Road foundation bearing characteristic analysis method and system

Abstract

The invention discloses a road foundation bearing characteristic analysis method and system, relates to the technical field of road foundation analysis, and realizes a fundamental breakthrough for the problems of high misjudgment rate, poor climate adaptability and the like caused by a traditional fixed threshold by constructing a dynamic self-adaptive risk judgment threshold mechanism. According to the mechanism, when a dynamic threshold value Tdy is calculated, spatial statistical characteristics of current foundation response strength are fully fused, and a comprehensive environment disturbance characteristic, namely a collapsible environment index, is combined, so that a dynamic judgment standard with environment sensitivity adjustment capability is formed. The system can automatically adjust the risk identification threshold value according to the actual rainwater infiltration intensity and the underground water dynamic jump condition, so that the system can actively reduce the risk tolerance boundary and identify and respond to an abnormal area in advance under the extreme working conditions of heavy rainfall, rapid underground water rising and the like, and moderately relaxes in a dry or stable period, so that the risk identification accuracy is improved. And the real-time performance and adaptability of risk identification are obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of highway foundation analysis, and specifically to a method and system for analyzing the bearing characteristics of highway foundations. Background Technique

[0002] In highway foundation engineering, especially in the directions of soft soil foundation treatment, collapsible loess foundation reinforcement, structural bearing characteristics evaluation and dynamic response control, etc., it has gradually become the focus of research and application in high-grade highways, mountain roads and newly built road networks under extreme climates. Specifically in the actual engineering situation, the problem of identifying the structural response and predicting the bearing capacity of potentially collapsible foundation structures during the rainy season has become one of the key bottlenecks restricting the safe and stable operation of the foundation.

[0003] In current engineering practices, the management of collapsible foundations mostly relies on static geological exploration reports or the setting of safety reserve coefficients formulated based on manual experience rules, and heavily depends on the theoretical estimates at the initial stage of construction. However, this approach is difficult to dynamically reflect the changing trends of the real-time impacts of factors such as "seasonal hydrological disturbances", "local heavy rainfall events", and "rapid rise of groundwater" on the foundation structure. In addition, most existing systems lack a dynamic monitoring and prediction mechanism that combines historical response behaviors with environmental conditions, and are unable to actively detect potential structural lag evolution processes, resulting in the inability to early warn of the collapsible behavior of the foundation under specific climate conditions.

[0004] Traditional foundation bearing assessment models ignore the "time lag" characteristics in the hydro-structure coupling relationship. There are obvious non-synchronous evolution paths during the processes of rainwater infiltration, seepage, groundwater level disturbance and structural relaxation, especially more significant in collapsible soil layers. However, due to the lack of a prediction mechanism and response model linked with time-series data, there often appears the phenomenon of "the ground remains stable, but in fact, the lower part has experienced structural loosening and sudden strength drop". Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method and system for analyzing the bearing characteristics of highway foundations, which solves the problems mentioned in the background technique.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A highway foundation bearing characteristic analysis system includes an environmental data collection and processing module, a collapsible lag factor calculation module, a lag response model construction module, a foundation state mapping and time-series analysis module, a dynamic risk identification and response decision-making module, and an evaluation output and reporting module; The environmental data collection and processing module collects the historical data of the highway foundation through sensors, fits it into the original data set W, and performs preprocessing to obtain the foundation data set DW; The collapsibility hysteresis factor calculation module extracts features from the foundation data set DW, including the permeability hysteresis coefficient Pr and the collapsibility response delay rate Tw; The hysteresis response model building module constructs a three-dimensional hysteresis response function S (t, x, y) according to the obtained seepage hysteresis coefficient Pr and collapsible response delay rate Tw; The foundation state mapping and time series analysis module performs time series derivative analysis on the acquired three-dimensional hysteresis response function S (t, x, y), obtains the sudden change speed of the foundation response, and generates a spatial mapping diagram ΔS (x, y); The dynamic risk identification and response decision module compares the acquired spatial mapping ΔS (x, y) with the dynamic threshold Tdy to generate a collapsible risk level map Λ (x, y); The assessment output and report module visualizes the obtained collapsible risk level map Λ(x, y), spatial mapping map ΔS(x, y) and three-dimensional hysteresis response function S(t,x,y).

[0007] Preferably, the environmental data acquisition and processing module includes a raw data acquisition unit and a data preprocessing unit; The original data acquisition unit collects historical data of the highway foundation through sensors, including daily rainfall R, groundwater level G and foundation settlement D, and fits them into the original data set W; Among them, the daily rainfall R is collected by the tipping bucket rain gauge in the elevated meteorological station around the roadbed; The groundwater level G is collected by a static pressure groundwater level meter buried at a depth of 10 to 20 m below the foundation; The foundation settlement D is obtained through ground displacement monitoring points arranged along the foundation cross section and laser leveling; The data preprocessing unit performs missing value processing, outlier cleaning and normalization processing on the acquired original data set W to obtain the ground-based data set DW; Missing value processing is done by filling the missing values of the original data set W using linear interpolation; Outlier cleaning is done by removing outliers from the original data set W using the triple standard deviation method; The normalization process includes normalizing the data in the original data set W using a normalization method to obtain a ground-based data set DW; The ground-based dataset DW is obtained by the following formula: ; Wherein, DWo represents the o-th data item in the ground-based dataset DW, Wo represents the o-th data item in the original dataset W, minWo represents the valley value of the o-th data item in the original dataset W, and maxWo represents the peak value of the o-th data item in the original dataset W.

[0008] Preferably, the collapsibility hysteresis factor calculation module includes a multi-scale hysteresis modeling unit and a structural relaxation response analysis unit; The multi-scale hysteresis modeling unit extracts features from the foundation dataset DW, extracts the delay time variation rate during the infiltration process, and constitutes the infiltration hysteresis coefficient Pr; The infiltration hysteresis coefficient Pr is obtained through the following formula: ; In the formula, Pr(t) represents the infiltration hysteresis coefficient at time t, G(t) represents the groundwater level height at time t, R(t) represents the daily rainfall at time t, α represents the adjustment parameter for controlling the rainfall response influence degree, d represents the derivative symbol, dG(t) / dt represents the groundwater level change rate, d 2 R(t) / dt 2 represents the rainfall acceleration; By using exponential moving average, noise removal is performed on the obtained infiltration hysteresis coefficient Pr; The noise removal formula is: ; In the formula, λ represents the smoothing factor, and λ ∈ (0,1), Pr(t - 1) represents the infiltration hysteresis coefficient at time t - 1.

