A dangerous rock collapse monitoring, analysis and early warning system and method

Abstract

The present invention relates to the field of dangerous rock collapse monitoring technology, and proposes a dangerous rock collapse monitoring analysis and early warning system and method, comprising the following steps: obtaining displacement data and geometric morphology data of the dangerous rock body, performing format conversion, data cleaning and standardization processing on the displacement data and geometric morphology data, and storing them in a data lake; extracting displacement data and geometric morphology data from the data lake, identifying the displacement data and geometric morphology data based on a collapse monitoring model, and obtaining the current collapse state of the dangerous rock body; building a collapse analysis prediction model based on an LSTM model, predicting the current collapse state through the collapse analysis prediction model, and obtaining the estimated change trend of the geometric morphology data; predicting the collapse impact range and collapse duration of the dangerous rock body based on the collapse analysis evaluation model combined with the current collapse state and the estimated change trend; evaluating the collapse warning level of the dangerous rock body based on the collapse warning model combined with the collapse impact range and collapse duration; and starting the corresponding early warning response plan according to the collapse warning level of the dangerous rock body. The present invention realizes real-time monitoring and accurate identification of the collapse state of the dangerous rock body, improves the prediction accuracy of the change trend of the collapse risk of the dangerous rock body over time, and completes the dynamic evaluation and timely response of the collapse warning level.

Description

Technical Field

[0001] The present invention relates to the technical field of dangerous rock mass collapse monitoring, and in particular to a dangerous rock mass collapse monitoring, analysis and early warning system and method. Background Art

[0002] Dangerous rock refers to unstable rock mass on steep slopes or cliffs that is divided by various structural surfaces. Although it has not collapsed yet, it actually has the main conditions for collapse, so collapse may occur in the near future. Dangerous rock deformation and collapse occur from time to time, which will bring great harm to people and objects around the place of occurrence, and often cause serious disasters. In order to improve the safety of roads, especially the collapse problem caused by dangerous rock deformation on both sides of the road, the prediction of dangerous rock deformation becomes particularly important.

[0003] Under natural conditions, the occurrence of dangerous rock collapse is a slow process. The rock exposed on the surface gradually cracks under various weathering and erosion effects, and eventually falls and rolls from the initial position to the foot of the slope. Due to the great uncertainty of the occurrence of dangerous rock collapse, the complexity and suddenness of the collapse movement itself, it will cause incalculable consequences after it occurs. In order to avoid the occurrence of collapse and reduce unnecessary human and economic losses caused by dangerous rock collapse, it is extremely important to predict the deformation law of dangerous rocks and quantitatively predict the deformation trend of dangerous rocks.

[0004] Existing technologies generally use convolutional neural networks to identify dangerous rock image data to determine whether there is a risk of collapse of dangerous rocks. However, the results of convolutional neural network recognition can only determine the current collapse risk of dangerous rocks, and cannot predict the collapse state of dangerous rock bodies in future periods. The accuracy is low when predicting the changing trend of the collapse risk of dangerous rock bodies over time, which poses a safety hazard. Summary of the invention

[0005] In view of this, the present invention proposes a dangerous rock mass collapse monitoring, analysis and early warning system and method, which solves the problem of low prediction accuracy of the changing trend of dangerous rock mass collapse risk over time in the prior art.

[0006] The technical solution of the present invention is implemented as follows: In a first aspect, the present invention provides a dangerous rock mass collapse monitoring, analysis and early warning method, comprising the following steps:

[0007] S1, obtain the displacement data and geometric data of the dangerous rock mass, convert the format of the displacement data and geometric data, perform data cleaning and standardization, and store them in the data lake;

[0008] S2, extracting displacement data and geometric morphology data from the data lake, identifying the displacement data and geometric morphology data based on the collapse monitoring model, and obtaining the current collapse state of the dangerous rock mass;

[0009] S3, build a collapse analysis prediction model based on the LSTM model, predict the current collapse state through the collapse analysis prediction model, and obtain the estimated change trend of the geometric morphology data;

[0010] S4, based on the collapse analysis and assessment model combined with the current collapse status and estimated change trend, predict the collapse impact range and collapse duration of the dangerous rock mass;

[0011] S5, based on the collapse warning model combined with the collapse impact range and collapse duration, evaluate the collapse warning level of the dangerous rock mass;

[0012] S6: According to the collapse warning level of the dangerous rock mass, initiate the corresponding warning response plan.

