A method for realizing remote monitoring of saline soil road foundation stability

Abstract

The application discloses a kind of methods for realizing the stability remote monitoring of saline soil road foundation, and the application relates to the technical field of data processing, and the application is based on historical failure record, combined with maintenance complexity and traffic influence to construct influence radius model, and screen monitoring section;Again, according to the division of important section of salinity and deformation characteristics, the risk mutation boundary is identified by OPTICS algorithm and the transition section is generated, and the three-dimensional grid unit is divided and bound to unique identification;Then, the grid is integrated with multiple sensors, and the differentiated data transmission method is used according to the risk level;Finally, the salt expansion and dissolution risk index is calculated, and each three-dimensional grid unit is marked by classification and the corresponding warning measures are executed.The method improves the monitoring pertinence and accuracy, and guarantees the road safety.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method for remote monitoring of the stability of road foundations in saline soil. Background Technology

[0002] Saline soil roads are prone to problems such as salt swelling and subsidence due to salt migration and moisture changes, seriously threatening the stability of the road structure. Remote monitoring of these roads is crucial for ensuring traffic safety. However, existing methods for remote monitoring of saline soil road foundation stability have several shortcomings:

[0003] First, the selection of monitored road sections lacks specificity, relying heavily on experience to define the scope without dynamically adjusting based on the complexity of historical faults and their impact on traffic. This can easily lead to missed detections in high-risk areas or over-monitoring in low-risk areas.

[0004] Secondly, the risk classification and spatial division are crude, often using a uniform grid size to cover the entire road section without considering the differences in different degrees of salinization, and ignoring the risk abrupt change boundaries between road sections, resulting in vague risk assessment and difficulty in accurately locating potential hazards.

[0005] Third, risk assessment and early warning lack multi-parameter coupling analysis, relying heavily on single indicators for judgment, failing to integrate the synergistic effects of salinity, moisture, temperature, and deformation, and the early warning measures are not graded, making it difficult to achieve differentiated treatment and affecting the effectiveness of monitoring. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method for remote monitoring of the stability of road foundations in saline soil, solving the problems of insufficient accuracy and effectiveness in existing remote monitoring methods for the stability of road foundations in saline soil.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for remote monitoring of the stability of road foundations in saline soil, comprising:

[0008] Based on the historical foundation failure records of saline soil roads, the precise station numbers of the failure points were extracted. A dynamic influence radius model was constructed based on the single maintenance time Ti and the average daily traffic volume Vi, and the monitored road sections and non-monitored road sections were selected.

[0009] For the monitored road sections, they are divided into first-level, second-level, and third-level important sections based on the initial salt concentration and historical annual deformation. Different sizes of three-dimensional grid cells are selected for each section, and the OPTICS algorithm is used to identify abrupt boundary and generate transition sections. Each three-dimensional grid cell is bound with a unique identifier.

[0010] Each 3D grid cell integrates four types of sensors. For 3D grid cells in the first-level important segment, the second-level important segment, and the transition segment, wired transmission is selected, while for 3D grid cells in the third-level important segment, wireless transmission is selected.

[0011] Based on sensor data, the salt swelling risk index SI and the solution collapse risk index DI of the three-dimensional grid cells are calculated, and high, medium and no risk are marked, and corresponding early warning measures are implemented.

[0012] As a further aspect of the present invention, the specific operation for constructing the dynamic influence radius model is as follows:

[0013] For each fault point Pi, i∈[1,n], collect the single repair time Ti and the average daily traffic volume Vi, both of which are historical averages, where n is the number of fault points;

[0014] The radius of influence is obtained using the formula Ri = α × Ti × log(Vi + 1), where α is a correction factor.

[0015] As a further aspect of the present invention, the specific steps for screening monitored road sections and non-monitored road sections are as follows:

[0016] Centered on the fault point Pi, the affected road segments [Pi-Ri, Pi+Ri] are obtained by combining the influence radius Ri;

[0017] If the affected road sections of adjacent fault points overlap, they need to be merged.

