Mountain highway deep foundation pit and high slope multi-source heterogeneous data monitoring and early warning method

Abstract

The application discloses a mountainous area highway deep foundation pit and high slope multi-source heterogeneous data monitoring and early warning method, relates to the technical field of mountainous area highway construction, and acquires an electronic distribution map of a mountainous area highway deep foundation pit and high slope, divides the electronic distribution map into multiple monitoring areas, collects relevant geological data, meteorological environment data and structural data, and establishes corresponding data sets. Then, the geological stability coefficient is calculated, and the geological instability risk is evaluated, and if the geological instability risk exists, a first risk signal is sent. After receiving the first signal, the groundwater inrush index is further calculated in combination with the meteorological environment data, and the permeation channel risk is evaluated. Then, the deep foundation pit collapse risk coefficient and the high slope inclination deformation risk coefficient are calculated, and the emergency strategy is generated according to the preset threshold. Finally, the convolutional neural network is used to construct a risk dynamic early warning model, training and testing are carried out, and real-time risk prediction is provided.

Description

Technical Field

[0001] This invention relates to the field of mountain highway construction technology, specifically to a method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountain highways. Background Technology

[0002] With the acceleration of urbanization and the continuous development of infrastructure construction, the safety of engineering projects such as mountain roads, tunnels, deep foundation pits, and high slopes has increasingly attracted attention. Especially in complex terrain areas such as mountain roads, risk management of deep foundation pits and high slopes has become particularly important. These engineering projects not only face uncertainties during construction but are also affected by factors such as climate change, geological conditions, and the external environment. Particularly under extreme weather conditions such as rainfall and temperature fluctuations, the stability of deep foundation pits and high slopes may be severely threatened.

[0003] Traditional methods for monitoring the risks of deep foundation pits and high slopes typically focus on monitoring a single factor, such as soil pressure, water level changes, or slope displacement. However, these individual monitoring data cannot comprehensively assess and predict the overall risk of deep foundation pits and high slopes, potentially causing important risk signals to be missed in complex environments. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a multi-source heterogeneous data monitoring and early warning method for deep foundation pits and high slopes in mountainous highways, in order to solve the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention provides a method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways, comprising the following steps:

[0006] Preferably, in step one, an electronic distribution map of deep foundation pits and high slopes of mountain roads is obtained in advance, and the deep foundation pits and high slopes of mountain roads are divided into several monitoring areas. Real-time geological data of deep foundation pits and high slopes of mountain roads in the monitoring areas are collected, and a first dataset is established; meteorological environmental data of the monitoring areas are collected, and a second dataset is established; structural data of deep foundation pits and high slopes of mountain roads in the monitoring areas are collected, and a third dataset is established.

[0007] Step 2: Analyze the first dataset, monitor the geological conditions of the deep foundation pit and high slope in the i-th monitoring area, and calculate the geological stability coefficient DZxs for the i-th monitoring area. i A first standard threshold P is preset and the geological stability coefficient DZxs of the i-th monitoring area is set in advance. i Comparative analysis is conducted to assess the geological risk of the i-th monitoring area. If there is a risk of geological instability, a first risk signal is issued.

[0008] Step 3: After receiving the first risk signal, further monitoring is conducted on the real-time rainfall Prain, real-time groundwater level Dxw, groundwater pressure T, groundwater flow Q, water flow path length L, groundwater flow cross-sectional area Ad, hydraulic head difference Δh, and groundwater permeability Flz of the i-th monitoring area, combined with the data from the second dataset. The groundwater inrush index QXxs for the i-th monitoring area is then calculated. i A second standard threshold Q is pre-set for comparative analysis to assess whether there are seepage channels forming in geological areas that could induce water inrush risk in monitoring areas with geological instability risks. If there are geological instability risks and seepage channels forming in geological areas that could induce water inrush risk, a second risk signal is issued.

[0009] Step 4: After receiving the second risk signal, combine it with the geological stability coefficient DZxs of the i-th monitoring area. i The groundwater inrush index QXxs in the i-th monitoring area i Using the data from the third dataset, calculate the deep foundation pit collapse risk coefficient Sfx for the i-th monitoring area. i , and the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i And by comparing and analyzing the preset third standard threshold Z1 and fourth standard threshold Z2, a corresponding strategy is generated.

[0010] Step 5: Using a convolutional neural network, construct an initial convolutional neural network model. Train and test the risk dynamic early warning model using the acquired feature information. Use the trained risk dynamic early warning model as data to run predictions and perform real-time risk predictions for deep foundation pits and high slopes on mountain roads.

[0011] Preferably, step one includes:

[0012] S11. Using GIS geographic information system technology and related surveying and mapping data, obtain electronic distribution maps of deep foundation pits and high slopes along mountain roads;

[0013] The area of ​​deep foundation pits and high slopes along mountain roads was divided into several monitoring zones, and these zones were marked as Jcq1, Jcq2, ..., Jcq on the electronic distribution map of deep foundation pits and high slopes along mountain roads. n n represents the number of monitoring areas;

[0014] S12. Collect real-time geological data of deep foundation pits and high slopes of mountain roads in the monitoring area, and establish the first dataset including: compressive strength σ of soil and rock layers in the monitoring area. cThe dataset includes: soil hardness (H), soil thickness (d), and soil moisture content (w) in the soil layer; meteorological environmental data of the monitoring area, and a second dataset including: real-time rainfall (Prain), real-time groundwater level (Dxw), groundwater pressure (T), and groundwater infiltration (FLz) in the monitoring area; and structural data of deep foundation pits and high slopes in mountainous areas of the monitoring area, and a third dataset including: bearing capacity (P) of the deep foundation pit support structure in the monitoring area. support The load value W of deep foundation pit structures build Area A of the deep foundation pit; shear strength σ of the soil on the high slope of the monitoring area. soil The height Hg of the high slope, the angle θ of the high slope, and the load value W of the high slope. load And the base area Ag of the high slope.

