A self-balanced static load displacement detection system and method for a foundation pile
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-07
- Publication Date
- 2026-06-02
Smart Images
Figure CN122129049A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of pile foundation testing technology, specifically a pile self-balancing static load displacement testing system and method. Background Technology
[0002] As the core load-bearing structure of a building project, the bearing capacity and displacement characteristics of foundation piles directly determine the overall safety of the project. Self-balancing static load testing has become the mainstream method for foundation pile performance testing because it eliminates the need for bulky reaction devices and is adaptable to complex sites. This method provides loading force through a pre-embedded load cell within the pile, combined with data collected by sensors to evaluate the stress and deformation patterns of the pile. It is widely used in major projects such as bridges and high-rise buildings.
[0003] Existing self-balancing detection technologies have significant limitations: Sensor deployment often relies on engineering experience or standard settings, lacking personalized adaptation to pile structure and geological conditions, which can easily lead to blind spots in displacement monitoring or data redundancy. Interference signals generated by temperature fluctuations and environmental vibrations during the testing process can easily cause deviations in displacement data.
[0004] Traditional data processing relies heavily on centralized backend analysis, which suffers from problems such as lagging time alignment and insufficient real-time performance. It cannot capture sudden deformation of the pile body during the loading process in a timely manner, making it difficult to achieve risk warning.
[0005] Therefore, a system and method for detecting the self-balancing static load displacement of foundation piles are provided, which can realize real-time, accurate and efficient detection of the self-balancing static load displacement of foundation piles, and greatly meet the high standard requirements of engineering for foundation pile performance evaluation. Summary of the Invention
[0006] To address the aforementioned technical problems, the present invention aims to provide a self-balancing static load displacement detection system and method for foundation piles, which can achieve real-time, accurate, and efficient detection of the self-balancing static load displacement of foundation piles, greatly meeting the high standards required for foundation pile performance evaluation in engineering projects.
[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting the static load displacement of a foundation pile under self-balancing load, the method comprising: Based on digital twin models and multi-condition finite element simulations, the placement of distributed fiber optic displacement sensors on the pile body is determined through machine learning and multi-objective optimization algorithms, and temperature and vibration sensors are also deployed. During the self-balancing loading process, load signals, displacement signals, temperature signals and vibration signals are collected simultaneously, and time sequence alignment and delay compensation are performed to form a standardized data sequence. By calling temperature and vibration signals, and using an environmental adaptive calibration algorithm to compensate for errors in the displacement signal, a compensated displacement signal is obtained. The compensation displacement signal is analyzed in real time by edge computing nodes, load-displacement curves are generated and key parameters are calculated, and an early warning is triggered when the parameters exceed the threshold.
[0008] Preferably, the determination of the placement location of the distributed fiber optic displacement sensor on the pile body based on the digital twin model and multi-condition finite element simulation, through machine learning and multi-objective optimization algorithms, includes: Based on building information modeling, geological survey and hydrogeological information, a digital twin pile model is constructed by integrating the geometric structure, material properties, soil parameters and seepage field of the pile; the uncertain parameters in the model are identified and quantified to form an adjustable parameter set; Based on the digital twin pile model and its adjustable parameter set, multi-condition parametric finite element simulation is performed, outputting massive amounts of simulation data; machine learning algorithms are used to analyze the simulation data to identify engineering events that have a key impact on the pile body response and their corresponding spatial distribution. Based on the spatial distribution of various engineering events, a dynamic monitoring demand hotspot map is generated along the pile body, which is marked with different physical quantity monitoring priorities. Based on the dynamic monitoring demand heat map, an improved multi-objective optimization algorithm is used to generate the optimal sensor deployment scheme; Based on the optimal sensor deployment scheme, corresponding distributed fiber optic displacement sensors are deployed on the pile body.
