An online health monitoring system for buildings

Abstract

This invention relates to the field of building monitoring technology, and more particularly to an online building health monitoring system, comprising an acquisition module, an identification module, a transmission determination module, a risk determination module, and an early warning module. This invention analyzes the temporal changes in normal stress and horizontal displacement within a slope and identifies the force-variable synergy patterns between them, enabling it to capture early mechanical precursors of slope instability. It introduces a support aging assessment to evaluate the response efficiency of the anchor system to stress changes, and adjusts the weights of stress, displacement, and axial force in the risk transmission assessment accordingly. Simultaneously, an adjustment factor intelligently lowers the early warning threshold for building deformation as slope risk increases, quantifying the correlation and spatial attenuation effect between slope displacement and building settlement. Furthermore, it integrates multiple parameters to generate a comprehensive correlation risk index, effectively solving the problems of low early warning accuracy and difficulty in timely detection of correlated risks due to isolated static monitoring and the lack of a risk transmission mechanism.

Description

Technical Field

[0001] This invention relates to the field of building monitoring technology, and in particular to an online building health monitoring system. Background Technology

[0002] In complex geological environments such as hilly and mountainous areas, slope-cutting for housing construction is a common land development practice. However, the resulting slope instability risks pose a continuous threat to the safety of adjacent buildings. Existing building health monitoring technologies mostly focus on the deformation, settlement, or vibration response of the building structure itself, often using isolated threshold judgments for early warning. Meanwhile, traditional slope monitoring systems often operate independently, making it difficult to effectively characterize the dynamic physical mechanisms of risk transmission from the slope to the building in time and space. Therefore, a system integrating multi-dimensional monitoring, real-time data analysis, and intelligent early warning is needed to ensure the comprehensive safety of both buildings and slopes.

[0003] Chinese Patent Publication No. CN113237885A discloses a method for evaluating building performance based on structural health monitoring data. This method includes acquiring the original three-dimensional drawings of the building to be evaluated, identifying and marking the original load-bearing nodes within the original three-dimensional drawings, and acquiring the three-dimensional coordinates of each original load-bearing node; constructing the current three-dimensional structural model of the building to be evaluated and acquiring the current three-dimensional coordinates of each load-bearing node; acquiring beam displacement parameters, beam deformation parameters, and floor slab displacement parameters based on the changes in the three-dimensional coordinates of each load-bearing node; and evaluating the building performance based on the beam displacement parameters, beam deformation parameters, and floor slab displacement parameters.

[0004] Therefore, the existing technology has the following problems: the method relies on the accuracy of the original 3D drawings, which can easily affect the evaluation results due to poor drawing quality or design negligence; the method identifies external defects through UAV images, which is easily affected by environmental interference and cannot detect internal damage; the method fails to link the building with the monitoring of its surrounding environment, which can easily ignore the related impact of the changing external environment on the building's health. Summary of the Invention

[0005] The purpose of this invention is to provide an online building health monitoring system to solve the problems mentioned in the background art.

[0006] This invention provides an online building health monitoring system, which overcomes the problems of low early warning accuracy and difficulty in timely detection of related risks in the prior art due to isolated and static monitoring and the lack of risk transmission mechanisms, by using multi-source data analysis and dynamic coupling of slope and building risks.

[0007] To achieve the above objectives, the present invention provides an online building health monitoring system, comprising: The acquisition module is used to acquire the tilt angle of the target building in the hilly and mountainous area, the differential settlement value of the central load-bearing column on the adjacent slope and the central load-bearing column on the back slope, the comprehensive axial force of the anchor rod on the adjacent slope, and the normal stress of the preset target point inside the slope soil and rock and the horizontal displacement of the preset monitoring point on the vertical profile. The identification module is used to identify the force-variable coordination mode based on the temporal changes of the normal stress and the horizontal displacement; The transmission determination module is used to determine the risk transmission index based on the temporal evolution characteristics and weight adjustment amount of the horizontal displacement, the normal stress and the comprehensive axial force. The weight adjustment amount is determined based on the support duration and the preset adjustment coefficient. The support duration is determined based on the temporal relative change of the normal stress and the comprehensive axial force under the force-variable synergy mode. The risk determination module is used to determine the building risk index based on the temporal changes of the differential settlement value and the tilt angle and the adjustment factor, and to determine the associated risk index based on the geometric mean of the building risk index, the risk transmission index, the temporal lag degree and the spatial coupling degree. The adjustment factor is determined based on the risk transmission index, and the temporal lag degree and the spatial coupling degree are determined based on the temporal characteristics and spatial characteristics of the horizontal displacement and the differential settlement value, respectively. The early warning module is used to issue early warnings based on the associated risk index and the force-change synergy mode.

[0008] Furthermore, the identification module includes: The change calculation unit is used to construct stress change sequence and displacement change sequence respectively based on the instantaneous rate of change of the normal stress and the horizontal displacement within a preset identification time period; The relevant determination unit is used to calculate the Pearson correlation coefficient of the stress change sequence and the displacement change sequence under different preset force hysteresis times, so as to obtain several sliding force correlation degrees. The identification unit is used to identify the force-variable cooperative mode based on all the sliding force-variable correlations.

[0009] Furthermore, the identification unit includes: The recording subunit is used to record the maximum value of all the sliding force correlations as the sliding correlation maximum value, and to record the preset force hysteresis time corresponding to the sliding correlation maximum value as the sliding offset maximum value; The identification subunit is used to identify the force-variable cooperative mode based on the threshold comparison results of the sliding correlation extrema and the sliding offset extrema, respectively.

[0010] Furthermore, the conduction determination module includes: An aging determination unit is used to determine the aging degree of the axial force based on the same-direction ratio and the average hysteresis, wherein the same-direction ratio and the average hysteresis are determined based on the coordinated change of the normal stress and the combined axial force; A comprehensive determination unit is used to determine the support aging based on the coordinated changes of the normal stress and the comprehensive axial force, as well as the axial aging aging degree. The transmission determination unit is used to determine the risk transmission index based on the horizontal displacement, the normal stress, the comprehensive axial force, and their respective adjustment weights, wherein the respective adjustment weights are determined based on the support aging period, the preset adjustment coefficient, and the weight adjustment amount.

[0011] Furthermore, the timeliness determination unit includes: A co-directional determination subunit is used to determine the co-directional ratio based on the ratio of differential co-directional moments within the preset identification time, wherein the differential co-directional moments are determined based on the co-directional moments of stress difference and axial force difference, and the stress difference and axial force difference are determined based on the first-order differences of the normal stress and the comprehensive axial force, respectively. A time-determining sub-unit is used to determine the hysteresis based on the time interval between the differential strain time and the axial force co-directional time, wherein the differential strain time is determined based on the stress difference and the differential co-directional time, and the axial force co-directional time is determined based on the differential strain time, the axial force difference, and the differential co-directional time. A timeliness determination subunit is used to determine the timeliness of the applied axis based on the average lag and the same-direction ratio, wherein the average lag is determined based on the average value of all the lags.

[0012] Furthermore, the comprehensive determination unit includes: The total amount determination subunit is used to determine the total stress change and the total axial force change based on the sum of the squares of all stress deviations and the sum of the squares of all axial force deviations, respectively. The stress deviation and axial force deviation are determined based on the relative mean deviations of the normal stress and the comprehensive axial force, respectively. A collaborative determination subunit is used to determine the axial force contribution based on the ratio of the load sharing degree to the preset sharing degree, wherein the load sharing degree is determined based on the collaborative change amount and the total stress change, and the collaborative change amount is determined based on the stress deviation, the axial force deviation and the preset identification time. A comprehensive sub-unit is determined, which is used to calculate the geometric average of the stress-axis effectiveness and the stress-axis contribution to obtain the support effectiveness.