[0015] Preferably, step two includes:

[0016] S21. Using the data from the first dataset, monitor the geological conditions of the deep foundation pit height and slope. After dimensionless processing, calculate the geological stability coefficient DZxs for the i-th monitoring area. i The formula is as follows:

[0017]

[0018] In the formula, n i σ represents the number of soil and rock layers in the i-th monitoring area. cj,i σ represents the compressive strength of the j-th soil layer in the i-th monitoring area. max H represents the maximum compressive strength of the soil and rock layer. j,i d represents the soil hardness of the j-th soil layer in the i-th monitoring area. j,i w represents the thickness of the j-th soil layer in the i-th monitoring area. j,i denoted by k, represents the soil moisture content of the j-th soil layer in the i-th monitoring area, k1 represents the influence coefficient of soil moisture content, and k2 represents the influence coefficient of soil pressure. k1 and k2 are obtained through experimental data.

[0019] Preferably, step two also includes:

[0020] S22. Pre-set the first standard threshold P and set the geological stability coefficient DZxs of the i-th monitoring area. i A comparative analysis is performed with the first standard threshold P to obtain the first evaluation result, including:

[0021] When the geological stability coefficient DZxs of the i-th monitoring area i When the value is ≥ the first standard threshold P, it indicates that there is no risk of geological instability in the current monitoring area;

[0022] When the geological stability coefficient DZxs of the i-th monitoring areai When the value is less than the first standard threshold P, it indicates that there is a risk of geological instability in the current monitoring area, triggering the first early warning instruction and generating the first risk signal.

[0023] Preferably, step three includes:

[0024] S31. After receiving the first risk signal, and combining the data from the second dataset on the precipitation, groundwater level, temperature, and wind speed of the i-th monitoring area, further monitoring of meteorological and environmental interference is conducted for areas with geological instability risk. After dimensionless processing, the groundwater inrush index QXxs of the i-th monitoring area is calculated and obtained. i The formula is as follows:

[0025]

[0026] In the formula, Flz i This represents the groundwater permeability value of the i-th monitoring area, where K is the permeability coefficient, including: for sandy soil: K = 10. -4 ~10 -3 cm / s; for clay: K = 10 -8 ~10 -8 cm / s;

[0027] Q is the groundwater flow rate, L is the flow path length, i.e. the distance from the inflow point to the outflow point; Ad is the cross-sectional area of ​​the groundwater flow, measured by vertical permeability testing, and Δh is the head difference.

[0028] n represents the number of monitoring areas, i represents the i-th monitoring area, and Prain i Prain represents the real-time rainfall in the i-th monitoring area. max Dxw represents the maximum rainfall that the geology can withstand. i Dxw represents the real-time groundwater level height value of the i-th monitoring area. max T represents the highest groundwater level that the geology can withstand. i T represents the groundwater pressure value in the i-th monitoring area. max Flz represents the maximum groundwater pressure that the geology can withstand. i Flz represents the groundwater permeability value of the i-th monitoring area. max This represents the maximum groundwater infiltration value that the geology can withstand, and w1, w2, w3, and w4 represent weighting coefficients.

[0029] Preferably, step three also includes:

[0030] S32. Pre-set the second standard threshold K and set the groundwater inrush index QXxs of the i-th monitoring area. iA second evaluation result is obtained by comparing the result with a second standard threshold K, including:

[0031] When the groundwater inrush index of the i-th monitoring area is QXxs i When the value is ≤ the second standard threshold K, it indicates that there is no risk of geological instability in the current monitoring area. A seepage channel has formed in the geological area, which could induce a water inrush risk. Continuous monitoring is required.

[0032] When the groundwater inrush index of the i-th monitoring area is QXxs i When the threshold K is greater than the second standard threshold, it indicates that there is a risk of geological instability in the current monitoring area and that a seepage channel has formed in the geological area, which may induce a risk of water inrush, triggering a second early warning instruction and generating a second risk signal.

[0033] Preferably, step four includes:

[0034] S41. After receiving the second risk signal, combine it with the geological stability coefficient DZxs of the i-th monitoring area. i The groundwater inrush index QXxs in the i-th monitoring area i After dimensionless processing of the data from the third dataset, the collapse risk coefficient Sfx of the deep foundation pit in the i-th monitoring area is calculated and obtained. i , and the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i The formula is as follows:

[0035]

[0036] In the formula, n represents the number of monitoring areas, and P supporti DZxs represents the bearing capacity value of the deep foundation pit support structure in the i-th monitoring area. i Let Q represent the geological stability coefficient of the i-th monitoring area. i PS represents the groundwater inrush index for the i-th monitoring area. loadi W represents the load factor of the deep foundation pit in the i-th monitoring area. buildi A represents the load value of the deep foundation pit structure in the i-th monitoring area. i This represents the area of ​​the deep foundation pit in the i-th monitoring area;

[0037]

[0038] In the formula, σ soili Hg represents the shear strength of the high slope soil in the i-th monitoring area. i θ represents the height of the high slope in the i-th monitoring area. i PG represents the angle of the high slope in the i-th monitoring area. loadi W represents the load coefficient of the high slope in the i-th monitoring area. loadiAg represents the load value of the i-th monitoring area and the high slope. i This represents the base area of ​​the high slope in the i-th monitoring area.

[0039] Preferably, step four also includes:

[0040] S42. Pre-set the third standard threshold Z1 and set the deep foundation pit collapse risk coefficient Sfx for the i-th monitoring area. i A comparative evaluation is performed with the third standard threshold Z1 to obtain the third evaluation result, including:

[0041] When the deep foundation pit collapse risk coefficient Sfx in the i-th monitoring area i When the value is less than the third standard threshold Z1, it indicates that there is no risk of collapse in the deep foundation pit, and monitoring should continue.

[0042] When the deep foundation pit collapse risk coefficient Sfx in the i-th monitoring area i When the value is ≥ the third standard threshold Z1, it indicates that the deep foundation pit is at risk of collapse, triggering the third early warning instruction and generating the first strategy, which includes: increasing the compressive strength of the soil around the foundation pit by 10% and improving the soil layer stability by 20% through deep mixing; reinforcing the loose soil layer through grouting technology and increasing the soil bearing capacity by 30%; and building temporary shelters and setting up two drainage points to reduce the impact of precipitation on the foundation pit by 70%.

