Method and system for multi-source deformation monitoring and early warning of super high-rise large cantilever concrete structure in construction stage

Abstract

This invention discloses a method and system for multi-source deformation monitoring and early warning during the construction phase of ultra-high-rise large cantilever concrete structures; belonging to the field of early warning technology for concrete structure construction. The early warning method of this application abandons the traditional fixed threshold method based on the Gaussian distribution assumption. This invention utilizes a generalized Pareto distribution to model the tail characteristics of the shaping residuals; it can adaptively adjust the early warning threshold according to the statistical characteristics of real-time data; combined with a hierarchical mechanism of trend early warning and extreme value alarm, it effectively reduces the risk of missed reporting of sudden and hidden damage, significantly improving the safety assurance level of the construction process.

Description

Technical Field

[0001] This invention belongs to the field of early warning technology for concrete structure construction, specifically relating to a method and system for multi-source deformation monitoring and early warning during the construction phase of super high-rise large cantilever concrete structures. Background Technology

[0002] As a crucial component of modern urban landmark buildings, super high-rise cantilevered concrete structures involve complex construction processes, long construction periods, and significant environmental influences. These structures undergo several critical phases during construction, including formwork erection, layered concrete pouring, support system conversion and unloading, and secondary loading, accompanied by continuous changes in load paths and frequent switching of boundary conditions. Furthermore, concrete itself exhibits time-varying characteristics such as early-age shrinkage and creep, and the construction site environment, including temperature, sunlight, and wind loads, is complex and variable.

[0003] Existing construction monitoring technologies mainly rely on traditional manual measurement methods such as total stations and levels, or use a single type of sensor for point monitoring. These traditional methods have significant shortcomings when dealing with super high-rise cantilever structures: First, existing monitoring is often based on simple comparisons with static design benchmark values, ignoring the sudden changes in structural stiffness caused by changes in construction conditions such as support unloading and normal deformation drift caused by time-varying material properties, which can easily lead to a large number of false alarms or missed alarms.

[0004] Secondly, monitoring data is severely affected by environmental noise, such as periodic deformation caused by temperature. Traditional filtering methods struggle to effectively decouple environmental effects from minor anomalies caused by actual structural damage, leading to delayed early warnings. Thirdly, existing early warning thresholds are typically fixed values ​​based on experience or Gaussian distribution assumptions, which cannot adapt to the non-stationary characteristics of structural responses during construction and are insufficient to detect extreme anomalies with low probability but severe consequences.

[0005] Therefore, there is an urgent need for a multi-source deformation monitoring and early warning method and system for the construction stage of ultra-high-rise large cantilever concrete structures that can adapt to complex time-varying working conditions and accurately eliminate environmental interference. Summary of the Invention

[0006] To address the deficiencies in the aforementioned technical solutions, the present invention aims to provide a method and system for multi-source deformation monitoring and early warning during the construction phase of ultra-high-rise large cantilever concrete structures.

[0007] To achieve the above objectives, the first aspect of this invention provides a method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure, comprising the following steps: S1: Establish a refined finite element model of the super high-rise cantilever structure, and construct a stiffness topology state space based on the construction condition sequence to generate a dynamic physical reference sequence that changes with the construction condition.

[0008] The refined finite element model needs to accurately define the geometric dimensions, material properties, and boundary conditions of the components. Especially for concrete components, a material constitutive model reflecting their time-varying characteristics is required. The construction condition sequence... The continuous construction process is discretized into a series of quasi-static structural states. Using the birth and death element technique, each working condition is addressed... Activate or deactivate the corresponding unit to calculate the theoretical displacement field under that working condition. and stress field .

[0009] The dynamic physical reference sequence The calculation formula combines the sudden stiffness changes caused by the switching of working conditions and the early-age shrinkage and creep effects of concrete: in, The time correlation coefficient reflects the effects of creep and shrinkage.

[0010] S2: Real-time acquisition of multi-source sensor data of the super high-rise cantilever structure, combined with dynamic physical reference sequence to calculate physical deviation, and use CEEMDAN algorithm to perform modal decomposition of physical deviation to extract environmental trend term and high frequency residual term.

[0011] The multi-source sensor data The physical deviation is calculated after time synchronization and resampling of data including displacement, strain, temperature, and wind speed. .

[0012] Using the Adaptive Noise Complete Set Empirical Mode Decomposition (CEEMDAN) algorithm to... Decompose: Based on the frequency characteristics of each intrinsic mode function (IMF) component and its correlation with environmental data, the decomposition results are reconstructed into environmental trend terms. and high-frequency residuals .

