold If the value is 3 times the noise standard deviation, it is considered an outlier and replaced with the linear interpolation result of the valid data before and after it. If 5 consecutive data points are considered outliers, the sensor fault flag is triggered, the data upload of that channel is suspended, and the fault time and location are recorded.

[0049] Specifically, when a soil layer has fewer than two effective sections due to sensor failure, the system automatically marks the soil layer as a data degradation zone. During subsequent cloud reconstruction, the system will use cross-soil layer linear interpolation for this section and mark the confidence level reduction in the results to avoid single-point failures causing the entire pile reconstruction to be interrupted.

[0050] 2.2: Based on the cleaned stress data sequence, the edge computing nodes further perform adaptive baseline drift correction.

[0051] During the interval between two pile driving operations (automatically identified by detecting when the rate of stress change is below a set threshold), the average data value during this period is used as the new dynamic zero point. In the dynamic loading phase of continuous pile driving, the most recent interval is used to determine the value. Based on the temperature change, superimposed Multiply by the factory voltage temperature drift factor The amount of compensation obtained.

[0052] Corrected pile axial force In the formula, all terms in the numerator are dimensionlessly expressed as mV, the denominator is mV / kN, and the result is kN. The corrected axial force data accurately reflects the mechanical response of the pile under external loads.

[0053] 2.3: Based on the corrected axial force value output, the edge computing nodes perform event-driven feature extraction.

[0054] Three types of triggering events are defined: axial force extreme value event (exceeding the safety threshold or changing by more than 10% compared to the previous extreme value), rate change event (axial force change rate exceeds the critical slope), and periodic integrity event (forced triggering at the end of each pile driving stroke).

[0055] Each time an event is triggered, a feature vector is extracted: event type code, occurrence time, peak / valley / average axial force, maximum rate of change of axial force, and duration.

[0056] During non-event periods, a statistical summary is reported at a frequency of 1Hz, including four fields: mean, variance, sample count, and a globally monotonically increasing sequence number. This sequence number uses a 64-bit unsigned integer, uniformly assigned by the edge nodes, spanning all channels and priority queues, ensuring that any data packet can be uniquely identified and sorted during retransmission, completely eliminating synchronization ambiguity and overflow risks in multi-channel / multi-queue scenarios.

[0057] 2.4: Based on feature vectors and statistical summaries, edge computing nodes complete adaptive data encapsulation and priority scheduling for wireless channels.

[0058] Data is divided into three priority queues (high, medium, and low) based on event type, and the send scheduler always prioritizes serving the high-priority queue. Each data packet header includes a link quality indicator field, and the receiver adjusts the transmit power and coding rate based on the bit error rate feedback.

[0059] When three consecutive transmission failures are detected, the current data packet is automatically stored in the local flash memory and retransmitted in an append-only manner during the next successful transmission cycle. During retransmission, deduplication and sorting are performed based on the global sequence number. The encapsulated data packet is then sent via LoRa or NB-IoT module, completing the transition from the edge of the charging station to the wide area network.

[0060] Step 3: Spatiotemporal alignment of multi-source data in the cloud and reconstruction of pile internal forces; After the structured feature stream output by the edge computing node arrives at the cloud server, the system performs spatiotemporal alignment on data from multiple sensing units of the same pile and different time segments. Based on this, it reconstructs the complete axial internal force distribution curve of the pile, transforming scattered point observations into a continuous field description, specifically including: 3.1: After receiving the timestamped data packet uploaded by the edge node, the cloud server first performs multi-sensor time synchronization correction. This is achieved using bidirectional message exchange and linear regression fitting. The cloud periodically sends synchronization requests to each edge node and records the time when the request is sent. With response reception time Edge nodes carry their local reception time in their responses. With the time of transmission .

[0061] At least 20 sets of paired data were continuously collected. Least squares linear fitting was performed with cloud time as the independent variable and local time as the dependent variable to obtain the slope k and intercept b. The timestamps of all subsequent reported data were based on... The correction ensures that the time consistency of data from each sensing unit reaches the millisecond level.

[0062] 3.2: Based on the multi-section axial force data aligned with the completion time, piecewise linear spatial interpolation considering pile-soil interaction is performed in the cloud.