[0009] Preferably, the structural relaxation response analysis unit identifies the time delay path of the structural response according to the obtained infiltration hysteresis coefficient Pr and combines it with the groundwater level height G, and extracts the collapsibility response delay rate Tw; The collapsibility response delay rate Tw is obtained through the following formula: ; In the formula, represents the infiltration response rate adjustment coefficient, represents the groundwater level influence coefficient, represents the foundation structural relaxation potential energy coefficient, Ψ represents the dynamic foundation relaxation potential energy function, represents the structural deformation trend, Hre represents the reference foundation depth, and ln represents the logarithmic function; The dynamic foundation relaxation potential energy function Ψ is obtained through the following formula: ; In the formula, A represents the spatial region integration area, D(t, x, y) represents the settlement amount at the spatial position (x, y) at time t, ∂ represents the partial derivative, and dxdy represents the two-dimensional integral operation.

[0010] Preferably, the hysteresis response model construction module outputs the obtained infiltration hysteresis coefficient Pr and collapsibility response delay rate Tw as dynamic factors, and constructs a three-dimensional hysteresis response function S(t,x,y) in the form of time-weighted integration; The three-dimensional lag response function S(t, x, y) is obtained by the following formula: ; In the formula, tx represents the integration time variable, and tx ∈ [0, t], e represents a constant, represents the memory decay factor, represents the spatial diffusion weight function; Spatial diffusion weight function is obtained by the following formula: ; In the formula, π represents the pi, with a value of 3.14, σ(tx) represents the perturbation diffusion standard deviation, exp represents the exponential function, (xo, yo) represents the perturbation source, such as the position coordinates of the heavy rainfall center and the infiltration concentration point; Perform multi-scale reconstruction analysis on the obtained three-dimensional lag response function S(t, x, y), extract the response trends from different time windows, including the average response value of the time window and the response fluctuation intensity coefficient, and use them to distinguish the response mutation growth region, the response long-term slow change region, and the abnormal points of the repeated perturbation response frequency; The average response value of the time window is obtained by integrating the three-dimensional lag response function S(t, x, y) at the spatial position (x, y) within T time units and then dividing by the time interval length T; The response fluctuation intensity coefficient is obtained by calculating the absolute value average of the time change rate of the three-dimensional lag response function S(t, x, y) within T time units.

[0011] Preferably, the foundation state mapping and time series analysis module includes a time series change rate calculation unit and a response spatial map generation unit; The time series change rate calculation unit calculates the response change rate ∂S(t, x, y) / ∂t of the foundation response per unit time by performing derivative analysis on the three-dimensional lag response function S(t, x, y) at each fixed spatial position (x, y); The response change rate ∂S(t, x, y) / ∂t is obtained by the following formula: ; In the formula, S(t - Δt, x, y) represents the three-dimensional lag response function at time t - Δt, and Δt represents the time interval.

[0012] Preferably, the response spatial map generation unit performs spatial integration processing on the obtained response change rate ∂S(t, x, y) / ∂t, generates a spatial map ΔS(x, y) of the foundation response rate, and obtains the risk response gradient index ∇S through the spatial map ΔS(x, y); The spatial mapping diagram ΔS(x, y) is obtained through the following formula: ; The risk response gradient index ∇S is obtained through the following formula: ; Compare the obtained risk response gradient index ∇S with the preset risk threshold Ts to judge the change rate difference state of the boundary area; The change rate difference state of the boundary area is obtained by matching in the following way: When the risk response gradient index ∇S ≤ the risk threshold Ts, it means that the boundary area is in a normal state; When the risk response gradient index ∇S > the risk threshold Ts, it means that the boundary area is in an abnormal state, indicating that the response change near this area is steep, belonging to the "response jump area" or "structural transition zone".

[0013] Preferably, the dynamic risk identification and response decision-making module compares the obtained spatial mapping diagram ΔS(x, y) with the dynamic threshold Tdy to generate the collapsibility risk level map Λ(x, y); The dynamic threshold Tdy is not a fixed constant, but is adaptively adjusted according to the current spatial mapping diagram ΔS(x, y), the daily rainfall R, and the groundwater level height G; The dynamic threshold Tdy is obtained through the following formula; ; In the formula, μΔ represents the spatial average value of the spatial mapping diagram ΔS(x, y), σΔ represents the spatial standard deviation of the spatial mapping diagram ΔS(x, y), p1 represents the risk enhancement adjustment coefficient, p2 represents the environmental disturbance suppression adjustment coefficient, represents the collapsibility environment index; The collapsibility environment index is obtained through the following formula: ; In the formula, ηR represents the rainfall disturbance intensity factor, ηG represents the abnormal jump factor of the groundwater level; The rainfall disturbance intensity factor ηR is obtained by integrating the rainfall change speed from time t - Tr to time t; The abnormal jump factor of the groundwater level ηG is obtained by the ratio of the difference between the groundwater level height at time t and the groundwater level height at time t - Δt to Δt; The collapsibility risk level map Λ(x, y) is obtained through the following formula: .

[0014] Preferably, the evaluation output and reporting module generates a two-dimensional color heat map according to the collapsibility risk level map Λ(x, y) to display different risk level partitions; Common color scheme: green indicates stable, yellow indicates medium risk, and red indicates high risk; The risk distribution and high-risk areas are displayed through the spatial mapping diagram ΔS(x, y); Through the three-dimensional lag response function S(t, x, y), at a fixed spatial position (x, y), a response curve varying with time is established; at time t0, a ground surface response map is established; The obtained two-dimensional color heat map, the display of risk distribution, high-risk areas, and the ground surface response map are summarized into a structured document, including PDF format and HTML format, and output.

