A loess collapsible area water conveyance engineering disaster risk real-time regulation method and system

Abstract

The present application relates to the technical field of loess collapsible area disaster prevention and control, especially relates to a loess collapsible area water conveyance engineering disaster risk real-time regulation method and system, first, environmental monitoring data is acquired, then the random forest algorithm is used to predict the rainfall infiltration mechanical change trend, the prediction result is used to determine the potential mechanical change area, then the foundation settlement prediction model is established, the support vector machine algorithm is used to identify the high-risk area from the settlement amplitude distribution map, then the historical engineering deformation data is matched with the current settlement amplitude distribution map, then the correlation analysis is carried out, then the priority sequence of disaster intervention is obtained based on the distribution of weak points affecting the path in the potential propagation direction and combined with historical disaster records, then the targeted numerical simulation scene is established, and the foundation reinforcement parameters are optimized through the combination of numerical simulation and historical data, the stability and reliability of the engineering are improved, the safety of the water conveyance engineering is improved, and the ability and efficiency of disaster response are improved.

Description

Technical Field

[0001] This invention relates to the field of disaster prevention and control technology in loess subsidence areas, specifically to a method and system for real-time control of disaster risks in water conveyance projects in loess subsidence areas. Background Technology

[0002] Effectively addressing disasters arising from the interaction between the natural environment and engineering activities is directly related to the safe operation of infrastructure and the sustainable development of the regional economy. This is especially true for projects like water conveyance projects, which require extremely high foundation stability; the importance of disaster prevention and control is self-evident. However, current disaster prevention measures for this region often suffer from systemic deficiencies, failing to fully adapt to complex environmental changes and engineering needs. Particularly when faced with the superposition of multiple disaster factors, existing methods often prove inadequate, lacking a holistic understanding of the disaster occurrence process and targeted responses.

[0003] Specifically, existing solutions, when faced with the unique environmental challenges of loess collapsing areas, neglect the interrelationships among multiple stages in the disaster formation process, leading to a disconnect between prevention and control measures and actual problems. This is especially true in water conveyance projects, where phenomena such as rainfall, seepage, and foundation deformation are not isolated but interconnected and gradually evolving. Without a holistic understanding of the relationships between these stages, it is difficult to take effective intervention measures before a disaster occurs. Consequently, many projects are unable to adjust their strategies in a timely manner when faced with emergencies, thus exacerbating disaster risks.

[0004] The core technical challenge in prevention and control lies in accurately grasping the impact of rainfall infiltration on the stability of loess foundations and how this impact gradually leads to deformation risks in engineering structures. Rainfall infiltration alters the mechanical properties of loess, reducing its bearing capacity. This change, in turn, further triggers uneven settlement of the foundation, ultimately threatening the overall safety of the water conveyance project. For example, in a water pipeline project, localized subsidence of the foundation after rainfall caused misalignment at pipe connections, leading to leakage and further structural damage. This chain reaction from environmental change to engineering failure is precisely the problem that current technology urgently needs to overcome.

[0005] Therefore, how to systematically identify and address the disaster risks in water conveyance projects in loess collapsing areas, from rainfall to foundation deformation, has become a key issue that this study urgently needs to solve. Summary of the Invention

[0006] To address the multiple logical problems in the prevention of water conveyance projects in loess subsidence areas, such as insufficient real-time environmental monitoring data, low prediction accuracy, unclear risk propagation paths, and difficulty in determining intervention priorities, this invention provides a method and system for real-time control of disaster risks in water conveyance projects in loess subsidence areas.

[0007] This invention is achieved through the following technical solution: A method for real-time control of disaster risks in water conveyance projects in loess collapsing areas includes: Rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone were collected and integrated to generate a comprehensive environmental monitoring and analysis dataset. Based on the comprehensive analysis dataset of environmental monitoring, the random forest algorithm is used to predict the trend of rainfall infiltration mechanics changes. The prediction results are then fused with the time series analysis of historical rainfall infiltration data to obtain high-priority monitoring areas. Based on the soil moisture content of high-priority monitoring areas, a foundation settlement prediction model was established using the finite element simulation method combined with the stress distribution changes in the loess layer. The settlement amplitude distribution map is extracted from the foundation settlement prediction model. The support vector machine algorithm is used to classify the high-risk areas in the settlement amplitude distribution map. Based on the stress distribution state of the pipeline, the deformation possibility of all pipelines in the high-risk area is judged, and a list of high-risk pipeline sections is drawn. For the list of high-risk pipeline sections, historical engineering deformation data of pipeline sections with high deformation potential are matched with the current settlement amplitude distribution map to obtain quantitative indicators of deformation risk; Based on the quantitative indicators of deformation risk, a correlation analysis of deformation risk and leakage risk is conducted to obtain the potential propagation direction of leakage risk. Based on the distribution of vulnerable points along the impact path in the potential propagation direction, a priority sequence for disaster intervention is obtained. The highest priority section is selected from the priority sequence of disaster intervention. The high-risk section simulation scenario is obtained through evaluation, weight assignment and numerical simulation. The parameters of the simulation scenario are adjusted based on the stress distribution to obtain the optimized foundation reinforcement parameter configuration.

[0008] Preferably, rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone are collected and integrated to generate a comprehensive environmental monitoring and analysis dataset, including: Real-time data on rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone are collected and stored as raw monitoring datasets. Based on the original monitoring dataset, preprocessing methods were used to denoise and standardize the rainfall intensity, surface humidity and soil moisture content data to obtain a standardized dataset. Data fusion methods are used to integrate standardized datasets and generate a unified fused dataset. If the rainfall intensity data of a certain area in the fused dataset exceeds a preset threshold, then the surface humidity and soil moisture content data of that area are subjected to weighted analysis to obtain the distribution of high-risk areas; Extract the soil moisture content change trend corresponding to the distribution of high-risk areas to determine whether there is a risk of subsidence. Based on the assessment results of the risk of subsidence and the rainfall intensity data of the area, the safety assessment results of the water conveyance project in the area were obtained; Based on the safety assessment results, comprehensive environmental monitoring analysis data for water conveyance projects in loess subsidence areas is generated.

[0009] Preferably, based on the comprehensive environmental monitoring analysis dataset, the random forest algorithm is used to analyze the changing trends of rainfall infiltration mechanics in high-risk areas, including: The random forest algorithm was used to predict the trend of rainfall infiltration mechanics, and the predicted trend of rainfall infiltration was obtained. Extract the data portion where the penetration intensity exceeds a preset threshold from the predicted trend results to determine the distribution of potential areas of mechanical change; For the distribution of potential mechanical changes, historical rainfall infiltration records for the corresponding areas are obtained from the comprehensive analysis dataset of environmental monitoring, and the infiltration intensity change sequence for the corresponding areas is obtained. By performing time-series analysis on the infiltration intensity variation sequence, it can be determined whether there is a continuous mechanical change tendency in the region; If the preliminary assessment results indicate that the mechanical changes are likely to continue, the environmental monitoring data for that area will be weighted to determine high-priority monitoring areas.

[0010] Preferably, using the soil moisture content of high-priority monitoring areas as a benchmark, a foundation settlement prediction model is established using the finite element simulation method combined with changes in loess layer stress distribution, including: Collect soil moisture data in high-priority monitoring areas, obtain real-time changes in soil moisture increase, and determine whether the increase exceeds the preset threshold. If the increase in soil moisture content exceeds the preset threshold, the stress distribution is calculated using the finite element simulation method based on the loess layer data in the area to obtain the change state of the stress distribution in the loess layer. Based on the changing state of stress distribution in the loess layer, sub-regions with stress concentration or abnormal distribution are extracted to determine whether there is a potential trend of foundation settlement. If a potential trend of foundation settlement is identified, historical settlement data for the corresponding sub-area is obtained from environmental monitoring records to determine the persistence characteristics of the settlement trend. Based on the persistent nature of the settlement trend and combined with current stress distribution change data, a foundation settlement prediction model is constructed. Based on the output of the prediction model, the distribution of settlement risk in each sub-region within the potential mechanical change area is analyzed, and the priority ranking of high-risk sub-regions is determined. Based on the priority ranking of high-risk sub-regions, real-time environmental monitoring data is obtained to determine whether there are abnormal fluctuations in stress distribution in the short term, thus obtaining the final risk distribution status.