[0013] S3: Establish a gray prediction GM(1,1) model for metabolism to perform rolling prediction of environmental trend terms, calculate the shaping residuals, and eliminate non-damaging drift caused by environmental and time-varying factors.

[0014] Select the most recent A metabolic GM(1,1) model is constructed using data points. The time response function is solved by accumulating data, establishing the whitening differential equation, and estimating parameters. The predicted environmental impact value for the next time step is obtained by cumulative subtraction and restoration. As new data is acquired, the model updates its parameters in real time by removing the oldest data. and Residual deformation Defined as physical deviation minus the predicted environmental trend term: The shaping residual is approximately a stationary random sequence under ideal, damage-free conditions, and can sensitively reflect the statistical property drift caused by structural damage.

[0015] S4: Model the tail features of the shaping residual based on the generalized Pareto distribution (GPD), calculate the dynamic hierarchical early warning threshold, and trigger the corresponding early warning response based on the comparison result between the real-time shaping residual and the threshold.

[0016] Selecting the shaping residual Exceeding the baseline noise level Using the data as a super-threshold sequence, the scaling parameter of the generalized Pareto distribution (GPD) is estimated in real time using the maximum likelihood estimation (MLE) method. and shape parameters Based on a preset extremely high risk probability The inverse solution yields the warning threshold that dynamically changes over time. : Based on real-time shaping residuals With dynamic threshold The comparison results and the significance test results of the residual mean trigger a graded early warning response, including trend warning and extreme value alarm.

[0017] Furthermore, in step S2, the multi-source sensing data is acquired through a sensor network deployed at key stress-bearing parts of the cantilever structure. The sensor network includes a GNSS receiver, a fiber optic grating sensor, a temperature sensor, and an anemometer.

[0018] Further, in step S3, the sequence length of the metabolic GM(1,1) model is... The optimal value is 10 to balance the sensitivity and stability of the prediction.

[0019] Further, in step S4, the shape parameters Changes directly reflect the evolution of damage risk, when The time corresponds to a heavy-tailed distribution, indicating a higher probability of extreme anomalies occurring.

[0020] The second aspect of this invention provides a multi-source deformation monitoring and early warning system for the construction stage of a super high-rise large cantilever concrete structure, comprising: a physical benchmark construction module for establishing a refined finite element model, constructing a stiffness topology state space based on the construction condition sequence, and generating a dynamic physical benchmark sequence; a data acquisition and decoupling module for real-time acquisition of multi-source sensor data, calculation of physical deviations, and extraction of environmental trend terms and high-frequency residual terms using the CEEMDAN algorithm; a residual shaping and trend elimination module for establishing a metabolic grey prediction GM(1,1) model to perform rolling prediction of environmental trend terms and calculate the shaped residuals; and a dynamic early warning decision module for modeling the tail features of the shaped residuals based on GPD, calculating dynamic hierarchical early warning thresholds, and triggering an early warning response based on the comparison results.

[0021] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention introduces stiffness topology state space construction technology, which generates a dynamic physical benchmark sequence that changes with working conditions based on a refined finite element model and birth and death element technology. This enables the monitoring system to predict and adapt to normal stiffness changes caused by working condition switching such as pouring and unloading, avoids false alarms caused by working condition changes in the traditional static threshold method, and ensures the physical authenticity and real-time performance of the comparison benchmark. 2. This invention employs a combination of CEEMDAN modal decomposition and a gray prediction model of metabolism to deeply decouple physical deviations. This not only effectively filters out high-frequency measurement noise, but more importantly, it accurately removes non-destructive trend drift caused by temperature, sunlight, and concrete shrinkage and creep from non-stationary monitoring data, extracting pure shaping residuals. This significantly improves the system's sensitivity to minor structural damage, solving the problem of delayed early warning under strong environmental interference. 3. The technical solution of this invention abandons the traditional fixed threshold method based on the Gaussian distribution assumption. This invention uses the generalized Pareto distribution to model the tail characteristics of the shaping residual; it can adaptively adjust the warning threshold according to the statistical characteristics of real-time data; and with the hierarchical mechanism of trend warning and extreme value alarm, it effectively reduces the risk of missed reporting of sudden and hidden damage, and significantly improves the safety assurance level of the construction process. Attached Figure Description

[0022] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of a multi-source deformation monitoring and early warning method for a super high-rise large cantilever concrete structure during the construction stage, provided as an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. The illustrative embodiments and descriptions of this invention are for illustrative purposes only and are not intended to limit the invention. The embodiments described below are some, but not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0024] In the following description, numerous specific details are set forth to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other embodiments, well-known structures, materials, or methods are not specifically described to avoid obscuring the invention. Unless otherwise specified, the materials, instruments, and reagents used in the following embodiments are commercially available. Unless otherwise specified, the techniques used in the embodiments are conventional methods well known to those skilled in the art.