[0015] A method for analyzing the bearing characteristics of a highway foundation includes the following steps: Step 1: The environmental data collection and processing module collects the historical data of the highway foundation through sensors, fits it into the original data set W, and performs preprocessing to obtain the foundation data set DW; Step 2: The collapsibility lag factor calculation module extracts the characteristics of the foundation data set DW, including the infiltration lag coefficient Pr and the collapsibility response delay rate Tw; Step 3: The lag response model construction module constructs a three-dimensional lag response function S(t, x, y) according to the obtained infiltration lag coefficient Pr and collapsibility response delay rate Tw; Step 4: The foundation state mapping and time series analysis module performs time series derivative analysis on the obtained three-dimensional lag response function S(t, x, y), obtains the sharp change speed of the foundation response, and generates a spatial mapping diagram ΔS(x, y); Step 5: The dynamic risk identification and response decision-making module compares the obtained spatial mapping diagram ΔS(x, y) with the dynamic threshold Tdy to generate a collapsibility risk level map Λ(x, y); Step 6: The evaluation output and reporting module visually outputs the obtained collapsibility risk level map Λ(x, y), spatial mapping diagram ΔS(x, y), and three-dimensional lag response function S(t, x, y).

[0016] The present invention provides a method and system for analyzing the bearing characteristics of a highway foundation, having the following beneficial effects: (1)During system operation, through the systematic design of the original data acquisition method, the acquisition paths of three types of key environmental and structural data, namely daily rainfall R, groundwater level height G, and foundation settlement D, were clarified. The data was collected based on existing sensing devices such as tipping bucket rain gauges, static groundwater level gauges, and laser levels, ensuring the real-time, representative, and engineering applicability of the data sources. Different from the traditional method of using single survey data or static charts as criteria, this system established a high-density original dataset suitable for "seasonal response trend identification" for the first time by continuously collecting long-term historical environmental data of the foundation, providing a solid data foundation for dynamic modeling and prediction of collapsibility evolution.

[0017] (2)By introducing a structural response delay modeling mechanism, the dynamic parameter of collapsibility response delay rate was systematically defined. Different from the previous simple threshold judgment that only relied on groundwater level or settlement rate, this system incorporated the "time delay path" of structural response into the analysis scope for the first time, and constructed a dynamic foundation relaxation potential energy function in combination with the foundation settlement gradient, enabling the accurate identification of the hysteretic deformation trend of the foundation structure after hydrological excitation. This design effectively filled the gap that traditional bearing capacity evaluation methods were difficult to identify the "non-instantaneous response process", and significantly improved the identification depth of collapsibility precursors.

[0018] (3)By constructing a dynamic adaptive risk discrimination threshold mechanism, a fundamental breakthrough was achieved in the problems of high misjudgment rate and poor climate adaptability brought by traditional fixed thresholds. When calculating the dynamic threshold Tdy, this mechanism fully integrated the spatial statistical characteristics of the current foundation response intensity and combined with the comprehensive environmental disturbance characteristic of the collapsibility environment index to form a dynamic discrimination standard with environmental sensitive adjustment ability. The system can automatically adjust the risk identification threshold according to the actual rainfall infiltration intensity and dynamic jump of groundwater level, so that in extreme working conditions such as heavy rainfall and rapid rise of groundwater, the system can actively reduce the risk tolerance boundary and identify abnormal response areas in advance, while moderately relaxing during dry or stable periods, significantly improving the real-time and adaptability of risk identification.

[0019] (4) By means of a time-varying integral and a spatial diffusion weight mechanism, the evolution trend of the perturbation propagating from the source point to different regions of the foundation is simulated. This model can comprehensively characterize the response accumulation and spatial evolution process of the structure under long-term loads and complex environmental actions, greatly improving the accuracy and predictability of the bearing trend identification. By analyzing the derivative of the three-dimensional response function to obtain the spatial mapping diagram ΔS(x, y), the system can accurately identify the regions with sudden increases in response and the spatially non-uniform deformation zones, which is more time-effective compared to traditional static judgment methods and has the ability to identify boundary anomalies, crack development, and load concentration areas. Through the introduction in the fifth step, the method no longer uses a single static judgment benchmark, but combines the spatial mapping diagram ΔS(x, y), rainfall intensity, and groundwater fluctuations to generate a dynamic threshold Tdy, thus realizing an intelligent control mechanism for self-adaptive adjustment of the risk identification sensitivity with environmental perturbations, effectively improving the adaptability and accuracy of risk discrimination. Brief Description of the Drawings

[0020] Figure 1 It is a schematic flowchart of the block diagram of a highway foundation bearing characteristic analysis system according to the present invention; Figure 2 It is a schematic diagram of the steps of a highway foundation bearing characteristic analysis method according to the present invention; Figure 3 It is a schematic flowchart of the collapsibility risk level map according to the present invention; Figure 4 It is a bar chart of the collapsibility response delay rate according to the present invention. Detailed Description of the Embodiments

[0021] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0022] Embodiment 1 The present invention provides a highway foundation bearing characteristic analysis system. Please refer to Figures 1 to 4 , which includes an environmental data acquisition and processing module, a collapsibility lag factor calculation module, a lag response model construction module, a foundation state mapping and time series analysis module, a dynamic risk identification and response decision module, and an evaluation output and reporting module; The environmental data acquisition and processing module collects the historical data of the highway foundation through sensors, fits it into the original data set W, and performs preprocessing to obtain the foundation data set DW; The collapsibility lag factor calculation module extracts features from the foundation data set DW, including the infiltration lag coefficient Pr and the collapsibility response delay rate Tw; The hysteresis response model building module constructs a three-dimensional hysteresis response function S (t, x, y) according to the obtained seepage hysteresis coefficient Pr and collapsible response delay rate Tw; The foundation state mapping and time series analysis module performs time series derivative analysis on the acquired three-dimensional hysteresis response function S (t, x, y), obtains the sudden change speed of the foundation response, and generates a spatial mapping diagram ΔS (x, y); The dynamic risk identification and response decision module compares the acquired spatial mapping ΔS (x, y) with the dynamic threshold Tdy to generate a collapsible risk level map Λ (x, y); The assessment output and report module visualizes the obtained collapsible risk level map Λ(x, y), spatial mapping map ΔS(x, y) and three-dimensional hysteresis response function S(t,x,y).

