Lower reservoir high slope landslide vibration monitoring and early warning system

Abstract

The invention discloses a lower reservoir high slope landslide vibration monitoring and early warning system, belongs to the technical field of vibration monitoring, and aims to solve the problems that an existing monitoring and early warning system is poor in comprehensiveness and low in risk estimation accuracy. The system comprises a monitoring unit, a comprehensive coefficient calculation unit and a risk level judgment unit, and the comprehensive coefficient calculation unit comprises an index quantification module, a matrix construction and weight distribution module and a comprehensive coefficient calculation module. A plurality of factors such as stratum characteristics, side slope geological conditions, rock mass characteristics, rock mass structures, side slope heights and side slope gradients are comprehensively considered, a dynamic matrix is constructed, a comprehensive coefficient is calculated by adopting a scientific weight distribution method, and a risk grade division standard is automatically adjusted according to a risk index change trend after fusion. The system adapts to the dynamic change of slope stability, timely gives out accurate early warning, overcomes the problem that an existing monitoring and early warning system only pays attention to a single index or a few indexes, and improves the comprehensiveness of evaluation.

Description

Technical Field

[0001] The invention relates to the technical field of vibration monitoring, and in particular to a vibration monitoring and early warning system for landslides on a high slope of a lower reservoir. Background Art

[0002] With the continuous expansion of the scale of water conservancy project construction, slope blasting excavation has been widely used in engineering. Slope blasting excavation can efficiently realize large-scale earth and stone excavation. However, the vibration generated during the blasting excavation process may have adverse effects on the surrounding environment and structures, such as causing instability of the surrounding geological structure and even damaging buildings in nearby areas.

[0003] The commonly used monitoring and early warning systems currently only focus on a single indicator or a few indicators. They lack a systematic and comprehensive evaluation system and find it difficult to fully consider the combined impact of multiple factors. The vibration problem caused by slope blasting and excavation in water conservancy projects is a complex dynamic process. Traditional evaluation methods make it difficult to quantitatively describe the vibration characteristics and impacts, which greatly limits the accuracy and reliability of risk assessment. Summary of the invention

[0004] The purpose of the present invention is to provide a vibration monitoring and early warning system for landslides on high slopes of lower reservoirs, which solves the problem that the existing monitoring and early warning systems in the background technology have poor comprehensiveness and low accuracy in risk estimation.

[0005] To achieve the above object, the present invention provides the following technical solution: a vibration monitoring and early warning system for landslides on high slopes of lower reservoirs, comprising:

[0006] A monitoring unit, which is used to obtain relevant data of the slope in a comprehensive and real-time manner, and includes a monitoring point layout module, a sensor configuration module, and a data acquisition and transmission module;

[0007] A comprehensive coefficient calculation unit, which is used to comprehensively evaluate the stability of the slope, includes:

[0008] Index quantification module: used to quantify stratum characteristics, slope geological conditions, rock mass characteristics, rock mass structure, slope height and slope gradient;

[0009] Matrix construction and weight distribution module: used to construct a dynamic matrix to reflect the values ​​of monitoring points under different indicators, and consider the mutual influence between different monitoring points. The spatial correlation matrix is ​​introduced. The spatial correlation matrix should accurately reflect the spatial relationship and interaction between monitoring points. The weight of each indicator is determined by combining principal component analysis with genetic algorithm, and the importance of each indicator to slope stability is comprehensively considered.

[0010] Comprehensive coefficient calculation module: using formula Calculate the comprehensive coefficient of each monitoring point, where represents the comprehensive coefficient of the i-th monitoring point, This part mainly reflects the traditional comprehensive calculation based on indicator weights and the values ​​of monitoring points under each indicator. Spatial correlation is partially introduced to consider the differences between different monitoring points and the impact of their relationship on the comprehensive coefficient of the current monitoring point;

[0011] Risk level judgment unit: used to judge the risk level of the slope. The risk level judgment unit includes a multimodal data fusion module, a dynamic risk level classification module and an early warning decision module.

[0012] Furthermore, the monitoring point layout module arranges multiple monitoring points according to the slope topography, geological structure and areas that may be affected by blasting, focusing on potential landslide areas, weak rock structures, key parts of the top and bottom of the slope, and adjusts the density of monitoring points according to the slope risk level.

[0013] Furthermore, the sensor configuration module selects high-precision vibration sensors, displacement sensors and stress sensors to monitor the vibration, displacement and stress changes of the slope in real time.