[0009] Preferably, the step of generating a dynamic monitoring demand hotspot map along the pile body, based on the spatial distribution of various engineering events and marked with different physical quantity monitoring priorities, includes: Based on each identified engineering event, the spatial and probabilistic features corresponding to the engineering event are extracted and quantified to obtain the feature vector of the engineering event. The feature vector includes the core depth range of the event occurrence, the probability density function of occurrence distributed along the pile depth, the spatial influence radius of the mechanical effect of the event, and the set of key physical quantities used to characterize the process event. Based on the aforementioned feature vectors and massive simulation data, an event occurrence probability factor field, a mechanical response mutation factor field, and a model prediction divergence factor field are constructed respectively. Based on the constructed event occurrence probability factor field, mechanical response mutation factor field, and model prediction divergence factor field, as well as the preset dynamic weight function related to the loading stage, a three-dimensional dynamic decision field is obtained by weighted fusion. The dynamic weight function is configured such that: in the early stage of loading, the model prediction divergence factor field dominates the fusion; in the middle and late stages of loading, the mechanical response mutation factor field dominates the fusion. Based on the three-dimensional dynamic decision field, slices and projections are performed according to preset key load stages and physical quantity types to obtain a series of two-dimensional dynamic monitoring demand hotspot maps distributed along the pile body. The dynamic monitoring demand hotspot maps are used to identify the monitoring priority of different physical quantities at different depth locations.
[0010] Preferably, the step of generating the optimal sensor deployment scheme based on the dynamic monitoring demand heatmap and using an improved multi-objective optimization algorithm includes: Based on the detection priority of each physical quantity corresponding to each position in the dynamic monitoring demand hotspot map, an information potential energy field along the pile depth-physical quantity distribution is constructed. Based on the information potential field, and based on a weighted random sampling strategy where the probability density is proportional to the potential energy, a population containing multiple initial deployment schemes is generated, provided that the preset range of the number of sensors is met. Based on each initial deployment scheme and the preset multi-dimensional fitness index quantification formula, the corresponding weighted comprehensive evaluation value is calculated, and the roulette wheel selection method is used to select excellent individuals from the current population. Based on the information potential field, a spatial gradient field is generated to identify areas where monitoring demand changes abruptly. Based on the spatial gradient field, the selected individuals are guided to undergo crossover and mutation operations to obtain the offspring population. After obtaining the offspring population, continue to calculate the weighted comprehensive evaluation value until the preset number of iterations is met, and output the offspring population as the optimal solution set for multi-objective optimization. Based on the optimal solution set, the individual with the largest weighted comprehensive rating value is selected as the optimal sensor deployment scheme.
[0011] Preferably, during the self-balancing loading process, the simultaneous acquisition of load signals, displacement signals, temperature signals, and vibration signals, followed by time alignment and delay compensation to form a standardized data sequence, includes: Based on the loading start command, the synchronous acquisition unit triggers and receives load signals from the hydraulic servo loading module, displacement signals from the distributed fiber optic displacement sensor network, and temperature and vibration signals from the temperature and vibration sensors in parallel. Each sampling point's load signal, displacement signal, temperature signal, and vibration signal is labeled with a microsecond-level time tag with a unified reference, thus obtaining a set of multi-source raw data streams with time tags; Based on the multi-source raw data stream with time tags, all its time tags are mapped to a common relative time axis with the loading and startup time as the zero point; Based on the fixed delay parameters of each signal transmission channel pre-calibrated, deterministic time offset compensation is performed on each data stream mapped to the common time axis, thereby obtaining a standardized data sequence that is initially aligned on the time axis.
[0012] Preferably, the step of calling the temperature signal and vibration signal, and using an environmental adaptive calibration algorithm to perform error compensation on the displacement signal to obtain the compensated displacement signal includes: Based on the load signal data sequence in the standardized data sequence, data segments with load change rate lower than the preset change threshold are identified as the zero stress change reference period; Based on the displacement signal sequence, temperature signal sequence, and vibration signal sequence corresponding to the zero stress change reference period, statistical analysis is performed to obtain the effective temperature compensation coefficient and vibration interference sensitivity. Throughout the loading process, based on the effective temperature compensation coefficient, the temperature offset component of each sensor is subtracted in real time from the displacement signal acquired in real time to obtain the displacement signal after preliminary temperature compensation. Based on the vibration interference sensitivity and the vibration signal acquired in real time, the amplitude of the displacement signal noise floor introduced by vibration is calculated, and then subtracted from the displacement signal initially compensated for by temperature to obtain the final compensated displacement signal.
[0013] Preferably, the method further includes: when an early warning is triggered, restoring the deformation state of the pile body based on three-dimensional visualization technology, and intelligently verifying the detection results in combination with preset parameter models for different pile types, and outputting a detection report.
