A highway foundation bearing characteristics analysis method and system

Abstract

The present invention discloses a method and system for analyzing the bearing characteristics of highway foundations, which relates to the technical field of highway foundation analysis. By constructing a dynamic adaptive risk discrimination threshold mechanism, a fundamental breakthrough is achieved in solving the problems of high misjudgment rate and poor climate adaptability brought about by traditional fixed thresholds. When calculating the dynamic threshold Tdy, this mechanism fully integrates the spatial statistical characteristics of the current foundation response intensity, and combines the comprehensive environmental disturbance characteristics of the wetting environment index to form a dynamic discrimination standard with environmentally sensitive adjustment capabilities. The system can automatically adjust the risk identification threshold according to the actual rainwater infiltration intensity and the dynamic jump of groundwater, so that under extreme conditions such as heavy rainfall and rapid groundwater rise, the system can actively lower the risk tolerance boundary and identify abnormal response areas in advance, while moderately relaxing it during dry or stable periods, significantly improving the real-time and adaptability of risk identification.

Description

Technical Field

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

[0002] In highway foundation engineering, areas such as soft soil treatment, collapsible loess foundation reinforcement, structural load-bearing characteristics assessment, and dynamic response control are becoming key research and application areas for high-grade highways, mountain roads, and new road networks in extreme climates. Specifically, identifying the structural response and predicting the bearing capacity of potentially collapsible foundations during the rainy season has become a key bottleneck restricting the safe and stable operation of foundations.

[0003] In current engineering practice, the management of collapsible foundations mostly relies on static geological survey reports or the setting of safety reserve factors based on manual experience rules, and is heavily reliant on theoretical estimates in the early stages of construction. However, this approach makes it difficult to dynamically reflect the changing trends of the real-time impact of factors such as "seasonal hydrological disturbances," "local heavy rainfall events," and "rapid groundwater rise" on foundation structures. In addition, most existing systems lack a dynamic monitoring and prediction mechanism based on the combination of historical response behavior and environmental conditions, making it impossible to proactively discover potential structural hysteresis evolution processes, resulting in the inability to provide early warning of foundation collapsible behavior under specific climatic conditions.

[0004] Traditional foundation bearing capacity assessment models ignore the time lag inherent in the hydrological-structural coupling relationship. Rainwater exhibits distinct asynchronous evolutionary paths in infiltration, seepage, groundwater level disturbance, and structural relaxation, particularly in collapsible soils. However, due to the lack of predictive mechanisms and response models linked to time-series data, the phenomenon of "the ground remaining stable while the underlying structure is actually destabilizing and experiencing a sudden drop in strength" is often observed. Summary of the Invention

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

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a highway foundation bearing characteristics analysis system, including an environmental data acquisition and processing module, a collapsibility hysteresis factor calculation module, a hysteresis 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;

[0007] The environmental data acquisition 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;

[0008] The collapsibility hysteresis factor calculation module extracts features from the foundation dataset DW, including the seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw;

[0009] The hysteresis response model construction module constructs a three-dimensional hysteresis response function S (t, x, y) based on the obtained seepage hysteresis coefficient Pr and collapsibility response delay rate Tw;

[0010] 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);

[0011] 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);

[0012] 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).

[0013] Preferably, the environmental data acquisition and processing module includes a raw data acquisition unit and a data pre-processing unit;

[0014] The raw 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 raw data set W;

[0015] The daily rainfall R is collected by the tipping bucket rain gauges in the elevated meteorological stations around the roadbed;

[0016] The groundwater level G is obtained by a static pressure groundwater level meter buried at a depth of 10 to 20 m below the foundation;

[0017] The foundation settlement D is obtained by using ground displacement monitoring points arranged along the foundation cross section and a laser leveler;

[0018] 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;

[0019] Missing value processing is done by filling in the missing values ​​of the original data set W using linear interpolation;

[0020] Outlier cleaning is done by removing outliers from the original data set W using the triple standard deviation method;

[0021] Normalization processing includes normalizing the data in the original data set W using a normalization method to obtain a ground-based data set DW;

[0022] The ground-based dataset DW is obtained using the following formula:

[0023] ;

[0024] Where DWo represents the oth data item in the ground-based dataset DW, Wo represents the oth data item in the original dataset W, minWo represents the valley value of the oth data item in the original dataset W, and maxWo represents the peak value of the oth data item in the original dataset W.

[0025] Preferably, the collapsible hysteresis factor calculation module includes a multi-scale hysteresis modeling unit and a structural relaxation response analysis unit;

[0026] 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;

[0027] The permeation hysteresis coefficient Pr is obtained by the following formula:

[0028] ;

[0029] Where Pr(t) represents the seepage hysteresis 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 that controls the rainfall response, d represents the derivative symbol, dG(t) / dt represents the groundwater level change rate, and d 2 R(t) / dt 2 represents the acceleration of rainfall;

[0030] The obtained permeability hysteresis coefficient Pr was de-noised by using exponential sliding average;

[0031] The noise removal formula is:

[0032] ;

[0033] where λ represents the smoothing factor and λ∈(0,1), and Pr(t-1) represents the permeation hysteresis coefficient at time t-1.

