River safety monitoring method and system

Abstract

The present application provides a river safety monitoring method and system, which relate to the technical field of river flow monitoring. The method includes: obtaining riverbed elevation distribution data of a target monitoring section area, dividing the target monitoring section area into multiple flow measurement partitions, obtaining section parameters of each flow measurement partition, and setting a radar flow velocity sensor for collecting river water velocity data in each flow measurement partition, and setting a radar water level sensor for measuring first river water level data in any flow measurement partition; constructing a hydraulic dynamics model, inputting river water velocity data to obtain average flow velocity data of the corresponding flow measurement partition; calculating second river water level data of the remaining flow measurement partitions based on the first river water level data and the riverbed elevation distribution data, calculating instantaneous flow and cumulative flow based on the section parameters, the first river water level data, the second river water level data and the average flow velocity data, and issuing an alarm signal when the cumulative flow exceeds a preset flow threshold. The present application can improve the accuracy of the monitoring results.

Description

Technical Field

[0001] The present application relates to the technical field of river flow monitoring, and in particular to a river safety monitoring method and system. Background Art

[0002] As a vital carrier of water resources, rivers play an irreplaceable and crucial role in flood control, irrigation, navigation, water supply, and ecological conservation. The safety of rivers is directly related to the safety of life and property, economic development, and ecological stability in surrounding areas.

[0003] Traditional river monitoring methods often rely on manual measurements or relatively simple equipment, such as buoys. Traditional river monitoring methods present numerous problems during data collection. Manual measurements are not only inefficient but also significantly affected by human factors, making it difficult to ensure data accuracy and consistency. Simple equipment has a limited monitoring range, typically only capturing flow data at a specific location or within a limited area of the river. Comprehensive and detailed monitoring of the entire monitoring section is impossible, resulting in a one-sided understanding of river flow conditions and difficulty accurately reflecting the overall flow dynamics of the river. Summary of the Invention

[0004] In order to improve the accuracy of river safety monitoring results, the present application provides a river safety monitoring method and system.

[0005] In the first aspect, the present application provides a river safety monitoring method, which adopts the following technical solutions:

[0006] A river safety monitoring method comprises the following steps:

[0007] Obtain riverbed elevation distribution data of a target monitoring section area, divide the target monitoring section area into a plurality of flow measurement subareas, obtain section parameters of each flow measurement subarea, each flow measurement subarea is provided with a radar flow velocity sensor for collecting river flow velocity data, and a radar water level sensor for measuring first river water level data is provided in any flow measurement subarea;

[0008] Constructing a hydraulic dynamics model, inputting the river flow velocity data of each flow measurement zone into the hydraulic dynamics model, and obtaining the average flow velocity data of each flow measurement zone;

[0009] The second river water level data of the remaining flow measurement sections is calculated based on the first river water level data and the riverbed elevation distribution data; the instantaneous flow is calculated based on the section parameters, the first river water level data, the second river water level data and the average flow velocity data; the cumulative flow is calculated based on the instantaneous flow, and an alarm signal is issued when the cumulative flow exceeds the preset flow threshold.

[0010] This application obtains the riverbed elevation distribution data of the target monitoring section area and divides it into multiple flow measurement zones. Subsequently, this application constructs a hydraulic dynamics model and inputs the river water velocity data into the model to obtain the average velocity data of the corresponding flow measurement zones. The hydraulic dynamics model can comprehensively consider the various physical characteristics of the river, process and analyze the velocity data collected by the sensor, and obtain a more representative average velocity. This can reduce the problem of large deviations in flow monitoring data caused by uneven flow velocity and irregular cross-sections in the river cross section, making the analysis of water flow conditions at different locations in the river more accurate.

[0011] This application also installs a radar water level sensor in any flow measurement sub-area to measure the first river water level data, and then calculates the second river water level data for the remaining flow measurement sub-areas based on the first river water level data and the riverbed elevation distribution data. Through the above technical solution, this application can infer the water level conditions of flow measurement sub-areas that are not equipped with water level sensors, expand the monitoring range of water level data, and make the water level information of the entire monitoring area more complete. This solves the problem of being unable to fully monitor the water level due to the limited number of sensors, and achieves comprehensive coverage of the water level data in the monitoring area, providing complete water level parameters for flow calculation.

[0012] Subsequently, the present application calculates the instantaneous flow rate based on the section parameters, the first river water level data, the second river water level data and the average flow rate data, and calculates the cumulative flow rate based on the instantaneous flow rate. By adopting the above scheme, the present application can accurately grasp the flow conditions of the river in different time periods. When the cumulative flow rate exceeds the preset flow threshold, an alarm signal is issued to promptly remind relevant personnel to take measures. The present application improves the traditional method of measuring the water velocity at a certain position in the target monitoring section area to divide the target monitoring section area into multiple flow measurement partitions, and measures the river water velocity data of each flow measurement partition, obtains the average flow velocity data based on these river water flow rate data, and finally calculates the cumulative flow rate based on the above data, so that the calculation results are more accurate, which improves the accuracy of river safety monitoring to a certain extent.