[0014] A second aspect of the present invention also provides a pile self-balancing static load displacement detection system, using the above-described pile self-balancing static load displacement detection method, comprising: The deployment module, based on digital twin model and multi-condition finite element simulation, uses machine learning and multi-objective optimization algorithms to determine the deployment location of distributed fiber optic displacement sensors on the pile, and deploys temperature sensors and vibration sensors. The loading acquisition module synchronously acquires load signals, displacement signals, temperature signals, and vibration signals during the self-balancing loading process, and performs time sequence alignment and delay compensation to form a standardized data sequence. The compensation module calls the temperature signal and vibration signal, and uses an environmental adaptive calibration algorithm to perform error compensation on the displacement signal to obtain the compensated displacement signal. The early warning module analyzes the compensated displacement signal in real time through edge computing nodes, generates load-displacement curves and calculates key parameters, and triggers an early warning when the parameters exceed the threshold.
[0015] Compared with the prior art, the beneficial effects of the present invention are: This solution generates a dynamic monitoring demand heat map through digital twin modeling, multi-condition finite element simulation, and improved multi-objective optimization algorithm. This enables precise matching of sensor deployment with the key monitoring areas of the pile body and the needs of the loading stage, avoiding redundant deployment. While ensuring the integrity of key data, it maximizes the efficiency of monitoring resources and solves the problem of blind deployment based on traditional experience.
[0016] This solution achieves microsecond-level timing alignment and delay compensation of multi-source signals, combined with an environmental adaptive calibration algorithm, to simultaneously eliminate temperature and vibration interference, significantly improving displacement detection accuracy and ensuring the reliability of load-displacement curves and core parameters, thus overcoming the limitations of single-factor compensation in existing technologies.
[0017] This solution leverages edge computing to achieve real-time on-site data processing and over-threshold early warning, coupled with 3D visualization deformation restoration and intelligent verification functions to improve the timeliness of risk response; the digital twin model can be adapted to different foundation piles and geological conditions, possessing strong versatility and scenario adaptability, broadening the engineering application scope of detection methods, and solving the problems of insufficient versatility and scenario adaptability, and low risk response timeliness of traditional detection methods. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.
[0019] Figure 1 This is a schematic diagram of a method for detecting static load displacement of foundation piles under self-balancing load.
[0020] Figure 2 This is a schematic diagram of a self-balancing static load displacement detection system for foundation piles. Detailed Implementation
[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.
[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0023] Example 1 like Figure 1 As shown in the figure, this embodiment discloses a method for detecting the static load displacement of a foundation pile under self-balancing load. The method includes: Based on digital twin models and multi-condition finite element simulations, the placement of distributed fiber optic displacement sensors on the pile body is determined through machine learning and multi-objective optimization algorithms, and temperature and vibration sensors are also deployed. It should be noted that the determination of the placement of distributed fiber optic displacement sensors on the pile body based on digital twin models and multi-condition finite element simulations, through machine learning and multi-objective optimization algorithms, includes: Based on building information modeling, geological survey and hydrogeological information, a digital twin pile model is constructed by integrating the geometric structure, material properties, soil parameters and seepage field of the pile; the uncertain parameters in the model are identified and quantified to form an adjustable parameter set; In this embodiment, the geometric structure and material property parameters of the foundation piles in the building information model are compared with the soil layer distribution and soil physical and mechanical parameters (including cohesion) in the geological survey report. internal friction angle Compression modulus Poisson's ratio This data is then integrated with groundwater level and seepage field parameters from hydrogeological information. A comprehensive geological-structural model continuously distributed in three-dimensional space is generated using a spatial interpolation algorithm, where soil parameters are expressed as functions of depth z. The uncertainties in the model input parameters are mainly identified, including: spatial variability of soil parameters, statistical distribution characteristics of pile-soil interface parameters, manufacturing tolerances of material properties, and simplified assumptions regarding boundary conditions; each uncertain parameter... Quantified into a probability distribution: ,in The mean, For variance, The distribution type (e.g., normal distribution, log-normal distribution, uniform distribution, etc.). All uncertainty parameters constitute the adjustable parameter set. .
[0024] Based on the digital twin pile model and its adjustable parameter set, multi-condition parametric finite element simulation is performed, outputting massive amounts of simulation data; machine learning algorithms are used to analyze the simulation data to identify engineering events that have a key impact on the pile body response and their corresponding spatial distribution. In this embodiment, the Latin hypercube sampling method is used in the parameter space. N sets of parameter combinations are generated internally. For each set of parameters, a corresponding finite element model is established to simulate the entire process from initial loading to ultimate failure. Load is applied in stages: ,in This is the m-th load level.