[0014] Furthermore, the data acquisition and transmission module uses a data collector to collect sensor data in real time. The data collector adopts LoRa wireless communication technology, configures corresponding gateways and terminal nodes, and transmits the collected data to the monitoring center.

[0015] Furthermore, the formation properties include lithology, hardness, permeability and formation age. The geological conditions of the slope increase the quantification of groundwater activity, including the impact of groundwater level, water flow velocity and water quality on slope stability. The rock mass properties include strength, elastic modulus and degree of anisotropy. The rock mass properties are comprehensively determined through laboratory testing, field monitoring and numerical simulation methods. The rock mass structure includes the density, openness, connectivity and direction and distribution of joints and fissures. The slope height is quantified in intervals, and the risk growth rate in different intervals is different. The slope gradient considers the inclination angle and slope change rate.

[0016] Furthermore, in the comprehensive coefficient calculation module, the formula middle is the weight optimized by genetic algorithm, is the value of the i-th monitoring point under the j-th indicator. In the comprehensive coefficient calculation module, the formula Where k is the spatial correlation coefficient, which is used to adjust the influence of spatial correlation in the calculation of comprehensive coefficient. Indicates that all other monitoring points except the i-th monitoring point are summed. It means the sum of six indicators: stratum characteristics, slope geological conditions, rock mass characteristics, rock mass structure, slope height and slope gradient. is the spatial correlation coefficient between the mth and nth monitoring points, reflecting the similarity between different monitoring points in various indicators. It represents the difference between the i-th monitoring point and other monitoring points in the n-th indicator.

[0017] Furthermore, the multimodal data fusion module introduces the slope surface temperature change, acoustic emission signal monitoring data, vibration monitoring quantity and comprehensive coefficient for multimodal data fusion, uses the convolutional neural network in deep learning for data fusion, automatically extracts features and integrates the relationship between different types of data. The calculation formula of the multimodal fusion module is: ; R is the risk index, which is used to quantitatively assess the risk level of slope landslide; n represents the number of monitoring points; Indicates that the accumulation operation is performed on n monitoring points, i takes values ​​from 1 to n in sequence, covering all monitoring points; represents the weight coefficient of the i-th monitoring point, which is determined by the importance and location of the monitoring point; V i represents the vibration monitoring value of the ith monitoring point, C i represents the comprehensive coefficient of the i-th monitoring point, T i represents the temperature change of the i-th monitoring point, S i represents the acoustic emission intensity of the ith monitoring point, is the weight coefficient of the corresponding data, which is determined by the hierarchical analysis method based on historical data and expert experience.

[0018] Furthermore, the dynamic risk level classification module uses the fuzzy logic system to automatically adjust the risk level classification criteria according to the changing trend of the fused risk indicators. The risk level is set to four levels: low risk, medium risk, high risk, and extremely high risk. The threshold interval is set to divide and the mean of the fused risk indicators is calculated. and standard deviation ;

[0019] when Low risk when

[0020] when The risk is medium.

[0021] when When the risk is high;

[0022] when Very high risk.

[0023] Furthermore, the early warning decision module adopts a reinforcement learning algorithm to optimize the early warning decision according to the actual effect after the early warning. When the risk level increases, the system automatically analyzes the cause and provides corresponding recommended measures.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The present invention provides a vibration monitoring and early warning system for landslides on high slopes of lower reservoirs, which comprehensively considers multiple factors such as stratum characteristics, slope geological conditions, rock characteristics, rock structure, slope height and slope gradient, constructs a dynamic matrix and adopts a scientific weight distribution method to calculate the comprehensive coefficient, automatically adjusts the risk level classification standard according to the changing trend of the integrated risk indicators, adapts to the dynamic changes of slope stability, and issues accurate early warnings in a timely manner, overcoming the problem that the existing monitoring and early warning systems only focus on a single indicator or a few indicators, and improves the comprehensiveness of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram of a module of the present invention;

[0027] Figure 2 It is a structural diagram of the safety risk factors of the indicator quantification module of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in 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.

[0029] In order to solve the technical problems of poor comprehensiveness of existing monitoring and early warning systems and low accuracy of risk estimation, such as Figure 1-Figure 2 As shown, the following preferred technical solutions are provided:

[0030] A vibration monitoring and early warning system for landslides on a high slope of a lower reservoir comprises a monitoring unit, a comprehensive coefficient calculation unit and a risk level judgment unit.