[0013] Furthermore, the conduction determination unit includes: The displacement determination subunit is used to determine the displacement adjustment weight based on the weight adjustment amount and the preset displacement weight of the horizontal displacement. Axial force determination subunit is used to determine the axial force adjustment weight based on the weight adjustment amount and the preset axial force weight of the comprehensive axial force; The transmission determination subunit is used to determine the stress adjustment weight based on the displacement adjustment weight and the axial force adjustment weight, and to determine the risk transmission index by combining the weighted fusion result of the horizontal displacement, the normal stress and the comprehensive axial force.

[0014] Furthermore, the risk determination module includes: A threshold determination unit is used to determine a dynamic settlement threshold and a dynamic tilt threshold based on the ratios of a preset settlement threshold and a preset tilt threshold to the adjustment factor, respectively, wherein the adjustment factor is determined based on the risk transmission index; The building determination unit is used to determine the building risk index based on the change of the tilt angle relative to the dynamic settlement threshold and the change of the differential settlement value relative to the dynamic tilt threshold. A sliding calculation unit is used to determine the displacement-settlement correlation degree based on the Pearson correlation coefficient of the displacement deviation sequence and the settlement deviation sequence at different preset position lag times, wherein the displacement deviation sequence and the settlement deviation sequence are determined based on the instantaneous relative deviations of the horizontal displacement and the differential settlement value, respectively. The correlation determination unit is used to determine the correlation risk index based on the geometric mean of the building risk index, the risk transmission index, the temporal lag degree, and the spatial coupling degree, wherein the temporal lag degree is determined based on the displacement settlement correlation degree.

[0015] Furthermore, the building determination unit includes: The change extraction subunit is used to determine the average settlement change based on the historical average transient rate of the differential settlement value, and to determine the average angle change based on the historical average transient rate of the tilt angle. The overshoot calculation subunit is used to determine the overshoot degree based on the average settlement change and the dynamic settlement threshold, and to determine the angle overshoot degree based on the average angle change and the dynamic tilt threshold. The building is defined by a sub-unit, which is used to calculate the geometric mean of the settlement exceedance and the angle exceedance to obtain the building risk index.

[0016] Furthermore, the association determination unit includes: The lag determination subunit is used to determine the time lag degree based on the maximum correlation of the subsidence and the maximum lag time, wherein the maximum correlation of the subsidence is determined based on the displacement settlement correlation, and the maximum lag time is determined based on the maximum correlation of the subsidence and the preset subsidence lag time. The distance determination subunit is used to calculate the spatial distance between the coordinates of the preset monitoring point and the coordinates of the target building in order to obtain the risk distance; The associated determination subunit is used to determine the spatial coupling degree based on the risk distance, and to calculate the geometric mean of the building risk index, the risk transmission index, the temporal lag degree, and the spatial coupling degree to obtain the associated risk index.

[0017] Compared with existing technologies, the advantages of this invention lie in its ability to keenly capture early mechanical precursors of slope instability by analyzing the temporal changes in normal stress and horizontal displacement within the slope and identifying the force-variable synergy patterns, thus improving the predictability of early warnings. By introducing a support aging assessment anchor system to evaluate its response to stress changes and adaptively adjusting the weights of stress, displacement, and axial force in risk transmission assessment, the risk assessment model can reflect the working status and degradation process of the support system in real time. By adjusting factors, the early warning threshold for building deformation is intelligently lowered as slope risk increases, achieving dynamic and sensitive perception of risk linkage. The temporal correlation and spatial attenuation effect between slope displacement and building settlement changes are quantified, reflecting the spatiotemporal transmission characteristics of risk. The generation of a comprehensive correlation risk index using multiple parameters avoids misjudgment by a single indicator, effectively solving the problems of low early warning accuracy and difficulty in timely perception of correlated risks due to isolated static monitoring and the lack of a risk transmission mechanism.

[0018] Furthermore, by calculating the rate of change sequence of normal stress and horizontal displacement, long-term trends and low-frequency noise in the data were effectively filtered out. By calculating the Pearson correlation coefficient under different preset force-variable lag times, not only was a statistical correlation between force and deformation identified, but the characteristic timescale in which stress change precedes displacement change was also quantified, thereby capturing the physical essence of internal stress redistribution driving macroscopic deformation before slope instability. By setting dual thresholds for pattern discrimination, it was ensured that the identified force-variable co-modulation patterns had significant statistical significance and clear physical meaning, i.e., high correlation and short response delay.

[0019] Furthermore, by recording the maximum correlation coefficient and its corresponding lag time, the most significant dynamic correlation patterns and their phase characteristics between stress and deformation can be captured. By employing a dual discrimination criterion of preset sliding threshold and preset offset threshold, high correlation coefficients ensure that the identified patterns are not random noise and have clear physical meaning, while short lag times indicate that slope deformation is being rapidly driven by current stress changes, and the system is in a sensitive unstable precursor stage. This achieves the mapping of abstract time-series data into slope state indicators with clear engineering significance.

[0020] Furthermore, by analyzing the co-directional and hysteretic relationship between stress and axial force changes, the response efficiency and timeliness of the anchor system to stress fluctuations were quantified, enabling early identification of potential signs of loosening or failure in the support. By calculating the actual contribution of the support system to slope load changes and integrating it with response timeliness into a support health index, the weights of various parameters used to assess risk transmission were dynamically adjusted. Specifically, when the support system is healthy, the model emphasizes the stress signal reflecting the driving source; when support effectiveness declines, the weights of displacement signals reflecting deformation accumulation and axial force signals reflecting the support's own state are automatically increased. This allows risk assessment to closely follow the shift of dominant mechanical factors during slope instability, improving the model's robustness and accuracy under different support conditions.

[0021] Furthermore, by calculating the unidirectional ratio, the synergistic effect of the two forces can be determined from the trend. By introducing preset strain and axial force thresholds, significant stress change events and the substantial mechanical response of the anchor bolts can be captured, focusing on the load transfer process with engineering significance. By calculating the hysteresis between each significant stress change and the corresponding anchor bolt force response and taking the average value, the average delay time of the support system response is objectively quantified. By using an exponential decay model to couple the unidirectional ratio and the average hysteresis to the axial force timeliness, this model can clearly express that even if the response directions are mostly consistent, if the response is too slow, i.e., the hysteresis is large, its overall timeliness will be significantly reduced.

[0022] Furthermore, by calculating the sum of squares of the deviations of stress and axial force from their mean values, the overall intensity of slope load fluctuations and the overall amplitude of internal force changes in the support system were quantified, respectively. The load sharing ratio was calculated by taking the time-series average of the product of stress deviation and axial force deviation, and then using this ratio to the total stress change to determine the actual proportion of load sharing by the anchor system on the slope load changes. The support aging was generated by geometrically averaging the stress-axial contribution and the stress-axial aging timeliness, which characterizes the response speed. This ensures that a high support aging can only be achieved when the support system possesses both fast response and high load sharing characteristics; any significant deficiency in either aspect will significantly reduce the support aging.

