Geotechnical engineering monitoring and early warning method and system

Abstract

The invention discloses a geotechnical engineering monitoring and early warning method and system, and relates to the field of geotechnical engineering, and the method comprises a construction module which is used for constructing an initial reference model for geotechnical engineering deformation instability monitoring according to geotechnical engineering design parameters and an initial structure state, extracting reference parameters, and storing the reference parameters to create a reference database; the sensing module is used for deploying a distributed sensing unit in a geotechnical engineering monitoring area, collecting real-time state data of a geotechnical structure, comparing the real-time state data with the reference model parameters in real time and outputting structure state deviation data; according to the method, real-time data is collected through a distributed hybrid sensing architecture, structural state deviation is accurately recognized by combining space-time coupling deviation calculation, multi-domain deviation characteristics are extracted, structural weakness is quantified, weak part orientation is defined, environment dynamic parameters and operation loads are fused to construct a risk coupling model, and the risk coupling accuracy is improved. And quantifying an instability risk coupling degree by using a dynamic index.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geotechnical engineering, in particular to a geotechnical engineering monitoring and early warning method and system. BACKGROUND

[0002] Geotechnical engineering safety monitoring and early warning is the core work content in engineering construction and operation management, and the stability of geotechnical engineering structures such as slopes, foundation pits, and foundations directly relates to the safety of the project and the surrounding environment. With the continuous expansion of China's infrastructure construction scale, the demand for geotechnical engineering disaster prevention and control is increasingly urgent, and the implementation of whole-process, intelligent monitoring and early warning is of great significance for disaster prevention and reduction.

[0003] The existing geotechnical engineering monitoring and early warning technology mainly uses sensor networks to collect displacement, strain, crack, and other monitoring data, and judges the safety state of the structure by setting thresholds or simple data comparison. Some monitoring systems introduce numerical simulation, statistical analysis, and other methods to assist risk assessment, which to some extent improves the monitoring accuracy. However, the stress mechanism of geotechnical engineering structures is complex, the monitoring data presents significant spatio-temporal coupling characteristics, and is dynamically affected by multiple factors such as rainfall, groundwater, and ground load. The single parameter or fixed threshold early warning method cannot fully reflect the structure instability risk, resulting in insufficient early warning accuracy, and false positives and false negatives often occur.

[0004] Chinese patent application CN112818567A discloses a geotechnical engineering intelligent monitoring and early warning method based on probability theory. The method obtains the coordinates of the monitoring points and the deep displacement monitoring data of the soil body, judges the trend structure of the spatial distribution of the deep displacement, uses the maximum likelihood estimation method combined with the Matérn equation to perform autocorrelation judgment on the deep displacement data, performs Kriging interpolation prediction on the unmonitored area according to the coordinates of the monitoring points and the unmonitored points, solves the prediction error, and then models the conditional random field of the unmonitored point displacement and performs Monte Carlo simulation, calculates the hazard warning occurrence probability, and generates a real-time warning signal. This method predicts the displacement of the unmonitored area by unbiased interpolation estimation of real-time data, and includes the prediction error in the early warning index calculation, which expands the early warning monitoring area and reduces the monitoring cost. However, this method mainly focuses on the spatial interpolation prediction and probability evaluation of a single monitoring parameter (deep displacement of soil body), and fails to systematically integrate the spatio-temporal coupling characteristics of multiple types of monitoring parameters (strain, crack, vibration, etc.), lacks hierarchical identification and positioning ability for weak parts of the geotechnical structure, and does not fully consider the dynamic coupling effect of rainfall, groundwater, ground load, and other environmental and load factors on the instability risk, resulting in insufficient comprehensiveness and real-time adaptability of risk assessment.

[0005] Therefore, we propose a geotechnical engineering monitoring and early warning method and system. SUMMARY

[0006] In view of the above-mentioned defects of the prior art, the present application provides a geotechnical engineering monitoring and early warning method and system, which can effectively solve the problems of the prior art.

[0007] To achieve the above object, the present application is implemented by the following technical solutions: The present application discloses a geotechnical engineering monitoring and early warning system, comprising: The construction module is used for constructing an initial benchmark model for geotechnical engineering deformation instability monitoring and extracting benchmark parameters according to geotechnical engineering design parameters and initial structure state, and storing the benchmark parameters to create a benchmark database; the perception module is used for deploying distributed perception units in the geotechnical engineering monitoring area, collecting real-time state data of the geotechnological structure, and comparing the real-time state data with the benchmark model parameters in real time to output structure state deviation data; the association module is used for extracting features of the structure state deviation data, calculating structure weakness indexes to quantify the weakness degree of each monitoring profile / zone, identifying the weakness level of the weak parts, and establishing an association relationship between the deviation features and the weak parts to form an association feature data set; the judgment module is used for constructing a risk coupling model by applying the association feature data set, judging the coupling degree of the deviation features and the instability risk, and outputting the risk coupling result; the mapping module is used for establishing a mapping logic of the risk and the early warning level according to the risk coupling result, mapping the risk coupling result into the corresponding early warning level, and storing the established mapping logic; and the early warning module is used for obtaining the risk coupling result mapped into the corresponding early warning level in the mapping module, selecting an early warning feedback target in a preset early warning feedback group based on the early warning level, and feeding back early warning information to the selected early warning feedback target. The construction module is interactively connected with the perception module through a wireless network, the perception module is interactively connected with the association module through a wireless network, the association module is interactively connected with the judgment module through a wireless network, and the judgment module is interactively connected with the mapping module and the early warning module through a wireless network.

[0008] Further, the geotechnical engineering design parameters in the construction module comprise: Slope engineering: slope gradient, slope height, soil layering parameters, supporting pile spacing and pile diameter, anchor design prestress and spacing, groundwater level design value, soil shear strength parameters; Foundation pit engineering: foundation pit excavation depth, supporting structure type and parameters, soil physical and mechanical parameters, design groundwater level, surrounding building load design value; Foundation engineering: foundation bearing capacity design value, pile foundation type and parameters, soil layer compression modulus and foundation settlement control value; The initial structure state includes: initial strain field distribution of the soil body after 30 days after construction is completed, initial value of horizontal displacement of the slope top, initial value of vertical displacement of the slope foot, initial width and coordinates of the slope surface crack, initial prestress value of the anchor cable, initial strain value of the supporting pile; initial displacement field of the pit wall after 7 days after excavation to the design depth, initial strain distribution of the supporting pile, initial heave of the pit bottom, initial value of the surrounding building settlement, initial value of the support axial force; initial settlement of the foundation, initial displacement of the pile top, initial value of the soil layer layered settlement after 30 days after loading is completed; In the construction module running stage, the consistency of the collected geotechnical engineering design parameters and the initial structure state data is checked, and abnormal data beyond the reasonable range is removed. Based on the mechanical correlation of the geotechnical engineering design parameters and the initial structure state, a calculation model of the synergistic action of the geotechnical body and the supporting structure is constructed. The calculation model is subjected to an operation condition of no external load disturbance. Through multiple rounds of iterative operation, the output parameters of the model in the stable state of the structure are obtained. The output parameters include the allowable deformation threshold field of each monitoring point within the design service life, the allowable stress interval of the supporting structure, and the allowable rate of crack propagation. The measured data of the initial structure state is compared with the stable operation output parameters of the model. The parameter item with the largest deviation from the measured data in the model is adjusted until the deviation between the output parameters of the model and the measured data is within the preset allowable deviation threshold. The calibrated model output parameters are used as the reference parameters. The corresponding geotechnical engineering design parameters and the initial structure state data are stored in the reference database together to complete the construction of the initial reference model.