[0025] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0026] Example 1 This embodiment provides a method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure. Figure 1 The overall flowchart of the method is shown, as follows: Figure 1 As shown, the method includes the following steps: Step S1: Establish a refined finite element model of the super high-rise cantilever structure, and construct a stiffness topology state space based on the construction condition sequence to generate a dynamic physical reference sequence that changes with the construction conditions.

[0027] The core objective of this step is to construct a digital twin benchmark that can evolve in real time with the construction progress, thereby addressing the benchmark drift problem caused by frequent changes in construction conditions in traditional monitoring. In the construction of super high-rise buildings, the physical boundary conditions and stiffness matrix of the structure are not static but undergo drastic changes with each concrete pour and each removal of temporary supports. If static design values ​​are used as the benchmark, these normal stiffness changes will be misjudged as anomalies.

[0028] Specifically, based on the design drawings of the super high-rise cantilever structure, the finite element analysis software, specifically ABAQUS, is used in this embodiment to establish a three-dimensional refined model that includes key components such as the core tube, cantilever frame, and temporary support system.

[0029] Furthermore, the model needs to accurately define the geometric dimensions of the components, material properties such as elastic modulus, Poisson's ratio, density, and boundary conditions.

[0030] Specifically, for concrete components, a material constitutive model that reflects their time-varying characteristics should be used, such as considering the growth function of the elastic modulus over time.

[0031] Subsequently, the construction condition sequence was defined. Each operating condition This represents a specific stage in the construction process, which is determined based on the actual construction plan.

[0032] For example: the The first layer of concrete pouring was completed. Temporary support unloading, curtain wall installation, etc.; this discretized working condition definition divides the continuous construction process into a series of quasi-static structural states, so that the structural stiffness matrix of each state is determined.

[0033] Then, a stiffness topology state space is constructed; this is achieved using the birth and death element technique in ABAQUS finite element software; for each working condition... In the finite element model, corresponding elements are activated or deactivated to simulate the actual construction process of the structure. For example, activating newly poured concrete elements and formwork support elements during the pouring stage will increase the stiffness matrix of the structure by the corresponding terms; deactivating temporary support elements during the unloading stage will decrease the stiffness matrix of the structure by the corresponding terms.

[0034] In this way, each working condition can be calculated. Theoretical displacement field of the substructure and stress field .

[0035] Finally, a dynamic physical reference sequence is generated. ; Specifically, this requires a focus on the shrinkage and creep effects of concrete in its early stages. In actual engineering, concrete is not an ideal elastic body; under continuous load, it will undergo creep deformation that increases over time, and this deformation is particularly significant in its early stages. If this physical characteristic is ignored, the monitored creep displacement will be misinterpreted as structural stiffness degradation.

[0036] Therefore, this embodiment introduces time-varying time... Correction factor for change This coefficient is calculated based on the concrete creep prediction model; in this embodiment, the CEB-FIP model is used. The dynamic physical reference sequence is expressed as: in: :time The dynamic physical reference value, whose physical meaning is the normal response value that the structure should theoretically exhibit under the current working conditions and the current age of the concrete; : No. The finite element theoretical displacement under each working condition was calculated by the birth and death element method; this reflects the sudden change in stiffness caused by structural topology changes such as the removal of supports, which is an acceptable and legal structural deformation. : The time correlation coefficient reflecting the effects of creep and shrinkage, with a value range of arrive The specific values ​​depend on the age of the concrete and the ambient humidity; this reflects the slow drift caused by the rheological properties of the material, which is also an acceptable physical phenomenon.

[0037] : The current moment, corresponding to the operating condition The duration.

[0038] The above formula The construction is mainly based on time-varying structural mechanics; among which Corresponding to the stiffness matrix in the structural dynamics equations The step change solved the problem of false alarms during operating condition switching; and This corresponds to the viscoelastic component in the material constitutive relation, solving the problem of false alarms related to creep drift. By multiplying these two physical terms, this method successfully decomposes the structural response into deterministic physical components and detectable anomalous components, laying a solid physical foundation for subsequent residual analysis.

[0039] Specifically, it can be deduced that any deviation The signal must originate from factors not included in the model, such as environmental interference or actual damage, thus eliminating false alarms caused by normal construction activities.