[0023] In this embodiment, a complete analysis chain from environmental data collection, factor identification, response modeling to risk judgment and visual output is established through a modular six-level structure, ensuring that information flow is smooth, timing logic is rigorous, and response efficiency is higher. All functions of this system are processed and modeled based on data collected by existing sensors, without changing the on-site hardware structure, reflecting the technical advantages of low cost, convenient deployment, and easy integration and upgrading, and are suitable for the intelligent transformation of newly built roads or existing roads.

[0024] The system innovatively proposes the permeability hysteresis coefficient Pr and the collapsible response delay rate Tw as key characteristic quantities. From the perspective of the asynchronous response to hydrological disturbances, it can capture potential response delay behaviors that traditional systems cannot identify and effectively identify the precursors of collapsible water. By modeling the dynamic evolution of foundation response in time and space, the system can perceive the foundation stress relaxation trend or settlement trend in advance during high-risk periods such as the rainy season, significantly enhancing the system's seasonal adaptability and risk prevention capabilities.

[0025] The foundation state mapping and time series analysis module can perform derivative analysis on the local area response rate and display it graphically, so as to achieve real-time locking of the foundation structure mutation area and the risk active area, and provide a quantitative basis for precise governance. By introducing an adaptive dynamic threshold mechanism, the system can automatically adjust the recognition standard, so that the early warning system has the ability to adapt to different weather, geological and load conditions, and realize the discrimination strategy of "tolerance in stable period and sensitivity in abnormal period".

[0026] Example 2 This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the environmental data acquisition and processing module includes a raw data acquisition unit and a data pre-processing unit; The original data acquisition unit collects historical data of the highway foundation through sensors, including the daily rainfall R, the groundwater level height G, and the foundation settlement D, and fits them into the original data set W; Among them, the daily rainfall R is collected by a tipping bucket rain gauge in the elevated weather station around the roadbed; The groundwater level height G is collected by a static piezometer buried at a depth of 10 - 20m under the foundation; The foundation settlement D is collected by ground displacement monitoring points and a laser level instrument arranged along the cross-section of the foundation; The data preprocessing unit performs missing value processing, outlier cleaning, and normalization processing on the obtained original data set W to obtain the foundation data set DW; The missing value processing fills the missing values in the original data set W by using the linear interpolation method; The outlier cleaning removes the outliers in the original data set W by the three - standard - deviation method; The normalization processing includes normalizing the data in the original data set W by using the normalization method to obtain the foundation data set DW; The foundation data set DW is obtained through the following formula: ; In the formula, DWo represents the o - th data in the foundation data set DW, Wo represents the o - th data in the original data set W, minWo represents the minimum value of the o - th data in the original data set W, and maxWo represents the maximum value of the o - th data in the original data set W.

[0027] The collapsible hysteresis factor calculation module includes a multi - scale hysteresis modeling unit and a structural relaxation response analysis unit; The multi - scale hysteresis modeling unit extracts features from the foundation data set DW, extracts the delay time variation rate during the infiltration process, and forms the infiltration hysteresis coefficient Pr; The infiltration hysteresis coefficient Pr is obtained through the following formula: ; In the formula, Pr(t) represents the infiltration hysteresis coefficient at time t, G(t) represents the groundwater level height at time t, R(t) represents the daily rainfall at time t, α represents the adjustment parameter controlling the rainfall response influence degree, d represents the derivative symbol, dG(t) / dt represents the groundwater level change rate, d 2 R(t) / dt 2 represents the rainfall acceleration; By using the exponential moving average, noise removal is performed on the obtained infiltration hysteresis coefficient Pr; The noise removal formula is: ; Where λ represents the smoothing factor, and λ∈(0,1), and Pr(t-1) represents the permeability hysteresis coefficient at time t-1.

[0028] This embodiment systematically designs the original data collection method, clarifies the acquisition path of three types of key environmental and structural data, namely daily rainfall R, groundwater level G and foundation settlement D, and deploys and collects data based on existing sensor equipment such as tipping bucket rain gauges, static pressure groundwater level gauges and laser levels, ensuring the real-time, representative and engineering applicability of data sources. Different from the traditional method of using single survey data or static charts as criteria, this system continuously collects historical foundation environmental data for a long time, and for the first time establishes a high-density original data set suitable for "seasonal response trend identification", providing a solid data foundation for dynamic modeling and prediction of subsidence evolution.

[0029] This example introduces a combination strategy of linear interpolation and triple standard deviation method to repair and remove missing values and outliers, respectively, significantly improving the integrity and anti-interference ability of the ground-based data set. In addition, through normalization, the data dimensions of different physical quantities are unified, making various environmental factors comparable in the same evaluation system, thereby constructing a high-quality ground-based data set that can directly participate in modeling and analysis, effectively solving the problems of traditional data clutter, single processing methods, and inability to support dynamic modeling.

[0030] The calculation method of the seepage hysteresis coefficient introduced in this embodiment breaks through the previous single factor analysis mode based only on water level or rainfall intensity, and establishes a dynamic factor that can simultaneously characterize the impact of groundwater response speed to rainfall and rainfall rate change trend, and further removes high-frequency fluctuations through exponential sliding average technology to form a response characteristic sequence with time series stability. This design not only realizes the accurate identification of the hydrological-structural asynchronous relationship, but also can quantitatively characterize the potential delayed response of foundation structural factors during heavy rainfall periods, thereby having higher dynamic adaptability and the ability to capture the precursors of collapse.

[0031] Example 3 This embodiment is explained in Example 2. Please refer to Figure 1 and Figure 4 ,Specifically: the structural relaxation response analysis unit identifies the time delay path of the structural response and extracts the wetting response delay rate Tw based on the obtained seepage hysteresis coefficient Pr and the groundwater level height G; The collapsibility response delay rate Tw is obtained by the following formula: ; In the formula, represents the permeation response rate adjustment coefficient, represents the groundwater level influence coefficient, where ξ represents the coefficient of the relaxation potential energy of the foundation structure, Ψ represents the dynamic foundation relaxation potential energy function, Hre represents the reference foundation depth, and ln represents the logarithmic function; Specific example: Table 1: Example table for calculating the delay rate of collapsibility response;

[0032] The dynamic foundation relaxation potential energy function Ψ is obtained through the following formula: ; In the formula, A represents the integral area of the spatial region, D(t, x, y) represents the settlement amount at the spatial position (x, y) at time t, ∂ represents the partial derivative, and dxdy represents the two-dimensional integral operation.

