Intelligent surveying and mapping method for geographic information of water conservancy project

Abstract

The invention discloses an intelligent surveying and mapping method for geographic information of a water conservancy project, which relates to the technical field of water conservancy projects and comprises the following steps: identifying a core risk point of the water conservancy project, dividing a monitoring area, deploying surveying and mapping equipment for the divided monitoring area, setting parameters, and collecting surveying and mapping data. The method comprises the following steps: performing primary verification on surveying and mapping data and a risk threshold library, performing secondary verification in combination with a historical database, when a real-time risk is triggered, triggering graded early warning according to the risk level of the real-time risk, generating a key retest scheme according to the triggered early warning level, performing retest, and adjusting the early warning level. According to the invention, different types of surveying and mapping equipment are deployed in a targeted manner, differential parameters are set, primary verification is carried out on surveying and mapping data and a risk threshold library, secondary verification is carried out in combination with a historical database, and a dynamic retest scheme is automatically generated by means of a perfect risk grading early warning mechanism. And operation and maintenance personnel are assisted to quickly and clearly master the engineering safety state.

Description

Technical Field

[0001] This invention relates to the field of water conservancy engineering technology, and in particular to an intelligent mapping method for geographic information of water conservancy projects. Background Technology

[0002] In recent years, with the continuous expansion of water conservancy project construction, the safe operation of facilities such as reservoirs, dams, and river channels has been increasingly affected by factors such as extreme weather and geological changes. Dynamic monitoring and risk warning of their geographic information have become a key link in ensuring the safety of people's lives and property and maintaining regional water security. In the current management of water conservancy projects, the demand for monitoring key indicators such as structural deformation, seepage pressure changes, and crack development is increasing. However, existing monitoring methods are difficult to fully match the requirements of modern management in terms of coverage, response speed, and data accuracy. The rapid iteration of technologies such as the Internet of Things, UAV remote sensing, and big data analysis has provided more technological possibilities for water conservancy project monitoring. The industry urgently needs a solution to improve the timeliness of risk identification and the scientific nature of risk handling in water conservancy projects, and to ensure the long-term stable operation of facilities.

[0003] In existing technologies, most monitoring methods adopt a uniform mode and fail to differentiate deployment according to the inherent risk levels of different areas of water conservancy projects. The equipment configuration and collection frequency are the same for high-risk points and regular areas, which can easily lead to missed early warning opportunities in high-risk areas and wasted resources in regular areas. Data processing and verification rely on single threshold comparisons and lack secondary verification of historical data. They are prone to misjudgment or omission due to accidental anomalies or environmental interference, and cannot distinguish between real risks and data deviations. The early warning classification mechanism is imperfect, there is no dynamic retesting scheme, it is difficult to adjust retesting parameters, and the early warning push is singular, making it difficult to ensure that the responsible parties can deal with it in a timely manner. In some schemes, the data transmission is not pre-processed, and the direct transmission of raw data to the cloud increases bandwidth pressure and reduces efficiency. Insufficient data visualization is also not conducive to operation and maintenance personnel grasping the safety status of the project. Summary of the Invention

[0004] The technical problems solved by this invention are: existing monitoring mostly adopts a uniform mode, without differentiating deployment according to the inherent risk levels of different areas of water conservancy projects. The equipment configuration and collection frequency of high-risk points are the same as those of regular areas, which easily leads to missed early warning opportunities in high-risk areas and waste of resources in regular areas. Data processing and verification rely on a single threshold comparison, lacking secondary verification of historical data, which is prone to misjudgment or omission due to accidental anomalies or environmental interference. It is impossible to distinguish between real risks and data deviations. The early warning classification mechanism is imperfect, there is no dynamic retesting scheme, it is difficult to adjust retesting parameters, and the early warning push is singular, making it difficult to ensure timely handling by the responsible parties. In some schemes, data transmission is not pre-processed, and direct transmission of raw data to the cloud increases bandwidth pressure and reduces efficiency. Insufficient data visualization is also not conducive to operation and maintenance personnel grasping the safety status of the project.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an intelligent mapping method for geographic information of water conservancy projects, comprising the following steps:

[0006] Step S1: Identify the core risk points of the water conservancy project and divide the monitoring area according to the inherent risk level of the core risk points;

[0007] Step S2: Deploy surveying equipment and set parameters for the divided monitoring area;

[0008] Step S3: Use the surveying equipment to collect surveying data and establish a data transmission link to transmit the surveying data to the cloud platform;

[0009] Step S4: The survey data is verified against the risk threshold database once, and then verified again in conjunction with the historical database. When a real-time risk is triggered, a graded warning is triggered according to the risk level of the real-time risk.

[0010] Step S5: Based on the triggered warning level, generate a key retest plan, conduct the retest, and adjust the warning level.

[0011] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, step S1 specifically includes:

[0012] The project utilizes water conservancy engineering designers and operation and maintenance teams to conduct on-site surveys, and combines project types with historical accident data to identify core risk points. The project types include reservoirs, dams, and rivers, and the historical accident data includes historical accident data for reservoirs, dams, and rivers.

[0013] The mapping area of ​​the core risk point is delineated according to the inherent risk level of the core risk point. The inherent risk level includes high-risk points, medium-risk points and low-risk points. The mapping area of ​​high-risk points is designated as a high-frequency monitoring area, and the mapping areas of medium-risk points and low-risk points are designated as regular monitoring areas.

[0014] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, step S2 specifically includes:

[0015] Surveying equipment is deployed in the high-frequency monitoring area and the conventional monitoring area respectively, and the parameters of the surveying equipment in the high-frequency monitoring area and the conventional monitoring area are set respectively;

[0016] The mapping equipment in the high-frequency monitoring area includes fixed-wing UAVs, ground displacement monitoring stations, and embedded pressure sensors;

[0017] The surveying equipment in the conventional monitoring area includes multi-rotor drones and mobile ground surveying robots.

