Rockfall risk early warning method and system based on static-dynamic-environmental quantity fusion

Abstract

The present application provides a dangerous rock risk early warning method and system based on static-dynamic-environmental quantity fusion, relates to the technical field of geological disaster monitoring and early warning, and the method comprises the following steps: obtaining rock mass monitoring data and environmental monitoring data of a dangerous rock to be measured, wherein the rock mass monitoring data comprises a rock mass vibration time domain signal and rock mass static parameters; extracting a dynamic index of the rock to be measured according to the rock mass vibration time domain signal; constructing a dangerous rock risk fine evaluation model based on a deep learning algorithm according to the dynamic index or the rock mass static parameters; inputting each environmental monitoring data into the dangerous rock risk fine evaluation model; and finely evaluating and early warning the dangerous rock risk. The present application not only considers single physical index monitoring and no longer relies on experience threshold for stability discrimination, but also fully considers the comprehensive action of rock mass dynamic response characteristics and multi-source environmental information, so that high-precision and early instability early warning can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster monitoring and early warning, and in particular to a dangerous rock risk early warning method and system based on the fusion of static-dynamic-environmental quantities. Background Art

[0002] With the continuous expansion of mountain engineering and water conservancy and hydropower development, the construction of highways and railways in mountainous areas has created high and steep slopes. The repeated rise and fall of reservoir water levels has also created drawdown slopes with high water level differences. Coupled with the impact of extreme weather conditions such as rainstorms and floods, as well as human activities, these slopes are exposed to a complex multi-factor synergistic environment, resulting in the formation of high, steep, and dangerous rock masses. Due to the changing hydro-meteorological conditions at the surface and underground, these dangerous rock masses are constantly exposed to a complex hydraulic-mechanical coupling environment, significantly increasing the risk of instability in these high, steep, and dangerous rock masses. These natural disasters pose a serious threat to Yangtze River navigation safety and to the lives and property of the people.

[0003] In the existing technology, the traditional dangerous rock risk warning method only considers a single physical indicator for monitoring or only relies on empirical thresholds for stability judgment, ignoring the combined effect of the dynamic response characteristics of the dangerous rock mass and multi-source environmental information, and is difficult to achieve high-precision and early instability warning. The present invention provides a dangerous rock risk warning based on the fusion of static-dynamic-environmental quantities. Summary of the Invention

[0004] In order to solve the technical problem that the traditional dangerous rock risk warning method in the existing technology only considers a single physical indicator for monitoring or only relies on empirical thresholds for stability judgment, ignores the dynamic response characteristics of the dangerous rock mass and the comprehensive effect of multi-source environmental information, and is difficult to achieve high-precision and early instability warning, the present invention provides a dangerous rock risk warning method and system based on the fusion of static-dynamic-environmental quantities.

[0005] The technical solutions provided by the embodiments of the present invention are as follows:

[0006] First aspect:

[0007] The dangerous rock risk early warning method based on the fusion of static, dynamic and environmental quantities provided by the embodiment of the present invention includes:

[0008] S1: Acquire multiple rock mass monitoring data and multiple environmental monitoring data of the rock mass to be measured, wherein the rock mass monitoring data includes rock mass vibration time domain signals and rock mass static parameters;

[0009] S2: Extract multiple dynamic indicators of the rock mass to be tested based on the rock mass vibration time domain signal;

[0010] S3: Based on various dynamic indicators or rock mass static parameters, a preliminary assessment of the risk of dangerous rock is conducted to preliminarily determine whether the rock mass to be tested has dangerous rock risk; if so, proceed to S4; otherwise, return to S1;

[0011] S4: Construct a detailed assessment model of dangerous rock risks based on deep learning algorithms;

[0012] S5: Input various environmental monitoring data into the dangerous rock risk fine assessment model, conduct a fine assessment of the dangerous rock risk, and finely judge whether the rock mass to be tested has dangerous rock risk; if so, issue an early warning signal; otherwise, return to S1.

[0013] Second aspect:

[0014] The dangerous rock risk early warning system based on the fusion of static, dynamic and environmental quantities provided by the embodiment of the present invention includes:

[0015] processor;

[0016] A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities as described in the first aspect is implemented.

[0017] The third aspect:

[0018] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities as described in the first aspect is implemented.

[0019] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0020] In an embodiment of the present invention, a plurality of dynamic indicators of the rock mass to be tested are extracted, and a preliminary assessment of the dangerous rock risk is performed based on the various dynamic indicators and the static parameters of the rock mass. The various environmental monitoring data are input into a fine assessment model of the dangerous rock risk to perform a fine assessment of the dangerous rock risk. This not only considers a single physical indicator for monitoring and no longer relies on empirical thresholds for stability judgment, but also fully considers the comprehensive effect of the dynamic response characteristics of the rock mass and multi-source environmental information, and can achieve high-precision and early instability warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 A schematic flow chart of a dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities provided in an embodiment of the present invention;

[0023] Figure 2 A schematic structural diagram of a dangerous rock risk warning system based on the fusion of static-dynamic-environmental quantities provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0025] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0026] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0027] In the embodiment of the present invention, sometimes a subscript such as W1 may be written as a non-subscript form such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0028] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0029] Reference Manual Figure 1 , which shows a flow chart of a dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities provided by an embodiment of the present invention.