[0031] The monitoring unit is used to obtain relevant data of the slope in a comprehensive and real-time manner;

[0032] The monitoring unit includes:

[0033] Monitoring point layout module: The monitoring point layout module rationally arranges multiple monitoring points according to the slope topography, geological structure and areas that may be affected by blasting, focusing on potential landslide areas, weak rock structure, and key parts of the top and bottom of the slope. The density of monitoring points is adjusted according to the risk level of the slope to ensure the comprehensiveness and representativeness of the monitoring data. At the same time, the monitoring points are carefully arranged by level and region, taking into account the geological differences and the degree of impact of blasting in different regions, so as to improve the accuracy and effectiveness of monitoring;

[0034] Sensor configuration module: The sensor configuration module uses high-precision vibration sensors, displacement sensors and stress sensors to monitor the vibration, displacement and stress changes of the slope in real time;

[0035] Data acquisition and transmission module: The data acquisition and transmission module uses a data collector to collect sensor data in real time. The data collector uses LoRa wireless communication technology, configures corresponding gateways and terminal nodes, and transmits the collected data to the monitoring center.

[0036] The comprehensive coefficient calculation unit is used to comprehensively evaluate the stability of the slope; the comprehensive coefficient calculation unit includes:

[0037] Index quantification module: used to quantify formation characteristics, slope geological conditions, rock mass characteristics, rock mass structure, slope height and slope gradient. Formation characteristics include lithology, hardness, permeability and formation age. Slope geological conditions increase groundwater activity, including the impact of groundwater level, water flow velocity and water quality on slope stability. Rock mass characteristics include strength, elastic modulus and degree of anisotropy. Rock mass characteristics are determined comprehensively through laboratory testing, on-site monitoring and numerical simulation methods. Rock mass structure includes density, openness, connectivity and direction and distribution of joints and fissures. Slope height is quantified in intervals, and the risk growth rate is different in different intervals. Slope gradient considers inclination angle and slope change rate.

[0038] Matrix construction and weight distribution module: used to construct a dynamic matrix to reflect the values ​​of monitoring points under different indicators, and consider the mutual influence between different monitoring points. The spatial correlation matrix is ​​introduced. The spatial correlation matrix should accurately reflect the spatial relationship and interaction between monitoring points. The weight of each indicator is determined by combining principal component analysis with genetic algorithm, and the importance of each indicator to slope stability is comprehensively considered.

[0039] Comprehensive coefficient calculation module: using formula Calculate the comprehensive coefficient of each monitoring point, where represents the comprehensive coefficient of the i-th monitoring point, This part mainly reflects the traditional comprehensive calculation based on indicator weights and the values ​​of monitoring points under each indicator, among which is the weight optimized by genetic algorithm, is the value of the i-th monitoring point under the j-th indicator, The spatial correlation is partially introduced, taking into account the differences between different monitoring points and the impact of the relationship between them on the comprehensive coefficient of the current monitoring point, where k is the spatial correlation coefficient, which is used to adjust the influence of spatial correlation in the calculation of the comprehensive coefficient. Indicates that all other monitoring points except the i-th monitoring point are summed. It means the sum of six indicators: stratum characteristics, slope geological conditions, rock mass characteristics, rock mass structure, slope height and slope gradient. is the spatial correlation coefficient between the mth and nth monitoring points, reflecting the similarity between different monitoring points in various indicators. If the numerical change trends of two monitoring points in a certain indicator are similar, then the spatial correlation coefficient between them will be larger. It represents the difference between the i-th monitoring point and other monitoring points in the n-th indicator. By considering this difference and combining it with the spatial correlation coefficient, it can reflect the relative relationship between the current monitoring point and other monitoring points in space, thereby further affecting the calculation of the comprehensive coefficient.

