A dangerous rock mass real-time instability early warning method and system based on multi-source monitoring data

Abstract

The present application relates to the field of highway safety technology, and particularly relates to a dangerous rock mass real-time instability early warning method and system based on multi-source monitoring data. The method comprises the following steps: analyzing a dynamic rock mass integrity coefficient according to dangerous rock mass geological structure data, and extracting surface displacement response characteristics through dangerous rock mass surface displacement monitoring data; obtaining a first stability evaluation index by fusing a dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient and the surface displacement response characteristics using DS evidence theory; obtaining a three-dimensional stress concentration distribution coefficient by analyzing a stress state of the dangerous rock mass, and obtaining a second stability evaluation index by combining the three-dimensional stress concentration distribution coefficient and a dynamic rainfall infiltration coefficient; and completing dangerous rock mass real-time instability early warning through the first stability evaluation index, the second stability evaluation index and a dynamic traffic vibration influence coefficient. The present application provides accurate basis for dangerous rock mass disaster early warning by constructing a multi-source data fusion real-time early warning model.

Description

Technical Field

[0001] This invention relates to the field of highway safety technology, specifically to a method and system for real-time early warning of unstable rock masses based on multi-source monitoring data. Background Technology

[0002] Instability of unstable rock masses on highway slopes is a significant hidden danger leading to highway traffic disruptions and casualties. Its instability process is influenced by a combination of factors, including geological conditions, environmental factors, and human disturbance, exhibiting complexity, dynamism, and scenario dependence. Existing rock mass instability early warning technologies have the following shortcomings:

[0003] Poor scenario adaptability: Traditional early warning models use fixed weights to allocate features (such as rainfall, vibration, and stability), which cannot adapt to different scenarios such as "normal operating conditions", "rainfall-dominated", and "vibration-dominated".

[0004] Missing parameters for real-world scenarios: The existing model does not consider real-world factors such as long-term weathering and decay (reduction in surface rock strength) and short-term continuous rainfall accumulation (nonlinear increase in pore water pressure) unique to unstable rock masses on highway slopes. For short-term heavy rainfall, the seepage-stress coupling theory is not introduced, resulting in the calculated value of pore water pressure being lower than the actual value, and the stability coefficient calculation being seriously out of touch with engineering practice.

[0005] Insufficient risk correction mechanism: Traditional methods ignore the hysteresis effect of crack seepage during rainfall infiltration, resulting in stress-rainfall coupling risk calculations that are ahead of the actual mechanical response.

[0006] To address the aforementioned issues, there is an urgent need for a method for early warning of unstable rock mass risks that can integrate features from multiple scenarios, adapt to actual engineering parameters, and is computationally simple and highly accurate. Summary of the Invention

[0007] To address the shortcomings of existing methods and the needs of practical applications, and to solve the aforementioned problems, this invention provides a real-time instability early warning method for unstable rock masses based on multi-source monitoring data. The method includes the following steps: analyzing the dynamic rock mass integrity coefficient based on the geological structure data of the unstable rock mass; extracting surface displacement response characteristics from the surface displacement monitoring data of the unstable rock mass; fusing the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics using DS evidence theory to obtain a first stability assessment index; analyzing the stress state of the unstable rock mass to obtain a three-dimensional stress concentration distribution coefficient; combining the three-dimensional stress concentration distribution coefficient and the dynamic rainfall infiltration coefficient to obtain a second stability assessment index; and using the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration influence coefficient to complete the real-time instability early warning of the unstable rock mass.

[0008] This invention breaks down information barriers by integrating multi-source data. Through the linkage of data on geological structure, displacement, temperature, stress, rainfall, and traffic vibration, it avoids risk misjudgment caused by single data and comprehensively covers the dimensions of internal characteristics and external influencing factors of unstable rock masses. DS evidence theory enables reliable fusion, effectively handling the uncertainties between temperature and thermal stress, rock mass integrity, and displacement characteristics, improving the credibility of the first stability indicator, and resolving conflicts when fusing multiple types of data. A dual-stability indicator focuses on a layered approach: the first indicator anchors the basic stability of the rock mass (integrity, displacement, thermal stress), while the second indicator relates to external risk (stress, rainfall). This layered assessment better aligns with the mechanism of internal deterioration and external triggering of unstable rock masses. Real-time dynamic early warning enhances disaster prevention timeliness. Based on real-time monitoring data, indicators and early warning results are dynamically updated, significantly shortening response time compared to traditional offline analysis. This provides accurate evidence for early warning and timely handling of unstable rock mass disasters, effectively improving the scientific nature and timeliness of geological disaster prevention and control, and ensuring the safety of surrounding projects (such as highways and buildings) and the safety of people's lives and property.

[0009] Optionally, the analysis of the dynamic rock mass integrity coefficient based on the geological structure data of the unstable rock mass includes the following steps:

[0010] The lithology-joint weight coefficient is set based on the joint density; the dynamic rock mass integrity coefficient is analyzed by using the fractal dimension of the joint network and the lithology-joint weight coefficient.