[0018] For the identified affected road segments, the total number of historical fault points F within their range is counted. If F ≥ Fth, the segment is classified as a monitored road segment; otherwise, it is classified as a non-monitored road segment. Here, Fth is the threshold for the number of fault points.

[0019] As a further aspect of the present invention, the specific rules for classifying the importance level of monitored road sections are as follows:

[0020] Set initial salt concentration thresholds Miumin and Miumax, and historical annual deformation thresholds Niumin and Niumax;

[0021] Obtain the initial salt concentration miu and the historical annual deformation niu of the monitored road section;

[0022] If miu≥Miumax or niu≥Niumax, it is classified as a first-level important paragraph; if Miumin≤miu<Miumax and Niumin≤niu<Niumax, it is classified as a second-level important paragraph; if miu<Miumin and niu<Niumin, it is classified as a third-level important paragraph.

[0023] As a further aspect of the present invention, the specific steps for selecting three-dimensional mesh elements of different sizes for each segment are as follows:

[0024] The selection rules for horizontal and vertical directions are as follows:

[0025] First-level important section: 1m / grid horizontally, 2m / grid vertically;

[0026] Secondary important sections: 3m / grid laterally, 5m / grid longitudinally;

[0027] Level 3 important section: 5m / grid horizontally, 15m / grid vertically;

[0028] The selection rules for depth are as follows:

[0029] The depth needs to cover the 0-2m foundation, which is divided into three parts: 0-0.5m, 0.5-1.5m, and 1.5-2m, corresponding to depths of 0.5m / grid, 1m / grid, and 0.5m / grid, respectively.

[0030] As a further aspect of the present invention, the specific steps for identifying mutation boundaries and generating transition segments using the OPTICS algorithm are as follows:

[0031] Extract the boundary station numbers of each important section and convert them into plane rectangular coordinates. Take one coordinate point every 1m along the longitudinal direction of the road, and label each point with the corresponding important section level label to form a point set database.

[0032] Initialize the core distance ε and the minimum number of points MinPts of the OPTICS algorithm;

[0033] Calculate the reachable distance of each point. When the difference in reachable distance between two adjacent points is greater than the threshold dth, it is marked as a mutation boundary.

[0034] Centered on the mutation boundary, extend tr meters to both sides as a transition section.

[0035] As a further aspect of the present invention, after dividing the three-dimensional mesh units, it is necessary to bind a unique identifier to each mesh unit. The unique identifier adopts four-dimensional coding: precise station number + horizontal interval + vertical interval + depth layer, and the specific format is: station number KXXX+XXX_horizontal X-Xm_vertical X-Xm_depth X-Xm.

[0036] As a further aspect of the present invention, the four types of sensors specifically include salt sensors, moisture sensors, temperature sensors, and displacement sensors.

[0037] As a further aspect of the present invention, according to the formula The salt swelling risk index SI was calculated, where C0 is the critical salt content for salt swelling of the saline soil section, T0 is the critical crystallization temperature of the saline soil section, and W0 is the optimum salt swelling moisture content of the saline soil section.

[0038] According to the formula The solution risk index DI was calculated, where W sat C represents the saturated water content of the saline soil in this section. thr V represents the critical salinity for dissolution of this section of saline soil. ref λ represents the reference deformation rate of the saline soil in this section, and λ, β, and γ are weighting indices.

[0039] As a further aspect of the present invention, the specific rules for dividing the three-dimensional mesh elements into high, medium, and no-risk categories are as follows:

[0040] If SI≥SIth and DI≥DIth, then the 3D mesh element is marked as a high-risk mesh.

[0041] If SI≥SIth or DI≥DIth is satisfied, then the three-dimensional mesh element is marked as a medium-risk mesh.

[0042] If SI < SIth and DI < DIth, mark the 3D mesh element as a risk-free mesh.

[0043] Among them, SIth and DIth are the threshold values ​​for the salt swelling risk index and the solution collapse risk index, respectively.