[0018] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, the step of setting parameters for the mapping equipment in the high-frequency monitoring area and the conventional monitoring area specifically includes:

[0019] For high-frequency monitoring areas: set the flight altitude of the fixed-wing UAV to the first preset flight altitude, adopt the first preset data acquisition cycle and enable real-time image transmission, adjust the ground displacement monitoring station to the first preset positioning accuracy, and set the embedded pressure sensor to the preset acquisition frequency.

[0020] For routine monitoring areas: set the flight altitude of the multi-rotor UAV to the second preset flight altitude, adopt the second preset data acquisition cycle and turn off real-time image transmission, and adjust the mobile ground mapping robot to the second preset positioning accuracy;

[0021] The first preset flight altitude is less than the second preset flight altitude, the first preset data acquisition period is less than the second preset data acquisition period, and the first preset positioning accuracy is greater than the second preset positioning accuracy.

[0022] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, step S3 specifically includes:

[0023] Surveying data is collected using surveying equipment in the high-frequency monitoring area and the conventional monitoring area. The surveying data includes displacement, seepage pressure, elevation change, crack width, aerial imagery, crack images, and distribution of leakage points. A data transmission link is constructed to transmit the surveying data to a cloud platform, and the visualization module of the cloud platform is used to visualize the surveying data.

[0024] The construction of the data transmission link includes:

[0025] The surveying data is received by edge computing nodes deployed at the water conservancy project surveying site, and the surveying data is preprocessed to obtain processed data. The processed data is then transmitted to the cloud platform through dual-mode communication technology. The preprocessing includes outlier handling, spatiotemporal coordinate unification, data format standardization, and data compression.

[0026] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, step S4 specifically includes:

[0027] The system acquires surveying data from the cloud platform and performs a first verification against the risk threshold database to determine whether a real-time risk has occurred. If a real-time risk is determined, a second verification is performed using the historical database to determine whether a real-time risk has occurred. If a real-time risk is determined, the risk level of the real-time risk is determined according to the preset risk level determination rules. A graded early warning is triggered based on the risk level of the real-time risk, and the location of the real-time risk is marked. The risk levels include Level 1 risk, Level 2 risk, and Level 3 risk.

[0028] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, the step of verifying the mapping data against a risk threshold database to determine whether a real-time risk has occurred specifically includes:

[0029] Based on the attributes of the surveying data, the surveying data is divided into quantitative data and qualitative data;

[0030] The quantitative data includes displacement, seepage pressure, elevation change, and crack width; the qualitative data includes aerial images, crack images, and distribution of leakage points.

[0031] For quantitative data, a direct numerical comparison method is used to obtain the specific value of the quantitative data, and the specific value is compared with a preset threshold in the risk threshold library.

[0032] If the specific value of one of the quantitative data exceeds the preset threshold, it is determined that a risk has occurred;

[0033] If all the specific values ​​of the quantitative data are less than the preset threshold, it is determined that no risk has occurred;

[0034] For qualitative data, a feature matching comparison method is used to obtain key feature parameters of the surveying and mapping images in the qualitative data, and the key feature parameters are compared with preset feature thresholds of the corresponding scenes of the surveying and mapping images in the risk threshold library.

[0035] If a key feature parameter of a mapping image in one of the qualitative data exceeds a preset feature threshold, it is determined that a risk has occurred.

[0036] If all key feature parameters of the surveyed images in the qualitative data are less than the preset feature threshold, then it is determined that no risk has occurred.

[0037] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, the further step of combining historical databases for secondary verification to determine whether a real-time risk has occurred specifically includes:

[0038] For quantitative data, a trend similarity verification method is used to extract the trend features of the quantitative data, and the trend features are compared with the trend features of historical normal working conditions and historical risk working conditions in the historical database.

[0039] If the similarity between the trend characteristics of the quantitative data and the trend characteristics of historical normal operating conditions is greater than or equal to the first threshold, it is determined that no risk has occurred.

[0040] If the similarity between the trend characteristics of the quantitative data and the trend characteristics of historical risk conditions is greater than or equal to the first threshold, then it is determined that a risk has occurred.

[0041] For qualitative data, a feature correlation verification method is used to extract key features of the surveying images in the qualitative data, and these key features are compared with the risk features of historical normal working conditions and historical risk working conditions in the historical database.

[0042] If the overlap rate between the key features of the mapping images in the qualitative data and the risk features of historical normal operating conditions is greater than or equal to the second threshold, it is determined that no risk has occurred.

[0043] If the overlap rate between the key features of the mapping images in the qualitative data and the risk features of historical risk conditions is greater than or equal to the second threshold, then it is determined that a risk has occurred.

[0044] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, the step of triggering graded early warning based on the risk level of the real-time risk specifically includes:

[0045] When the real-time risk is classified as Level 1 risk, a Level 1 warning is triggered. Pop-ups and voice calls are simultaneously pushed to engineering operation and maintenance personnel and local water conservancy management departments through the cloud platform, and on-site audible and visual alarms are activated.

[0046] When the real-time risk is a level 2 risk, a level 2 warning is triggered, and a pop-up window and a text message are sent to the engineering operation and maintenance personnel through the cloud platform.

[0047] When the real-time risk is classified as Level 3, a Level 3 warning is triggered, and a pop-up window is pushed to the engineering and maintenance personnel through the cloud platform.

[0048] As a preferred embodiment of the intelligent mapping method for geographic information of water conservancy projects according to the present invention, step S5 specifically includes:

[0049] Based on the triggered warning level, the cloud platform automatically generates a key retesting plan. The key retesting plan includes the retesting scope, retesting resources, and retesting frequency. Retesting is carried out according to the key retesting plan, retesting data is obtained, and the surveying data and retesting data are compared to determine the risk change trend. The warning level is adjusted according to the risk change trend, and disposal suggestions are generated.