[0023] Furthermore, by increasing the weight adjustment as the support effectiveness decreases, the weights of horizontal displacement and comprehensive axial force will increase synchronously and linearly, while the weight of normal stress will be compressed accordingly. This simulates the physical transformation of the risk-dominant factors during slope instability: when the support system is robust, the risk mainly stems from the stress anomaly at the driving source; when the support effectiveness declines, the risk is more directly manifested as uncontrollable deformation development and the sudden change or failure of the axial force of the support structure itself. Through this adaptive weight adjustment, the calculation model of the risk transmission index can intelligently follow the slope's instability evolution process from stress-driven to deformation support failure-dominated, enabling risk identification to dynamically focus on the most sensitive and direct risk characterization indicators.

[0024] Furthermore, by introducing an adjustment factor, the warning thresholds for building settlement and tilt are automatically lowered when slope risk transmission intensifies, achieving intelligent adjustment of the building risk benchmark according to the state of the risk source. By analyzing the correlation and lag time between the slope horizontal displacement change rate and the building differential settlement change rate, the temporal process and statistical correlation of slope deformation transmission to the building are revealed, providing an objective basis for determining the common origin of damage between the two. By generating a comprehensive correlation risk index through multi-parameter fusion, the early warning decision-making process fully incorporates risk information from various independent dimensions, while the mathematical properties of geometric mean suppress false alarms that may be caused by anomalies in a single indicator, achieving a comprehensive, balanced, and robust risk assessment.

[0025] Furthermore, by calculating the average instantaneous rate of change of differential settlement and tilt angle, the assessment focus shifts from absolute deformation to deformation rate, enabling earlier and more sensitive identification of the accelerating trend of uneven settlement and overall tilt of building foundations. By comparing the average rate of change with dynamic thresholds and calculating the exceedance, and using a max function to ensure that only the exceeding portion is quantified for risk, the risk assessment results clearly reflect both the existence of risk and its severity level. By employing a geometric mean method to integrate the exceedance of settlement and tilt, it is ensured that a significant exceedance of any single indicator will significantly increase the overall risk value, improving the comprehensive ability to capture and balance the risks of multiple potential deformation modes of buildings.

[0026] Furthermore, the statistical correlation strength between slope displacement and building settlement changes is characterized by the maximum correlation coefficient in time series, and the delay in risk transmission is measured by the maximum lag time. These two factors are then combined into a temporal lag degree, indicating that high correlation and short lag together lead to high temporal coupling, while long lag significantly weakens its effective value. Spatially, a negative exponential model is used to quantify the attenuation of risk with distance. Simultaneously, the correlation risk index is calculated through multi-source data fusion, ensuring that the comprehensive risk index not only reflects the independent states of the source and receiver but also forcibly incorporates the constraint of the spatiotemporal correlation strength between them, reducing the possibility of misassociating independent risk events with no substantial correlation.

[0027] By adopting the above technical solutions, early monitoring and early warning of instability of buildings in hilly and mountainous areas have been achieved.

[0028] Compared with existing technologies, the advantages of this invention are as follows: By analyzing the temporal changes in normal stress and horizontal displacement within the slope and identifying the force-variable synergy patterns, it can keenly capture early mechanical precursors of slope instability, improving the predictability of early warnings. By introducing the response efficiency of the support time-based assessment anchor system to stress changes, and adaptively adjusting the weights of stress, displacement, and axial force in risk transmission assessment, the risk assessment model can reflect the working status and degradation process of the support system in real time. By adjusting the factor, the early warning threshold for building deformation is intelligently lowered as the slope risk increases, achieving dynamic and sensitive perception of risk linkage. By quantifying the temporal correlation and spatial attenuation effect between slope displacement and building settlement changes, the spatiotemporal transmission characteristics of risk are reflected. By generating a comprehensive correlation risk index using multiple parameters, misjudgment by a single indicator is avoided, effectively solving the problems of low early warning accuracy and difficulty in timely perception of correlated risks due to isolated static monitoring and the lack of a risk transmission mechanism. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the building online health monitoring system in this embodiment; Figure 2 This is a logic diagram for determining the force-variable cooperative mode of the identification subunit in this embodiment; Figure 3 This is a logic diagram for determining the differential strain timing of the timing determination subunit in this embodiment. Figure 4 This is a schematic diagram of the structure of the comprehensive determination unit in this embodiment. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Please see Figure 1 As shown, this is a structural schematic diagram of the building online health monitoring system of this embodiment. This embodiment provides a building online health monitoring system, including: The acquisition module is used to acquire the tilt angle of the target building in the hilly and mountainous area, the differential settlement value of the central load-bearing column on the adjacent slope and the central load-bearing column on the back slope, the comprehensive axial force of the anchor rod on the adjacent slope, and the normal stress of the preset target point inside the slope soil and rock and the horizontal displacement of the preset monitoring point on the vertical profile. An identification module, connected to the acquisition module, is used to identify the force-variable coordination mode based on the temporal changes of the normal stress and the horizontal displacement. A transmission determination module, which is connected to the acquisition module and the identification module respectively, is used to determine the risk transmission index based on the temporal evolution characteristics and weight adjustment amount of the horizontal displacement, the normal stress and the comprehensive axial force. The weight adjustment amount is determined based on the support duration and the preset adjustment coefficient. The support duration is determined based on the temporal relative change of the normal stress and the comprehensive axial force under the force-variable synergy mode. A risk determination module, connected to both the acquisition module and the transmission determination module, is used to determine a building risk index based on the temporal changes of the differential settlement value and the tilt angle and an adjustment factor, and to determine an associated risk index based on the geometric mean of the building risk index, the risk transmission index, the temporal lag, and the spatial coupling degree. The adjustment factor is determined based on the risk transmission index, and the temporal lag and spatial coupling degree are determined based on the temporal and spatial characteristics of the horizontal displacement and the differential settlement value, respectively. An early warning module, which is connected to the risk determination module and the identification module, is used to issue an early warning based on the associated risk index and the force-change synergy mode.

[0032] In this embodiment, the building online health monitoring system is applied to common hazard points in hilly and mountainous towns where houses are built on slopes. The target building in this scenario is a brick-concrete residential building adjacent to a 15-meter-high slope. Although the slope is supported by a grid beam and anchor bolts, there is still a potential risk of landslides and even endangering the building's safety under conditions of continuous heavy rainfall. This embodiment acquires multi-source parameters of the building and its adjacent slope to form an integrated monitoring network. Among them, differential settlement refers to the difference in vertical displacement between adjacent load-bearing columns of the target building, used to determine whether the building structure has additional stress, cracking, or tilting due to foundation deformation. It can be obtained by a high-precision static leveling system installed at the base of adjacent load-bearing columns or walls. Tilt angle refers to the tilt angle of the target building as a whole or part of its components relative to the gravity plumb line or horizontal reference plane. It can be obtained by a high-precision biaxial inclinometer installed on the load-bearing structure of the building, such as columns or walls, based on MEMS or electrolyte principles. Comprehensive axial force is a characterization of the tensile or compressive force on the entire support system. Changes in these forces can reflect slope stability and the overall load level of the support system. It can be obtained by selecting the most representative key monitoring anchors in the potential sliding surface crossing area of ​​the slope, the support structure such as retaining walls, and the upper, middle, and lower rows of anchors distributed along the slope height, and installing anchor cable force gauges on them to obtain independent axial force time series data at each point. By assigning different weights to the axial forces of anchors at different locations, the comprehensive axial force is obtained by weighted summation, where W i =D max +c / Di +c, where W i It is the weight at a certain position, D i This refers to the minimum horizontal distance between the monitoring anchor and the preset potential sliding surface. This preset potential sliding surface is the most dangerous sliding surface of the slope determined during the design phase. In this embodiment, it is an arc shape, with its deepest point approximately 8 meters from the slope surface, corresponding to the position on the vertical profile. c is a preset weighting constant, set to c=1 meter in this embodiment. D max D is the one among all monitored anchor bolts. i The maximum value; normal stress refers to the compressive stress of the soil perpendicular to the plane at the preset target point, which can be obtained by burying an earth pressure cell, such as a vibrating wire or fiber optic grating type, at the target point; horizontal displacement refers to the horizontal displacement of preset fixed monitoring points at different depths in the vertical borehole of the slope, which can be obtained by burying a slewing tube with a guide groove in the borehole of the slope, fixing a series of sensor probes at fixed intervals, such as 0.5 meters or 1 meter, in the guide groove, continuously measuring the inclination angle of each depth point, and calculating the horizontal displacement of each point relative to the reference point at the bottom of the hole through geometric integration.

