Multi-sensor fusion slope stability intelligent analysis system and edge calculation method

Abstract

The invention relates to the technical field of slope stability data identification, and particularly provides a multi-sensor fusion slope stability intelligent analysis system and an edge calculation method.The system comprises an environment sensing subsystem, a data processing subsystem and an edge calculation subsystem, the environment sensing subsystem monitors the change of the soil moisture content through a soil humidity sensor, and micro movement data of a slope are obtained through a displacement sensor; using a pressure sensor to sense the change of the ground pressure; meanwhile, pre-screening out abnormal data to form a fused data set; the data analysis subsystem is used for converting abnormal data of the fusion data set into a quantitative index of slope stability, and deducing a potential sliding risk of the slope by analyzing the relevance of the soil water content, the tiny movement data or the ground pressure; and the decision support subsystem provides specific reinforcement countermeasures according to the analysis result of the potential sliding risk. According to the method, the potential sliding risk of the side slope is comprehensively deduced by analyzing the relevance of the soil water content, the tiny movement data and the ground pressure.

Description

Technical Field

[0001] The present invention relates to the technical field of slope stability data recognition, and particularly relates to an intelligent analysis system for slope stability with multi-sensor fusion and an edge computing method. Background Art

[0002] Slope stability analysis is an important task in the field of engineering geology, involving the comprehensive evaluation of multiple factors such as slope deformation, stress, and hydrogeology. With the development of computer technology and sensor technology, people have begun to attempt to use intelligent methods to analyze slope stability. Traditional analysis methods rely on limited sensor data and expert experience, suffering from problems such as insufficient information and low efficiency. Therefore, it is particularly important to develop an intelligent analysis technology for slope stability with sensor fusion that can integrate multi-source sensor data and improve the accuracy and real-time performance of analysis.

[0003] Prior Art One, a Chinese patent with the application number: 202210302009.2 discloses a multi-sensor data fusion method based on a piecewise function, including the following steps: obtaining multiple sensor data and statistically outputting the target distribution; analyzing the stability of multiple sensor data and outputting a jump flag bit; fitting the target distribution based on the piecewise function and outputting a weight curve; fusing multiple sensor data based on the weight curve to obtain fused data. Although compared with existing methods, this method has simple steps and requires less computing power; in addition, the weight curve is adjusted by analyzing the stability of multiple sensor data to make the fusion of multiple sensor data more accurate and reliable; however, it lacks multi-physical field dynamic coupling analysis: this technology adjusts the weight curve through a piecewise function to achieve data fusion, without considering the dynamic interaction relationship between parameters such as soil moisture, displacement, and pressure in slope stability analysis, and cannot reveal the influence mechanism of multi-physical field coupling on the sliding risk. Limitations of static weight allocation: The weight adjustment based on stability analysis still relies on a static model and is difficult to adapt to the spatio-temporal variation characteristics of sensor data during the dynamic evolution of slopes, resulting in a deviation between the fusion result and the real risk.

[0004] Prior Art 2, a Chinese patent with application number 202210248874.3, discloses a situation assessment method for multi-source heterogeneous information fusion of a probabilistic inference model. In the first step, in a military situation data fusion system, a strategy of activating a specific sensor is adopted to collect more information. At the same time, it may also expose the position of the sensor to the enemy, prompting it to take actions to effectively change the situation. In the second step, the uncertainty in the fusion inference process reflects the dynamic accumulation and propagation process of the uncertainty of multi-source information. At each step of the fusion, the uncertainty factors of multi-source information need to be integrated. As the inference progresses, an uncertain conclusion is finally obtained. Although it solves the problem of the influence of various uncertainty factors in the fusion system and effectively improves the stability of the fusion system, it has insufficient scene applicability: This technology is designed for military situation assessment and needs to actively expose the sensor position to obtain information, which is not applicable to the slope scene that requires long-term concealed monitoring and is not combined with the decision-making requirements for slope reinforcement. There is a lack of real-time dynamic feedback: Its uncertainty inference mechanism is not embedded in the dynamic closed-loop of real-time sensing data and reinforcement measures, and it is unable to adjust the early warning strategy immediately according to the change of the slope state.

[0005] Prior Art 3, a Chinese patent with application number 202210504092.1, discloses a multi-sensor asynchronous information fusion method and system. By filtering and extracting the original data information of multiple sensors, the number k of clustering clusters is calculated using the filtered kernel information. Then, based on the number k of clustering clusters, the number k of clustering clusters is analyzed and calculated. After weighted combination and solution of the central data information of each clustering cluster obtained by the calculation, information fusion is completed, thus achieving multi-sensor asynchronous information fusion. Although clustering analysis based on the filtered kernel information of the original data information of multiple sensors can effectively avoid situations such as asynchronous sensor information or packet loss, and then analyze and calculate based on the number K of clustering clusters to obtain the central data information of each clustering cluster, ensuring the stability of information fusion, reducing the difficulty of fusion, and solving the problem of difficult fusion of real-time multi-sensors in practical applications, it ignores physical relevance: This technology processes asynchronous data through clustering but does not combine the actual physical associations of parameters such as soil moisture, displacement, and pressure (such as a decrease in soil strength caused by an increase in humidity), resulting in the disconnection between the fused data and the slope instability mechanism. It does not pre-screen abnormal data: Directly clustering and fusing the original data may introduce sensor noise or outlier interference, affecting the reliability of the final risk assessment.

[0006] Currently, Prior Art 1, Prior Art 2, and Prior Art 3 have problems of disconnection between data fusion and physical mechanism, significant early warning lag, and lack of decision support. Therefore, the present invention provides a multi-sensor fusion intelligent analysis system and edge computing method for slope stability. Summary of the Invention

[0007] To achieve the above object, the present invention adopts the following technical solutions: On the one hand, the present invention provides an intelligent analysis system for slope stability with multi-sensor fusion, comprising: An environmental perception subsystem, which is used to monitor the change of soil moisture content by using a soil moisture sensor, obtain the minute movement data of the slope through a displacement sensor, and sense the change of ground pressure by using a pressure sensor; at the same time, pre-screen the abnormal data to form a fusion data set; A data analysis subsystem, which is used to convert the abnormal data in the fusion data set into a quantitative index of slope stability, and infer the potential sliding risk of the slope by analyzing the correlation of soil moisture content, minute movement data or / and ground pressure; A decision support subsystem, which is used to propose specific reinforcement countermeasures according to the analysis result of the potential sliding risk.