[0043] Preferably, step four also includes:

[0044] S43. Pre-set the fourth standard threshold Z2, and set the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area. i A comparative evaluation is performed with the fourth standard threshold Z2 to obtain the fourth evaluation result, including:

[0045] When the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i <The fourth standard threshold Z2 indicates that there is no risk of tilting or deformation of the high slope, and continuous monitoring is required;

[0046] When the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i When the value is ≥ the fourth standard threshold Z2, it indicates that the high slope is at risk of tilting and deformation, triggering the fourth early warning instruction and generating the second strategy, which includes: reinforcing and increasing the support bearing capacity of the deep foundation pit retaining wall by 20%, increasing the anchor density by 25%, and increasing the strength of the reinforced concrete retaining wall by 10%; reducing water accumulation by 40% and reducing water seepage pressure on the soil by 25% by setting up two drainage points; and strengthening the bearing capacity of loose soil and increasing the tensile strength of the soil by 35% through shotcreting technology.

[0047] Preferably, step five includes:

[0048] S51. Using a convolutional neural network, construct an initial model of the convolutional neural network, and use the deep foundation pit collapse risk coefficient Sfx of the i-th monitoring area. i The tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i The initial convolutional neural network model was trained and tested, and the trained initial convolutional neural network model was used as the dynamic risk early warning model. Simultaneously, the deep foundation pit collapse risk coefficient Sfx for the i-th monitoring area was used. i The tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i The intermediate layer output is used as a feature vector to identify feature information. The acquired feature information is used to train and test the dynamic risk warning model. The trained dynamic risk warning model is then used as data to run predictions and perform real-time risk predictions for deep foundation pits and high slopes on mountain roads.

[0049] This invention provides a multi-source heterogeneous data monitoring and early warning method for deep foundation pits and high slopes in mountainous highways. It has the following beneficial effects:

[0050] (1) The multi-source heterogeneous data monitoring and early warning method for deep foundation pits and high slopes of mountain highways uses multi-source heterogeneous data, including geological data, meteorological environmental data and structural data, to comprehensively assess the multiple risks of deep foundation pits and high slopes of mountain highways. It can accurately determine the risk level of different monitoring areas, thereby achieving more efficient and accurate early warning and avoiding the omission of potential safety hazards.

[0051] (2) The multi-source heterogeneous data monitoring and early warning method for deep foundation pits and high slopes in mountainous areas dynamically adjusts risk assessment parameters based on different geological environments and meteorological conditions. It calculates key indicators such as the groundwater inrush index and geological stability coefficient by combining multiple factors such as precipitation, groundwater pressure, and soil moisture content, thereby comprehensively assessing the risks of high slopes and deep foundation pits. Especially under extreme weather or complex geological conditions, it can promptly identify potential water inrush risks or geological instability risks and take preventative measures in advance.

[0052] (3) The multi-source heterogeneous data monitoring and early warning method for deep foundation pits and high slopes in mountainous highways comprehensively analyzes real-time data from the monitoring area, calculates the collapse risk coefficient of deep foundation pits and the tilt deformation risk coefficient of high slopes respectively, and performs analysis and evaluation to automatically generate targeted early warning strategies. When a collapse risk occurs in a deep foundation pit, the system can automatically suggest ways to improve the stability of the foundation pit, such as deep mixing, grouting reinforcement technology, and temporary shielding facilities, and can provide specific reinforcement values. This intelligent strategy generation not only improves the efficiency of protective measures but also allows for flexible adjustments based on actual conditions.

[0053] (4) The multi-source heterogeneous data monitoring and early warning method for deep foundation pits and high slopes in mountainous highways, through the establishment of a convolutional neural network model, enables real-time dynamic monitoring and prediction of risks. By combining real-time acquisition of feature information and training to continuously optimize the early warning model, the accuracy of prediction is improved, the reliance on manual intervention is reduced, and the automation and efficiency of risk management are ensured. Through systematic risk assessment and early warning measures, the safety management level of the project is improved, and data support and reference are provided for subsequent project maintenance and upgrades, demonstrating strong sustainability. Attached Figure Description

[0054] Figure 1 This is a schematic diagram illustrating the steps of the multi-source heterogeneous data monitoring and early warning method for deep foundation pits and high slopes in mountainous highways according to the present invention.

[0055] Figure 2 A block diagram of a computing device capable of implementing several embodiments of the present disclosure is shown. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] Example 1

[0058] Please see Figure 1 This invention provides a method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways, comprising the following steps:

[0059] Step 1: First, obtain electronic distribution maps of deep foundation pits and high slopes along mountain roads, and divide the deep foundation pits and high slopes along mountain roads into several monitoring areas. Collect real-time geological data of deep foundation pits and high slopes along mountain roads in the monitoring areas and establish the first dataset. Collect meteorological environmental data of the monitoring areas and establish the second dataset. Collect structural data of deep foundation pits and high slopes along mountain roads in the monitoring areas and establish the third dataset.

[0060] Step 2: Analyze the first dataset, monitor the geological conditions of the deep foundation pit and high slope in the i-th monitoring area, and calculate the geological stability coefficient DZxs for the i-th monitoring area. i A first standard threshold P is preset and the geological stability coefficient DZxs of the i-th monitoring area is set in advance. i Comparative analysis is conducted to assess the geological risk of the i-th monitoring area. If there is a risk of geological instability, a first risk signal is issued.

[0061] Step 3: After receiving the first risk signal, further monitoring is conducted on the real-time rainfall Prain, real-time groundwater level Dxw, groundwater pressure T, groundwater flow Q, water flow path length L, groundwater flow cross-sectional area Ad, hydraulic head difference Δh, and groundwater permeability Flz of the i-th monitoring area, combined with the data from the second dataset. The groundwater inrush index QXxs for the i-th monitoring area is then calculated. i A second standard threshold Q is pre-set for comparative analysis to assess whether there are seepage channels forming in geological areas that could induce water inrush risk in monitoring areas with geological instability risks. If there are geological instability risks and seepage channels forming in geological areas that could induce water inrush risk, a second risk signal is issued.

[0062] Step 4: After receiving the second risk signal, combine it with the geological stability coefficient DZxs of the i-th monitoring area. i The groundwater inrush index QXxs in the i-th monitoring area i Using the data from the third dataset, calculate the deep foundation pit collapse risk coefficient Sfx for the i-th monitoring area. i , and the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i And by comparing and analyzing the preset third standard threshold Z1 and fourth standard threshold Z2, a corresponding strategy is generated.