[0009] Further, the distributed sensing unit adopts a hybrid deployment architecture of fiber grating arrays and wireless sensing nodes. The fiber grating arrays are laid along the slope profile direction or the foundation pit depth direction for collecting soil body / supporting structure strain field and temperature compensation data. The wireless sensing nodes are integrated by laser displacement sensors, piezoelectric vibration sensors, capacitive crack meters, and soil moisture sensors. The wireless sensing nodes are deployed according to the slope monitoring profile or the foundation pit monitoring partition. The number of wireless sensing nodes at the slope foot / pit bottom area is doubled for collecting real-time displacement, vibration response, crack width, and soil moisture content. In the sensing module, when the reference model parameters are compared in real time, the spatiotemporal coupling structure deviation degree is calculated first, and then the deviation data is output based on the deviation degree. The spatiotemporal coupling structure deviation degree calculation formula is: ; In the formula: is the number of monitoring parameter types and the number of geotechnical engineering monitoring profiles / partitions; is the risk weight coefficient of the i-th monitoring parameter; is the site risk coefficient of the j-th monitoring profile / partition; is the real-time deviation value of the i-th monitoring parameter of the j-th monitoring profile / partition. The static baseline threshold for the i-th type of monitoring parameter in the j-th monitoring profile / region; This represents the time trend weighting coefficient. Let be the rate of change of the deviation of the i-th type of monitoring parameter in the j-th monitoring profile / region; Let be the maximum allowable rate of change of deviation for the i-th type of monitoring parameter in the j-th monitoring profile / region; Spatiotemporal coupling structure deviation The calculation results are compared with the static relative deviation rate of each monitoring profile / zone and each monitoring parameter. Deviation change rate Perform correlation and integration to generate a data set including monitoring profile / zone number, parameter type, The structured deviation data report, which includes numerical values ​​and out-of-tolerance parameter identifiers, is simultaneously pushed to the relevant modules.

[0010] Furthermore, the features extracted by the association module during runtime include time-domain features, frequency-domain features, and time-frequency-domain features; The correlation module calculates a structural weakness index based on the deviation data to quantify the weakness of each monitoring profile / zone: ; In the formula: This is a structural weakness index used to quantify the structural weakness of a monitoring profile / zone; a larger value indicates a higher degree of weakness. This refers to the number of monitoring profiles / zones for geotechnical engineering. The part weighting coefficient for the j-th monitoring profile / zone; The number of monitored parameter types; The importance weights for the k-th type of monitoring parameters; The measured deviation value of the k-th type of monitoring parameter for the j-th monitoring profile / region; Let be the warning threshold for the k-th type of monitoring parameter; This is the nonlinear amplification factor; This is the spatial gradient adjustment coefficient; Let be the spatial gradient of the deviation between the j-th monitoring profile / section and its adjacent monitoring profiles / sections.

[0011] Furthermore, the risk coupling model in the judgment module includes a core input layer and an auxiliary input layer: the core input layer is the associated feature dataset output by the association module, and the auxiliary input layer includes dynamic parameters of the geotechnical environment and dynamic operational loads; When the judgment module determines the degree of coupling between deviation characteristics and instability risk, it uses a dynamic risk coupling index for quantitative characterization: ; In the formula: is a dynamic risk coupling index, the higher the value represents the stronger the coupling effect of the bias feature and the environment and load, and the higher the instability risk is; is a correlation feature weight coefficient; is an environmental parameter coupling weight; is a real-time environmental parameter comprehensive value; is an environmental parameter reference value; is an environmental parameter nonlinear amplification coefficient; is a load parameter coupling weight; is a real-time operating load comprehensive value; is a load parameter reference value; is a load parameter nonlinear amplification coefficient.

[0012] Further, the mapping logic of risk and early warning level in the mapping module is: based on the dynamic risk coupling index and the preset historical risk-accident correlation database to determine the early warning level: first-level early warning, indicating immediate instability risk: when ≥ , or the growth rate in a preset short time is not less than a preset growth threshold, and the corresponding instability probability of the interval in the historical database is not less than 80%; second-level early warning, indicating high instability risk: when ≤ < , and the corresponding structure damage probability of the interval in the historical database is not less than 60%; third-level early warning, indicating medium instability risk: when ≤ < , and the corresponding abnormal development probability of the interval in the historical database is not less than 40%; fourth-level early warning, indicating low risk: when < , and the corresponding structure stability probability of the interval in the historical database is not less than 90%; wherein, , , is a dynamic threshold and > > , the mapping module iteratively corrects the threshold based on the newly added and the actual structure state data every preset period, and the correction formula is , represents the original threshold, This represents the correction factor. This indicates the amount of abnormal data within the period. This indicates the total amount of data within the period.

[0013] Furthermore, the warning feedback target is preset by the system user, and is preset by the system user as a level four warning feedback target that is adapted to the warning level; Among them, when a Level 1 warning is issued, the corresponding warning feedback target is all warning feedback targets; When a Level 2 warning is issued, the corresponding warning feedback target is all warning feedback targets at Level 2 and below. The system user will preset the decision on whether to send the warning to the Level 1 warning feedback target. When a Level 3 warning is issued, the corresponding warning feedback target is all warning feedback targets at Level 3 and below. The system user presets the decision on whether to provide feedback to the Level 2 warning feedback target. When a Level 4 warning is issued, the corresponding warning feedback target is all Level 4 warning feedback targets. The system user makes a preset decision on whether to send feedback to Level 3 warning feedback targets.

[0014] Furthermore, the initial baseline model and the risk coupling model will trigger a system reset when any of the following conditions are met: The output of the perception module The error exceeded the preset deviation threshold for three consecutive data collection cycles. Geotechnical engineering has completed structural modification work, which includes slope reinforcement, retaining pile reinforcement, anchor cable reinforcement, and foundation pit support reinforcement. The amount of new data added to the benchmark database has reached 50% of the initial data volume; The deviation rate from the actual structural state exceeds the preset error rate.

[0015] On the other hand, a geotechnical engineering monitoring and early warning method includes: According to the geotechnical engineering design parameters and the initial structure state, the initial benchmark model of geotechnical engineering deformation instability monitoring is constructed through data consistency verification and model iteration calibration, and the benchmark parameters are extracted and stored to create a benchmark database; the distributed sensing unit of the mixed architecture of the fiber grating array and the wireless sensing node is deployed in the geotechnical engineering monitoring area, the real-time structure state data is collected, the spatial and temporal coupling structure deviation degree is calculated, and the structure state deviation data compared with the benchmark model parameters are output; the time domain, frequency domain and time-frequency domain features of the structure state deviation data are extracted, the structure weakness index is calculated to quantify the weakness degree of each monitoring profile / zone, the weakness part is identified, and the correlation between the deviation features and the weakness part is established to form a correlation feature data set; the risk coupling model is constructed with the correlation feature data set as the core input and the geotechnical environment dynamic parameters and the operation dynamic load as the auxiliary input, the dynamic risk coupling index is calculated to judge the coupling degree of the deviation features and the instability risk; the mapping logic of the risk and the early warning level is established based on the dynamic risk coupling index and the historical risk-accident correlation database, the risk coupling result is mapped to the corresponding early warning level, and the early warning threshold is periodically iterated and corrected; the early warning information of the geotechnical engineering instability risk corresponding to the early warning feedback target is pushed to the adaptive target selected from the preset early warning feedback group according to the early warning level obtained by mapping.