[0025] Each simulation outputs the corresponding displacement, stress, and strain fields, as well as the pile internal force distribution, pile-soil interaction forces, and soil condition indicators. Each event... Features identified from simulation data using machine learning algorithms include the probability of occurrence in the parameter space. And typical distribution patterns in physical space.
[0026] Based on the spatial distribution of various engineering events, a dynamic monitoring demand hotspot map is generated along the pile body, which is marked with different physical quantity monitoring priorities. The process of generating a dynamic monitoring demand hotspot map along the pile body, based on the spatial distribution of various engineering events and marked with different physical quantity monitoring priorities, includes: Based on each identified engineering event, the spatial and probabilistic features corresponding to the engineering event are extracted and quantified to obtain the feature vector of the engineering event. The feature vector includes the core depth range of the event occurrence, the probability density function of occurrence distributed along the pile depth, the spatial influence radius of the mechanical effect of the event, and the set of key physical quantities used to characterize the event. In this embodiment, based on the depth records of events occurring in the simulation data, cluster analysis is used to determine the depth range where events occur most densely, thus obtaining the core depth interval. A kernel density estimation method is employed to construct a probability density function based on sample data representing the event's depth. Sensitivity analysis is then used to determine the sensitivity of the data to represent the event. The most critical physical quantities form a set of key physical quantities.
[0027] Based on the aforementioned feature vectors and massive simulation data, an event occurrence probability factor field, a mechanical response mutation factor field, and a model prediction divergence factor field are constructed respectively. In this embodiment, the event occurrence probability factor field is obtained by weighted fusion based on the importance weights and probability density functions of each event; the mechanical response mutation factor field is obtained by normalizing the simulation data of each physical quantity after performing gradient analysis or higher-order derivative calculation along the depth direction; the model prediction divergence factor field is calculated based on the spatial variation coefficients of the simulation prediction values of each physical quantity under the adjustable parameter set, and the above factor fields are normalized.
[0028] Based on the constructed event occurrence probability factor field, mechanical response mutation factor field, and model prediction divergence factor field, as well as the preset dynamic weight function related to the loading stage, a three-dimensional dynamic decision field is obtained by weighted fusion. The dynamic weight function is configured such that: in the early stage of loading, the model prediction divergence factor field dominates the fusion; in the middle and late stages of loading, the mechanical response mutation factor field dominates the fusion. In this embodiment, the dynamic weighting function is: ; ; ; , , For the corresponding benchmark weight coefficient, , For shape control parameters, , For loading phase parameters, , where t=0 indicates the start of loading, and t=1 indicates the achievement of the ultimate load.
[0029] The three-dimensional dynamic decision field is: ;in , , These are the corresponding normalized event occurrence probability factor field, mechanical response mutation factor field, and model prediction divergence factor field, respectively.
[0030] Based on the three-dimensional dynamic decision field, slices and projections are performed according to preset key load stages and physical quantity types to obtain a series of two-dimensional dynamic monitoring demand hotspot maps distributed along the pile body. The dynamic monitoring demand hotspot maps are used to identify the monitoring priority of different physical quantities at different depth locations.
[0031] Specifically, slicing and projection involve slicing by load stage, and generating a heat map for a specific load stage. , Where K represents the preset number of critical load stages. Slicing is done by physical quantity type, and heatmaps are generated for specific physical quantities. , , where L is the number of physical quantity types of interest.
[0032] Based on the dynamic monitoring demand heat map, an improved multi-objective optimization algorithm is used to generate the optimal sensor deployment scheme; The step of generating the optimal sensor deployment scheme based on a dynamic monitoring demand hotspot map and using an improved multi-objective optimization algorithm includes: Based on the detection priorities of each physical quantity corresponding to each location in the dynamic monitoring demand hotspot map, an information potential energy field along the pile depth-physical quantity distribution is constructed; in this embodiment, the dynamic monitoring demand hotspot map is assumed to be at a depth of... Physical quantities The monitoring priority at the location is The information potential energy field is defined as: , The potential energy scale parameter is greater than zero. For cases where multiple physical quantities need to be considered, a comprehensive information potential energy field is generated. The definition is obtained by weighted summation of the various information potential energy fields.