[0011] This invention sets a lithology-joint weighting coefficient based on joint density, which can differentiate the contribution of inherent lithological strength and joint development to rock mass integrity, avoiding assessment bias caused by a single factor. The second step combines the fractal dimension of the joint network (reflecting joint complexity) with the weighting coefficient to capture in real-time the impact of dynamic changes such as joint expansion and new fracture formation on rock mass integrity. The overall method provides accurate basic mechanical parameters for subsequent stability analysis and instability risk warning of unstable rock masses, supporting real-time monitoring and safety decision-making in projects such as highway slopes, and reducing the risk of engineering accidents caused by inaccurate rock mass integrity assessments.

[0012] Optionally, the step of extracting surface displacement response features from the surface displacement monitoring data of the unstable rock mass includes the following steps:

[0013] The gain coefficient of the Kalman filter is dynamically updated based on the adaptive noise covariance adjustment mechanism to optimize the surface displacement monitoring data of the unstable rock mass; the dynamic displacement trend coefficient is extracted based on the optimized data.

[0014] By dynamically updating the gain coefficient of the Kalman filter through adaptive noise covariance adjustment, it adapts to complex noise environments such as multipath effects and traffic interference in highway scenarios, effectively optimizing displacement monitoring data, eliminating invalid interference, and improving data accuracy. Based on the optimized data, dynamic displacement trend coefficients are extracted, which can accurately capture the minute displacement changes and evolution patterns on the surface of unstable rock masses, providing high-quality data support for subsequent stability analysis and instability risk early warning, helping to identify early signs of instability, avoiding misjudgment or omission, and ensuring the operational safety of highway slopes.

[0015] Optionally, obtaining the dynamic temperature thermal stress coefficient includes the following steps:

[0016] Based on the rock mass depth and temperature gradient, the depth temperature gradient correction factor is analyzed; combined with the depth temperature gradient correction factor and the physical properties of the rock mass, the dynamic temperature thermal stress coefficient is obtained.

[0017] By introducing a depth temperature gradient correction factor, the limitations of traditional calculations that ignore the temperature difference at depth in the rock mass are overcome. By combining the physical properties of the rock mass, the calculation of the dynamic temperature thermal stress coefficient is made to fit the actual scenario of exposed surface and closed interior of dangerous rock masses on highways, and the interlayer thermal stress is accurately quantified. Finally, reliable thermal stress parameters are provided for the stability analysis and instability risk warning of dangerous rock masses, which enhances the engineering adaptability and warning accuracy of subsequent warning models and avoids risk misjudgment caused by the distortion of thermal stress calculation.

[0018] Optionally, the step of using DS evidence theory to integrate the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics to obtain the first stability assessment index includes the following steps:

[0019] Basic probability assignment functions are set for the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics, respectively; a first fusion model is constructed based on the DS evidence theory and the basic probability assignment functions, and a first stability evaluation index is obtained through the first fusion model.

[0020] The basic probability assignment function can quantify the contribution weight of each feature to stability, avoiding subjective bias; the first fusion model realizes the effective integration of multi-source heterogeneous features, breaks the limitations of single feature evaluation, and allows the first stability evaluation index to comprehensively reflect the synergistic effects of thermal stress, rock mass structure, and displacement response; the final output index is more comprehensive and reliable, providing an accurate basis for subsequent stability analysis of dangerous rock masses, reducing evaluation errors caused by the one-sidedness of single features, and improving the scientific nature and engineering reference value of stability evaluation.

[0021] Optionally, the analysis of the stress state of the unstable rock mass to obtain the three-dimensional stress concentration distribution coefficient includes the following steps:

[0022] Based on the stress state of the unstable rock mass, a stress field is constructed by Kriging interpolation; the stress concentration factor is calculated based on the interpolated data to obtain the three-dimensional stress concentration distribution factor.

[0023] By using Kriging interpolation to fill the spatial gaps in stress monitoring, a continuous and accurate three-dimensional stress field is constructed, overcoming the limitations of discrete single-point monitoring data that cannot reflect the spatial distribution of stress. Based on this, the stress concentration factor is calculated, transforming the abstract stress state into a quantified three-dimensional distribution index, which can accurately locate high-stress risk areas and avoid blind spots in planar analysis.

[0024] Optionally, extracting the dynamic rainfall infiltration coefficient includes the following steps:

[0025] The permeability coefficient is dynamically corrected based on the joint opening and cumulative rainfall; the dynamic rainfall infiltration coefficient is extracted by combining the corrected permeability coefficient and the surface condition correction coefficient.

[0026] The dynamic infiltration coefficient is matched with the joint evolution and surface effects under rainfall, which is more suitable for highway slope scenarios. The accurate dynamic rainfall infiltration coefficient provides reliable parameters for subsequent instability risk early warning, improving the accuracy and engineering practicality of highway unstable rock mass monitoring and early warning.

[0027] Optionally, obtaining the second stability assessment index by combining the three-dimensional stress concentration distribution coefficient and the dynamic rainfall infiltration coefficient includes the following steps:

[0028] The effective stress is dynamically corrected based on pore water pressure; the second stability assessment index is calculated by combining the corrected effective stress, dynamic rainfall infiltration coefficient, and fracture seepage hysteresis correction coefficient.

[0029] A dynamic rainfall infiltration coefficient is introduced to reflect the impact of real-time changes in rainfall intensity and duration on infiltration, avoiding static errors caused by fixed parameters. A crack seepage lag correction coefficient is superimposed to account for the time difference in rainwater transport within cracks, preventing risk assessment from being premature or delayed. The overall system achieves coupled calculation of three-dimensional stress and rainfall infiltration, making the second stability assessment index more closely reflect the actual working conditions of unstable rock masses on highway slopes. This provides accurate and dynamic intermediate parameter support for subsequent instability risk early warning, enhancing the scientific rigor and engineering guidance value of stability assessment.