[0063] Step 5: Using a convolutional neural network, construct an initial convolutional neural network model. Train and test the risk dynamic early warning model using the acquired feature information. Use the trained risk dynamic early warning model as data to run predictions and perform real-time risk predictions for deep foundation pits and high slopes on mountain roads.

[0064] In this embodiment, a comprehensive assessment of multiple risks associated with deep foundation pits and high slopes in mountainous highways is conducted using geological, meteorological, and structural data. This allows for precise determination of risk levels in different monitoring areas, leading to more efficient and accurate early warning systems and preventing the omission of potential safety hazards. Risk assessment parameters are dynamically adjusted based on varying geological environments and meteorological conditions. By combining factors such as precipitation, groundwater pressure, and soil moisture content, key indicators like the groundwater inrush index and geological stability coefficient are calculated to comprehensively assess the risks of high slopes and deep foundation pits. The collapse risk coefficient of deep foundation pits and the tilt deformation risk coefficient of high slopes are then calculated, analyzed, and evaluated, automatically generating targeted early warning strategies. When a collapse risk occurs in a deep foundation pit, the system automatically suggests methods to improve its stability, such as deep mixing, grouting reinforcement, and temporary shielding facilities, and provides specific reinforcement values. A convolutional neural network model is established to enable real-time dynamic monitoring and prediction of risks. By combining real-time acquisition of feature information and continuous training to optimize the early warning model, we can ensure that timely alerts are issued and effective response strategies are provided when sudden risk events occur, thereby reducing the occurrence of disasters and losses.

[0065] Example 2

[0066] This embodiment is an explanation based on Embodiment 1. Please refer to [link / reference]. Figure 1 Specifically, step one includes:

[0067] S11. Using GIS geographic information system technology and related surveying and mapping data, obtain electronic distribution maps of deep foundation pits and high slopes along mountain roads;

[0068] The area of ​​deep foundation pits and high slopes along mountain roads was divided into several monitoring zones, and these zones were marked as Jcq1, Jcq2, ..., Jcq on the electronic distribution map of deep foundation pits and high slopes along mountain roads. n n represents the number of monitoring areas;

[0069] S12. Collect real-time geological data of deep foundation pits and high slopes of mountain roads in the monitoring area, and establish the first dataset, including: monitoring the compressive strength σ of the soil and rock layers by installing a soil mechanics testing instrument in the monitoring area. cThe hardness H of the soil in the rock and soil layer was measured by installing electromagnetic wave radar to measure the thickness d of the rock and soil layer, and the water content w of the soil in the rock and soil layer was measured by installing a soil moisture meter; meteorological environmental data of the monitoring area were collected, and a second dataset was established, including: monitoring rainfall Prain by installing tipping bucket rain gauges in the monitoring area, and monitoring real-time groundwater level Dxw, groundwater flow Q, water flow path length L, groundwater flow cross-sectional area Ad, hydraulic head difference Δh, groundwater pressure T, and groundwater permeability FLz by installing groundwater pressure sensors; structural data of deep foundation pits and high slopes of mountain roads in the monitoring area were collected, and a third dataset was established, including: monitoring the bearing capacity P of the deep foundation pit support structure by installing load sensors in the deep foundation pits in the monitoring area. support The load value W of deep foundation pit structures build The area A of the deep foundation pit was collected using an electronic distribution map, and the shear strength σ of the soil on the high slope was monitored by installing a shear strength measuring instrument on the high slope. soil The height Hg, angle θ, and base area Ag of the high slope are collected through electronic distribution maps. The load value W of the high slope is monitored by installing load sensors at the bottom of the slope. load .

[0070] In this embodiment, by utilizing GIS (Geographic Information System) technology to construct an electronic distribution map of deep foundation pits and high slopes along mountain roads, precise delineation of complex terrain can be achieved. Dividing the monitoring area into multiple specific regions allows for more detailed and independent risk data for each region, avoiding the inaccurate or missed risks caused by the broad coverage of traditional monitoring methods, thereby improving overall monitoring efficiency and accuracy. This method not only collects geological data such as soil compressive strength, soil hardness, thickness, and moisture content, but also incorporates meteorological data such as rainfall, groundwater pressure, and permeability, as well as structural data such as the bearing capacity and load values ​​of supporting structures. This multi-dimensional data collection provides a comprehensive understanding of the environment, structure, and meteorological conditions of deep foundation pits and high slopes along mountain roads, providing more accurate input for subsequent risk assessment and effectively improving the comprehensiveness and completeness of the monitoring data.

[0071] Example 3

[0072] This embodiment is an explanation based on Embodiment 2. Please refer to [link / reference]. Figure 1 Specifically, step two includes:

[0073] S21. Using the data from the first dataset, monitor the geological conditions of the deep foundation pit height and slope. After dimensionless processing, calculate the geological stability coefficient DZxs for the i-th monitoring area. i The formula is as follows:

[0074]

[0075] In the formula, n i σ represents the number of soil and rock layers in the i-th monitoring area. cj,i σ represents the compressive strength of the j-th soil layer in the i-th monitoring area. max H represents the maximum compressive strength of the soil and rock layer. j,i d represents the soil hardness of the j-th soil layer in the i-th monitoring area. j,i w represents the thickness of the j-th soil layer in the i-th monitoring area. j,i Let k represent the soil moisture content of the j-th soil layer in the i-th monitoring area, k1 represent the influence coefficient of soil moisture content, and k2 represent the influence coefficient of soil pressure. k1 and k2 are obtained through experimental data, and commonly used values ​​are shown in the table below:

[0076]

[0077]

[0078] In this embodiment, by comprehensively considering multiple geological indicators such as the compressive strength, hardness, thickness, and soil moisture content of the soil and rock layers, the calculated geological stability coefficient can comprehensively reflect the geological stability of each monitoring area. This method can effectively identify potential risk areas and predict possible slope landslides or foundation pit collapses in advance. Accurate calculation of the influence coefficients of soil moisture content and soil pressure based on experimental data allows the monitoring system to account for complex changes in soil conditions. Specific influence coefficients for different moisture content ranges and pressure ranges ensure more accurate monitoring results under different environments, providing a reliable scientific basis for engineering decisions.