[0050] Based on the triggered warning level, the cloud platform automatically generates a key retest plan, including:

[0051] If a Level 1 warning is triggered, the retesting scope includes the real-time risk location and the surrounding area that is greater than the third threshold but less than the fourth threshold. Retesting resources are prioritized for fixed-wing UAVs and ground displacement monitoring stations, and the retesting frequency is set to the fifth threshold.

[0052] If a Level 2 warning is triggered, the retesting scope includes the real-time risk location and the surrounding area that is greater than the sixth threshold but less than the third threshold. Retesting resources will be prioritized for multi-rotor drones and mobile ground mapping robots, and the retesting frequency will be set to the seventh threshold.

[0053] If a Level 3 warning is triggered, the retesting scope includes real-time risk locations, and mobile ground mapping robots will be prioritized for retesting resources. The retesting frequency will be set to the eighth threshold.

[0054] The adjustment of the early warning level based on the risk change trend includes:

[0055] If the risk change trend matches the preset safety change trend, then the warning level will be lowered by one level;

[0056] If the risk change trend does not conform to the preset safety change trend, the warning level will be raised by one level, and handling suggestions will be pushed to the cloud platform.

[0057] The beneficial effects of this invention are as follows: By dividing high-frequency monitoring areas and routine monitoring areas according to the inherent risk levels of different regions of water conservancy projects, and by deploying different types of surveying equipment and setting differentiated parameters, this invention ensures that high-risk areas are monitored with higher accuracy and shorter cycles, effectively avoiding missed early warning opportunities, while also preventing the waste of monitoring resources in routine areas, thus achieving a rational allocation of monitoring resources. Through primary verification of surveying data against a risk threshold database, and secondary verification using a historical database, misjudgments and omissions caused by accidental data anomalies or environmental interference are effectively eliminated, significantly improving the accuracy of risk assessment and accurately distinguishing true risks. In response to data deviations, a robust risk grading and early warning mechanism, coupled with a dynamic retesting plan automatically generated based on the early warning level, allows for flexible adjustment of the retesting scope, resources, and frequency. Furthermore, differentiated early warning push methods are adopted according to risk levels to ensure that risks of different levels are received and addressed in a timely manner, thereby improving risk response efficiency. By preprocessing surveying data through edge computing nodes before transmitting it to the cloud, the network bandwidth pressure caused by direct transmission of raw data is significantly reduced, improving data processing efficiency and the real-time nature of risk early warning. At the same time, the visualization module of the cloud platform can intuitively present surveying data and risk distribution, helping maintenance personnel to quickly and clearly grasp the safety status of the project. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the basic process of an intelligent mapping method for geographic information of water conservancy projects, provided as an embodiment of the present invention. Detailed Implementation

[0059] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0060] Example, refer to Figure 1 As an embodiment of the present invention, an intelligent mapping method for geographic information of water conservancy projects is provided, comprising the following steps:

[0061] Step S1: Identify the core risk points of the water conservancy project and divide the monitoring area according to the inherent risk level of the core risk points.

[0062] Step S2: Deploy surveying equipment and set parameters for the divided monitoring area.

[0063] Step S3: Use surveying equipment to collect surveying data and build a data transmission link to transmit the surveying data to the cloud platform.

[0064] Step S4 involves verifying the survey data against the risk threshold database once, and then performing a second verification using the historical database. When a real-time risk is triggered, a tiered warning is issued based on the risk level of the real-time risk.

[0065] Step S5: Based on the triggered warning level, generate a key retest plan, conduct the retest, and adjust the warning level.

[0066] Step S1 specifically includes:

[0067] The project utilizes on-site surveys conducted by water conservancy engineering designers and operation and maintenance teams. At the same time, it combines project types and historical accident data to identify core risk points. Project types include reservoirs, dams, and rivers, while historical accident data includes historical accident data for reservoirs, dams, and rivers.

[0068] The mapping areas of core risk points are delineated based on their inherent risk levels. The inherent risk levels include high-risk points, medium-risk points, and low-risk points. The mapping areas of high-risk points are designated as high-frequency monitoring areas, while the mapping areas of medium-risk points and low-risk points are designated as regular monitoring areas.

[0069] Core risk point identification is achieved through "manual investigation + data support": In reservoir projects, designers investigate dam cracks and spillway conditions, while the operation and maintenance team supplements seepage observation data. Combined with historical accident data on piping and spillway siltation in similar reservoirs, the back slope of the dam and the silted section of the spillway are identified as core risk points. In dam projects, designers check the density of the dam body and the seepage prevention curtain, while the operation and maintenance team reports the location of leakage. Referring to historical cases of seepage due to dam cracks and scour at the dam toe, the loose fill section and the scour area at the dam toe are listed as core risk points. In river projects, designers check the scour of bends and the siltation of the riverbed, while the operation and maintenance team supplements the records of mainstream line deviation. Based on historical data on the impact of bend collapse and siltation on flood control, the concave bank of the bend and the highly silted river section are identified as core risk points.

[0070] The inherent risk level is defined by combining the probability of risk occurrence and the extent of loss: high-risk points are those with a probability ≥30% and the accident causing flooding of downstream villages and casualties, such as obvious cracks on the upstream slope of the reservoir dam and similar cracks that have previously caused leakage; medium-risk points are those with a probability of 10%-30% and the accident causing damage to local facilities, such as the dam body with slightly loose fill but no history of leakage and no important facilities nearby; low-risk points are those with a probability <10% and the accident has a small impact, such as no important buildings on both sides of the river, light siltation and no history of collapse, in order to improve the objectivity of the classification.