[0030] An embodiment of the present invention provides a dangerous rock risk warning method based on the fusion of static, dynamic, and environmental quantities. This method can be implemented by a dangerous rock risk warning device based on the fusion of static, dynamic, and environmental quantities. The dangerous rock risk warning device based on the fusion of static, dynamic, and environmental quantities can be a terminal or a server. The processing flow of the dangerous rock risk warning method based on the fusion of static, dynamic, and environmental quantities can include the following steps:

[0031] S1: Acquire multiple rock mass monitoring data and multiple environmental monitoring data of the rock mass to be measured, wherein the rock mass monitoring data includes rock mass vibration time domain signals and rock mass static parameters.

[0032] It should be noted that environmental monitoring data refers to external environmental information related to dangerous rock collapse collected by various sensors deployed around the rock mass. This data can reveal the hydrological and meteorological conditions of the environment in which the dangerous rock mass is located and is an external driving factor for the occurrence of geological disasters. The rock mass vibration time domain signal refers to the original time series data collected by a laser Doppler vibrometer that reflects the vibration response of the rock mass under external disturbances. It is usually in the form of a continuous change curve of the rock mass surface velocity or displacement within a certain time period and can be used to reflect the dynamic characteristics and potential anomalies of the rock mass structure. The static parameters of the rock mass refer to the shear performance indicators of the rock mass obtained through triaxial shear tests. They are used to characterize the ability of the rock mass to resist slippage or rupture under the action of force and are the basic basis for rock mass stability analysis and critical state judgment.

[0033] Optionally, the environmental monitoring data includes: rainfall, temperature and fracture water pressure.

[0034] It should be noted that rainfall refers to the depth of precipitation falling per unit area over a specific time period and is an important external driver of regional hydrological processes and rock collapse mechanisms. Temperature refers to the temperature at the surface and deep within the rock mass, directly reflecting the impact of seasonal changes and diurnal temperature differences on thermal expansion and contraction, or frost heave, of the rock mass. Fracture water pressure refers to the water pressure within the cracks within the rock mass and is used to reveal the evolution of deep infiltration, water storage, and fracture aperture.

[0035] Rock mass static parameters include: rock mass cohesion and rock mass internal friction angle.

[0036] It should be noted that rock cohesion refers to the ability of a rock mass to maintain its integrity in the absence of external normal stress and is one of the fundamental parameters for measuring rock mass shear strength. The internal friction angle of a rock mass refers to the angle between the resultant normal stress and internal friction on the shear plane and the normal stress when the rock mass reaches shear failure limit equilibrium. It reflects the magnitude of friction between rock particles and is a core parameter in rock mass shear strength models, such as the Mohr-Coulomb criterion.

[0037] Dynamic indicators include: time domain indicators, frequency domain indicators, energy indicators and modal indicators.

[0038] It should be noted that time domain indicators refer to parameters that describe signal strength and impact extracted directly from the rock mass vibration time domain signal (i.e., the original waveform). They are used to determine the severity of the vibration response and the characteristics of transient instability. Frequency domain indicators refer to frequency distribution characteristic parameters calculated by performing a fast Fourier transform on the vibration signal to obtain the spectrum. They are used to reflect the concentration and change trend of the rock mass vibration energy in the frequency dimension. Energy indicators refer to parameters related to mechanical energy release estimated from the vibration signal. They are used to assess the degree of local fracture or energy accumulation in the rock mass. Modal indicators refer to parameters that describe the damping and energy dissipation capacity of the rock mass or its structural surface during vibration. They are used to indirectly assess the expansion of structural cracks and the level of stiffness degradation.

[0039] Time domain indicators include: pulse indicators and margin indicators.

[0040] It should be noted that the impulse index refers to the ratio of the peak value of the rock mass vibration time-domain signal to the effective value (RMS) of the signal. It is used to measure the strength of the impact component in the signal. A larger value indicates more intense transient vibration, which may indicate a precursor to instability. The margin index refers to the ratio of the peak value of the rock mass vibration to the root mean square amplitude of the signal. It reflects the concentration of fluctuations in the signal. A smaller value tends to indicate poor system stability.

[0041] Frequency domain indicators include: center of gravity frequency and root mean square frequency.

[0042] It should be noted that the center of gravity frequency refers to the average center frequency of the energy distribution in the rock mass vibration spectrum. Its decrease often indicates weakening of rock mass stiffness or increased damage. The root mean square frequency refers to the root mean square value of the frequencies in the spectrum. It comprehensively reflects the distribution characteristics of vibration energy in the high and low frequency ranges. Its decrease often indicates that the structure is tending to become unstable.

[0043] Energy indicators include: impact energy and relative energy of the first frequency band.

[0044] It should be noted that impact energy refers to the kinetic energy released per unit mass of rock due to instantaneous high-amplitude vibration during rock mass vibration. It is used to determine the extent of local crack initiation or development. The relative energy in the first frequency band refers to the ratio of low-frequency energy to total energy in the vibration spectrum. An increase in this ratio generally indicates increased rock damage and stronger low-frequency response.