[0079] Example 4

[0080] This embodiment is an explanation based on Embodiment 1. Please refer to [link / reference]. Figure 1 Specifically, step two also includes:

[0081] S22. Pre-set the first standard threshold P and set the geological stability coefficient DZxs of the i-th monitoring area. i A comparative analysis is performed with the first standard threshold P to obtain the first evaluation result, including:

[0082] When the geological stability coefficient DZxs of the i-th monitoring area i When the value is ≥ the first standard threshold P, it indicates that there is no risk of geological instability in the current monitoring area;

[0083] When the geological stability coefficient DZxs of the i-th monitoring area i When the value is less than the first standard threshold P, it indicates that there is a risk of geological instability in the current monitoring area, triggering the first early warning instruction and generating the first risk signal.

[0084] In this embodiment, the stability of the monitored area can be assessed by calculating the geological stability coefficient and comparing it with a pre-set threshold, thus providing scientific guidance for resource allocation and protective measures during construction. For areas with poor stability, an early warning signal is issued, allowing for increased reinforcement or enhanced dewatering measures to ensure the safety of personnel and equipment during construction.

[0085] Example 5

[0086] This embodiment is an explanation based on Embodiment 1. Please refer to [link / reference]. Figure 1 Specifically, step three includes:

[0087] S31. After receiving the first risk signal, and combining the data from the second dataset on the precipitation, groundwater level, temperature, and wind speed of the i-th monitoring area, further monitoring of meteorological and environmental interference is conducted for areas with geological instability risk. After dimensionless processing, the groundwater inrush index QXxs of the i-th monitoring area is calculated and obtained. i The formula is as follows:

[0088]

[0089] In the formula, Flz i This represents the groundwater permeability value of the i-th monitoring area, where K is the permeability coefficient, including: for sandy soil: K = 10. -4 ~10 -3 cm / s; for clay: K = 10 -8 ~10 -8 cm / s;

[0090] Q is the groundwater flow rate, L is the flow path length, i.e. the distance from the inflow point to the outflow point; Ad is the cross-sectional area of ​​the groundwater flow, measured by vertical permeability testing, and Δh is the head difference.

[0091] n represents the number of monitoring areas, i represents the i-th monitoring area, and Prain i Prain represents the real-time rainfall in the i-th monitoring area. max Dxw represents the maximum rainfall that the geology can withstand. i Dxw represents the real-time groundwater level height value of the i-th monitoring area. max T represents the highest groundwater level that the geology can withstand. i T represents the groundwater pressure value in the i-th monitoring area. max Flz represents the maximum groundwater pressure that the geology can withstand. i Flz represents the groundwater permeability value of the i-th monitoring area. maxThe value represents the maximum groundwater infiltration that the geology can withstand. w1, w2, w3, and w4 represent weighting coefficients, where 0 < w1 < 1, 0 < w2 < 1, 0 < w3 < 1, and 0 < w4 < 1, and w1 + w2 + w3 + w4 = 1.

[0092] In this embodiment, real-time meteorological environmental data can reflect the impact of the current meteorological environment on geological stability. When abnormal weather is detected that causes a rise in groundwater level or an increase in groundwater flow, the groundwater inrush index is calculated. This accurately reflects the impact of groundwater on the stability of deep foundation pits and high slope areas. The system can issue timely warnings and identify potential risk areas with unstable geology in advance, thereby taking necessary protective measures to avoid accidents.

[0093] Example 6

[0094] This embodiment is an explanation based on Embodiment 1. Please refer to [link / reference]. Figure 1 Specifically, step three also includes:

[0095] S32. Pre-set the second standard threshold K and set the groundwater inrush index QXxs of the i-th monitoring area. i A second evaluation result is obtained by comparing the result with a second standard threshold K, including:

[0096] When the groundwater inrush index of the i-th monitoring area is QXxs i When the value is ≤ the second standard threshold K, it indicates that there is no risk of geological instability in the current monitoring area. A seepage channel has formed in the geological area, which could induce a water inrush risk. Continuous monitoring is required.

[0097] When the groundwater inrush index of the i-th monitoring area is QXxs i When the threshold K is greater than the second standard threshold, it indicates that there is a risk of geological instability in the current monitoring area and that a seepage channel has formed in the geological area, which may induce a risk of water inrush, triggering a second early warning instruction and generating a second risk signal.

[0098] In this embodiment, by comparing and analyzing a pre-set threshold with the groundwater inrush index, specific risk values ​​can be provided to on-site engineers and decision-makers. These values ​​directly influence the formulation of emergency measures. An early warning signal is issued when groundwater pressure or permeability reaches a critical value. Measures such as timely reinforcement of supporting structures, enhancement of drainage facilities, or temporary sealing can be implemented to prevent sudden groundwater flow from threatening the foundation pit and slopes.

[0099] Example 7

[0100] This embodiment is an explanation based on Embodiment 1. Please refer to [link / reference]. Figure 1 Specifically, step four includes:

[0101] S41. After receiving the second risk signal, combine it with the geological stability coefficient DZxs of the i-th monitoring area. i The groundwater inrush index QXxs in the i-th monitoring area i After dimensionless processing of the data from the third dataset, the collapse risk coefficient Sfx of the deep foundation pit in the i-th monitoring area is calculated and obtained. i , and the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i The formula is as follows:

[0102]

[0103] In the formula, n represents the number of monitoring areas. DZxs represents the bearing capacity value of the deep foundation pit support structure in the i-th monitoring area. i Let Q represent the geological stability coefficient of the i-th monitoring area. i This represents the groundwater inrush index for the i-th monitoring area. This represents the load factor of the deep foundation pit in the i-th monitoring area. A represents the load value of the deep foundation pit structure in the i-th monitoring area. i This represents the area of ​​the deep foundation pit in the i-th monitoring area;

[0104]

[0105] In the formula, Hg represents the shear strength of the high slope soil in the i-th monitoring area. i θ represents the height of the high slope in the i-th monitoring area. i This represents the angle of the high slope in the i-th monitoring area. This represents the load coefficient of the high slope in the i-th monitoring area. Ag represents the load value of the i-th monitoring area and the high slope. i This represents the base area of ​​the high slope in the i-th monitoring area.