[0033] Based on the aforementioned information potential energy field, and using a weighted random sampling strategy where probability density is proportional to potential energy, a population containing multiple initial deployment schemes is generated, provided that the preset number of sensors is met. In this embodiment, a probability density function distributed along the pile depth is constructed based on the comprehensive information potential energy. ,in[ , [This refers to the depth range of the pile body.]
[0034] Based on each initial deployment scheme and a preset multi-dimensional fitness index quantification formula, the corresponding weighted comprehensive evaluation value is calculated, and excellent individuals are selected from the current population using the roulette wheel selection method. In this embodiment, the multi-dimensional fitness includes information capture efficiency index, network spatial resolution index, and event collaborative monitoring capability index.
[0035] The information acquisition efficiency index is calculated based on the sum of the corresponding potential energy values of all sensor points in the information potential energy field in this scheme. The network spatial resolution metric is based on the Euclidean distance between any two sensors in the scheme. An information utility penalty function that decays exponentially with decreasing distance is introduced. It is obtained by aggregating the penalty values of all sensor pairs to actively suppress spatial clustering redundancy of sensors. The event collaborative monitoring capability index is determined based on the core depth range, influence radius, and key physical quantity set of each engineering event. The index is calculated by counting the number of events in which the solution can completely monitor all key physical quantities within its influence radius. The ratio of this number to the total number of events is the value of this index.
[0036] Based on the comprehensive information potential energy field, a spatial gradient field is generated to identify areas of abrupt changes in monitoring demand; wherein, based on the first derivative or difference calculation of the comprehensive information potential energy field along the pile depth direction, a spatial gradient field is generated to identify areas of abrupt changes in monitoring demand; in this embodiment, the calculation of the spatial gradient field is specifically as follows: If differential calculation is used, then Then, the spatial gradient field is normalized, and the gradient magnitude field is defined as follows: .
[0037] Based on the spatial gradient field, the selected individuals are guided to undergo crossover and mutation operations to obtain the offspring population. In this embodiment, during the directional crossover operation, sensor individuals with complementary distributions in high gradient regions are preferentially selected for pairing. After pairing, the sensor individuals in high gradient regions are more likely to be retained on the individual.
[0038] During mutation operations, a gradient-guided mechanism is introduced, whereby for each individual, a gradient is applied with probability. Add a sensor at position z, with probability The sensor at position z is removed, and the sensor is moved from position z to a neighboring position. The probability is proportional to .
[0039] After obtaining the offspring population, continue to calculate the weighted comprehensive evaluation value until the preset number of iterations is met, and output the offspring population as the optimal solution set for multi-objective optimization. Based on the optimal solution set, the individual with the largest weighted comprehensive rating value is selected as the optimal sensor deployment scheme.
[0040] Based on the optimal sensor deployment scheme, corresponding distributed fiber optic displacement sensors are deployed on the pile body.
[0041] During the self-balancing loading process, load signals, displacement signals, temperature signals and vibration signals are collected simultaneously, and time sequence alignment and delay compensation are performed to form a standardized data sequence. It should be noted that, during the self-balancing loading process, the simultaneous acquisition of load signals, displacement signals, temperature signals, and vibration signals, followed by time alignment and delay compensation to form a standardized data sequence, includes: Based on the loading start command, the synchronous acquisition unit triggers and receives load signals from the hydraulic servo loading module, displacement signals from the distributed fiber optic displacement sensor network, and temperature and vibration signals from the temperature and vibration sensors in parallel. Each sampling point is labeled with a microsecond-level time tag with a unified reference, thus obtaining a set of multi-source raw data streams with time tags. In this embodiment, the load sensor built into the hydraulic servo loading module, the demodulator of the distributed fiber optic displacement sensor network, and the temperature and vibration sensors connected to the synchronous acquisition unit are configured with a unified precision clock source or connected to the same precise clock protocol network. Through the setting of a high-precision clock, each frame of displacement data, load data, temperature data, and vibration data is labeled with a microsecond-level time tag based on its internal clock during data acquisition.
[0042] Based on the multi-source raw data stream with time tags, all its time tags are mapped to a common relative time axis with the loading start time as the zero point. In this embodiment, for the fixed microsecond-level delay caused by different signal transmission paths, deterministic correction is performed on the common time axis through preset channel delay compensation parameters.