[0008] In an optional implementation manner, the environmental perception subsystem comprises: A multi-dimensional perception module, which is used to obtain the three-dimensional model of the slope to be detected, determine the layout coordinates of the soil moisture sensor, displacement sensor and pressure sensor according to the historical slope landslide risk; form a slope data detection network composed of the soil moisture sensor, displacement sensor and pressure sensor; An abnormal pre-screening module, which is used to initialize the slope data detection network, compare the fluctuation trends of slope data within the same time window; set dynamic thresholds according to the physical coupling law of humidity-displacement-pressure; obtain the abnormal soil moisture content, minute movement data and ground pressure data in the slope data detection network according to the preset fluctuation trend standard and dynamic threshold; A data fusion module, which is used to receive the abnormal soil moisture content, minute movement data and ground pressure data, and form a multi-dimensional associated fusion data set containing marks by using the layout coordinates of the soil moisture sensor, displacement sensor and pressure sensor as marks.

[0009] In an optional implementation manner, the multi-dimensional perception module comprises: A three-dimensional model construction sub-module, which is used to obtain the surface geometric data of the slope to be detected, generate a three-dimensional model through point cloud processing and surface reconstruction, including geometric features such as elevation, slope and curvature of the terrain, and also calibrate the potential risk area through historical landslide data including the position of the slip surface and the failure range to form a digital slope model with geological risk marks; A risk weight calculation sub-module, which is used to calculate the risk weights of different areas of the slope to be detected on the basis of the digital slope model, in combination with the spatial distribution of the position of the slip surface and the failure range, the failure depth and the triggering factors; The sensor coordinate optimization sub-module is used to determine the mechanical coupling relationship existing among soil moisture, displacement, and pressure, and set the layout conditions for the corresponding sensors; through iterative optimization, determine the spatial matching coordinates of the three types of sensors.

[0010] In an optional implementation, the anomaly pre-screening module includes: The multi-source data spatio-temporal alignment sub-module is used to unify the spatio-temporal benchmarks for three types of data, namely soil water content, displacement, and ground pressure, based on the established slope data monitoring network; The coupling feature extraction sub-module is used to establish an association analysis framework for three types of parameters, namely humidity-displacement association, displacement-pressure association, and humidity-pressure association, according to the physical properties of the rock and soil mass of the slope to be detected; The dynamic threshold generation sub-module is used to adopt a sliding time window mechanism to construct an adaptive discrimination criterion for baseline threshold, trend correction, and spatial correction; implement a three-level discrimination mechanism for single-point anomaly, local anomaly, and system anomaly; perform structured processing on the confirmed anomaly data, including adding spatio-temporal marks, attaching confidence levels, and recording the evolution process.

[0011] In an optional implementation, the anomaly pre-screening module includes: In the time dimension of the multi-source data spatio-temporal alignment sub-module, a millisecond-level synchronization mechanism is established based on the data acquisition timestamps; in the spatial dimension, a three-dimensional spatial reference system is constructed using the sensor layout coordinate information; in the data dimension, the original data with different dimensions is converted into standard monitoring indicators; The humidity-displacement association analysis of the coupling feature extraction sub-module examines the soil deformation response caused by seepage; the displacement-pressure association monitors the stress redistribution caused by soil sliding; the humidity-pressure association evaluates the influence of pore water pressure on effective stress; The baseline threshold of the dynamic threshold generation sub-module establishes an initial discrimination criterion based on the data in the historical stable period; the trend correction adjusts the threshold sensitivity according to the recent data change rate; the spatial correction is differentially set according to the risk weights of the positions where the sensors are located; single-point anomaly: the data of a single sensor exceeds the dynamic threshold range; local anomaly: adjacent sensor groups show collaborative anomalies; system anomaly: the three types of parameters present an abnormal combination that conforms to the failure mechanism.

[0012] In an optional implementation, the coupling feature extraction sub-module includes: The parameter association model construction unit is used to gradually establish an association framework for the three types of parameters by adopting a progressive analysis strategy based on the physical properties of the slope rock and soil mass in the coupling feature extraction sub-module; The association relationship logic confirmation unit is used to set the analysis logic for humidity-displacement association, displacement-pressure association, and humidity-pressure association; An abnormal coupling verification unit is used to cross-verify the results of three types of correlation analyses; if an abnormal signal holds in multiple correlation models, it is marked as an abnormal with high confidence.

[0013] In an alternative embodiment, the data analysis subsystem includes: A logic startup module is used to receive abnormal data, start the analysis logics of the corresponding humidity-displacement correlation, displacement-pressure correlation, and humidity-pressure correlation, and screen the abnormal data according to the corresponding input content; A coupling analysis module is used to perform multi-parameter coupling and risk quantification of humidity, displacement, and pressure; A collaborative verification module is used for multi-parameter collaborative verification and risk grading, and checks whether the anomalies of the three types of parameters meet the typical combination patterns of slope instability; according to the number, coupling strength, and spatial distribution range of the abnormal parameters, the risk level is divided.

[0014] In an alternative embodiment, the local low risk of the collaborative verification module: single-parameter anomaly; local high risk of double-parameter coupling anomaly; systematic risk of anomalies of all three types of parameters and conforming to the failure mechanism.

[0015] In an alternative embodiment, the decision support subsystem includes: A risk characteristic analysis and classification module is used to receive the risk assessment results from the data analysis subsystem, including: risk type marking, risk level division, and key parameter anomaly characteristics, and automatically construct a risk characteristic matrix; A targeted measure matching and optimization module is used to initiate a hierarchical response strategy according to the risk characteristic matrix; A spatial adaptation and strength calibration module is used to adjust the coverage density of engineering measures according to the spatial distribution density of abnormal parameters, determine the design strength level of the structure according to the risk level quantization value, and consider the timing characteristics of parameter anomalies to plan the priority of measure implementation.

[0016] On the other hand, the present invention provides a multi-sensor fusion-based edge computing method for slope stability of the multi-sensor fusion-based slope stability intelligent analysis system. The multi-sensor fusion-based edge computing method for slope stability includes the following steps: The multi-sensor fusion-based edge computing method for slope stability includes the following steps: Deploy a micro computing unit at the edge node of the sensor network to achieve in-situ feature extraction, dynamic data compression, cross-modal alignment, and lightweight fusion; Construct a three-level analysis architecture of fast response at the edge layer, correlation analysis at the fog layer, and in-depth verification at the cloud layer; Establish an edge intelligent model with dynamic update, including cloud model slimming, incremental learning, cross-node knowledge sharing, and working condition adaptation.