[0013] On the basis of the above technical solution, preferably, step S1 includes:

[0014] Converting the displacement data and the geometric data into a matrix data set format to obtain a displacement data set and a geometric data set, deleting missing values ​​of the displacement data set, and interpolating the missing values ​​of the geometric data set to obtain a geometric data set;

[0015] A K-means-based clustering algorithm is used to denoise the displacement data set and the geometric morphology data set. A file similarity comparison algorithm is used to remove duplicate and redundant file data, and data standardization is completed. The data is then stored in a data lake.

[0016] Based on the above technical solution, preferably, step S2 includes:

[0017] The collapse monitoring model is constructed by combining a CNN model, an LSTM model, a multimodal fusion network and a Softmax output layer. The CNN model includes a standard convolution layer, a hole convolution and a U-Net architecture. The CNN model extracts image features from a geometric morphology data set to obtain image features.

[0018] The LSTM model is a two-layer LSTM structure, the LSTM structure includes 1024 neurons, and the LSTM model predicts the time series changes of the displacement data group to obtain time series features;

[0019] Image features and time series predictions are fused through a multimodal fusion network to obtain image time series fusion features;

[0020] The image temporal fusion features are classified through the Softmax output layer to obtain the current collapse state of the dangerous rock mass.

[0021] On the basis of the above technical solution, preferably, step S3 includes:

[0022] The collapse analysis prediction model is constructed based on the LSTM model. The current collapse state is predicted by the collapse analysis prediction model to obtain the estimated change trend of the geometric morphology data. The collapse analysis prediction model includes:

[0023]

[0024] Among them, α(t) is the estimated change trend of geometric data, L t is the actual geometric data value at time t, is the geometric data value predicted at time t when the current collapse state is state α, is the standard deviation of the actual geometric data value at time t, is the standard deviation of the geometric data values ​​predicted at time t when the current collapse state is state α.

[0025] Based on the above technical solution, preferably, step S4 includes:

[0026] The calculation formula of the collapse analysis evaluation model is:

[0027]

[0028] Among them, L(t) is the impact range of collapse, T(t) is the duration of collapse, S is the number of types of current collapse state, β s is the weight coefficient of the impact of the sth current collapse state on the collapse of dangerous rock mass, F s (e) is the e-th geometric data of the s-th current collapse state, E is the number of geometric data of the s-th current collapse state, α(t) is the estimated change trend of the geometric data, sign(·) is the signal function, δ L is the leveling parameter for the collapse impact range, δ T Leveling parameters for collapse duration.

[0029] Based on the above technical solution, preferably, step S5 includes:

[0030] Based on the collapse warning model combined with the collapse impact range and collapse duration, the collapse warning level of the dangerous rock mass is comprehensively evaluated. According to multiple preset thresholds of the collapse impact range and multiple preset thresholds of the collapse duration, the collapse warning level of the dangerous rock mass is divided into the first collapse warning level, the second collapse warning level, the third collapse warning level, the fourth collapse warning level, the fifth collapse warning level, the sixth collapse warning level and the seventh collapse warning level according to their severity. Among them, the seventh collapse warning level is the collapse warning level with the highest severity.