[0034] Preferably, the structural relaxation response analysis unit identifies the time delay path of the structural response based on the obtained seepage hysteresis coefficient Pr and combined with the groundwater level height G, and extracts the collapsible response delay rate Tw;

[0035] The collapsibility response delay rate Tw is obtained by the following formula:

[0036] ;

[0037] Where, represents the osmotic response rate adjustment coefficient, represents the groundwater level influence coefficient, represents the relaxation potential energy coefficient of the foundation structure, 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;

[0038] The dynamic foundation relaxation potential energy function Ψ is obtained by the following formula:

[0039] ;

[0040] Where A represents the integrated area of ​​the spatial region, D(t, x, y) represents the amount of settlement at the spatial position (x, y) at time t, ∂ represents the partial derivative, and dxdy represents the two-dimensional integral operation.

[0041] Preferably, the hysteresis response model construction module outputs the obtained seepage hysteresis coefficient Pr and collapsible response delay rate Tw as dynamic factors, and constructs a three-dimensional hysteresis response function S (t, x, y) through a time-weighted integral form;

[0042] The three-dimensional hysteresis response function S(t,x,y) is obtained by the following formula:

[0043] ;

[0044] Where 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;

[0045] Spatial diffusion weight function Obtained by the following formula:

[0046] ;

[0047] Where π represents the circumference of a circle and its value is 3.14, σ(tx) represents the standard deviation of the disturbance diffusion, exp represents the exponential function, and (xo, yo) represents the location coordinates of the disturbance source, such as the center of heavy rainfall and the infiltration concentration point.

[0048] Multi-scale reconstruction analysis is performed on the obtained three-dimensional lagged response function S(t,x,y), and response trends are extracted from different time windows, including the average response value and response fluctuation intensity coefficient of the time window. These trends are used to distinguish areas of sudden response growth, areas of long-term slow response, and abnormal points of repeated perturbation response frequency.

[0049] The time window average response value is obtained by integrating the three-dimensional lag response function S(t, x, y) of the spatial position (x, y) within T time units and then dividing it by the time interval length T;

[0050] The response fluctuation intensity coefficient is obtained by calculating the average absolute value of the time change rate of the three-dimensional lag response function S (t, x, y) within T time units.

[0051] Preferably, the foundation state mapping and time series analysis module includes a time series change rate calculation unit and a response space map generation unit;

[0052] The time series change rate calculation unit calculates the response change rate ∂S(t,x,y) / ∂t of the foundation 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);

[0053] The response change rate ∂S(t,x,y) / ∂t is obtained by the following formula:

[0054] ;

[0055] Where S(t-Δt,x,y) represents the three-dimensional hysteresis response function at time t-Δt, and Δt represents the time interval.

[0056] Preferably, the response space map generation unit performs spatial integration processing on the acquired response change rate ∂S(t,x,y) / ∂t to generate a spatial mapping map ΔS(x,y) of the foundation response rate, and obtains the risk response gradient index ∇S through the spatial mapping map ΔS(x,y);

[0057] The spatial mapping ΔS(x, y) is obtained by the following formula:

[0058] ;

[0059] The risk response gradient index ∇S is obtained by the following formula:

[0060] ;

[0061] Compare the obtained risk response gradient index ∇S with the preset risk threshold Ts to determine the change rate difference state of the boundary area;

[0062] The change rate difference state of the boundary area is obtained by matching:

[0063] When the risk response gradient index ∇S ≤ the risk threshold Ts, it means that the boundary area is in a normal state;

[0064] When the risk response gradient index ∇S>risk threshold Ts, it means that the boundary area is in an abnormal state, indicating that the response changes steeply near the area and belongs to the "response jump zone" or "structural transition zone".

[0065] Preferably, the dynamic risk identification and response decision module compares the acquired spatial mapping image ΔS (x, y) with the dynamic threshold value Tdy to generate a collapsible risk level map Λ (x, y);

[0066] The dynamic threshold Tdy is not a fixed constant, but is adaptively adjusted according to the current spatial map ΔS (x, y), daily rainfall R and groundwater level G;

[0067] The dynamic threshold Tdy is obtained by the following formula;

[0068] ;

[0069] Where μΔ represents the spatial mean of the spatial mapping graph ΔS(x, y), σΔ represents the spatial standard deviation of the spatial mapping graph ΔS(x, y), p1 represents the risk enhancement adjustment coefficient, and p2 represents the environmental disturbance suppression adjustment coefficient. represents the collapsible environment index;

[0070] Collapsible Environment Index Obtained by the following formula:

[0071] ;

[0072] Where, ηR represents the rainfall disturbance intensity factor, ηG represents the groundwater level abnormal jump factor;

[0073] The rainfall disturbance intensity factor ηR is obtained by integrating the rainfall change rate from time t-Tr to time t;

[0074] The abnormal jump factor ηG of groundwater level is obtained by the ratio of the difference between the groundwater level at time t and the groundwater level at time t-Δt to Δt;

[0075] The collapsibility risk level map Λ(x, y) is obtained by the following formula:

[0076] .