[0033] In this embodiment, the preset target points refer to the points pre-selected within the slope, located 0.5-1.5 meters above and approximately 0.5-1.5 meters below the preset potential sliding surface, to directly monitor the original mechanical driving force that causes the sliding. Since the soil above the sliding surface directly bears the sliding thrust, its stress exhibits a sensitive monotonic increase, serving as a leading indicator of instability. Therefore, the normal stress above the sliding surface is a representative input for calculation by each module. The soil below is in a complex state of passive compression or even tension, resulting in a slow and unstable stress signal. Therefore, the normal stress above the sliding surface is used for auxiliary trend judgment and self-verification of the monitoring system. The preset monitoring points are the stable strata at the sensor installation depth, set at fixed intervals within the vertical boreholes of the slope, extending downwards from the slope surface to 20 meters. Together, they constitute a three-dimensional sensing network from the source of mechanical driving force to the deformation response result.

[0034] In this embodiment, the warning module ultimately outputs a three-level linkage warning signal: when the comprehensive risk assessment value is lower than the first threshold (set to 0.65 in this embodiment), the system determines it to be in a safe state and does not trigger a warning; when the assessment value is between the first and second thresholds, the system issues a yellow warning at the attention level, prompting relevant personnel to increase monitoring frequency and conduct preliminary hazard investigation; when the assessment value exceeds the second threshold (set to 0.85 in this embodiment), the system immediately triggers a red warning at the danger level, indicating that slope instability has posed a significant threat to nearby buildings and that an emergency response mechanism needs to be activated, including but not limited to personnel evacuation, traffic control, and engineering emergency measures. This warning result is simultaneously pushed to management departments and on-site personnel through the platform interface, SMS, and audible and visual alarm devices to ensure timely and accurate transmission of risk information.

[0035] The preset adjustment coefficient is a dimensionless proportional coefficient used to adjust the degree of influence of support duration on the weight adjustment amount. It depends on the specific slope geological conditions, support design safety factor and system early warning sensitivity requirements. It is usually set between 0.1 and 0.5. In this embodiment, it is set to 0.3, which can be evaluated and quantified as the weight adjustment range, so as to realize intelligent correction of the risk model based on the health status of the defense line.

[0036] By analyzing the temporal changes in normal stress and horizontal displacement within the slope and identifying the force-variance synergy patterns, early mechanical precursors of slope instability can be keenly detected, improving the predictability of early warnings. By introducing a support aging assessment to evaluate the response efficiency of the anchor system to stress changes, and adaptively adjusting the weights of stress, displacement, and axial force in risk transmission assessment, the risk assessment model can reflect the working status and degradation process of the support system in real time. Adjustment factors intelligently lower the warning threshold for building deformation as slope risk increases, achieving dynamic and sensitive perception of risk linkage. Quantifying the temporal correlation and spatial attenuation effect between slope displacement and building settlement changes reflects the spatiotemporal transmission characteristics of risk. Generating a comprehensive correlation risk index using multiple parameters avoids misjudgment based on a single indicator, effectively solving the problems of low early warning accuracy and difficulty in timely detection of correlated risks due to isolated static monitoring and the lack of a risk transmission mechanism.

[0037] Specifically, the identification module includes: The change calculation unit is used to calculate the relative deviation of the normal stress and the horizontal displacement at any two adjacent moments based on the previous preset identification time, so as to obtain the stress change rate and the displacement change rate, and construct the stress change sequence and the displacement change sequence according to the stress change rate and the displacement change rate respectively. A related determination unit, which is connected to the change calculation unit, is used to calculate the Pearson correlation coefficients of the stress change sequence and the displacement change sequence under different preset force hysteresis times, so as to obtain several sliding force correlation degrees. An identification unit, connected to the correlation determination unit, is used to identify the force-variable cooperative mode based on the threshold comparison results of the sliding correlation maximum and minimum values ​​and the sliding offset maximum and minimum values, wherein the sliding correlation maximum and minimum values ​​are determined based on all the sliding force-variable correlations, and the sliding offset maximum and minimum values ​​are determined based on the sliding correlation maximum and minimum values.

[0038] In this embodiment, the maximum value of the horizontal displacement of all preset monitoring points at each time point of the preset identification time is taken as the horizontal displacement at that moment of the preset identification time. In this way, the displacement and stress data scattered at different monitoring points can be integrated to reflect the maximum horizontal displacement and maximum normal stress that the building structure bears at that moment as a whole.

[0039] The preset identification duration is the length of the forward-looking data time window on which the calculation of stress and displacement change rates is based. It depends on the sampling frequency of the monitoring data and the typical development stage of the slope deformation process, and is usually set between 24 hours and 7 days. In this embodiment, it can be set to 72 hours, which can provide a time-series data basis for subsequent correlation analysis. The preset force hysteresis duration is the set of time offsets set when calculating the Pearson correlation coefficient between the stress change sequence and the displacement change sequence. It depends on the stress transmission rate and deformation response characteristics of the slope soil and rock medium, and is usually set between 0.5 hours and 12 hours. In this embodiment, a set of offsets [0 hours, 0.5 hours, 1 hour, 2 hours, 4 hours, 6 hours, 8 hours, 12 hours] is preset for sliding calculation, which can systematically quantify the delay effect of stress change driving deformation response.

[0040] By calculating the rate of change sequences of normal stress and horizontal displacement, long-term trends and low-frequency noise in the data were effectively filtered out. By calculating the Pearson correlation coefficient under different preset force-variable lag times, not only was a statistical correlation between force and deformation identified, but the characteristic timescale in which stress change precedes displacement change was quantified. This allowed the capture of the physical essence of internal stress redistribution driving macroscopic deformation before slope instability. By setting dual thresholds for pattern discrimination, it was ensured that the identified force-variable co-modulation patterns possessed significant statistical meaning and clear physical connotations—that is, high correlation and short response delay.

[0041] Please see Figure 2 As shown, this is the logic diagram for determining the force-variable cooperative mode of the identification subunit in this embodiment. In this embodiment, the identification unit includes: A recording subunit is used to record the maximum value of all the sliding force correlations as the sliding correlation maximum value, and to record the preset force hysteresis time corresponding to the sliding correlation maximum value as the sliding offset maximum value; An identification subunit, connected to the recording subunit, is used to identify the force-variable cooperative mode when the maximum value of the sliding correlation is greater than a preset sliding threshold and the maximum value of the sliding offset is less than or equal to a preset offset threshold. Otherwise, it will be identified as a non-cooperative mode.