[0013] Optionally, the method further includes:

[0014] Based on all river water velocity data, the surface velocity data of the target monitoring section area is calculated using the interpolation method. The vertical velocity data of each position in the target monitoring section area is calculated according to the preset velocity distribution calculation model. A vertical velocity data change curve is drawn based on the vertical velocity data. The horizontal axis of the vertical velocity data change curve is the shortest distance to any river bank, and the vertical axis is the vertical velocity data.

[0015] Calculate the first-order derivative of the vertical flow velocity data change curve with respect to the distance, discretize the first-order derivative to obtain a discretization result, normalize the discretization result to obtain first data, calculate a vertical average value based on the first data and the vertical flow velocity data, and update the vertical average value to the river water flow velocity data.

[0016] This application is based on all river water velocity data and uses the interpolation method to calculate the surface velocity data of the target monitoring section area. The interpolation method can use the known river water velocity data to reasonably estimate the surface velocity at other locations in the monitoring area, so as to have a more comprehensive understanding of the surface velocity distribution of the entire section area, help to more accurately grasp the water flow conditions on the river surface, and improve the accuracy of overall monitoring and analysis.

[0017] Subsequently, the present application calculated vertical velocity data for each location within the target monitoring section based on the velocity distribution calculation model. This velocity distribution calculation model comprehensively considers the physical characteristics of the river and the principles of hydrodynamics, and is capable of calculating vertical velocity at different locations. By adopting this approach, the present application was able to expand a single surface velocity to multiple vertical levels, more realistically reflecting the internal flow structure of the river and contributing to a deeper understanding of the river's vertical velocity distribution characteristics.

[0018] Subsequently, the present application plots a vertical velocity data change curve based on the vertical velocity data and calculates the first-order derivative of the vertical velocity data change curve. The first-order derivative can reflect the rate of change of the vertical velocity. Subsequently, the present application discretizes the first-order derivative value, converting the continuous derivative value into discrete data points to obtain the first-order derivative value, and then normalizes the first-order derivative value to obtain the first data, thereby converting the discretization result to the range of 0-1. Finally, the present application calculates the vertical average value based on the first data and the vertical velocity data. The first data reflects the characteristics of the vertical velocity change. Calculating the vertical average value in combination with the vertical velocity data can comprehensively consider the size and change of the velocity itself, so that the calculated vertical average value can more accurately represent the overall level of the entire vertical velocity, providing a more accurate and comprehensive basis for updating river water velocity data. By adopting the above scheme, the present application can optimize and update river water velocity data so that the data is more consistent with the actual flow velocity characteristics of the river.

[0019] Optionally, the method further includes:

[0020] At least one reference point, multiple working reference points, and multiple observation points are set on the riverbanks on both sides of the target monitoring section area, a three-dimensional coordinate system is constructed with the location of the reference point as the coordinate origin, and the original three-dimensional coordinates of each working reference point and each observation point are determined;

[0021] Select at least one reference point and at least one working reference point to construct a monitoring reference network, retest the three-dimensional coordinates of each working reference point after a preset time period, obtain the remeasured three-dimensional coordinates of each working reference point, and obtain the horizontal displacement change and vertical displacement change of each working reference point based on the original three-dimensional coordinates and the remeasured three-dimensional coordinates of each working reference point;

[0022] The three-dimensional displacement of each working base point is calculated based on the displacement change in the horizontal direction and the displacement change in the vertical direction of each working base point, and the three-dimensional displacement of each working base point is judged in turn to see whether it is greater than the preset three-dimensional displacement threshold. If so, a working base point alarm signal is issued; if not, no processing is performed.

[0023] This application sets benchmark points, working base points and observation points on both sides of the river banks in the target monitoring section area, and constructs a three-dimensional coordinate system with the benchmark points as the coordinate origin to determine the original three-dimensional coordinates of each point, thereby achieving precise spatial positioning of key points in the monitoring area. It can clearly describe the positional relationship of each point in three-dimensional space, which helps to fully understand the spatial layout of the monitoring area.

[0024] Subsequently, the present application constructs a monitoring benchmark network with benchmark points and at least one working benchmark point, forming a relatively stable monitoring system. The benchmark points serve as the reference base for the entire monitoring system, and the working benchmark points serve as key monitoring nodes. This combination enables long-term, stable monitoring of displacement changes in the monitoring area. The three-dimensional coordinates of each working benchmark point are retested after a preset period of time to obtain remeasured three-dimensional coordinates. By comparing them with the original three-dimensional coordinates, the horizontal and vertical displacement changes of the working benchmark point can be captured. This regular remeasurement method can promptly detect small displacements of the riverbed, provide dynamic data for assessing riverbed stability, and help identify potential safety hazards in advance. Subsequently, the present application calculates the three-dimensional displacement based on the horizontal and vertical displacement changes of each working benchmark point, and sequentially determines whether the three-dimensional displacement of each working benchmark point exceeds a preset three-dimensional displacement threshold. Based on the judgment result, an alarm signal is issued or no action is taken. The present application uses a threshold-based alarm mechanism to promptly identify working benchmark points with abnormal displacement, issue an alarm signal in advance, and remind relevant personnel to take measures to address them, thereby ensuring the safety of the monitoring area and the surrounding environment.