[0077] Preferably, the assessment output and reporting module generates a two-dimensional color heat map based on the collapsible risk level map Λ(x, y), showing different risk level zones;

[0078] Common color scheme: green for stable, yellow for medium risk, and red for high risk;

[0079] The risk distribution and high-risk areas are displayed through the spatial map ΔS(x, y);

[0080] Through the three-dimensional hysteresis response function S (t, x, y), a response-time curve is established at a fixed spatial position (x, y); at time t0, a foundation surface response diagram is established;

[0081] The acquired two-dimensional color heat map, risk distribution display, high-risk area and foundation surface response map are summarized into structured documents, including PDF and HTML formats, and output.

[0082] A method for analyzing the bearing characteristics of a highway foundation comprises the following steps:

[0083] Step 1: The environmental data acquisition 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;

[0084] Step 2: The collapsibility hysteresis factor calculation module extracts features from the foundation dataset DW, including the seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw;

[0085] Step 3: The hysteresis response model construction module constructs a three-dimensional hysteresis response function S (t, x, y) based on the obtained seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw;

[0086] Step 4: The foundation state mapping and time series analysis module performs time series derivative analysis on the obtained three-dimensional hysteresis response function S (t, x, y) to obtain the sudden change speed of the foundation response and generate a spatial mapping diagram ΔS (x, y);

[0087] Step 5: The dynamic risk identification and response decision module compares the acquired spatial mapping image ΔS (x, y) with the dynamic threshold value Tdy to generate a collapsible risk level map Λ (x, y);

[0088] Step 6. 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).

[0089] The present invention provides a method and system for analyzing the bearing characteristics of a highway foundation, which has the following beneficial effects:

[0090] (1) During system operation, the acquisition paths of three key environmental and structural data, namely daily rainfall R, groundwater level G, and foundation settlement D, were systematically designed through the design of the original data collection method. These data were then deployed and collected 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 the data sources. Unlike the traditional method of using single survey data or static charts as criteria, this system continuously collects historical foundation environmental data over a long period of time, establishing for the first time 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.

[0091] (2) By introducing a structural response delay modeling mechanism, a dynamic parameter, the collapsible response delay rate, is systematically defined. Unlike previous simple threshold judgments that rely solely on groundwater level or settlement rate, this system, for the first time, incorporates the "time delay path" of structural response into the analysis scope and constructs a dynamic foundation relaxation potential energy function in combination with the foundation settlement gradient, thereby accurately identifying the hysteresis deformation trend of the foundation structure after hydrological stimulation. This design effectively fills the gap in traditional bearing capacity evaluation methods that are difficult to identify "non-instantaneous response processes" and significantly improves the depth of identification of collapsible precursors.

[0092] (3) By constructing a dynamic adaptive risk identification threshold mechanism, a fundamental breakthrough has been achieved in solving the problems of high misjudgment rate and poor climate adaptability brought about by traditional fixed thresholds. When calculating the dynamic threshold Tdy, this mechanism fully integrates the spatial statistical characteristics of the current foundation response intensity and combines the comprehensive environmental disturbance characteristics of the collapsible environmental index to form a dynamic identification standard with environmentally sensitive adjustment capabilities. The system can automatically adjust the risk identification threshold according to the actual rainwater infiltration intensity and the dynamic jump of groundwater. Under extreme conditions such as heavy rainfall and rapid groundwater rise, the system can actively lower the risk tolerance boundary and identify abnormal response areas in advance, while moderately relaxing it during dry or stable periods, significantly improving the real-time and adaptability of risk identification.

[0093] (4) Through the time-varying integral and spatial diffusion weight mechanism, the evolution trend of the disturbance propagating from the source point to different areas 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 environments, greatly improving the accuracy and predictability of load 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 response surge area and the spatial non-uniform deformation zone. Compared with the traditional static judgment method, it is more timely 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 fluctuation to generate a dynamic threshold Tdy, thereby realizing an intelligent control mechanism in which the risk identification sensitivity is adaptively adjusted with environmental disturbances, effectively improving the adaptability and accuracy of risk identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 This is a schematic diagram of the flow chart of a highway foundation bearing characteristics analysis system according to the present invention;

[0095] Figure 2 This is a schematic diagram of the steps of a method for analyzing the bearing characteristics of a highway foundation according to the present invention;

[0096] Figure 3 It is a schematic diagram of the process of the collapsible risk level map of the present invention;

[0097] Figure 4 It is a bar graph of the collapsibility response delay rate of the present invention. DETAILED DESCRIPTION

[0098] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0099] Example 1

[0100] The present invention provides a highway foundation bearing characteristics analysis system, please refer to Figures 1 to 4 , including environmental data acquisition and processing module, collapsibility 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 reporting module;

[0101] The environmental data acquisition 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;

[0102] The collapsibility hysteresis factor calculation module extracts features from the foundation dataset DW, including the seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw;

[0103] The hysteresis response model construction module constructs a three-dimensional hysteresis response function S (t, x, y) based on the obtained seepage hysteresis coefficient Pr and collapsibility response delay rate Tw;

[0104] 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);

[0105] 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);

[0106] 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).