[0042] The preset sliding threshold is a standard value for judging whether the stress change sequence and the displacement change sequence have a significant statistical correlation. It depends on the early warning system's tolerance for false alarm rate and the required statistical confidence level. It is usually set between 0.6 and 0.9. In this embodiment, it can be set to 0.75, which can effectively screen out force-change correlations with clear statistical significance. The preset offset threshold is a standard value for judging whether the response speed of stress change driving displacement deformation is instantaneous. It depends on the rheological characteristics of the slope soil and rock and the early warning system's sensitivity requirements for response delay. It is usually set between 2 hours and 8 hours. In this embodiment, it can be set to 4 hours, which can ensure the identification of high-risk synergistic states with rapid response and direct mechanism.

[0043] By recording the maximum correlation coefficient and its corresponding lag time, the most significant dynamic correlation patterns and phase characteristics between stress and deformation can be captured. Using a dual discrimination criterion of preset sliding threshold and preset offset threshold, high correlation coefficients ensure that the identified patterns are not random noise and have clear physical meaning. Short lag times indicate that slope deformation is rapidly driven by current stress changes, and the system is in a sensitive, pre-unstable stage. This achieves the mapping of abstract time-series data into slope state indicators with clear engineering significance.

[0044] Specifically, the conduction determination module includes: An aging determination unit is used to determine the aging degree of the axial force based on the same-direction ratio and the average hysteresis, wherein the same-direction ratio and the average hysteresis are determined based on the coordinated change of the normal stress and the combined axial force; A comprehensive determination unit, connected to the aging determination unit, is used to determine the support aging based on the geometric average of the axial aging degree and the axial contribution degree, wherein the axial contribution degree is determined based on the coordinated variation trend of the normal stress and the comprehensive axial force; A transmission determination unit, connected to the comprehensive determination unit, is used to determine the risk transmission index based on the horizontal displacement, the normal stress, the comprehensive axial force, and their respective adjustment weights. The respective adjustment weights are determined based on the weight adjustment amount, which is determined based on the coupling characteristics of the support duration and the preset adjustment coefficient. Here, Q = μ × (1 - Z), where Q is the weight adjustment amount, μ is the preset adjustment coefficient, and Z is the support duration. By analyzing the co-directional and hysteretic relationship between stress and axial force changes, the response efficiency and timeliness of the anchor system to stress fluctuations were quantified, enabling early identification of potential signs of loosening or failure in the support. The actual contribution of the support system to slope load changes was calculated and integrated with response timeliness into a support health index. The weights of various parameters used to assess risk transmission were dynamically adjusted: when the support system is healthy, the model emphasizes the stress signal of the driving source; when support effectiveness declines, the weights of displacement signals reflecting deformation accumulation and axial force signals reflecting the support's own state are automatically increased. This allows risk assessment to closely follow the shift of dominant mechanical factors during slope instability, improving the model's robustness and accuracy under different support conditions.

[0045] Please see Figure 3 As shown, this is the logic diagram for determining the differential strain time by the timing determination subunit in this embodiment. In this embodiment, the timing determination unit includes: A co-directional determination subunit is used to calculate the proportion of the number of differential co-directional moments to the total number of moments in the preset identification time, so as to obtain the co-directional proportion. The differential co-directional moments are determined based on the co-directional moments of stress difference and axial force difference. The stress difference and axial force difference are determined based on the first-order differences of the normal stress and the comprehensive axial force within the preset identification time, respectively. A time-determining subunit, connected to the same-direction-determining subunit, is used to calculate the time interval between the differential strain time and the axial force same-direction time to obtain the hysteresis. The differential strain time is determined based on the differential same-direction time corresponding to the stress difference being greater than a preset strain threshold, and the axial force same-direction time is determined based on the differential strain time and the first differential same-direction time corresponding to the axial force difference being greater than a preset axial force threshold after it. A timeliness determination subunit, connected to both the direction-of-motion determination subunit and the timing determination unit, is used to calculate the average value of all the lags to obtain the average lag, and to determine the response axis timeliness based on the coupling characteristics of the average lag and the direction-of-motion ratio, where Y = T × e (-λ×H） Where Y is the aging time of the axial force, T is the same-direction ratio, λ is the preset axial force attenuation coefficient, and H is the average hysteresis.

[0046] The preset strain threshold is an incremental threshold value used to filter out significant changes in normal stress. It depends on the measurement accuracy of the earth pressure cell and the sensitivity of the slope to small stress fluctuations, and is usually set between 1 kPa and 5 kPa. In this embodiment, it is set to 2 kPa, which can effectively filter out small stress fluctuations caused by sensor noise or soil creep. The preset axial force threshold is an incremental threshold value used to determine whether the axial force of the anchor bolt has a substantial response. It depends on the design bearing capacity of the anchor bolt, the initial prestress level, and the noise level of the monitoring system. It is usually set between 5% and 15% of the design bearing capacity of a single anchor bolt. In this embodiment, it is set to 8% of the design bearing capacity of a single anchor bolt, which can effectively eliminate interference from axial force fluctuations caused by temperature changes or small slippage. The preset axial force attenuation coefficient is an attenuation constant that controls the rate of negative impact of the average hysteresis on the axial timeliness. It depends on the sensitivity and tolerance of the early warning system to the support response delay, and is usually set between 0.1 and 0.5. In this embodiment, it is set to 0.2, which can reasonably quantify the weakening effect of the response delay on the overall timeliness of the support system using an exponential attenuation model.

[0047] In this embodiment, the formula for calculating the time-sensitive validity is Y=T×e. (-λ×H） It is a composite attenuation model that integrates trend synergy and response timeliness. The basic synergy reflects the overall synergistic trend between the support system and slope mechanical changes; the velocity penalty factor, e... (-λ×H） It is an exponential decay term with the natural constant as the base, where the average hysteresis represents the average delay time of the anchor bolt response stress change. The preset decay coefficient controls the degree of weakening of timeliness due to the delay. The axial timeliness is the product of these two factors. Its underlying logic is that an ideal support system should not only have coordinated trends (high T value) but also rapid response (low H value). The multiplicative structure reflects a strict "AND" logic, meaning that a serious defect in any aspect will lead to a significant reduction in overall timeliness. The exponential decay term e (-λ×H） This indicates that when the lag time is very short, the value of this item is close to 1, which hardly causes any loss to the degree of coordination; however, as the average lag time increases, the value of this item will decrease sharply, thus producing an amplified penalty effect on the timeliness. This reflects that in the scenario of slope instability early warning, slow response is often more dangerous than asynchronous trend, because the delayed support force may not be able to stop the accelerated development of deformation.

[0048] By calculating the unidirectional ratio, the synergistic effect of the two forces can be determined from the trend. By introducing preset strain and axial force thresholds, significant stress change events and the resulting substantial mechanical responses of the anchor bolts can be captured, focusing on the load transfer process with engineering significance. By calculating the hysteresis between each significant stress change and the corresponding anchor bolt force response and taking the average value, the average delay time of the support system response is objectively quantified. By using an exponential decay model to couple the unidirectional ratio and the average hysteresis to the axial force timeliness, this model can clearly express that even if the response directions are mostly consistent, if the response is too slow (i.e., the hysteresis is large), its overall timeliness will be significantly reduced.