[0031] On the basis of the above technical solution, preferably, step S6 includes:

[0032] For the first collapse warning level, a monitoring team is dispatched to regularly check the monitoring data;

[0033] For the second collapse warning level, the frequency of regular inspection of monitoring data should be increased, and relevant personnel should be reminded to be vigilant;

[0034] For the third collapse warning level, more frequent monitoring data checks will be conducted, along with detailed data analysis and safety publicity for people in the affected areas;

[0035] For the fourth collapse warning level, non-essential personnel are restricted from entering the collapse-affected area and emergency equipment is prepared;

[0036] For the fifth collapse warning level, non-essential personnel in the area affected by the collapse will be evacuated, and a 24-hour standby will be implemented through the emergency command center;

[0037] For the sixth landslide warning level, all personnel in the area affected by the landslide are forced to evacuate and the area affected by the landslide is blocked;

[0038] For the seventh collapse warning level, a comprehensive emergency response procedure is initiated, public warnings are continuously issued, public information communication is maintained, and post-disaster recovery resources and disaster support resources are deployed.

[0039] In a second aspect, the present invention further provides a dangerous rock mass collapse monitoring, analysis and early warning system, the system comprising:

[0040] The collapse data processing module is used to obtain the displacement data and geometric data of the dangerous rock mass, convert the format of the displacement data and geometric data, clean and standardize the data, and store them in the data lake;

[0041] The collapse state identification module is used to extract displacement data and geometric morphology data from the data lake, identify the displacement data and geometric morphology data based on the collapse monitoring model, and obtain the current collapse state of the dangerous rock mass;

[0042] The geometric change prediction module is used to build a collapse analysis prediction model based on the LSTM model, predict the current collapse state through the collapse analysis prediction model, and obtain the estimated change trend of the geometric morphology data;

[0043] The collapse impact prediction module is used to predict the collapse impact range and collapse duration of dangerous rock masses based on the collapse analysis and assessment model combined with the current collapse status and estimated change trend;

[0044] The collapse warning assessment module is used to assess the collapse warning level of dangerous rock masses based on the collapse warning model combined with the collapse impact range and collapse duration;

[0045] The collapse warning response module is used to initiate the corresponding warning response plan according to the collapse warning level of the dangerous rock mass.

[0046] In a third aspect, the present invention further provides an electronic device, comprising: at least one processor, at least one memory, a communication interface and a bus;

[0047] The processor, memory and communication interface communicate with each other via the bus, the memory stores program instructions executable by the processor, and the processor calls the program instructions to implement steps of a dangerous rock collapse monitoring, analysis and early warning method.

[0048] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions enable a computer to implement steps of a dangerous rock mass collapse monitoring, analysis and early warning method.

[0049] The dangerous rock mass collapse monitoring, analysis and early warning system and method of the present invention have the following beneficial effects compared with the prior art:

[0050] (1) By acquiring displacement data and geometric data of dangerous rock masses, performing format conversion, data cleaning and standardization, identifying collapse states, predicting the changing trend of geometric data, predicting the impact range and duration of collapse, evaluating the collapse warning level, and initiating corresponding warning response plans, the real-time monitoring and accurate identification of the collapse state of dangerous rock masses are achieved, the prediction accuracy of the changing trend of the collapse risk of dangerous rock masses over time is improved, and the dynamic evaluation and timely response of the collapse warning level are completed;

[0051] (2) Through the data processing process, including format conversion, data cleaning, standardization and denoising, the high quality of the input data is ensured. The K-means clustering algorithm is used for denoising and the file similarity comparison algorithm is used to remove redundant data, which provides a basis for collapse analysis and significantly improves the consistency of the data;

[0052] (3) By combining the CNN model, LSTM model and multimodal fusion network, the collapse monitoring model can simultaneously process image features and time series features to accurately identify the current collapse state of the dangerous rock mass. The collapse analysis and prediction model can be used to make high-precision predictions on the future geometric changes of the dangerous rock mass, thus improving the early warning capability of the system.