[0044] This invention provides a method for remote monitoring of the stability of road foundations in saline soil, which has the following advantages compared with the prior art:

[0045] (1) This invention constructs a dynamic impact radius model and combines the maintenance complexity of historical faults with the traffic impact to dynamically screen monitoring road sections, avoiding the blindness of empirical delineation, realizing the precise tilt of monitoring resources towards high-risk areas, and improving the targeting of monitoring and the efficiency of resource utilization.

[0046] (2) This invention achieves refined risk classification and precise spatial positioning by using hierarchical grid division and OPTICS algorithm to identify risk mutation boundaries and transition sections, and calculates salt swelling and dissolution risk indices based on multi-parameter coupling. This solves the problem of fuzzy assessment caused by traditional coarse division and improves the accuracy of risk identification.

[0047] (3) The present invention adopts differentiated sensor transmission and early warning strategies for different risk levels, ensuring real-time response in high-risk areas and controlling costs in low-risk areas. Combined with the unique grid identifier, it achieves rapid positioning and handling, taking into account both the effectiveness and economy of monitoring, and improving the road foundation safety guarantee capability. Attached Figure Description

[0048] Figure 1 This is a flowchart of the steps of the present invention;

[0049] Figure 2 This is a road segmentation diagram for the present invention;

[0050] Figure 3 This is the grid risk map of the present invention. Detailed Implementation

[0051] 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.

[0052] like Figures 1 to 3 As shown, this invention provides a method for remote monitoring of the stability of road foundations in saline soil, comprising:

[0053] Based on historical foundation failure records of saline soil roads, a dynamic influence radius model was constructed to identify monitored and non-monitored road sections.

[0054] Screen records of foundation failures caused by saline soil from road maintenance files, such as salt swelling collapse and solution cracking, and extract the precise station number of each failure point as the core positioning point, which can be denoted as Pi, i∈[1,n], where n is the number of failure points;

[0055] Precise stationing starts at kilometer 0 (denoted as K0+000) of the road project and proceeds along the centerline of the road towards the end point. Every 1,000 meters is marked as one kilometer station. For example, K1+000 is 1 kilometer from the starting point. Between kilometers, 100-meter stations are set in units of 10 meters and meter stations in units of 1 meter, forming a continuous stationing system.

[0056] The specific steps for constructing a dynamic influence radius model are as follows:

[0057] For each fault point Pi, two parameters are collected: single repair time Ti and average daily traffic volume Vi, both of which are historical averages.

[0058] The longer the repair time is, the more complex the saline soil problem at the fault point is, such as deep salt swelling, continuous dissolution, which makes repair difficult and the surrounding soil is more likely to be affected by salinization, because the salinity characteristics of the fault point are easy to spread to the surrounding area.

[0059] Using Ti as a parameter allows the radius of influence to expand dynamically with the complexity of the fault. For example, the radius of a fault point that takes 10 hours to repair is larger than that that that takes 5 hours, thus avoiding the omission of potential risks in the surrounding area.

[0060] For road sections with high daily traffic volume (Vi), the impact of malfunctions on traffic is more significant, requiring an expanded monitoring range for early warning. For example, a road section with an average of 10,000 vehicles per day has a larger impact radius than a road section with 1,000 vehicles per day. This ensures more comprehensive monitoring coverage in high-traffic areas and reduces the disruption of traffic caused by malfunctions.

[0061] The radius of influence is defined by the formula Ri = α × Ti × log(Vi + 1), where α is a correction factor, which can be taken as 0.5.

[0062] The linear term α×Ti allows the radius to increase proportionally with the maintenance time, intuitively demonstrating the logic that the more difficult the problem, the larger the scope of its impact.

[0063] The greater the traffic volume, the more significant the potential impact of the fault on traffic. It is necessary to appropriately expand the monitoring range. However, the physical and chemical effects of saline soil are limited by soil properties, and the range will not expand infinitely with the traffic volume. By log(Vi+1) of several terms, the increase in traffic volume can be compressed to avoid abnormal expansion of the radius due to excessive Vi. The 1 in the logarithmic terms is to eliminate the problem of log(0) being meaningless when Vi=0.

[0064] For each fault point Pi, the affected road segment [Pi-Ri, Pi+Ri] is obtained based on the dynamic influence radius Ri. The upper and lower ends of the interval can be converted according to the road station number. For example, if Pi=K12+350 and Ri=20 meters, then the affected road segment is [K12+330, K12+370].