[0033] The hysteretic response model construction module outputs the obtained seepage hysteretic coefficient Pr and the delay rate of collapsibility response Tw as dynamic factors, and constructs a three-dimensional hysteretic response function S(t, x, y) in the form of time-weighted integration; The three-dimensional hysteretic response function S(t, x, y) is obtained through the following formula: ; In the formula, tx represents the integral time variable, and tx ∈ [0, t], e represents a constant, represents the memory decay factor, represents the spatial diffusion weight function; The spatial diffusion weight function is obtained through the following formula: ; In the formula, π represents the pi, σ(tx) represents the standard deviation of perturbation diffusion, exp represents the exponential function, and (xo, yo) represents the position coordinates of the perturbation source; Perform multi-scale reconstruction analysis on the obtained three-dimensional hysteretic response function S(t, x, y), extract the response trends from different time windows, including the average response value of the time window and the response fluctuation intensity coefficient, and use them to distinguish the response mutation growth region, the response long-term slow change region, and the abnormal points of the repeated perturbation response frequency; The average response value of the time window is obtained by integrating the three-dimensional hysteretic response function S(t, x, y) at the spatial position (x, y) within T time units and then dividing by the length T of the time interval; The response fluctuation intensity coefficient is obtained by calculating the average value of the absolute value of the time change rate of the three-dimensional hysteretic response function S(t, x, y) within T time units.

[0034] Based on the calculation of the previous seepage hysteresis coefficient Pr, this embodiment systematically defines the dynamic parameter of the collapsible response delay rate by introducing a structural response delay modeling mechanism. Different from the previous simple threshold judgment that only relied on the groundwater level or settlement rate, this system for the first time included the "time delay path" of the structural response into the analysis scope, and constructed a dynamic foundation relaxation potential energy function in combination with the foundation settlement gradient, so as to accurately identify the hysteresis deformation trend of the foundation structure after hydrological excitation. This design effectively fills the gap that traditional bearing capacity evaluation methods are difficult to identify the "non-instant response process", and effectively improves the identification depth of collapsible precursors.

[0035] This embodiment constructs a three-dimensional hysteresis response function model with the infiltration hysteresis coefficient Pr and the collapsible response delay rate Tw as dual core driving factors, integrates the historical disturbance effect through time-weighted integration, and introduces the spatial diffusion weight function to achieve the real propagation simulation of the disturbance source in time and space. In particular, the introduction of the disturbance memory attenuation mechanism and the Gaussian spatial diffusion kernel function can simulate the non-uniformity and path attenuation effect of rainwater diffusion in the foundation, greatly improving the response prediction accuracy of the system under complex climatic conditions, and making up for the key shortcomings of the traditional model, such as insufficient spatiotemporal coupling and the lack of disturbance diffusion mechanism.

[0036] The system also conducts in-depth mining of the three-dimensional hysteresis response function S (t, x, y) through multi-scale response reconstruction analysis, and constructs two types of advanced features: the average response value of the time window and the response fluctuation intensity coefficient, which correspond to the identification capabilities of long-term slow-changing settlement trends and sudden response jump behaviors. This analysis method not only realizes the partition identification of "chronic structural deformation", "sudden response growth" and "abnormal points of repeated disturbance frequency", but also gives the system the ability to identify the continuous state from stable → fluctuation → risk → critical, providing data support for subsequent risk level assessment and active intervention strategies.

[0037] This embodiment comprehensively enhances the adaptive recognition and evolutionary perception capabilities of foundation structures to seasonal and sudden subsidence disturbances by constructing a spatiotemporal fusion response prediction model and a multi-scale behavior trend assessment mechanism without relying on additional sensing hardware. It effectively solves the pain points of traditional methods such as "insufficient recognition dimensions, unquantifiable response delays, and lack of data support for trend judgment", and provides an intelligent solution for early warning and zoning management of subsidence foundations during the rainy season.

[0038] Example 4 This embodiment is explained in Example 3, please refer to Figure 3 ,Specifically: the foundation state mapping and time series analysis module includes a ,time series change rate calculation unit and a response space ,map generation unit; The time series change rate calculation unit calculates the response change rate ∂S(t, x, y) / ∂t of the ground response change per unit time by performing derivative analysis on the three-dimensional lag response function S(t, x, y) at each fixed spatial position (x, y). The response change rate ∂S(t, x, y) / ∂t is obtained through the following formula: ; In the formula, S(t - Δt, x, y) represents the three-dimensional lag response function at time t - Δt, and Δt represents the time interval.

[0039] The response space map generation unit performs spatial integration processing on the obtained response change rate ∂S(t, x, y) / ∂t to generate a spatial map ΔS(x, y) of the ground response rate, and obtains the risk response gradient index ∇S through the spatial map ΔS(x, y). The spatial map ΔS(x, y) is obtained through the following formula: ; The risk response gradient index ∇S is obtained through the following formula: ; Compare the obtained risk response gradient index ∇S with the preset risk threshold Ts to judge the change rate difference state of the boundary region. The change rate difference state of the boundary region is obtained by matching in the following way: When the risk response gradient index ∇S ≤ the risk threshold Ts, it indicates that the boundary region is in a normal state; When the risk response gradient index ∇S > the risk threshold Ts, it indicates that the boundary region is in an abnormal state.

[0040] Based on the existing three-dimensional lag response function S(t, x, y) modeling, this embodiment deeply explores the spatial mutation characteristics in the response evolution process, and for the first time introduces the derivative behavior of "response change rate" into the ground lag response analysis system, establishing a time series change rate calculation unit, which can perform derivative estimation in the time direction on the three-dimensional lag response function S(t, x, y) at any fixed spatial position (x, y), so as to obtain the speed of the ground structure response change per unit time.

[0041] Through the response space map generation unit, the response rate values of each point are integrated into a full-region distribution map spatial map ΔS(x, y), realizing the high-resolution spatial visualization of the ground response rate, and effectively depicting the continuity and discreteness of the response distribution. This embodiment introduces the risk response gradient index ∇S to judge the gradient jump of the response rate in adjacent regions, and quantitatively describes the mutation intensity of the ground structure boundary region from a mathematical perspective.