[0053] (4) By considering the impact range and duration of collapse, a comprehensive and objective assessment of collapse risk is conducted, and a seven-level collapse warning level and corresponding warning response plan are set to provide targeted response measures for different degrees of collapse risk, effectively improving the efficiency of disaster prevention and mitigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0055] Figure 1 A flow chart of a dangerous rock mass collapse monitoring, analysis and early warning method of the present invention;

[0056] Figure 2 The present invention is a structural diagram of a dangerous rock mass collapse monitoring, analysis and early warning system. DETAILED DESCRIPTION

[0057] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] See also Figure 1 This embodiment provides a dangerous rock mass collapse monitoring, analysis and early warning method, including the following steps:

[0059] S1, obtain the displacement data and geometric data of the dangerous rock mass, convert the format of the displacement data and geometric data, perform data cleaning and standardization, and store them in the data lake;

[0060] S2, extracting displacement data and geometric morphology data from the data lake, identifying the displacement data and geometric morphology data based on the collapse monitoring model, and obtaining the current collapse state of the dangerous rock mass;

[0061] S3, build a collapse analysis prediction model based on the LSTM model, predict the current collapse state through the collapse analysis prediction model, and obtain the estimated change trend of the geometric morphology data;

[0062] S4, based on the collapse analysis and assessment model combined with the current collapse status and estimated change trend, predict the collapse impact range and collapse duration of the dangerous rock mass;

[0063] S5, based on the collapse warning model combined with the collapse impact range and collapse duration, evaluate the collapse warning level of the dangerous rock mass;

[0064] S6: According to the collapse warning level of the dangerous rock mass, initiate the corresponding warning response plan.

[0065] Specifically, this embodiment obtains displacement data and geometric data of dangerous rock masses, performs format conversion, data cleaning and standardization processing, identifies collapse states, predicts changing trends of geometric data, predicts the impact range and duration of collapse, evaluates collapse warning levels, and initiates corresponding warning response plans, thereby achieving real-time monitoring and accurate identification of collapse states of dangerous rock masses, improving the accuracy of predicting the changing trends of collapse risks of dangerous rock masses over time, and completing dynamic evaluation and timely response of collapse warning levels.

[0066] Step S1 includes:

[0067] Converting the displacement data and the geometric data into a matrix data set format to obtain a displacement data set and a geometric data set, deleting missing values ​​of the displacement data set, and interpolating the missing values ​​of the geometric data set to obtain a geometric data set;

[0068] A K-means-based clustering algorithm is used to denoise the displacement data set and the geometric morphology data set. A file similarity comparison algorithm is used to remove duplicate and redundant file data, and data standardization is completed. The data is then stored in a data lake.

[0069] Specifically, this embodiment ensures the integrity and consistency of the data by converting the data into a matrix format, deleting missing values ​​and performing interpolation, adopts a K-means-based clustering algorithm for denoising, effectively reduces noise interference in the data, removes duplicate and redundant file data through a file similarity comparison algorithm, reduces data redundancy, and improves storage efficiency and the effectiveness of data management.

[0070] The geometric data of the dangerous rock mass include the length, width, height, surface roughness and crack volume ratio of the dangerous rock mass.

[0071] This embodiment stores the processed data in the data lake, realizes standardized processing and centralized management of data, and provides a convenient data access and usage environment.

[0072] Step S2 includes:

[0073] The collapse monitoring model is constructed by combining a CNN model, an LSTM model, a multimodal fusion network and a Softmax output layer. The CNN model includes a standard convolution layer, a hole convolution and a U-Net architecture. The CNN model extracts image features from a geometric morphology data set to obtain image features.

[0074] The LSTM model is a two-layer LSTM structure, the LSTM structure includes 1024 neurons, and the LSTM model predicts the time series changes of the displacement data group to obtain time series features;

[0075] Image features and time series predictions are fused through a multimodal fusion network to obtain image time series fusion features;

[0076] The image temporal fusion features are classified through the Softmax output layer to obtain the current collapse state of the dangerous rock mass.

[0077] Specifically, this embodiment uses the CNN model to extract image features of geometric morphology data, and uses the LSTM model to extract time series features of displacement data, thereby realizing comprehensive feature capture of the state of dangerous rock masses. The CNN model uses a combination of standard convolutional layers, hole convolutions, and U-Net architectures to enhance the depth and breadth of image feature extraction; the LSTM model uses a two-layer structure of 1024 neurons to improve the accuracy of time series feature extraction.