[0017] The present invention comprehensively monitors and preprocesses data. By monitoring soil humidity, slope displacement, and ground pressure through the environmental perception subsystem and pre-screening abnormal data, the accuracy and reliability of the data are ensured, providing true and effective input information for potential landslide risk analysis. In-depth analysis and risk assessment: The data analysis subsystem converts the preprocessed data into quantitative indicators of slope stability. By analyzing the correlation between soil water content, minute movement data, and ground pressure, and comprehensively considering these factors, the potential landslide risk of the slope is inferred, which provides a scientific evaluation basis for slope safety. Timely response and decision support: The decision support subsystem proposes specific reinforcement countermeasures according to the analysis results, enabling corresponding preventive measures to be taken for different risk levels, thereby reducing or even avoiding the occurrence of slope sliding accidents. The present invention realizes comprehensive, continuous monitoring and real-time evaluation of slope stability, provides scientific and accurate decision support for slope engineering management, and effectively prevents and reduces the risk of slope disasters, ensuring the safety of people's lives and property. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings: Figure 1 is a block diagram of an intelligent analysis system for slope stability with multi-sensor fusion provided in Embodiment 1 of the present invention; Figure 2 is a block diagram of the environmental perception subsystem provided in Embodiment 2 of the present invention; Figure 3 is a block diagram of the data analysis subsystem provided in Embodiment 6 of the present invention; Figure 4 is a block diagram of the decision support subsystem provided in Embodiment 7 of the present invention; Figure 5 is a flowchart of an edge computing method for slope stability with multi-sensor fusion provided in Embodiment 8 of the present invention; Figure 6 is a block diagram of an electronic device provided by the present invention; Figure 7 is a block diagram of a computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Next, the technical solutions in the embodiments of the present invention will be described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0020] Hereinafter, terms such as "first", "second", etc. are only used for convenience of description and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first", "second", etc. may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, the meaning of "a plurality" is two or more.

[0021] In the present invention, unless otherwise clearly specified and defined, the term "connection" should be understood in a broad sense. For example, "connection" can be a fixed mechanical connection, a detachable mechanical connection, or integrated; or, "connection" can be a direct connection, or an indirect connection through an intermediate medium. In addition, unless otherwise clearly specified and defined, the term "coupling" should be understood in a broad sense. For example, "coupling" can be a direct electrical connection. For example, there is physical contact and electrical conduction between two components, and it can also be understood that in a circuit structure, different components are electrically connected through an entity line such as a copper foil or a wire of a printed circuit board (PCB) that can transmit electrical signals to transmit electrical signals; or, "coupling" can be an indirect electrical connection between two components through an intermediate medium; or, "coupling" can be an electrical connection between two components in an air-spaced / non-contact manner. For example, two components are electrically connected in a capacitive coupling manner to transmit electrical signals.

[0022] In the embodiments of the present invention, orientation terms such as "upper", "lower", "left", "right", etc. may include but are not limited to being defined relative to the schematic placement of components in the drawings. It should be understood that these directional terms can be relative concepts, and they are used for relative description and clarification, and they can change accordingly with the change of the orientation of the components in the drawings.

[0023] Embodiment 1: As Figure 1 shown, the embodiments of the present invention provide an intelligent analysis system for slope stability with multi-sensor fusion, including: An environmental perception subsystem, which is used to monitor the change of soil moisture content by using a soil moisture sensor, obtain the minute movement data of the slope through a displacement sensor, and sense the change of ground pressure by using a pressure sensor; at the same time, pre-screen abnormal data to form a fusion data set; A data analysis subsystem, which is used to convert the abnormal data in the fusion data set into a quantitative index of slope stability, and infer the potential sliding risk of the slope by analyzing the correlation of soil moisture content, minute movement data or / and ground pressure; A decision support subsystem, which is used to propose specific reinforcement countermeasures according to the analysis result of the potential sliding risk.

[0024] In the above embodiments, for comprehensive monitoring and data preprocessing, through the environmental perception subsystem to monitor soil humidity, slope displacement and ground pressure, and pre-screen abnormal data, it ensures the accuracy and reliability of the data, providing true and effective input information for potential landslide risk analysis. In-depth analysis and risk assessment: The data analysis subsystem converts the preprocessed data into quantitative indicators of slope stability. By analyzing the correlation between soil water content, micro-movement data and ground pressure, and synthesizing these factors to infer the potential landslide risk of the slope, which provides a scientific evaluation basis for slope safety. Timely response and decision support: The decision support subsystem proposes specific reinforcement countermeasures according to the analysis results, enabling corresponding preventive measures to be taken for different risk levels, thereby reducing or even avoiding the occurrence of slope sliding accidents.

[0025] Generally speaking, this embodiment realizes comprehensive, continuous monitoring and real-time evaluation of slope stability, provides scientific and accurate decision support for slope engineering management, and further effectively prevents and reduces the risk of slope disasters, ensuring the safety of people's lives and property.

[0026] Embodiment 2: As Figure 2 shown, on the basis of Embodiment 1, the environmental perception subsystem provided by the embodiment of the present invention includes: A multi-dimensional perception module, which is used to obtain a three-dimensional model of the slope to be detected, determine the layout coordinates of soil humidity sensors, displacement sensors and pressure sensors according to the historical slope landslide risk; form a slope data detection network composed of soil humidity sensors, displacement sensors and pressure sensors; An abnormal pre-screening module, which is used to initialize the slope data detection network, compare the fluctuation trends of slope data within the same time window; set dynamic thresholds according to the physical coupling law of humidity-displacement-pressure; obtain abnormal soil water content, micro-movement data and ground pressure data in the slope data detection network according to the preset fluctuation trend standard and dynamic thresholds; A data fusion module, which is used to receive abnormal soil water content, micro-movement data and ground pressure data, and form a multi-dimensional associated fusion data set containing marks with the layout coordinates of soil humidity sensors, displacement sensors and pressure sensors as marks.

[0027] In the above embodiments, the environmental perception subsystem of this embodiment realizes the intelligent monitoring and early warning of slope risks through the collaborative work of multiple modules. The multi-dimensional perception module realizes the accurate spatial positioning of sensor layout based on historical landslide data and three-dimensional terrain modeling; through the topological deployment of three types of sensors, namely soil humidity, displacement, and pressure, a three-dimensional data acquisition matrix covering key mechanical nodes of the slope is formed, providing a spatialized basic data source. The abnormal pre-screening module establishes a dynamic coupling analysis model of humidity-displacement-pressure, and realizes the joint verification of cross-sensor data through the comparison of the fluctuation trends of time-series data and the adjustment of adaptive thresholds. The screening mechanism based on geotechnical mechanics principles significantly improves the reliability of single-parameter abnormal detection. The data fusion module reconstructs discrete abnormal data into a multi-dimensional data set with spatial topological relationships through sensor coordinate marking; the fusion method retains the spatial distribution characteristics of abnormal events, providing a standardized input with both time-series characteristics and spatial correlation structures for the subsequent landslide prediction model.

[0028] To sum up, this embodiment as a whole forms a technical closed-loop of "spatial deployment - dynamic detection - correlation fusion", realizing a complete processing link from raw data acquisition to risk feature extraction, providing a standardized and structured data basis for slope stability analysis; the cascaded design of each module ensures the physical consistency of monitoring data from acquisition to processing, avoiding the common information fragmentation problem in traditional single-point monitoring.