[0011] Preferably, a settlement amplitude distribution map is extracted from the foundation settlement prediction model, and a support vector machine algorithm is used to classify high-risk areas in the settlement amplitude distribution map. Based on the stress distribution state of the pipelines, the deformation probability of all pipelines in the high-risk areas is obtained, including: The settlement amplitude distribution map is obtained by using the foundation settlement prediction model. The data points in the distribution map are initially screened to obtain the preliminary range of the high-risk area. Based on the preliminary scope of the high-risk area, the support vector machine algorithm is used to classify the data in the distribution graph to determine the specific boundaries of the high-risk area; For the specific boundaries of high-risk areas, obtain the corresponding underground pipeline layout information from the environmental monitoring database to determine whether the pipeline is located within the high-risk area; If the pipeline is located in a high-risk area, its geometric parameters can be extracted through pipeline layout information, and combined with settlement amplitude data, the stress distribution of the pipeline can be calculated to obtain the potential trend of pipeline deformation.

[0012] Preferably, historical engineering deformation data of pipeline sections with high deformation potential are matched with the current settlement amplitude distribution map to obtain quantitative indicators of deformation risk, including: Deformation records related to pipeline deformation are obtained from historical engineering archives. Pipeline sections with high deformation risk are initially screened to determine the key areas of concern. For the selected key areas, the corresponding settlement amplitude data is obtained and visualized through distribution graphics to obtain the settlement amplitude distribution status within the area; Based on the settlement amplitude distribution, a comparative analysis method is used to match historical deformation records with the current state to determine the similarity distribution between the two. If the similarity distribution exceeds the preset threshold, the segment is marked as a high-risk area, and the specific location distribution of the pipeline is obtained based on the pipeline layout information of the area. By analyzing the specific location distribution of the pipeline and the distribution of settlement amplitude, the distribution of stress points in the high-risk area of ​​the pipeline is analyzed, and the concentrated area of ​​stress points is obtained. For areas with concentrated stress points, obtain corresponding geological stability data from the environmental monitoring database; By comprehensively analyzing geological stability data and areas of concentrated stress points, a priority ranking of pipeline deformation risks is constructed.

[0013] Preferably, a correlation analysis of deformation risk and leakage risk is conducted based on quantitative indicators of deformation risk to obtain the potential propagation direction of leakage risk. Based on the distribution of vulnerable points along the impact path in the potential propagation direction, a priority sequence for disaster intervention is obtained, including: Correlation analysis was performed on deformation risk and leakage risk to obtain historical monitoring data related to deformation risk. Combined with current environmental variables, a preliminary path model for risk propagation was constructed to obtain the potential propagation direction of leakage risk. Based on the potential propagation direction of leakage risk, network topology analysis is used to analyze the connection relationship between nodes, identify key nodes and paths for risk propagation, and determine the key areas that will trigger a chain reaction. For key areas with cascading effects, corresponding stability parameters are obtained from the geological environment database to analyze the external constraints on risk propagation within the region and determine the distribution of vulnerable points along the impact path. If the density of vulnerable points affecting the path exceeds a preset threshold, the vulnerable points are prioritized and combined with historical disaster records to obtain a preliminary sequence of disaster interventions. Based on the preliminary sequence of disaster intervention, obtain pipeline layout data for relevant areas, analyze the spatial overlap between pipelines and vulnerable points, and determine the pipeline sections that require key protection. For key pipeline sections requiring protection, corresponding flow change data are obtained from the real-time monitoring system, and the final disaster intervention sequence is determined based on the flow change.

[0014] Preferably, the highest priority section is selected from the disaster intervention priority sequence. A high-risk section simulation scenario is obtained through assessment, weight assignment, and numerical simulation. The simulation scenario parameters are adjusted based on the stress distribution and historical settlement records to obtain an optimized foundation reinforcement parameter configuration, including: Data on the highest priority sections are obtained from the disaster intervention priority sequence. Combined with settlement data of foundation settlement, the geological stability of the sections is preliminarily assessed through information processing to obtain the distribution range of high-risk sections. For the distribution range of high-risk areas, data on the variation of soil moisture content is obtained. If the variation exceeds the preset threshold, the variation is weighted through information processing to determine the allocation ratio of weight factors. Based on the weighting ratio and settlement data, a numerical simulation method is used to construct a simulation scenario for high-risk sections, and the stress distribution of each sub-region within the scenario is obtained. Based on the stress distribution, historical settlement records for the corresponding sections are obtained from a pre-established geological database. If the historical settlement records show an abnormal trend, the abnormal trend is marked through information processing to identify the sub-areas that need to be focused on. Based on the sub-regions that require special attention, obtain the initial configuration scheme of their reinforcement parameters. By adjusting the parameters in the simulated scenario of the initial configuration scheme, the optimized combination of reinforcement parameters is obtained.

[0015] Preferably, this also includes multi-dimensional identification of key areas of focus: Based on the distribution of high-priority monitoring areas, key areas of concern are identified by checking whether there is a surge in infiltration intensity in the real-time rainfall infiltration data of the relevant areas. Based on the prediction results of the foundation settlement prediction model, the key areas of concern are identified by the abnormal fluctuations in stress distribution; In high-risk sections of pipeline deformation, key areas of concern are identified by whether the pipeline stress exceeds a preset threshold.

[0016] A system for real-time control of disaster risks in water conveyance projects in loess collapsing areas includes: The data acquisition and fusion module is used to collect and fuse rainfall intensity, surface humidity and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone to generate a comprehensive environmental monitoring and analysis dataset. The monitoring area analysis module is used to predict the trend of rainfall infiltration mechanics changes based on the comprehensive environmental monitoring analysis dataset and the random forest algorithm. The prediction results are then fused with the time series analysis of historical rainfall infiltration data to obtain high-priority monitoring areas. The prediction model building module is used to establish a foundation settlement prediction model based on the soil moisture content of high-priority monitoring areas and by combining the finite element simulation method with the stress distribution changes of loess layers. The pipeline deformation analysis module is used to extract the settlement amplitude distribution map from the foundation settlement prediction model, use the support vector machine algorithm to classify the high-risk areas in the settlement amplitude distribution map, determine the deformation probability of all pipelines in the high-risk areas based on the pipeline stress distribution state, and draw a list of high-risk pipeline sections. The risk matching and sorting module matches historical engineering deformation data of pipeline sections with high deformation probability with the current settlement amplitude distribution map for the list of high-risk pipeline sections to obtain quantitative indicators of deformation risk. The disaster intervention analysis module is used to conduct correlation analysis between deformation risk and leakage risk based on the deformation risk quantification index, obtain the potential propagation direction of leakage risk, and obtain the priority sequence of disaster intervention based on the distribution of vulnerable points along the impact path in the potential propagation direction. The reinforcement parameter optimization module is used to select the highest priority section from the priority sequence of disaster intervention, obtain the simulation scenario of high-risk section through evaluation, weight assignment and numerical simulation, and adjust the parameters of the simulation scenario based on the stress distribution to obtain the optimized foundation reinforcement parameter configuration.

[0017] An electronic device includes a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a real-time disaster risk control method for water conveyance projects in loess subsidence areas. Addressing a typical business scenario where rainfall-induced soil subsidence leads to a series of chain reactions such as foundation settlement, pipeline deformation, and leakage in water conveyance projects in loess subsidence areas, this method utilizes multi-source data fusion, advanced algorithm application, and scientific risk correlation analysis to achieve accurate identification, effective prediction, reasonable intervention, and parameter optimization of disaster risks in water conveyance projects in loess subsidence areas. Furthermore, by adopting differentiated control measures for regions with different risk levels and characteristics, it can maximize the efficiency and effectiveness of disaster risk prevention and control, improve the accuracy of disaster prediction and intervention efficiency in water conveyance projects in loess subsidence areas, reduce project losses, and thus achieve intelligent risk management. Specifically, the process begins by acquiring environmental monitoring data. Then, a random forest algorithm is used to predict the trend of changes in rainfall infiltration mechanics. The prediction results are then used to identify potential areas of mechanical change. Next, a foundation settlement prediction model is established, and a support vector machine algorithm is used to identify high-risk areas from the settlement amplitude distribution map. Historical engineering deformation data is then matched with the current settlement amplitude distribution map. Following this, correlation analysis is conducted, and based on the distribution of vulnerable points along the potential propagation path, combined with historical disaster records, a priority sequence for disaster intervention is obtained. A targeted numerical simulation scenario is then established, and foundation reinforcement parameters are optimized by combining numerical simulation with historical data. This improves the stability and reliability of the project, thereby enhancing the safety of the water conveyance project and ultimately improving the ability and efficiency to respond to disasters.