[0025] Optionally, when the three-dimensional displacement of the working base point is not greater than a preset three-dimensional displacement threshold, the method further includes:

[0026] Obtaining remeasured three-dimensional coordinates of the observation point, wherein the remeasured three-dimensional coordinates of the observation point and the remeasured three-dimensional coordinates of the working base point have the same timestamp;

[0027] Based on the horizontal and vertical displacement changes of each working base point, the re-measured three-dimensional coordinates of the observation point are updated to obtain the absolute three-dimensional coordinates of the observation point. The three-dimensional displacement of each observation point is calculated according to the original three-dimensional coordinates and the absolute three-dimensional coordinates of the observation point. It is judged in turn whether the three-dimensional displacement of each observation point is greater than the preset three-dimensional displacement threshold. If so, an observation point alarm signal is issued; if not, no processing is performed.

[0028] When the 3D displacement of the working base point is no greater than a preset 3D displacement threshold, the present application obtains the remeasured 3D coordinates of the observation point, and the remeasured 3D coordinates of the observation point and the remeasured 3D coordinates of the working base point have the same timestamp. This method of acquiring data with the same timestamp ensures that the working base point and observation point data are collected at the same time, eliminating measurement errors caused by time differences. This makes subsequent displacement calculations and analysis based on this data more accurate and reliable, and can truly reflect the positional relationship between the working base point and the observation point at the same point in time.

[0029] Subsequently, the present application updates the remeasured three-dimensional coordinates of the observation point based on the horizontal and vertical displacement changes of each working base point to obtain the absolute three-dimensional coordinates of the observation point. By updating the observation point coordinates based on the displacement changes of the working base point, the relative coordinates of the observation point can be converted into absolute coordinates, so that the position information of the observation point can be accurately represented in a unified coordinate system, reducing the impact of the working base point displacement on the description of the observation point position, and making the position information of the observation point more accurate.

[0030] Subsequently, the present application calculates the three-dimensional displacement of each observation point based on the original three-dimensional coordinates and absolute three-dimensional coordinates of the observation point. Through the above scheme, the present application can accurately calculate the displacement of the observation point in three-dimensional space, thereby being able to fully reflect the displacement changes of the observation point in the horizontal and vertical directions. It then determines in turn whether the three-dimensional displacement of each observation point is greater than the preset three-dimensional displacement threshold, and takes corresponding measures based on the judgment results. The present application realizes the real-time monitoring and early warning function of the displacement of the observation point, which can promptly detect potential safety hazards.

[0031] Optionally, when the three-dimensional displacement of the observation point is not greater than a preset three-dimensional displacement threshold, the method further includes:

[0032] Construct a statistical analysis model, input the remeasured three-dimensional coordinates of each working base point and the three-dimensional displacement of each working base point into the statistical analysis model in sequence, and obtain the predicted value of the three-dimensional displacement of each working base point at the next moment; input the absolute three-dimensional coordinates of each observation point and the three-dimensional displacement of each observation point into the statistical analysis model in sequence, and obtain the predicted value of the three-dimensional displacement of each observation point at the next moment;

[0033] Determine whether the predicted value of the three-dimensional displacement of each working base point at the next moment and the predicted value of the three-dimensional displacement of each observation point at the next moment are consistent with expectations. If so, no processing is performed; if not, an early warning signal is output.

[0034] This application constructs a statistical analysis model and inputs the remeasured three-dimensional coordinates and three-dimensional displacement of the working base point or observation point into the model to obtain the predicted value of the three-dimensional displacement of each working base point or each observation point at the next moment. The above scheme mines the potential laws and trends in the data through the statistical analysis model. The statistical analysis model can learn the displacement change patterns of different base points and observation points under different coordinates and displacement conditions.

[0035] Subsequently, this application predicts the three-dimensional displacement of the working base point and the observation point at the next moment respectively, and judges whether the predicted values of the three-dimensional displacement of each working base point and the observation point at the next moment are in line with expectations. It can quickly identify those displacement changes that deviate from the normal range, and realize a preliminary estimate of future displacement changes. Compared with relying solely on real-time monitoring data, the introduction of predicted values can discover potential abnormal displacement trends in advance.

[0036] Optionally, the method further includes:

[0037] Obtain the historical three-dimensional displacement of the observation point and the real-time three-dimensional displacement at the current moment, arrange the historical three-dimensional displacements in chronological order of acquisition time to obtain a historical three-dimensional displacement sequence, and draw a historical three-dimensional displacement curve based on the historical three-dimensional displacement sequence;

[0038] Establish a safety monitoring model for fitting the historical three-dimensional displacement curve, and obtain the theoretical three-dimensional displacement at the current moment based on the safety monitoring model;

[0039] Determine whether the difference between the real-time three-dimensional displacement and the theoretical three-dimensional displacement is greater than a preset difference threshold. If so, output a deviation alarm signal; if not, do nothing.