[0065] If the affected road segments of multiple fault points overlap, they need to be merged into continuous road segments. For example, if the affected road segments of P1 [K12+330, K12+370] and the affected road segments of P2 [K12+360, K12+400] overlap, they should be merged to obtain continuous road segments [K12+330, K12+400].

[0066] For each consecutive road segment after merging, the total number F of all historical fault points within its range is counted. If F≥Fth, it is classified as a monitored road segment; otherwise, it is classified as a non-monitored road segment. Here, Fth is the threshold for the number of fault points.

[0067] For the monitored road sections, the monitored road sections are divided into first-level important sections, second-level important sections, and third-level important sections according to the initial salt concentration miu and the historical annual deformation niu;

[0068] The method for obtaining the initial salt concentration miu and historical annual deformation niu of the monitored road section is as follows:

[0069] For the initial salt concentration miu, soil samples were taken from three points in the center and both ends of each segment, and the salt concentration was measured using an ion chromatograph. The average value was taken as the miu of that segment.

[0070] For the historical annual deformation niu, retrieve the total station monitoring data for that period of history and calculate the annual average deformation niu;

[0071] The initial salt concentration thresholds Miumin and Miumax, and the historical annual deformation thresholds Niumin and Niumax can be set with reference to the "Technical Specifications for Highway Design and Construction in Saline Soil Areas" and engineering practice.

[0072] The importance level of monitored road sections is determined according to the following rules:

[0073] If miu≥Miumax or niu≥Niumax, then it is classified as a first-level important segment;

[0074] miu≥Miumax indicates that the soil salinity has far exceeded the safety threshold, and is prone to severe dissolution or salt swelling due to changes in moisture, which directly leads to road surface cracking; niu≥Niumax indicates that the roadbed is deforming too quickly, which will rapidly damage the road structure and threaten driving safety; any exceeding the standard can prevent it from being ignored due to the extreme of a single indicator, and ensure that high-risk road sections are given priority for monitoring.

[0075] If Miumin≤miu<Miumax and Niumin≤niu<Niumax, then it is classified as a second-level important segment;

[0076] Miumin≤miu<Miumax indicates that the salinity is high but not extreme, while Niumin≤niu<Niumax indicates that the deformation is slow but continuous. The two will promote each other (salt accelerates the loosening of soil structure, and deformation intensifies salt migration), which may escalate to a high risk in the long term.

[0077] If miu < Miumin and niu < Niumin, then it is divided into three levels of importance.

[0078] miu < Miumin indicates extremely low salinity, niu < Niumin indicates extremely slow deformation, and the impact of salinization on roads is negligible (soil structure is stable, deformation is negligible), so high-frequency monitoring is not required.

[0079] Different sizes of 3D mesh units are selected for the first, second, and third levels of importance segments, and the specific division rules are as follows:

[0080] In a two-dimensional plane, i.e., choosing between horizontal and vertical directions:

[0081] First-level important section: 1m / grid horizontally, 2m / grid vertically;

[0082] Salinization damage in high-risk areas is characterized by its suddenness and large local differences. Dense grids can accurately locate the spatial distribution of risk points and avoid missing key changes due to sparse grids.

[0083] Secondary important sections: 3m / grid laterally, 5m / grid longitudinally;

[0084] The damage in medium-risk areas is gradual, and there is no need to capture the subtle local differences as in high-risk areas, but sufficient density is required to track the changing trends.

[0085] Level 3 important section: 5m / grid horizontally, 15m / grid vertically;

[0086] The salinization impact in low-risk areas is minimal and stable, and sparse meshes are sufficient to meet the requirements.