[0045] Modal indicators include: damping ratio.

[0046] It should be noted that the damping ratio refers to the ability of rock structures or joints to dissipate vibration energy. It is a function of the frequency ratio of the damped system to the undamped system and is used to evaluate the degree of energy attenuation caused by cracks or weak surfaces in the rock mass.

[0047] In an embodiment of the present invention, by acquiring monitoring data of the rock mass to be tested and environmental monitoring data, high-quality input features are provided for the subsequent dangerous rock risk identification algorithm, which helps to improve the early warning model's ability to identify instability precursors and overall prediction accuracy, and enhances the system's deployability, interpretability and engineering practicality.

[0048] In a possible implementation, S1 specifically includes sub-steps S101 to S103:

[0049] S101: Obtain rock static parameters of the rock mass to be tested through triaxial shear test.

[0050] It should be noted that the triaxial shear test refers to a standard mechanical test method used to determine the shear parameters of rock materials. By applying axial load to cylindrical rock specimens under controlled confining pressure conditions, its failure behavior under actual ground stress states is simulated, thereby obtaining the strength and failure characteristics of the rock mass. It is one of the basic tests commonly used in rock mechanics analysis and rock stability assessment.

[0051] S102: Obtaining a rock mass vibration time domain signal of the rock mass to be measured by a laser Doppler vibrometer.

[0052] It should be noted that the laser Doppler vibrometer refers to a non-contact, high-precision vibration measurement device based on the principle of the laser Doppler effect. It emits a laser beam to the surface of the object being measured and receives its reflected light, and uses the optical frequency shift to calculate the instantaneous vibration velocity or displacement of the target surface. It is often used to obtain the dynamic response of rock masses or structures under external disturbances and is an important means of extracting dynamic indicators.

[0053] S103: Obtain environmental monitoring data through the rainfall sensor, temperature sensor and water pressure sensor.

[0054] Specifically, rainfall sensors are used to measure rainfall, temperature sensors are used to measure the temperature of the rock surface and deep rock layers, and water pressure sensors are used to measure the water pressure in the cracks within the rock mass, i.e., the fracture water pressure.

[0055] It should be noted that a rainfall sensor is a meteorological measurement device used to monitor and record atmospheric precipitation in real time. It converts the precipitation depth per unit time into an electrical signal output. A temperature sensor is a sensor device used to measure the temperature of rock masses. It uses an inserted probe to sense changes in the rock's dielectric constant and thus infer the rock's temperature. A water pressure sensor is a sensor device used to measure the water pressure within cracks or structural surfaces within rock and soil. It is used to reflect changes in the fissure water pressure caused by water-rock coupling within the rock mass.

[0056] In an embodiment of the present invention, rock static parameters, vibration response data and environmental inducements are obtained through triaxial shear tests, laser Doppler vibrometers and multiple sensors, respectively. This not only achieves a systematic perception of the three elements of disaster formation, but also provides a structured, high-quality data basis for the construction and interpretation of subsequent early warning models.

[0057] S2: Extract multiple dynamic indicators of the rock mass to be tested based on the rock mass vibration time domain signal.

[0058] In this embodiment of the present invention, by extracting structured dynamic indicators, effective modeling and risk identification of the dynamic response of complex rock masses are achieved. This process not only enhances the model's expressive power and predictive accuracy, but also provides a solid data foundation for building an interpretable early warning mechanism and an engineering-implementable disaster monitoring system.

[0059] In a possible implementation, S2 specifically includes sub-steps S201 to S205:

[0060] S201: Constructing a vibration waveform diagram based on the rock mass vibration time domain signal.

[0061] S202: Convert the rock mass vibration time domain signal into a rock mass vibration frequency domain signal through a fast Fourier transform algorithm.

[0062] It should be noted that the Fast Fourier Transform (FFT) algorithm is an efficient computational method for converting time-domain signals into frequency-domain signals. It can quickly decompose the signal's frequency components, thereby revealing the signal's periodicity, dominant frequency, and energy distribution. This algorithm is widely used in fields such as vibration analysis, acoustic wave testing, and rock mass dynamic response feature extraction, and is a core tool in signal spectrum analysis.

[0063] Specifically, the Fast Fourier Transform (FFT) algorithm is used to discretize the collected rock vibration time-domain signal, representing it as a finite-length time series. This series is then calculated using the FFT algorithm, expanding the energy components contained in the time-domain signal along the frequency axis to obtain the amplitude distribution of the rock vibration at different frequencies, i.e., the frequency-domain signal.

[0064] S203: Constructing a vibration spectrum diagram based on the rock mass vibration frequency domain signal.

[0065] S204: Extracting pulse index, margin index and impact energy according to the vibration waveform.

[0066] Optionally, the calculation formula of the pulse indicator is specifically as follows:

[0067]

[0068]

[0069] in, I f Represents the pulse indicator, X p represents the peak value of the rock mass vibration time domain signal, X rms represents the effective value of the rock mass vibration time domain signal, N represents the total number of sampling points, X i Indicates the i The rock mass vibration time domain signal of the sampling points.

[0070] Optionally, the calculation formula of the margin index is specifically:

[0071]

[0072]

[0073] in, CL f represents the margin index, X r represents the root square amplitude of the rock vibration time domain signal, and | | represents the absolute value.