[0040] The present application obtains the historical three-dimensional displacement of the observation point and arranges it in chronological order of acquisition time to form a historical three-dimensional displacement sequence. Subsequently, the present application draws a historical three-dimensional displacement curve based on the historical three-dimensional displacement sequence, and establishes a safety monitoring model for fitting the historical three-dimensional displacement curve. By fitting, the present application can find a mathematical expression to describe the characteristics of the historical displacement curve, so as to better understand the law of displacement change of the observation point. By establishing a safety monitoring model for fitting the historical three-dimensional displacement curve, the present application can predict the theoretical three-dimensional displacement at the current moment based on historical data. The safety monitoring model provides a theoretical three-dimensional displacement for the current moment by learning the law of historical displacement changes. Subsequently, the present application promptly discovers abnormalities in the displacement of the observation point by judging whether the difference between the real-time three-dimensional displacement and the theoretical three-dimensional displacement is greater than a preset threshold.

[0041] Optionally, the method also includes: using a laser radar carried by a drone to obtain three-dimensional point cloud data of the river channel, constructing a three-dimensional point cloud model of the river channel based on the three-dimensional point cloud data, formulating an initial inspection route of the drone according to the three-dimensional point cloud model and the inspection target and using a path planning algorithm, controlling the drone to inspect the river channel according to the initial inspection route, and collecting meteorological data and water quality data during the inspection process, and uploading the meteorological data and water quality data to the cloud.

[0042] This application uses a laser radar mounted on a drone to obtain three-dimensional point cloud data of the river. The laser radar can quickly and accurately obtain three-dimensional spatial information of the river surface. A three-dimensional point cloud model of the river is then constructed based on the three-dimensional point cloud data, converting the abstract river data into an intuitive three-dimensional visualization model. The three-dimensional point cloud model can clearly display the shape, structure, and spatial relationship of different areas of the river, allowing relevant personnel to more intuitively observe and analyze the characteristics of the river. Subsequently, this application uses a path planning algorithm to formulate the initial inspection route of the drone based on the three-dimensional point cloud model and inspection targets. The three-dimensional point cloud model provides detailed terrain information of the river. Combined with the inspection targets (such as specific areas, key facilities, etc.), the path planning algorithm can comprehensively consider factors such as flight distance, obstacle avoidance, and task priority to formulate a reasonable and efficient inspection route, enabling the drone to comprehensively and accurately cover the inspection area, improving inspection efficiency and quality.

[0043] Subsequently, the drone is controlled to patrol the river along the initial inspection route. This application enables dynamic monitoring of the river during real-time flight. The drone can quickly reach the designated area and obtain the latest river information. This allows for rapid response to emergencies (such as floods and pollution incidents), providing timely data support for emergency response. During the inspection process, the drone's onboard sensors collect meteorological and water quality data, enabling comprehensive monitoring of the river environment. The collection of meteorological data (such as wind speed, direction, temperature, and humidity) and water quality data (such as pH, dissolved oxygen, and turbidity) provides a comprehensive understanding of the river's ecological and environmental status, providing a comprehensive data basis for river management and water resource protection.

[0044] Optionally, the method also includes: using sensors carried by the unmanned boat to perform perception modeling on the water surface and boundary environment of the water area to obtain a river channel model, marking the water area contour and the location of obstacles in the river channel model to obtain the marked river channel model, dividing the river channel into multiple sub-areas based on the marked river channel model, and performing water area cleaning in each sub-area respectively.

[0045] This application uses sensors aboard unmanned vessels to model the water surface and its boundaries. These sensors can accurately and in real time capture a wide range of water information, such as water depth, current velocity, surface width, and topography. Compared to traditional manual measurement methods, unmanned vessel sensors offer greater efficiency and accuracy, covering a wider area of water, and providing a reliable data foundation for building accurate river channel models.

[0046] Subsequently, this application obtains a river model based on the sensory data. This model intuitively displays the overall shape and structure of the waterway. The application then marks the waterway outline and the locations of obstacles in the river model, enriching and clarifying the information in the river model. Marking the waterway outline clearly defines the area of the waterway, while marking the locations of obstacles helps identify potential hazardous areas or factors that affect navigation, cleaning, and other operations. Based on the marked river model, this application then divides the river into multiple sub-areas, breaking down the originally complex river cleaning task into multiple relatively independent small tasks. This task decomposition makes cleaning operations more organized and efficient, facilitates the arrangement of cleaning equipment and personnel, and improves operational flexibility and operability. By performing water cleaning in each sub-area separately, this application can adopt appropriate cleaning methods and equipment based on the specific conditions of each sub-area (such as the degree of pollution and the distribution of obstacles), achieving precise cleaning. This targeted cleaning approach can more effectively remove pollutants from the waterway, improve cleaning results, and reduce resource waste.

[0047] Secondly, this application provides a river safety monitoring system, which adopts the following technical solutions:

[0048] A river safety monitoring system, comprising: a processor and a memory,

[0049] The memory stores program code;

[0050] When the processor calls the program code in the memory, the steps of the method described in the first aspect are executed.