[0019] Furthermore, by collecting and integrating multi-source data such as rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone, the environmental conditions of the region can be comprehensively and holistically reflected, which helps to improve the ability to identify potential disaster risks.

[0020] Furthermore, by using the random forest algorithm to predict the trend of rainfall infiltration mechanics changes in high-risk areas and combining it with time series analysis of historical rainfall data to obtain high-priority monitoring areas, monitoring resources can be concentrated on key areas, thereby improving monitoring efficiency and the timeliness of risk warnings.

[0021] Furthermore, using the soil moisture content of high-priority monitoring areas as a benchmark, a foundation settlement prediction model is established using the finite element method combined with the stress distribution changes in the loess layer. The settlement amplitude distribution map is extracted from the foundation settlement prediction model, and the high-risk areas are classified using the support vector machine algorithm. Based on the stress distribution state of the pipeline, the deformation probability of the pipeline in the high-risk area is judged, and a list of high-risk pipeline sections is drawn. Thus, high-risk areas can be quickly and accurately identified from the settlement amplitude distribution map.

[0022] Furthermore, historical engineering deformation data of pipeline sections with high deformation potential are matched with the current settlement amplitude distribution map to obtain quantitative indicators of deformation risk. By combining historical data with current monitoring data, the degree of pipeline deformation risk can be scientifically quantified.

[0023] Furthermore, by associating deformation risk and leakage risk, the potential propagation direction of leakage risk can be obtained, and the priority sequence of disaster intervention can be obtained based on the distribution of vulnerable points along the impact path in the potential propagation direction, thereby improving the efficiency and effectiveness of disaster prevention and control.

[0024] Furthermore, based on the priority sequence of disaster intervention and the settlement data mapping results of the foundation settlement prediction model, the highest priority section is selected. The soil moisture content increase judgment of the highest priority section is integrated as a weighting factor to generate a targeted numerical simulation scenario. Based on the stress distribution of the simulation scenario and the historical settlement record of the section, the optimized foundation reinforcement parameter configuration is obtained. Through simulation and optimization, the most suitable foundation reinforcement scheme is formulated for the most critical high-risk section, which improves the stability of the foundation, helps to reduce the disaster risk caused by loess subsidence in the water conveyance project, and ensures the safe operation of the project. Attached Figure Description

[0025] Figure 1 This is a flowchart of a method for real-time control of disaster risks in water conveyance projects in loess collapsing areas according to the present invention.

[0026] Figure 2 This is a schematic diagram of a real-time disaster risk control method for water conveyance projects in loess collapsing areas according to the present invention.

[0027] Figure 3 This is another schematic diagram of a real-time disaster risk control method for water conveyance projects in loess collapsing areas according to the present invention.

[0028] Figure 4 This is a schematic diagram of the disaster risk control system for water conveyance projects in loess collapsing areas according to the present invention. Detailed Implementation

[0029] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification.

[0030] Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. These embodiments are provided to fully and completely disclose the invention and to fully convey its scope to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the drawings is not intended to limit the invention. In the drawings, the same units / elements are referred to by the same reference numerals.

[0031] Unless otherwise stated, the terms used herein (including technical terms) have their common meaning as understood by one of ordinary skill in the art. Furthermore, it is understood that terms defined in commonly used dictionaries should be understood to have a meaning consistent with the context of their relevant field, and not to be interpreted as having an idealized or overly formal meaning.

[0032] The present invention will be further described in detail below with reference to specific embodiments. These descriptions are for explanation purposes only and are not intended to limit the scope of the invention.

[0033] This invention discloses a method for real-time control of disaster risks in water conveyance projects in loess subsidence areas, referring to... Figures 1-3 ,include: S100 collects and integrates rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone to generate a comprehensive environmental monitoring and analysis dataset. Specifically: S101, real-time data collection of rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone, storing this data as a raw monitoring dataset. For example, in the environmental monitoring of the water conveyance project in the Loess Collapse Zone, real-time data collection of environmental data is carried out using satellite remote sensing equipment and ground sensor networks. Satellite remote sensing is used to acquire rainfall intensity data; for example, a rainfall event is recorded when the hourly rainfall reaches 10 mm or more. Ground sensors accurately measure surface humidity and soil moisture content; for example, a soil moisture content increase from 20% to 35% at a certain point.

[0034] S102, based on the original monitoring dataset, preprocessing methods are used to denoise and standardize the rainfall intensity, surface moisture, and soil moisture content data to obtain a standardized dataset. For example, preprocessing the original data improves data quality and reduces the impact of errors on risk assessment. Specifically, rainfall intensity data may contain outliers due to signal interference; for instance, a record might show a sudden increase in rainfall to 100 mm, significantly deviating from the normal range. This can be corrected by using mean filtering to remove noise and adjust the data to a reasonable value. Surface moisture and soil moisture content data are standardized to ensure consistency in dimensions across different data sources, facilitating subsequent fusion analysis.

[0035] S103 utilizes data fusion methods to integrate standardized datasets and generate a unified fused dataset. This fused dataset comprehensively reflects the regional environmental status, providing a reliable basis for risk identification. For example, rainfall intensity, surface moisture, and soil moisture content data are fused using a weighted average method to generate a unified fused dataset. For instance, the weights for rainfall intensity, soil moisture content, and surface moisture are set to 0.4, 0.3, and 0.3 respectively, ensuring a balanced contribution of each indicator to the final result.

[0036] S104. If the rainfall intensity data for a certain area in the fused dataset exceeds a preset threshold, then a weighted analysis is performed on the surface moisture and soil moisture content data for that area to obtain the distribution of high-risk areas. Weighted processing highlights the impact of key factors, identifies high-priority monitoring areas, and provides a basis for resource allocation. For example, if the rainfall intensity in the fused data exceeds a preset threshold (e.g., 15 mm per hour), then a weighted analysis is performed on the surface moisture and soil moisture content data to accurately locate potential threat areas and improve monitoring efficiency. Assuming that a certain area has a surface moisture content of 80% and a soil moisture content of 40%, with a comprehensive score higher than the risk threshold, it can be identified as a high-risk area. Overall score = w1 × surface moisture + w2 × soil moisture content Here, w1 and w2 are preset weights, and usually satisfy w1+w2=1.

[0037] S105 extracts the soil moisture content change trend corresponding to the distribution of high-risk areas to determine whether there is a risk of subsidence, thereby achieving early warning and reducing engineering safety hazards. For example, if the soil moisture content in a certain area increases from 30% to 45% within three days, combined with historical data analysis, the risk of subsidence can be predicted.

[0038] S106. Combining the assessment results of the subsidence risk and the rainfall intensity data of the area, the safety assessment results of the water conveyance project in the area are obtained, providing a scientific basis for project management. If the rainfall intensity remains high and the subsidence risk is significant, the assessment results indicate that the project is under considerable threat and protective measures need to be strengthened.

[0039] S107 generates comprehensive environmental monitoring analysis data for water conveyance projects in loess subsidence areas based on safety assessment results, identifies key monitoring areas for subsequent monitoring, and thus optimizes resource allocation and improves the targeting and effectiveness of monitoring. For example, if the soil moisture content around a section of the water conveyance channel changes drastically and rainfall is frequent, it can be listed as a key monitoring target.