[0074] Optionally, the calculation formula of the impact energy is specifically:

[0075]

[0076]

[0077] in, E i Indicates the i The impact energy of each sampling point, c represents the conversion factor, β represents the kurtosis index, E t Represents the total vibration energy of the rock mass to be tested, m Indicates the quality of the rock mass to be tested, X av represents the mean value of the rock mass vibration time domain signal, D X Represents the variance of the rock mass vibration time domain signal.

[0078] S205: Extracting the center of gravity frequency, root mean square frequency, relative energy of the first frequency band, and damping ratio based on the vibration spectrum, thereby completing the extraction of multiple dynamic indicators of the rock mass to be measured.

[0079] Optionally, the calculation formula of the center of gravity frequency is specifically:

[0080]

[0081] in, f avg represents the center of gravity frequency, f Indicates frequency, p ( f ) represents the frequency domain signal of rock vibration at frequency.

[0082] Optionally, the calculation formula of the root mean square frequency is specifically:

[0083]

[0084] in, f b represents the rms frequency.

[0085] Optionally, the calculation formula of the relative energy of the first frequency band is specifically:

[0086]

[0087] in, E r1 Indicates the relative energy of the first frequency band, B f Indicates the upper frequency limit of the first frequency band, F s Indicates full-band bandwidth.

[0088] Optionally, the calculation formula of the damping ratio is specifically:

[0089]

[0090] in, x represents the damping ratio, f d Indicates damped vibration frequency.

[0091] In an embodiment of the present invention, by processing the rock vibration time domain signal according to steps S201 to S205, constructing a vibration waveform diagram and generating a frequency domain signal and a spectrum diagram using the fast Fourier transform algorithm, and further extracting a number of dynamic indicators including pulse index, margin index, impact energy, center of gravity frequency, root mean square frequency, relative energy of the first frequency band and damping ratio, it is possible to achieve efficient conversion from the original unstructured signal to structured and quantitative features, which not only effectively improves the data availability and model input quality, but also covers multiple physical dimensions of the rock dynamic response, which helps to accurately identify instability precursors such as vibration anomalies, stiffness degradation and energy accumulation before disasters occur.

[0092] S3: Based on various dynamic indicators or rock mass static parameters, a preliminary assessment of the rock mass risk is performed to determine whether the rock mass to be tested is at risk. If so, proceed to S4. Otherwise, return to S1.

[0093] It should be noted that the preliminary assessment of dangerous rock risks includes two methods: preliminary assessment of dangerous rock risks based on various dynamic indicators and preliminary assessment of dangerous rock risks based on rock static parameters. In actual application, technical personnel in this field can choose one of them to conduct preliminary assessment of dangerous rock risks according to actual needs, or can use both methods to conduct preliminary assessment of dangerous rock risks at the same time. The present invention does not limit this.

[0094] In this embodiment of the present invention, by dividing the initial assessment of dangerous rock risks into two complementary pathways, one based on dynamic indicators and the other on rock mass static parameters, the comprehensiveness and accuracy of risk identification are enhanced while also providing good data adaptability and engineering practicality. This dual-channel mechanism not only improves the system's adaptability to various application scenarios but also effectively enhances the model's interpretability, the efficiency of the initial screening phase, and the stability of the overall early warning system, providing a high-quality input foundation for the subsequent refined identification phase.

[0095] In a possible implementation, performing a preliminary assessment of dangerous rock risks based on various dynamic indicators in S3 specifically includes sub-steps S301A to S303A:

[0096] S301A: Normalize various kinetic indicators.

[0097] S302A: Construct a preliminary assessment model for dangerous rock risks based on support vector machine algorithm.

[0098] It's important to note that the support vector machine algorithm is a machine learning method commonly used for classification and regression analysis. Its core concept is to construct an optimal hyperplane in the feature space to separate data samples of different categories with the greatest possible separation. This algorithm has good generalization capabilities when dealing with small sample sizes, high dimensions, and nonlinear problems, making it suitable for scenarios requiring accurate determination of stability, such as rock risk identification.

[0099] S303A: Input the normalized dynamic indicators into the preliminary assessment model of dangerous rock risks to conduct a preliminary assessment of dangerous rock risks.

[0100] Optionally, the normalized dynamic indicators are input into the preliminary assessment model for dangerous rock risk according to the following formula to perform a preliminary assessment of dangerous rock risk:

[0101]

[0102] in, It represents the preliminary assessment result of dangerous rock risk output by the preliminary assessment model of dangerous rock risk, where When the initial assessment shows that the rock mass to be tested does not have dangerous rock risk, When the rock mass to be tested is preliminarily assessed to have a dangerous rock risk, sign represents the symbolic function, where hour, sign =+1, when hour, sign =-1, n represents the total number of support vectors, Indicates the Lagrange multipliers of support vectors, Indicates the The true category labels corresponding to the support vectors are K ( ) represents the kernel function, x represents the kinetic index after normalization of the input, Indicates the The kinetic index after normalization of the support vectors, b represents the bias term.