[0063] The pile body is divided into sections according to the soil layers defined in the geological survey report. Within each section, the corrected axial force value of two adjacent sensing sections is used. , As the interpolation endpoint, for any depth z located in the interval Internal, reconstructed axial force The reconstructed continuous internal force curve S(z) accurately depicts the attenuation mode of the pile axial force with depth.

[0064] 3.3: Based on the reconstructed continuous pile internal force distribution curve S(z), the cloud performs automatic location of abnormal areas based on gradient analysis.

[0065] First, calculate the first derivative of S(z). When its absolute value deviates from the expected lateral resistance range of the soil layer by more than 30%, it is marked as a candidate anomaly. Subsequently, the fluctuation energy E of the continuous internal force curve is calculated within a window centered on the candidate point and with a length of 2m, and compared with the average energy of the entire pile. In comparison, if If a segment is identified as a valid anomaly, its start and end depths and severity index are recorded. The list of located anomaly segments includes the precise depth range and quantitative indicators.

[0066] 3.4: Based on the list of abnormal sections and the reconstructed continuous internal force curve S(z), the three-dimensional visualization mapping and structured storage of the internal force state of the pile body are completed in the cloud.

[0067] The visualization uses color-coded rendering in a cylindrical coordinate system. Abnormal sections are highlighted with flashing borders and overlaid with geological columnar charts as background references.

[0068] In terms of structured storage, each reconstruction result is encapsulated into a standard JSON object, which includes fields such as station number, timestamp, array of sampling points for continuous internal force curves, list of abnormal sections, interpolation parameters used, data quality identifier and confidence label of degraded area, and stored in time series database with multi-dimensional indexes.

[0069] Step 4: Dynamic assessment of pile integrity based on internal force reconstruction curve; After reconstructing the internal force distribution curve of the pile and locating abnormal sections, the system uses the reconstructed continuous internal force field to quantitatively and dynamically determine the structural integrity of the pile, distinguishing different damage types such as joint welding defects, concrete cracks, necking, or segregation, and providing quantitative indicators, specifically including: 4.1: Based on the continuous internal force curve S(z) and the list of abnormal sections, the cloud analysis engine first performs a joint connection quality evaluation based on the internal force transmission coefficient.

[0070] Extracting axial force at the upstream section of the joint within the abnormal section Axial force at downstream section (Interpolation results taken from 0.3m on both sides of the joint location) Calculate the measured transfer coefficient. .

[0071] Combined with the historical maximum pressure value of the joint when it passed through the pile driver's clamp recorded in step two. Establish a reference for the relationship between the transfer coefficient and the clamping force. To address the cold start issue caused by the lack of historical data for the first pile or a new batch of pipe piles, a dual-mode mechanism is adopted for benchmark acquisition: When historical data of intact connectors from the same batch (sample size ≥ 10) exists in the cloud database, statistical regression is used to generate... Otherwise, it will automatically switch to theoretical benchmark mode and calculate the ideal transfer coefficient based on the joint design drawings and material mechanical parameters. A safety reduction factor of 0.95 is applied as a temporary baseline, which is then automatically and smoothly transitioned to the statistical baseline as subsequent data on intact joints accumulates. Joint Damage Index The value ranges from 0 to 1, where 0 indicates that the device is intact and 1 indicates that it is completely ineffective.

[0072] 4.2: Based on the joint damage index Based on the reconstructed continuous internal force curve morphology characteristics, the cloud platform further performs pile body damage type identification. The identification logic employs a multi-feature fusion decision tree. First, check if the abnormal section is located at the interface of a known soil layer. If so, If so, it is classified as a soil layer effect; otherwise, it enters the bulk damage branch and extracts the local gradient deviation. With energy decay rate The two features are used for discrimination.

[0073] when Exceeding the threshold and A sudden increase is identified as a transverse crack or fracture; when Smaller but When the level is consistently high and accompanied by signs of reduction in the equivalent cross-sectional area, it is judged as necking or segregation; when both are slightly exceeded and Damage values ​​between 0.1 and 0.3 are considered longitudinal microcracks or insufficient concrete strength. The identified damage type is labeled accordingly. The values ​​together constitute the integrity state vector.