[0030] Optionally, the step of completing the real-time instability warning of the unstable rock mass through the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration influence coefficient includes the following steps:

[0031] Based on the second stability assessment index and the dynamic traffic vibration impact coefficient, a scene identification factor is determined; a real-time instability risk index model for unstable rock mass is constructed by combining the scene identification factor, the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration impact coefficient, and a real-time instability warning for unstable rock mass is achieved through the real-time instability risk index model for unstable rock mass.

[0032] By using the second stability assessment index and the dynamic traffic vibration impact coefficient as the basis for determining scene identification factors, different risk conditions of unstable rock masses can be accurately classified, avoiding scene adaptation bias caused by fixed weights. Furthermore, by integrating the first stability assessment index to construct a risk index model, key factors such as foundation stability, risk superposition, and disturbance impact can be considered, achieving multi-dimensional feature synergistic quantification. This improves both the real-time nature and accuracy of instability early warning, providing a scientific basis for targeted treatment of unstable rock masses on highway slopes and effectively reducing traffic safety risks caused by instability.

[0033] Secondly, to efficiently execute the real-time rock mass instability early warning method based on multi-source monitoring data provided by this invention, this invention also provides a real-time rock mass instability early warning system based on multi-source monitoring data, including a processor, an input device, an output device, and a memory. The processor, input device, output device, and memory are interconnected. The memory stores a computer program containing program instructions. The processor is configured to call the program instructions to execute the real-time rock mass instability early warning method based on multi-source monitoring data as described in the first aspect of this invention. The real-time rock mass instability early warning system based on multi-source monitoring data of this invention has a compact structure and stable performance, and can stably execute the real-time rock mass instability early warning method based on multi-source monitoring data provided by this invention, further improving the overall applicability and practical application capability of this invention. Attached Figure Description

[0034] Figure 1 This is a flowchart of a real-time instability early warning method for unstable rock masses based on multi-source monitoring data, provided by an embodiment of the present invention.

[0035] Figure 2 This is a framework diagram of a real-time instability early warning system for unstable rock masses based on multi-source monitoring data, provided in an embodiment of the present invention. Detailed Implementation

[0036] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.

[0037] Throughout this specification, references to "an embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "in an embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0038] Please see Figure 1 To address the aforementioned problems, this invention provides a real-time early warning method for unstable rock masses based on multi-source monitoring data, such as... Figure 1 As shown, in one embodiment, the method includes the following steps:

[0039] S1. Analyze the dynamic rock mass integrity coefficient based on the geological structure data of the unstable rock mass, and extract the surface displacement response characteristics through the surface displacement monitoring data of the unstable rock mass.

[0040] In this embodiment, the analysis of dynamic rock mass integrity coefficient based on geological structure data of unstable rock masses includes the following steps:

[0041] S111. Set the lithology-joint weight coefficient based on joint density.

[0042] The degree of joint development is a key factor affecting the mechanical properties of rock mass. The ratio of real-time joint density to initial joint density is introduced as a core indicator, which intuitively reflects the development and changes of joints over time or due to engineering disturbances.

[0043] Specifically, the lithology-joint weighting coefficient is set based on the joint density. ,satisfy:

[0044]

[0045] in, This represents the measured joint density of the unstable rock mass at time t. This indicates the initial joint density of the unstable rock mass.

[0046] When the real-time joint density increases by more than 20% compared to the initial state, it indicates that the joint network has developed significantly (such as crack propagation and the generation of new cracks). At this point, the influence of joints on the integrity of the rock mass exceeds that of the lithology itself, so joints are given a higher weight, reflecting the geological law that joints dominate stability.

[0047] This means that the degree of joint development is in a stable or slightly changing state, and the inherent strength of the lithology (such as uniaxial compressive strength) plays a dominant role in the integrity of the rock mass. Therefore, the lithology is given a higher weight, which is consistent with the mechanical property that the lithology controls the stability of the foundation.

[0048] S112. Analyze the dynamic rock mass integrity coefficient using the fractal dimension of the joint network and the lithology-joint weight coefficient.

[0049] In this embodiment, the dynamic rock mass integrity coefficient is analyzed using the fractal dimension of the joint network and the lithology-joint weight coefficient. It satisfies the following formula:

[0050]

[0051] in, This represents the initial value of rock mass integrity, serving as a quantitative indicator of rock mass integrity in its initial state. Let represent the fractal dimension of the joint network at time t. The box-counting method can accurately quantify the geometric complexity of the joint network. These parameters represent lithological strength and directly reflect the mechanical properties of rocks in resisting axial compressive failure. It represents the uniaxial compressive strength of a standard rock sample, used to eliminate the scale effect caused by lithological differences.