[0078] Image features and time series features are fused through a multimodal fusion network to obtain comprehensive image and time series fusion features, which fully utilizes the complementary advantages of different types of data. The Softmax output layer is used to classify the fused features, thus achieving accurate identification of the current collapse state of the dangerous rock mass.

[0079] Step S3 includes:

[0080] The collapse analysis prediction model is constructed based on the LSTM model. The current collapse state is predicted by the collapse analysis prediction model to obtain the estimated change trend of the geometric morphology data. The collapse analysis prediction model includes:

[0081]

[0082] Among them, α(t) is the estimated change trend of geometric data, L t is the actual geometric data value at time t, is the geometric data value predicted at time t when the current collapse state is state α, is the standard deviation of the actual geometric data value at time t, is the standard deviation of the geometric data values ​​predicted at time t when the current collapse state is state α.

[0083] Specifically, this embodiment can accurately predict the current collapse state through the collapse analysis and prediction model built based on the LSTM model, especially in capturing the changing trend of geometric morphological data, which significantly improves the prediction accuracy. The LSTM model is good at processing time series data. By analyzing the time series changes of geometric morphological data, it can better understand and predict the dynamic changes of data over time.

[0084] By calculating the estimated changing trends of geometric data, the model can provide reliable predictions about future states and important trend information for decision makers. By analyzing actual data values, predicted data values ​​and their standard deviations, the model can provide quantitative information about the accuracy and uncertainty of the predictions.

[0085] Step S4 includes:

[0086] The calculation formula of the collapse analysis evaluation model is:

[0087]

[0088]

[0089] Among them, L(t) is the impact range of collapse, T(t) is the duration of collapse, S is the number of types of current collapse state, β s is the weight coefficient of the impact of the sth current collapse state on the collapse of dangerous rock mass, F s (e) is the e-th geometric data of the s-th current collapse state, E is the number of geometric data of the s-th current collapse state, α(t) is the estimated change trend of the geometric data, sign(·) is the signal function, δ L is the leveling parameter for the collapse impact range, δ T Leveling parameters for collapse duration.

[0090] Specifically, this embodiment quantifies the impact range and duration of collapse through calculation of the collapse analysis and evaluation model. The model can comprehensively consider the impact of various collapse states on dangerous rock masses by introducing impact weight coefficients of different collapse states.

[0091] By introducing the collapse impact range leveling parameters and the collapse duration leveling parameters, the model can dynamically adjust the evaluation results according to the actual situation. The collapse impact range leveling parameters and the collapse duration leveling parameters are used to smooth and adjust the prediction results to reflect changes in the actual situation.

[0092] Through the combination of signal functions and leveling parameters, the model can adapt to different collapse states and geometric morphology change trends, improving the applicability and flexibility of the model in different situations.

[0093] The collapse impact range leveling parameter and collapse duration leveling parameter are used as adjustment factors to adjust the sensitivity of the model to the collapse impact range and duration, and the degree of influence of the model on different factors can also be adjusted.

[0094] Step S5 includes:

[0095] Based on the collapse warning model combined with the collapse impact range and collapse duration, the collapse warning level of the dangerous rock mass is comprehensively evaluated. According to multiple preset thresholds of the collapse impact range and multiple preset thresholds of the collapse duration, the collapse warning level of the dangerous rock mass is divided into the first collapse warning level, the second collapse warning level, the third collapse warning level, the fourth collapse warning level, the fifth collapse warning level, the sixth collapse warning level and the seventh collapse warning level according to their severity. Among them, the seventh collapse warning level is the collapse warning level with the highest severity.