[0040] S200, based on the comprehensive environmental monitoring analysis dataset, uses the random forest algorithm to predict the trend of rainfall infiltration mechanics changes. The prediction results are then fused with time-series analysis of historical rainfall infiltration data to obtain high-priority monitoring areas. Specifically: S201 uses the random forest algorithm to predict the trend of rainfall infiltration mechanics changes, obtaining the predicted trend of rainfall infiltration. In the prediction process, a prediction model is first constructed using historical rainfall data and soil property data as input. The stability of the prediction is improved by ensembling multiple decision trees of the random forest algorithm. For example, in a certain prediction, the prediction model infers the trend of rainfall infiltration intensity (increasing or decreasing) in the next 7 days based on the rainfall data and soil moisture content changes of the past 30 days.

[0041] S202, extract the data portion where the infiltration intensity exceeds a preset threshold from the predicted trend results, and determine the distribution of potential mechanical change areas. For example, for the data portion in the predicted trend results where the infiltration intensity exceeds a preset threshold, a threshold (such as an infiltration rate exceeding 5 mm per hour) is set as an outlier. Suppose a certain area shows an infiltration intensity of 6 mm per hour in the prediction, which is significantly higher than the normal range, then it is marked as a potential mechanical change area.

[0042] S203. For a given distribution of mechanical changes, historical rainfall infiltration records for the corresponding area are obtained from environmental monitoring data to obtain a sequence of infiltration intensity changes within the area. For example, after determining the distribution of mechanical changes, when extracting historical rainfall infiltration records for the corresponding area from environmental monitoring data, a sequence of infiltration intensity changes over the past 90 days can be obtained. Suppose that the historical records for a certain area show that the infiltration intensity gradually increases from 2 mm per hour to 5 mm per hour, exhibiting a continuous upward trend. Time series analysis can be used to determine whether there is a persistent tendency for mechanical changes in the area, capturing potential risk signals from long-term data and improving the accuracy of the judgment.

[0043] S204. By performing time-series analysis on the infiltration intensity change sequence, it is determined whether there is a continuous mechanical change tendency in the region.

[0044] S205. If the preliminary assessment results show that the mechanical change trend continues, the environmental monitoring data of the area shall be weighted to determine the high-priority monitoring area. For example, if the preliminary assessment results show that the mechanical change trend continues, when weighting the environmental monitoring data of the area, the weights of rainfall intensity and soil moisture content data can be increased, for example, set to 0.5 and 0.3 respectively, while the weights of other indicators are appropriately reduced to balance the influence of various factors.

[0045] S206. Based on the distribution of high-priority monitoring areas, obtain real-time rainfall infiltration data for relevant areas, determine whether there is a surge in infiltration intensity in a short period of time, and identify key areas of concern.

[0046] For example, based on the distribution of high-priority monitoring areas, real-time data analysis can promptly detect anomalies. When acquiring real-time rainfall infiltration data, ground sensors can update the data hourly to determine if there is a surge in infiltration intensity within a short period. If the infiltration intensity around a section of a water conveyance channel consistently exceeds a threshold and experiences a short-term surge, it is designated as a high-risk area of ​​focus. For instance, if the infiltration intensity in a certain area suddenly increases from 3 mm / hour to 7 mm / hour within 24 hours, combined with historical trend analysis, it can be inferred that this area carries a high risk in the short term and is therefore a key area of ​​focus.

[0047] Furthermore, based on basic predictions and analysis, more environmental factors (such as topographic slope data) can be introduced to further refine the distribution of areas of mechanical change, thereby enriching the dimensions of risk assessment, improving the comprehensiveness of monitoring, and providing more solid data support for the safety management of water conveyance projects in loess subsidence areas.

[0048] S300, using the soil moisture content of high-priority monitoring areas as a benchmark, establishes a foundation settlement prediction model using the finite element method combined with changes in loess layer stress distribution. Specifically: S301: Collect soil moisture data in high-priority monitoring areas to obtain real-time changes in soil moisture content and determine whether the increase exceeds a preset threshold. For example, in the environmental monitoring scenario of a water conveyance project in the Loess Collapse Zone, soil moisture data in areas with potential mechanical changes can be collected by humidity sensors deployed at key locations, recording changes in soil moisture content hourly. Assuming the preset threshold is an increase in moisture content exceeding 2.5% per hour, if an increase of 3.2% is detected in a certain area, it indicates a possible abnormal infiltration situation, requiring further analysis.

[0049] S302. If the increase in soil moisture content exceeds a preset threshold, the finite element method (FEM) is used to calculate the stress distribution in the loess layer within that area, thus obtaining the stress distribution variation. For example, when calculating the loess layer stress distribution in an area exceeding the threshold using FEM, the soil layer is divided into multiple grid cells, and parameters such as soil moisture content and layer thickness are input to simulate the stress distribution at different locations. If the stress value in a certain sub-region is significantly higher than the surrounding area, reaching 500 kPa per square meter, while the surrounding area only reaches 300 kPa, then it can be preliminarily determined that there is a stress concentration phenomenon at that location.

[0050] S303, based on the changing state of stress distribution in the loess layer, extracts sub-regions with stress concentration or abnormal distribution, and combines historical monitoring data with the current stress state to determine whether there is a potential trend of foundation settlement. For example, assuming that stress concentration persists in a certain sub-region and the cumulative ground settlement reaches 5 mm in the past 30 days, it can be inferred that it has a settlement trend and needs to be closely monitored.

[0051] S304. If a potential trend of foundation settlement is identified, historical settlement data for the corresponding sub-region is obtained from environmental monitoring records to determine the persistence characteristics of the settlement trend. For example, when obtaining historical settlement data from environmental monitoring records, the settlement change sequence over the past 90 days is extracted and its persistence characteristics are analyzed. Suppose that the settlement in a certain sub-region gradually increases from 2 mm per month to 4 mm per month, showing an accelerating trend, this indicates that the settlement risk may be escalating, providing a basis for subsequent predictions.

[0052] S305, considering the persistent nature of settlement trends, constructs a foundation settlement prediction model by combining current stress distribution change data. For example, when constructing the foundation settlement prediction model, current stress distribution data is combined with historical settlement trends, and a time window of the next 30 days is set to predict the possible settlement magnitude. If the model outputs that the future settlement of a certain sub-region may reach 8 mm, exceeding the safe range, then it is classified as a high-risk area.

[0053] S306. Based on the output of the prediction model, analyze the settlement risk distribution of each sub-region within the potential mechanical change area and determine the priority ranking of high-risk sub-regions. For example, after analyzing the settlement risk distribution and determining the priority ranking, when acquiring real-time environmental monitoring data for high-risk sub-regions, the stress value and settlement amount are updated hourly by sensors. Suppose that the stress value in a certain sub-region suddenly increases from 400 kPa to 600 kPa within 24 hours, indicating abnormal fluctuations in the short term, requiring increased vigilance.

[0054] S307: Based on the priority ranking of high-risk sub-regions, real-time environmental monitoring data is obtained to determine whether there are abnormal fluctuations in stress distribution in the short term, thus obtaining the final risk distribution status. For example, the determination of the final risk distribution status can be based on the data from the above-mentioned stages, designating sub-regions with continuous settlement, abnormal stress, and significant short-term fluctuations as the highest risk level. This multi-dimensional analysis method accurately locates problem areas, providing a reliable basis for subsequent protective measures, while improving monitoring efficiency and the rationality of resource allocation.

[0055] S400 extracts the settlement amplitude distribution map from the foundation settlement prediction model, uses the support vector machine algorithm to classify the high-risk areas in the settlement amplitude distribution map, obtains the deformation probability of all pipelines in the high-risk areas based on the pipeline stress distribution state, and draws a list of high-risk pipeline sections.

[0056] Specifically: S401. A settlement amplitude distribution map is obtained using a foundation settlement prediction model. Preliminary screening of data points in this map yields a preliminary range of high-risk areas. The distribution map visually reflects potential high-risk locations, providing a basis for subsequent screening. For example, when obtaining the settlement amplitude distribution map using the foundation settlement prediction model, the monitoring area is divided into multiple grid cells, each corresponding to a settlement amplitude value. Assume that the settlement amplitude in a certain area gradually decreases from the center to the periphery, reaching 10 mm at the center and only 2 mm at the periphery. During preliminary screening, a settlement amplitude threshold of 8 mm is set. Areas exceeding this value are marked as preliminary high-risk areas. If the settlement amplitude of a group of grid cells is all between 8 and 12 mm, it is included in the preliminary range.