[0107] In this embodiment, a modular six-level structure establishes a complete analysis chain from environmental data collection, factor identification, response modeling, to risk assessment and visualization output. This ensures a seamless information flow, rigorous timing logic, and more efficient response. All functions of this system process and model data collected by existing sensors, without changing the on-site hardware structure. This system demonstrates the technical advantages of low cost, easy deployment, and ease of integration and upgrade. It is suitable for both new highways and the intelligent transformation of existing roads.

[0108] The system innovatively proposes the seepage hysteresis coefficient (Pr) and the collapsibility response delay rate (Tw) as key characteristic quantities. By addressing the asynchrony of hydrological perturbation responses, it can capture potential response delays that traditional systems cannot identify, effectively identifying precursors to collapsibility. By modeling the dynamic evolution of foundation responses in time and space, the system can proactively detect foundation stress relaxation or settlement trends during high-risk periods such as the rainy season, significantly enhancing the system's seasonal adaptability and risk prevention capabilities.

[0109] The foundation state mapping and time series analysis module performs derivative analysis of local response rates and displays them graphically, enabling real-time identification of areas with sudden structural changes and active risk zones, providing a quantitative basis for precise governance. By introducing an adaptive dynamic threshold mechanism, the system automatically adjusts identification criteria, enabling the early warning system to adapt to varying weather, geological, and load conditions, achieving a "tolerant during stable periods, sensitive during abnormal periods" discrimination strategy.

[0110] Example 2

[0111] 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;

[0112] The raw 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 raw data set W;

[0113] The daily rainfall R is collected by the tipping bucket rain gauges in the elevated meteorological stations around the roadbed;

[0114] The groundwater level G is obtained by a static pressure groundwater level meter buried at a depth of 10 to 20 m below the foundation;

[0115] The foundation settlement D is obtained by using ground displacement monitoring points arranged along the foundation cross section and a laser leveler;

[0116] 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;

[0117] Missing value processing is done by filling in the missing values ​​of the original data set W using linear interpolation;

[0118] Outlier cleaning is done by removing outliers from the original data set W using the triple standard deviation method;

[0119] Normalization processing includes normalizing the data in the original data set W using a normalization method to obtain a ground-based data set DW;

[0120] The ground-based dataset DW is obtained using the following formula:

[0121] ;

[0122] Where DWo represents the oth data item in the ground-based dataset DW, Wo represents the oth data item in the original dataset W, minWo represents the valley value of the oth data item in the original dataset W, and maxWo represents the peak value of the oth data item in the original dataset W.

[0123] The collapsible hysteresis factor calculation module includes a multi-scale hysteresis modeling unit and a structural relaxation response analysis unit;

[0124] 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;

[0125] The permeation hysteresis coefficient Pr is obtained by the following formula:

[0126] ;

[0127] Where Pr(t) represents the seepage hysteresis 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 that controls the rainfall response, d represents the derivative symbol, dG(t) / dt represents the groundwater level change rate, and d 2 R(t) / dt 2 represents the acceleration of rainfall;

[0128] The obtained permeability hysteresis coefficient Pr was de-noised by using exponential sliding average;

[0129] The noise removal formula is:

[0130] ;

[0131] where λ represents the smoothing factor and λ∈(0,1), and Pr(t-1) represents the permeation hysteresis coefficient at time t-1.

[0132] This embodiment systematically designs the raw data collection method, clarifying the acquisition paths for three key environmental and structural data types: daily rainfall R, groundwater level G, and foundation settlement D. These data are then deployed and collected using existing sensing equipment such as tipping bucket rain gauges, hydrostatic groundwater level gauges, and laser levels, ensuring the real-time, representative, and engineering-appropriate nature of the data sources. Unlike traditional methods that rely on single-shot survey data or static charts, this system continuously collects historical foundation environmental data over a long period of time, establishing for the first time a high-density raw data set suitable for "seasonal response trend identification," providing a solid data foundation for dynamic modeling and collapsible evolution prediction.

[0133] This example introduces a combined strategy of linear interpolation and triple standard deviation to repair and remove missing values ​​and outliers, respectively, significantly improving the integrity and anti-interference capabilities of the ground-based dataset. Furthermore, through normalization, the data dimensions of different physical quantities are unified, making various environmental factors comparable within the same evaluation system. This constructs a high-quality ground-based dataset that can directly participate in modeling and analysis, effectively addressing the problems of traditional data clutter, limited processing methods, and inability to support dynamic modeling.

[0134] The method for calculating the seepage hysteresis coefficient introduced in this example breaks away from previous single-factor analysis models based solely on water level or rainfall intensity. Instead, it establishes a dynamic factor that simultaneously characterizes the impact of groundwater's response speed to rainfall and the changing trend of rainfall rate. Furthermore, an exponential moving average technique is used to remove high-frequency fluctuations, forming a response characteristic sequence with temporal stability. This design not only accurately identifies asynchronous hydrological-structural relationships but also quantitatively characterizes the potential delayed response of foundation structural factors during periods of heavy rainfall, thereby enhancing dynamic adaptability and the ability to capture precursors of subsidence.