[0049] Please see Figure 4 As shown, this is a schematic diagram of the structure of the comprehensive determination unit in this embodiment. In this embodiment, the comprehensive determination unit includes: The total amount determination subunit is used to determine the total stress change based on the sum of the squares of all stress deviations and the total axial force change based on the sum of the squares of all axial force deviations. The stress deviation is determined based on the relative mean deviation of the normal stress at each time point of the preset identification time, and the axial force deviation is determined based on the relative mean deviation of the comprehensive axial force at each time point of the preset identification time. A collaborative determination subunit, connected to the total determination subunit, is used to determine the axial force contribution based on the ratio of the load sharing degree to the preset sharing degree, wherein the load sharing degree is determined based on the ratio of the collaborative change amount to the total stress change amount, and the collaborative change amount is determined based on the ratio of the sum of the products of the stress deviation and the axial force deviation at each moment within the preset identification time to the total number of moments in the preset identification time. A comprehensive determination subunit, connected to the collaborative determination subunit, is used to calculate the geometric average of the stress-axis effectiveness and the stress-axis contribution to obtain the support effectiveness.

[0050] The preset load-sharing ratio is a benchmark ratio used to evaluate the load-sharing effectiveness of the support system. It depends on the design safety factor of the slope support, the density of the anchor bolts, and the design bearing capacity. It is usually set between 0.5 and 0.8. In this embodiment, it can be set to 0.65, which can quantify the gap between the actual effectiveness of the support system and the design expectation.

[0051] By calculating the sum of squares of the deviations of stress and axial force from their mean values, the overall intensity of slope load fluctuations and the overall amplitude of internal force changes in the support system were quantified, respectively. The load sharing ratio was calculated by taking the time-series average of the product of stress deviation and axial force deviation, and then using this ratio to the total stress change to determine the actual proportion of load sharing by the anchor system on the slope load changes. The support aging was generated by geometrically averaging the stress-axial contribution and the stress-axial aging timeliness, which characterizes the response speed. This ensures that a high support aging can only be achieved when the support system possesses both fast response and high load sharing characteristics; any significant deficiency in either aspect will significantly reduce the support aging.

[0052] Specifically, the conduction determination unit includes: A displacement determination subunit is used to determine the displacement adjustment weight based on the weight adjustment amount and the preset displacement weight of the horizontal displacement, wherein A w =A w0 +Q, where A w It is the displacement adjustment weight, A w0 It is a preset displacement weight; Axial force determination subunit is used to determine the axial force adjustment weight based on the weight adjustment amount and the preset axial force weight of the comprehensive axial force, wherein A Z =A Z0 +Q, where A Z It is the axial force adjustment weight, A Z0 It is a preset axial force weight; The transmission determination subunit is used to determine the stress adjustment weight based on the displacement adjustment weight and the axial force adjustment weight, where AY = 1 - Aw - AZ, and AY is the stress adjustment weight. It also determines the risk transmission index by combining the weighted fusion result of the horizontal displacement, the normal stress, and the comprehensive axial force, where M = A w ×P+A Z ×L+A Y ×G, where M is the risk transmission index, P is the horizontal displacement, L is the combined axial force, and G is the normal stress.

[0053] The preset displacement weight is the initial weight coefficient assigned to the horizontal displacement index in the risk transmission index calculation when the support system is in an intact state. It depends on the reliability of slope deformation monitoring and its early warning importance in a stable state, and is usually set between 0.2 and 0.4. In this embodiment, it can be set to 0.25, which can reasonably reflect the role of deformation as a long-term or secondary risk indicator when the support is effective. The preset axial force weight is the initial weight coefficient assigned to the comprehensive axial force index in the risk transmission index calculation when the support system is in an intact state. It depends on the directness and importance of anchor force monitoring in representing the overall state of the support system, and is usually set between 0.2 and 0.4. In this embodiment, it can be set to 0.35, which can highlight the core contribution of the internal force state of the support structure itself to risk assessment when the support is effective.

[0054] By increasing the weight adjustment as the support effectiveness decreases, the weights of horizontal displacement and comprehensive axial force will increase synchronously and linearly, while the weight of normal stress will be compressed accordingly. This simulates the physical transformation of the risk-dominant factors during slope instability: when the support system is robust, the risk mainly stems from the stress anomaly at the driving source; when the support effectiveness declines, the risk is more directly manifested as uncontrollable deformation development and the sudden change or failure of the axial force of the support structure itself. Through this adaptive weight adjustment, the calculation model of the risk transmission index can intelligently follow the slope's instability evolution process from stress-driven to deformation support failure-dominated, enabling risk identification to dynamically focus on the most sensitive and direct risk characterization indicators.

[0055] Specifically, the risk determination module includes: A threshold determination unit is used to calculate the ratios of a preset settlement threshold and a preset tilt threshold to the adjustment factor, respectively, to obtain the dynamic settlement threshold and the dynamic tilt threshold. The adjustment factor is determined based on the ratio of the risk transmission index to the preset transmission threshold, where K = 1 + η × (N / N0), where K is the adjustment factor, η is the preset sensitivity coefficient, N is the risk transmission index, and N0 is the preset transmission threshold. The building determination unit, which is connected to the threshold determination unit, is used to determine the building risk index based on the change of the tilt angle relative to the dynamic settlement threshold and the change of the differential settlement value relative to the dynamic tilt threshold. The sliding calculation unit is used to calculate the relative deviations of the horizontal displacement and the differential settlement value at adjacent times within a preset time period, so as to obtain the displacement deviation sequence and the settlement deviation sequence respectively, and to calculate the Pearson correlation coefficient of the displacement deviation sequence and the settlement deviation sequence under different preset lag times, so as to obtain several displacement-settlement correlations. The association determination unit is connected to the building determination unit and the sliding calculation unit respectively, and is used to determine the association risk index based on the geometric mean of the building risk index, the risk transmission index, the temporal lag degree and the spatial coupling degree, wherein the temporal lag degree is determined based on the displacement settlement correlation degree.

[0056] The preset settlement threshold is the basic threshold for judging whether the differential settlement of a building is abnormal. It depends on the building structure type, foundation type, and allowable deformation standard, and is usually set between 3mm and 10mm. In this embodiment, it can be set to 5mm, which can provide an initial, static evaluation benchmark for building settlement safety. The preset tilt threshold is the basic threshold for judging whether the tilt of a building is abnormal. It depends on the building structure type, height, and relevant code allowable values, and is usually set between 0.001rad and 0.003rad. In this embodiment, it can be set to 0.002rad, which can provide an initial, static evaluation benchmark for building tilt safety. The preset transmission threshold is the benchmark value used to normalize the risk transmission index in the calculation of the adjustment factor. It depends on the historical statistical value of the slope risk level or the design safety threshold, and is usually set between 0.5 and 1.0. In this embodiment, it can be set to 0.8, which can normalize the risk transmission index and provide a reasonable scale for the calculation of the adjustment factor. The preset sensitivity coefficient is the coefficient used to control the intensity of the influence of the risk transmission index in the calculation of the adjustment factor. The threshold value depends on the system's sensitivity to slope risk transmission and is typically set between 0.3 and 0.7. In this embodiment, it is set to 0.5 to balance the variation of the adjustment factor and avoid over- or under-adjustment of the building threshold. The preset determination time is the length of the time window on which the relative deviations of horizontal displacement and differential settlement values ​​are calculated. It depends on the sampling frequency of the monitoring data and the building's response cycle to slope deformation and is typically set between 24 hours and 7 days. In this embodiment, it can be set to 48 hours to provide sufficient data points for calculating the rate of change of displacement and settlement. The preset hysteresis time is the set of time offsets set when calculating the Pearson correlation coefficient between the displacement deviation sequence and the settlement deviation sequence. It depends on the time scale of slope deformation transmission to the building foundation and is typically set between 0 hours and 24 hours. In this embodiment, an offset set of [0 hours, 2 hours, 4 hours, 6 hours, 8 hours, 12 hours, 18 hours, 24 hours] is set to systematically quantify the delay time in which slope deformation precedes building settlement.