[0052] Introducing fractal dimension to dynamically model the joint evolution process, fractal theory, through the quantitative index of fractal dimension, can effectively characterize the irregular geometric morphology of joint networks. An increase in the fractal dimension indicates an increase in the complexity of the joint network, leading to intensified rock mass fracturing, and consequently... A decrease reflects a reduction in the integrity and bearing capacity of the unstable rock mass; while an increase in lithological strength parameters will... An increase in the coefficient indicates an enhancement in the inherent bearing capacity of the unstable rock mass. This dynamic coefficient, by coupling joint evolution with lithological characteristics, can quantify the dynamic changes in the bearing capacity of the unstable rock mass in real time, providing a theoretical basis for subsequent stability assessment and instability early warning.

[0053] Furthermore, the extraction of surface displacement response features from the surface displacement monitoring data of the unstable rock mass includes the following steps:

[0054] S121. The gain coefficient of the Kalman filter is dynamically updated based on the adaptive noise covariance adjustment mechanism to optimize the surface displacement monitoring data of the unstable rock mass.

[0055] The surface displacement monitoring data of the unstable rock mass adopts a dual-modal sensing scheme, including real-time GNSS data and InSAR time-series data. The GNSS module is based on the BeiDou-3 system, with a high-frequency sampling capability of 1Hz and a positioning accuracy of ±2mm, which is suitable for capturing instantaneous displacement. The InSAR technology, through synthetic aperture radar interferometry, can achieve sub-millimeter-level deformation monitoring, effectively compensating for the monitoring blind spots of GNSS in obstructed environments.

[0056] In this embodiment, based on the adaptive noise covariance adjustment mechanism, the gain coefficient is dynamically updated to effectively suppress noise interference such as GNSS multipath effect and ionospheric delay. Data optimization is achieved through a dual process of state prediction and measurement update, satisfying the following:

[0057]

[0058] in, This represents the filtered displacement at time k. This indicates the dynamically adjusted filter gain. This represents real-time GNSS data. This represents the covariance of the estimation error at time k-1. This represents the variance of the measurement noise.

[0059] Furthermore, initialization is based on the nominal accuracy of the GNSS equipment being ±2mm. To ensure initial filter stability; real-time residual feedback is introduced, using the measured value at the current time. and predicted value residual Dynamically update R:

[0060] when hour, If the initial noise standard deviation is identified as sudden strong noise (such as multipath effect caused by heavy vehicles), the R value is automatically increased (to 1.5-2 times the initial value), and the filter gain is reduced. This reduces the interference of abnormal measurement values ​​on the results;

[0061] when When the noise level is stable, the R value is reduced (to 0.6-0.8 times the initial value), and the [other values] are increased. This enhances the sensitivity to actual displacement changes.

[0062] S122. Extract the dynamic displacement trend coefficient based on the optimized data.

[0063] In this embodiment, an LSTM network with 50 time-step input layers is constructed, and 128 neurons with attention mechanisms are deployed in the hidden layers to learn the nonlinear evolution law through historical displacement sequences. The network outputs the predicted displacement value at time t. Based on this, the dynamic displacement trend coefficient is calculated. ,satisfy:

[0064]

[0065] in, This indicates that a 1-hour sliding time window is set to match the high-frequency monitoring needs of highway scenarios. , making Using mm / h as the unit to intuitively represent the rate of displacement change, it can effectively quantify the creep speed of unstable rock masses, and its positive and negative values ​​and the gradient of change directly reflect the level of instability risk.

[0066] Furthermore, dynamic early warning thresholds can be constructed based on the principles of statistical process control. ,satisfy:

[0067]

[0068] in, This represents the average trend coefficient over the previous 24 hours. This corresponds to the standard deviation. This threshold is dynamically updated according to environmental conditions, adapting to different geological conditions and climatic factors, effectively reducing the false alarm rate.

[0069] By quantifying the rate of displacement change, a dynamic correlation between displacement evolution and instability risk is established, especially with the ability to identify the continuous acceleration phase in advance. Then utilize The principle defines the boundary of abnormal fluctuations. When the real-time trend coefficient exceeds the threshold, a three-level early warning mechanism can be triggered to provide a basis for decision-making in highway disaster prevention and mitigation.

[0070] S2. Using the DS evidence theory, the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics are integrated to obtain the first stability assessment index.

[0071] In one optional embodiment, obtaining the dynamic temperature thermal stress coefficient includes the following steps:

[0072] First, based on the rock mass depth and temperature gradient, the depth-temperature gradient correction factor is analyzed.

[0073] Temperature sensors were installed at intervals of 50-100 meters along the surface and key internal nodes of the dangerous rock masses on both sides of the highway to collect real-time temperature monitoring data T(t) (unit: °C) along the highway, with a sampling frequency of once per 10 minutes. This data sequence fully records the dynamic changes in the surface temperature of the rock mass over time, including information such as diurnal temperature difference and seasonal temperature fluctuations, providing basic data support for subsequent analysis.

[0074] The surface of the unstable rock mass is significantly affected by diurnal temperature variations and solar radiation, while internal temperature changes are slower and smaller. This temperature difference along the depth direction generates interlayer thermal stress, exacerbating the risk of joint propagation. Based on the rock mass depth and temperature gradient, a depth temperature gradient correction factor is analyzed. It satisfies the following formula:

[0075]

[0076] in, express Temperature difference between depth and surface Indicates the thermal conductivity of the rock mass. This indicates the depth difference between the temperature monitoring points. This represents the average temperature of the unstable rock mass at time t.

[0077] Secondly, the dynamic temperature thermal stress coefficient is obtained by combining the depth temperature gradient correction factor and the physical properties of the rock mass.