[0051] In summary, this application includes at least one of the following beneficial technical effects:

[0052] 1. This application improves the traditional method of measuring water velocity at a certain location in the target monitoring section area to dividing the target monitoring section area into multiple flow measurement partitions, and measuring the river water velocity data of each flow measurement partition, obtaining the average flow velocity data based on these river water velocity data, and finally calculating the cumulative flow based on the above data, making the calculation results more accurate and improving the accuracy of river safety monitoring to a certain extent.

[0053] 2. This application accurately calculates the displacement of observation points in three-dimensional space, thereby comprehensively reflecting the horizontal and vertical displacement changes of the observation points. It then determines whether the 3D displacement of each observation point exceeds a preset 3D displacement threshold and takes appropriate measures based on the judgment results. This application implements real-time monitoring and early warning of observation point displacement, enabling timely identification of potential safety hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart from S11 data collection to S13 flow calculation and alarm in Example 1 of the present application;

[0055] Figure 2 This is a flow chart from S11 data collection to S14 data update in Example 1 of the present application;

[0056] Figure 3 This is a flowchart of Example 2 of the present application. DETAILED DESCRIPTION

[0057] The following combination Figures 1 to 3 This application is described in further detail.

[0058] Example 1: This example discloses a river safety monitoring method, referring to Figure 1The method includes: S11 data acquisition, S12 modeling, S13 flow calculation and alarm. In this embodiment, the riverbed elevation distribution data of the target monitoring section area is first obtained, and the target monitoring section area is divided into multiple flow measurement partitions, and the cross-sectional parameters of each flow measurement partition are obtained. A radar flow velocity sensor for collecting river water velocity data is set in each flow measurement partition, and a radar water level sensor for measuring first river water level data is set in any flow measurement partition; then a hydraulic dynamics model is constructed, and the river water velocity data is input to obtain the average flow velocity data of the corresponding flow measurement partition; then, the second river water level data of the remaining flow measurement partitions is calculated based on the first river water level data and the riverbed elevation distribution data, and then the instantaneous flow and the cumulative flow are calculated based on the cross-sectional parameters, the first river water level data, the second river water level data and the average flow velocity data. Finally, an alarm signal is issued when the cumulative flow exceeds a preset flow threshold. The execution process of each step of this embodiment is as follows:

[0059] S11 data acquisition: Riverbed elevation distribution data collection is carried out in the target monitoring section area. The riverbed elevation distribution data includes the riverbed elevation data at each location in the target monitoring section area. The riverbed elevation data collection method is as follows: a multi-beam echo sounder transmits multiple sound beams to the bottom of the water and receives the reflected sound wave signals, thereby quickly and accurately obtaining the elevation data of different positions on the bottom of the water, and then constructing the three-dimensional distribution data of the riverbed elevation.

[0060] In other embodiments, the elevation data of each point on the riverbed in the target monitoring section area may also be measured by a single-beam echo sounder.

[0061] The target monitoring section area is divided into multiple flow measurement zones. For example, if a river channel has different terrain features such as shallows or deep troughs, these zones can be divided according to these features. Alternatively, areas with similar flow characteristics can be divided into the same flow measurement zone based on changes in water velocity and direction. In this embodiment, the river confluence is used as the target monitoring section area and is divided into multiple flow measurement zones.

[0062] For each designated flow measurement zone, detailed cross-sectional parameters are obtained. These parameters include cross-sectional shape and area. Cross-sectional shape can be determined through field measurements combined with riverbed elevation data. For example, a drone equipped with a high-definition camera can be used to photograph the riverbed cross section. Combined with the riverbed elevation data, the cross-sectional contour information can be extracted to determine the shape of each flow measurement zone. Cross-sectional area can be calculated by measuring the coordinates of each point on the cross section and using methods such as mathematical integration.

[0063] Radar flow sensors are installed in each flow measurement zone to collect river flow data. Radar flow sensors operate based on the Doppler effect, emitting electromagnetic waves onto the water surface and receiving the reflected signals. The sensors calculate river flow data by analyzing the frequency changes of the reflected signals.

[0064] At the same time, a radar water level sensor is installed in a selected flow measurement zone to measure the first river water level data. The radar water level sensor also uses electromagnetic wave technology to calculate the first river water level data by measuring the time interval between the transmission and reception of electromagnetic waves and combining it with the propagation speed of electromagnetic waves.

[0065] S12 modeling, hydraulic dynamics model is an important means to simulate the movement of river water flow and obtain more accurate water flow information.

[0066] The hydraulic dynamics model of this embodiment is based on the basic principles of fluid mechanics. It takes into account various factors such as the topography, roughness, and water flow boundary conditions of the river channel. It uses mathematical equations to describe the movement of water in the river channel. The river water velocity data is input into the hydraulic dynamics model to obtain the average flow velocity data of the flow measurement partition corresponding to the river water velocity data.