[0071] Combining the on-site survey experience of water conservancy engineering designers and operation and maintenance teams, this approach fully leverages the professionals' familiarity with the actual conditions of engineering structures and geological conditions, reducing potential biases from relying solely on data. Furthermore, it incorporates historical accident data corresponding to different engineering types (reservoirs, dams, rivers), allowing for the identification of high-risk aspects of different engineering types from past accident patterns. This makes the identification of core risk points more targeted and reliable, avoiding the omission of critical hidden danger areas. Based on inherent risk levels (high, medium, low), core risk points are divided into high-frequency monitoring areas and routine monitoring areas. This directly guides the selection of surveying equipment, parameter settings, and monitoring frequency planning in subsequent steps, ensuring that high-risk areas receive denser and more accurate monitoring resources, while routine areas adopt appropriate monitoring schemes, achieving the monitoring goal of "focused monitoring of key areas and efficient coverage of general areas."

[0072] Step S2 specifically includes:

[0073] Surveying equipment was deployed in both the high-frequency monitoring area and the conventional monitoring area, and the parameters of the surveying equipment in the high-frequency monitoring area and the conventional monitoring area were set separately.

[0074] The mapping equipment in the high-frequency monitoring area includes fixed-wing UAVs, ground displacement monitoring stations, and embedded pressure sensors.

[0075] By leveraging the collaboration of multiple devices, it enables large-scale and rapid aerial photography, continuous tracking of minute structural displacements, and real-time monitoring of internal seepage pressure, comprehensively covering key safety indicators in high-risk areas and meeting the high demands of high-risk areas for monitoring accuracy and real-time performance.

[0076] The surveying equipment for the regular monitoring area includes multi-rotor drones and mobile ground surveying robots.

[0077] It can meet basic monitoring needs through flexible short-to-medium distance surveys and detailed ground mapping, while avoiding redundant investment in high-cost equipment and reducing resource waste. At the same time, targeted parameter settings further ensure that equipment in different monitoring areas can output effective data according to actual needs, providing accurate and adaptable hardware support for subsequent survey data collection and risk verification, ensuring that the entire monitoring system operates efficiently while also being economical.

[0078] The parameters for the surveying equipment in the high-frequency monitoring area and the conventional monitoring area were set separately, specifically including:

[0079] For high-frequency monitoring areas: set the flight altitude of the fixed-wing UAV to the first preset flight altitude, adopt the first preset data acquisition cycle and enable real-time image transmission, adjust the ground displacement monitoring station to the first preset positioning accuracy, and set the embedded pressure sensor to the preset acquisition frequency.

[0080] For routine monitoring areas: set the flight altitude of the multi-rotor UAV to the second preset flight altitude, adopt the second preset data acquisition cycle and turn off real-time image transmission, and adjust the mobile ground mapping robot to the second preset positioning accuracy.

[0081] The first preset flight altitude is less than the second preset flight altitude, the first preset data acquisition period is less than the second preset data acquisition period, and the first preset positioning accuracy is greater than the second preset positioning accuracy.

[0082] For high-frequency monitoring areas, fixed-wing UAVs, flying at a lower altitude (first preset flight altitude), ensure the clarity of aerial photography details. A shorter data acquisition cycle (first preset data acquisition cycle) enables real-time dynamic risk tracking. Real-time image transmission ensures that anomalies can be detected immediately. Combined with ground displacement monitoring stations with higher positioning accuracy (first preset positioning accuracy) and embedded pressure sensors with preset acquisition frequencies, the accuracy, real-time performance, and continuity of monitoring in high-risk areas are comprehensively enhanced. For conventional monitoring areas, multi-rotor UAVs, flying at a higher altitude (second preset flight altitude), expand the scope of a single survey. A longer data acquisition cycle (second preset data acquisition cycle) reduces unnecessary repetitive work. Turning off real-time image transmission reduces resource consumption. Mobile ground mapping robots employ positioning accuracy adapted to basic monitoring (second preset positioning accuracy), meeting the monitoring needs of conventional areas while avoiding resource waste. This parameter differentiation ensures that high-frequency monitoring areas can capture minute risk signals while controlling costs in conventional monitoring areas, laying a solid foundation for subsequent accurate data acquisition and risk identification.

[0083] Step S3 specifically includes:

[0084] Surveying data is collected using surveying equipment in high-frequency and conventional monitoring areas. The surveying data includes displacement, seepage pressure, elevation changes, crack width, aerial images, crack images, and leakage point distribution. A data transmission link is established to transmit the surveying data to a cloud platform, and the cloud platform's visualization module is used to visualize the surveying data.

[0085] Building a data transmission link includes:

[0086] The edge computing nodes deployed at the water conservancy project surveying site receive surveying data, preprocess the surveying data to obtain processed data, and transmit the processed data to the cloud platform through dual-mode communication technology. The preprocessing includes outlier handling, spatiotemporal coordinate unification processing, data format standardization processing, and data compression processing.

[0087] Unified standards for edge computing preprocessing: Spatiotemporal coordinates are uniformly based on the 2000 National Geodetic Coordinate System to prevent coordinate deviations from being difficult to integrate. Outliers are processed according to clear thresholds (such as seepage pressure exceeding 3 times the normal value) to avoid subjective deletion or retention of invalid data.

[0088] Dual-mode communication adapts to the field network: satellite communication is prioritized to supplement weak signals in remote reservoirs, while terrestrial communication is the main method for urban waterways, with satellite as a backup, to ensure uninterrupted data transmission and prevent early warning delays caused by poor signal.

[0089] Outlier handling employs the 3σ principle or box plot method: First, the mean and standard deviation of similar historical data are statistically analyzed. When newly collected data exceeds the mean ± 3σ range, it is marked as an anomaly. For example, if the historical mean of seepage pressure in a certain area is 0.3 MPa and the standard deviation is 0.1 MPa, and the new data shows 1.2 MPa, it is judged as an anomaly. After manual verification confirms that it is a sensor malfunction, it is removed. For outliers with instantaneous fluctuations (such as a single elevation jump caused by airflow affecting a drone), the moving average method is used to replace it with the average value of the surrounding time.