[0057] S402. Based on the preliminary scope of the high-risk area, a support vector machine (SVM) algorithm is used to classify the data in the distribution map to determine the specific boundaries of the high-risk area. For example, when using the SVM algorithm to classify the distribution map data, subsidence amplitude and soil parameters are used as input features to train the model to distinguish between high-risk and low-risk areas. Assuming that after classification, the subsidence amplitude of a certain boundary area is 8.5 mm, and it forms a clear boundary with the surrounding low-risk areas, then the specific boundary of the high-risk area can be accurately determined.

[0058] S403 specifies the boundaries of high-risk areas, requiring the acquisition of corresponding underground pipeline layout information (such as pipeline burial depth, material, and direction data) from the environmental monitoring database to determine whether the pipeline is located within a high-risk area. For example, assuming a pipeline is buried at a depth of 2 meters and happens to be located within a high-risk area, further analysis of its potential impact is needed.

[0059] S404. If the pipeline is located in a high-risk area, its geometric parameters are extracted using pipeline layout information. Combined with settlement amplitude data, the stress distribution of the pipeline is calculated to obtain the potential trend of pipeline deformation. For example, when calculating the stress distribution of the pipeline, the stress conditions at different locations are simulated by combining settlement amplitude data and pipeline geometric parameters to identify potential problem points. Suppose that uneven settlement causes one end of a pipeline section to experience a stress of 200 kPa, while the other end only experiences 50 kPa, then it can be inferred that there is a deformation trend.

[0060] S405. Based on the potential trend of pipeline deformation, settlement change data for the pipeline area is obtained from historical monitoring records. The matching degree between the change data and the current settlement magnitude is analyzed to determine the persistence of the deformation trend. In this step, matching analysis enhances the reliability of the judgment. When analyzing the matching degree between the settlement change data for the pipeline area and the current settlement magnitude, historical records of the past 60 days are extracted. Assuming that historical data shows that the settlement increases by 3 mm per month, consistent with the current trend, it indicates that the deformation trend is persistent.

[0061] S406, by analyzing the persistence of deformation trends and combining them with boundary data of high-risk areas, constructs a pipeline deformation risk distribution map to determine the risk priority of each pipeline segment and identify those with higher risk priority. The distribution map provides an intuitive basis for priority ranking. When constructing the pipeline deformation risk distribution map, the boundary data of high-risk areas is combined with deformation trends to classify risk levels. If a pipeline segment is located within the boundary and exhibits a significant deformation trend, it is marked as a high-risk segment.

[0062] S407: For pipeline sections with high risk priority, real-time stress monitoring data is acquired. If the stress monitoring data exceeds a preset threshold, the section is marked as a key focus area, thus obtaining the final risk distribution status. For example, when acquiring real-time stress monitoring data for pipeline sections with high risk priority, sensors are deployed to record data once per hour. Suppose the stress value of a certain pipeline section suddenly increases from 150 kPa to 250 kPa, exceeding the preset threshold of 200 kPa, then it is listed as a key focus area.

[0063] S408: Based on the final risk distribution status, a list of high-risk pipeline sections and their corresponding risk levels are drawn. For example, the determination of the final risk distribution status can be based on a comprehensive analysis of pipeline deformation trends, stress data, and boundary ranges, clearly marking high-risk sections. This multi-dimensional analysis approach can effectively improve the accuracy of risk identification and provide reliable support for subsequent protective measures.

[0064] S500, for a list of high-risk pipeline sections, matches historical engineering deformation data of pipeline sections with high deformation potential with the current settlement amplitude distribution map to obtain quantitative indicators of deformation risk. Specifically: S501: Obtain deformation record data related to pipeline deformation from historical engineering archives, conduct preliminary screening of pipeline sections with high deformation risk priority, and determine the key areas of concern. For example, when obtaining deformation record data related to pipeline deformation from historical engineering archives, priority can be given to screening pipeline sections that have experienced significant deformation within the past five years. Suppose a pipeline section deformed by 5 mm three years ago due to foundation settlement, and similar records exist in the surrounding area, then this section is listed as a section of preliminary concern.

[0065] S502. For the selected key areas, acquire the corresponding settlement amplitude data and visualize it using a distribution graph to obtain the settlement amplitude distribution within the area. For example, when acquiring and visualizing settlement amplitude data for the selected key areas, the area can be divided into multiple small grids, each grid corresponding to a settlement value. Assuming the settlement value of the central grid in a certain area is 9 mm, gradually decreasing to 3 mm around the perimeter, the distribution graph can intuitively show the areas of concentrated settlement, providing a basis for subsequent analysis.

[0066] S503. Based on the settlement amplitude distribution, a comparative analysis method is used to match historical deformation records with the current state to determine the similarity distribution between the two. For example, when matching historical deformation records with the current settlement state using the comparative analysis method, the deformation trend in the historical data and the changing pattern of the current settlement amplitude can be extracted. Suppose the historical record shows that the settlement of a certain section increases by 2 mm per month, and the current data also shows a similar trend, then the similarity is high. If it exceeds a preset threshold of 80%, it is marked as a high-risk area.

[0067] S504. If the similarity distribution exceeds a preset threshold, the segment is marked as a high-risk area, and the specific location distribution of the pipelines is obtained based on the pipeline layout information of that area. For example, when obtaining the pipeline layout information of a high-risk area, the burial depth, length, and direction data of the pipelines can be extracted from the archives. Suppose a pipeline is buried at a depth of 1.5 meters and happens to pass through a settlement concentration area, then the correlation between its location distribution and the impact of settlement needs to be focused on, laying the foundation for subsequent analysis.

[0068] S505 analyzes the distribution of stress points in high-risk areas of pipelines by combining their specific location with the distribution of settlement amplitude, thus identifying areas of concentrated stress. For example, when analyzing the distribution of stress points in conjunction with the settlement amplitude distribution, attention can be paid to areas of concentrated stress caused by uneven settlement. If one end of a pipeline experiences greater stress due to settlement differences, while the other end is relatively stable, the area of ​​concentrated stress may be located in the settlement transition zone, requiring further analysis of its potential impact.

[0069] S506 specifies that for areas with concentrated stress points, corresponding geological stability data should be obtained from the environmental monitoring database. For example, when obtaining geological stability data from the environmental monitoring database, soil density and groundwater level information can be extracted. If the soil in a certain area is relatively loose and the groundwater level fluctuates significantly, it may exacerbate the risk of pipeline deformation, and the degree of impact needs to be comprehensively assessed in conjunction with the distribution of stress points.

[0070] S507, through comprehensive analysis of geological stability data and areas of concentrated stress points, establishes a priority ranking of pipeline deformation risks and determines monitoring schemes for high-priority sections. For example, when establishing the pipeline deformation risk priority ranking, a comprehensive assessment can be conducted based on settlement magnitude, stress concentration, and geological conditions to achieve reasonable resource allocation and ensure the safety of key areas. If a section has high settlement and unstable geological conditions, it is classified as high priority, and a detailed monitoring plan is developed, such as increasing monitoring frequency or deploying more sensors.

[0071] S600 uses deformation risk quantification indicators to conduct correlation analysis between deformation risk and leakage risk, obtains the potential propagation direction of leakage risk, and obtains the priority sequence of disaster intervention based on the distribution of vulnerable points along the impact path in the potential propagation direction. Specifically: S601, Correlation analysis of deformation risk and leakage risk, obtaining historical monitoring data related to deformation risk, and combining current environmental variables to construct a preliminary path model for risk propagation, thereby obtaining the potential propagation direction of leakage risk.

[0072] S602. Based on the potential propagation direction of leakage risk, network topology analysis is used to analyze the connection relationships between nodes, identify key nodes and paths for risk propagation, and determine the key areas that may trigger cascading effects. For example, when using network topology analysis, the pipeline system is abstracted as a network structure of nodes and edges. Assume a pipeline system has 5 key nodes, where node A connects to 3 main pipelines. If node A leaks, it may affect downstream nodes B and C. By analyzing the connection relationships, node A is identified as the key node for risk propagation, and the surrounding area is listed as a key area that may trigger cascading effects.