[0067] The calculation model of the hydrodynamic model of this embodiment is as follows:

[0068] ;

[0069] in, is the river flow velocity data of the i-th flow measurement area, is the average flow velocity data for the i-th flow measurement zone. a and b are empirical coefficients affected by channel morphology (rectangular, trapezoidal, parabolic, etc.), flow conditions (laminar, turbulent), and riverbed roughness. The values of a and b are determined by conducting extensive flow velocity measurements in different channels and employing regression analysis and other methods to determine appropriate coefficient values for a specific channel. The values of a and b both range from 1.0 to 1.5. In this example, after fitting, the values of a and b were determined to be 1.23 and 1.01, respectively.

[0070] In other embodiments, the hydrodynamic model may also adopt the following calculation model:

[0071] ;

[0072] ;

[0073] ;

[0074] in, The distance between the i-th flow measurement zone and the riverbed The flow velocity at the point, K is the Karman constant, which is 0.41. is the riverbed roughness height, is the river water height of the i-th flow measurement area, and its value is equal to the first river water level data or the second river water level data; is the river flow velocity data of the i-th flow measurement area.

[0075] In other embodiments, if the accuracy of the calculation result is not required to be high, the river water velocity data of each flow measurement partition may be used instead of the average flow velocity data of each flow measurement partition to perform calculations in subsequent steps.

[0076] S13 Flow Calculation and Alarm: Based on the first river water level data measured in a particular flow measurement sub-area, combined with previously acquired riverbed elevation distribution data, the second river water level data for the remaining flow measurement sub-areas is calculated. Since the flow in a river channel is hydraulically connected to each flow measurement sub-area during its flow, the water level will vary with changes in the riverbed topography and flow conditions. By combining the known first water level data and riverbed elevation distribution data (which includes the riverbed elevation data for each flow measurement sub-area), the water level data for the remaining flow measurement sub-areas can be inferred.

[0077] ;

[0078] in, The river water level data of the j-th flow measurement area, that is, the j-th second river water level data; is the riverbed elevation data of the jth flow measurement area; is the riverbed elevation data of the i-th flow measurement area; is the river water level data of the i-th flow measurement area, that is, the first river water level data.

[0079] The instantaneous flow is calculated based on the cross-sectional parameters (such as cross-sectional area) of each flow measurement zone, the first river water level data, the calculated second river water level data and the average flow velocity data obtained from the hydraulic dynamics model.

[0080] The formula for calculating instantaneous flow is: flow equals cross-sectional area multiplied by the average flow velocity. In practice, this calculation is performed separately for each flow measurement section, and then the instantaneous flow rates for each section are summed to obtain the total instantaneous flow rate for the entire target monitoring section. This calculation process is well established and will not be repeated here. The instantaneous flow calculation in this embodiment fully considers the differences in flow characteristics across different river sections, and can more accurately reflect the actual flow conditions in the river at a given moment.

[0081] Based on the calculated instantaneous flow data, the cumulative flow is calculated by integration and other methods. The cumulative flow reflects the total amount of water passing through the target monitoring section area within a certain period of time.

[0082] During real-time monitoring, an alarm is automatically issued when the accumulated flow exceeds a preset threshold. This alarm can be communicated to relevant personnel through various means, such as text messages, emails, and audible and visual alarms, enabling timely action, such as strengthening flood prevention inspections and adjusting water resource allocation plans, to ensure safety and normal water demand in the river and surrounding areas.

[0083] Reference Figure 2 In other embodiments, the method further comprises:

[0084] S14 data update. In actual water flow monitoring, due to limitations in equipment installation location, cost, and other factors, it may not be possible to directly measure river flow velocity data at every location within the target monitoring section. Interpolation is a method of estimating data at unknown locations using known data points. Its purpose is to use limited measured flow velocity data to more accurately estimate the surface flow velocity across the entire target monitoring section, thereby obtaining a more comprehensive surface flow velocity distribution within the area.

[0085] This example uses the inverse distance weighted interpolation method. This method assumes that the flow velocity at an unknown location is inversely proportional to the flow velocity at known data points, with closer data points having a greater influence on the unknown location. During the calculation, different weights are assigned based on the distance between the known data points and the unknown location, and the flow velocity at the unknown location is estimated using a weighted average.

[0086] Suppose there are five measured velocity data points located at different locations within the target monitoring section. The coordinates of these data points and their corresponding velocity values are known. Using the inverse distance weighted interpolation method as an example, for any unmeasured point in the area, the distance between it and the five measured points is calculated. Weights are assigned based on the inverse of the distances, and then a weighted average is performed to obtain an estimated surface velocity value for that unmeasured point. This process is repeated throughout the entire target monitoring section, ultimately yielding the surface velocity data for that area.

[0087] In other embodiments, the water level data of the remaining flow measurement subareas may also be calculated using a linear interpolation method.

[0088] The vertical distribution of flow velocity is usually affected by many factors, such as the dynamic characteristics of the water flow, riverbed topography, boundary conditions, etc. Common flow velocity distribution calculation models include logarithmic flow velocity distribution model and exponential flow velocity distribution model.