[0103] In an embodiment of the present invention, the normalized kinetic indicators are input into the support vector machine-based evaluation model, which not only improves the accuracy and robustness of preliminary risk identification, but also significantly enhances the system's ability to handle nonlinearity, fuzzy boundaries and small sample problems.

[0104] In a possible implementation, the preliminary assessment of dangerous rock risks based on rock mass static parameters in S3 specifically includes sub-steps S301B to S303B:

[0105] S301B: Calculate the critical slope height of the rock mass to be tested based on the rock mass cohesion and internal friction angle using the Kullmann model.

[0106] It should be noted that the Kullmann model is a simplified rock mass stability analysis method based on limit equilibrium theory. It is primarily used to assess the potential slip conditions of rock slopes under their own gravity. By geometrically constructing a sliding wedge and combining parameters such as rock mass cohesion, internal friction angle, and slope angle, the model derives the critical slope height or safety factor under the sliding limit state. This model is often used to identify potential rockfall sources and provide preliminary assessments of slope stability.

[0107] Alternatively, the critical slope height of the rock mass to be measured can be calculated according to the following formula:

[0108]

[0109] in, H c represents the critical slope height of the rock mass to be measured,C represents the rock mass cohesion, c represents the rock mass density, sin represents the sine function, cos represents the cosine function, d Indicates the actual slope angle, f It represents the friction angle in the rock mass.

[0110] S302B: When the actual slope height of the rock mass to be measured is greater than or equal to the critical slope height, it is preliminarily assessed that the rock mass to be measured has a dangerous rock risk.

[0111] S303B: When the actual slope height of the rock mass to be measured is less than the critical slope height, it is preliminarily assessed that there is no risk of dangerous rock in the rock mass to be measured.

[0112] In an embodiment of the present invention, the preliminary assessment of dangerous rock risks based on rock static parameters and the Kullmann model is incorporated into the overall early warning process, which not only builds a bridge between traditional physical criteria and modern intelligent identification, but also provides a preliminary risk assessment path with strong engineering feasibility, high interpretability and flexible deployment. It is an important component of supporting the practical application and engineering implementation of the system.

[0113] S4: Construct a detailed assessment model for dangerous rock risks based on deep learning algorithms.

[0114] Optionally, the deep learning algorithm is specifically: a temporal convolutional neural network.

[0115] It's important to note that the Temporal Convolutional Neural Network (TCN) is a deep learning model specifically designed for processing time series data. It employs a one-dimensional convolutional structure, extracting local temporal features from a sequence using sliding convolution kernels, and uses dilated convolutions to expand the receptive field to capture long-term dependencies. While maintaining the consistency of input and output sequence lengths, this network achieves deep feature extraction by stacking multiple layers of residual blocks. It is widely used in fields such as time series prediction, signal recognition, and geological disaster warning, boasting the advantages of high efficiency, stability, and parallel computing.

[0116] In an embodiment of the present invention, a refined assessment model based on a temporal convolutional neural network (TCN) is constructed. The long-term and short-term characteristics of landslide precursor signals are captured in parallel through dilation and causal convolution. The residual block is combined to ensure the stability of the deep network. The input / output attention mechanism is used to automatically focus on key timing channels, thereby achieving end-to-end dynamic risk probability prediction.

[0117] S5: Input the environmental monitoring data into the rock-risk assessment model to conduct a detailed assessment of the rock-risk risk and determine whether the rock mass under test is at risk. If so, issue a warning signal. Otherwise, return to S1.

[0118] In an embodiment of the present invention, by inputting multi-source environmental parameters and prior risk levels into a refined assessment model, not only is real-time dynamic, integrated and complementary, graded warning, and closed-loop adaptive full-process intelligent monitoring achieved, but also a standardized and scalable interface design is provided for engineering deployment, greatly improving the system's response speed, recognition accuracy, and operation and maintenance convenience.

[0119] In a possible implementation, the step of inputting the environmental monitoring data into the dangerous rock risk fine assessment model in S5 to perform fine assessment of dangerous rock risk specifically includes sub-steps S501 to S505:

[0120] S501: normalize each environmental monitoring data.

[0121] Optionally, each environmental monitoring data is normalized according to the following formula:

[0122]

[0123] in, Z j Represents the first j Environmental monitoring data, z j Indicates the j Environmental monitoring data, J Indicates the total number of environmental monitoring data.

[0124] In an embodiment of the present invention, the environmental monitoring data are normalized so that data of different dimensions, such as rainfall, temperature, and fissure water pressure, are all mapped to the same numerical range, eliminating the difference in characteristic scales. This not only ensures the numerical stability of model training, but also facilitates fair comparison of information from each channel by subsequent entropy weights and attention mechanisms.

[0125] S502: Calculate multiple weight coefficients of each normalized environmental monitoring data using an entropy weight method.

[0126] It should be noted that the entropy weighting method is an objective weighting method for determining the weights of multiple indicators. Its basic principle is to use information entropy to measure the degree of uncertainty of each indicator in the sample data. The smaller the entropy value, the more concentrated the information of the indicator, the stronger its discriminatory ability, and the larger the corresponding weight. This method does not require human intervention and can reflect the inherent differences in data characteristics. It is often used in multi-source information fusion scenarios such as comprehensive evaluation, risk analysis, and landslide early warning.