[0106] In this embodiment, by comprehensively analyzing the real-time data of the monitoring area and performing dimensionless processing, the collapse risk coefficient of deep foundation pit and the tilt deformation risk coefficient of high slope are calculated respectively. This method is more comprehensive than the traditional single-factor analysis and can comprehensively assess the potential risks under complex geological and environmental conditions, thereby improving the accuracy of risk prediction.

[0107] Example 8

[0108] This embodiment is an explanation based on Embodiment 1. Please refer to [link / reference]. Figure 1 Specifically, step four also includes:

[0109] S42. Pre-set the third standard threshold Z1 and set the deep foundation pit collapse risk coefficient Sfx for the i-th monitoring area. i A comparative evaluation is performed with the third standard threshold Z1 to obtain the third evaluation result, including:

[0110] When the deep foundation pit collapse risk coefficient Sfx in the i-th monitoring area i When the value is less than the third standard threshold Z1, it indicates that there is no risk of collapse in the deep foundation pit, and monitoring should continue.

[0111] When the deep foundation pit collapse risk coefficient Sfx in the i-th monitoring area i When the value is ≥ the third standard threshold Z1, it indicates that the deep foundation pit is at risk of collapse, triggering the third early warning instruction and generating the first strategy, which includes: increasing the compressive strength of the soil around the foundation pit by 10% and improving the soil layer stability by 20% through deep mixing; reinforcing the loose soil layer through grouting technology and increasing the soil bearing capacity by 30%; and building temporary shelters and setting up two drainage points to reduce the impact of precipitation on the foundation pit by 70%.

[0112] In this embodiment, by setting a standard threshold Z1 and comparing it with a risk coefficient, the system can issue a timely warning when the risk of a deep foundation pit exceeds the set threshold. This warning mechanism has a highly efficient response capability, can identify potential risks in advance, avoid sudden disasters, and ensure the safety of construction personnel and the project. When the risk coefficient reaches the warning standard, the system can automatically generate a first strategy and propose targeted reinforcement and protection measures, such as deep mixing, grouting technology, and the construction of temporary shelters. These measures help to directly improve the compressive strength, bearing capacity, and stability of the soil, reduce the probability of foundation pit collapse, and effectively reduce safety hazards during construction.

[0113] Example 9

[0114] This embodiment is an explanation based on Embodiment 1. Please refer to [link / reference]. Figure 1 Specifically, step four also includes:

[0115] S43. Pre-set the fourth standard threshold Z2, and set the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area. i A comparative evaluation is performed with the fourth standard threshold Z2 to obtain the fourth evaluation result, including:

[0116] When the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i <The fourth standard threshold Z2 indicates that there is no risk of tilting or deformation of the high slope, and continuous monitoring is required;

[0117] When the tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area iWhen the value is ≥ the fourth standard threshold Z2, it indicates that the high slope is at risk of tilting and deformation, triggering the fourth early warning instruction and generating the second strategy, which includes: reinforcing and increasing the support bearing capacity of the deep foundation pit retaining wall by 20%, increasing the anchor density by 25%, and increasing the strength of the reinforced concrete retaining wall by 10%; reducing water accumulation by 40% and reducing water seepage pressure on the soil by 25% by setting up two drainage points; and strengthening the bearing capacity of loose soil and increasing the tensile strength of the soil by 35% through shotcreting technology.

[0118] In this embodiment, by pre-setting a standard threshold Z2 and comparing it with the tilt deformation risk coefficient, the system can promptly identify the tilt deformation risk of high slopes. Once the risk coefficient exceeds the set threshold, the system automatically triggers an early warning mechanism, effectively preventing potential landslides or tilt deformation accidents and providing early warnings for construction and the surrounding environment. When the risk coefficient exceeds the standard threshold, the system generates a second strategy, including targeted protective measures such as reinforcing the deep foundation pit retaining wall, increasing the anchor density, and improving the strength of the retaining wall. These measures can effectively enhance the compressive strength of the soil, improve the bearing capacity of the support structure, reduce water accumulation and seepage pressure, and reduce the possibility of slope tilt deformation.

[0119] Example 10

[0120] This embodiment is an explanation based on Embodiment 1. Please refer to [link / reference]. Figure 1 Specifically, step five includes:

[0121] S51. Using a convolutional neural network, construct an initial model of the convolutional neural network, and use the deep foundation pit collapse risk coefficient Sfx of the i-th monitoring area. i The tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i The initial convolutional neural network model was trained and tested, and the trained initial convolutional neural network model was used as the dynamic risk early warning model. Simultaneously, the deep foundation pit collapse risk coefficient Sfx for the i-th monitoring area was used. i The tilt deformation risk coefficient Gfx of the high slope in the i-th monitoring area i The intermediate layer output is used as a feature vector to identify feature information. The acquired feature information is used to train and test the dynamic risk warning model. The trained dynamic risk warning model is then used as data to run predictions and perform real-time risk predictions for deep foundation pits and high slopes on mountain roads.

[0122] In this embodiment, a convolutional neural network model is established to enable real-time dynamic monitoring and prediction of risks. By combining real-time acquisition of feature information with continuous training and optimization of the early warning model, the accuracy of predictions is improved, reliance on manual intervention is reduced, and the automation and efficiency of risk management are ensured. Through systematic risk assessment and early warning measures, the safety management level of engineering projects is improved, and data support and reference are provided for subsequent engineering maintenance and upgrades, demonstrating strong sustainability.

[0123] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value, it is acceptable.

[0124] The above formulas are all derived from software simulation using a large amount of data and are selected to be close to the actual values. The coefficients in the formulas are set by those skilled in the art according to the actual situation. The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any equivalent substitutions or changes made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the protection scope of the present invention.

[0125] Figure 2 A schematic block diagram of an example device 200 that can be used to implement embodiments of the present disclosure is shown. Device 200 can be used to implement... Figure 1 One or more operations in the method. As shown, device 200 includes a central processing unit (CPU) 201, which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) 202 or loaded from storage unit 208 into random access memory (RAM) 203. RAM 203 may also store various programs and data required for the operation of device 200. CPU 201, ROM 202, and RAM 203 are interconnected via bus 204. Input / output (I / O) interface 205 is also connected to bus 204.