[0016] Compared with the known prior art, the technical scheme provided by the present application has the following beneficial effects: The present application provides a geotechnical engineering monitoring and early warning method and system. In the execution process, a precise initial benchmark model is constructed in combination with the geotechnical engineering design parameters and the initial structure state, the benchmark parameters are ensured to be reliable through data verification and iteration calibration, real-time data is collected through a distributed mixed sensing architecture, the structure state deviation is accurately identified in combination with the spatial and temporal coupling deviation degree, the multi-domain deviation features are extracted and the structure weakness is quantified to clearly point to the weakness part, the environmental dynamic parameters and the operation load are integrated to construct a risk coupling model, the instability risk coupling degree is dynamically quantified, the early warning level mapping logic is established based on the historical data and the dynamic threshold to realize graded early warning, and the system can be reset and optimized according to specific conditions, effectively improving the accuracy, timeliness and adaptability of the geotechnical engineering instability risk monitoring and early warning, providing reliable protection for the safety construction and operation of geotechnical engineering, reducing safety accidents caused by risk misjudgment or omission, and reducing the operation and maintenance cost of geotechnical engineering. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical scheme in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings from these drawings without creating any inventive labor.

[0018] Figure 1 Fig. 1 is a structural schematic diagram of a geotechnical engineering monitoring and early warning system; Figure 2 Fig. 2 is a flow schematic diagram of a geotechnical engineering monitoring and early warning method. DETAILED DESCRIPTION

[0019] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0020] The present application will be further described below with reference to the embodiments.