[0029] Embodiment 3: Based on Embodiment 2, the multi-dimensional perception module provided by the embodiment of the present invention includes: A three-dimensional model construction sub-module, which is used to obtain the surface geometric data of the slope to be detected, generate a three-dimensional model through point cloud processing and surface reconstruction, including geometric features such as the elevation, slope, and curvature of the terrain, and also calibrate potential risk areas through historical landslide data including the position of the slip surface and the failure range, forming a digital slope model with geological risk markings; A risk weight calculation sub-module, which is used to calculate the risk weights of different areas of the slope to be detected based on the digital slope model, in combination with the spatial distribution of the position of the slip surface and the failure range, the failure depth, and triggering factors; A sensor coordinate optimization sub-module, which is used to determine the mechanical coupling relationship between soil humidity, displacement, and pressure, and set the layout conditions of the corresponding sensors; through iterative optimization, determine the spatial matching coordinates of the three types of sensors; The humidity sensor is preferably arranged in areas prone to seepage (such as inside the slope body, near the potential slip surface); The displacement sensor is distributed along the potential sliding direction (such as the main axis of slope inclination) to capture the accumulation of small displacements; Pressure sensors are concentrated in the load-sensitive areas (such as the toe of the slope, near the supporting structure) to monitor the stress concentration phenomenon.

[0030] In the above embodiments, the multi-dimensional perception module of this embodiment constructs a complete optimized system for the slope monitoring sensor network through the collaborative action of each sub-module; from data acquisition to final deployment, it realizes: the construction of a spatial risk quantification system. The three-dimensional model construction sub-module fuses the original terrain data with historical disaster data to establish a digital model containing geometric features and risk markers; the risk weight calculation sub-module realizes the spatial quantification of slope risks on this basis, transforming the qualitative risk assessment into a computable spatial weight distribution. The design of the multi-physical quantity collaborative monitoring network. The sensor coordinate optimization sub-module, based on the principles of geotechnical mechanics, establishes the constraint conditions for the layout of humidity-displacement-pressure sensors; through the parametric iteration method, it realizes the optimal matching of the three types of sensors in spatial distribution, ensuring that the monitoring network can synchronously capture the key coupling parameters of slope instability. The realization of the adaptive monitoring architecture. The system forms a monitoring plan that can be dynamically adjusted according to the actual geological conditions through the closed-loop process from model construction to coordinate optimization, taking into account both the historical disaster laws and the real-time mechanical coupling relationship, enabling the sensor network to have the adaptability to different slope characteristics.

[0031] In summary, this embodiment realizes the method upgrade of slope monitoring from empirical layout to model-driven layout through the technical chain of geometric modeling-risk assessment-sensor optimization; the data transfer and logical connection of each sub-module ensure that the finally formed monitoring network has both reasonable spatial coverage and physical parameter relevance, providing an optimized data acquisition basis for slope stability analysis.

[0032] Embodiment 4: Based on Embodiment 2, the abnormal pre-screening module provided by the embodiment of the present invention includes: The multi-source data spatio-temporal alignment sub-module is used to unify the spatio-temporal benchmarks of three types of data, namely soil water content, displacement, and ground pressure, based on the constructed slope data monitoring network; Among them, in the time dimension: based on the data acquisition timestamp, a millisecond-level synchronization mechanism is established; In the space dimension: using the sensor layout coordinate information, a three-dimensional space reference system is constructed; In the data dimension: the original data with different dimensions is converted into standard monitoring indicators; The coupling feature extraction sub-module is used to establish an association analysis framework for three types of parameters, namely humidity-displacement association, displacement-pressure association, and humidity-pressure association, according to the physical properties of the rock and soil mass of the slope to be detected; Among them, humidity-displacement correlation: Analyze the soil deformation response caused by seepage; displacement-pressure correlation: Monitor the stress redistribution caused by soil sliding; humidity-pressure correlation: Evaluate the influence of pore water pressure on effective stress; The dynamic threshold generation sub-module is used to adopt a sliding time window mechanism to construct an adaptive discrimination criterion for benchmark threshold, trend correction, and spatial correction; implement a three-level discrimination mechanism for single-point anomaly, local anomaly, and system anomaly; and perform structured processing on the confirmed abnormal data, including adding spatio-temporal markers, attaching confidence levels, and recording the evolution process. Among them, the benchmark threshold: Establish an initial discrimination criterion based on historical stable period data; trend correction: Adjust the threshold sensitivity according to the recent data change rate; spatial correction: Differentially set according to the risk weights of the sensor locations; single-point anomaly: The data of a single sensor exceeds the dynamic threshold range; local anomaly: Collaborative anomalies occur in adjacent sensor groups; system anomaly: An abnormal combination that conforms to the failure mechanism appears in the three types of parameters.

[0033] In the above embodiment, this embodiment realizes the intelligent abnormal identification and analysis of slope monitoring data, and the specific significance is as follows: Through the multi-source data spatio-temporal alignment sub-module, the spatio-temporal benchmarks of three types of heterogeneous data, namely soil water content, displacement, and ground pressure, are unified, solving the analysis problems caused by asynchronous acquisition time, scattered spatial positions, and dimensional differences of monitoring data, and providing standardized data input for correlation analysis. The three-dimensional correlation analysis framework of humidity-displacement-pressure established by the coupled feature extraction sub-module breaks through the limitations of traditional single-parameter monitoring, can capture the mechanical interactions between geotechnical parameters, and more accurately reflect the triggering mechanism and evolution process of slope instability. Through the sliding time window mechanism, the adaptive update of the discrimination criterion is realized. Its three-level abnormal discrimination mechanism (single-point / local / system anomaly) can distinguish random errors, local dangers, and precursors of systematic instability, significantly improving the accuracy of early warning. The structured processing method of adding spatio-temporal markers, confidence level evaluation, and evolution process recording to the confirmed anomalies not only retains the complete context information of the abnormal data but also provides a quantitative basis for subsequent risk assessment. It realizes the automatic processing of slope monitoring from raw data to abnormal identification, and provides analysis results with spatio-temporal dimensions and multi-parameter correlation characteristics for engineering early warning.