[0087] Regarding depth selection:

[0088] The depth needs to cover the 0-2m foundation, therefore, it is divided into three parts to cover 0-0.5m (surface layer), 0.5-1.5m (middle layer), and 1.5-2m (deep layer), corresponding to depths of 0.5m / cell, 1m / cell, and 0.5m / cell;

[0089] The damage caused by saline soil to roads is mainly concentrated in the 0-2m depth range: In the surface layer, which is in direct contact with the atmosphere and is most affected by precipitation and evaporation, salts rise or fall with water, making it the most active area for salinization and prone to causing surface soil salt swelling / collapse; In the middle layer, which is the main distribution layer of the subgrade fill material and bears most of the road load, salinization will cause soil particles to loosen and bearing capacity to decrease, leading to road settlement; In the deep layer, which is close to the original subgrade soil, salinization will affect the overall stability of the subgrade.

[0090] The risk of saline soil along the road is not evenly distributed. If grids are directly divided, the grids at the abrupt boundary will contain both high-risk and medium-risk areas, making the risk level of the grid unclear.

[0091] Therefore, the OPTICS algorithm is needed to locate the abrupt boundary between segments and obtain the transition segment mesh. The specific operation is as follows:

[0092] Extract the boundary station numbers of each important section and convert them into plane rectangular coordinates. Take one coordinate point every 1m along the longitudinal direction (length direction) of the road, and label each point with the corresponding important section level label to form a point set database. For example, the point corresponding to K120+000 is labeled as level one, and the point corresponding to K120+100 is labeled as level two.

[0093] Initialize the parameters of the OPTICS algorithm, including the core distance ε and the minimum number of points MinPts;

[0094] ε can be taken as 2m. The risk of saline soil in the longitudinal direction of the road has local continuity. The soil characteristics within a 2m range are usually similar. This setting allows the algorithm to determine points of the same level within 2m as closely related.

[0095] MinPts can be 3. Three consecutive points must be of the same level to be considered a stable cluster, so as to avoid the clustering results being affected by a single outlier.

[0096] Calculate the reachable distance for each point, which is the distance from that point to the nearest core point. A core point is defined as a point within an ε-range of at least MinPts points of the same level. For example, point A (K120+098) in a first-level important segment has 5 first-level points within 2m, satisfying MinPts=3, and is denoted as a core point; adjacent point B (K120+100) has only 1 second-level point within 2m, which does not satisfy MinPts=3, and its reachable distance is the distance to the nearest second-level core point.

[0097] When the difference in reachability between two adjacent points is greater than the threshold dth, it is marked as a mutation boundary;

[0098] Centered on the mutation boundary, extend tr meters to both sides (a total range of 2tr meters) as a transition segment. For example, if the boundary is at K120+100, then the transition segment is [K120+100-tr, K120+100+tr].

[0099] Take the geometric mean of the dimensions of two adjacent grid levels, and calculate them separately for the horizontal and vertical directions. For example, if two adjacent grids are a first-level important segment grid (1m horizontally, 2m vertically) and a second-level important segment grid (3m horizontally, 5m vertically), then the horizontal dimension of the transition grid is... Rounded to 2, the vertical dimension is Rounded down to 3;

[0100] After dividing the three-dimensional mesh into units, each mesh unit needs to be bound with a unique identifier. The unique identifier adopts a four-dimensional code of "precise station number + horizontal interval + vertical interval + depth layer", and the specific format is "KXXX+XXX_horizontal X-Xm_vertical X-Xm_depth X-Xm".

[0101] For example, K120+050_Horizontal 0-1m_Vertical 0-2m_Depth 0-0.5m represents a three-dimensional grid cell of a first-level important segment at K120+50m, with a horizontal dimension of 0-1m, a vertical dimension of 0-2m, and a depth of 0-0.5m below the ground surface.

[0102] Four types of sensors are integrated into each 3D mesh cell, with the sensors embedded at the center of the mesh cell to ensure that the real data of that mesh is directly collected.

[0103] The four types of sensors specifically include salt sensors, moisture sensors, temperature sensors, and displacement sensors;

[0104] Salt sensors, such as the FJA-10 soil salt sensor, reflect the degree of salinization by measuring the salt concentration;

[0105] Moisture sensors, such as the TDR-300, monitor soil moisture content because the salt swelling / collapse of saline soils is closely related to moisture.

[0106] Temperature sensors, such as the DS18B20, record soil temperature, which affects the rate of salt crystallization / dissolution.