[0135] Example 3

[0136] 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 based on the obtained seepage hysteresis coefficient Pr and combined with the groundwater level height G, and extracts the collapsible response delay rate Tw;

[0137] The collapsibility response delay rate Tw is obtained by the following formula:

[0138] ;

[0139] Where, represents the osmotic response rate adjustment coefficient, represents the groundwater level influence coefficient, represents the relaxation potential energy coefficient of foundation structure, Ψ represents the dynamic foundation relaxation potential energy function, Hre represents the reference foundation depth, and ln represents the logarithmic function;

[0140] Specific examples:

[0141] Table 1: Example table for calculating the delay rate of collapsible response;

[0142]

[0143] The dynamic foundation relaxation potential energy function Ψ is obtained by the following formula:

[0144] ;

[0145] Where A represents the integrated area of ​​the spatial region, D(t, x, y) represents the amount of settlement at the spatial position (x, y) at time t, ∂ represents the partial derivative, and dxdy represents the two-dimensional integral operation.

[0146] The hysteresis response model construction module outputs the obtained seepage hysteresis coefficient Pr and collapsibility response delay rate Tw as dynamic factors, and constructs a three-dimensional hysteresis response function S (t, x, y) through time-weighted integration.

[0147] The three-dimensional hysteresis response function S(t,x,y) is obtained by the following formula:

[0148] ;

[0149] Where 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;

[0150] Spatial diffusion weight function Obtained by the following formula:

[0151] ;

[0152] Where π represents the circumference of a circle, σ(tx) represents the standard deviation of the disturbance diffusion, exp represents the exponential function, and (xo, yo) represents the position coordinates of the disturbance source.

[0153] Multi-scale reconstruction analysis is performed on the obtained three-dimensional lagged response function S(t,x,y), and response trends are extracted from different time windows, including the average response value and response fluctuation intensity coefficient of the time window. These trends are used to distinguish areas of sudden response growth, areas of long-term slow response, and abnormal points of repeated perturbation response frequency.

[0154] The time window average response value is obtained by integrating the three-dimensional lag response function S(t, x, y) of the spatial position (x, y) within T time units and then dividing it by the time interval length T;

[0155] The response fluctuation intensity coefficient is obtained by calculating the average absolute value of the time change rate of the three-dimensional lag response function S (t, x, y) within T time units.

[0156] Based on the previous calculation of the 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. Unlike previous simple threshold judgments that only rely on groundwater level or settlement rate, this system for the first time incorporates the "time delay path" of structural response into the analysis scope, and constructs a dynamic foundation relaxation potential energy function based on the foundation settlement gradient, thereby accurately identifying the hysteretic deformation trend of the foundation structure after hydrological stimulation. This design effectively fills the gap in traditional bearing capacity evaluation methods that are difficult to identify "non-instantaneous response processes" and greatly improves the depth of identification of collapsible precursors.

[0157] This example constructs a three-dimensional hysteresis response function model with the seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw as dual core driving factors. This model incorporates historical disturbance effects through a time-weighted integral and introduces a spatial diffusion weighting function to simulate the realistic propagation of disturbance sources across time and space. In particular, the introduction of a disturbance memory decay mechanism and a Gaussian spatial diffusion kernel function simulates the heterogeneity and path attenuation effects of rainwater diffusion in the foundation, significantly improving the system's response prediction accuracy under complex climate conditions and addressing key shortcomings of traditional models, such as insufficient spatiotemporal coupling and the absence of a disturbance diffusion mechanism.

[0158] The system also conducts in-depth analysis of the three-dimensional hysteresis response function S(t,x,y) through multi-scale response reconstruction analysis, constructing two advanced features: the time window average response value and the response fluctuation intensity coefficient, which correspond to the ability to identify long-term slowly varying settlement trends and sudden response jump behaviors, respectively. This analytical approach not only enables the regional identification of "chronic structural deformation," "sudden response growth," and "abnormal points in the frequency of repeated disturbances," but also enables the system to identify a continuous state from stable to fluctuating to risky to critical, providing data support for subsequent risk level assessment and proactive intervention strategies.

[0159] 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. This effectively addresses 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 achieving early warning and zoning management of subsidence foundations during the rainy season.

[0160] Example 4

[0161] 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;

[0162] The time series change rate calculation unit calculates the response change rate ∂S(t,x,y) / ∂t of the foundation 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);

[0163] The response change rate ∂S(t,x,y) / ∂t is obtained by the following formula:

[0164] ;

[0165] Where S(t-Δt,x,y) represents the three-dimensional hysteresis response function at time t-Δt, and Δt represents the time interval.

[0166] The response space map generation unit spatially integrates the acquired response change rate ∂S(t,x,y) / ∂t to generate a spatial mapping map ΔS(x,y) of the foundation response rate, and obtains the risk response gradient index ∇S through the spatial mapping map ΔS(x,y);

[0167] The spatial mapping ΔS(x, y) is obtained by the following formula:

[0168] ;

[0169] The risk response gradient index ∇S is obtained by the following formula:

[0170] ;

[0171] Compare the obtained risk response gradient index ∇S with the preset risk threshold Ts to determine the change rate difference state of the boundary area;

[0172] The change rate difference state of the boundary area is obtained by matching:

[0173] When the risk response gradient index ∇S ≤ the risk threshold Ts, it means that the boundary area is in a normal state;

[0174] When the risk response gradient index ∇S>risk threshold Ts, it means that the boundary area is in an abnormal state.