[0078] The dynamic temperature thermal stress coefficient is obtained by combining the depth temperature gradient correction factor and the physical properties of the rock mass. This refers to accurately capturing the dynamic impact of temperature fluctuations on rock masses by quantifying the rate of temperature change per unit time, thus more realistically reflecting the actual mechanical state of unstable rock masses under temperature influence, and satisfying the following:

[0079]

[0080] in, The coefficient of thermal expansion of rock mass reflects its volumetric expansion or contraction characteristics as temperature changes. This represents the elastic modulus of the rock mass. This represents the initial elastic modulus. Setting the time interval to 1 hour as the calculation time interval for the temperature change rate can both avoid noise interference introduced by too short an interval and effectively capture the temperature change trend. This represents the Poisson's ratio of the rock mass.

[0081] The dynamic temperature thermal stress coefficient establishes a quantitative relationship between temperature change and rock mass thermal stress. When the temperature change rate... An increase in temperature indicates that the rock mass has undergone a significant thermal expansion and contraction process in a short period of time, resulting in an increase in thermal stress, which in turn leads to… The value is rising, and the high value is... This indicates that the joint surfaces inside the rock mass may open or close due to thermal stress, thereby altering the stability of the rock mass structure. This is further supported by the elastic modulus E and Poisson's ratio. The correction can dynamically and accurately reflect the real-time impact of thermal stress on rock mass structure, providing key characteristic indicators for early warning of unstable rock masses.

[0082] In this embodiment, the step of using DS evidence theory to integrate the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics to obtain the first stability assessment index includes the following steps:

[0083] S21. Set basic probability allocation functions for the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics, respectively.

[0084] For each feature, a basic probability assignment function is constructed, and its physical meaning is transformed into a probabilistic form for stability judgment.

[0085] Specifically, for This value is directly taken as the probability of the unstable rock mass being in a "stable" state, i.e. The probability of an "unstable" state is This mapping method is based on the positive correlation between rock mass integrity and stability; the more intact the rock mass, the higher the probability of it being in a stable state.

[0086] for , , , This refers to the displacement change rate threshold set based on highway engineering experience and real-time environmental conditions. Exceed At that time, the probability of unstable rock mass will increase with the increase of displacement change rate.

[0087] for , , , A critical thermal stress value of 0.5 MPa / h is taken. If the rate of change of thermal stress exceeds this critical value, the probability of the unstable rock mass due to temperature factors will increase significantly.

[0088] A critical thermal stress value of 0.5 MPa / h is taken. If the rate of change of thermal stress exceeds this critical value, the probability of the unstable rock mass due to temperature factors will increase significantly.

[0089] S22. Construct a first fusion model based on the DS evidence theory and the basic probability allocation function, and obtain a first stability evaluation index through the first fusion model.

[0090] After calculating the basic probability allocation functions for each feature, a first fusion model is constructed based on the DS evidence theory and the basic probability allocation functions, satisfying:

[0091]

[0092] This formula effectively corrects potential conflicts of evidence between different features. For example, when The geological conditions appear stable, but When a sudden change occurs, the conflict correction mechanism reallocates weights based on the reliability of each piece of evidence, preventing a single outlier from excessively interfering with the final result. The resulting first stability assessment index... The value range is [0,1]. The closer the value is to 1, the higher the stability of the unstable rock mass at that moment; conversely, the closer it is to 0, the greater the risk of instability.

[0093] This invention uses the DS evidence theory as its core framework, fully leveraging its unique advantages in handling uncertain information. Specifically... , and By constructing a conflict resolution mechanism, three types of data with different physical properties and dimensions are integrated and processed to resolve the uncertain information they contain. This transforms multi-source, heterogeneous monitoring data that may contain logical contradictions into a quantitative assessment index of rock mass stability with physical meaning. This effectively overcomes the drawbacks of information bias and provides a more rigorous and comprehensive theoretical basis and quantitative analysis method for real-time instability early warning of rock masses in highway engineering scenarios.

[0094] S3. Analyze the stress state of the unstable rock mass to obtain the three-dimensional stress concentration distribution coefficient. Combine the three-dimensional stress concentration distribution coefficient and the dynamic rainfall infiltration coefficient to obtain the second stability assessment index.

[0095] In this embodiment, the step of analyzing the stress state of the unstable rock mass to obtain the three-dimensional stress concentration distribution coefficient includes the following steps:

[0096] S311. Based on the stress state of the unstable rock mass, a stress field is constructed by Kriging interpolation.

[0097] The internal stress monitoring data of the unstable rock mass was collected using a high-precision fiber optic grating sensor, denoted as Within the monitoring area of ​​the unstable rock mass, sensors were arranged with a grid spacing of 3m×3m×5m, and a total of 20 monitoring points were set up.

[0098] Kriging spatial interpolation technology is used to construct a continuous stress field from discrete monitoring point data, thereby reflecting the overall stress distribution of the unstable rock mass. Specifically, 20 key monitoring points form a basic data network. These monitoring points collect stress data on the surface of the unstable rock mass in real time through stress sensors, covering key structural surfaces and vulnerable areas of the unstable rock mass. Through the Kriging interpolation algorithm, the discrete stress monitoring data is transformed into a continuous three-dimensional stress field cloud map, which can not only intuitively present the spatial gradient changes of stress distribution, but also effectively identify stress concentration areas, providing a quantitative basis for the risk assessment of unstable rock mass instability.