[0090] First, a benchmark is established for unified spatiotemporal coordinate processing: the spatial coordinates of all data collected by all devices are converted into the 2000 National Geodetic Coordinate System. For example, the latitude and longitude (such as 118°E and 32°N) obtained by the UAV GPS are converted into plane rectangular coordinates through Gauss-Kruger projection. Beijing time is used uniformly in terms of time, and the timestamps of different devices are calibrated. For example, if the ground monitoring station shows the collection time as 14:05:20, while the UAV shows 14:05:18 (due to network delay), it is uniformly corrected to the intermediate value of 14:05:19 to ensure that the displacement and image data at the same time can accurately correspond.

[0091] Data format standardization and unified specifications were established: TIFF, PNG, and other image formats captured by drones were uniformly converted to JPEG, with a fixed resolution of 1920×1080. XML, binary, and other numerical formats output by sensors were converted to JSON, with field names standardized as "timestamp, location X, location Y, monitoring value". For example, seepage pressure data was standardized from "{timestamp:1620000000,pressure:0.35}" to "{collection time:2021-05-03 12:00:00,X coordinate:352000,Y coordinate:486000,seepage pressure value:0.35}", facilitating direct parsing in the cloud.

[0092] Data compression processing employs adaptive algorithms for different data types: JPEG2000 lossless compression is used for aerial images and crack images, preserving pixel details while compressing to 30%-50% of the original volume; for numerical data such as displacement and seepage pressure values, compression is achieved by removing redundant decimal points (e.g., retaining two decimal places) and using differential encoding (recording the difference between adjacent data), such as compressing "0.3215, 0.3218, 0.3220" into "0.32, +0.0003, +0.0002", reducing the amount of data transmitted without affecting the accuracy of the analysis.

[0093] By collecting various types of surveying data, such as displacement and seepage pressure, the safety status of water conservancy projects can be reflected from different dimensions, including structural deformation, hydraulic state, and appearance characteristics. This avoids monitoring blind spots caused by single data types. By preprocessing data through edge computing nodes (outlier handling, spatiotemporal coordinate unification, etc.) and combining it with dual-mode communication technology for transmission, the redundancy and transmission pressure of the original data can be significantly reduced, improving data transmission efficiency and accuracy. Meanwhile, the visualization display on the cloud platform can transform abstract data into intuitive charts or images, making it convenient for staff to quickly grasp the real-time status of the project.

[0094] Step S4 specifically includes:

[0095] The system acquires surveying data from the cloud platform and performs a first verification against the risk threshold database to determine whether a real-time risk has occurred. If a real-time risk is determined, a second verification is performed using the historical database. If a real-time risk is determined, the risk level of the real-time risk is determined according to the preset risk level determination rules. A graded early warning is triggered based on the risk level of the real-time risk, and the location of the real-time risk is marked. The risk levels include Level 1 risk, Level 2 risk, and Level 3 risk.

[0096] The pre-defined risk level assessment rules, based on the scope of impact, degree of hazard, and speed of development, classify risks into Level 1, Level 2, and Level 3: Level 1 risks affect critical structures of the project (such as the core section of a reservoir dam), potentially causing serious accidents such as dam failure and threatening lives within 24 hours. Examples include a reservoir dam's water-facing slope with an average daily displacement exceeding 5 mm, seepage pressure exceeding the threshold by 1.2 times, and the appearance of transverse cracks exceeding 10 meters. Level 2 risks affect localized areas of the project (such as non-core sections of a dam), potentially causing localized leakage, small collapses, and economic losses within 72 hours. Examples include a river bend with a daily scour depth of 0.8 meters (exceeding the threshold by 0.5 meters) and a history of small-scale collapses. Level 3 risks have a small impact area and slow development (no deterioration within one week), causing only surface damage. Examples include a dam in a regular monitoring area with a 2-meter-long, 0.5-millimeter-wide surface crack and normal displacement and seepage pressure. Quantitative indicators ensure objective assessment and provide a basis for graded early warning.

[0097] First, the cloud-based surveying data is verified against a risk threshold database to quickly screen for abnormal data exceeding the threshold and preliminarily determine whether there is a real-time risk. If the first verification determines that there is a risk, a second verification is conducted using a historical database. By comparing the data characteristics of historical normal operating conditions and risky operating conditions, misjudgments caused by accidental anomalies or environmental interference are eliminated to ensure the accuracy of risk identification. Once the second verification confirms the existence of a real-time risk, the risk is classified into Level 1, Level 2, and Level 3 risks according to preset risk level determination rules (such as combining the scope of risk impact and the degree of harm). Then, a graded warning is triggered according to the corresponding level, allowing different responsible parties to clearly understand the response priority. At the same time, the location of the real-time risk is marked, which makes it easier for staff to quickly locate the hidden danger area, buy time for subsequent re-measurement and handling, and effectively prevent the risk from escalating.

[0098] The process of verifying the survey data against a risk threshold database to determine whether a real-time risk has occurred includes:

[0099] Based on the attributes of the surveying data, the surveying data is divided into quantitative data and qualitative data.

[0100] Quantitative data includes displacement, seepage pressure, elevation changes, and crack width, while qualitative data includes aerial images, crack images, and the distribution of leakage points.

[0101] For quantitative data, a direct numerical comparison method is used to obtain the specific values ​​of the quantitative data, and then the specific values ​​are compared with the preset thresholds in the risk threshold library.

[0102] If the specific value of one of the quantitative data exceeds the preset threshold, it is determined that a risk has occurred.

[0103] If all the specific values ​​of the quantitative data are less than the preset threshold, it is determined that no risk has occurred.

[0104] For qualitative data, a feature matching comparison method is used to obtain key feature parameters of the surveying and mapping images in the qualitative data. The key feature parameters are then compared with the preset feature thresholds of the corresponding scenes in the risk threshold library.

[0105] If the key feature parameter of the survey image in one of the qualitative data exceeds the preset feature threshold, it is determined that a risk has occurred.