[0175] Based on the existing three-dimensional hysteresis response function S(t,x,y) modeling, this embodiment conducts in-depth exploration of the spatial mutation characteristics in the response evolution process, introduces the derivative behavior of the "response change rate" into the foundation hysteresis response analysis system for the first time, and establishes a time series change rate calculation unit. It can estimate the derivative of the three-dimensional hysteresis response function S(t,x,y) at any fixed spatial position (x,y) in the time direction, thereby obtaining the speed of change of the foundation structure response per unit time.

[0176] The response space map generation unit integrates the response rate values ​​at each point into a spatial map of the full regional distribution, ΔS(x, y). This enables high-resolution spatial visualization of the foundation response rate, effectively depicting the continuity and heterogeneity of the response distribution. This embodiment introduces a risk response gradient index, ∇S, to determine the gradient jumps in the response rates of adjacent regions, mathematically quantifying the magnitude of the sudden change at the boundary of the foundation structure.

[0177] By comparing the risk response gradient index ∇S with the preset risk threshold Ts, the system can not only determine whether a certain location has reached the standard line of response intensity, but more importantly, it can identify whether the mutation trend of the regional boundary response is abnormal, and then distinguish between the spatial difference transition zone and the active structural imbalance zone, realizing the automatic identification and state classification of abnormal behavior in the structural transition zone.

[0178] This embodiment establishes a linkage analysis system in the two dimensions of time derivative and spatial gradient, thereby achieving accurate capture of foundation response mutation points, boundary jump zones, and dynamic discontinuity areas, effectively making up for the shortcomings of existing technologies in terms of insufficient accuracy in boundary change recognition and difficulty in continuous tracking of dynamic responses, and further improving the system's intelligence level and engineering adaptability in foundation structure continuity recognition, sensitive area positioning, and structural safety zoning.

[0179] Example 5

[0180] This embodiment is explained in Example 4. Please refer to Figure 1 and Figure 3 ,Specifically: the dynamic risk identification and response decision module compares the ,acquired spatial mapping ΔS(x, y) with the dynamic threshold Tdy to generate the ,collapse risk level map Λ(x, y);

[0181] The dynamic threshold Tdy is not a fixed constant, but is adaptively adjusted according to the current spatial map ΔS (x, y), daily rainfall R and groundwater level G;

[0182] The dynamic threshold Tdy is obtained by the following formula;

[0183] ;

[0184] Where μΔ represents the spatial mean of the spatial mapping graph ΔS(x, y), σΔ represents the spatial standard deviation of the spatial mapping graph ΔS(x, y), p1 represents the risk enhancement adjustment coefficient, and p2 represents the environmental disturbance suppression adjustment coefficient. represents the collapsible environment index;

[0185] Collapsible Environment Index Obtained by the following formula:

[0186] ;

[0187] Where, ηR represents the rainfall disturbance intensity factor, ηG represents the groundwater level abnormal jump factor;

[0188] The rainfall disturbance intensity factor ηR is obtained by integrating the rainfall change rate from time t-Tr to time t;

[0189] The abnormal jump factor ηG of groundwater level is obtained by the ratio of the difference between the groundwater level at time t and the groundwater level at time t-Δt to Δt;

[0190] The collapsibility risk level map Λ(x, y) is obtained by the following formula:

[0191] .

[0192] The assessment output and reporting module generates a two-dimensional color heat map based on the collapse risk level map Λ(x, y), showing different risk level zones;

[0193] The risk distribution and high-risk areas are displayed through the spatial map ΔS(x, y);

[0194] Through the three-dimensional hysteresis response function S (t, x, y), a response-time curve is established at a fixed spatial position (x, y); at time t0, a foundation surface response diagram is established;

[0195] The acquired two-dimensional color heat map, risk distribution display, high-risk area and foundation surface response map are summarized into structured documents, including PDF and HTML formats, and output.

[0196] This embodiment achieves a fundamental breakthrough in solving the problems of high misjudgment rate and poor climate adaptability brought about by traditional fixed thresholds by constructing a dynamic adaptive risk discrimination threshold mechanism. When calculating the dynamic threshold Tdy, this mechanism fully integrates the spatial statistical characteristics of the current foundation response intensity and combines the comprehensive environmental disturbance characteristics of the collapsible environmental index to form a dynamic discrimination standard with environmentally sensitive adjustment capabilities. The system can automatically adjust the risk identification threshold according to the actual rainwater infiltration intensity and the dynamic jump of groundwater. Under 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 it in dry or stable periods, significantly improving the real-time and adaptability of risk identification.

[0197] This example proposes a joint modeling of the rainfall disturbance intensity factor and the groundwater level abnormal jump factor, comprehensively quantifying the sudden change trend of the structure's external environment. By integrating the rainfall rate change over a time window and calculating the change ratio of the groundwater fluctuation gradient, the system can identify potential hazardous conditions in real time when the structure remains untouched but the environment changes first, providing accurate advance information for foundation structure response modeling. This mechanism effectively identifies the potential collapsibility triggering mechanism, which is easily overlooked in traditional methods, further expanding the time window for risk identification.