[0127] In an embodiment of the present invention, the entropy weight method is used to automatically calculate the weight coefficient of each normalized environmental indicator, and the uncertainty of data distribution is measured by information entropy. This not only eliminates the need for manual subjective weighting, but also highlights key driving parameters that change dramatically and have concentrated information, thereby improving the sensitivity and reliability of multi-source fusion.

[0128] In a possible implementation, S502 specifically includes sub-steps S5021 to S5023:

[0129] S5021: Calculate multiple entropy values ​​of each environmental monitoring data after normalization, where the entropy values ​​include: information entropy value, fuzzy entropy value, and permutation entropy value.

[0130] Optionally, the information entropy value is calculated as follows:

[0131]

[0132] in, f 1j Represents the first j The information entropy value of environmental monitoring data, ln represents the natural logarithm function, p ( Z j ) represents the first j The probability distribution of environmental monitoring data.

[0133] Optionally, the calculation formula of the fuzzy entropy value is specifically:

[0134]

[0135] in, f 2j Represents the first j The fuzzy entropy value of environmental monitoring data, m represents the embedding dimension, H ( ) represents the Heaviside step function, r represents the tolerance threshold, Represents the first j Environmental monitoring data and The distance between environmental monitoring data, Represents the first Environmental monitoring data.

[0136] Optionally, the calculation formula of the permutation entropy value is specifically:

[0137]

[0138] in, f3j Represents the first j The permutation entropy value of environmental monitoring data, Indicates the first j Environmental monitoring data k The probability of a pattern appearing.

[0139] S5022: Calculate multiple coefficients of variation of each environmental monitoring data after normalization based on each entropy value.

[0140] The coefficient of variation includes a first coefficient of variation, a second coefficient of variation, and a third coefficient of variation.

[0141] Optionally, multiple coefficients of variation of each environmental monitoring data after normalization are calculated based on each information entropy value:

[0142]

[0143] in, d 1j Represents the first j The first coefficient of variation of environmental monitoring data, d 2j Represents the first j The second coefficient of variation of environmental monitoring data, d 3j Represents the first j The third coefficient of variation of environmental monitoring data.

[0144] S5023: Calculate multiple weight coefficients of each environmental monitoring data after normalization based on each coefficient of variation.

[0145] The weight coefficients include a first weight coefficient, a second weight coefficient, and a third weight coefficient.

[0146] Optionally, multiple weight coefficients of each environmental monitoring data after normalization are calculated based on each coefficient of variation:

[0147]

[0148] in, oh 1j Represents the first j The first weight coefficient of environmental monitoring data, oh 2j Represents the first j The second weight coefficient of environmental monitoring data, oh 3j Represents the first jThe third weight coefficient of environmental monitoring data.

[0149] In this embodiment of the present invention, three methods, information entropy, fuzzy entropy, and permutation entropy, are first used to characterize the overall uncertainty, noise robustness, and mutation characteristics of environmental monitoring data from different perspectives. Each entropy value is then converted into a coefficient of variation to highlight parameters with high information content. Finally, through normalization, the coefficient of variation is converted into an objective weight, achieving dynamic and adaptive allocation of environmental parameter weights. This process not only ensures the complementarity and robustness of multi-source information but also provides stable, interpretable, and efficient input features for subsequent risk assessment models.

[0150] S503: Calculate multiple entropy weight scores of each normalized environmental monitoring data according to each weight coefficient.

[0151] The entropy weight score includes a first entropy weight score, a second entropy weight score, and a third entropy weight score.

[0152] Optionally, multiple entropy weight scores of each environmental monitoring data after normalization are calculated according to each weight coefficient:

[0153]

[0154] in, s 1j Represents the first j The first entropy weight score of environmental monitoring data, s 2j Represents the first j The second entropy weight score of environmental monitoring data, s 3j Represents the first j The third entropy weight score of environmental monitoring data.

[0155] In an embodiment of the present invention, the above-mentioned weight coefficients are used to perform weighted summation on the environmental parameters to generate several entropy weight score curves, thereby realizing dimensionality reduction and aggregation of multidimensional data. This not only retains the main features of each channel, but also greatly reduces the input dimension, simplifies the model calculation burden, and enhances the centralized expression of early warning signals.

[0156] S504: Calculate the dangerous rock risk index based on each entropy weight score through the attention mechanism.

[0157] In an embodiment of the present invention, multiple entropy weight scores are dynamically weighted and fused through the attention mechanism, which can automatically adjust the attention to different scores according to time series changes, effectively suppress the interference of abnormal fluctuations of a single indicator, and output a more stable and robust landslide risk degree (LHD), providing key prior information for detailed assessment.

[0158] In a possible implementation, S504 specifically includes sub-steps S5041 and S5042:

[0159] S5041: Concatenate the entropy weight scores to calculate the comprehensive entropy weight score.

[0160] Optionally, the individual entropy weight scores are concatenated to calculate a comprehensive entropy weight score according to the following formula:

[0161]

[0162]

[0163] in, Q represents the query matrix, K represents the bond matrix, V represents the value matrix, W represents the weight matrix, S represents the comprehensive entropy weight score, softmax ( ) represents the softmax activation function, T represents the transpose operation, d Indicates the dimensions of the bond matrix.