[0074] 4.3: Damage type label and joint damage index A unified damage level quantification mapping is performed in the cloud. For joint damage, it is directly applied... As a component score; For damage to the body, based on the length of the abnormal segment The ratio to pile diameter d and the severity weight corresponding to the damage type (Transverse crack 1.0, necking 0.8, longitudinal microcrack 0.4), calculate the damage components of the body: ,in, Minimum thresholds related to damage type (0.15 for transverse cracks, 0.05 for necking, and 0.02 for longitudinal microcracks) are set to ensure that brittle damage is not underestimated due to small geometric dimensions. All damage components within the same depth range are synthesized according to the worst-case scenario, and the maximum value is taken as the local integrity index of that section. Overall score for pile integrity By analyzing each cross section The result is obtained by integrating along the pile length and dividing by the pile length.

[0075] 4.4: Comprehensive scoring based on the integrity of the entire pile and the local integrity index of each section The sequence, completed in the cloud, represents the engineering aspects of the pile integrity status and compares them with early warning thresholds. The integrity status is divided into four levels: Level 1 ( If it is intact, the pile driving can continue; Level 2 ( For minor defects, it is recommended to reduce speed and increase monitoring. Level 3 ( For obvious damage, construction should be suspended and the work re-tested; Level 4 ( Severe damage necessitates scrapping or special reinforcement.

[0076] Simultaneously for any cross section All of these situations trigger local high-risk alerts. The warning information, along with the integrity level, damage type, quantitative indicators, and handling recommendations, is packaged into a standard message and pushed to the construction site terminal.

[0077] Step 5: Dynamically extrapolate the bearing capacity based on the integrated pile driving process parameters; After quantifying the integrity status of the pile body, the system deeply integrates the integrity information with real-time pile driving process parameters to dynamically extrapolate the vertical compressive bearing capacity of a single pile. Simultaneously with construction, it estimates the equivalent bearing capacity considering the impact of actual damage, specifically including: 5.1: Acceptance of local integrity index of each section The cloud-based analysis engine performs back-calculation of pile side friction and separation of end resistance based on integrity correction, along with the reconstructed continuous internal force curve S(z).

[0078] At depth z, the corrected unit side friction resistance is: In the formula, d is the pile diameter (m). Take 0.5m, N(z) and The value of the axial force (kN) at both ends of the reconstructed curve is given.

[0079] This is a nonlinear damage loss term, and its value is equal to... ,in, To predict the ultimate axial force, The nonlinear indices are (concrete cracks 1.5, joint loosening 0.8, necking 1.0, obtained from indoor loading tests of specimens with similar damage based on regression).

[0080] To prevent losses from exceeding the actual axial force difference under overload conditions, resulting in a negative calculated side resistance, a protective cutoff is added: Cumulative side resistance Through the Along the depth of the soil The points are obtained through integration.

[0081] Instantaneous end resistance ,in, This is the total pile driving force recorded in real time by the pile driver.

[0082] Effective end resistance End resistance transfer efficiency factor The lower limit of 0.1 is a numerical protection threshold to avoid computational singularities and does not represent the physical residual end bearing capacity.

[0083] 5.2: Based on effective end resistance With cumulative side resistance The bearing capacity aging correction, which takes into account the recovery effect of construction disturbance, is performed in the cloud. The soil consolidation parameters required for the correction are... and An adaptive acquisition mechanism is adopted: laboratory test values ​​are read first from the geological survey report; if they are not available, the median value is taken from the empirical range of the corresponding soil type in the "Technical Specification for Building Pile Foundations" JGJ94, and the confidence level of the parameter source is marked in the output results with three levels of labels: high, medium and low (high = laboratory test, medium = standard empirical value, low = inversion estimation).

[0084] For the lateral resistance recovery coefficient of cohesive soil layers Sandy soil layer The opposite applies to the aging correction of the terminal resistance.

[0085] Equivalent static bearing capacity To avoid the impact of damage being factored into both the load-bearing capacity value and the safety factor, a comprehensive safety factor is defined. ( (For design characteristic loads), and simultaneously output the structural reliability reduction factor independently. .

[0086] when and The load-bearing capacity and integrity requirements are deemed to be met at that time. when but The system indicates that the bearing capacity meets the standard, but there are significant defects in the pile body. when Regardless of time How can we determine that the load-bearing capacity is insufficient?