[0089] Logarithmic flow velocity distribution model: This model is based on fluid mechanics theory. It assumes that water flow is affected by the combined effects of viscous force and gravity in the vertical direction. The flow velocity is low near the riverbed due to the influence of viscous force. As the height increases, the flow velocity gradually increases and conforms to a logarithmic relationship.

[0090] The calculation model of the logarithmic flow velocity distribution model is as follows:

[0091] ;

[0092] ;

[0093] in, The distance between the i-th flow measurement zone and the riverbed The flow velocity at the point, K is the Karman constant, which is 0.41. is the riverbed roughness height, is the river flow velocity data of the i-th flow measurement area.

[0094] Exponential velocity distribution model: This model assumes that the vertical velocity distribution conforms to the exponential law.

[0095] The calculation model of the exponential flow velocity distribution model is as follows:

[0096] ;

[0097] in, The distance between the i-th flow measurement zone and the riverbed The flow rate at is the river flow velocity data of the ith flow measurement zone, a is the empirical coefficient, which is the same as that in the S12 modeling and is set to 1.23.

[0098] A scatter plot is created within the target monitoring section, using the shortest distance from each location to any riverbank as the horizontal axis and the corresponding vertical velocity data as the vertical axis. Based on the distribution trend of these scatter plots, an appropriate curve fitting method (such as polynomial fitting or exponential fitting) is selected to plot a vertical velocity data curve.

[0099] Subsequently, the first derivative of the vertical velocity data curve is calculated. The first derivative represents the rate of change of vertical velocity with distance. By calculating the first derivative, the speed of change of vertical velocity in different flow measurement zones can be understood. In this embodiment, numerical differentiation is used to calculate the first derivative of the vertical velocity data curve.

[0100] Since first-order derivatives are continuous data obtained through numerical calculations, discretization can convert these first-order derivatives into discrete forms that are easier to process and analyze. At the same time, discretization can also reduce the amount of data and improve computational efficiency.

[0101] This embodiment divides the range of first-order derivative values into several intervals according to the division method of each flow measurement zone, and then corresponds each first-order derivative value to a corresponding interval, and uses the interval number or representative value to represent the first-order derivative value.

[0102] For example, if there are 10 flow measurement subareas (the target monitoring section is 50m long, and each 5m section is a flow measurement subarea, for a total of 10 flow measurement subareas), the range of the first-order derivative value is also divided into 10 segments (0-5m corresponds to one segment, and so on, resulting in 10 segments). The first-order derivative value of each segment is assigned to the corresponding flow measurement subarea to obtain the discretization processing result. Determine whether there is a first-order derivative value of 0 in each flow measurement subarea. If so, the flow measurement subarea is further divided to obtain multiple new flow measurement subareas, and S11 data collection is re-executed; if not, no processing is performed.

[0103] For example, if there are c first-order derivative values of 0 in the i-th flow measurement partition, the i-th flow measurement partition needs to be divided into c+1 new flow measurement partitions.

[0104] The normalization process may map the data range of the discretization process result to [0, 1], and the discretization process result after the normalization process is recorded as the first data.

[0105] The vertical average value can be obtained by multiplying the first data with the corresponding vertical velocity data as a weight, then adding all the products and dividing them by the sum of the first data.

[0106] The vertical average value is updated to the river water velocity data, and the vertical average value can reflect the overall characteristics of the river water velocity in the entire target monitoring section area.

[0107] Example 2: Reference Figure 3 The difference between this embodiment and embodiment 1 is that the method further includes:

[0108] S21 determines the original coordinates and sets at least one reference point, multiple working base points and multiple observation points on the river banks on both sides of the target monitoring section area.

[0109] Benchmarks: Select at least one benchmark point on the riverbank on both sides of the target monitoring section. The benchmark point serves as the fundamental reference point for the entire coordinate system and must be highly stable, unaffected by external factors such as erosion and riverbed deformation. For example, a benchmark point can be set in an area with a hard rock and stable geological structure. Deep-buried concrete piles with embedded metal markers are typically used to ensure their long-term stability.

[0110] Working Base Points: Multiple working base points are set up near the benchmark and within the monitoring section. These serve as transition points between the benchmark and observation points, transmitting coordinate information. They also serve as important reference points for monitoring riverbed deformation. The setting of working base points requires consideration of their uniformity and rationality, ensuring comprehensive coverage of the monitoring section. For example, working base points can be set up at regular intervals upstream and downstream, and on both sides of the monitoring section, using shallowly buried concrete piles or metal markers.

[0111] Observation Points: Multiple observation points are set up at key locations within the target monitoring section. These observation points are used to directly monitor riverbed deformation, and their locations should be selected based on the monitoring objectives and priorities. For example, in areas where deformation such as landslides and subsidence are likely, observation points should be more frequently set up. Surface markers or embedded sensors can be used.