[0099] S312. Calculate the stress concentration factor based on the interpolated data to obtain the three-dimensional stress concentration distribution factor.

[0100] In this embodiment, the stress concentration factor is calculated based on the interpolated data to obtain the three-dimensional stress concentration distribution factor, which satisfies:

[0101]

[0102] in, Indicates the stress concentration factor. This represents the allowable stress of the rock mass at time t. Indicates the ultimate compressive strength of the rock mass. This represents the interpolated stress at spatial coordinates (x, y, z) in the time dimension t.

[0103] when When the stress concentration exceeds the allowable stress value of the rock mass under its current condition, it indicates that the stress in that area exceeds the allowable stress value of the rock mass. Under such circumstances, the unstable rock mass is highly susceptible to crack propagation in that area, which may lead to instability. The three-dimensional stress concentration distribution coefficient can intuitively present the stress concentration distribution within the entire spatial range of the unstable rock mass, helping to accurately locate the core area of ​​stress concentration. This provides a key basis for the rapid identification of the instability source of the unstable rock mass, so that corresponding protective measures can be taken in a timely manner to ensure the safety of highway traffic.

[0104] Further, the dynamic rainfall infiltration coefficient is extracted, including the following steps:

[0105] S321. Dynamically adjust the permeability coefficient based on joint opening and cumulative rainfall.

[0106] Specifically, the permeability coefficient is dynamically adjusted based on the joint opening and cumulative rainfall. ,satisfy:

[0107]

[0108] in, This represents the initial value of the rock mass permeability coefficient, determined based on previous geological surveys and tests. It reflects the permeability of the rock mass in its initial state. This represents the joint opening at time t. The initial joint opening is represented by a basic parameter obtained from geological exploration. This represents the cumulative rainfall at time t, reflecting the cumulative effect of rainfall over time. The greater the cumulative rainfall, the more significant its impact on the permeability of the rock mass. Indicates the intensity of radar rainfall.

[0109] S322. Combine the corrected permeability coefficient and the surface condition correction coefficient to extract the dynamic rainfall infiltration coefficient.

[0110] Along highways, unstable rock masses often have vegetation such as weeds and shrubs on their surfaces (intercepting some rainwater), and are mostly distributed in slope areas (the steeper the slope, the greater the rainwater runoff loss). Both of these factors contribute to the actual amount of rainfall infiltrating into the rock mass being less than the theoretically calculated value. A surface condition correction coefficient is fitted based on the vegetation cover on the unstable rock mass surface and the slope gradient of the slope where the unstable rock mass is located. This makes the calculation of the effective infiltration coefficient more consistent with the actual surface conditions in highway scenarios.

[0111] Furthermore, by combining the corrected permeability coefficient and the surface condition correction coefficient, the dynamic rainfall infiltration coefficient is extracted. It satisfies the following formula:

[0112]

[0113] in, Indicates the duration of rainfall. This represents the average thickness of the unstable rock mass. The magnitude of the dynamic rainfall infiltration coefficient directly characterizes the degree to which rainfall infiltration weakens the rock mass's strength. Specifically, the deeper the rainfall infiltration, the greater the weakening effect on the rock mass's strength. The higher the value, the lower the shear strength of the rock mass. During rainfall, rainwater infiltration increases the pore water pressure inside the rock mass and softens the joint filling material, thereby reducing the stability of the unstable rock mass. The dynamic rainfall infiltration coefficient can quantify this weakening effect. Combining rainfall factors with rock mass mechanical properties is beneficial for achieving accurate early warning of the risk of unstable rock mass.

[0114] The process of obtaining a second stability assessment index by combining the three-dimensional stress concentration distribution coefficient and the dynamic rainfall infiltration coefficient includes the following steps:

[0115] S331. Dynamically correct the effective stress based on pore water pressure.

[0116] Specifically, the effective stress is dynamically corrected based on the pore water pressure to satisfy:

[0117]

[0118] in, This represents the corrected effective stress. This indicates pore water pressure.

[0119] S332. Calculate the second stability assessment index by combining the corrected effective stress, dynamic rainfall infiltration coefficient, and fracture seepage hysteresis correction coefficient.

[0120] In highway scenarios, the more complex the fracture network of unstable rock masses (higher joint fractal dimension) and the faster the seepage rate (larger dynamic permeability coefficient), the more tortuous the transport path of rainwater within the fractures, and the more significant the time lag from surface infiltration to the generation of pore water pressure in the stress concentration zone. Introducing a fracture seepage lag correction coefficient makes the stability coefficient more closely resemble the actual mechanical response process. ,satisfy:

[0121]

[0122] Furthermore, by combining the corrected effective stress, dynamic rainfall infiltration coefficient, and fracture seepage hysteresis correction coefficient, a second stability assessment index is calculated. It satisfies the following formula:

[0123]

[0124] S4. Real-time instability warning of unstable rock mass is completed by using the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration influence coefficient.

[0125] The method of achieving real-time instability early warning for unstable rock masses through the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration impact coefficient includes the following steps:

[0126] S41. Determine the scene identification factor based on the second stability assessment index and the dynamic traffic vibration influence coefficient.