[0087] Step Six: Engineering integration of multi-dimensional monitoring results and output of construction decisions; After completing the load-bearing capacity safety margin assessment, the system integrates the integrity status of all the aforementioned outputs, load-bearing capacity projection results, and original monitoring data in an engineering manner, forming a comprehensive decision-making output that can directly guide on-site construction, acceptance, and archiving. This ensures that the monitoring data truly translates into practical effectiveness in construction quality and safety management, specifically including: 6.1: Undertaking the comprehensive safety factor Structural reliability reduction factor Based on the integrity classification results, the cloud-based system performs multi-dimensional decision fusion to generate the final construction and disposal instructions. A matrix-style decision-making logic is employed. when and Output normal acceptance command in time; when but Output conditional acceptance instructions along with details of integrity defects; when but If the output indicates insufficient bearing capacity, it is recommended to add piles or review the design. when and If a serious abnormality command is output, work must be stopped immediately and a special investigation initiated.

[0088] The decision matrix is ​​perfectly aligned with the decoupling criteria in step 5.2.

[0089] When there is semantic overlap between the integrity warning level in step 4.4 and the handling instruction in step 6.1, the handling instruction in step 6.1 shall be the final arbitration basis, and the integrity warning shall only be displayed as auxiliary reference information to eliminate the flaws in the connection between multi-level decisions.

[0090] 6.2: Based on the construction handling instructions generated in step 6.1, the cloud platform executes real-time information push and interactive confirmation for the construction site. Push channels include three methods: on-site large screen, mobile terminal, and audible and visual alarms, automatically selected according to the urgency of the instruction.

[0091] In addition to the instructions themselves, the pushed content includes a 3D internal force visualization screenshot from step three, a damage location map from step four, and a bearing capacity projection curve and decoupling indicators from step five. The receiving operator needs to click to confirm; the confirmation information is sent back to the cloud, and a timestamp and operator identity are recorded to form a closed loop. If no confirmation is received within 5 minutes, the system automatically escalates the notification to the project management personnel.

[0092] 6.3: Based on the disposal instructions confirmed in step 6.2 and the full-process monitoring data, the cloud automatically generates electronic files that meet the engineering acceptance specifications.

[0093] The archive includes basic information on the pile foundation, sensor layout diagram, raw data summary, internal force reconstruction curve, integrity evaluation report, and bearing capacity calculation sheet (including...). and (Item and parameter confidence labels), construction handling instructions and confirmation records.

[0094] All data is formatted using a unified template and supports PDF export and digital signatures. Data integrity check codes are embedded in the files, generated using a chained hash algorithm. Arrange the key data from each step (steps one through five) in chronological order, and calculate the SHA-256 hash value for the first data block. Calculate for each subsequent data block i ,final It is embedded as a file verification code. To balance security and performance, an intermediate hash value is generated and persistently stored for every 100 data blocks. During verification, it can be traced from any intermediate point, avoiding the overhead of full serial computation.

[0095] 6.4: Utilizing electronic archives and historical project databases, perform group-based quality trend analysis and risk warnings in the cloud. This involves analyzing all completed piles within the same project. , , The system statistically analyzes damage types to identify common quality issues and trigger corresponding alerts. Analysis results are automatically pushed to project management in the form of weekly or monthly reports, along with trend charts and recommended actions.

[0096] The above description is merely an exemplary embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention specification and drawings under the technical concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for real-time monitoring of stress in a static pressure pipe pile, characterized in that, include: The encapsulated sensing unit is implanted inside the pipe pile. The number of monitoring sections is determined based on geological stratification, and the sensing units are evenly arranged along the circumference of each section. The embedded sensing unit is subjected to in-situ signal conditioning and calibration, and a standardized digital signal stream is output. The system receives the standardized digital signal stream, performs outlier removal and baseline drift correction to obtain the corrected axial force value, extracts feature vectors and statistical summaries based on the corrected axial force value, performs adaptive data encapsulation and priority scheduling, and outputs a structured feature data packet. Receive the structured feature data packet, perform multi-sensor time synchronization correction, reconstruct the continuous pile internal force distribution curve based on the time-aligned data, locate abnormal sections based on the continuous pile internal force distribution curve, and complete three-dimensional visualization mapping and structured storage. Based on the continuous pile internal force distribution curve and the abnormal section, the joint damage index is calculated and the pile body damage type is identified. The degree of damage is quantified to obtain the overall pile integrity score and the local integrity index of each section. Based on the local integrity index of each section and the internal force distribution curve of the continuous pile body, the unit side friction resistance after correction is back-calculated and the effective end resistance is separated. The bearing capacity is then corrected by the soil consolidation parameters to obtain the equivalent static bearing capacity. The comprehensive safety factor and structural reliability reduction factor are then output. Based on the comprehensive safety factor, the structural reliability reduction factor, and the comprehensive score of the whole pile integrity, a construction disposal instruction is generated and pushed to the construction site terminal, an electronic file is automatically generated, and a group quality trend analysis is performed.