[0198] This example implements spatial hierarchical management of different risk areas by constructing a collapsible risk level map Λ(x, y). The distribution of high-risk areas is visually displayed through a two-dimensional color heat map. Furthermore, the system incorporates a three-dimensional hysteresis response function S(t, x, y) to construct response-time curves for fixed spatial points and spatial response profiles for fixed time points, respectively. This transforms the evolution of foundation conditions into a visual trajectory. Finally, by compiling all analysis results into a structured report, a closed-loop process is achieved, from data analysis to chart output to document archiving. This significantly enhances the interpretability and verifiability of the results, as well as their practical value for project scheduling.

[0199] Example 6

[0200] A highway foundation bearing characteristics analysis method, please refer to Figure 2 , specifically: including the following steps:

[0201] Step 1: The environmental data acquisition 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;

[0202] Step 2: The collapsibility hysteresis factor calculation module extracts features from the foundation dataset DW, including the seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw;

[0203] Step 3: The hysteresis response model construction module constructs a three-dimensional hysteresis response function S (t, x, y) based on the obtained seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw;

[0204] Step 4: The foundation state mapping and time series analysis module performs time series derivative analysis on the obtained three-dimensional hysteresis response function S (t, x, y) to obtain the sudden change speed of the foundation response and generate a spatial mapping diagram ΔS (x, y);

[0205] Step 5: The dynamic risk identification and response decision module compares the acquired spatial mapping image ΔS (x, y) with the dynamic threshold value Tdy to generate a collapsible risk level map Λ (x, y);

[0206] Step 6. 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).

[0207] In this embodiment, through the synchronous collection and preprocessing of multi-source data such as highway foundation rainfall, groundwater and settlement in the first step, a systematic foundation operation history dataset DW was established. This effectively overcomes the problem of traditional methods relying solely on static geological survey data or artificial experience parameters, and enables the foundation bearing characteristic analysis to have high temporal resolution and full-process perception capabilities.

[0208] The second step, collapsibility hysteresis factor extraction, constructs the permeability hysteresis coefficient Pr and the collapsibility response delay rate Tw, respectively, revealing the asynchronous and delayed dynamic evolution path of the foundation structure in response to hydrological disturbances. These two parameters not only fill the gap in traditional static indicators in characterizing the dynamic behavior of structures, but also provide key driving factors for subsequent predictive modeling.

[0209] Through a time-varying integral and spatial diffusion weighting mechanism, the evolutionary trend of disturbances propagating from the source to different areas of the foundation is simulated. This model comprehensively characterizes the cumulative and spatial evolution of a structure's response under long-term loads and complex environments, significantly improving the accuracy and predictability of load-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 areas of sudden response increases and spatially inhomogeneous deformation zones. Compared to traditional static judgment methods, this method is more timely and capable of identifying boundary anomalies, crack development, and load concentration areas. With the introduction of the fifth step, the method no longer uses a single static judgment benchmark. Instead, it combines the spatial mapping ΔS(x, y), rainfall intensity, and groundwater fluctuations to generate a dynamic threshold Tdy. This implements an intelligent control mechanism that adaptively adjusts risk identification sensitivity to environmental disturbances, effectively improving the adaptability and accuracy of risk identification.

[0210] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A highway foundation bearing characteristics analysis system, characterized by: 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 reporting module; The environmental data acquisition 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 dataset DW, including the seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw; The permeation hysteresis coefficient Pr is obtained by the following formula: ； Where Pr(t) represents the seepage hysteresis 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 that controls the rainfall response, d represents the derivative symbol, dG(t) / dt represents the groundwater level change rate, and d 2 R(t) / dt 2 represents the acceleration of rainfall; The collapsibility response delay rate Tw is obtained by the following formula: ； Where, represents the osmotic response rate adjustment coefficient, represents the groundwater level influence coefficient, represents the relaxation potential energy coefficient of foundation structure, 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 by the following formula: ； Where A represents the integrated area of ​​the spatial region, D(t, x, y) represents the amount of settlement at the spatial position (x, y) at time t, ∂ represents the partial derivative, and dxdy represents the two-dimensional integral operation; The hysteresis response model construction module constructs a three-dimensional hysteresis response function S (t, x, y) based on the obtained seepage hysteresis coefficient Pr and collapsibility response delay rate Tw; The three-dimensional hysteresis response function S(t,x,y) is obtained by the following formula: ； Where 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; Spatial diffusion weight function Obtained by the following formula: ； Where π represents the circumference of a circle, σ(tx) represents the standard deviation of the disturbance diffusion, exp represents the exponential function, and (xo, yo) represents the position coordinates of the disturbance source. 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 dynamic threshold Tdy is obtained by the following formula; ； Where μΔ represents the spatial mean of the spatial mapping graph ΔS(x, y), σΔ represents the spatial standard deviation of the spatial mapping graph ΔS(x, y), p1 represents the risk enhancement adjustment coefficient, and p2 represents the environmental disturbance suppression adjustment coefficient. represents the collapsible environment index; Collapsible Environment Index Obtained by the following formula: ； Where, ηR represents the rainfall disturbance intensity factor, ηG represents the groundwater level abnormal jump factor; 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 groundwater level is obtained by the ratio of the difference between the groundwater level at time t and the groundwater level at time t-Δt to Δt; 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. A highway foundation bearing characteristics 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 pre-processing unit; The raw 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 raw data set W; The daily rainfall R is collected by the tipping bucket rain gauges in the elevated meteorological stations around the roadbed; The groundwater level G is obtained 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 by using ground displacement monitoring points arranged along the foundation cross section and a laser leveler; 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 in 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; Normalization processing 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 using the following formula: ； Where DWo represents the oth data item in the ground-based dataset DW, Wo represents the oth data item in the original dataset W, minWo represents the valley value of the oth data item in the original dataset W, and maxWo represents the peak value of the oth data item in the original dataset W.