[0164] S5042: Normalize the comprehensive entropy weight score and calculate the dangerous rock risk index.

[0165] It should be noted that the dangerous rock risk index refers to a quantitative indicator obtained through comprehensive analysis of multiple influencing factors, which is used to measure the possibility of geological disasters (such as landslides and collapses) occurring in the rock mass to be tested.

[0166] Optionally, the comprehensive entropy weight score is normalized according to the following formula to calculate the dangerous rock risk index:

[0167]

[0168] in, LHD represents the dangerous rock risk index, Normalize ( ) indicates normalization processing.

[0169] In an embodiment of the present invention, by inputting multi-path entropy weight scores into the attention fusion module and generating a unified normalized landslide hazard index, it is possible to dynamically allocate the weights of each indicator according to real-time data, adaptively highlight key information, suppress invalid noise, and output it in a unified scale for subsequent deep model recognition and alarm decision-making, thereby significantly improving the accuracy and reliability of landslide risk assessment.

[0170] S505: Inputting the normalized environmental monitoring data and dangerous rock risk index into a dangerous rock risk detailed assessment model to perform a detailed assessment of dangerous rock risk.

[0171] Specifically, the normalized environmental monitoring parameters and the dangerous rock risk index are combined to form a time-series input feature matrix and fed into a temporal convolutional neural network model. The model extracts time-dependent features through causal and dilated convolutions, while incorporating a residual structure to ensure stable training of the deep network. Finally, a fully connected layer and activation function output the landslide risk probability. Based on a set decision threshold, the presence of a dangerous rock risk is determined (if the landslide risk probability exceeds the decision threshold, the rock mass is deemed to be at risk; otherwise, it is deemed to be at risk). This allows for a fine-scale classification of risk status.

[0172] In an embodiment of the present invention, the normalized environmental data and the calculated LHD sequence are jointly constructed into a time series feature matrix and input into a temporal convolutional neural network. The network automatically extracts multi-scale spatiotemporal dependency features through causal convolution and dilated convolution, and outputs the landslide risk probability, realizing online real-time, end-to-end high-precision graded warning.

[0173] Reference Manual Figure 2 , which shows a structural diagram of the dangerous rock risk warning system based on the fusion of static-dynamic-environmental quantities provided by the present invention.

[0174] The present invention also provides a dangerous rock risk warning system 20 based on the fusion of static-dynamic-environmental quantities, which is applied to the above-mentioned dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities, and includes:

[0175] Processor 201;

[0176] The memory 202 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 201, the dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities as described in the method embodiment is implemented.

[0177] The dangerous rock risk warning system 20 based on the fusion of static-dynamic-environmental quantities provided by the present invention can execute the above-mentioned dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities and achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0178] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0179] In an embodiment of the present invention, a plurality of dynamic indicators of the rock mass to be tested are extracted, and a preliminary assessment of the dangerous rock risk is performed based on the various dynamic indicators and the static parameters of the rock mass. The various environmental monitoring data are input into a fine assessment model of the dangerous rock risk to perform a fine assessment of the dangerous rock risk. This not only considers a single physical indicator for monitoring and no longer relies on empirical thresholds for stability judgment, but also fully considers the comprehensive effect of the dynamic response characteristics of the rock mass and multi-source environmental information, and can achieve high-precision and early instability warning.

[0180] It should be understood that the processor in the embodiments of the present invention may be a central processing unit (CPU), but may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or 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, etc.

[0181] It should also be understood that the memory in the embodiments of the present invention may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0182] The above embodiments can be implemented in whole or in part via software, hardware (e.g., circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0183] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0184] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0185] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0186] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0187] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0188] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0189] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0190] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0191] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion 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 enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical disks.

[0192] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities as described in the method embodiment is implemented.

[0193] The computer-readable storage medium provided by the present invention can realize the steps and effects of the dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities in the above method embodiment. To avoid repetition, the present invention will not go into details.

[0194] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0195] In an embodiment of the present invention, a plurality of dynamic indicators of the rock mass to be tested are extracted, and a preliminary assessment of the dangerous rock risk is performed based on the various dynamic indicators and the static parameters of the rock mass. The various environmental monitoring data are input into a fine assessment model of the dangerous rock risk to perform a fine assessment of the dangerous rock risk. This not only considers a single physical indicator for monitoring and no longer relies on empirical thresholds for stability judgment, but also fully considers the comprehensive effect of the dynamic response characteristics of the rock mass and multi-source environmental information, and can achieve high-precision and early instability warning.

[0196] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0197] There are a few points to note:

[0198] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0199] (2) For the sake of clarity, the thickness of layers or regions in the drawings used to describe the embodiments of the present invention are exaggerated or reduced, that is, these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element may be "directly on" or "under" the other element or intervening elements may be present.

[0200] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to form new embodiments.