[0057] By introducing adjustment factors, the warning thresholds for building settlement and tilt are automatically lowered when slope risk transmission intensifies, achieving intelligent adjustment of the building risk benchmark according to the state of the risk source. Analysis of the correlation and lag time between the slope horizontal displacement change rate and the building differential settlement change rate reveals the temporal process and statistical correlation of slope deformation transmission to buildings, providing an objective basis for determining the common origin of damage between the two. The generation of a comprehensive correlation risk index through multi-parameter fusion ensures that early warning decisions fully incorporate risk information from various independent dimensions, while the mathematical properties of geometric mean suppress false alarms that may result from anomalies in a single indicator, achieving a comprehensive, balanced, and robust risk assessment.

[0058] Specifically, the building determination unit includes: The change extraction subunit is used to extract the average value of the instantaneous change rate of the differential settlement value at each time based on the previous preset extraction time, so as to obtain the average value of settlement change, and to extract the average value of the instantaneous change rate of the tilt angle at each time, so as to obtain the average value of angle change. An exceedance calculation subunit, connected to the change extraction subunit, is used to calculate the ratio of the difference between the mean settlement change and the dynamic settlement threshold to the dynamic settlement threshold to determine the settlement exceedance degree, where E c =max[0,(B c -B c0 ) / B c0 ], where E c It is the degree of settlement exceedance, B c It is the average settlement change, B c0 It is a dynamic settlement threshold, and the angle exceedance is determined by calculating the ratio of the difference between the mean angle change and the dynamic tilt threshold to the dynamic tilt threshold, where E J =max[0,(B J -B J0 ) / B J0 ], where E J It is the degree of angular transcendence, B J It is the average value of the angle change, B J0 It is a dynamic tilt threshold; The building determination subunit is connected to the exceedance calculation subunit to calculate the geometric mean of the settlement exceedance and the angle exceedance to obtain the building risk index.

[0059] The preset extraction duration is the length of the forward-looking data time window on which the average value of building settlement and tilt change rate is based. It depends on the sampling frequency of the monitoring data and the short-term fluctuation characteristics of building deformation. It is usually set between 12 hours and 24 hours. In this embodiment, it is set to 24 hours, which can provide a stable average value reflecting the recent deformation trend for calculating the building risk index.

[0060] By calculating the average instantaneous rate of change of differential settlement and tilt angle, the assessment focus shifts from absolute deformation to deformation rate, enabling earlier and more sensitive identification of the accelerating trend of uneven settlement and overall tilt of building foundations. By comparing the average rate of change with dynamic thresholds and calculating the exceedance level, and using a max function to ensure that only the exceeding portion is quantified, the risk assessment results clearly reflect both the existence of risk and its precise severity level. By employing a geometric mean method to integrate the exceedance levels of settlement and tilt, it ensures that a significant exceedance of any single indicator will significantly increase the overall risk value, improving the comprehensive ability to capture and balance the risks of multiple potential deformation modes in buildings.

[0061] Specifically, the association determination unit includes: The lag determination subunit is used to determine the time series lag based on the maximum correlation of the sinking and the maximum lag time, where U = max(0, ρmax) × e -γ×r Wherein, U is the time lag degree, ρmax is the maximum correlation of subsidence, γ is the preset lag attenuation coefficient, and r is the maximum lag time. The maximum correlation of subsidence is determined based on the maximum value of all the displacement settlement correlations, and the maximum lag time is determined based on the preset subsidence lag time corresponding to the maximum correlation of subsidence. The distance determination subunit is used to obtain the coordinates of the preset monitoring point and the coordinates of the target building, respectively, so as to obtain the risk source coordinates and the risk receptor coordinates, and calculate the spatial distance between the risk source coordinates and the risk receptor coordinates to obtain the risk distance; An association determination subunit, which is connected to both the hysteresis determination subunit and the distance determination subunit, is used to determine the spatial coupling degree based on the risk distance, wherein...

[0062] Among them, J K The spatial coupling degree is f, the preset coupling coefficient is D0, and the risk distance is D0. The geometric mean of the building risk index, the risk transmission index, the temporal lag degree, and the spatial coupling degree is calculated to obtain the associated risk index.

[0063] The preset hysteresis attenuation coefficient is an attenuation constant that controls the rate attenuation of the negative impact of the maximum hysteresis time on the time-series hysteresis. It depends on the sensitivity and tolerance of the early warning system to the risk transmission delay and is usually set between 0.05 and 0.2. In this embodiment, it is set to 0.1, which allows the calculated time-series hysteresis to more accurately reflect the timeliness of risk transmission. The preset coupling coefficient is an influence coefficient that controls the attenuation rate of risk coupling strength on the spatial distance. It depends on the transmission and attenuation characteristics of deformation or stress waves by the foundation medium and is usually set between 0.01 and 0.1. In this embodiment, it is set to 0.03, which can quantify the attenuation law of risk with spatial distance through a negative exponential model.

[0064] In this embodiment, the formula for calculating spatial coupling degree is... This is a mathematical model based on exponential decay, containing two core parameters: the risk distance, i.e., the actual straight-line spatial distance between the slope risk source and the building risk receptor, and a preset coupling coefficient closely related to the physical properties of the foundation medium, used to quantify the rate attenuation of risk with distance in a specific soil and rock mass. This formula originates from the simulation of the physical essence of engineering risk transmission; that is, when the mechanical effects of slope instability, such as stress redistribution and deformation transmission, propagate in the soil and rock medium, their intensity typically decreases nonlinearly with increasing propagation distance, rather than simply weakening linearly. The exponential decay model is used to... It depicts the general pattern that the impact is severe when it is near and decreases sharply when it is far away. The preset coupling coefficient acts as a medium filter. This formula transforms the abstract spatial relationship into a continuous quantitative value between 0 and 1, so that the risk assessment model can objectively reflect the geographical constraints of risk decay with distance.

[0065] This study characterizes the statistical correlation strength between slope displacement and building settlement changes using the maximum correlation coefficient over time, and measures the delay in risk transmission using the maximum lag time. These two factors are then combined into a temporal lag value, indicating that high correlation and short lags together lead to high temporal coupling, while long lags significantly weaken its effective value. Spatially, a negative exponential model is used to quantify the attenuation of risk with distance. Furthermore, the correlation risk index is calculated through multi-source data fusion, ensuring that the comprehensive risk index not only reflects the independent states of the source and receiver but also forcibly incorporates the constraint of the spatiotemporal correlation strength between them, reducing the possibility of misassociating independent risk events with no substantial connection.