[0096] Specifically, the calculation formula for the classification of dangerous rock mass collapse warning levels is:

[0097]

[0098] Wherein, C is the collapse warning level of the dangerous rock mass, C1, C2, C3, C4, C5, C6 and C7 are the first collapse warning level, the second collapse warning level, the third collapse warning level, the fourth collapse warning level, the fifth collapse warning level, the sixth collapse warning level and the seventh collapse warning level respectively; L(t) is the collapse impact range, T(t) is the collapse duration, and are the first collapse impact range threshold and the second collapse impact range threshold, T1 C and They are the first collapse duration threshold and the second collapse duration threshold respectively.

[0099] This embodiment provides accurate graded warning of the collapse risk of dangerous rock masses by combining the collapse impact range and duration. The warning levels range from the first to the seventh, which refines the levels of risk assessment and makes the warning more targeted.

[0100] By setting multiple preset thresholds, the model can dynamically adjust the collapse warning level, improve the ability to identify different risk levels, and enhance the targeted nature of risk management and response measures. The division of warning levels provides decision makers with clear risk level information, helping them to take appropriate response measures according to different warning levels, thereby improving the efficiency and effectiveness of decision-making.

[0101] By using multiple thresholds to assess the impact scope and duration of collapse, the model can adapt to different geological conditions and environmental changes, providing more adaptive early warning information.

[0102] Step S6 includes:

[0103] For the first collapse warning level, a monitoring team is dispatched to regularly check the monitoring data;

[0104] For the second collapse warning level, the frequency of regular inspection of monitoring data should be increased, and relevant personnel should be reminded to be vigilant;

[0105] For the third collapse warning level, more frequent monitoring data checks will be conducted, along with detailed data analysis and safety publicity for people in the affected areas;

[0106] For the fourth collapse warning level, non-essential personnel are restricted from entering the collapse-affected area and emergency equipment is prepared;

[0107] For the fifth collapse warning level, non-essential personnel in the area affected by the collapse will be evacuated, and a 24-hour standby will be implemented through the emergency command center;

[0108] For the sixth landslide warning level, all personnel in the area affected by the landslide are forced to evacuate and the area affected by the landslide is blocked;

[0109] For the seventh collapse warning level, a comprehensive emergency response procedure is initiated, public warnings are continuously issued, public information communication is maintained, and post-disaster recovery resources and disaster support resources are deployed.

[0110] Specifically, this embodiment establishes a hierarchical response mechanism by taking corresponding response measures for different collapse warning levels. The hierarchical response mechanism can gradually upgrade response measures according to the severity of the risk, thereby effectively managing resources and improving action efficiency.

[0111] Each warning level has specific response measures, from regular monitoring to comprehensive emergency response. The system can flexibly adapt to different risk scenarios. By conducting safety publicity, restricting entry, evacuation and issuing public warnings at higher warning levels, the system not only protects public safety, but also improves the public's awareness of and ability to respond to collapse risks. At the highest warning level, a comprehensive emergency response procedure is initiated, and post-disaster recovery resources and disaster support resources are deployed to ensure that resources can be effectively utilized in times of crisis and minimize losses and impacts.

[0112] See also Figure 2 The present invention also provides a dangerous rock collapse monitoring, analysis and early warning system, the system comprising:

[0113] The collapse data processing module is used to obtain the displacement data and geometric data of the dangerous rock mass, convert the format of the displacement data and geometric data, clean and standardize the data, and store them in the data lake;

[0114] The collapse state identification module is used to extract displacement data and geometric morphology data from the data lake, identify the displacement data and geometric morphology data based on the collapse monitoring model, and obtain the current collapse state of the dangerous rock mass;

[0115] The geometric change prediction module is used to build a collapse analysis prediction model based on the LSTM model, predict the current collapse state through the collapse analysis prediction model, and obtain the estimated change trend of the geometric morphology data;

[0116] The collapse impact prediction module is used to predict the collapse impact range and collapse duration of dangerous rock masses based on the collapse analysis and assessment model combined with the current collapse status and estimated change trend;

[0117] The collapse warning assessment module is used to assess the collapse warning level of dangerous rock masses based on the collapse warning model combined with the collapse impact range and collapse duration;

[0118] The collapse warning response module is used to initiate the corresponding warning response plan according to the collapse warning level of the dangerous rock mass.