[0073] S603, for key areas of cascading impacts, retrieve corresponding stability parameters from the geological environment database, analyze external constraints on risk propagation within the region, and determine the distribution of vulnerable points along the impact path. For example, when analyzing geological stability parameters in key areas of cascading impacts, soil moisture content and formation compressibility data can be obtained from the database. If a region has high soil moisture content and easily compressible formations, it may become an external constraint on risk propagation, thus identifying that region as a vulnerable point along the impact path.

[0074] S604. If the density of vulnerable points affecting the path exceeds a preset threshold, the vulnerable points are prioritized and, combined with historical disaster records, a preliminary sequence of disaster interventions is obtained. For example, if the density of vulnerable points exceeds the preset threshold, such as more than 3 vulnerable points per square kilometer, they need to be prioritized and, combined with historical disaster records, such as past pipeline rupture events caused by loose soil, a preliminary sequence of disaster interventions is generated.

[0075] S605: Based on the preliminary sequence of disaster intervention, obtain pipeline layout data for relevant areas, analyze the spatial overlap between pipelines and vulnerable points, and determine the pipeline sections to be protected.

[0076] S606 specifies that for key protected pipeline sections, corresponding flow change data is obtained from the real-time monitoring system. Based on these flow changes, the final disaster intervention sequence is determined according to priority. For example, if a pipeline section directly passes through an area with concentrated vulnerability points and its burial depth is only 1.2 meters, it is designated as a key protected section. Further flow change data is obtained from the real-time monitoring system. If a sudden drop in flow exceeding 20% ​​of the normal range is detected, an automatic early warning mechanism is triggered, and the final intervention sequence is determined.

[0077] S607 generates corresponding resource scheduling instructions based on the final disaster intervention sequence. Through the system's automated allocation module, scheduling signals are sent to relevant equipment to complete the resource allocation for disaster intervention. For example, resource scheduling refers to the coordinated instructions issued by the system to associated monitoring equipment, emergency rescue equipment, automatic control units, and personnel dispatch platforms based on the final disaster intervention sequence. The aim is to complete enhanced monitoring, physical intervention, or process control of designated high-risk areas. When generating resource scheduling instructions, signals can be sent to monitoring equipment through the system's automated allocation module, thereby quickly responding to potential risks and ensuring efficient resource utilization. For instance, if a key area requires increased inspection frequency, the system will automatically allocate the nearest inspection equipment to that area, completing the resource allocation for disaster intervention.

[0078] S700 selects the highest priority section from the disaster intervention priority sequence. Through assessment, weighting, and numerical simulation, it obtains a simulation scenario for the high-risk section. Based on the stress distribution and historical settlement records of the simulation scenario, it adjusts the simulation scenario parameters to obtain an optimized foundation reinforcement parameter configuration. Specifically: S701: Data for the highest priority sections is obtained from the disaster intervention priority sequence. Combined with foundation settlement data, a preliminary assessment of the geological stability of these sections is conducted through information processing to determine the distribution range of high-risk sections. For example, when obtaining data for the highest priority sections from the disaster intervention priority sequence, attention can be paid to the geological environment of a critical section within the pipeline system. Suppose a section is located in a low-lying area. Based on foundation settlement data, it is found that the settlement of this section reached 5 cm in the past year, far exceeding the average of 2 cm in the surrounding area. A preliminary assessment of its geological stability is conducted through information processing, determining that this section is a high-risk area, with a distribution range covering a radius of 500 meters.

[0079] S702, for the distribution range of high-risk areas, acquire historical soil moisture content variation data. If the variation exceeds a preset threshold, the variation amplitude is weighted in the information processing stage to determine the allocation ratio of weighting factors. For example, for acquiring soil moisture content variation data in high-risk areas, data records from the past three months are extracted from monitoring equipment, and weighting helps to more accurately identify the dominant risk factors. Assuming the moisture content variation increases from the normal value of 10% to 25%, exceeding the preset threshold of 15%, the variation amplitude is given a higher weight in the information processing stage, determining its weight ratio to be 40%, while other factors such as settlement data account for 60%.

[0080] S703, based on the weighting ratios of various factors and combined with settlement data, uses numerical simulation methods to construct a simulation scenario for high-risk sections, obtaining the stress distribution in each sub-region within the scenario. When constructing simulation scenarios for high-risk sections, combining settlement data and weighting factors with numerical simulation analysis of stress distribution helps staff intuitively understand the risk distribution. For example, assuming the simulation results show that the stress concentration in a certain sub-region within the section is higher than in other areas, manifested as an abnormally high stress value 1 meter below the surface, this sub-region is marked as a potential risk point.

[0081] S704, regarding stress distribution, retrieves historical settlement records for the corresponding section from a pre-established geological database. If the historical settlement records show an abnormal trend, the abnormal trend is marked through the information processing stage to identify sub-areas requiring key attention. For example, assuming that the sub-area has experienced two abnormal settlements in the past five years, each exceeding 3 cm, accompanied by slight deformation of the pipeline, it is marked as a key area of ​​concern through the information processing stage.

[0082] S705, based on the sub-regions requiring key attention, obtains the initial configuration scheme of reinforcement parameters. By adjusting the parameters in a simulated scenario of the initial configuration scheme, an optimized combination of reinforcement parameters is obtained. For example, when determining the initial configuration scheme for the reinforcement parameters of the key sub-regions, the initial setting is a reinforcement material thickness of 20 cm and a support point spacing of 2 meters. Subsequently, through parameter adjustments in a simulated scenario, it was found that increasing the thickness to 25 cm and shortening the spacing to 1.5 meters resulted in a more uniform stress distribution, thus determining the optimized parameter combination. This simulation scenario adjustment process ensures the scientific validity of the reinforcement scheme, providing technical support for the reinforcement of key sub-regions in water conveyance projects in loess collapsing areas, and guaranteeing the long-term stable operation of the project.

[0083] The S706, for the optimized combination of hardening parameters, generates corresponding equipment scheduling instructions through the system automation module, sends instruction signals to relevant equipment, and completes the deployment of the hardening parameters. For example, for the optimized combination of hardening parameters, the system automation module generates equipment scheduling instructions. Suppose that the two hardening devices closest to the section need to be dispatched to the site, the system will automatically send signals and arrange for the equipment to be deployed within 24 hours. This automation significantly improves response speed and resource allocation efficiency.

[0084] This invention provides a real-time disaster risk control method for water conveyance projects in loess collapsing areas. Through data fusion analysis, it accurately identifies high-risk areas. Considering multiple influencing factors, a random forest algorithm is used to filter out areas requiring key attention, improving the targeting and effectiveness of monitoring. A foundation settlement prediction model is then established, and a support vector machine algorithm is used to identify high-risk areas from the settlement amplitude distribution map. The deformation probability of pipelines in high-risk areas is determined based on the pipeline stress distribution, further refining the risk assessment and making the prediction of foundation settlement and pipeline deformation more accurate. Then, historical engineering deformation data of pipeline sections with high deformation probability are matched with the current settlement amplitude distribution map to obtain high-risk pipeline deformation sections ranked by priority, allowing for targeted measures. Subsequently, by considering the mutual influence between pipeline deformation and leakage, as well as the correlation analysis of environmental factors on leakage propagation, and using the distribution of vulnerable points along the potential propagation path as a benchmark, combined with historical disaster records, a priority sequence for disaster intervention is obtained. This scientifically determines the key areas and order of disaster intervention, improving the efficiency and effectiveness of disaster intervention. Then, in targeted numerical simulation scenarios, the foundation reinforcement parameters are optimized by combining numerical simulation with historical data to improve the stability and reliability of the project, thereby enhancing the safety of the water conveyance project and improving the ability and efficiency to cope with disasters.