[0107] Displacement sensors, such as vibrating wire seam gauges, are used to measure vertical / lateral deformation, directly capturing the physical response of salt swelling / collapse;

[0108] The damage to saline soil is the result of the coupling of multiple factors such as salt content, moisture, temperature, and deformation. A single sensor cannot reflect the whole picture. For example, measuring only salt content may be misjudged as high risk, but combining moisture and temperature can determine that the actual risk of salt swelling is low.

[0109] Before collecting data, all sensors must be synchronized with GPS to ensure that the salinity, moisture, temperature, and deformation data collected by the same three-dimensional grid unit correspond to the same point in time, thus avoiding misjudgment caused by asynchronous data.

[0110] Different transmission methods need to be selected for sensor data collected at the first, second, and third levels of critical sections and transition sections:

[0111] For the primary and secondary important sections and transition sections, wired transmission, such as buried cables, should be selected because they have strong anti-interference capabilities and can avoid the attenuation of wireless signals in complex terrain.

[0112] For critical segments at level three, choosing wireless transmission, such as LoRa wireless transmission, can reduce cabling costs.

[0113] For each three-dimensional grid cell, based on the four core parameters collected in real time, namely salt concentration C, water content W, temperature T, and vertical deformation rate V, the corresponding salt swelling risk index SI and solution collapse risk index DI are calculated.

[0114] According to the formula The salt swelling risk index SI was calculated, where C0 is the critical salt content for salt swelling of the saline soil section, T0 is the critical crystallization temperature of the saline soil section, and W0 is the optimum salt swelling moisture content of the saline soil section.

[0115] In the formula, This indicates that the closer the salt content is to the critical value of this region, the greater the crystallization potential; This indicates that when the temperature is below the crystallization critical value, the ratio is greater than 1, because low temperature will accelerate crystallization, thus amplifying the risk; This indicates that when the moisture content deviates from the optimal value, the salt swelling capacity decreases. If W is too high, it will fill the pores, and if W is too low, there will be insufficient crystallization space. By constructing a quadratic function, the nonlinear relationship between excessively high or low moisture content and the suppression of salt swelling can be characterized.

[0116] According to the formula The solution risk index DI was calculated, where W sat C represents the saturated water content of the saline soil in this section. thr V represents the critical salinity for dissolution of this section of saline soil. ref λ represents the reference deformation rate of the saline soil in this section, and λ, β, and γ are weighting indices determined by the characteristics of the saline soil in the region.

[0117] λ quantifies the sensitivity of different saline soils to water. For example, clay saline soils are more sensitive to water subsidence, so λ can be 1.5, while sandy soils are less sensitive, so λ can be 1.0.

[0118] β quantifies the sensitivity of different saline soils to salt. For example, β can be taken as 1.2 for sulfate saline soils and 1.0 for chloride saline soils.

[0119] γ quantifies the sensitivity of different saline soils to deformation. For example, newly filled roadbeds are more sensitive to deformation, so γ can be taken as 1.5, while γ of old roadbeds can be taken as 1.0.

[0120] After calculating the SI and DI of each 3D mesh element, a risk threshold needs to be set and risk meshes need to be marked. The specific partitioning rules are as follows:

[0121] If SI≥SIth and DI≥DIth are satisfied, then the three-dimensional mesh element is marked as a high-risk mesh, where SIth and DIth are the salt swelling risk index threshold and the solution collapse risk index threshold, respectively.

[0122] When the risks of salt swelling and subsidence exceed the standard at the same time, the road structure faces two-way damage, which can easily lead to serious accidents such as pavement cracking and roadbed instability. Moreover, the repair difficulty is far greater than that of a single risk. For example, grouting in the salt swelling area may aggravate subsidence, and replacement in the subsidence area may induce salt swelling.

[0123] If SI≥SIth or DI≥DIth is satisfied, then the three-dimensional mesh element is marked as a medium-risk mesh.