[0127] In highway traffic scenarios, the instability of unstable rock masses is closely related to dynamic traffic loads. A short-time Fourier transform combined with the Hanning window function is employed. For the original vibration signal Frequency domain analysis is performed to satisfy:

[0128]

[0129] in, express Time Frequency The power spectral density at a given point quantifies the energy distribution of different frequency components; the Hanning window function effectively suppresses spectral leakage, ensuring the accuracy of the analysis results; the time window T is set to 10s, balancing time resolution and spectral stability.

[0130] Furthermore, the dynamic traffic vibration influence coefficient is calculated. ,satisfy:

[0131]

[0132] in, Indicates the resonant frequency of the rock mass. This represents the mean power spectral density of background vibration (statistical value during periods without vehicles). Indicates standard vehicle load. This represents vehicle load data. When the dominant frequency of traffic vibration is close to... hour, It will increase significantly, leading to The rapid rise directly reflects the risk of increased fracture propagation due to forced vibration of the rock mass. Simultaneously, this is combined with real-time vehicle load data. The dynamic correction makes the assessment results closer to the disturbance characteristics under actual highway operating conditions, providing a scientific basis for early warning of unstable rock masses.

[0133] Furthermore, based on the second stability assessment index and the dynamic traffic vibration impact coefficient, the scene identification factor is determined to satisfy:

[0134]

[0135] in, Represents normal operating conditions and meets the following requirements: and ;

[0136] Representing the dominant rainfall-related operating condition, satisfying: and ;

[0137] This represents the dominant vibration condition, satisfying the following: and .

[0138] S42. Construct a real-time instability risk index model for unstable rock mass by combining the scene identification factor, the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration influence coefficient. Complete the real-time instability warning of unstable rock mass through the real-time instability risk index model for unstable rock mass.

[0139] Specifically, dimensional differences are eliminated through critical threshold normalization to ensure... They participate in calculations on the same order of magnitude.

[0140] Furthermore, based on the scene recognition factor, weights are assigned to the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration impact coefficient, satisfying:

[0141] For normal operating conditions , , ;

[0142] For rainfall-dominated operating conditions , , ;

[0143] For vibration-dominated operating conditions , , ;

[0144] Furthermore, a real-time instability risk index model for unstable rock masses is constructed by combining the scene identification factor, the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration influence coefficient, satisfying the following:

[0145]

[0146] in, They are The result after normalization.

[0147] Furthermore, based on experience in highway rock mass monitoring projects, The alert level is divided into five levels, and specific response measures are defined.

[0148] Please see Figure 2 In this embodiment, to efficiently execute the real-time rock mass instability early warning method based on multi-source monitoring data provided by the present invention, the present invention also provides a real-time rock mass instability early warning system based on multi-source monitoring data, comprising: an input device, an output device, a processor, and a memory, wherein the input device, output device, processor, and memory are interconnected, and the memory contains program instructions for the steps of the real-time rock mass instability early warning method based on multi-source monitoring data. The real-time rock mass instability early warning system based on multi-source monitoring data of the present invention has a compact structure and stable performance, and can stably execute the real-time rock mass instability early warning method based on multi-source monitoring data of the present invention, further improving the overall applicability and practical application capability of the present invention.

[0149] In this embodiment, the processor may be a central processing unit, but it can also be other general-purpose processors, digital signal processors, application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (OPGs), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor. Input devices can be used to acquire data. Output devices can be used to output the results obtained by storing program instructions contained in a computer program in the memory provided by this invention. The memory may include read-only memory and random access memory (RAM), and provides instructions and data to the processor. A portion of the memory may also include non-volatile random access memory (RAM).

[0150] In one possible implementation, the memory may include a stored program area and a stored data area. The stored program area may store the operating system and applications required for at least one function; the stored data area may store data created during use. Furthermore, the memory may include read-only memory and random access memory, and provides instructions and data to the processor. The memory stores the operating system and operating instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof. The operating instructions may include various operation instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and handling hardware-based tasks.

[0151] The embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for real-time early warning of unstable rock masses based on multi-source monitoring data.

[0152] The storage medium can include various media that can store program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0153] In summary, this invention utilizes multi-source data integration to break down information barriers. By linking data on geological structure, displacement, temperature, stress, rainfall, and traffic vibration, it avoids misjudgments caused by single data points and comprehensively covers the dimensions of internal characteristics and external influencing factors of unstable rock masses. The DS evidence theory enables reliable fusion, effectively handling uncertainties between temperature and thermal stress, rock mass integrity, and displacement characteristics, enhancing the credibility of the first stability indicator, and resolving conflicts when fusing multiple types of data. A dual-stability indicator system focuses on layered approaches: the first indicator anchors the basic stability of the rock mass (integrity, displacement, thermal stress), while the second indicator correlates with external risk factors (stress, rainfall). This layered assessment better aligns with the mechanism of internal deterioration and external triggering of unstable rock masses. Real-time dynamic early warning enhances disaster prevention timeliness. Based on real-time monitoring data, indicators and early warning results are dynamically updated, significantly shortening response time compared to traditional offline analysis. This provides accurate evidence for early warning and timely handling of unstable rock mass disasters, effectively improving the scientific nature and timeliness of geological disaster prevention and control, and ensuring the safety of surrounding projects (such as highways and buildings) and the safety of people's lives and property.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the present invention.