[0042] By comparing the risk response gradient index ∇S with a preset risk threshold Ts, the system can not only determine whether a certain position reaches the standard line of response intensity, but more importantly, it can identify whether the mutation trend of the boundary response in the region is abnormal, thereby distinguishing the spatial difference transition region from the structure imbalance active region, and realizing the automatic identification and status classification of the abnormal behavior of the structure transition zone.

[0043] In this embodiment, by establishing a linkage analysis system in two dimensions of time derivative and spatial gradient, precise capture of the ground response mutation points, boundary jump zones, and dynamic discontinuity regions is achieved, effectively making up for the deficiencies of the prior art in aspects such as insufficient accuracy in boundary change identification and difficulty in continuously tracking dynamic responses, and further improving the intelligent level and engineering adaptability of the system in ground structure continuity identification, sensitive area positioning, and structural safety zoning.

[0044] Embodiment 5 This embodiment is an explanatory description based on Embodiment 4. Please refer to Figure 1 and Figure 3 , specifically: The dynamic risk identification and response decision module compares the obtained spatial mapping map ΔS(x, y) with the dynamic threshold Tdy to generate the collapsible risk level map Λ(x, y); The dynamic threshold Tdy is not a fixed constant, but is adaptively adjusted according to the current spatial mapping map ΔS(x, y), the daily rainfall R, and the groundwater level height G; The dynamic threshold Tdy is obtained through the following formula; ; In the formula, μΔ represents the spatial average value of the spatial mapping map ΔS(x, y), σΔ represents the spatial standard deviation of the spatial mapping map ΔS(x, y), p1 represents the risk enhancement adjustment coefficient, p2 represents the environmental disturbance suppression adjustment coefficient, represents the collapsible environment index; Collapsible environment index is obtained through the following formula: ; In the formula, ηR represents the rainfall disturbance intensity factor, ηG represents the abnormal jump factor of the groundwater level; The rainfall disturbance intensity factor ηR is obtained by integrating the rainfall change rate from time t - Tr to time t; The abnormal jump factor ηG of the groundwater level is obtained by the ratio of the difference between the groundwater level height at time t and the groundwater level height at time t - Δt to Δt; The collapsible risk level map Λ(x, y) is obtained through the following formula: .

[0045] The evaluation output and reporting module generates a two-dimensional color heat map based on the collapsibility risk level map Λ(x, y) to display different risk level partitions; The risk distribution and high-risk areas are shown through the spatial mapping diagram ΔS(x, y); Through the three-dimensional lag response function S(t, x, y), at a fixed spatial position (x, y), a curve of the response varying with time is established; at time t0, a ground surface response map is established; The obtained two-dimensional color heat map, the display of risk distribution, high-risk areas, and the ground surface response map are summarized into a structured document, including PDF format and HTML format, and then output.

[0046] In this embodiment, by constructing a dynamic adaptive risk discrimination threshold mechanism, fundamental breakthroughs have been achieved in problems such as high misjudgment rate and poor climate adaptability brought by traditional fixed thresholds. When calculating the dynamic threshold Tdy, this mechanism fully integrates the spatial statistical characteristics of the current ground response intensity and combines the comprehensive environmental disturbance characteristic of the collapsibility environment index to form a dynamic discrimination criterion with environmental sensitivity adjustment ability. The system can automatically adjust the risk identification threshold according to the actual rainwater infiltration intensity and the dynamic jump of groundwater level, so that in extreme working conditions such as heavy rainfall and rapid rise of groundwater, the system can actively lower the risk tolerance boundary and identify abnormal response areas in advance, while moderately relaxing during dry or stable periods, significantly improving the real-time and adaptability of risk identification.

[0047] This embodiment proposes to jointly model the rainfall disturbance intensity factor and the abnormal jump factor of the groundwater level to comprehensively quantify the mutation trend of the external environment of the structure. By performing time-window integration on the change rate of rainfall and calculating the change ratio of the groundwater fluctuation gradient, the system can real-time identify potential dangerous conditions where the structure remains unchanged and the environment changes first, providing accurate pre-information for the ground structure response modeling. This mechanism can effectively identify the latent collapsibility triggering mechanism that is easily overlooked in traditional methods, further expanding the time window of risk identification.

[0048] In this embodiment, by constructing the collapsibility risk level map Λ(x, y), spatial hierarchical management of different risk areas is realized, and the distribution of high-risk areas is intuitively displayed through a two-dimensional color heat map. At the same time, the system combines the three-dimensional lag response function S(t, x, y) to construct a response time curve of a fixed spatial point and a spatial response profile diagram of a fixed time point respectively, thus transforming the evolution process of the ground state into a visual trajectory. Finally, by summarizing all analysis results into a structured report, a full-process closed-loop from data analysis to chart output and then to document archiving is realized, greatly improving the interpretability, reviewability of the results and the practical value of engineering scheduling.

[0049] Embodiment 6 A method for analyzing the bearing characteristics of highway subgrades, please refer to Figure 2 , specifically: including the following steps: Step 1: The environmental data acquisition and processing module collects the historical data of the highway subgrade through sensors, fits it into the original data set W, and performs preprocessing to obtain the subgrade data set DW; Step 2: The collapsibility lag factor calculation module extracts the characteristics of the subgrade data set DW, including the infiltration lag coefficient Pr and the collapsibility response delay rate Tw; Step 3: The lag response model construction module constructs a three-dimensional lag response function S(t, x, y) based on the obtained infiltration lag coefficient Pr and collapsibility response delay rate Tw; Step 4: The subgrade state mapping and time series analysis module performs time series derivative analysis on the obtained three-dimensional lag response function S(t, x, y), obtains the sharp change speed of the subgrade response, and generates a spatial mapping diagram ΔS(x, y); Step 5: The dynamic risk identification and response decision-making module compares the obtained spatial mapping diagram ΔS(x, y) with the dynamic threshold Tdy to generate a collapsibility risk level map Λ(x, y); Step 6: The evaluation output and reporting module visually outputs the obtained collapsibility risk level map Λ(x, y), spatial mapping diagram ΔS(x, y), and three-dimensional lag response function S(t, x, y).