[0040] Step S2: Collect multi-source sensor data of the super high-rise cantilever structure in real time, and calculate the physical deviation by combining it with the dynamic physical reference sequence. Use the CEEMDAN algorithm to perform modal decomposition on the physical deviation and extract the environmental trend term and high-frequency residual term.

[0041] Step S2 aims to extract the true structural response deviation from the measured data that is severely affected by environmental disturbances.

[0042] At super high-rise construction sites, sensor data often includes not only structural deformation, but also strong temperature effects such as expansion and contraction caused by sunlight, wind load vibration, and measurement noise; if decoupling is not performed, these environmental noises will mask minute damage signals.

[0043] Specifically, in step S2, a multi-source sensor network is first deployed. The multi-source sensor data is acquired through a sensor network deployed at key stress-bearing parts of the cantilever structure. The sensor network includes a GNSS receiver, a fiber optic grating sensor, a temperature sensor, and an anemometer. GNSS receivers are installed at key stress-bearing parts of the cantilever structure, such as the root and end, to obtain three-dimensional displacement data. Fiber optic grating sensors are installed to obtain strain data of key components. Temperature sensors and anemometers are also deployed to obtain environmental data.

[0044] Furthermore, all data needs to be time-synchronized and resampled by a data acquisition instrument to ensure the consistency of the time reference, with the sampling frequency set to 1Hz.

[0045] Then calculate the physical deviation. Physical deviation is defined as the difference between the measured value and the theoretical reference value. in: :time Physical deviation; :time Measured data of displacement or strain; :time The dynamic physical reference value is calculated in step S1; because The effects of sudden changes in operating conditions and creep have been eliminated, therefore The remaining components mainly include: reversible deformation caused by environmental factors such as temperature, sunlight, and wind load, high-frequency noise of the measurement system, and abnormal deformation caused by potential structural damage.

[0046] Then, in order to decouple environmental effects and noise, physical deviations are... Adaptive noise-complete ensemble empirical mode decomposition (EMD) is performed. The CEEMDAN algorithm, by adding adaptive white noise to assist decomposition, effectively solves the mode aliasing problem in traditional EMD methods, and can decompose non-stationary signals into a series of intrinsic mode functions (IMFs). The decomposition process is as follows: in: : No. Each intrinsic mode function component represents signal fluctuations at different time scales from high frequency to low frequency. Residual components represent the extremely low frequency trend of the signal; The total number of IMFs obtained from the decomposition; Finally, based on the frequency characteristics of each IMF component and the correlation analysis with environmental data, the decomposition results were reconstructed into two parts: environmental trend items. and high-frequency residuals .

[0047] The specific steps are as follows: First, calculate the components obtained from each decomposition and compare them with the measured ambient temperature data. Pearson correlation coefficient between : in, Describing covariance, It represents the standard deviation.

[0048] Then, define a set of IMF indexes that are strongly correlated with environmental data. The term "strong correlation" refers to a correlation coefficient whose absolute value is greater than a preset correlation threshold. Specifically, in this embodiment The value is 0.6, that is: Based on the above index set Construct environmental trend items : The remaining high-frequency, low-energy IMF components contain measurement noise and possible sudden damage signals, and are classified as high-frequency residual terms. : In the above formula, the physical basis for using CEEMDAN to decompose signals is that physical signals from different sources have different characteristic time scales; thermal deformation caused by temperature and solar radiation usually has obvious diurnal periodicity, i.e. low frequency, which corresponds to the low-order IMF components decomposed by CEEMDAN; while wind load vibration, measurement noise and sudden stress release caused by structural damage usually manifest as high-frequency or transient signals, corresponding to high-order IMF components.

[0049] By linking specific IMF components to measured temperature data through correlation analysis, a physical source tracing is essentially completed, confirming which part of the deformation is caused by temperature. This blind source separation based on physical characteristic scale is more adaptable than traditional frequency domain filtering because it does not assume that the signal is stationary and can adapt to non-stationary environmental interference at the construction site, thereby accurately separating environmental effects from the total deviation and retaining residual components containing damage information.

[0050] Step S3: Establish a gray prediction model for metabolism, GM(1,1), to perform rolling predictions on environmental trend terms, calculate the shaped residuals, and eliminate non-damaging drift caused by environmental and time-varying factors.

[0051] Step S3 aims to further eliminate nonlinear drift in the environmental trend term to obtain a pure damage characteristic signal. This is because, although CEEMDAN isolates the environmental term, environmental impacts themselves are non-stationary, such as seasonal temperature variations. Furthermore, the accuracy of the traditional GM(1,1) model decays over time in long-term predictions, while the metabolic model adapts to the dynamic changes of the system by continuously updating the data sequence, making it highly suitable for handling non-stationary data during construction.