[0126] Multiple components in device 200 are connected to I / O interface 205, including: input unit 206, such as keyboard, mouse, etc.; output unit 207, such as various types of monitors, speakers, etc.; storage unit 208, such as disk, optical disk, etc.; and communication unit 209, such as network card, modem, wireless transceiver, etc. Communication unit 209 allows device 200 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0127] Processing unit 201 executes the various methods and processes described above, such as Figure 1 The various operations within. For example, in some embodiments, Figure 1 The various operations described above can be implemented as computer software programs, which are tangibly contained in a machine-readable medium, such as storage unit 208. In some embodiments, part or all of the computer program can be loaded and / or installed on device 400 via ROM 202 and / or communication unit 209. When the computer program is loaded into RAM 203 and executed by CPU 201, the operations described above can be performed. Figure 1 The various operations within. Alternatively, in other embodiments, the CPU 201 can be configured to perform these operations by any other suitable means (e.g., by means of firmware). Figure 1 The various operations within.

[0128] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload programmable logic devices (CPLDs), and so on.

[0129] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0130] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0131] Furthermore, although the operations are described in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0132] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A multi-source heterogeneous data monitoring and early warning method for deep foundation pits and high slopes in mountainous highways, characterized in that, Includes the following steps: Step 1: First, obtain electronic distribution maps of deep foundation pits and high slopes along mountain roads, and divide the deep foundation pits and high slopes along mountain roads into several monitoring areas. Collect real-time geological data of deep foundation pits and high slopes along mountain roads in the monitoring areas and establish the first dataset. Collect meteorological environmental data of the monitoring areas and establish the second dataset. Collect structural data of deep foundation pits and high slopes along mountain roads in the monitoring areas and establish the third dataset. Step 2: Analyze the first dataset, monitor the geological conditions of the deep foundation pit and high slope in the i-th monitoring area, and calculate the geological stability coefficient of the i-th monitoring area. A first standard threshold P is pre-set and the geological stability coefficient of the i-th monitoring area is set. Comparative analysis is conducted to assess the geological risk of the i-th monitoring area. If there is a risk of geological instability, a first risk signal is issued. Step 3: After receiving the first risk signal, combine the data from the second dataset to analyze the real-time rainfall (Prain), real-time groundwater level (Dxw), groundwater pressure (T), groundwater flow rate (Q), flow path length (L), and cross-sectional area of ​​the groundwater flow in the i-th monitoring area. Water head difference Further monitoring was conducted on the groundwater infiltration value Flz, and the groundwater inrush index for the i-th monitoring area was calculated. A second standard threshold Q is pre-set for comparative analysis to assess whether there are seepage channels forming in geological areas that could induce water inrush risk in monitoring areas with geological instability risks. If there are geological instability risks and seepage channels forming in geological areas that could induce water inrush risk, a second risk signal is issued. Step 4: After receiving the second risk signal, combine it with the geological stability coefficient of the i-th monitoring area. Groundwater inrush index in the i-th monitoring area Using data from the third dataset, calculate the deep foundation pit collapse risk coefficient for the i-th monitoring area. and the tilt deformation risk coefficient of the high slope in the i-th monitoring area And by comparing and analyzing the preset third standard threshold Z1 and fourth standard threshold Z2, a corresponding strategy is generated. Step 5: Using a convolutional neural network, construct an initial convolutional neural network model. Train and test the risk dynamic early warning model using the acquired feature information. Use the trained risk dynamic early warning model as data to run predictions and perform real-time risk predictions for deep foundation pits and high slopes on mountain roads.

2. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 1, characterized in that, Step one includes: S11. Using GIS geographic information system technology and related surveying and mapping data, obtain electronic distribution maps of deep foundation pits and high slopes along mountain roads; The area of ​​deep foundation pits and high slopes along mountain roads was divided into several monitoring zones, and these zones were marked as Jcq1, Jcq2, ..., Jcq on the electronic distribution map of deep foundation pits and high slopes along mountain roads. n n represents the number of monitoring areas; S12. Collect real-time geological data of deep foundation pits and high slopes along mountain roads in the monitoring area, and establish the first dataset, including: the compressive strength of the soil and rock layers in the monitoring area. The dataset includes: soil hardness (H), soil thickness (d), and soil moisture content (w) in the soil layer; meteorological environmental data of the monitoring area, and a second dataset including: real-time rainfall (Prain), real-time groundwater level (Dxw), groundwater pressure (T), and groundwater infiltration (FLz) in the monitoring area; and structural data of deep foundation pits and high slopes of mountain roads in the monitoring area, and a third dataset including: bearing capacity of deep foundation pit support structures in the monitoring area. Load values ​​of deep foundation pit structures Area A of the deep foundation pit; shear strength of the soil on the high slope of the monitoring area. Height of high slope Angle of high slope Load values ​​of high slopes and the base area of ​​high slope .

3. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 2, characterized in that, Step two includes: S21. Using the data from the first dataset, monitor the geological conditions of the deep foundation pit height and slope. After dimensionless processing, calculate the geological stability coefficient of the i-th monitoring area. The formula is as follows: ； In the formula, This represents the number of soil and rock layers in the i-th monitoring area. This represents the compressive strength of the j-th soil layer in the i-th monitoring area. Indicates the maximum compressive strength of the soil and rock layer. This represents the soil hardness of the j-th soil layer in the i-th monitoring area. This represents the thickness of the j-th soil layer in the i-th monitoring area. This represents the soil moisture content of the j-th soil layer in the i-th monitoring area. This represents the influence coefficient of soil moisture content. This represents the soil pressure influence coefficient. and Obtained through experimental data.

4. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 3, characterized in that, Step two also includes: S22. Pre-set the first standard threshold P, and set the geological stability coefficient of the i-th monitoring area. A comparative analysis is performed with the first standard threshold P to obtain the first evaluation result, including: When the geological stability coefficient of the i-th monitoring area When the value is ≥ the first standard threshold P, it indicates that there is no risk of geological instability in the current monitoring area; When the geological stability coefficient of the i-th monitoring area When the value is less than the first standard threshold P, it indicates that there is a risk of geological instability in the current monitoring area, triggering the first early warning instruction and generating the first risk signal.

5. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 4, characterized in that, Step three includes: S31. Upon receiving the first risk signal, and combining the data from the second dataset on the precipitation, groundwater level, temperature, and wind speed of the i-th monitoring area, further monitoring of meteorological and environmental interference is conducted for areas with geological instability risk. After dimensionless processing, the groundwater inrush index for the i-th monitoring area is calculated. The formula is as follows: ； ； In the formula, This represents the groundwater infiltration value of the i-th monitoring area; Q is the groundwater flow rate, and L is the water flow path length, i.e., the distance from the inflow point to the outflow point. The cross-sectional area of ​​the groundwater flow is measured through a vertical permeability test. Due to head difference; n represents the number of monitoring areas, and i represents the i-th monitoring area. This represents the real-time rainfall in the i-th monitoring area. This indicates the maximum amount of rainfall the geology can withstand. This represents the real-time groundwater level height in the i-th monitoring area. This indicates the highest groundwater level that the geology can withstand. This represents the groundwater pressure value in the i-th monitoring area. This indicates the maximum groundwater pressure that the geology can withstand. This represents the groundwater permeability value of the i-th monitoring area. This indicates the maximum groundwater permeability that the geology can withstand. , 2. and This represents the weighting coefficient.

6. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 5, characterized in that, Step three also includes: S32. Pre-set the second standard threshold K and set the groundwater inrush index of the i-th monitoring area. A second evaluation result is obtained by comparing the result with a second standard threshold K, including: When the groundwater inrush index of the i-th monitoring area When the value is ≤ the second standard threshold K, it indicates that there is no risk of geological instability in the current monitoring area. A seepage channel has formed in the geological area, which could induce a water inrush risk. Continuous monitoring is required. When the groundwater inrush index of the i-th monitoring area When the threshold K is greater than the second standard threshold, it indicates that there is a risk of geological instability in the current monitoring area and that a seepage channel has formed in the geological area, which may induce a risk of water inrush, triggering a second early warning instruction and generating a second risk signal.

7. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 6, characterized in that, Step four includes: S41. After receiving the second risk signal, combine it with the geological stability coefficient of the i-th monitoring area. Groundwater inrush index in the i-th monitoring area After dimensionless processing of the data from the third dataset, the collapse risk coefficient of the deep foundation pit in the i-th monitoring area is calculated and obtained. , and the tilt deformation risk coefficient of the high slope in the i-th monitoring area The formula is as follows: ； ； In the formula, n represents the number of monitoring areas. This represents the bearing capacity value of the deep foundation pit support structure in the i-th monitoring area. This represents the geological stability coefficient of the i-th monitoring area. This represents the groundwater inrush index for the i-th monitoring area. This represents the load factor of the deep foundation pit in the i-th monitoring area. This represents the load value of the deep foundation pit structure in the i-th monitoring area. This represents the area of ​​the deep foundation pit in the i-th monitoring area; ； ； In the formula, This represents the shear strength of the high slope soil in the i-th monitoring area. This represents the height of the high slope in the i-th monitoring area. This represents the angle of the high slope in the i-th monitoring area. This represents the load coefficient of the high slope in the i-th monitoring area. This represents the load value of the i-th monitoring area and the high slope. This represents the base area of ​​the high slope in the i-th monitoring area.

8. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 7, characterized in that, Step four also includes: S42. Pre-set the third standard threshold Z1, and set the deep foundation pit collapse risk coefficient for the i-th monitoring area. A comparative evaluation is performed with the third standard threshold Z1 to obtain the third evaluation result, including: When the deep foundation pit collapse risk coefficient of the i-th monitoring area When the value is less than the third standard threshold Z1, it indicates that there is no risk of collapse in the deep foundation pit, and monitoring should continue. When the deep foundation pit collapse risk coefficient of the i-th monitoring area When the value is ≥ the third standard threshold Z1, it indicates that the deep foundation pit is at risk of collapse, triggering the third early warning instruction and generating the first strategy, which includes: increasing the compressive strength of the soil around the foundation pit by 10% and improving the soil layer stability by 20% through deep mixing; reinforcing the loose soil layer through grouting technology and increasing the soil bearing capacity by 30%; and building temporary shelters and setting up two drainage points to reduce the impact of precipitation on the foundation pit by 70%.

9. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 8, characterized in that, Step four also includes: S43. Pre-set the fourth standard threshold Z2, and set the tilt deformation risk coefficient of the high slope in the i-th monitoring area. A comparative evaluation is performed with the fourth standard threshold Z2 to obtain the fourth evaluation result, including: When the tilt deformation risk coefficient of the high slope in the i-th monitoring area <The fourth standard threshold Z2 indicates that there is no risk of tilting or deformation of the high slope, and continuous monitoring is required; When the tilt deformation risk coefficient of the high slope in the i-th monitoring area When the value is ≥ the fourth standard threshold Z2, it indicates that the high slope is at risk of tilting and deformation, triggering the fourth early warning instruction and generating the second strategy, which includes: reinforcing and increasing the support bearing capacity of the deep foundation pit retaining wall by 20%, increasing the anchor density by 25%, and increasing the strength of the reinforced concrete retaining wall by 10%; reducing water accumulation by 40% and reducing water seepage pressure on the soil by 25% by setting up two drainage points; and strengthening the bearing capacity of loose soil and increasing the tensile strength of the soil by 35% through shotcreting technology.

10. The method for monitoring and early warning of multi-source heterogeneous data on deep foundation pits and high slopes in mountainous highways according to claim 9, characterized in that, Step five includes: S51. Using a convolutional neural network, construct an initial model of the convolutional neural network, and use the deep foundation pit collapse risk coefficient of the i-th monitoring area as the starting point. and the tilt deformation risk coefficient of the high slope in the i-th monitoring area The initial convolutional neural network model was trained and tested, and the trained initial convolutional neural network model was used as a dynamic risk warning model. Simultaneously, the collapse risk coefficient of the deep foundation pit in the i-th monitoring area was used. and the tilt deformation risk coefficient of the high slope in the i-th monitoring area The intermediate layer output is used as a feature vector to identify feature information. The acquired feature information is used to train and test the dynamic risk warning model. The trained dynamic risk warning model is then used as data to run predictions and perform real-time risk predictions for deep foundation pits and high slopes on mountain roads.

Images