[0119] Specifically, a dangerous rock collapse monitoring, analysis and early warning system of this embodiment realizes comprehensive data processing, accurate state identification and prediction, dynamic impact assessment, and hierarchical early warning and response mechanism. The system starts with data acquisition, collapse state identification, geometric change prediction, impact range and duration assessment, and finally to early warning level assessment and response plan initiation, forming a complete monitoring-analysis-early warning-response chain. The system not only improves the accuracy and timeliness of collapse risk prediction, but also enhances the efficiency and effectiveness of emergency management. By providing a reliable data foundation, accurate risk assessment and targeted response measures, the system significantly improves the risk management capabilities of dangerous rock collapse, provides strong technical support for disaster prevention and mitigation work, and effectively improves the level of public safety.

[0120] The present invention also discloses an electronic device, comprising: at least one processor, at least one memory communication interface and a bus: wherein the processor, memory and communication interface communicate with each other through the bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement a dangerous rock collapse monitoring, analysis and early warning method.

[0121] The present invention also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, wherein the computer instructions enable the computer to implement all or part of the steps of a dangerous rock mass collapse monitoring, analysis and early warning method described in an embodiment of the present invention. The storage medium includes: a U disk, a mobile hard disk, a read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk, and other media that can store program codes.

[0122] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A dangerous rock mass collapse monitoring, analysis and early warning method, characterized in that: The following steps are involved: S1, obtain the displacement data and geometric data of the dangerous rock mass, convert the format of the displacement data and geometric data, perform data cleaning and standardization, and store them in the data lake; S2, extract displacement data and geometric data from the data lake, identify the displacement data and geometric data based on the collapse monitoring model, and obtain the current collapse state of the dangerous rock mass; S3, build a collapse analysis prediction model based on the LSTM model, predict the current collapse state through the collapse analysis prediction model, and obtain the estimated change trend of the geometric morphology data; S4, based on the collapse analysis and assessment model combined with the current collapse status and estimated change trend, predict the collapse impact range and collapse duration of the dangerous rock mass; S5, based on the collapse warning model combined with the collapse impact range and collapse duration, evaluate the collapse warning level of the dangerous rock mass; S6, according to the collapse warning level of the dangerous rock mass, initiate the corresponding warning response plan; Step S1 includes: Converting the displacement data and the geometric data into a matrix data set format to obtain a displacement data set and a geometric data set, deleting missing values ​​of the displacement data set, and interpolating the missing values ​​of the geometric data set to obtain a geometric data set; A K-means-based clustering algorithm is used to denoise the displacement data set and the geometric morphology data set. A file similarity comparison algorithm is used to remove duplicate and redundant file data, complete data standardization, and use data lake storage. Step S2 includes: The collapse monitoring model is constructed by combining a CNN model, an LSTM model, a multimodal fusion network and a Softmax output layer. The CNN model includes a standard convolution layer, a hole convolution and a U-Net architecture. The CNN model extracts image features from a geometric morphology data set to obtain image features. The LSTM model is a two-layer LSTM structure, the LSTM structure includes 1024 neurons, and the LSTM model predicts the time series changes of the displacement data group to obtain time series features; Image features and time series predictions are fused through a multimodal fusion network to obtain image time series fusion features; The image temporal fusion features are classified through the Softmax output layer to obtain the current collapse state of the dangerous rock mass.

2. A dangerous rock mass collapse monitoring, analysis and early warning method as claimed in claim 1, characterized in that: Step S3 includes: The collapse analysis prediction model is constructed based on the LSTM model. The current collapse state is predicted by the collapse analysis prediction model to obtain the estimated change trend of the geometric morphology data. The collapse analysis prediction model includes: Among them, α(t) is the estimated change trend of geometric data, L t is the actual geometric data value at time t, is the geometric data value predicted at time t when the current collapse state is state α, σ Lt is the standard deviation of the actual geometric data value at time t, is the standard deviation of the geometric data values ​​predicted at time t when the current collapse state is state α.