[0021] Embodiment 1 The geotechnical engineering monitoring and early warning system of the present embodiment, as shown in Fig. 1, comprises: Figure 1 A construction module for constructing an initial benchmark model of geotechnical engineering deformation instability monitoring and extracting benchmark parameters according to geotechnical engineering design parameters and initial structure state, and storing the benchmark parameters to create a benchmark database; The geotechnical engineering design parameters in the construction module comprise: Slope engineering: slope gradient, slope height, soil layering parameters (thickness of each soil layer, internal friction angle, cohesion), support pile spacing and pile diameter, anchor design prestress and spacing, design value of groundwater level (depth from the ground surface), soil shear strength parameters (including saturated state and unsaturated state); Foundation pit engineering: foundation pit excavation depth, support structure type (such as cantilever, support, anchor) and parameters (support pile diameter, corbel size, support system stiffness), soil physical and mechanical parameters (compression modulus, permeability coefficient), design groundwater level, peripheral building load design value; Foundation engineering: design value of foundation bearing capacity, pile foundation type (such as precast pile, cast-in-place pile) and parameters (pile length, pile diameter, pile tip resistance), soil compression modulus and foundation settlement control value; The initial structure state comprises: ​The initial structure state includes: initial strain field distribution of the soil body 30 days after construction is completed, initial value of horizontal displacement of the slope top, initial value of vertical displacement of the slope foot, initial width and coordinates of the slope surface crack, initial prestress value of the anchor cable, and initial strain value of the supporting pile; initial displacement field of the pit wall 7 days after excavation to the design depth, initial strain distribution of the supporting pile, initial heave of the pit bottom, initial value of the surrounding building settlement, and initial value of the support axial force; initial settlement of the foundation, initial displacement of the pile top, and initial value of the soil layer settlement after loading is completed for 30 days; In the running phase of the construction module, the consistency of the collected geotechnical engineering design parameters and the initial structure state data is checked, and abnormal data beyond the reasonable range is removed. Based on the mechanical correlation of the geotechnical engineering design parameters and the initial structure state, a calculation model of the cooperative action of the geotechnical body and the supporting structure is constructed. The model needs to cover the full-structure of the geotechnical engineering: slope engineering: including the slope top, slope body, slope foot, supporting pile, anchor cable and soil layer, and representing the stress distribution of the soil body, the pile-soil interaction of the supporting pile, and the stress transmission of the anchor cable; foundation pit engineering: including the pit wall, pit bottom, supporting pile, support system and surrounding influence area, and representing the interaction of the pit wall soil body and the supporting structure, the support constraint action, and the pit bottom heave mechanism; foundation engineering: including each soil layer, pile foundation and upper load transmission path, and representing the pile-soil interaction, the pile end resistance and the pile side friction resistance distribution. The calculation model is subjected to an operation condition without external load disturbance (i.e. only considering the self-weight of the geotechnical body, the initial ground stress and the pre-stress of the supporting structure, without considering external factors such as ground surcharge, construction disturbance, rainfall, etc.), and through multiple rounds of iterative operation, the output parameters of the model in the stable state of the structure are obtained. The output parameters include: the allowable deformation threshold field of each monitoring point within the design service life, the allowable stress range of the supporting structure, and the allowable rate of crack propagation. The measured data of the initial structure state is compared with the output parameters of the model in the stable state, and the calculation parameter item with the largest deviation from the measured data in the model (such as the elastic modulus of the soil body, the stiffness coefficient of the supporting structure, the boundary constraint condition, etc.) is adjusted through the parameter inversion method. The soil body shear strength parameters (internal friction angle and cohesion) are subject to the geotechnical engineering investigation test data and are not adjusted as the iteration calibration object. Until the deviation between the output parameters of the model and the measured data is within the preset allowable deviation threshold (±5% to ±8%), the calibrated model output parameters are used as the reference parameters, which are associated with the corresponding geotechnical engineering design parameters and the initial structure state data, and are stored in the reference database together to complete the construction of the initial reference model. The perception module is used to deploy distributed perception units in the geotechnical engineering monitoring area, collect real-time state data of the geotechnical structure, and compare it with the reference model parameters in real time to output the structure state deviation data. The distributed perception unit adopts a hybrid deployment architecture of fiber grating array and wireless sensing node: The fiber grating array is arranged along a slope monitoring profile (from the top of the slope to the foot of the slope) or a foundation pit depth direction (from the top of the pit wall to the bottom of the pit), and is used to collect soil / supporting structure strain field and temperature compensation data; The wireless sensing node is integrated by a laser displacement sensor, a piezoelectric vibration sensor, a capacitive crack meter, and a soil moisture content sensor, and is deployed according to a slope monitoring profile (slope top-slope waist-slope foot logic) or a foundation pit monitoring partition, and the slope foot area / pit bottom area is encrypted by 2 times, and is used to collect real-time displacement, vibration response, crack width and soil moisture content; In the perception module, when compared with the reference model parameters in real time, the spatiotemporal coupling structure deviation degree is first calculated, and then the deviation data is output based on the deviation degree; The spatiotemporal coupling structure deviation degree calculation formula is: ; In the formula, is the number of monitoring parameter types, and the number of geotechnical engineering monitoring profiles / partitions; is the risk weight coefficient of the i th monitoring parameter; is the site risk coefficient of the j th monitoring profile / partition; is the real-time deviation value of the i th monitoring parameter of the j th monitoring profile / partition; is the static reference threshold value of the i th monitoring parameter of the j th monitoring profile / partition; is the time trend weight coefficient; is the deviation change rate of the i th monitoring parameter of the j th monitoring profile / partition; is the allowed maximum deviation change rate of the i th monitoring parameter of the j th monitoring profile / partition; The above formula realizes comprehensive coverage of the monitoring parameters and profiles / partitions through double summation, and introduces risk weight coefficients and site risk coefficients, which can be differentiated according to the influence degree of the monitoring parameters on the instability risk and the risk level difference of different profiles / partitions of geotechnical engineering; The calculation result of the spatiotemporal coupling structure deviation degree is associated and integrated with the static relative deviation rate of each monitoring profile / partition and each monitoring parameter, the deviation change rate to generate a structured deviation data report containing the monitoring profile / partition number, parameter type, value and out-of-tolerance parameter identification, and the report is pushed to the associated module at the same time; In the formula, , and the weight coefficient and the risk coefficient are greater than zero and less than one, the weight coefficient of the monitoring parameter with a higher influence degree on the geotechnical engineering instability is greater, and vice versa, the risk coefficient of the geotechnical structure risk level is greater, and vice versa, and the risk level of the geotechnical structure is defined by the system end user, ∈ (0, 1], when the deviation change rate of the structure state deviation data is closer to the allowed maximum deviation change rate of the corresponding monitoring parameter, the greater the value is, and vice versa; The association module is used for feature extraction of the structure state deviation data, mining of the association features of the deviation data and the weak parts of the geotechnical structure, and formation of an association feature data set. The features extracted by the association module in the structure state deviation data include time domain features, frequency domain features and time-frequency domain features, the time domain features include sliding average, root mean square error and abnormal duration of the deviation data, the frequency domain features include power spectrum peak value and main frequency drift of the deviation data, and the time-frequency domain features include wavelet coefficient energy entropy and singular value entropy, wherein the time-frequency domain features are extracted by wavelet transform, and the abnormal duration is the continuous time from the first time of exceeding the threshold value to the time of returning to the threshold value; The association module calculates a structure weakness index of the deviation data to quantify the weakness degree of each monitoring profile / zone: ; In the formula, is the structure weakness index, which is used for quantifying the structure weakness degree of the monitoring profile / zone, and the greater the value is, the higher the weakness degree is; is the number of geotechnical engineering monitoring profiles / zone; is the part weight coefficient of the jth monitoring profile / zone; is the number of monitoring parameter types; is the importance weight of the kth monitoring parameter; is the measured deviation value of the kth monitoring parameter of the jth monitoring profile / zone; is the early warning threshold value of the kth monitoring parameter; is a nonlinear amplification coefficient; is a spatial gradient adjustment coefficient; is the deviation spatial gradient of the jth monitoring profile / zone and the adjacent monitoring profile / zone; wherein the deviation spatial gradient The calculation formula is: ; In the formula, is the number of monitoring parameters participating in gradient calculation; is the deviation value of the kth monitoring parameter of the jth monitoring profile / zone; This represents the deviation value of the k-th type of monitoring parameter for the (j-1)-th monitoring profile / region. The spatial distance between the j-th monitoring profile / section and the (j-1)-th monitoring profile / section; The first term in the above formula quantifies the degree to which the deviation values ​​of various monitoring parameters in each monitoring profile / zone exceed the warning threshold through double summation, reflecting the degree to which the structural state deviates from the safety benchmark, and introduces a nonlinear amplification factor. The first term reflects the physical mechanism of accelerated enhancement of weakness when approaching a critical state; the second term reflects the influence of the rate of change of deviation between adjacent monitoring profiles / regions through a spatial gradient correction term, when the spatial gradient... A large value indicates that the spatial distribution of deviations in this area is uneven, stress concentration is significant, and the structure is more vulnerable, which is consistent with the classic principle of geotechnical engineering for identifying weak points based on displacement abrupt changes and stress concentration; the third item uses the location weight coefficient. Highlight the contribution of mechanically sensitive areas (such as the sliding start-up zone at the toe of the slope and the bulging zone at the bottom of the pit); in, , Furthermore, all weighting coefficients are values ​​greater than zero and less than one; the monitoring parameters that have a higher impact on geotechnical engineering instability, The larger the value, the lower the value; the higher the risk level of the geotechnical structure, the better. The larger the value, the smaller the value; When a certain monitoring parameter approaches or exceeds the warning threshold, its contribution to the assessment of structural weakness is nonlinearly amplified. When the spatial gradient of the deviation has a greater auxiliary effect on the assessment of structural weakness, The larger the value; Calculate the values ​​of each monitoring profile / zone After summing the values ​​(only for the j-th profile / region), sort the values ​​to identify the level of weak points and establish the correlation between deviation characteristics and weak points: Slope engineering (r=3 levels): The core weak zone (q=1) is The monitoring profile with the largest value is set as The secondary weak zone (q=2) is The monitoring profile with the second largest value is set as The potential weak zone (q=3) is Profiles with relatively small values ​​but still requiring monitoring are set as follows: ; The method for identifying the hierarchy of weak points in foundation pit engineering and foundation engineering is similar, with the core weak point corresponding to... The largest area (such as the pit bottom heave zone, the soft soil layer at the pile tip), the secondary weak area corresponds to the second largest area, and the potential weak area corresponds to the area with a smaller value but needs attention. Through the above structural weakness assessment and weak point identification, deviation characteristics and structural weakness indices are identified. Weak points are hierarchically linked and integrated to form a related feature dataset, which serves as the core input to the risk coupling model. The judgment module is used to construct a risk coupling model using the associated feature dataset, judge the degree of coupling between deviation features and instability risk, and output the risk coupling result; The risk coupling model in the judgment module includes a core input layer and an auxiliary input layer: the core input layer is the hierarchical associated feature dataset output by the association module, and the auxiliary input layer includes dynamic parameters of the soil and rock environment and dynamic loads of operation; the dynamic parameters of the soil and rock environment include: atmospheric precipitation, soil moisture content change rate, groundwater level change rate, and vegetation root system soil stabilization coefficient; the dynamic loads of operation include: ground load (such as soil piles on slope tops and loads around pits), construction disturbance vibration amplitude (such as pile foundation construction and blasting vibration), surrounding building loads, and duration of temporary loads; When determining the degree of coupling between deviation characteristics and instability risk, the judgment module uses a dynamic risk coupling index for quantitative characterization: ; In the formula: It is a dynamic risk coupling index. The higher the value, the stronger the coupling effect between the deviation characteristics and the environment and load, and the higher the risk of instability. These are the weight coefficients for the associated features; for; Weights are used to couple environmental parameters; This represents the combined value of real-time environmental parameters. These are baseline values ​​for environmental parameters; This refers to the nonlinear amplification factor for environmental parameters. For load parameter coupling weights; This represents the comprehensive value of real-time operational loads. These are the baseline values ​​for the load parameters; This is the nonlinear amplification factor for the load parameters; The above formula is based on Based on this foundation, it reflects the inherent fragility of the geotechnical structure. It strengthens the impact of monitoring profiles / zones with high