[0085] This invention provides a real-time risk control method for water conveyance projects in loess subsidence areas. By combining multi-source data fusion, intelligent algorithm prediction, and engineering mechanics simulation, it achieves dynamic identification, risk quantification, and proactive intervention for water pipeline disasters induced by loess subsidence. It integrates machine learning prediction results with historical time-series data to dynamically identify key areas most likely to experience subsidence deformation. Support vector machines are used to intelligently classify settlement amplitude distribution maps, and the possibility of deformation is judged based on the pipeline's stress state, generating a list of high-risk pipeline sections. By matching historical deformation data with the current settlement state, a deformation risk quantification index is constructed, and leakage risk is further correlated to identify potential leakage propagation directions and vulnerable points along the path, providing a basis for systemic risk prevention and control. Through proactive and intelligent risk control, it reduces sudden pipeline ruptures, leaks, or ground collapses, extends the service life of water conveyance projects, ensures water supply safety, and avoids the high costs and social impact of emergency repairs.

[0086] This invention also discloses a system for real-time control of disaster risks in water conveyance projects in loess subsidence areas, referring to... Figure 4 ,include: The data acquisition and fusion module is used to collect and fuse rainfall intensity, surface humidity and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone to generate a comprehensive environmental monitoring and analysis dataset. The monitoring area analysis module is used to predict the trend of rainfall infiltration mechanics changes based on the comprehensive environmental monitoring analysis dataset and the random forest algorithm. The prediction results are then fused with the time series analysis of historical rainfall infiltration data to obtain high-priority monitoring areas. The prediction model building module is used to establish a foundation settlement prediction model based on the soil moisture content of high-priority monitoring areas and by combining the finite element simulation method with the stress distribution changes of loess layers. The pipeline deformation analysis module is used to extract the settlement amplitude distribution map from the foundation settlement prediction model, use the support vector machine algorithm to classify the high-risk areas in the settlement amplitude distribution map, determine the deformation probability of all pipelines in the high-risk areas based on the pipeline stress distribution state, and draw a list of high-risk pipeline sections. The risk matching and sorting module matches historical engineering deformation data of pipeline sections with high deformation probability with the current settlement amplitude distribution map for the list of high-risk pipeline sections to obtain quantitative indicators of deformation risk. The disaster intervention analysis module is used to conduct correlation analysis between deformation risk and leakage risk based on the deformation risk quantification index, obtain the potential propagation direction of leakage risk, and obtain the priority sequence of disaster intervention based on the distribution of vulnerable points along the impact path in the potential propagation direction. The reinforcement parameter optimization module is used to select the highest priority section from the priority sequence of disaster intervention, obtain the simulation scenario of high-risk section through evaluation, weight assignment and numerical simulation, and adjust the parameters of the simulation scenario based on the stress distribution to obtain the optimized foundation reinforcement parameter configuration.

[0087] This invention discloses a system for real-time control of disaster risks in water conveyance projects in loess subsidence areas. This system integrates data acquisition and fusion modules to automatically and continuously monitor and analyze key environmental factors such as rainfall, humidity, and moisture content. It comprehensively utilizes random forest algorithms for trend prediction and time-series analysis to select high-priority monitoring areas, and constructs a high-fidelity foundation settlement prediction model based on finite element simulation. The pipeline deformation analysis module and risk matching and ranking module incorporate support vector machine classification and historical data matching techniques to intelligently analyze the specific impact of settlement on the pipeline structure. The disaster intervention analysis module correlates deformation and leakage risks, analyzes risk propagation paths and vulnerabilities, and generates a priority sequence based on risk assessment. Management personnel then implement reinforcement measures based on the system-generated intervention priorities and recommended reinforcement parameter configurations to ensure water conveyance safety.

[0088] This invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, and is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions from a computer storage medium to implement the corresponding method flow or corresponding function. This invention also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method. The computer-readable storage medium is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both built-in storage media in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space containing the terminal's operating system. Furthermore, this storage space also contains one or more instructions suitable for loading and execution by a processor; these instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium.

[0089] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0090] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0091] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0092] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0093] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

[0094] The above description is merely a preferred embodiment of the present invention and is not intended to limit the technical solution of the present invention in any way. Those skilled in the art should understand that, without departing from the spirit and principles of the present invention, the technical solution can be modified and replaced in several simple ways, and these modifications and replacements are all within the scope of protection covered by the claims.

Claims

1. A method for real-time control of disaster risk in water conveyance projects in loess collapsing areas, characterized in that, include: Rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone were collected and integrated to generate a comprehensive environmental monitoring and analysis dataset. Based on the comprehensive analysis dataset of environmental monitoring, the random forest algorithm is used to predict the trend of rainfall infiltration mechanics changes. The prediction results are then fused with the time series analysis of historical rainfall infiltration data to obtain high-priority monitoring areas. Based on the soil moisture content of high-priority monitoring areas, a foundation settlement prediction model was established using the finite element simulation method combined with the stress distribution changes in the loess layer. The settlement amplitude distribution map is extracted from the foundation settlement prediction model. The support vector machine algorithm is used to classify the high-risk areas in the settlement amplitude distribution map. Based on the stress distribution state of the pipeline, the deformation possibility of all pipelines in the high-risk area is judged, and a list of high-risk pipeline sections is drawn. For the list of high-risk pipeline sections, historical engineering deformation data of pipeline sections with high deformation potential are matched with the current settlement amplitude distribution map to obtain quantitative indicators of deformation risk; Based on the quantitative indicators of deformation risk, a correlation analysis of deformation risk and leakage risk is conducted to obtain the potential propagation direction of leakage risk. Based on the distribution of vulnerable points along the impact path in the potential propagation direction, a priority sequence for disaster intervention is obtained. The highest priority section is selected from the priority sequence of disaster intervention. The high-risk section simulation scenario is obtained through evaluation, weight assignment and numerical simulation. The parameters of the simulation scenario are adjusted based on the stress distribution in the simulation scenario to obtain the optimized foundation reinforcement parameter configuration.

2. The method for real-time control of disaster risk in water conveyance projects in loess collapsing areas according to claim 1, characterized in that, Rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone were collected and integrated to generate a comprehensive environmental monitoring and analysis dataset, including: Real-time data on rainfall intensity, surface humidity, and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone are collected and stored as raw monitoring datasets. Based on the original monitoring dataset, preprocessing methods were used to denoise and standardize the rainfall intensity, surface humidity and soil moisture content data to obtain a standardized dataset. Data fusion methods are used to integrate standardized datasets and generate a unified fused dataset. If the rainfall intensity data of a certain area in the fused dataset exceeds a preset threshold, then the surface humidity and soil moisture content data of that area are subjected to weighted analysis to obtain the distribution of high-risk areas; Extract the soil moisture content change trend corresponding to the distribution of high-risk areas to determine whether there is a risk of subsidence. Based on the assessment results of the risk of subsidence and the rainfall intensity data of the area, the safety assessment results of the water conveyance project in the area were obtained; Based on the safety assessment results, comprehensive environmental monitoring analysis data for water conveyance projects in loess subsidence areas is generated.

3. The method for real-time control of disaster risk in water conveyance projects in loess collapsing areas according to claim 1, characterized in that, Based on the comprehensive environmental monitoring analysis dataset, the random forest algorithm was used to analyze the trends in rainfall infiltration mechanics in high-risk areas, including: The random forest algorithm was used to predict the trend of rainfall infiltration mechanics, and the predicted trend of rainfall infiltration was obtained. Extract the data portion where the penetration intensity exceeds a preset threshold from the predicted trend results to determine the distribution of potential areas of mechanical change; For the distribution of potential mechanical changes, historical rainfall infiltration records for the corresponding areas are obtained from the comprehensive environmental monitoring analysis dataset, and the infiltration intensity change sequence for the corresponding areas is obtained. By performing time-series analysis on the infiltration intensity variation sequence, it can be determined whether there is a continuous mechanical change tendency in the region; If the preliminary assessment results indicate that the mechanical changes are likely to continue, the environmental monitoring data for that area will be weighted to determine the high-priority monitoring areas.