[0050] In this embodiment, through the synchronous acquisition and preprocessing of multi-source data such as rainfall, groundwater, and settlement of the highway subgrade in the first step, a systematic historical data set DW of subgrade operation is established, effectively overcoming the problem that traditional methods only rely on static geological exploration data or artificial experience parameters, and enabling the analysis of subgrade bearing characteristics to have high time resolution and full-process perception ability.

[0051] Through the extraction of the collapsibility lag factor in the second step, the infiltration lag coefficient Pr and the collapsibility response delay rate Tw are respectively constructed, revealing the asynchronous and lagging dynamic evolution path of the subgrade structure in the face of hydrological disturbances. These two parameters not only fill the gap that traditional static indicators are difficult to characterize the dynamic behavior of the structure, but also provide key driving factors for subsequent predictive modeling.

[0052] Through a time-varying integration and spatial diffusion weight mechanism, the evolution trend of the perturbation propagating from the source point to different regions of the foundation is simulated. This model can comprehensively characterize the response accumulation and spatial evolution process of the structure under long-term loads and complex environmental effects, greatly improving the accuracy and predictability of the bearing trend identification. By analyzing the derivative of the three-dimensional response function to obtain the spatial mapping ΔS(x, y), the system can accurately identify the regions of sudden response increase and spatial non-uniform deformation bands, which is more time-effective compared to traditional static judgment methods and has the ability to identify boundary anomalies, crack development, and load concentration areas. Through the introduction in the fifth step, the method no longer uses a single static judgment benchmark, but combines the spatial mapping ΔS(x, y), rainfall intensity, and groundwater fluctuations to generate a dynamic threshold Tdy, thus realizing an intelligent control mechanism for self-adaptive adjustment of the risk identification sensitivity with environmental perturbations, effectively improving the adaptability and accuracy of risk discrimination.

[0053] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A highway foundation bearing characteristic analysis system, characterized in that: It includes environmental data acquisition and processing module, collapsible hysteresis factor calculation module, hysteresis response model construction module, foundation state mapping and time series analysis module, dynamic risk identification and response decision module and assessment output and report module; The environmental data collection and processing module collects historical data of the highway foundation through sensors, fits it into the original data set W, and performs preprocessing to obtain the foundation data set DW; The collapsibility hysteresis factor calculation module extracts features from the foundation data set DW, including the permeability hysteresis coefficient Pr and the collapsibility response delay rate Tw; The hysteresis response model building module constructs a three-dimensional hysteresis response function S (t, x, y) according to the obtained seepage hysteresis coefficient Pr and collapsible response delay rate Tw; The foundation state mapping and time series analysis module performs time series derivative analysis on the acquired three-dimensional hysteresis response function S (t, x, y), obtains the sudden change speed of the foundation response, and generates a spatial mapping diagram ΔS (x, y); The dynamic risk identification and response decision module compares the acquired spatial mapping ΔS (x, y) with the dynamic threshold Tdy to generate a collapsible risk level map Λ (x, y); The assessment output and report module visualizes the obtained collapsible risk level map Λ(x, y), spatial mapping map ΔS(x, y) and three-dimensional hysteresis response function S(t,x,y).

2. The highway foundation bearing characteristic analysis system according to claim 1, characterized in that: The environmental data acquisition and processing module includes a raw data acquisition unit and a data preprocessing unit; The original data acquisition unit collects historical data of the highway foundation through sensors, including daily rainfall R, groundwater level G and foundation settlement D, and fits them into the original data set W; Among them, the daily rainfall R is collected by the tipping bucket rain gauge in the elevated meteorological station around the roadbed; The groundwater level G is collected by a static pressure groundwater level meter buried at a depth of 10 to 20 m below the foundation; The foundation settlement D is obtained through ground displacement monitoring points arranged along the foundation cross section and laser leveling; The data preprocessing unit performs missing value processing, outlier cleaning and normalization processing on the acquired original data set W to obtain the ground-based data set DW; Missing value processing is done by filling the missing values of the original data set W using linear interpolation; Outlier cleaning is done by removing outliers from the original data set W using the triple standard deviation method; The normalization process includes normalizing the data in the original data set W using a normalization method to obtain a ground-based data set DW; The ground-based dataset DW is obtained by the following formula: ； Wherein, DWo represents the o-th data item in the ground-based dataset DW, Wo represents the o-th data item in the original dataset W, minWo represents the valley value of the o-th data item in the original dataset W, and maxWo represents the peak value of the o-th data item in the original dataset W.

3. The analysis system for the bearing characteristics of a highway foundation according to claim 2, wherein: The collapsible hysteresis factor calculation module includes a multi-scale hysteresis modeling unit and a structural relaxation response analysis unit; The multi-scale hysteresis modeling unit performs feature extraction on the foundation dataset DW, extracts the delay time variation rate during the infiltration process, and constructs the infiltration hysteresis coefficient Pr; The permeability hysteresis coefficient Pr is obtained by the following formula: ； Where Pr(t) represents the osmotic lag coefficient at time t, G(t) represents the groundwater level at time t, R(t) represents the daily rainfall at time t, α represents the adjustment parameter for controlling the influence degree of rainfall response, d represents the derivative symbol, dG(t) / dt represents the groundwater level change rate, d 2 R(t) / dt 2 represents the rainfall acceleration; The obtained penetration lag coefficient Pr is denoised by using the exponential moving average. The denoising formula is: ； In the formula, λ represents the smoothing factor, and λ ∈ (0, 1), and Pr(t - 1) represents the penetration lag coefficient at time t - 1.

4. The highway foundation bearing characteristic analysis system according to claim 3, wherein: The structural relaxation response analysis unit identifies the time delay path of the structural response and extracts the collapsibility response delay rate Tw according to the obtained penetration lag coefficient Pr and in combination with the groundwater level height G. The collapsibility response delay rate Tw is obtained through the following formula: ； In the formula, represents the osmotic response rate adjustment coefficient, represents the groundwater level influence coefficient, represents the foundation structure relaxation potential energy coefficient, Ψ represents the dynamic foundation relaxation potential energy function, Hre represents the reference foundation depth, and ln represents the logarithmic function; The dynamic foundation relaxation potential energy function Ψ is obtained through the following formula: ； In the formula, A represents the integral area of the spatial region, D(t, x, y) represents the settlement amount at the spatial position (x, y) at time t, ∂ represents the partial derivative, and dxdy represents the two-dimensional integral operation.