[0201] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A dangerous rock risk warning method based on the fusion of static, dynamic and environmental quantities is characterized by: include: S1: Acquire multiple rock mass monitoring data and multiple environmental monitoring data of the rock mass to be measured, wherein the rock mass monitoring data includes rock mass vibration time domain signals and rock mass static parameters; S2: extracting multiple dynamic indicators of the rock mass to be measured based on the rock mass vibration time domain signal; S3: Based on the dynamic indicators or the static parameters of the rock mass, a preliminary assessment of the risk of dangerous rock is performed to preliminarily determine whether the rock mass to be tested has dangerous rock risk; if so, proceed to S4; otherwise, return to S1; S4: Construct a detailed assessment model of dangerous rock risks based on deep learning algorithms; S5: Input each of the environmental monitoring data into the dangerous rock risk fine assessment model, perform fine assessment of the dangerous rock risk, and finely judge whether the rock mass to be tested has dangerous rock risk; if so, issue an early warning signal; otherwise, return to S1.

2. The dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities according to claim 1 is characterized in that: The environmental monitoring data include: rainfall, temperature and fissure water pressure; The rock mass static parameters include: rock mass cohesion and rock mass internal friction angle; The dynamic indicators include: time domain indicators, frequency domain indicators, energy indicators and modal indicators; The time domain indicators include: pulse indicators and margin indicators; The frequency domain indicators include: center of gravity frequency and root mean square frequency; The energy indicators include: impact energy and relative energy of the first frequency band; The modal index includes: damping ratio.

3. The dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities according to claim 2 is characterized in that: Said S1 specifically includes: S101: Obtaining rock mass static parameters of the rock mass to be tested through a triaxial shear test; S102: Obtaining a rock mass vibration time domain signal of the rock mass to be measured by a laser Doppler vibrometer; S103: Acquire the environmental monitoring data through a rainfall sensor, a temperature sensor, and a water pressure sensor.

4. The dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities according to claim 2 is characterized in that: The S2 specifically includes: S201: constructing a vibration waveform diagram based on the rock mass vibration time domain signal; S202: converting the rock mass vibration time domain signal into a rock mass vibration frequency domain signal through a fast Fourier transform algorithm; S203: constructing a vibration spectrum diagram based on the rock mass vibration frequency domain signal; S204: extracting the pulse index, the margin index, and the impact energy according to the vibration waveform; S205: extracting the center of gravity frequency, the root mean square frequency, the relative energy of the first frequency band, and the damping ratio according to the vibration spectrum, thereby completing the extraction of multiple dynamic indicators of the rock mass to be measured.

5. The dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities according to claim 4 is characterized in that: The preliminary assessment of dangerous rock risks according to the various dynamic indicators in S3 specifically includes: S301A: normalizing each of the kinetic indices; S302A: Construct a preliminary assessment model for dangerous rock risks based on support vector machine algorithm; S303A: Inputting the normalized dynamic indicators into the dangerous rock risk preliminary assessment model to perform a preliminary assessment of the dangerous rock risk.

6. The dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities according to claim 2 is characterized in that: The preliminary assessment of dangerous rock risks based on the rock mass static parameters in S3 specifically includes: S301B: Calculating the critical slope height of the rock mass to be measured based on the rock mass cohesion and the rock mass internal friction angle using a Kullmann model; S302B: When the actual slope height of the rock mass to be measured is greater than or equal to the critical slope height, a preliminary assessment is made that the rock mass to be measured has a risk of dangerous rock formation; S303B: When the actual slope height of the rock mass to be measured is less than the critical slope height, it is preliminarily assessed that there is no risk of dangerous rock in the rock mass to be measured.

7. The dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities according to claim 1 is characterized in that: The deep learning algorithm is specifically: temporal convolutional neural network; Inputting the environmental monitoring data into the dangerous rock risk fine assessment model in S5 to perform fine assessment of the dangerous rock risk specifically includes: S501: performing normalization processing on each of the environmental monitoring data; S502: Calculating multiple weight coefficients of each normalized environmental monitoring data using an entropy weight method; S503: Calculating multiple entropy weight scores of each environmental monitoring data after normalization according to each weight coefficient; S504: Calculating a dangerous rock risk index based on each of the entropy weight scores through an attention mechanism; S505: Inputting the normalized environmental monitoring data and the dangerous rock risk index into the dangerous rock risk fine assessment model to perform a fine assessment of the dangerous rock risk.

8. The dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities according to claim 7 is characterized in that: The S502 specifically includes: S5021: Calculating multiple entropy values ​​of each normalized environmental monitoring data, wherein the entropy values ​​include: information entropy value, fuzzy entropy value, and permutation entropy value; S5022: Calculating multiple coefficients of variation of each normalized environmental monitoring data according to each of the entropy values; S5023: Calculate the multiple weight coefficients of each normalized environmental monitoring data according to each coefficient of variation.

9. The dangerous rock risk early warning method based on the fusion of static-dynamic-environmental quantities according to claim 7 is characterized in that: The S504 specifically includes: S5041: concatenate the entropy weight scores to calculate a comprehensive entropy weight score; S5042: Normalize the comprehensive entropy weight score and calculate the dangerous rock risk index.

10. The dangerous rock risk warning system based on the fusion of static-dynamic-environmental quantities is characterized by: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the dangerous rock risk warning method based on the fusion of static-dynamic-environmental quantities as described in any one of claims 1 to 9 is implemented.

Images