4. The method for real-time control of disaster risk in water conveyance projects in loess collapsing areas according to claim 1, characterized in that, Using the soil moisture content of high-priority monitoring areas as a benchmark, a foundation settlement prediction model was established using the finite element method combined with changes in loess stress distribution, including: Collect soil moisture data in high-priority monitoring areas, obtain real-time changes in soil moisture increase, and determine whether the increase exceeds the preset threshold. If the increase in soil moisture content exceeds the preset threshold, the stress distribution is calculated using the finite element simulation method based on the loess layer data in the area to obtain the change state of the stress distribution in the loess layer. Based on the changing state of stress distribution in the loess layer, extract sub-regions with stress concentration or abnormal distribution to determine whether there is a potential trend of foundation settlement. If a potential trend of foundation settlement is identified, historical settlement data for the corresponding sub-area is obtained from environmental monitoring records to determine the persistence characteristics of the settlement trend. Based on the persistent nature of the settlement trend and combined with current stress distribution change data, a foundation settlement prediction model is constructed. Based on the output of the prediction model, the distribution of settlement risk in each sub-region within the potential mechanical change area is analyzed, and the priority ranking of high-risk sub-regions is determined. Based on the priority ranking of high-risk sub-regions, real-time environmental monitoring data is obtained to determine whether there are abnormal fluctuations in stress distribution in the short term, thus obtaining the final risk distribution status.

5. The method for real-time control of disaster risk in water conveyance projects in loess collapsing areas according to claim 1, characterized in that, Settlement amplitude distribution maps are extracted from the foundation settlement prediction model. A support vector machine algorithm is used to classify high-risk areas in the settlement amplitude distribution maps. Based on the stress distribution state of the pipelines, the deformation probability of all pipelines in the high-risk areas is obtained, including: The settlement amplitude distribution map is obtained by using the foundation settlement prediction model. The data points in the distribution map are initially screened to obtain the preliminary range of the high-risk area. Based on the preliminary scope of the high-risk area, the support vector machine algorithm is used to classify the data in the distribution graph to determine the specific boundaries of the high-risk area; For the specific boundaries of high-risk areas, obtain the corresponding underground pipeline layout information from the environmental monitoring database to determine whether the pipeline is located within the high-risk area; If the pipeline is located in a high-risk area, its geometric parameters can be extracted through pipeline layout information, and combined with settlement amplitude data, the stress distribution of the pipeline can be calculated to obtain the potential trend of pipeline deformation.

6. The method for real-time control of disaster risk in water conveyance projects in loess collapsing areas according to claim 1, characterized in that, By matching historical engineering deformation data of pipeline sections with high deformation potential with the current settlement amplitude distribution map, quantitative indicators of deformation risk are obtained, including: Deformation records related to pipeline deformation were obtained from historical engineering archives. Pipeline sections with high deformation risk priority were initially screened to determine the key areas of concern. For the selected key areas, the corresponding settlement amplitude data is obtained and visualized through distribution graphics to obtain the settlement amplitude distribution status within the area; Based on the settlement amplitude distribution, a comparative analysis method is used to match historical deformation records with the current state to determine the similarity distribution between the two. If the similarity distribution exceeds the preset threshold, the segment is marked as a high-risk area, and the specific location distribution of the pipeline is obtained based on the pipeline layout information of the area. By analyzing the specific location distribution of the pipeline and the distribution of settlement amplitude, the distribution of stress points in the high-risk area of ​​the pipeline is analyzed, and the concentrated area of ​​stress points is obtained. For areas with concentrated stress points, obtain corresponding geological stability data from the environmental monitoring database; By comprehensively analyzing geological stability data and areas of concentrated stress points, a priority ranking of pipeline deformation risks is constructed.

7. The method for real-time control of disaster risk in water conveyance projects in loess collapsing areas according to claim 1, characterized in that, Based on quantitative indicators of deformation risk, a correlation analysis of deformation risk and leakage risk is conducted to obtain the potential propagation direction of leakage risk. Based on the distribution of vulnerable points along the impact path in the potential propagation direction, a priority sequence for disaster intervention is obtained, including: Correlation analysis was performed on deformation risk and leakage risk to obtain historical monitoring data related to deformation risk. Combined with current environmental variables, a preliminary path model for risk propagation was constructed to obtain the potential propagation direction of leakage risk. Based on the potential propagation direction of leakage risk, network topology analysis is used to analyze the connection relationship between nodes, identify key nodes and paths for risk propagation, and determine the key areas that will trigger a chain reaction. For key areas with cascading effects, corresponding stability parameters are obtained from the geological environment database, external constraints on risk propagation within the region are analyzed, and the distribution of vulnerable points along the impact path is determined. If the density of vulnerable points affecting the path exceeds a preset threshold, the vulnerable points are prioritized and combined with historical disaster records to obtain a preliminary sequence of disaster interventions. Based on the preliminary sequence of disaster intervention, obtain pipeline layout data for relevant areas, analyze the spatial overlap between pipelines and vulnerable points, and determine the pipeline sections that require key protection. For key pipeline sections requiring protection, corresponding flow change data are obtained from the real-time monitoring system, and the final disaster intervention sequence is determined based on the flow change.

8. The method for real-time control of disaster risk in water conveyance projects in loess collapsing areas according to claim 1, characterized in that, The highest priority section is selected from the disaster intervention priority sequence. Through assessment, weighting, and numerical simulation, a simulation scenario for the high-risk section is obtained. Based on the stress distribution and historical settlement records of the simulation scenario, the simulation scenario parameters are adjusted to obtain the optimized foundation reinforcement parameter configuration, including: Data on the highest priority sections are obtained from the disaster intervention priority sequence. Combined with settlement data of foundation settlement, the geological stability of the sections is preliminarily assessed through information processing to obtain the distribution range of high-risk sections. For the distribution range of high-risk areas, data on the variation of soil moisture content is obtained. If the variation exceeds the preset threshold, the variation is weighted through information processing to determine the allocation ratio of weight factors. Based on the weighting ratio and settlement data, a numerical simulation method is used to construct a simulation scenario for high-risk sections, and the stress distribution of each sub-region within the scenario is obtained. Based on the stress distribution, historical settlement records for the corresponding sections are obtained from a pre-established geological database. If the historical settlement records show an abnormal trend, the abnormal trend is marked through information processing to identify the sub-areas that need to be focused on. Based on the sub-regions that require special attention, obtain the initial configuration scheme of their reinforcement parameters. By adjusting the parameters in the simulated scenario of the initial configuration scheme, the optimized combination of reinforcement parameters is obtained.

9. A system for implementing the real-time disaster risk control method for water conveyance projects in loess collapsing areas as described in any one of claims 1 to 8, characterized in that, include: The data acquisition and fusion module is used to collect and fuse rainfall intensity, surface humidity and soil moisture content within the water conveyance project area of ​​the Loess Collapse Zone to generate a comprehensive environmental monitoring and analysis dataset. The monitoring area analysis module is used to predict the trend of rainfall infiltration mechanics changes based on the comprehensive environmental monitoring analysis dataset and the random forest algorithm. The prediction results are then fused with the time series analysis of historical rainfall infiltration data to obtain high-priority monitoring areas. The prediction model building module is used to establish a foundation settlement prediction model based on the soil moisture content of high-priority monitoring areas and by combining the finite element simulation method with the stress distribution changes of loess layers. The pipeline deformation analysis module is used to extract the settlement amplitude distribution map from the foundation settlement prediction model, use the support vector machine algorithm to classify the high-risk areas in the settlement amplitude distribution map, determine the deformation probability of all pipelines in the high-risk areas based on the pipeline stress distribution state, and draw a list of high-risk pipeline sections. The risk matching and sorting module matches historical engineering deformation data of pipeline sections with high deformation probability with the current settlement amplitude distribution map for the list of high-risk pipeline sections to obtain quantitative indicators of deformation risk. The disaster intervention analysis module is used to conduct correlation analysis between deformation risk and leakage risk based on the deformation risk quantification index, obtain the potential propagation direction of leakage risk, and obtain the priority sequence of disaster intervention based on the distribution of vulnerable points along the impact path in the potential propagation direction. The reinforcement parameter optimization module is used to select the highest priority section from the priority sequence of disaster intervention, obtain the simulation scenario of high-risk section through evaluation, weight assignment and numerical simulation, and adjust the parameters of the simulation scenario based on the stress distribution to obtain the optimized foundation reinforcement parameter configuration.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images