3. A dangerous rock mass collapse monitoring, analysis and early warning method as claimed in claim 2, characterized in that: Step S4 includes: The calculation formula of the collapse analysis evaluation model is: Among them, L(t) is the impact range of collapse, T(t) is the duration of collapse, S is the number of types of current collapse state, β s is the weight coefficient of the impact of the sth current collapse state on the collapse of dangerous rock mass, F s (e) is the e-th geometric data of the s-th current collapse state, E is the number of geometric data of the s-th current collapse state, α(t) is the estimated change trend of the geometric data, sign(·) is the signal function, δ L is the leveling parameter for the collapse impact range, δ T Leveling parameters for collapse duration.

4. A dangerous rock mass collapse monitoring, analysis and early warning method as claimed in claim 3, characterized in that: Step S5 includes: Based on the collapse warning model combined with the collapse impact range and collapse duration, the collapse warning level of the dangerous rock mass is comprehensively evaluated. According to multiple preset thresholds of the collapse impact range and multiple preset thresholds of the collapse duration, the collapse warning level of the dangerous rock mass is divided into the first collapse warning level, the second collapse warning level, the third collapse warning level, the fourth collapse warning level, the fifth collapse warning level, the sixth collapse warning level and the seventh collapse warning level according to their severity. Among them, the seventh collapse warning level is the collapse warning level with the highest severity.

5. A dangerous rock mass collapse monitoring, analysis and early warning method as claimed in claim 4, characterized in that: Step S6 includes: For the first collapse warning level, a monitoring team is dispatched to regularly check the monitoring data; For the second collapse warning level, the frequency of regular inspection of monitoring data should be increased, and relevant personnel should be reminded to be vigilant; For the third collapse warning level, more frequent monitoring data checks will be conducted, along with detailed data analysis and safety publicity for people in the affected areas; For the fourth collapse warning level, non-essential personnel are restricted from entering the collapse-affected area and emergency equipment is prepared; For the fifth collapse warning level, non-essential personnel in the area affected by the collapse will be evacuated, and a 24-hour standby will be implemented through the emergency command center; For the sixth landslide warning level, all personnel in the area affected by the landslide are forced to evacuate and the area affected by the landslide is blocked; For the seventh collapse warning level, a comprehensive emergency response procedure is initiated, public warnings are continuously issued, public information communication is maintained, and post-disaster recovery resources and disaster support resources are deployed.

6. A dangerous rock mass collapse monitoring, analysis and early warning system, used to execute a dangerous rock mass collapse monitoring, analysis and early warning method as claimed in any one of claims 1 to 5, characterized in that: The system comprises: The collapse data processing module is used to obtain the displacement data and geometric data of the dangerous rock mass, convert the format of the displacement data and geometric data, clean and standardize the data, and store them in the data lake; The collapse state identification module is used to extract displacement data and geometric morphology data from the data lake, identify the displacement data and geometric morphology data based on the collapse monitoring model, and obtain the current collapse state of the dangerous rock mass; The geometric change prediction module is used to build a collapse analysis prediction model based on the LSTM model, predict the current collapse state through the collapse analysis prediction model, and obtain the estimated change trend of the geometric morphology data; The collapse impact prediction module is used to predict the collapse impact range and collapse duration of dangerous rock masses based on the collapse analysis and assessment model combined with the current collapse status and estimated change trend; The collapse warning assessment module is used to assess the collapse warning level of dangerous rock masses based on the collapse warning model combined with the collapse impact range and collapse duration; The collapse warning response module is used to initiate the corresponding warning response plan according to the collapse warning level of the dangerous rock mass.

7. An electronic device, characterized in that: include: at least one processor, at least one memory, a communication interface, and a bus; The processor, memory, and communication interface communicate with each other via the bus, the memory stores program instructions executable by the processor, and the processor calls the program instructions to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions enable a computer to implement the method according to any one of claims 1 to 5.

Images