[0043] Based on the pre-calibrated fixed delay parameters of each signal transmission channel, deterministic time offset compensation is performed on each data stream mapped to a common time axis, thereby obtaining a standardized data sequence that is initially aligned on the time axis. In this embodiment, during the loading process, the relative drift of the clocks inside the load sensor, demodulator, and synchronous acquisition unit is periodically compared. When the relative drift of any clock exceeds a preset threshold, a non-uniform resampling algorithm based on cubic spline interpolation is used to resample the data sequence of the drifting signal channel onto a uniform, drift-free ideal time grid to eliminate accumulated timing errors.
[0044] By calling temperature and vibration signals, and using an environmental adaptive calibration algorithm to compensate for errors in the displacement signal, a compensated displacement signal is obtained. It should be noted that the temperature signal and vibration signal are called together, and the displacement signal is compensated for errors using an environmental adaptive calibration algorithm. The compensated displacement signal includes: Based on the load signal data sequence in the standardized data sequence, data segments with load change rate lower than the preset change threshold are identified as the zero stress change reference period; Based on the displacement signal sequence, temperature signal sequence, and vibration signal sequence corresponding to the zero stress change reference period, statistical analysis is performed to obtain the effective temperature compensation coefficient and vibration interference sensitivity. Throughout the loading process, based on the effective temperature compensation coefficient, the temperature offset component of each sensor is subtracted in real time from the displacement signal acquired in real time to obtain the displacement signal after preliminary temperature compensation. Based on the vibration interference sensitivity and the vibration signal acquired in real time, the amplitude of the displacement signal noise floor introduced by vibration is calculated, and then subtracted from the displacement signal initially compensated for by temperature to obtain the final compensated displacement signal.
[0045] In this embodiment, after completing the above two compensation steps, it is determined whether the fluctuation range of the corresponding compensated displacement signal exceeds the standard. If not, the compensation is effective; otherwise, recalibration is performed or a warning is issued to check the sensor.
[0046] The compensation displacement signal is analyzed in real time by edge computing nodes, load-displacement curves are generated and key parameters are calculated, and an early warning is triggered when the parameters exceed the threshold.
[0047] In this embodiment, the compensated displacement signal is analyzed in real time by edge computing nodes. Based on the compensated displacement and load information, the load-displacement relationship curve is plotted in real time, and key early warning parameters are calculated, including the displacement rate sequence and the graded rebound rate sequence. The load-displacement relationship curve is plotted and updated in real time based on the load value and the corresponding pile top displacement value at the load stabilization moment. The displacement rate sequence is calculated in real time based on the first-order difference of the displacement signal with respect to time. The graded rebound rate sequence is calculated in real time based on the ratio of the displacement recovery amount before and after unloading of each load level to the loading displacement amount.
[0048] In detail, based on the initial loading data, a statistical process control method is used to calculate the moving average and moving standard deviation of the displacement rate sequence and the graded rebound rate sequence, respectively; the warning threshold is set as follows: , where W is a coefficient preset according to the safety level of the foundation pile; this threshold is dynamically updated as the loading process progresses, adapting to the actual response level of the current test pile; By comparing the measured parameters with the dynamic early warning threshold in real time and combining the load-displacement curve shape, an anomaly warning is triggered when the following two conditions are met simultaneously: Condition 1 is that the displacement rate sequence continuously exceeds its upper warning threshold for a preset number of times, or the graded rebound rate sequence continuously falls below its lower warning threshold. Condition 2 is that the real-time load-displacement curve deviates significantly from the ideal elastic reference curve fitted based on the initial data. The degree of deviation is quantified by calculating the area difference between the two curves or the displacement deviation of key points, and exceeds the preset threshold. When the above collaborative diagnostic conditions are met, the system automatically determines that it is abnormal and triggers different levels of early warning actions according to the severity of the abnormal deviation. At the same time, the abnormal time, key parameter snapshots and diagnostic basis are recorded in the real-time report.
[0049] When an early warning is triggered, the deformation state of the pile body is restored based on three-dimensional visualization technology, and the detection results are intelligently verified by combining preset parameter models of different pile types, and a detection report is output.
[0050] like Figure 2 As shown, this embodiment also discloses a pile self-balancing static load displacement detection system, which uses the above-mentioned pile self-balancing static load displacement detection method, including: The deployment module, based on digital twin model and multi-condition finite element simulation, uses machine learning and multi-objective optimization algorithms to determine the deployment location of distributed fiber optic displacement sensors on the pile, and deploys temperature sensors and vibration sensors. The loading acquisition module synchronously acquires load signals, displacement signals, temperature signals, and vibration signals during the self-balancing loading process, and performs time sequence alignment and delay compensation to form a standardized data sequence. The compensation module calls the temperature signal and vibration signal, and uses an environmental adaptive calibration algorithm to perform error compensation on the displacement signal to obtain the compensated displacement signal. The early warning module analyzes the compensated displacement signal in real time through edge computing nodes, generates load-displacement curves and calculates key parameters, and triggers an early warning when the parameters exceed the threshold.