3. A highway foundation bearing characteristics analysis system according to claim 2, characterized in that: 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 obtained permeability hysteresis coefficient Pr was de-noised by using exponential sliding average; The noise removal formula is: ； where λ represents the smoothing factor and λ∈(0,1), and Pr(t-1) represents the permeation hysteresis coefficient at time t-1.

4. A highway foundation bearing characteristics analysis system according to claim 3, characterized in that: The structural relaxation response analysis unit identifies the time delay path of the structural response based on the obtained seepage hysteresis coefficient Pr and combined with the groundwater level height G, and extracts the collapsible response delay rate Tw.

5. A highway foundation bearing characteristics analysis system according to claim 4, characterized in that: The hysteresis response model construction module outputs the obtained seepage hysteresis coefficient Pr and collapsibility response delay rate Tw as dynamic factors, and constructs a three-dimensional hysteresis response function S (t, x, y) through time-weighted integration. Multi-scale reconstruction analysis is performed on the obtained three-dimensional lagged response function S(t,x,y), and response trends are extracted from different time windows, including the average response value and response fluctuation intensity coefficient of the time window. These trends are used to distinguish areas of sudden response growth, areas of long-term slow response, and abnormal points of repeated perturbation response frequency. The time window average response value is obtained by integrating the three-dimensional lag response function S(t, x, y) of the spatial position (x, y) within T time units and then dividing it by the time interval length T; The response fluctuation intensity coefficient is obtained by calculating the average absolute value of the time change rate of the three-dimensional lag response function S (t, x, y) within T time units.

6. A highway foundation bearing characteristics 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 space map generation unit; The time series change rate calculation unit calculates the response change rate ∂S(t,x,y) / ∂t of the foundation 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 by the following formula: ； Where S(t-Δt,x,y) represents the three-dimensional hysteresis response function at time t-Δt, and Δt represents the time interval.

7. A highway foundation bearing characteristics analysis system according to claim 6, characterized in that: The response space map generation unit spatially integrates the acquired response change rate ∂S(t,x,y) / ∂t to generate a spatial mapping map ΔS(x,y) of the foundation response rate, and obtains the risk response gradient index ∇S through the spatial mapping map ΔS(x,y); The spatial mapping ΔS(x, y) is obtained by the following formula: ； The risk response gradient index ∇S is obtained by the following formula: ； Compare the obtained risk response gradient index ∇S with the preset risk threshold Ts to determine the change rate difference state of the boundary area; The change rate difference state of the boundary area is obtained by matching: 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>risk threshold Ts, it means that the boundary area is in an abnormal state.

8. The highway foundation bearing characteristics analysis system according to claim 1, characterized in that: The dynamic risk identification and response decision module compares the acquired spatial mapping 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 map ΔS (x, y), daily rainfall R and groundwater level G; The collapsibility risk level map Λ(x, y) is obtained by the following formula: 。 9. A highway foundation bearing characteristics analysis system according to claim 8, characterized in that: The assessment output and reporting module generates a two-dimensional color heat map based on the collapse risk level map Λ(x, y), showing different risk level zones; The risk distribution and high-risk areas are displayed through the spatial map ΔS(x, y); Through the three-dimensional hysteresis response function S (t, x, y), a response-time curve is established at a fixed spatial position (x, y); at time t0, a foundation surface response diagram is established; The acquired two-dimensional color heat map, risk distribution display, high-risk area and foundation surface response map are summarized into structured documents, including PDF and HTML formats, and output.

10. A method for analyzing the bearing characteristics of a highway foundation, applied to a system for analyzing the bearing characteristics of a highway foundation according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: The environmental data acquisition 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; Step 2: The collapsibility hysteresis factor calculation module extracts features from the foundation dataset DW, including the seepage hysteresis coefficient Pr and the collapsibility response delay rate Tw; Step 3: The hysteresis response model construction module constructs a three-dimensional hysteresis response function S (t, x, y) based on the obtained seepage hysteresis coefficient Pr and the 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 hysteresis response function S (t, x, y) to obtain the sudden change speed of the foundation response and generate a spatial mapping diagram ΔS (x, y); Step 5: The dynamic risk identification and response decision module compares the acquired spatial mapping image ΔS (x, y) with the dynamic threshold value Tdy to generate a collapsible risk level map Λ (x, y); Step 6. 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).

Images