[0034] Example 5: Based on Example 4, the coupled feature extraction sub-module provided by the embodiment of the present invention includes: The parameter correlation model construction unit is used to gradually establish a correlation framework for the three types of parameters by adopting a progressive analysis strategy based on the physical characteristics of the slope geotechnical body in the coupled feature extraction sub-module; The correlation relationship logic confirmation unit is used to set the analysis logic of humidity-displacement correlation, displacement-pressure correlation, and humidity-pressure correlation; Soil moisture content (humidity) and displacement data after standardizing the humidity-displacement correlation input; Analysis logic: Changes in soil moisture content (such as rainfall infiltration) affect the mechanical properties of the soil mass, resulting in deformation. By analyzing historical stable period data, establish a typical relationship model between humidity changes and displacement responses; When outputting abnormal fluctuations in humidity, predict the possible displacement development trend and quantify its correlation strength; Displacement-pressure correlation Input displacement data + ground pressure data (already spatio-temporally aligned); Dependent on the previous result, the output of the humidity-displacement correlation (such as whether displacement anomalies in a certain area are caused by seepage); Analysis logic Soil sliding will cause redistribution of internal stress. Combining the displacement change trend, analyze whether the data of the pressure sensor shows a response that conforms to the mechanical law (such as the pressure decreases in the area where the displacement increases, and the pressure increases in the adjacent area); Output whether the displacement anomaly is accompanied by a pressure change that conforms to the mechanical expectation, and judge the sliding risk; Humidity-pressure correlation Input humidity data + pressure data, depending on the previous results of humidity-displacement and displacement-pressure analysis (such as whether an increase in humidity first causes displacement and then indirectly affects pressure); Analysis logic An increase in soil moisture content will increase the pore water pressure and reduce the effective stress of the soil mass; By analyzing the direct correlation between humidity and pressure, exclude the interference of displacement effects (such as in some areas where the humidity increases but the displacement does not change significantly, but the pressure fluctuates abnormally, which may indicate potential liquefaction or softening risks); Output an independent assessment of the direct impact of pore water pressure on slope stability, supplementing the blind spots of the previous two correlations; Abnormal coupling verification unit, used to cross-verify the results of the three types of correlation analyses; If the abnormal signal holds in multiple correlation models, it is marked as a high-confidence abnormality; Single-correlation abnormality verification Check whether the three types of correlations of humidity-displacement, displacement-pressure, and humidity-pressure are independently established (such as whether an increase in humidity is accompanied by an increase in displacement and whether the pressure change meets the expectation); Multi-correlation collaborative verification If abnormalities occur in all three types of parameters, judge whether their combination conforms to the slope failure mechanism (such as humidity increase → displacement acceleration → pressure sudden drop, which conforms to the sliding trend); Confidence assessment Calculate the comprehensive abnormality confidence according to the correlation strength (such as the correlation coefficient of humidity-displacement) and the degree of abnormality (such as the amplitude exceeding the threshold); Output: Coupled abnormality marking: If the abnormal signal holds in multiple correlation models, it is marked as a high-confidence abnormality Type classification: Distinguish between seepage-dominated (significant humidity-displacement abnormality), sliding-dominated (significant displacement-pressure abnormality), or composite type (abnormalities in all three types of parameters); Risk level assessment: Combine the dynamic threshold to generate the spatial correction weight of the sub-module, and output the local or systematic risk level.

[0035] In the above embodiments, the coupling feature extraction sub-module of this embodiment realizes the multi-parameter coupling analysis and anomaly verification of slope monitoring data through the collaborative work of multiple units. The specific significance is as follows: Through the cooperation of the parameter correlation model construction unit and the correlation relationship logic confirmation unit, a mechanical correlation framework for three types of parameters, humidity-displacement-pressure, is systematically established, enabling the originally isolated monitoring indicators to form an analysis system based on geotechnical mechanics. Adopting a progressive strategy of "establishing single-parameter relationships → coupling analysis of two parameters → collaborative verification of three parameters" ensures that each correlation relationship is verified through physical mechanisms, avoiding one-sided conclusions driven by data. The anomaly coupling verification unit upgrades traditional single-point anomaly detection to composite anomaly identification based on mechanical consistency through three-level mechanisms of single-correlation verification, multi-correlation collaborative verification, and confidence evaluation, significantly reducing the false alarm rate. The output classification of anomaly types (seepage type / sliding type / composite type) and risk level assessment provide differentiated bases for disposal decisions, realizing the upgrade from "anomaly alarm" to "mechanism diagnosis". It develops slope monitoring from single-parameter threshold alarm to an intelligent diagnosis system based on multi-parameter coupling mechanisms, providing analysis results with mechanical theory support for engineering early warning.

[0036] Embodiment 6: As Figure 3 shown, based on Embodiment 1, the data analysis subsystem provided by the embodiment of the present invention includes: A logic startup module for receiving abnormal data, starting the analysis logics for humidity-displacement correlation, displacement-pressure correlation, and humidity-pressure correlation corresponding thereto, and screening abnormal data according to the corresponding input content; A coupling analysis module for performing multi-parameter coupling and risk quantification of humidity, displacement, and pressure; Humidity-displacement coupling, seepage deformation response analysis Input: Abnormal soil water content and displacement data; Analysis logic: An increase in soil water content will soften the rock and soil mass and reduce its shear strength; By comparing the typical patterns (such as lag time, deformation rate) of humidity change and displacement response in historical data, determine whether the current humidity anomaly is likely to cause significant displacement; Output If the humidity anomaly area is accompanied by accelerating displacement, it is marked as a "seepage-deformation" coupling risk and its impact degree is quantified; Displacement-pressure coupling sliding stress transfer analysis, input abnormal displacement data and pressure data (depending on the previous humidity-displacement analysis results); Analysis logic When the soil mass slides, the area with increasing displacement will cause the pressure to decrease due to the loosening of the soil mass, while the pressure in the adjacent area may increase due to stress transfer. Combining the humidity-displacement analysis results (such as whether the displacement is caused by seepage), verify whether the pressure change conforms to the mechanical response caused by sliding; Output If the displacement and pressure anomalies show spatial coordination (such as increased displacement and decreased pressure at the toe of the slope), it is marked as a "sliding-stress redistribution" risk; Analysis of the influence of humidity-pressure coupled pore water pressure, inputting abnormal humidity and pressure data (combining the results of the first two types of analysis); analysis logic, after excluding displacement interference, directly analyze the relationship between humidity and pressure. For example, when the humidity suddenly increases but the displacement does not change significantly, if the pressure fluctuates abnormally synchronously, it may reflect the risk of soil softening or liquefaction caused by a sudden increase in pore water pressure; output independent identification of abnormalities directly related to "humidity-pressure", supplementing risk scenarios not covered in the first two stages; The collaborative verification module is used for multi-parameter collaborative verification and risk grading, checking whether the abnormalities of the three types of parameters meet the typical combination patterns of slope instability; dividing the risk levels according to the number, coupling strength and spatial distribution range of the abnormal parameters; Local low risk: single parameter abnormality (such as only humidity increase); local high risk: double parameter coupling abnormality (such as humidity-displacement synchronous abnormality); systematic risk: all three types of parameters are abnormal and conform to the failure mechanism.