[0040] Risk level judgment unit: used to judge the risk level of the slope;

[0041] The risk level judgment unit includes:

[0042] Multimodal data fusion module: The multimodal data fusion module introduces the slope surface temperature change, acoustic emission signal monitoring data, vibration monitoring quantity and comprehensive coefficient for multimodal data fusion, and uses the convolutional neural network in deep learning for data fusion, automatically extracts features and integrates the relationship between different types of data. The calculation formula of the multimodal fusion module is: ; R is the risk index, which is used to quantitatively assess the risk level of slope landslide; n represents the number of monitoring points; Indicates that the accumulation operation is performed on n monitoring points, i takes values ​​from 1 to n in sequence, covering all monitoring points; represents the weight coefficient of the i-th monitoring point, which is determined by the importance and location of the monitoring point; V i represents the vibration monitoring value of the ith monitoring point, C i represents the comprehensive coefficient of the i-th monitoring point, T i represents the temperature change of the i-th monitoring point, S i represents the acoustic emission intensity of the ith monitoring point, is the weight coefficient of the corresponding data, which is determined by the hierarchical analysis method based on historical data and expert experience.

[0043] Dynamic risk level classification module: The dynamic risk level classification module uses a fuzzy logic system to automatically adjust the risk level classification standards according to the changing trend of the fused risk indicators. For example, when the monitoring data shows that the slope stability is gradually declining, the risk level classification standards are tightened; when the slope stability improves, the standards are appropriately relaxed. The dynamic adjustment process should be timely and accurate, and be able to adapt to the dynamic changes in the slope status. The risk level is set to four levels: low risk, medium risk, high risk, and extremely high risk. The threshold interval is set to divide and the mean of the fused risk indicator is calculated. and standard deviation ;

[0044] when Low risk when

[0045] when The risk is medium.

[0046] when When the risk is high;

[0047] when When the risk is extremely high;

[0048] Early warning decision module: The early warning decision module adopts reinforcement learning algorithm to optimize the early warning decision according to the actual effect after the early warning. When the risk level increases, the system automatically analyzes the cause and provides corresponding recommended measures.

[0049] Specifically, firstly, according to the topography and geomorphology of the slope, the geological structure and the areas that may be affected by blasting, multiple monitoring points are reasonably arranged, focusing on potential landslide areas, weak rock structure, and key parts of the top and bottom of the slope. The density of monitoring points is adjusted according to the risk level of the slope. At the same time, the geological differences and the degree of blasting impact in different areas are considered, and a detailed layout is carried out by level and region. High-precision vibration sensors, displacement sensors and stress sensors are selected to monitor the vibration, displacement and stress changes of the slope in real time. The sensor data is collected in real time using a data collector. By adopting LoRa wireless communication technology, the corresponding gateways and terminal nodes are configured to transmit the collected data to the monitoring center.

[0050] Then, the formation characteristics, slope geological conditions, rock mass characteristics, rock mass structure, slope height and slope gradient were quantified to construct a dynamic matrix to reflect the values ​​of the monitoring points under different indicators, and the mutual influence between different monitoring points was considered. The spatial correlation matrix was introduced, and the weight of each indicator was determined by combining the principal component analysis method with the genetic algorithm. The importance of each indicator to the slope stability was comprehensively considered, and the comprehensive coefficient of each monitoring point was calculated using the formula;

[0051] Finally, the slope surface temperature change, acoustic emission signal monitoring data, vibration monitoring quantity and comprehensive coefficient are introduced for multimodal data fusion, and the convolutional neural network in deep learning is used for data fusion to automatically extract features and integrate the relationship between different types of data. The fuzzy logic system is used to automatically adjust the risk level classification standard according to the changing trend of the fused risk indicators. The risk level is set to four levels: low risk, medium risk, high risk and extremely high risk. The mean and standard deviation of the fused risk indicators are calculated and the threshold interval is set for classification. The reinforcement learning algorithm is used to optimize the warning decision according to the actual effect after the warning. When the risk level increases, the system automatically analyzes the cause and provides corresponding recommended measures.

[0052] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0053] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A vibration monitoring and early warning system for landslides on high slopes of lower reservoirs, characterized in that: include: A monitoring unit, which is used to obtain relevant data of the slope in a comprehensive and real-time manner, and includes a monitoring point layout module, a sensor configuration module, and a data acquisition and transmission module; A comprehensive coefficient calculation unit, which is used to comprehensively evaluate the stability of the slope, includes: Index quantification module: used to quantify stratum characteristics, slope geological conditions, rock mass characteristics, rock mass structure, slope height and slope gradient; Matrix construction and weight distribution module: used to construct a dynamic matrix to reflect the values ​​of monitoring points under different indicators, and consider the mutual influence between different monitoring points. The spatial correlation matrix is ​​introduced. The spatial correlation matrix should accurately reflect the spatial relationship and interaction between monitoring points. The weight of each indicator is determined by combining principal component analysis with genetic algorithm, and the importance of each indicator to slope stability is comprehensively considered. Comprehensive coefficient calculation module: using formula Calculate the comprehensive coefficient of each monitoring point, where represents the comprehensive coefficient of the i-th monitoring point, This part mainly reflects the traditional comprehensive calculation based on indicator weights and the values ​​of monitoring points under each indicator. Spatial correlation is partially introduced to consider the differences between different monitoring points and the impact of their relationship on the comprehensive coefficient of the current monitoring point; Risk level judgment unit: used to judge the risk level of the slope. The risk level judgment unit includes a multimodal data fusion module, a dynamic risk level classification module and an early warning decision module.

2. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: The monitoring point layout module arranges multiple monitoring points according to the slope topography, geological structure and areas that may be affected by blasting, focusing on potential landslide areas, weak rock structure, key parts of the top and bottom of the slope, and adjusts the density of monitoring points according to the slope risk level.

3. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: The sensor configuration module uses high-precision vibration sensors, displacement sensors and stress sensors to monitor the vibration, displacement and stress changes of the slope in real time.

4. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: The data acquisition and transmission module uses a data collector to collect sensor data in real time. The data collector uses LoRa wireless communication technology, configures corresponding gateways and terminal nodes, and transmits the collected data to the monitoring center.

5. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: The formation properties include lithology, hardness, permeability and formation age. The geological conditions of the slope increase the quantification of groundwater activity, including the impact of groundwater level, water flow velocity and water quality on slope stability. The rock mass properties include strength, elastic modulus and degree of anisotropy. The rock mass properties are determined comprehensively through laboratory testing, on-site monitoring and numerical simulation methods. The rock mass structure includes the density, openness, connectivity of joints and fissures as well as their direction and distribution pattern. The slope height is quantified in intervals, and the risk growth rate in different intervals is different. The slope gradient considers the inclination angle and slope change rate.

6. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: In the comprehensive coefficient calculation module, the formula middle is the weight optimized by genetic algorithm, is the value of the i-th monitoring point under the j-th indicator.

7. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: In the comprehensive coefficient calculation module, the formula Where k is the spatial correlation coefficient, which is used to adjust the influence of spatial correlation in the calculation of comprehensive coefficient. Indicates that all other monitoring points except the i-th monitoring point are summed. It means the sum of six indicators: stratum characteristics, slope geological conditions, rock mass characteristics, rock mass structure, slope height and slope gradient. is the spatial correlation coefficient between the mth and nth monitoring points, reflecting the similarity between different monitoring points in various indicators. It represents the difference between the i-th monitoring point and other monitoring points in the n-th indicator.

8. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: The multimodal data fusion module introduces the slope surface temperature change, acoustic emission signal monitoring data, vibration monitoring quantity and comprehensive coefficient for multimodal data fusion, uses the convolutional neural network in deep learning for data fusion, automatically extracts features and integrates the relationship between different types of data. The calculation formula of the multimodal fusion module is: ; R is the risk index, which is used to quantitatively assess the risk level of slope landslide; n represents the number of monitoring points; Indicates that the accumulation operation is performed on n monitoring points, i takes values ​​from 1 to n in sequence, covering all monitoring points; represents the weight coefficient of the i-th monitoring point, which is determined by the importance and location of the monitoring point; V i represents the vibration monitoring value of the ith monitoring point, C i represents the comprehensive coefficient of the ith monitoring point, T i represents the temperature change of the i-th monitoring point, S i represents the acoustic emission intensity of the ith monitoring point, is the weight coefficient of the corresponding data, which is determined by the hierarchical analysis method based on historical data and expert experience.

9. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: The dynamic risk level classification module uses the fuzzy logic system to automatically adjust the risk level classification criteria according to the changing trend of the fused risk indicators. The risk level is set to four levels: low risk, medium risk, high risk, and extremely high risk. The threshold interval is set to divide and calculate the mean of the fused risk indicators. and standard deviation ; when Low risk when when The risk is medium. when When the risk is high; when Very high risk.

10. A vibration monitoring and early warning system for landslides on a lower reservoir high slope as claimed in claim 1, characterized in that: The early warning decision module adopts a reinforcement learning algorithm to optimize the early warning decision according to the actual effect after the early warning. When the risk level increases, the system automatically analyzes the cause and provides corresponding recommended measures.

Images