fragility and significant risk contribution through a fragility weighting coefficient. Simultaneously, it introduces environmental and operational load parameters as risk triggering factors. A binary fraction is used to quantify the deviation of environmental and load values ​​from the baseline. Combined with a nonlinear amplification coefficient, the influence of parameters is adjusted for different scenarios such as soil softening upon contact with water and large ground loads. This achieves multi-dimensional coupling quantification of structural fragility with environmental and load factors, conforming to the classic risk assessment framework of "risk = fragility × triggering factor." It comprehensively reflects the actual influencing factors of instability risk, improves the accuracy of risk judgment, and adheres to the principle of "risk = fragility × triggering factor." The higher the reflected monitoring profile / section structure weakness degree, the greater the contribution to instability risk, The greater the value, the lower the weakness degree, and the smaller the contribution to instability risk, The smaller the value; By atmospheric rainfall, soil moisture content rate, groundwater level change rate (mandatory) and vegetation root system soil fixation coefficient (optional) through weighted summation, By ground load, construction vibration amplitude, surrounding building load, temporary load duration through weighted summation, And Before calculation, each parameter is converted to [0,1] interval according to the preset normalization formula to eliminate dimensional difference, and then calculation is performed; The preset 0.2≤ ≤0.4, 0.3≤ ≤0.5, and , , The sum of the three is always not more than 6; When the soil is easy to soften when encountering water (such as clay soil, expansive soil), the groundwater level rises rapidly (rise rate> 0.5m / d) or the vegetation root system fails (ecological slope scene), The value increases; When the ground load is large (more than 20% of the design value), the construction vibration frequency is close to the inherent frequency of the soil or the surrounding building load continues to increase, The value increases; The mapping module is used to establish the mapping logic of risk and early warning level according to the risk coupling result, map the risk coupling result to the corresponding early warning level, and store the established mapping logic; The mapping logic of risk and early warning level in the mapping module is: Based on dynamic risk coupling index And the preset historical risk-accident correlation database to determine the early warning level: First-level warning, indicating immediate instability risk: when ≥ , or The growth rate in a preset short time is not less than the preset growth threshold, and the corresponding instability probability in the historical database Is not less than 80%; Second-level warning, indicating high instability risk: when ≤ < , and the corresponding structure damage probability in the historical database Is not less than 60%. Level 3 warning, indicating medium instability risk: when ≤ < , and the corresponding abnormal development probability in the historical database is not less than 40%; Level 4 warning, indicating low risk: when < < , and the corresponding structural stability probability in the historical database is not less than 90%; wherein , , , is a dynamic threshold value and > > , the mapping module iteratively corrects the threshold value based on the newly added and the actual structural state data every preset period, and the correction formula is , , wherein represents the original threshold value, represents the correction coefficient, represents the amount of abnormal data in the period, represents the total data amount in the period, is initially set to 0.05~0.3, and the higher the proportion of structural state abnormal data in the period, i.e. , the larger the value, and vice versa; The above formula is based on the original threshold value, introduces the correction coefficient and the proportion of abnormal data in the period, and adjusts the threshold correction amplitude through the product of the two. When the amount of abnormal data in the period increases, the threshold value will be adjusted upwards to avoid missing the new risk due to a fixed threshold. When the amount of abnormal data is small, the threshold correction amplitude is reduced to ensure the stability of the warning level determination and to ensure the dynamic iterative optimization of the threshold value with the accumulation of system operation data. This makes the mapping logic of risk and warning level continuously adapt to the changes in the state of the rock-soil structure, thereby improving the timeliness and accuracy of the warning determination; The mapping module also includes an index association mapping function when establishing the mapping logic of risk and warning level, which is used to establish a corresponding relationship between the dynamic risk coupling index calculated by the system and the standard safety index in the geotechnical engineering design specification, so that the warning information can present both the self-defined risk index and the industry-standard safety evaluation index, making it easier for engineering and technical personnel to understand the actual engineering significance of the warning results; The mapping module has a preset index mapping relationship table, which stores the corresponding relationship between the value interval and the standard safety index of geotechnical engineering, and the specific mapping index is determined according to the type of geotechnical engineering: Slope engineering: Mapped to slope stability safety factor The estimation range and safety factor definition refer to the "Technical Specification for Building Slope Engineering" GB50330, and the requirements for permanent slopes. Temporary slope requirements ; Foundation pit engineering: The deformation level of the foundation pit support structure and the estimated value of the stress safety of the support pile are mapped to the deformation level of the foundation pit support structure and the deformation level classification refers to the "Technical Specification for Foundation Pit Support" JGJ120, which is divided into Level 1 deformation (strict control), Level 2 deformation (general control) and Level 3 deformation (relaxed control). The stress safety of the support pile is defined as the ratio of the actual stress to the allowable stress. Foundation engineering: The mapping is to the foundation settlement state and settlement occupancy rate. The settlement occupancy rate is defined as the ratio of the actual cumulative settlement to the allowable settlement value specified in the "Code for Design of Building Foundations" GB50007. A settlement occupancy rate of less than 70% is a safe state, 70% to 90% is a monitoring state, and more than 90% is a warning state. During the initial deployment of the system, initial values ​​for the index mapping relationship table are preset according to the geotechnical engineering type and relevant design specifications. The initial mapping relationship is established based on the following: reference to the safety factor requirements and deformation control standards in "Technical Specification for Building Slope Engineering" GB50330, "Technical Specification for Foundation Pit Support" JGJ120, and "Code for Design of Building Foundations" GB50007; combined with engineering experience data, The fourth threshold range ( , , ) Corresponds to the safety factor or deformation level segment; for the same warning level, The closer to the upper limit threshold, the lower the corresponding safety factor estimate or the more severe the deformation level; the initial mapping relationship table for foundation pit engineering and foundation engineering is established in the same way, based on the deformation control requirements and settlement allowable values ​​in the relevant specifications respectively; During system operation, the indicator mapping table is gradually optimized in the following two ways to improve mapping accuracy: Method 1: Finite Element Inversion Analysis When the perception module calculates the spatiotemporal coupling structure deviation When the deviation exceeds the preset threshold continuously, or the warning level reaches level two or above, the mapping module triggers a finite element numerical simulation inversion analysis. The specific process is as follows: The finite element model parameters are updated based on the current monitoring data (displacement, strain, crack width, etc.); the safety factor of the current structure is calculated using the strength reduction method (slope engineering), internal force analysis method (foundation pit engineering), or settlement calculation method (foundation engineering). Actual values ​​of standard indicators such as support pile bending moment or foundation settlement; compare these actual values ​​with the corresponding time... Value pairing is performed, and the data is stored as a set of calibration data in the mapping relationship optimization dataset. When the accumulated optimization dataset reaches a preset threshold (e.g., more than 30 sets), regression analysis methods (e.g., multinomial regression, support vector regression) are used to establish... The quantitative functional relationship between the standard indicators and the interval divisions and corresponding indicator estimates in the mapping table are updated. Method 2: Expert Certification Method Every preset period (e.g., quarterly or semi-annually), geotechnical engineering experts assess the consistency between the warning level and the actual structural condition, based on on-site inspection results. The expert assessment includes: whether the warning level matches the actual conditions observed on-site, such as crack development, displacement growth, and support deformation; and the accuracy of the warning information. The consistency between the corresponding standard indicator estimates and the on-site structural safety assessment conclusions (such as visual stability of slopes, slight deformation of foundation pit support, etc.); if a systematic deviation is found in the mapping relationship (such as the actual safety factor being generally higher than the estimated value in the mapping table during a level-two warning), experts can manually adjust the interval division or corresponding indicator range in the mapping relationship table, and the corrected mapping relationship table will be stored as a new version and take effect; The two optimization methods mentioned above can be executed in parallel. The finite element inversion analysis method is preferred to obtain quantitative data. When the data accumulation is insufficient or the engineering conditions are special, the expert calibration method is used for calibration to ensure that the index mapping relationship table continuously adapts to the actual engineering conditions. It should be noted that: The historical risk-accident correlation database is built with the system's own data accumulation as the core and external data from the same source as the supplement: The core components originate from the entire chain of data continuously collected and stored during system operation, including real-time monitoring data of the geotechnical structure state acquired by the sensing module (such as strain, displacement, crack width, soil moisture content, etc.), the associated feature dataset output by the association module, and the dynamic risk coupling index calculated by the judgment module. The early warning module's early warning operation records, as well as subsequent data obtained through manual inspections and professional structural testing, such as the risk index corresponding to the actual structural state / accident outcome, are also included. Whether damage actually occurred in the geotechnical engineering within the numerical range, the degree of damage, and whether instability (slope landslide, foundation pit collapse, excessive foundation settlement) occurred; The supplementary section introduces publicly available accident case data from other geotechnical engineering projects with the same or similar engineering conditions as the target geotechnical engineering project (such as soil layer type, support structure type, groundwater level, slope gradient / foundation pit depth, and ground load level), as well as historical risk-structure status correlation data from authoritative geotechnical engineering monitoring databases in the industry (such as the China Geological Survey Geological Disaster Database and geotechnical engineering monitoring databases of various provinces and cities). All external data are included after being standardized in format. At the same time, during the database construction process, outliers and invalid data (such as outliers caused by sensor failures) will be removed from all the above-mentioned source data, risk data will be labeled with structural status tags (such as labeled as "stable", "minor damage", "severe damage", "instability"), and the data will be classified and stored according to engineering conditions and risk index ranges. The early warning module is used to obtain the risk coupling result mapping corresponding to the early warning level in the mapping module, select the early warning feedback target from the preset early warning feedback group based on the early warning level, and feed back early warning information to the selected early warning feedback target; The early warning feedback target is preset by the system user and is preset by the system user as a level four early warning feedback target that is adapted to the early warning level; Among them, when a Level 1 warning is issued, the corresponding warning feedback target is all warning feedback targets; When a Level 2 warning is issued, the corresponding warning feedback target is all warning feedback targets at Level 2 and below. The system user will preset the decision on whether to send the warning to the Level 1 warning feedback target. When a Level 3 warning is issued, the corresponding warning feedback target is all warning feedback targets at Level 3 and below. The system user presets the decision on whether to provide feedback to the Level 2 warning feedback target. When a Level 4 warning is issued, the corresponding warning feedback target is all Level 4 warning feedback targets. The system user presets and decides whether to provide feedback to the Level 3 warning feedback target. When the early warning module pushes early warning information to the early warning feedback target, in addition to the early warning level, it also includes the following related indicator information: (1) Current value of dynamic risk coupling index: Actual calculation results; (2) Standard safety index estimates: obtained by querying the index mapping relationship table, and different indices are presented according to different geotechnical engineering types: Slope engineering: Slope stability safety factor Estimated range; Foundation pit engineering: Deformation level and state description of support structure; Foundation engineering: Settlement state and settlement occupancy rate of foundation; (3) Weak part hierarchical positioning information: Based on the calculation of the associated module The identification results clearly indicate the specific locations of the core weak areas and secondary weak areas; (4) Current status of environmental and load parameters: list the key environmental or load factors that trigger risk coupling; (5) Disposal suggestions: automatically generated based on the warning level and weak parts; The initial baseline model and the risk-coupled model will trigger a system reset if any of the following conditions are met: The output of the perception module The error exceeded the preset deviation threshold for three consecutive data collection cycles. Geotechnical engineering completes structural reconstruction work, which includes slope engineering: slope reinforcement (such as foot pressing, slope cutting, anti-slide pile reinforcement), anchor reinforcement, support pile reinforcement; foundation pit engineering: foundation pit support reinforcement, support pile reinforcement, anchor reinforcement, pit bottom grouting reinforcement; foundation engineering: foundation reinforcement grouting, pile foundation reinforcement, foundation reinforcement treatment; The amount of new data added to the reference database reaches 50% of the initial data amount; The deviation rate from the actual structure state exceeds the preset error rate; The construction module is connected to the perception module through a wireless network, the perception module is connected to the association module through a wireless network, the association module is connected to the judgment module through a wireless network, and the judgment module is connected to the mapping module and the early warning module through a wireless network.