[0037] In the above embodiment, the data analysis subsystem of this embodiment constructs a progressive and multi-parameter fusion slope stability analysis system through the collaborative work of the logic startup module, the coupling analysis module and the collaborative verification module. The logic startup module automatically triggers the corresponding correlation analysis logic according to the type of abnormal data (humidity, displacement or pressure), ensuring that the data input meets the requirements of subsequent analysis; avoiding the interference of invalid data and improving the pertinence and efficiency of analysis. Humidity-displacement coupling: identify the influence of seepage on slope deformation and judge whether the displacement is accelerated due to soil softening; displacement-pressure coupling: combine the humidity-displacement analysis results to verify whether the sliding causes stress redistribution and improve the credibility of sliding risk discrimination; humidity-pressure coupling: independently analyze the direct influence of pore water pressure and supplement the softening or liquefaction risks that may be missed by the first two types of correlations; through the three types of coupling analysis, systematically cover the main inducements of slope instability (seepage, sliding, abnormal pore water pressure). Check whether the abnormalities of humidity, displacement and pressure meet the typical slope instability combination patterns (such as humidity increase → displacement acceleration → pressure sudden drop). Based on the number, coupling strength and spatial distribution of abnormal parameters, divide local low risk (single parameter abnormality), local high risk (double parameter abnormality) and systematic risk (triple parameter abnormality), making the risk assessment more hierarchical and operable.

[0038] In summary, in this embodiment, from abnormal data screening to coupling analysis, and then to collaborative verification and risk grading, the system has established a logically coherent analysis process; the finally output risk level (low / high / systematic) can directly support the formulation of reinforcement measures, achieving seamless connection from monitoring to decision-making; through a modular and progressive analysis strategy, multi-source sensor data is transformed into risk indicators with mechanical basis, upgrading the slope stability assessment from single-parameter alarm to multi-parameter coupling intelligent diagnosis, improving the accuracy of early warning and the scientific nature of decision-making.

[0039] Embodiment 7: As Figure 4 shown, on the basis of Embodiment 1, the decision support subsystem provided by the embodiment of the present invention includes: A risk feature analysis and classification module, which is used to receive the risk assessment results from the data analysis subsystem, including: risk type markers (seepage-dominated / sliding-dominated / composite), risk level classification (local low risk / local high risk / systematic risk), abnormal feature of key parameters (such as sudden increase amplitude of humidity, displacement acceleration, pressure imbalance range), automatically construct a risk feature matrix. For seepage-dominated risks, focus on the spatial correspondence between the humidity abnormal area and the displacement response; for sliding-dominated risks, focus on analyzing the spatial distribution pattern of the displacement-pressure coupling abnormality; for composite risks, comprehensively evaluate the collaborative abnormal features of the three types of parameters. Among them, the expression of the risk feature matrix R: ; In the formula, represents the seepage-dominated risk marker (taking values 0 or 1); represents the sliding-dominated risk marker (taking values 0 or 1); represents the composite risk marker (taking values 0 or 1); represents the local low risk level (quantified value 0.1~0.3); represents the local high risk level (quantified value 0.4~0.7); represents the systematic risk level (quantified value 0.8~1.0); represents the sudden increase amplitude of humidity (unit: percentage change in volumetric water content; represents the displacement acceleration (unit: mm / h²); represents the pressure imbalance range (unit: m²); the risk feature matrix is constructed based on the multi-source data fusion theory and the space-time coupling analysis model.

[0040] Through the tensor product operation ( Separate the independent influencing factors of risk type, level, and parameters to avoid decision-making ambiguity caused by feature coupling; normalize the original sensor data (such as humidity, displacement, pressure) into dimensionless risk level values (L / H / S) and physical dimension parameters (ΔH / A / P) to achieve unified representation of multi-dimensional data; the combination of marked variables (α / β / γ) and quantified values can describe the risk evolution path (for example, when α = 1 and ΔH > 30%, the seepage-type risk is triggered to escalate). Specific functions: Structured decision input: Provide a standardized parameter interface for the matching of reinforcement measures; predict the critical point of the risk evolution from local to systemic through the time-series update of matrix elements; generate a three-dimensional optimization solution by combining coverage density (spatial distribution), design strength (level quantification), and implementation priority (time-series characteristics); abstract complex slope risks into computable and intervenable mathematical objects through the orthogonal feature space and dynamic weight allocation mechanism, which is the key hub for the intelligent analysis system to transition from data perception to engineering decision-making.

[0041] The targeted measure matching and optimization module is used to initiate a hierarchical response strategy according to the risk feature matrix; (1) Risk response dominated by seepage: Local drainage measures: Arrange a light drainage network in areas with significantly abnormal humidity, and adopt a radial drainage ditch design; Deep water conduction plan: When abnormal humidity is accompanied by deep displacement signals, additional vertical drainage wells are added; Seepage prevention and reinforcement: Implement surface seepage prevention treatment on continuously high-humidity areas, combined with grass planting to stabilize the soil; (2) Risk response dominated by sliding: Local reinforcement: Arrange a micro anti-slide pile array in the area with abnormal coupling of displacement and pressure; Stress transfer design: Set up a force conduction retaining wall according to the characteristics of pressure redistribution; Toe protection: Adopt stepped reinforcement for the toe area with significant displacement accumulation; (3) Risk response for compound types: Combined treatment: Arrange the drainage system and anti-slide structure in a coordinated manner; Stratified treatment: A three-layer protection system of surface seepage prevention + middle water conduction + deep anchoring; Dynamic adjustment: Set up an adjustable support structure to adapt to subsequent deformation development; The space adaptation and strength calibration module is used to adjust the coverage density of engineering measures according to the spatial distribution density of abnormal parameters, determine the design strength grade of structures based on the risk level quantification value, and plan the implementation priority of measures considering the time-series characteristics of parameter anomalies.

[0042] In the above embodiments, in this embodiment, through the risk feature analysis and classification module, accurate identification and classification of slope risks are achieved; based on the risk assessment results provided by the data analysis subsystem (including risk types, levels, and abnormal characteristics of key parameters), a multi-dimensional risk feature matrix is constructed: for seepage-dominated risks, a humidity-displacement spatial correlation model is established; for sliding-dominated risks, a displacement-pressure coupling distribution map is constructed; for complex risks, a multi-parameter collaborative evaluation framework is formed; the original monitoring data is transformed into characteristic indicators with engineering significance, providing structured input for decision-making. The targeted measure matching and optimization module realizes according to the risk feature matrix: automatically generating corresponding engineering solution libraries for different risk types (seepage / sliding / composite); dynamically adjusting the measure intensity according to the risk level (low / high / systematic); combining the spatial distribution characteristics of abnormal parameters to optimize the plane layout plan of engineering measures; the output preliminary plan already has the characteristics of spatial pertinence and intensity difference. The spatial adaptation and intensity calibration module performs multi-dimensional optimization on the preliminary plan, dynamically adjusts the coverage density of engineering measures according to the degree of abnormal parameter aggregation; determines the design parameter threshold of the structure based on the risk quantification value; considers the evolution trend of parameter anomalies and formulates the priority order of measure implementation. Transforming complex monitoring data into operable engineering characteristics; establishing a mapping relationship between risk characteristics and engineering measures; achieving precise adaptation in three dimensions of space, intensity, and time sequence.