[0106] If all key feature parameters of the surveyed images in the qualitative data are less than the preset feature threshold, then it is determined that no risk has occurred.

[0107] This method quickly and accurately screens potential risks from surveying data, providing clear targets for subsequent secondary verification. For quantitative data (such as displacement), direct numerical comparison is used, allowing for intuitive judgment of whether a risk exceeds the limit by comparing the specific value with a preset threshold. For example, if the dam displacement exceeds the threshold by 0.5 mm, it can be quickly identified as a risk. For qualitative data (such as aerial images), feature matching comparison is used. By extracting key features from the images (such as crack length and seepage point area) and comparing them with preset feature thresholds, the problem of direct judgment of non-numerical data is solved, enabling effective risk screening of image-based data. The combination of the two methods ensures the objectivity of quantitative data judgment and improves the operability of qualitative data risk identification, significantly improving the efficiency of preliminary risk screening.

[0108] Further verification using historical databases is conducted to determine whether a real-time risk has occurred. This includes:

[0109] For quantitative data, a trend similarity verification method is used to extract the trend features of quantitative data and compare the trend features with the trend features of historical normal working conditions and historical risk working conditions in the historical database.

[0110] If the similarity between the trend characteristics of quantitative data and the trend characteristics of historical normal operating conditions is greater than or equal to the first threshold, it is determined that no risk has occurred.

[0111] If the similarity between the trend characteristics of quantitative data and the trend characteristics of historical risk conditions is greater than or equal to the first threshold, then it is determined that a risk has occurred.

[0112] For qualitative data, the feature correlation verification method is used to extract key features of the surveying images in the qualitative data, and compare the key features with the risk features of historical normal working conditions and historical risk working conditions in the historical database.

[0113] If the overlap rate between the key features of the mapping images in the qualitative data and the risk features of historical normal operating conditions is greater than or equal to the second threshold, it is determined that no risk has occurred.

[0114] If the overlap rate between the key features of the mapping images in the qualitative data and the risk features of historical risk conditions is greater than or equal to the second threshold, then it is determined that a risk has occurred.

[0115] The setting of the first and second thresholds is dynamically optimized based on the verification effect of historical data. The first threshold can be set to 80% (i.e., trend similarity ≥ 80% is considered a match), and the second threshold can be set to 70% (i.e., feature overlap rate ≥ 70% is considered a correlation). The historical database is stored in categories according to project type and risk type to ensure that the historical features of the corresponding scenario can be called during comparison. For example, the displacement trend of the reservoir dam needs to be compared with the historical data of the same type of reservoir. At the same time, when new features that do not match the historical normal and risk conditions appear, they are included in the manual review process to avoid misjudgment due to incomplete coverage of historical data. This ensures that the secondary verification can effectively filter false alarms without omitting new risks.

[0116] In-depth verification of potential risks identified in a single screening effectively eliminates misjudgments caused by accidental data anomalies or environmental interference, improving the accuracy of risk identification. For quantitative data, a trend similarity verification method is used. By comparing the current data change trend with the trend characteristics of historical normal and risk conditions (such as the growth slope of displacement over three consecutive days), it can be determined whether the data anomaly is a short-term fluctuation or a precursor to risk. For example, if the displacement of a dam exceeds the standard in a single instance but the trend is highly similar to historical normal conditions, it can be determined as non-risk. For qualitative data, a feature correlation verification method is used. By analyzing the overlap between key features of the current image (such as crack morphology and leakage point distribution) and historical features, it can distinguish between real risks and similar normal phenomena. For example, if a newly discovered crack has a high overlap rate with crack features in historical risk conditions, then a risk is confirmed.

[0117] Triggering tiered early warnings based on the real-time risk level specifically includes:

[0118] When the real-time risk is classified as Level 1, a Level 1 warning is triggered. Pop-ups and voice calls are simultaneously sent to engineering operation and maintenance personnel and local water conservancy management departments through the cloud platform, and on-site audible and visual alarms are activated.

[0119] When the real-time risk level is 2, a level 2 warning is triggered, and a pop-up window and text message are sent to the engineering operation and maintenance personnel through the cloud platform.

[0120] When the real-time risk is level three, a level three warning is triggered, and a pop-up window is pushed to the engineering operation and maintenance personnel through the cloud platform.

[0121] Based on the urgency of the risk, corresponding levels of response forces are mobilized to ensure that high-risk hazards are dealt with as quickly as possible, while avoiding excessive resource consumption caused by low-risk warnings. The combination of pop-up windows, voice calls, and on-site audible and visual alarms for Level 1 risks can synchronize the highest level of risk to maintenance personnel and local management departments in a timely manner, and quickly mobilize all parties in conjunction with on-site alarms; Level 2 risks are notified by pop-up windows and SMS reminders to ensure that the maintenance team is informed in a timely manner and arranges targeted handling; Level 3 risks are only pushed by pop-up windows to remind attention without interfering with routine work.

[0122] Step S5 specifically includes:

[0123] Based on the triggered warning level, the cloud platform automatically generates a key retest plan. The key retest plan includes the retest scope, retest resources, and retest frequency. Retests are conducted according to the key retest plan to obtain retest data. The survey data and retest data are compared to determine the risk change trend. The warning level is adjusted according to the risk change trend, and disposal suggestions are generated.

[0124] Based on the triggered warning level, the cloud platform automatically generates a key retest plan, including:

[0125] If a Level 1 warning is triggered, the retesting scope includes the real-time risk location and the surrounding area that is greater than the third threshold but less than the fourth threshold. Retesting resources will be prioritized for fixed-wing UAVs and ground displacement monitoring stations, and the retesting frequency will be set to the fifth threshold.

[0126] If a Level 2 warning is triggered, the retesting scope includes the real-time risk location and the surrounding area that is greater than the sixth threshold but less than the third threshold. Retesting resources will be prioritized for multi-rotor drones and mobile ground mapping robots, and the retesting frequency will be set to the seventh threshold.