[0066] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A building online health monitoring system, characterized in that, include: The acquisition module is used to acquire the tilt angle of the target building in the hilly and mountainous area, the differential settlement value of the central load-bearing column on the adjacent slope and the central load-bearing column on the back slope, the comprehensive axial force of the anchor rod on the adjacent slope, and the normal stress of the preset target point inside the slope soil and rock and the horizontal displacement of the preset monitoring point on the vertical profile. The identification module is used to identify the force-variable coordination mode based on the temporal changes of the normal stress and the horizontal displacement; The transmission determination module is used to determine the risk transmission index based on the temporal evolution characteristics and weight adjustment amount of the horizontal displacement, the normal stress and the comprehensive axial force. The weight adjustment amount is determined based on the support duration and the preset adjustment coefficient. The support duration is determined based on the temporal relative change of the normal stress and the comprehensive axial force under the force-variable synergy mode. The risk determination module is used to determine the building risk index based on the temporal changes of the differential settlement value and the tilt angle and the adjustment factor, and to determine the associated risk index based on the geometric mean of the building risk index, the risk transmission index, the temporal lag degree and the spatial coupling degree. The adjustment factor is determined based on the risk transmission index, and the temporal lag degree and the spatial coupling degree are determined based on the temporal characteristics and spatial characteristics of the horizontal displacement and the differential settlement value, respectively. The early warning module is used to issue early warnings based on the associated risk index and the force-change synergy mode.

2. The building online health monitoring system according to claim 1, characterized in that, The identification module includes: The change calculation unit is used to construct stress change sequence and displacement change sequence respectively based on the instantaneous change rate of the normal stress and the horizontal displacement within a preset identification time period; The relevant determination unit is used to calculate the Pearson correlation coefficient of the stress change sequence and the displacement change sequence under different preset force hysteresis times, so as to obtain several sliding force correlation degrees. The identification unit is used to identify the force-variable cooperative mode based on all the sliding force-variable correlations.

3. The building online health monitoring system according to claim 2, characterized in that, The identification unit includes: The recording subunit is used to record the maximum value of all the sliding force correlations as the sliding correlation maximum value, and to record the preset force hysteresis time corresponding to the sliding correlation maximum value as the sliding offset maximum value; The identification subunit is used to identify the force-variable cooperative mode based on the threshold comparison results of the sliding correlation extrema and the sliding offset extrema, respectively.

4. The building online health monitoring system according to claim 3, characterized in that, The conduction determination module includes: An aging determination unit is used to determine the aging degree of the axial force based on the same-direction ratio and the average hysteresis, wherein the same-direction ratio and the average hysteresis are determined based on the coordinated change of the normal stress and the combined axial force; A comprehensive determination unit is used to determine the support aging based on the coordinated changes of the normal stress and the comprehensive axial force, as well as the axial aging aging degree. The transmission determination unit is used to determine the risk transmission index based on the horizontal displacement, the normal stress, the comprehensive axial force, and their respective adjustment weights, wherein the respective adjustment weights are determined based on the support aging period, the preset adjustment coefficient, and the weight adjustment amount.

5. The building online health monitoring system according to claim 4, characterized in that, The timeliness determination unit includes: A co-directional determination subunit is used to determine the co-directional ratio based on the ratio of differential co-directional moments within the preset identification time, wherein the differential co-directional moments are determined based on the co-directional moments of stress difference and axial force difference, and the stress difference and axial force difference are determined based on the first-order differences of the normal stress and the comprehensive axial force, respectively. A time-determining sub-unit is used to determine the hysteresis based on the time interval between the differential strain time and the axial force co-directional time, wherein the differential strain time is determined based on the stress difference and the differential co-directional time, and the axial force co-directional time is determined based on the differential strain time, the axial force difference, and the differential co-directional time. A timeliness determination subunit is used to determine the timeliness of the applied axis based on the average lag and the same-direction ratio, wherein the average lag is determined based on the average value of all the lags.

6. The building online health monitoring system according to claim 5, characterized in that, The comprehensive determination unit includes: The total amount determination subunit is used to determine the total stress change and the total axial force change based on the sum of the squares of all stress deviations and the sum of the squares of all axial force deviations, respectively. The stress deviation and axial force deviation are determined based on the relative mean deviations of the normal stress and the comprehensive axial force, respectively. A collaborative determination subunit is used to determine the axial force contribution based on the ratio of the load sharing degree to the preset sharing degree, wherein the load sharing degree is determined based on the collaborative change amount and the total stress change, and the collaborative change amount is determined based on the stress deviation, the axial force deviation and the preset identification time. A comprehensive sub-unit is determined, which is used to calculate the geometric average of the stress-axis effectiveness and the stress-axis contribution to obtain the support effectiveness.

7. The building online health monitoring system according to claim 6, characterized in that, The conduction determination unit includes: The displacement determination subunit is used to determine the displacement adjustment weight based on the weight adjustment amount and the preset displacement weight of the horizontal displacement. Axial force determination subunit is used to determine the axial force adjustment weight based on the weight adjustment amount and the preset axial force weight of the comprehensive axial force; The transmission determination subunit is used to determine the stress adjustment weight based on the displacement adjustment weight and the axial force adjustment weight, and to determine the risk transmission index by combining the weighted fusion result of the horizontal displacement, the normal stress and the comprehensive axial force.

8. The building online health monitoring system according to claim 7, characterized in that, The risk determination module includes: A threshold determination unit is used to determine a dynamic settlement threshold and a dynamic tilt threshold based on the ratios of a preset settlement threshold and a preset tilt threshold to the adjustment factor, respectively, wherein the adjustment factor is determined based on the risk transmission index; The building determination unit is used to determine the building risk index based on the change of the tilt angle relative to the dynamic settlement threshold and the change of the differential settlement value relative to the dynamic tilt threshold. A sliding calculation unit is used to determine the displacement-settlement correlation degree based on the Pearson correlation coefficient of the displacement deviation sequence and the settlement deviation sequence at different preset position lag times, wherein the displacement deviation sequence and the settlement deviation sequence are determined based on the instantaneous relative deviations of the horizontal displacement and the differential settlement value, respectively. The correlation determination unit is used to determine the correlation risk index based on the geometric mean of the building risk index, the risk transmission index, the temporal lag degree, and the spatial coupling degree, wherein the temporal lag degree is determined based on the displacement settlement correlation degree.

9. The building online health monitoring system according to claim 8, characterized in that, The building determination unit includes: The change extraction subunit is used to determine the average settlement change based on the historical average transient rate of the differential settlement value, and to determine the average angle change based on the historical average transient rate of the tilt angle. The overshoot calculation subunit is used to determine the overshoot degree based on the average settlement change and the dynamic settlement threshold, and to determine the angle overshoot degree based on the average angle change and the dynamic tilt threshold. The building is defined by a sub-unit, which is used to calculate the geometric mean of the settlement exceedance and the angle exceedance to obtain the building risk index.

10. The building online health monitoring system according to claim 9, characterized in that, The association determination unit includes: The lag determination subunit is used to determine the time lag degree based on the maximum correlation of the subsidence and the maximum lag time, wherein the maximum correlation of the subsidence is determined based on the displacement settlement correlation, and the maximum lag time is determined based on the maximum correlation of the subsidence and the preset subsidence lag time. The distance determination subunit is used to calculate the spatial distance between the coordinates of the preset monitoring point and the coordinates of the target building in order to obtain the risk distance; The associated determination subunit is used to determine the spatial coupling degree based on the risk distance, and to calculate the geometric mean of the building risk index, the risk transmission index, the temporal lag degree, and the spatial coupling degree to obtain the associated risk index.

Images