[0052] Specifically, regarding the separated environmental trend items A gray prediction model for metabolism, GM(1,1), is established. The construction steps are as follows: Step 1: Sequence Selection: Select the nearest sequence. A sequence of data points As a modeling sequence, where Corresponding time Environmental trend items In this embodiment The value is set to 10 to balance the sensitivity and stability of the prediction.

[0053] Step 2: Accumulation Generation: Perform an accumulation generation (1-AGO) on the original sequence to obtain the sequence. ,in: The purpose of cumulative generation is to weaken the randomness of the original sequence, enhance its regularity, and make it exhibit an exponential growth pattern, thus making it suitable for differential equation modeling.

[0054] Step 3: Establish the whitening differential equation: in: The development coefficient reflects the development trend of the sequence, i.e., the rate of growth or decline. Grey action quantity reflects the endogenous change relationship between data and the combined effect of external inputs.

[0055] Step 4: Parameter estimation: Estimate parameters using the least squares method. : Among them, data matrix and data vector They are respectively: matrix The nearest neighbor mean of the cumulative generated sequence is constructed and used to approximate the background value in the differential equation.

[0056] Step 5: Solve the prediction model: Solve the differential equation to obtain the time response function. .

[0057] Step 6: Cumulative Subtraction Restoration: Obtain the predicted value of the original sequence through cumulative subtraction restoration. This is the predicted environmental impact value for the next time step. .

[0058] Step 7: Metabolic Renewal: When new measured data is obtained Then, it is added to the end of the sequence, while the oldest data is removed. Maintain sequence length as Repeat steps 2-6 above to update the model parameters in real time. and .

[0059] Next, the shaping residuals are calculated. The shaping residual is defined as the physical bias minus the predicted environmental trend term. Specifically, the physical significance of introducing the metabolic GM(1,1) model lies in constructing an adaptive environmental filter. Traditional regression models often assume that environmental influence parameters are constant, but in the construction of super high-rise buildings, the temperature and wind field characteristics change non-stationarily as the structural height increases. Grey system theory excels at handling uncertainties with small samples and limited information. The metabolic mechanism simulates the system's memory and adaptation to the latest state, capturing the local evolutionary trend of environmental influences in real time.

[0060] The residual is obtained by predicting and subtracting this trend. Physically, it represents the remaining energy that was not eliminated by normal operating conditions (step S1) and normal environmental trends (step S3).

[0061] In an ideal, damage-free state, this residual energy should consist solely of white noise, exhibiting a stationary sequence with zero mean and constant variance. However, once real damage occurs in the structure, such as slippage or crack propagation, it disrupts the original physical equilibrium, leading to... Non-stationary statistical property drift occurs, such as mean shift or a sharp increase in variance.

[0062] Therefore, this step transforms the complex problem of physical damage identification into a statistically non-stationary detection problem, greatly improving the detection sensitivity.

[0063] Step S4: Model the tail features of the shaping residual based on the generalized Pareto distribution, calculate the dynamic hierarchical early warning threshold, and trigger the corresponding early warning response based on the comparison result between the real-time shaping residual and the threshold.

[0064] Step S4 aims to establish a rigorous probability early warning mechanism. Compared with the traditional threshold setting based on Gaussian distribution, extreme value statistics pays more attention to the tail of the distribution, that is, extreme events with low probability of occurrence but serious consequences, which is highly consistent with the need for early warning of structural damage.

[0065] Specifically, regarding plastic surgery residuals The Peak Overthreshold (POT) model in EVS is used for analysis. First, the shaped residual is selected. Exceeding the baseline noise level Data as over-threshold sequence ; Base noise level Can be taken as The 90th percentile of the sequence. According to the Pickands-Balkema-deHaan theorem, when the threshold... When sufficiently high, the amount exceeding the threshold The distribution of approximately follows a generalized Pareto distribution.

[0066] The cumulative distribution function (CDF) of GPD is: in: : The amount exceeding the threshold, i.e. The range of values ​​is (when )or (when ); Scale parameters This determines the degree of dispersion of the distribution and reflects the amplitude of residual fluctuations; The shape parameter determines the thickness of the tail of the distribution.

[0067] Specifically, parameters The physical warning means: when When the distribution corresponds to the short-tailed Weibull domain, it means that the residual has a definite upper limit, which physically corresponds to the structure being in an elastically stable state where the deformation is bounded.