[0127] If a Level 3 warning is triggered, the retesting scope will include real-time risk locations, and mobile ground mapping robots will be prioritized for retesting resources. The retesting frequency will be set to the eighth threshold.

[0128] Adjusting early warning levels based on risk trends includes:

[0129] If the risk change trend matches the preset safety change trend, the warning level will be lowered by one level.

[0130] If the risk change trend does not conform to the preset safety change trend, the warning level will be raised by one level, and handling suggestions will be pushed to the cloud platform.

[0131] Based on the differentiated allocation of retesting resources according to the warning level (high-precision equipment such as fixed-wing UAVs for Level 1 warnings, and mobile robots for Level 3 warnings), retesting scope (Level 1 covers the risk point and a wider surrounding area), and retesting frequency (Level 1 retesting is more intensive), we ensure that high-risk areas receive stricter tracking and monitoring, while low-risk areas complete verification efficiently. By comparing the retesting data with the original surveying data, we determine the risk trend. If the trend is safe, the warning level is lowered; otherwise, it is raised and disposal suggestions are pushed out. This avoids excessive warnings that waste resources and prevents escalating risks from being ignored. It forms a complete response mechanism from warning to dynamic adjustment, improves the timeliness and accuracy of risk management, and ensures the scientific nature of safe handling of water conservancy projects.

[0132] This invention divides high-frequency monitoring zones and routine monitoring zones according to the inherent risk levels of different areas of water conservancy projects, and deploys different types of surveying equipment and sets differentiated parameters accordingly. This ensures that high-risk areas are monitored with higher accuracy and shorter cycles, effectively avoiding missed early warning opportunities, while also preventing the waste of monitoring resources in routine areas, thus achieving a rational allocation of monitoring resources. By verifying the surveying data against a risk threshold database in the first instance, and then conducting a second verification using a historical database, it effectively eliminates misjudgments and omissions caused by accidental data anomalies or environmental interference, significantly improving the accuracy of risk assessment and accurately distinguishing between real risks and data deviations. With the help of a comprehensive risk classification and early warning mechanism, coupled with a dynamic retesting plan automatically generated according to the early warning level, the scope, resources and frequency of retesting can be flexibly adjusted. Furthermore, a differentiated early warning push method is adopted according to the risk level to ensure that different levels of risks can be received and dealt with in a timely manner, thereby improving the efficiency of risk response. By preprocessing the surveying data through edge computing nodes before transmitting it to the cloud, the network bandwidth pressure caused by direct transmission of raw data is greatly reduced, improving data processing efficiency and the real-time nature of risk early warning. At the same time, the visualization module of the cloud platform can intuitively present the surveying data and risk distribution, helping operation and maintenance personnel to quickly and clearly grasp the safety status of the project.

[0133] 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 implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. 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.

[0134] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An intelligent mapping method for geographic information of water conservancy projects, characterized in that, Includes the following steps: Step S1: Identify the core risk points of the water conservancy project and divide the monitoring area according to the inherent risk level of the core risk points; Step S2: Deploy surveying equipment and set parameters for the divided monitoring area; Step S3: Use the surveying equipment to collect surveying data and establish a data transmission link to transmit the surveying data to the cloud platform; Step S4: The survey data is verified against the risk threshold database once, and then verified again in conjunction with the historical database. When a real-time risk is triggered, a graded warning is triggered according to the risk level of the real-time risk. Step S5: Based on the triggered warning level, generate a key retest plan, conduct the retest, and adjust the warning level.

2. The intelligent mapping method for geographic information of water conservancy projects as described in claim 1, characterized in that: Step S1 specifically includes: The project utilizes water conservancy engineering designers and operation and maintenance teams to conduct on-site surveys, and combines project types with historical accident data to identify core risk points. The project types include reservoirs, dams, and rivers, and the historical accident data includes historical accident data for reservoirs, dams, and rivers. The mapping area of ​​the core risk point is delineated according to the inherent risk level of the core risk point. The inherent risk level includes high-risk points, medium-risk points and low-risk points. The mapping area of ​​high-risk points is designated as a high-frequency monitoring area, and the mapping areas of medium-risk points and low-risk points are designated as regular monitoring areas.

3. The intelligent mapping method for geographic information of water conservancy projects as described in claim 2, characterized in that: Step S2 specifically includes: Surveying equipment is deployed in the high-frequency monitoring area and the conventional monitoring area respectively, and the parameters of the surveying equipment in the high-frequency monitoring area and the conventional monitoring area are set respectively; The mapping equipment in the high-frequency monitoring area includes fixed-wing UAVs, ground displacement monitoring stations, and embedded pressure sensors; The surveying equipment in the conventional monitoring area includes multi-rotor drones and mobile ground surveying robots.

4. The intelligent mapping method for geographic information of water conservancy projects as described in claim 3, characterized in that: The specific steps for setting parameters for the mapping equipment in the high-frequency monitoring area and the conventional monitoring area include: For high-frequency monitoring areas: set the flight altitude of the fixed-wing UAV to the first preset flight altitude, adopt the first preset data acquisition cycle and enable real-time image transmission, adjust the ground displacement monitoring station to the first preset positioning accuracy, and set the embedded pressure sensor to the preset acquisition frequency. For routine monitoring areas: set the flight altitude of the multi-rotor UAV to the second preset flight altitude, adopt the second preset data acquisition cycle and turn off real-time image transmission, and adjust the mobile ground mapping robot to the second preset positioning accuracy; The first preset flight altitude is less than the second preset flight altitude, the first preset data acquisition period is less than the second preset data acquisition period, and the first preset positioning accuracy is greater than the second preset positioning accuracy.