Claims

1. A method for real-time early warning of unstable rock mass based on multi-source monitoring data, characterized in that, Includes the following steps: The dynamic rock mass integrity coefficient is analyzed based on the geological structure data of the unstable rock mass, and the surface displacement response characteristics are extracted from the surface displacement monitoring data of the unstable rock mass. The first stability assessment index is obtained by integrating the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics using the DS evidence theory. The stress state of the unstable rock mass is analyzed to obtain the three-dimensional stress concentration distribution coefficient. Combined with the three-dimensional stress concentration distribution coefficient and the dynamic rainfall infiltration coefficient, a second stability assessment index is obtained. Real-time instability warning of unstable rock mass is achieved by using the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration impact coefficient. Extracting the dynamic rainfall infiltration coefficient includes the following steps: The permeability coefficient is dynamically adjusted based on joint opening and cumulative rainfall. The dynamic rainfall infiltration coefficient is extracted by combining the corrected permeability coefficient and the surface condition correction coefficient. The process of obtaining a second stability assessment index by combining the three-dimensional stress concentration distribution coefficient and the dynamic rainfall infiltration coefficient includes the following steps: Effective stress is dynamically corrected based on pore water pressure. The second stability assessment index is calculated by combining the corrected effective stress, dynamic rainfall infiltration coefficient and fracture seepage hysteresis correction coefficient. The method of completing real-time instability early warning of unstable rock masses through the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration influence coefficient includes the following steps: The scene identification factor is determined based on the second stability assessment index and the dynamic traffic vibration impact coefficient. A real-time instability risk index model for unstable rock mass is constructed by combining the scene identification factor, the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration impact coefficient. The real-time instability risk index model for unstable rock mass is used to complete the real-time instability warning of unstable rock mass. The real-time instability risk index model for the unstable rock mass satisfies: in, This indicates the real-time instability risk index of the unstable rock mass. They are The result after normalization This represents the second stability assessment index. Indicates the dynamic traffic vibration impact coefficient. This represents the first stability assessment indicator. This indicates the weight of the second stability assessment index. This indicates the weight of the dynamic traffic vibration impact coefficient. This indicates the weight of the first stability assessment indicator; Based on the second stability assessment index and the dynamic traffic vibration impact coefficient, the scene identification factor is determined to satisfy: in, Represents normal operating conditions and meets the following requirements: and ; Representing the dominant rainfall-related operating condition, satisfying: and ; This represents the dominant vibration condition, satisfying the following: and ; Based on the scene recognition factor, weights are assigned to the first stability assessment index, the second stability assessment index, and the dynamic traffic vibration impact coefficient, satisfying the following: For normal operating conditions , , ; For rainfall-dominated operating conditions , , ; For vibration-dominated operating conditions , , .

2. The method for real-time instability early warning of unstable rock masses based on multi-source monitoring data according to claim 1, characterized in that, The analysis of dynamic rock mass integrity coefficient based on geological structure data of unstable rock masses includes the following steps: The lithology-joint weighting coefficient is set based on the joint density; The dynamic rock mass integrity coefficient is analyzed by using the fractal dimension of the joint network and the lithology-joint weight coefficient.

3. The method for real-time instability early warning of unstable rock masses based on multi-source monitoring data according to claim 1, characterized in that, The extraction of surface displacement response features from unstable rock mass surface displacement monitoring data includes the following steps: The gain coefficient of the Kalman filter is dynamically updated based on the adaptive noise covariance adjustment mechanism to optimize the surface displacement monitoring data of the unstable rock mass. Dynamic displacement trend coefficients are extracted from the optimized data.

4. The method for real-time instability early warning of unstable rock masses based on multi-source monitoring data according to claim 1, characterized in that, Obtaining the dynamic temperature thermal stress coefficient includes the following steps: Analyze the depth-temperature gradient correction factor based on the rock mass depth and temperature gradient; The dynamic temperature thermal stress coefficient is obtained by combining the depth temperature gradient correction factor and the physical properties of the rock mass.

5. The method for real-time instability early warning of unstable rock masses based on multi-source monitoring data according to claim 1, characterized in that, The method of using DS evidence theory to integrate the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics to obtain the first stability assessment index includes the following steps: Basic probability assignment functions are set for the dynamic temperature thermal stress coefficient, the dynamic rock mass integrity coefficient, and the surface displacement response characteristics, respectively. A first fusion model is constructed based on the DS evidence theory and the basic probability allocation function, and a first stability evaluation index is obtained through the first fusion model.

6. The method for real-time instability early warning of unstable rock masses based on multi-source monitoring data according to claim 1, characterized in that, The analysis of the stress state of the unstable rock mass to obtain the three-dimensional stress concentration distribution coefficient includes the following steps: Based on the stress state of the unstable rock mass, a stress field is constructed by Kriging interpolation. The stress concentration factor is calculated based on the interpolated data to obtain the three-dimensional stress concentration distribution factor.

7. A real-time instability early warning system for unstable rock masses based on multi-source monitoring data, characterized in that, The real-time instability early warning system for unstable rock masses based on multi-source monitoring data includes: an input device, an output device, a processor, and a memory. The input device, output device, processor, and memory are interconnected. The memory includes program instructions, which are used to execute the real-time instability early warning method for unstable rock masses based on multi-source monitoring data according to any one of claims 1-6.

Images