[0124] Although a single risk may not reach the level of emergency, it may develop into a compound risk over time. For example, if a grid with only salt swelling is exposed to heavy rainfall, water intrusion may cause subsidence. Marking it as a medium risk can ensure that such grids are included in routine monitoring and prevent the risk from escalating.

[0125] Otherwise, mark the 3D mesh cell as a risk-free mesh;

[0126] When SI < SIth and DI < DIth, it indicates that the risks of salt swelling and solution collapse are within a safe range, such as low salt content and stable deformation, and no additional monitoring is required.

[0127] The specific methods for issuing corresponding early warnings for different risk network units are as follows:

[0128] If a 3D mesh element is deemed high-risk, the system will implement the following early warning measures:

[0129] The system automatically reads the unique identifier of high-risk grids, resolves the precise station number, horizontal direction, vertical direction, and depth, and simultaneously pushes the information to the road maintenance emergency command group (including maintenance engineers and construction team leaders) via SMS and platform pop-ups. The pushed information specifically includes: the unique identifier of the three-dimensional grid unit, the risk index (SI, DI), the core risk parameters (salt concentration C, moisture content W, temperature T, vertical deformation rate V), and the suggested treatment plan.

[0130] Increase the data acquisition frequency of the three-dimensional grid unit sensor and continuously track the changes of SI and DI. If any index rises above the specified threshold miu within an hour, the traffic police department should be contacted simultaneously to issue a traffic control notice. Tr and miu are the time threshold and the index threshold, respectively, and both need to be set according to the actual saline soil road conditions.

[0131] If the 3D mesh element is classified as medium risk, the system will implement the following early warning measures:

[0132] The system generates a daily trend analysis report for medium-risk grids, linking them to historical data based on the grid's unique identifier. For example, if the SI (Signal Index) rises from 0.9 to 1.0 in the past 7 days, the report is pushed to the daily maintenance team, highlighting the grids with increased risk.

[0133] A single risk (such as only a high SI) may evolve into a compound risk due to environmental changes. Strengthening monitoring can capture the deterioration trend. For example, the initial DI of a medium-risk grid is 0.75 (exceeding the threshold) and SI is 0.6 (not exceeding the threshold). Due to continuous rainfall, after 3 days, the DI of the grid cell rises to 0.9 and SI rises to 0.85, both exceeding the threshold. At this time, the system upgrades the warning in time to avoid missed detection.

[0134] If the 3D mesh element is risk-free, the system will implement the following early warning measures:

[0135] Reduce the frequency of sensor data acquisition, and the system will automatically compare the historical baseline ranges of SI and DI associated with the 3D grid cell and summarize them in the monthly report without sending out separate warnings;

[0136] These three-dimensional grid cells have low risk and slow change. Setting up low-frequency monitoring can reduce data redundancy. At the same time, by correlating the historical data of the three-dimensional grid cells, slow-deteriorating hidden risks can be identified.

[0137] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0138] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for realizing remote monitoring of saline soil road foundation stability, characterized in that, Comprise: According to the historical foundation failure record of the saline soil road, the accurate pile number of the failure point is extracted, a dynamic influence radius model is constructed based on the single maintenance time Ti and the daily traffic volume Vi, and the monitoring road section and the non-monitoring road section are screened out, and the specific operation of constructing the dynamic influence radius model is as follows: for each failure point Pi, i∈[1,n], the single maintenance time Ti and the daily traffic volume Vi are collected, both of which are the historical average values, wherein n is the number of failure points; according to the formula the influence radius is obtained, wherein, is a correction factor; For monitoring section, according to initial salt concentration and historical annual deformation, it is divided into first, second and third important sections, different size three-dimensional grid units are selected for each section, and OPTICS algorithm is used to identify mutation boundary and generate transition section, and each three-dimensional grid unit is bound with unique identification; Each three-dimensional grid unit integrates four types of sensors, for three-dimensional grid units of first, second and third important sections, wired transmission is selected, and for three-dimensional grid units of third important section, wireless transmission is selected; According to the sensor data, a salt expansion risk index SI and a dissolution collapse risk index DI of a three-dimensional grid unit are calculated, high, medium, and no risks are marked out, and corresponding warning measures are performed, the SI is calculated according to the formula The DI is calculated according to the formula Wherein, is the salt expansion critical salinity of the segment of saline soil, is the crystallization critical temperature of the segment of saline soil, is the optimal salt expansion moisture content of the segment of saline soil, is the saturated moisture content of the segment of saline soil, is the dissolution collapse critical salinity of the segment of saline soil, is the reference deformation rate of the segment of saline soil, , , is a weight index.