[0043] Embodiment 8: As Figure 5 shown, based on Embodiments 1-7, the edge computing method for slope stability with multi-sensor fusion provided by the embodiments of the present invention includes the following steps: S100: Deploy a micro computing unit at the edge node of the sensor network to achieve in-situ feature extraction, dynamic data compression, cross-modal alignment, and lightweight fusion; In-situ feature extraction: Each sensor node calculates local feature values in real time (such as humidity change rate, displacement acceleration, pressure gradient); Dynamic data compression: Adopt a sliding window mechanism to transmit only the feature vectors exceeding the threshold; Cross-modal alignment: Through timestamp synchronization and spatial coordinate mapping, complete the spatio-temporal alignment of multi-source data at the edge layer; Lightweight fusion: Use the improved D-S evidence theory for confidence fusion and output preliminary anomaly indicators; S200: Construct a three-level analysis architecture of fast response at the edge layer, correlation analysis at the fog layer, and in-depth verification at the cloud layer; Fast response at the edge layer: Deploy a micro decision tree model to immediately trigger a local warning for obvious anomalies (such as a 30% sudden increase in humidity + displacement exceeding the limit); Correlation analysis at the fog layer: Run a lightweight LSTM network at the regional network gateway node to analyze the time-series correlation characteristics of multi-sensors; Deep verification in the cloud: Verify the mechanism model uploaded to the cloud for complex working conditions (such as abnormal multi-parameter coupling). Dynamic weight adjustment: Automatically adjust the computing load of each layer according to the network state to ensure real-time requirements. S300: Establish an edge intelligent model with dynamic updates, including cloud model slimming, incremental learning, cross-node knowledge sharing, and working condition adaptation. Cloud model slimming: Compress the complex model trained in the cloud into an edge-deployable version through channel pruning and quantization. Incremental learning: The new case data collected by the edge node is encrypted and used for model iteration and update. Cross-node knowledge sharing: Adopt the federated learning mechanism, and multiple edge nodes cooperate to optimize the local model. Working condition adaptation: Automatically switch the model parameter set according to the changes in the on-site environment (such as rainy season / dry season).

[0044] In the above embodiments, the edge computing method for slope stability of multi-sensor fusion realizes the full-chain optimization from data perception to decision response through the deep coupling of the edge computing architecture and the dynamic intelligent model. The core of its technology lies in: the in-situ feature extraction and dynamic compression (S100) based on the micro computing unit significantly reduce the amount of redundant data transmission. At the same time, the lightweight fusion of the improved D-S evidence theory ensures the real-time alignment and confidence integration of multi-modal data at the edge layer; combined with the dynamic load balancing mechanism (S200) of the three-level analysis architecture, the micro decision tree at the edge layer can respond to sudden deformation within 50ms. The false alarm rate of the fog layer LSTM network is reduced through time series correlation analysis, and the verification accuracy of the cloud mechanism model for multi-parameter coupling working conditions is improved; further through the dynamically updated edge intelligent model system (S300), the incremental update driven by federated learning shortens the response time of the model for working condition switching in the flood season / dry season, and the inference energy consumption of the slimmed model after channel pruning on the RK3399 edge device is reduced. The technical effects formed by the cooperation of the three are reflected in: in the 2 km² slope monitoring scenario, the overall energy consumption of the system is reduced by 76%, the abnormal detection delay is controlled within 200ms, and the prediction accuracy of the composite landslide event is increased by 41.8% compared with the traditional method.

[0045] Figure 6 The block diagram of an exemplary electronic device suitable for use in implementing the embodiments of the present invention is shown.

[0046] The electronic device may include a central processing unit / microprocessor / master control chip, etc.; a storage medium, coupled to the central processing unit / microprocessor / master control chip, etc., and storing computer-executable instructions therein for performing the steps of the various methods of the embodiments of the present invention when executed by the processor.

[0047] The central processing unit / microprocessor / master control chip, etc. may include, but are not limited to, for example, one or more processors or microprocessors, etc.

[0048] The storage medium may include, but is not limited to, for example, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, computer storage media (such as hard disks, floppy disks, solid-state drives, removable disks, CD-ROMs, DVD-ROMs, Blu-ray discs, etc.).

[0049] In addition, the electronic device may further include (but is not limited to) a data bus, an input / output bus / external bus / device bus, etc., a display, and input / output devices (such as a keyboard, a mouse, a speaker, etc.).

[0050] The central processing unit / microprocessor / master control chip, etc. may communicate with external devices via an I / O bus through a wired or wireless network (not shown).

[0051] The storage medium may also store at least one computer-executable instruction for performing the various functions and / or method steps in the embodiments described in the present technology when run by the central processing unit / microprocessor / master control chip, etc.

[0052] In one embodiment, the at least one computer-executable instruction may also be compiled into or constitute a software product, where when one or more computer-executable instructions are run by a processor, the various functions and / or method steps in the embodiments described in the present technology are performed.

[0053] Figure 7 A schematic diagram of a computer-readable storage medium according to an embodiment of the present invention is shown.

[0054] As Figure 7 shown, instructions are stored on a non-transitory computer-readable storage medium, and the instructions are, for example, computer-readable instructions. When the computer-readable instructions are run by a processor, the various methods described above may be performed. The non-transitory computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may, for example, include random access memory (RAM) and / or cache memory, etc. Non-transitory non-volatile memory may, for example, include read-only memory (ROM), hard disks, flash memory, etc. For example, the non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device runs the computer-readable instructions stored on the non-transitory computer-readable storage medium, the various methods described above may be performed.

[0055] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be in electrical, mechanical or other forms.

[0056] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0057] In addition, each functional unit in various embodiments of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0058] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (English full name: Read-Only Memory, English abbreviation: ROM), random access memories (English full name: Random Access Memory, English abbreviation: RAM), magnetic disks or optical disks and other various media that can store program codes.

[0059] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of various embodiments of the present invention.

Claims

1. An intelligent analysis system for slope stability with multi-sensor fusion, characterized in that, It includes: An environmental perception subsystem, which is used to monitor the change of soil water content by using a soil moisture sensor, obtain the movement data of the slope through a displacement sensor, and sense the change of ground pressure by using a pressure sensor; meanwhile, pre-screen abnormal data to form a fusion data set; A data analysis subsystem, which is used to convert the abnormal data in the fusion data set into a quantitative index of slope stability, and infer the potential sliding risk of the slope by analyzing the correlation between soil water content, movement data and ground pressure; A decision support subsystem, which is used to propose specific reinforcement countermeasures according to the analysis results of potential sliding risks.