5. The analysis system for highway foundation bearing characteristics according to claim 4, characterized in that: The hysteretic response model construction module outputs the obtained penetration lag coefficient Pr and the collapsibility response delay rate Tw as dynamic factors, and constructs a three-dimensional hysteretic response function S(t, x, y) in the form of time-weighted integration; The three-dimensional hysteretic response function S(t, x, y) is obtained through the following formula: ； where \(t_x\) represents the integration time variable, and \(t_x\in[0,t]\), \(e\) represents a constant, represents the memory decay factor, represents the spatial diffusion weight function; Spatial diffusion weight function Obtained by the following formula: ； In the formula, π represents the pi, σ(tx) represents the standard deviation of disturbance diffusion, exp represents the exponential function, and (xo, yo) represents the position coordinates of the disturbance source; Perform multi-scale reconstruction analysis on the obtained three-dimensional hysteretic response function S(t, x, y), extract the response trends from different time windows, including the average response value of the time window and the response fluctuation intensity coefficient, and use them to distinguish the response mutation growth region, the long-term slow change region of the response, and the abnormal points of the repeated disturbance response frequency; The average response value of the time window is obtained by integrating the three-dimensional hysteretic response function S(t, x, y) at the spatial position (x, y) within T time units and then dividing by the length T of the time interval. The response fluctuation intensity coefficient is obtained by calculating the absolute value average of the time change rate of the three-dimensional hysteretic response function S(t, x, y) within T time units.

6. The highway foundation bearing characteristic analysis system according to claim 5, characterized in that: The foundation state mapping and time series analysis module includes a time series change rate calculation unit and a response spatial map generation unit; The time series change rate calculation unit calculates the response change rate ∂S(t, x, y) / ∂t of the foundation response per unit time by performing derivative analysis on the three-dimensional hysteretic response function S(t, x, y) at each fixed spatial position (x, y); The response change rate ∂S(t, x, y) / ∂t is obtained through the following formula: ； In the formula, S(t - Δt, x, y) represents the three-dimensional hysteretic response function at time t - Δt, and Δt represents the time interval.

7. The analysis system for the bearing characteristics of a highway foundation according to claim 6, wherein: The response spatial map generation unit performs spatial integration processing on the obtained response change rate ∂S(t, x, y) / ∂t, generates a spatial map ΔS(x, y) of the foundation response rate, and obtains the risk response gradient index ∇S through the spatial map ΔS(x, y); The spatial map ΔS(x, y) is obtained through the following formula: ； The risk response gradient index ∇S is obtained through the following formula: ； Compare the obtained risk response gradient index ∇S with the preset risk threshold Ts to judge the change rate difference state of the boundary region. The difference state of the change rate of the boundary region is obtained through the following matching method: When the risk response gradient index ∇S ≤ the risk threshold Ts, it indicates that the boundary region is in a normal state; When the risk response gradient index ∇S > the risk threshold Ts, it indicates that the boundary region is in an abnormal state.

8. The analysis system for the bearing characteristics of a highway foundation according to claim 1, wherein: The dynamic risk identification and response decision-making module compares the obtained spatial mapping diagram ΔS(x, y) with the dynamic threshold Tdy to generate the collapsible risk level map Λ(x, y); The dynamic threshold Tdy is not a fixed constant, but is adaptively adjusted according to the current spatial mapping diagram ΔS(x, y), the daily rainfall R, and the groundwater level height G; The dynamic threshold Tdy is obtained through the following formula; ； where μΔ represents the spatial average value of the spatial mapping diagram ΔS(x, y), σΔ represents the spatial standard deviation of the spatial mapping diagram ΔS(x, y), p1 represents the risk enhancement adjustment coefficient, p2 represents the environmental disturbance suppression adjustment coefficient, represents the collapsible environment index; Collapsible environment index Obtained by the following formula: ； In the formula, ηR represents the rainfall disturbance intensity factor, and ηG represents the abnormal jump factor of the groundwater level; The rainfall disturbance intensity factor ηR is obtained by integrating the rainfall change rate from time t - Tr to time t; The abnormal jump factor ηG of the groundwater level is obtained by the ratio of the difference between the groundwater level height at time t and the groundwater level height at time t - Δt to Δt; The collapsible risk level map Λ(x, y) is obtained through the following formula: 。 9. The analysis system for the bearing characteristics of a highway foundation according to claim 8, characterized in that: The evaluation output and reporting module generates a two-dimensional color heat map based on the collapsible risk level map Λ(x, y) to display different risk level partitions; The risk distribution and high-risk areas are displayed through the spatial mapping diagram ΔS(x, y); Through the three-dimensional lag response function S(t, x, y), at a fixed spatial position (x, y), a response curve changing with time is established; at time t0, a ground surface response map is established; The obtained two-dimensional color heat map, the display of risk distribution, high-risk areas, and the ground surface response map are summarized into a structured document, including PDF format and HTML format, and output.

10. A method for analyzing the bearing characteristics of a highway foundation, which is applied to a system for analyzing the bearing characteristics of a highway foundation according to any one of claims 1 to 9, and is characterized in that: It includes the following steps: Step 1: The environmental data collection and processing module collects the historical data of the highway foundation through sensors, fits it into the original data set W, and performs preprocessing to obtain the foundation data set DW; Step 2: The collapsible lag factor calculation module extracts the characteristics of the foundation data set DW, including the penetration lag coefficient Pr and the collapsible response delay rate Tw; Step 3: The lag response model construction module constructs the three-dimensional lag response function S(t, x, y) according to the obtained penetration lag coefficient Pr and collapsible response delay rate Tw; Step 4: The foundation state mapping and time series analysis module performs time series derivative analysis on the obtained three-dimensional lag response function S(t, x, y), obtains the sudden change speed of the foundation response, and generates the spatial mapping diagram ΔS(x, y); Step 5: The dynamic risk identification and response decision-making module compares the obtained spatial mapping diagram ΔS(x, y) with the dynamic threshold Tdy to generate the collapsible risk level map Λ(x, y); Step 6: The evaluation output and reporting module visually outputs the obtained collapsible risk level map Λ(x, y), spatial mapping diagram ΔS(x, y), and three-dimensional lag response function S(t, x, y).

Images