2. The method for realizing remote monitoring of the stability of a saline soil road foundation according to claim 1, characterized in that, The specific steps of screening monitored road section and non-monitored road section are: Taking fault point Pi as the center, combining with influence radius Ri to get influence section [Pi-Ri, Pi+Ri]; If the influence sections of adjacent fault points overlap, merging is required; For the obtained several influence sections, the total number F of all historical fault points in the range is counted, if F >= Fth, it is classified as a monitored road section, otherwise it is a non-monitored road section, wherein Fth is the fault point number threshold.

3. The method for realizing remote monitoring of the stability of a saline soil road foundation according to claim 1, characterized in that, The specific rules for dividing the importance level of the monitored road section are: Set initial salt concentration threshold Miumin, Miumax and historical annual deformation threshold Niumin, Niumax; Get the initial salt concentration m of the monitoring section and the historical annual deformation n; If m >= Miumax or n >= Niumax, it is classified as a first important section; if Miumin <= m < Miumax and Niumin <= n < Niumax, it is classified as a second important section; if m < Miumin and n < Niumin, it is classified as a third important section.

4. The method for realizing remote monitoring of the stability of a saline soil road foundation according to claim 1, characterized in that, The specific steps of selecting different size three-dimensional grid units for each section are: The selection rules in the horizontal and vertical directions are: For first important section, 1m per grid in horizontal direction and 2m per grid in vertical direction; For second important section, 3m per grid in horizontal direction and 5m per grid in vertical direction; For third important section, 5m per grid in horizontal direction and 15m per grid in vertical direction; The selection rule in depth is: The depth needs to cover 0-2m foundation, which is divided into three parts, covering 0-0.5m, 0.5-1.5m and 1.5-2m, corresponding to depth 0.5m per grid, depth 1m per grid and depth 0.5m per grid.

5. The method for realizing remote monitoring of the stability of a saline soil road foundation according to claim 1, characterized in that, The specific steps of using OPTICS algorithm to identify mutation boundary and generate transition section are: Extract the boundary stake number of each important section, and convert the stake number into plane rectangular coordinates, take 1 coordinate point every 1m along the road longitudinal direction, and label the corresponding important section level label for each point to form a point set database; the core distance of the OPTICS algorithm is initialized to MinPts. Calculate the reachable distance of each point, and when the difference of the reachable distance of adjacent two points is greater than threshold dth, mark it as a mutation boundary; Take the mutation boundary as the center and expand tr meters to both sides as the transition section.

6. The method for realizing remote monitoring of saline soil road foundation stability according to claim 1, characterized in that, After dividing the three-dimensional grid units, a unique identification needs to be bound for each grid unit, and the unique identification uses four-dimensional coding: accurate stake number+horizontal interval+vertical interval+depth layer, the specific format is: stake number KXXX+XXX_horizontal X-Xm_vertical X-Xm_depth X-Xm.

7. The method for realizing remote monitoring of saline soil road foundation stability according to claim 1, characterized in that, The four types of sensors specifically include salt sensor, moisture sensor, temperature sensor and displacement sensor. 8.The method for realizing remote monitoring of salinized soil road foundation stability according to claim 1, characterized in that, The specific rules for dividing the three-dimensional grid units into high, medium and no risk are as follows: If SI≥SIth and DI≥DIth, the three-dimensional grid unit is marked as a high-risk grid; If SI≥SIth or DI≥DIth, the three-dimensional grid unit is marked as a medium-risk grid; If SI<SIth and DI<DIth, the three-dimensional grid unit is marked as a no-risk grid; Wherein, SIth and DIth are salt expansion risk index threshold and dissolution collapse risk index threshold respectively.

Images