[0022] In this embodiment, the construction module runs to construct an initial reference model of geotechnical engineering deformation instability monitoring according to the geotechnical engineering design parameters and the initial structure state, and extracts reference parameters, stores the reference parameters to create a reference database, the perception module is deployed in the geotechnical engineering monitoring area to collect real-time state data of the geotechnical structure, and compares the real-time state data with the reference model parameters in real time, and outputs structure state deviation data, the association module further extracts features from the structure state deviation data, mines the association features of the deviation data and the weak parts of the geotechnical structure, forms an association feature data set, and then the judgment module applies the association feature data set to construct a risk coupling model, judges the coupling degree of the deviation features and the instability risk, outputs the risk coupling result, and through the mapping module, the mapping logic of risk and early warning level is established according to the risk coupling result, the risk coupling result is mapped to the corresponding early warning level and the established mapping logic is stored, finally the early warning module obtains the risk coupling result mapped to the corresponding early warning level in the mapping module, selects an early warning feedback target in a preset early warning feedback group based on the early warning level, and feeds back early warning information to the selected early warning feedback target.

[0023] In the above embodiment, the system can establish a precise reference according to the geotechnical engineering design parameters and the initial structure state, collect structure data in real time and compare and analyze, accurately mine the association of deviation and weak parts, quantify the risk coupling degree combined with environmental dynamic parameters and operation load, and push the adaptive early warning according to the risk level. It can also dynamically correct the early warning threshold and reset the model in time. It can identify instability hazards in advance, greatly reduce the probability of accidents, and effectively ensure the safety of geotechnical engineering construction and operation.

[0024] Regarding the construction of the initial reference model in this embodiment, an example of constructing an initial reference model is shown below: Taking the construction of the initial reference model of the highway slope as an example, first, collect the core data of the slope: First, the geotechnical engineering design parameters include: slope angle 45°, slope height 25m, soil layer parameters (first layer silty clay thickness 8m, internal friction angle φ1=18°, cohesion c1=25kPa; second layer strongly weathered mudstone thickness 12m, internal friction angle φ2=28°, cohesion c2=45kPa), support pile spacing 3.5m, pile diameter 1.0m, anchor cable design prestress 250kN, and groundwater level design value (12m down from the ground surface). Second, the initial structural state data are the measured data 30 days after the slope construction was completed, which include: the strain of the shallow soil of the slope body of 50~80με, the initial value of the horizontal displacement at the top of the slope of 12mm, the initial value of the vertical displacement at the toe of the slope of 8mm, the initial width of the crack on the slope surface of 0.15mm, the initial prestress value of the anchor cable of 242kN, and the initial strain value of the support pile of 150με. Next, the collected data were checked for consistency. An abnormal displacement of 35mm was found at a monitoring point at the top of the slope (exceeding the reasonable displacement range of 8-15mm for similar slopes), and this data was removed. Subsequently, based on the mechanical relationship between the design parameters and the initial structural state, a calculation model of the interaction between the soil and rock mass and the support structure was constructed. This model covers the entire slope structure (including the top, body, toe, support piles, and anchor cables).