[0068] when When the distribution corresponds to the heavy-tailed Fréchet domain, it means that the probability of extreme anomalies increases significantly. This physically corresponds to the structure entering the stage of damage accumulation or nonlinear instability. At this time, a small disturbance may trigger a huge response, indicating an extremely high risk of failure.

[0069] Therefore, through real-time monitoring Changes in these parameters can provide insights into the evolution of the structure's physical stability from a probabilistic and statistical perspective. Furthermore, by employing the maximum likelihood estimation (MLE) method, based on historical shaped residual data within a sliding window, the parameters of the GPD can be estimated in real time. and .

[0070] Next, based on the preset extremely high risk probability For example, in this embodiment, we take That is, a failure probability of one in ten thousand, from which the corresponding dynamic threshold can be solved. That is, to seek Make : The threshold It changes dynamically over time, based on the current statistical characteristics of the residuals, i.e., by... and It reflects automatic adjustment; when the fluctuation of the monitored data increases or heavy-tailed characteristics appear, that is, the structure tends to be unstable, the threshold will automatically tighten or loosen in order to maintain a constant risk level.

[0071] Finally, a tiered early warning mechanism is constructed: Level I trend early warning: monitoring the shaping residuals through a sliding T-test. The mean change; if the test statistic is significant (P value < 0.05), it indicates the existence of systematic bias, suggesting that there may be loosening or slow deformation of the support, and it is recommended to increase the monitoring frequency and check the on-site support condition.

[0072] Level II extreme value alarm: When a single real-time shaped residual value... Exceeding the dynamic threshold When this occurs, it indicates that an extremely rare abnormal event has occurred, which usually corresponds to sudden damage such as local buckling or weld cracking. It is recommended to immediately suspend construction and conduct a detailed on-site inspection.

[0073] This embodiment constructs a multi-source deformation monitoring and early warning system for the construction stage of a super high-rise large cantilever concrete structure through the above steps and methods. The system includes: a physical benchmark construction module, a data acquisition and decoupling module, a residual shaping and trend elimination module, and a dynamic early warning decision module. The physical benchmark construction module is used to establish a refined finite element model, construct a stiffness topology state space based on the construction condition sequence, and generate a dynamic physical benchmark sequence. The physical benchmark construction module includes a pre-set concrete creep prediction model, which is used to calculate the time correction coefficient that changes over time in order to correct the dynamic physical benchmark sequence.

[0074] The data acquisition and decoupling module is used to acquire multi-source sensor data in real time, calculate physical deviations, and use the CEEMDAN algorithm to extract environmental trend terms and high-frequency residual terms. The data acquisition and decoupling module is connected to a multi-source sensor network, which includes: a global navigation satellite system receiver for acquiring three-dimensional displacement data of key nodes of the cantilever structure; a fiber optic grating sensor for acquiring strain data of key components; and a temperature sensor and an anemometer for acquiring ambient temperature and wind speed data. The data acquisition and decoupling module is also used to perform time synchronization and resampling processing on all acquired data.

[0075] The residual shaping and trend elimination module is used to establish a metabolism grey prediction GM(1,1) model to perform rolling prediction of environmental trend items and calculate the shaped residuals. The dynamic early warning decision module is used to model the tail features of the shaped residual based on GPD, calculate the dynamic hierarchical early warning threshold, and trigger an early warning response based on the comparison results.

[0076] The dynamic early warning decision module is configured to: assess the risk of damage evolution based on the changing trend of the generalized Pareto distribution shape parameter estimated in real time, and dynamically adjust the dynamic graded early warning threshold according to the preset risk probability.

[0077] From step S1, eliminating the influence of working conditions, to step S2, separating physical signals, then to step S3, eliminating environmental trends to obtain pure residuals, and finally to step S4, making decisions based on probabilities to quantify risks, a multi-source deformation monitoring and early warning system for the construction stage of super high-rise cantilever concrete structures effectively solves the problems of delayed early warning and high false alarm rate caused by complex working conditions and variable environments during the construction of super high-rise cantilever structures.

[0078] It should be understood that the terms "system," "device," "unit," and / or "module" as used in this specification are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0079] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0080] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0081] It should be noted that the structures, proportions, sizes, etc., illustrated in the accompanying drawings are merely for illustrative purposes to aid those skilled in the art and are not intended to limit the scope of the invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in proportions, or adjustments to size, without affecting the effectiveness and purpose of the invention, should still fall within the scope of the disclosed technical content. Furthermore, terms such as "upper," "lower," "left," "right," and "middle" used in this specification are merely for clarity and not intended to limit the scope of the invention. Changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention.