2. The method for real-time monitoring of static pressure pipe pile stress according to claim 1, characterized in that, The process of embedding the encapsulated sensing unit inside the pipe pile includes: A multi-level thermal protection packaging structure is constructed. A polyimide flexible circuit board is selected as the signal acquisition substrate. The strain-sensitive grid wire and the temperature compensation resistor are soldered to the signal acquisition substrate in a full-bridge manner. An organosilicon-modified epoxy resin adhesive layer is coated on the circuit surface as primary insulation protection. A groove-shaped shell is formed by stamping titanium alloy sheets. The coated polyimide flexible circuit board is placed inside the groove-shaped shell. After the alumina ceramic filler is injected and cured, a sealed cavity is formed, resulting in a sensing unit whose size is controlled within a preset range.

3. The method for real-time monitoring of static pressure pipe pile stress according to claim 2, characterized in that, The method of determining the number of monitoring sections based on geological stratification and uniformly arranging the sensing units along the circumference of each section includes: The number of monitoring sections is determined based on the number of soil layers penetrated by the pile as revealed in the geological survey report, so that each soil layer segment has no less than two effective sensing sections. Four sensing units are evenly arranged along the circumference of each monitoring section. The sensing units are fastened to the inside of the main reinforcement using stainless steel clamps. The lead wire is laid longitudinally along the inside of the main reinforcement to the reserved wiring cavity at the pile end and fixed. Quick-drying cement slurry is applied to the outer surface of the sensing unit to ensure deformation coordination with the pile body material.

4. The method for real-time monitoring of static pressure pipe pile stress according to claim 3, characterized in that, The in-situ signal conditioning and calibration of the embedded sensing unit includes: Record the initial output voltage value of each channel under no external load conditions and the ambient temperature of the pile body. After applying the standard preload, record the change in output voltage, calculate the actual sensitivity coefficient of each sensing channel and write it into the non-volatile memory. The on-chip analog-to-digital converter is enabled to set the oversampling rate. A digital low-pass filter is used to filter out high-frequency noise. The conditioned digital signal is packaged according to a predefined frame format to generate the standardized digital signal stream. Each frame contains a channel number, timestamp, original code value, temperature value, and checksum field.

5. The method for real-time monitoring of static pressure pipe pile stress according to claim 1, characterized in that, The process of performing outlier removal and baseline drift correction to obtain the corrected axial force value includes: The latest sampled values ​​are stored in a ring buffer in real time. The absolute deviation between each new data point and the median of the historical data in the window is calculated. When the absolute deviation exceeds a preset threshold, it is judged as an outlier and replaced with a linear interpolation result. When multiple consecutive data points are judged as outliers, the sensor fault flag is triggered. During the interval between two pile driving actions, the average data value is used as the dynamic zero point. During the dynamic loading segment of continuous pile driving, the temperature compensation is superimposed based on the dynamic zero point, and the corrected axial force value is calculated in combination with the actual sensitivity coefficient.

6. The method for real-time monitoring of static pressure pipe pile stress according to claim 5, characterized in that, The extraction of feature vectors and statistical summaries based on the corrected axial force values ​​includes: Three types of triggering events are defined: extreme axial force events, rate change events, and periodic integrity events. Each time an event is triggered, a feature vector containing the event type code, occurrence time, peak axial force, valley axial force, average axial force, maximum axial force change rate, and duration is extracted. During non-event periods, a statistical summary containing the mean, variance, sample count, and a globally monotonically increasing sequence number is reported at a preset frequency. The globally monotonically increasing sequence number is uniformly allocated by the edge nodes to eliminate ambiguity in multi-channel synchronization.