[0051] Optionally, in this embodiment, those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.
[0052] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0053] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more electronic devices to execute all or part of the steps of the methods described in the various embodiments of this application.
[0054] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.
[0055] In the several embodiments provided in this application, it should be understood that the disclosed application can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of units or modules may be electrical or other forms.
[0056] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0057] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0058] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.
Claims
1. A method for detecting the static load displacement of a foundation pile under self-balancing load, characterized in that, The method includes: Based on digital twin models and multi-condition finite element simulations, the placement of distributed fiber optic displacement sensors on the pile body is determined through machine learning and multi-objective optimization algorithms, and temperature and vibration sensors are also deployed. During the self-balancing loading process, load signals, displacement signals, temperature signals and vibration signals are collected simultaneously, and time sequence alignment and delay compensation are performed to form a standardized data sequence. By calling temperature and vibration signals, and using an environmental adaptive calibration algorithm to compensate for errors in the displacement signal, a compensated displacement signal is obtained. The compensation displacement signal is analyzed in real time by edge computing nodes, load-displacement curves are generated and key parameters are calculated, and an early warning is triggered when the parameters exceed the threshold.
2. The method for detecting the static load displacement of a foundation pile according to claim 1, characterized in that, The determination of the placement of distributed fiber optic displacement sensors on the pile body based on digital twin models and multi-condition finite element simulations, using machine learning and multi-objective optimization algorithms, includes: Based on building information modeling, geological survey and hydrogeological information, a digital twin pile model is constructed by integrating the geometric structure, material properties, soil parameters and seepage field of the pile; the uncertain parameters in the model are identified and quantified to form an adjustable parameter set; Based on the digital twin pile model and its adjustable parameter set, multi-condition parametric finite element simulation is performed, outputting massive amounts of simulation data; machine learning algorithms are used to analyze the simulation data to identify engineering events that have a key impact on the pile body response and their corresponding spatial distribution. Based on the spatial distribution of various engineering events, a dynamic monitoring demand hotspot map is generated along the pile body, which is marked with different physical quantity monitoring priorities. Based on the dynamic monitoring demand heat map, an improved multi-objective optimization algorithm is used to generate the optimal sensor deployment scheme; Based on the optimal sensor deployment scheme, corresponding distributed fiber optic displacement sensors are deployed on the pile body.
3. The method for detecting the static load displacement of a foundation pile according to claim 2, characterized in that, The aforementioned method, based on the spatial distribution of various engineering events, integrates and generates a dynamic monitoring demand hotspot map along the pile body, marked with different physical quantity monitoring priorities, including: Based on each identified engineering event, the spatial and probabilistic features corresponding to the engineering event are extracted and quantified to obtain the feature vector of the engineering event. The feature vector includes the core depth range of the event occurrence, the probability density function of occurrence distributed along the pile depth, the spatial influence radius of the mechanical effect of the event, and the set of key physical quantities used to characterize the process event. Based on the aforementioned feature vectors and massive simulation data, an event occurrence probability factor field, a mechanical response mutation factor field, and a model prediction divergence factor field are constructed respectively. Based on the constructed event occurrence probability factor field, mechanical response mutation factor field, and model prediction divergence factor field, as well as the preset dynamic weight function related to the loading stage, a three-dimensional dynamic decision field is obtained by weighted fusion. The dynamic weight function is configured such that: in the early stage of loading, the model prediction divergence factor field dominates the fusion; in the middle and late stages of loading, the mechanical response mutation factor field dominates the fusion. Based on the three-dimensional dynamic decision field, slices and projections are performed according to preset key load stages and physical quantity types to obtain a series of two-dimensional dynamic monitoring demand hotspot maps distributed along the pile body. The dynamic monitoring demand hotspot maps are used to identify the monitoring priority of different physical quantities at different depth locations.