Claims

1. A method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure, characterized in that, Includes the following steps: S1. Establish a refined finite element model of the super high-rise cantilever structure, and construct a stiffness topology state space based on a preset construction condition sequence; wherein, the construction of the stiffness topology state space is achieved through the birth and death element technique in the finite element software; for each condition... In the finite element model, corresponding elements are activated or deactivated to simulate the actual construction process of the structure; the working conditions for each condition are calculated. Theoretical displacement field of the substructure and stress field A time correction factor reflecting the effects of concrete shrinkage and creep is introduced. The theoretical displacement field is corrected, and finally, a dynamic physical reference sequence is generated. The dynamic physical reference sequence is represented as follows: ;in: :time The dynamic physical reference value; : The time correlation coefficient reflecting the effects of creep and shrinkage; : The current moment, corresponding to the operating condition The duration; : No. The finite element theoretical displacement under each working condition is calculated by the birth and death element method; S2. Real-time acquisition of multi-source sensor data from super high-rise cantilever structures. The physical deviation is calculated in conjunction with the dynamic physical reference sequence. ;in: :time Physical deviation; :time Measured data of displacement or strain; :time The dynamic physical reference value is calculated in step S1; for physical deviations Adaptive noise-complete ensemble empirical mode decomposition is performed. The CEEMDAN algorithm is used to assist the decomposition by adding adaptive white noise, decomposing the non-stationary signal into a series of intrinsic mode functions (IMFs). The decomposition process is as follows: ;in: : No. Each intrinsic mode function component represents signal fluctuations at different time scales from high frequency to low frequency. Residual components represent the extremely low frequency trend of the signal; The total number of IMFs obtained from the decomposition; finally, based on the frequency characteristics of each IMF component and the correlation analysis with environmental data, the decomposition results are reconstructed into two parts: environmental trend items. and high-frequency residuals ; Calculate the result of each decomposition Components and measured ambient temperature data Pearson correlation coefficient between : ;in, Describing covariance, The standard deviation is represented; then, a set of IMF indexes strongly correlated with environmental data is defined. The term "strong correlation" refers to a correlation coefficient whose absolute value is greater than a preset correlation threshold. ,Right now: Based on index set Construct environmental trend items : The remaining high-frequency, low-energy IMF components contain measurement noise and possible sudden damage signals, and are classified as high-frequency residual terms. : ; S3. Establish a gray prediction GM(1,1) model for metabolism to perform rolling predictions on environmental trend items, calculate the shaped residuals, and then analyze the separated environmental trend items. A gray prediction model for metabolism, GM(1,1), is established. The construction steps are as follows: Step 1: Sequence selection: Select the nearest sequence. A sequence of data points As a modeling sequence, where Corresponding time Environmental trend items ; Step 2: Accumulation Generation: Perform an accumulation generation (1-AGO) on the original sequence to obtain the sequence. ,in: ; Step 3: Establish the whitening differential equation: ;in: The development coefficient reflects the development trend of the sequence, i.e., the rate of growth or decline. Grey action quantity reflects the endogenous change relationship between data and the combined effect of external inputs; Step 4: Parameter estimation: Estimate parameters using the least squares method. : Among them, the data matrix and data vector They are respectively: ;matrix The nearest neighbor mean of the cumulative generated sequence is constructed to approximate the background value in the differential equation; Step 5: Solve the prediction model: Solve the differential equation to obtain the time response function. ; Step 6: Cumulative Subtraction Restoration: Obtain the predicted value of the original sequence through cumulative subtraction restoration. This refers to the predicted environmental impact value at the next moment. ; Step 7: Metabolic Renewal: When new measured data is obtained Then, it is added to the end of the sequence, while the oldest data is removed. Maintain sequence length as Repeat steps 2-6 above to update the model parameters in real time. and Finally, calculate the shaping residuals. The shaping residual is defined as the physical bias minus the predicted environmental trend term. By predicting and subtracting this trend, non-destructive drift caused by environmental and time-varying factors is eliminated; the resulting residuals... Physically, it represents the remaining energy that was not eliminated by normal operating conditions (step S1) and normal environmental trends (step S3). S4. Model the tail features of the shaped residual sequence based on the extreme value statistical distribution, calculate the dynamic hierarchical early warning threshold, and trigger the corresponding early warning response based on the comparison result between the real-time shaped residual and the dynamic hierarchical early warning threshold.