7. The method for real-time monitoring of static pressure pipe pile stress according to claim 1, characterized in that, The reconstruction of the internal force distribution curve of the continuous pile body based on time-aligned data includes: Two-way message exchange and linear regression fitting are used to correct the timestamps of the data reported by each edge node, so that the data of each sensing unit can achieve millisecond-level time consistency. The pile body is divided into soil layers according to the geological survey report. Within each segment, the corrected axial force values ​​of two adjacent sensing sections are used as interpolation endpoints. Piecewise linear spatial interpolation is performed at any depth position to obtain a continuous pile body internal force distribution curve.

8. The method for real-time monitoring of static pressure pipe pile stress according to claim 7, characterized in that, Based on the internal force distribution curve of the continuous pile body, abnormal sections are located, including: Calculate the first derivative of the internal force distribution curve of the continuous pile body. When the absolute value of the first derivative deviates from the expected lateral resistance range of the soil layer by more than a preset proportion, it is marked as a candidate anomaly point. The fluctuation energy of the internal force curve is calculated within a preset length window centered on the candidate anomaly point. When the ratio of the fluctuation energy to the average energy of the entire pile exceeds a preset multiple, it is confirmed as an effective anomaly section, and the start and end depths and severity index are recorded.

9. The method for real-time monitoring of static pressure pipe pile stress according to claim 1, characterized in that, The calculation of the joint damage index and identification of the pile body damage type include: The measured transfer coefficient is calculated by extracting the axial force of the upstream section and the axial force of the downstream section of the joint in the abnormal section. The transfer coefficient is then established in conjunction with the historical maximum clamping force. The joint damage index is obtained based on the ratio of the measured transfer coefficient to the associated benchmark. Check whether the abnormal section is located at the soil layer interface. If not, extract the local gradient deviation and energy attenuation rate. Based on the combination relationship of the local gradient deviation, energy attenuation rate and joint damage index, determine the damage type of transverse crack, necking, segregation or longitudinal microcrack.

10. A real-time monitoring system for static pressure pipe pile stress, used to implement the real-time monitoring method for static pressure pipe pile stress as described in any one of claims 1-9, characterized in that, include: The sensor integration module is used to embed the packaged sensor unit inside the pipe pile, determine the number of monitoring sections based on geological stratification, and evenly arrange the sensor unit along the circumferential direction in each section. The embedded sensor unit is subjected to in-situ signal conditioning and calibration, and a standardized digital signal stream is output. The edge processing module is used to receive the standardized digital signal stream output by the sensing integration module, perform outlier removal and baseline drift correction to obtain the corrected axial force value, extract feature vectors and statistical summaries based on the corrected axial force value, perform adaptive data encapsulation and priority scheduling, and output structured feature data packets. The cloud reconstruction module is used to receive the structured feature data packets output by the edge processing module, perform multi-sensor time synchronization correction, reconstruct the internal force distribution curve of the continuous pile body based on the time-aligned data, locate abnormal sections according to the internal force distribution curve of the continuous pile body, and complete three-dimensional visualization mapping and structured storage. The integrity discrimination module is used to calculate the joint damage index and identify the damage type of the pile body based on the continuous pile internal force distribution curve and the abnormal section output by the cloud reconstruction module, and to quantify the degree of damage to obtain the overall integrity score of the whole pile and the local integrity index of each section. The bearing capacity derivation module is used to back-calculate the corrected unit side friction and separate the effective end resistance based on the local integrity index of each section output by the integrity discrimination module and the internal force distribution curve of the continuous pile body output by the cloud reconstruction module. It then combines the soil consolidation parameters to perform bearing capacity time correction to obtain the equivalent static bearing capacity and outputs the comprehensive safety factor and structural reliability reduction factor. The decision output module is used to generate construction disposal instructions and push them to the construction site terminal based on the comprehensive safety factor output by the bearing capacity deduction module, the structural reliability reduction factor, and the comprehensive integrity score of the whole pile output by the integrity discrimination module. It also automatically generates electronic files and performs group quality trend analysis.

Images