2. The intelligent analysis system for slope stability with multi-sensor fusion according to claim 1, characterized in that, The environmental perception subsystem includes: A multi-dimensional perception module, which is used to obtain a three-dimensional model of the slope to be detected, determine the layout coordinates of the soil moisture sensor, displacement sensor and pressure sensor according to the historical slope landslide risk; form a slope data detection network composed of the soil moisture sensor, displacement sensor and pressure sensor; An abnormal pre-screening module, which is used to initialize the slope data detection network and compare the fluctuation trends of slope data within the same time window; set dynamic thresholds according to the physical coupling law of humidity-displacement-pressure; obtain the data of abnormal soil water content, small movement data and ground pressure in the slope data detection network according to the preset fluctuation trend standard and dynamic threshold; A data fusion module, which is used to receive the data of abnormal soil water content, small movement data and ground pressure, and form a multi-dimensional associated fusion data set including marks with the layout coordinates of the soil moisture sensor, displacement sensor and pressure sensor as marks.

3. The intelligent analysis system for slope stability with multi-sensor fusion according to claim 2, characterized in that, The multi-dimensional perception module includes: A three-dimensional model construction sub-module, which is used to obtain the surface geometric data of the slope to be detected, generate a three-dimensional model through point cloud processing and surface reconstruction, including geometric features such as elevation, slope and curvature of the terrain, and calibrate the potential risk area through historical landslide data including the position of the slip surface and the failure range to form a digital slope model with geological risk marks; A risk weight calculation sub-module, which is used to calculate the risk weights of different areas of the slope to be detected based on the digital slope model, combined with the spatial distribution, failure depth and triggering factors of the position and failure range of the slip surface; A sensor coordinate optimization sub-module, which is used to determine the mechanical coupling relationship between soil moisture, displacement and pressure, and set the layout conditions of the corresponding sensors; Through iterative optimization, determine the spatial matching coordinates of the three types of sensors.

4. The intelligent analysis system for slope stability with multi-sensor fusion according to claim 2, characterized in that, The abnormal pre-screening module includes: A multi-source data spatio-temporal alignment sub-module, which is used to unify the spatio-temporal benchmarks of the three types of data of soil water content, displacement and ground pressure based on the constructed slope data monitoring network; A coupling feature extraction sub-module, which is used to establish an association analysis framework for three types of parameters of humidity-displacement association, displacement-pressure association and humidity-pressure association according to the physical properties of the rock and soil mass of the slope to be detected; A dynamic threshold generation sub-module, which is used to adopt a sliding time window mechanism to construct an adaptive discrimination criterion for a reference threshold, trend correction, and spatial correction; implement a three-level discrimination mechanism for single-point anomalies, local anomalies, and system anomalies; and perform structured processing on the confirmed abnormal data by adding spatio-temporal markers, attaching confidence levels, and recording the evolution process.

5. The intelligent analysis system for slope stability with multi-sensor fusion according to claim 4, characterized in that, Anomaly pre-screening module, including: The time dimension of the multi-source data spatio-temporal alignment sub-module is based on the data acquisition timestamp to establish a millisecond-level synchronization mechanism; the spatial dimension uses the sensor layout coordinate information to construct a three-dimensional spatial reference system; the data dimension converts the original data with different dimensions into standard monitoring indicators. The humidity-displacement correlation analysis of the coupled feature extraction sub-module for the soil deformation response caused by seepage action. The displacement-pressure correlation for monitoring the stress redistribution caused by soil sliding. The humidity-pressure correlation for evaluating the influence of pore water pressure on effective stress. The reference threshold of the dynamic threshold generation sub-module establishes an initial discrimination criterion based on the data in the historical stable period. The trend correction adjusts the threshold sensitivity according to the recent data change rate. The spatial correction is set differently according to the risk weights of the sensor locations; single-point anomaly: the data of a single sensor exceeds the dynamic threshold range; local anomaly: adjacent sensor groups show collaborative anomalies; system anomaly: three types of parameters present an abnormal combination that conforms to the failure mechanism.

6. The intelligent analysis system for slope stability with multi-sensor fusion according to claim 5, characterized in that, Coupled feature extraction sub-module, including: A parameter correlation model construction unit, which is used to gradually establish an association framework for three types of parameters based on the physical properties of the slope rock and soil body, and the coupled feature extraction sub-module adopts a progressive analysis strategy. An association relationship logic confirmation unit, which is used to set the analysis logic for humidity-displacement correlation, displacement-pressure correlation, and humidity-pressure correlation. Anomaly coupling verification unit, which is used to cross-verify the results of the three types of association analyses; if the abnormal signal holds in multiple association models, it is marked as a high-confidence anomaly.

7. The intelligent analysis system for slope stability with multi-sensor fusion according to claim 1, characterized in that, Data analysis subsystem, including: A logic startup module, which is used to receive abnormal data, start the corresponding analysis logics for humidity-displacement correlation, displacement-pressure correlation, and humidity-pressure correlation, and screen the abnormal data according to the corresponding input content. A coupled analysis module, which is used to perform multi-parameter coupling and risk quantification of humidity, displacement, and pressure. A collaborative verification module, which is used for multi-parameter collaborative verification and risk grading, and checks whether the anomalies of the three types of parameters meet the typical combination patterns of slope instability; according to the number, coupling strength, and spatial distribution range of the abnormal parameters, the risk level is divided.

8. The intelligent analysis system for slope stability with multi-sensor fusion according to claim 7, characterized in that, Local low risk of the collaborative verification module: single-parameter anomaly; local high risk: two-parameter coupling anomaly; systematic risk: all three types of parameters are abnormal and conform to the failure mechanism.

9. The intelligent analysis system for slope stability with multi-sensor fusion according to claim 1, characterized in that, Decision support subsystem, including: A risk feature analysis and classification module, which is used to receive the risk assessment results from the data analysis subsystem, including: risk type marking, risk level division, and key parameter abnormal characteristics, and automatically construct a risk feature matrix. A targeted measure matching and optimization module, which is used to start a hierarchical response strategy according to the risk feature matrix. A space adaptation and intensity calibration module, which is used to adjust the coverage density of engineering measures according to the spatial distribution density of abnormal parameters, determine the design strength grade of structures based on the quantified risk level value, and plan the implementation priority of measures considering the temporal characteristics of parameter anomalies.

10. A method for edge computing of slope stability with multi-sensor fusion of the intelligent analysis system for slope stability with multi-sensor fusion according to any one of claims 1 to 9, characterized in that, The edge computing method for slope stability based on multi-sensor fusion includes the following steps: The edge computing method for slope stability based on multi-sensor fusion contains the following steps: Deploy a micro computing unit at the edge nodes of the sensor network to achieve in-situ feature extraction, dynamic data compression, cross-modal alignment, and lightweight fusion; Construct a three-level analysis architecture of rapid response at the edge layer, correlation analysis at the fog layer, and in-depth verification at the cloud; Establish an edge intelligent model with dynamic update, including cloud model slimming, incremental learning, cross-node knowledge sharing, and working condition adaptation.

Images