2. The method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure according to claim 1, characterized in that, The step S1 of constructing the stiffness topology state space includes: defining a discretized sequence of construction conditions, each condition corresponding to a specific stage in the construction process; using the element activation and kill technique in finite element analysis, activating or killing the corresponding element for each condition, and calculating the theoretical displacement field under that condition; and introducing a time correction coefficient that reflects the influence of concrete shrinkage and creep to correct the theoretical displacement field to generate the dynamic physical reference sequence.

3. The method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure according to claim 1, characterized in that, Step S2, which extracts the environmental trend term and the high-frequency residual term, includes: calculating the difference between the real-time acquired multi-source sensor data and the dynamic physical reference sequence as the physical deviation; decomposing the physical deviation using an adaptive noise complete set empirical mode decomposition algorithm to obtain intrinsic mode function components and residual components; and calculating the correlation between each intrinsic mode function component and the ambient temperature data, reconstructing the components with a correlation higher than a preset threshold as the environmental trend term, and reconstructing the remaining components as the high-frequency residual term.

4. The method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure according to claim 1, characterized in that, Step S3, establishing a grey prediction model for metabolism, includes: selecting the most recent data points as the modeling sequence, establishing a whitening differential equation describing the development trend of the sequence; estimating the model parameters using the least squares method, solving the time response function to obtain the predicted environmental impact value at the next moment; when new data is acquired, removing the oldest data and adding new data, updating the model parameters in real time; and calculating the difference between the physical deviation and the predicted environmental impact value as the shaped residual.

5. The method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure according to claim 1, characterized in that, Step S4, calculating the dynamic hierarchical early warning threshold, includes: selecting data exceeding the basic noise level in the shaped residuals as the over-threshold sequence; using the maximum likelihood estimation method to estimate the scale parameter and shape parameter of the generalized Pareto distribution in real time; and, based on the preset extremely high risk probability, combining the scale parameter and shape parameter, solving out the dynamic early warning threshold that changes dynamically over time.

6. The method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure according to claim 5, characterized in that, The step S4 triggers the corresponding early warning response as follows: when monitoring the mean change of the real-time shaping residual, when the mean deviates significantly, a first-level trend early warning is triggered; when monitoring a single real-time shaping residual value, when it exceeds the dynamic early warning threshold, a second-level extreme value alarm is triggered.

7. A multi-source deformation monitoring and early warning system for the construction stage of a super high-rise large cantilever concrete structure, characterized in that, A method for multi-source deformation monitoring and early warning during the construction stage of a super high-rise large cantilever concrete structure according to any one of claims 1-6, the system comprising: a physical benchmark construction module, a data acquisition and decoupling module, a residual shaping and trend elimination module, and a dynamic early warning decision module; The physical benchmark construction module is used to establish a refined finite element model, construct a stiffness topology state space based on the construction condition sequence, and generate a dynamic physical benchmark sequence. The data acquisition and decoupling module is used to acquire multi-source sensor data in real time, calculate physical deviations, and use the CEEMDAN algorithm to extract environmental trend terms and high-frequency residual terms. The residual shaping and trend elimination module is used to establish a metabolism grey prediction GM(1,1) model to perform rolling prediction of environmental trend items and calculate the shaped residuals. The dynamic early warning decision module is used to model the tail features of the shaped residual based on GPD, calculate the dynamic hierarchical early warning threshold, and trigger an early warning response based on the comparison results.

8. The multi-source deformation monitoring and early warning system for the construction stage of super high-rise large cantilever concrete structures according to claim 7, characterized in that, The data acquisition and decoupling module is connected to a multi-source sensor network, which includes: a global navigation satellite system receiver for acquiring three-dimensional displacement data of key nodes of the cantilever structure; a fiber optic grating sensor for acquiring strain data of key components; a temperature sensor and an anemometer for acquiring ambient temperature and wind speed data; the data acquisition and decoupling module is also used to perform time synchronization and resampling processing on all acquired data.

9. The multi-source deformation monitoring and early warning system for the construction stage of super high-rise large cantilever concrete structures according to claim 7, characterized in that, The physical benchmark construction module includes a pre-set concrete creep prediction model, which is used to calculate the time correction coefficient that changes over time in order to correct the dynamic physical benchmark sequence.

10. The multi-source deformation monitoring and early warning system for the construction stage of super high-rise large cantilever concrete structures according to claim 7, characterized in that, The dynamic early warning decision module is configured to: assess the risk of damage evolution based on the changing trend of the generalized Pareto distribution shape parameter estimated in real time, and dynamically adjust the dynamic graded early warning threshold according to the preset risk probability.

Images