4. The method for detecting the static load displacement of a foundation pile according to claim 3, characterized in that, The optimal sensor deployment scheme, generated based on a dynamic monitoring demand heatmap and using an improved multi-objective optimization algorithm, includes: Based on the detection priority of each physical quantity corresponding to each position in the dynamic monitoring demand hotspot map, an information potential energy field along the pile depth-physical quantity distribution is constructed. Based on the information potential field, and based on a weighted random sampling strategy where the probability density is proportional to the potential energy, a population containing multiple initial deployment schemes is generated, provided that the preset range of the number of sensors is met. Based on each initial deployment scheme and the preset multi-dimensional fitness index quantification formula, the corresponding weighted comprehensive evaluation value is calculated, and the roulette wheel selection method is used to select excellent individuals from the current population. Based on the information potential field, a spatial gradient field is generated to identify areas where monitoring demand changes abruptly. Based on the spatial gradient field, the selected individuals are guided to undergo crossover and mutation operations to obtain the offspring population. After obtaining the offspring population, continue to calculate the weighted comprehensive evaluation value until the preset number of iterations is met, and output the offspring population as the optimal solution set for multi-objective optimization. Based on the optimal solution set, the individual with the largest weighted comprehensive rating value is selected as the optimal sensor deployment scheme.
5. The method for detecting the static load displacement of a foundation pile according to claim 4, characterized in that, During the self-balancing loading process, load signals, displacement signals, temperature signals, and vibration signals are simultaneously acquired, and time sequence alignment and delay compensation are performed to form a standardized data sequence, including: Based on the loading start command, the synchronous acquisition unit triggers and receives load signals from the hydraulic servo loading module, displacement signals from the distributed fiber optic displacement sensor network, and temperature and vibration signals from the temperature and vibration sensors in parallel. Each sampling point's load signal, displacement signal, temperature signal, and vibration signal is labeled with a microsecond-level time tag with a unified reference, thus obtaining a set of multi-source raw data streams with time tags; Based on the multi-source raw data stream with time tags, all its time tags are mapped to a common relative time axis with the loading and startup time as the zero point; Based on the fixed delay parameters of each signal transmission channel pre-calibrated, deterministic time offset compensation is performed on each data stream mapped to the common time axis, thereby obtaining a standardized data sequence that is initially aligned on the time axis.
6. The method for detecting the static load displacement of a foundation pile according to claim 5, characterized in that, The temperature and vibration signals are invoked, and the displacement signal is compensated for errors using an environmental adaptive calibration algorithm to obtain the compensated displacement signal, which includes: Based on the load signal data sequence in the standardized data sequence, data segments with load change rate lower than the preset change threshold are identified as the zero stress change reference period; Based on the displacement signal sequence, temperature signal sequence, and vibration signal sequence corresponding to the zero stress change reference period, statistical analysis is performed to obtain the effective temperature compensation coefficient and vibration interference sensitivity. Throughout the loading process, based on the effective temperature compensation coefficient, the temperature offset component of each sensor is subtracted in real time from the displacement signal acquired in real time to obtain the displacement signal after preliminary temperature compensation. Based on the vibration interference sensitivity and the vibration signal acquired in real time, the amplitude of the displacement signal noise floor introduced by vibration is calculated, and then subtracted from the displacement signal initially compensated for by temperature to obtain the final compensated displacement signal.
7. The method for detecting the static load displacement of a foundation pile according to claim 1, characterized in that, The method further includes: when an early warning is triggered, restoring the deformation state of the pile body based on three-dimensional visualization technology, and intelligently verifying the detection results by combining preset parameter models of different pile types, and outputting a detection report.
8. A pile self-balancing static load displacement detection system, implementing the pile self-balancing static load displacement detection method according to any one of claims 1 to 7, characterized in that, include: The deployment module, based on digital twin model and multi-condition finite element simulation, uses machine learning and multi-objective optimization algorithms to determine the deployment location of distributed fiber optic displacement sensors on the pile, and deploys temperature sensors and vibration sensors. The loading acquisition module synchronously acquires load signals, displacement signals, temperature signals, and vibration signals during the self-balancing loading process, and performs time sequence alignment and delay compensation to form a standardized data sequence. The compensation module calls the temperature signal and vibration signal, and uses an environmental adaptive calibration algorithm to perform error compensation on the displacement signal to obtain the compensated displacement signal. The early warning module analyzes the compensated displacement signal in real time through edge computing nodes, generates load-displacement curves and calculates key parameters, and triggers an early warning when the parameters exceed the threshold.