5. The intelligent mapping method for geographic information of water conservancy projects as described in claim 4, characterized in that: Step S3 specifically includes: Surveying data is collected using surveying equipment in the high-frequency monitoring area and the conventional monitoring area. The surveying data includes displacement, seepage pressure, elevation change, crack width, aerial imagery, crack images, and distribution of leakage points. A data transmission link is constructed to transmit the surveying data to a cloud platform, and the visualization module of the cloud platform is used to visualize the surveying data. The construction of the data transmission link includes: The surveying data is received by edge computing nodes deployed at the water conservancy project surveying site, and the surveying data is preprocessed to obtain processed data. The processed data is then transmitted to the cloud platform through dual-mode communication technology. The preprocessing includes outlier handling, spatiotemporal coordinate unification, data format standardization, and data compression.

6. The intelligent mapping method for geographic information of water conservancy projects as described in claim 5, characterized in that: Step S4 specifically includes: The system acquires surveying data from the cloud platform and performs a first verification against the risk threshold database to determine whether a real-time risk has occurred. If a real-time risk is determined, a second verification is performed using the historical database to determine whether a real-time risk has occurred. If a real-time risk is determined, the risk level of the real-time risk is determined according to the preset risk level determination rules. A graded early warning is triggered based on the risk level of the real-time risk, and the location of the real-time risk is marked. The risk levels include Level 1 risk, Level 2 risk, and Level 3 risk.

7. The intelligent mapping method for geographic information of water conservancy projects as described in claim 6, characterized in that: The step of verifying the mapping data against the risk threshold database to determine whether a real-time risk has occurred specifically includes: Based on the attributes of the surveying data, the surveying data is divided into quantitative data and qualitative data; The quantitative data includes displacement, seepage pressure, elevation change, and crack width; the qualitative data includes aerial images, crack images, and distribution of leakage points. For quantitative data, a direct numerical comparison method is used to obtain the specific value of the quantitative data, and the specific value is compared with a preset threshold in the risk threshold library. If the specific value of one of the quantitative data exceeds the preset threshold, it is determined that a risk has occurred; If all the specific values ​​of the quantitative data are less than the preset threshold, it is determined that no risk has occurred; For qualitative data, a feature matching comparison method is used to obtain key feature parameters of the surveying and mapping images in the qualitative data, and the key feature parameters are compared with preset feature thresholds of the corresponding scenes of the surveying and mapping images in the risk threshold library. If a key feature parameter of a mapping image in one of the qualitative data exceeds a preset feature threshold, it is determined that a risk has occurred. If all key feature parameters of the surveyed images in the qualitative data are less than the preset feature threshold, then it is determined that no risk has occurred.

8. The intelligent mapping method for geographic information of water conservancy projects as described in claim 7, characterized in that: The further verification using historical databases to determine whether a real-time risk has occurred specifically includes: For quantitative data, a trend similarity verification method is used to extract the trend features of the quantitative data, and the trend features are compared with the trend features of historical normal working conditions and historical risk working conditions in the historical database. If the similarity between the trend characteristics of the quantitative data and the trend characteristics of historical normal operating conditions is greater than or equal to the first threshold, it is determined that no risk has occurred. If the similarity between the trend characteristics of the quantitative data and the trend characteristics of historical risk conditions is greater than or equal to the first threshold, then it is determined that a risk has occurred. For qualitative data, a feature correlation verification method is used to extract key features of the surveying images in the qualitative data, and these key features are compared with the risk features of historical normal working conditions and historical risk working conditions in the historical database. If the overlap rate between the key features of the mapping images in the qualitative data and the risk features of historical normal operating conditions is greater than or equal to the second threshold, it is determined that no risk has occurred. If the overlap rate between the key features of the mapping images in the qualitative data and the risk features of historical risk conditions is greater than or equal to the second threshold, then it is determined that a risk has occurred.

9. The intelligent mapping method for geographic information of water conservancy projects as described in claim 8, characterized in that: The specific provisions of triggering a graded early warning based on the risk level of the real-time risk include: When the real-time risk is classified as Level 1 risk, a Level 1 warning is triggered. Pop-ups and voice calls are simultaneously pushed to engineering operation and maintenance personnel and local water conservancy management departments through the cloud platform, and on-site audible and visual alarms are activated. When the real-time risk is a level 2 risk, a level 2 warning is triggered, and a pop-up window and a text message are sent to the engineering operation and maintenance personnel through the cloud platform. When the real-time risk is classified as Level 3, a Level 3 warning is triggered, and a pop-up window is pushed to the engineering and maintenance personnel through the cloud platform.

10. The intelligent mapping method for geographic information of water conservancy projects as described in claim 9, characterized in that: Step S5 specifically includes: Based on the triggered warning level, the cloud platform automatically generates a key retesting plan. The key retesting plan includes the retesting scope, retesting resources, and retesting frequency. Retesting is carried out according to the key retesting plan, retesting data is obtained, and the surveying data and retesting data are compared to determine the risk change trend. The warning level is adjusted according to the risk change trend, and disposal suggestions are generated. Based on the triggered warning level, the cloud platform automatically generates a key retest plan, including: If a Level 1 warning is triggered, the retesting scope includes the real-time risk location and the surrounding area that is greater than the third threshold but less than the fourth threshold. Retesting resources are prioritized for fixed-wing UAVs and ground displacement monitoring stations, and the retesting frequency is set to the fifth threshold. If a Level 2 warning is triggered, the retesting scope includes the real-time risk location and the surrounding area that is greater than the sixth threshold but less than the third threshold. Retesting resources will be prioritized for multi-rotor drones and mobile ground mapping robots, and the retesting frequency will be set to the seventh threshold. If a Level 3 warning is triggered, the retesting scope includes real-time risk locations, and mobile ground mapping robots will be prioritized for retesting resources. The retesting frequency will be set to the eighth threshold. The adjustment of the early warning level based on the risk change trend includes: If the risk change trend matches the preset safety change trend, then the warning level will be lowered by one level; If the risk change trend does not conform to the preset safety change trend, the warning level will be raised by one level, and handling suggestions will be pushed to the cloud platform.

Images