[0025] The model was subjected to computational conditions without external load disturbance. After multiple rounds of iterative computation, the output parameters of the model under the structural stability state were obtained: the allowable horizontal displacement at the top of the slope is 18 mm, the allowable maximum bending moment of the support pile is 180 kN·m, and the allowable daily crack propagation rate is ≤0.01 mm / d.

[0026] Comparing the measured data of the initial structural state with the output parameters of the model's steady-state calculation, it was found that the "internal friction angle of the soil layer" parameter in the model deviated the most from the measured data. After adjusting this parameter, the deviation between the model output parameters and the measured data was reduced to within the preset allowable deviation threshold (±6%).

[0027] Finally, the calibrated model output parameters are used as reference parameters, and the above geotechnical engineering design parameters and initial structural state data are associated with them and stored together in the reference database to complete the construction of the initial reference model of the slope.

[0028] Example 2: At the implementation level, based on Example 1, this example refers to... Figure 2 A more detailed description of the geotechnical engineering monitoring and early warning system in Example 1 is provided below: A geotechnical engineering monitoring and early warning method includes: Based on the geotechnical engineering design parameters and the initial structural state, an initial benchmark model for monitoring deformation instability in geotechnical engineering was constructed through data consistency verification and model iteration calibration. Benchmark parameters were extracted and stored to create a benchmark database. The distributed sensing unit of the mixed architecture of the fiber grating array and the wireless sensing node is deployed in the geotechnical engineering monitoring area, real-time structure state data is collected, and structure state deviation data compared with the benchmark model parameters are output after calculating the space-time coupled structure deviation degree; The time domain, frequency domain and time-frequency domain features of the structure state deviation data are extracted, the associated features are calculated and mined, and the associated feature data set is formed; The risk coupling model is constructed with the associated feature data set as the core input and the geotechnical environment dynamic parameters and the operation dynamic load as the auxiliary input, the dynamic risk coupling index is calculated to judge the coupling degree of the deviation features and the instability risk; Based on the dynamic risk coupling index and the historical risk-accident association database, the mapping logic of risk and early warning level is established, the risk coupling result is mapped to the corresponding early warning level, and the early warning threshold is regularly iterated and corrected; According to the early warning level obtained by mapping, the adaptive target is selected in the preset early warning feedback group, and the corresponding geotechnical engineering instability risk early warning information is pushed to the early warning feedback target.

[0029] In summary, the method and system in the above embodiments, in the execution process, combine the geotechnical engineering design parameters and the initial structure state to construct a precise initial benchmark model, ensure the reliability of the benchmark parameters through data verification and iterative calibration, collect real-time data through the distributed mixed sensing architecture, accurately identify the structure state deviation by combining the space-time coupled deviation degree, extract multi-domain deviation features and quantify the structure weakness to clearly point out the weak parts, build a risk coupling model by integrating environmental dynamic parameters and operating loads, dynamically quantify the instability risk coupling degree, and establish an early warning level mapping logic based on historical data and dynamic thresholds to realize graded early warning. It can also trigger system reset and optimize the model according to specific conditions, effectively improve the accuracy, timeliness and adaptability of geotechnical engineering instability risk monitoring and early warning, provide reliable protection for geotechnical engineering safety operation, reduce safety accidents caused by risk misjudgment or omission, and reduce operation and maintenance costs.

[0030] The above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A geotechnical monitoring and early warning system, characterized in that, The method comprises the following steps: A construction module is used to construct an initial benchmark model of geotechnical engineering deformation instability monitoring and extract benchmark parameters according to geotechnical engineering design parameters and an initial structure state, and the benchmark parameters are stored to create a benchmark database; A perception module is used to deploy distributed perception units in a geotechnical engineering monitoring area, collect real-time state data of a geotechnological structure, and compare the real-time state data with benchmark model parameters in real time to output structure state deviation data; An association module is used to extract features of the structure state deviation data, calculate structure weakness indexes to quantify the weakness degree of each monitoring profile / zone, identify the grade of weak parts, and establish an association between the deviation features and the weak parts to form an association feature data set; A judgment module is used to construct a risk coupling model by using the association feature data set, judge the coupling degree of the deviation features and instability risks, and output a risk coupling result; A mapping module is used to establish a mapping logic of risks and early warning levels according to the risk coupling result, map the risk coupling result to a corresponding early warning level, and store the established mapping logic; An early warning module is used to obtain the risk coupling result mapped to the corresponding early warning level in the mapping module, select an early warning feedback target in a preset early warning feedback group based on the early warning level, and feed back early warning information to the selected early warning feedback target.

2. The geotechnical engineering monitoring and early warning system of claim 1, wherein, The geotechnical engineering design parameters in the construction module include: Slope engineering: slope gradient, slope height, soil layering parameters, support pile spacing and pile diameter, anchor design prestress and spacing, groundwater level design value, soil shear strength parameters; Foundation pit engineering: foundation pit excavation depth, support structure type and parameters, soil physical and mechanical parameters, design groundwater level, surrounding building load design value; Foundation engineering: foundation bearing capacity design value, pile foundation type and parameters, soil layer compression modulus and foundation settlement control value; The initial structure state includes: soil initial strain field distribution 30 days after construction, slope top horizontal displacement initial value, slope foot vertical displacement initial value, slope surface crack initial width and coordinates, anchor initial prestress value, support pile initial strain value; pit wall initial displacement field, support pile initial strain distribution, pit bottom initial heave, surrounding building settlement initial value, support shaft force initial value 7 days after excavation to the design depth; foundation initial settlement, pile top initial displacement, soil layering settlement initial value 30 days after loading; The construction module runs a stage, consistency check of collected geotechnical engineering design parameters and initial structure state data, elimination of abnormal data beyond reasonable range, construction of calculation model of rock-soil body and supporting structure synergistic effect based on mechanical correlation of geotechnical engineering design parameters and initial structure state, application of no external load disturbance operation condition to calculation model, acquisition of output parameters of model in stable state of structure through multiple rounds of iterative operation, the output parameters including allowable deformation threshold field of each monitoring point in design service life, allowable stress interval of supporting structure, allowable rate of crack propagation, comparison of measured data of initial structure state and model stable operation output parameters, adjustment of parameter items in model with largest deviation from measured data until deviation of model output parameters from measured data is within preset allowable deviation threshold, and storage of calibrated model output parameters as reference parameters, associated with corresponding geotechnical engineering design parameters and initial structure state data, to reference database to complete construction of initial reference model.

3. The geotechnical engineering monitoring and early warning system of claim 1, wherein, The distributed sensing unit adopts a hybrid deployment architecture of fiber grating array and wireless sensing node, the fiber grating array is arranged along the slope monitoring profile or the depth direction of the foundation pit, and is used for collecting soil body / supporting structure strain field and temperature compensation data, the wireless sensing node is integrated by a laser displacement sensor, a piezoelectric vibration sensor, a capacitive crack meter and a soil water content sensor, and is deployed according to the slope monitoring profile or the foundation pit monitoring partition, and the number of the wireless sensing node at the slope foot or the pit bottom area is doubled, and is used for collecting real-time displacement, vibration response, crack width and soil water content; In the sensing module, when the reference model parameters are compared in real time, the spatiotemporal coupling structure deviation degree is calculated first, and then the deviation data is output based on the deviation degree; The spatiotemporal coupling structure deviation degree calculation formula is: ； In the formula: The number of monitoring parameter types, the number of geotechnical engineering monitoring profiles / sections; The risk weight coefficient of the i-th monitoring parameter; The site risk coefficient of the j-th monitoring profile / section; The real-time deviation value of the i-th monitoring parameter of the j-th monitoring profile / section; The static reference threshold value of the i-th monitoring parameter of the j-th monitoring profile / section; The time trend weight coefficient; The deviation change rate of the i-th monitoring parameter of the j-th monitoring profile / section; The maximum allowable deviation change rate of the i-th monitoring parameter of the j-th monitoring profile / section; The calculation result of the spatiotemporal coupling structure deviation degree is associated and integrated with the static relative deviation rate of each monitoring profile / partition and each monitoring parameter, the deviation change rate , to generate a structured deviation data report containing the monitoring profile / partition number, parameter type, , numerical value and out-of-tolerance parameter identification, and the report is pushed to the associated module synchronously.

4. The geotechnical engineering monitoring and early warning system of claim 1, wherein, The features extracted by the correlation module include time domain features, frequency domain features and time-frequency domain features; The correlation module calculates the structure weakness index of the deviation data to quantify the weakness degree of each monitoring profile / partition: ； In the formula: is a structural weakness index, used to quantify the structural weakness of the monitoring profile / zone, the larger the value, the higher the degree of weakness; is the number of geotechnical engineering monitoring profiles / zones; is the site weight coefficient of the jth monitoring profile / zone; is the number of monitoring parameters; is the importance weight of the kth monitoring parameter; is the measured deviation value of the kth monitoring parameter of the jth monitoring profile / zone; is the early warning threshold of the kth monitoring parameter; is a nonlinear amplification coefficient; is a spatial gradient adjustment coefficient; is the deviation spatial gradient of the jth monitoring profile / zone and the adjacent monitoring profile / zone.

5. A geotechnical monitoring and warning system according to claim 4, wherein, The risk coupling model in the judgment module includes a core input layer and an auxiliary input layer: the core input layer is the correlation feature data set output by the correlation module, and the auxiliary input layer includes geotechnical environment dynamic parameters and operation dynamic load; When the judgment module judges the coupling degree of deviation features and instability risk, the dynamic risk coupling index is used for quantitative representation: ； In the formula: is a dynamic risk coupling index, and the higher the value represents the stronger the coupling effect of the bias feature with the environment and load, and the higher the risk of instability; is a correlation feature weight coefficient; is an environment parameter coupling weight; is a real-time environment parameter comprehensive value; is an environment parameter reference value; is an environment parameter nonlinear amplification coefficient; is a load parameter coupling weight; is a real-time operation load comprehensive value; is a load parameter reference value; is a load parameter nonlinear amplification coefficient.

6. A geotechnical monitoring and warning system according to claim 5, wherein, The mapping logic of risk and early warning level in the mapping module is: Dynamic risk coupling index Alert level determination with a pre-set historical risk-incident correlation database: A Level 1 warning indicates an immediate risk of instability: when ≥ ,or The growth rate within a preset short period is not less than a preset growth threshold, and this interval is in the historical database. The corresponding probability of instability is not less than 80%; A level-two warning indicates a high risk of instability: when ≤ < And this interval in the historical database The corresponding structural damage probability is not less than 60%; A Level 3 warning indicates a risk of instability in the middle: when ≤ < And this interval in the historical database The corresponding probability of abnormal development is no less than 40%; Level 4, indicating low risk: when < , and the corresponding structural stability probability in the historical database for this interval is not less than 90%; Wherein, , , is a dynamic threshold and , the mapping module iteratively corrects the threshold based on the newly added and the actual structure state data every preset period, and the correction formula is , represents the original threshold, represents the correction coefficient, represents the amount of abnormal data in the period, represents the total amount of data in the period.​​ 7. The geotechnical engineering monitoring and early warning system of claim 1, wherein, The early warning feedback target is preset by the system end user, and the four-level early warning feedback target is preset by the system end user to adapt to the early warning level; Among them, the corresponding early warning feedback target of the first level early warning is all early warning feedback targets; The corresponding early warning feedback target of the second level early warning is all early warning feedback targets of the second level and below, and whether to feedback to the first level early warning feedback target is decided by the system end user; The corresponding early warning feedback target of the third level early warning is all early warning feedback targets of the third level and below, and whether to feedback to the second level early warning feedback target is decided by the system end user. When the fourth-level early warning is triggered, the corresponding early warning feedback target is all fourth-level early warning feedback targets, and the system end user presets whether to feed back to the third-level early warning feedback target.

8. The geotechnical engineering monitoring and early warning system according to claim 3 or 5, characterized in that, The initial benchmark model and the risk coupling model are triggered to reset the system when any of the following conditions is met: the perception module outputs exceeds a preset deviation threshold for 3 consecutive acquisition cycles Geotechnical engineering completes structural reconstruction work, which includes slope reinforcement, support pile reinforcement, anchor reinforcement, foundation pit support reinforcement; The amount of new data in the benchmark database reaches 50% of the initial data amount; a deviation rate from an actual structural state exceeds a preset error rate.

9. The geotechnical monitoring and warning system of claim 1, wherein, The construction module is connected to the sensing module through a wireless network, the sensing module is connected to the association module through a wireless network, the association module is connected to the judgment module through a wireless network, and the judgment module is connected to the mapping module and the warning module through a wireless network.

10. A method for monitoring and early warning of geotechnical engineering, the method is a method for implementing the geotechnical engineering monitoring and early warning system according to any one of claims 1-9, characterized in that, It comprises: According to the geotechnical engineering design parameters and the initial structure state, the geotechnical engineering deformation instability monitoring initial benchmark model is constructed through data consistency verification and model iteration calibration, and the benchmark parameters are extracted and stored to create a benchmark database; Deploy a distributed sensing unit with a hybrid architecture of fiber grating arrays and wireless sensing nodes in the geotechnical engineering monitoring area, collect real-time structure state data, calculate the spatiotemporal coupling structure deviation degree, and output the structure state deviation data compared with the benchmark model parameters; Extract the time domain, frequency domain and time-frequency domain features of the structure state deviation data, calculate the structure weakness index to quantify the weakness degree of each monitoring profile / zone, identify the weakness level and establish the association between the deviation features and the weakness positions, and form the associated feature data set; Take the associated feature data set as the core input, the geotechnical environment dynamic parameters and the operation dynamic load as the auxiliary input to construct the risk coupling model, calculate the dynamic risk coupling index to judge the coupling degree of the deviation features and the instability risk; Based on the dynamic risk coupling index and the historical risk-accident association database, the mapping logic of risk and early warning level is established, the risk coupling result is mapped to the corresponding early warning level, and the early warning threshold is regularly iterated and corrected; According to the early warning level obtained by mapping, select the adaptive target in the preset early warning feedback group, and push the corresponding geotechnical engineering instability risk early warning information to the early warning feedback target.

Images