[0112] A three-dimensional rectangular coordinate system is constructed, with the location of the reference point as the coordinate origin. The directions of the X, Y, and Z axes are determined based on geographic orientation to construct the coordinate system. In this embodiment, the X axis is set to the direction of the river flow, the Y axis is perpendicular to the river flow, and the Z axis is perpendicular to the riverbed and points upward. In other embodiments, the directions of the coordinate axes can be set as needed.

[0113] Using high-precision surveying instruments (such as total stations and GPS), each working base point and observation point is measured to determine their original 3D coordinates within the constructed 3D coordinate system. Multiple observations and data processing are required during the measurement process to improve coordinate accuracy. For example, when using a total station, multiple measurements are performed at different time periods and under different weather conditions, and the average value is then taken as the final original coordinate.

[0114] S22 determines the amount of change, and constructs the reference point and at least one working base point into a monitoring reference network. The monitoring reference network is used to provide a stable reference frame for monitoring the displacement change of the working base point.

[0115] After a preset period of time (such as half a year), the three-dimensional coordinates of each working base point are retested using the same measuring instrument and method to obtain the remeasured three-dimensional coordinates of each working base point.

[0116] Based on the original three-dimensional coordinates (X0, Y0, Z0) and the remeasured three-dimensional coordinates (X1, Y1, Z1) of each work base point, the displacement change ΔX=X1-X0 and ΔY=Y1-Y0 in the horizontal direction (X-axis and Y-axis direction) and the displacement change ΔZ=Z1-Z0 in the vertical direction (Z-axis direction) of each work base point are calculated.

[0117] S23 analyzes the three-dimensional displacement, calculates the three-dimensional displacement of each working base point based on the displacement change in the horizontal direction and the displacement change in the vertical direction of each working base point, and judges in turn whether the three-dimensional displacement of each working base point is greater than the preset three-dimensional displacement threshold. If so, a working base point alarm signal is issued; if not, execute S24 to remeasure the observation point coordinates.

[0118] According to the horizontal displacement changes ΔX and ΔY of each working base point and the vertical displacement change ΔZ, the three-dimensional displacement D of each working base point is calculated. The calculation model of the three-dimensional displacement D is as follows:

[0119] .

[0120] The 3D displacement D of each working base point is determined in turn to see if it exceeds a preset 3D displacement threshold. This threshold is determined based on factors such as the riverbed's geological conditions, monitoring objectives, and safety requirements. If the 3D displacement of a working base point exceeds the threshold, a working base point alarm signal is issued, prompting personnel to inspect and analyze the working base point and its surrounding area. If the 3D displacement of all working base points is within the threshold, step S24 is executed to remeasure the observation point coordinates.

[0121] S24 remeasures the coordinates of the observation point to obtain the remeasured three-dimensional coordinates (X1', Y1', Z1') of the observation point. The remeasured three-dimensional coordinates of the observation point and the remeasured three-dimensional coordinates of the working base point have the same timestamp.

[0122] Based on the displacement changes of each working base point in the horizontal direction and the vertical direction, the remeasured three-dimensional coordinates of the observation point are updated to obtain the absolute three-dimensional coordinates (X2, Y2, Z2) of the observation point.

[0123] ;

[0124] Based on the original 3D coordinates (X3, Y3, Z3) and absolute 3D coordinates (X2, Y2, Z2) of the observation point, the 3D displacement D' corresponding to each observation point is calculated. The calculation model of the 3D displacement D' is as follows:

[0125] ;

[0126] The 3D displacement D' of each observation point is determined in turn to determine whether it is greater than a preset 3D displacement threshold. If the 3D displacement of a particular observation point is greater than the preset 3D displacement threshold, an observation point alarm signal is issued, prompting relevant personnel to focus on monitoring and handling the observation point and its surrounding area. If the 3D displacement of all observation points is within the threshold, the prediction step S25 is executed.

[0127] S25 prediction, build statistical analysis models, which can adopt time series analysis models (such as ARIMA model, BI-LSTM model), regression analysis models, etc. The above models can predict the changing trend of the next moment based on historical data and current data.

[0128] The remeasured 3D coordinates and 3D displacement of each work base point are sequentially input into the constructed statistical analysis model. The remeasured 3D coordinates provide spatial position information for the work base point, while the 3D displacement reflects the positional changes of the work base point at different times. The model is trained on this input data, learning patterns and regularities in the data and adjusting model parameters to better fit historical data. Once training is complete, the model calculates a predicted 3D displacement value for each work base point at the next moment based on the current data and the learned regularities.

[0129] Similarly, the absolute three-dimensional coordinates of each observation point and the three-dimensional displacement of each observation point are sequentially input into the statistical analysis model to obtain the predicted value of the three-dimensional displacement of each observation point at the next moment.

[0130] Determine whether the predicted values of the three-dimensional displacement of each working base point at the next moment and the predicted values of the three-dimensional displacement of each observation point at the next moment are consistent with expectations. If the predicted values do not meet expectations, a warning signal is output to prompt relevant personnel to take appropriate measures; if the predicted values meet expectations, execute S26 to calculate the deviation.

[0131] The predicted value of the three-dimensional displacement of each working base point at the next moment and the predicted value of the three-dimensional displacement of each observation point at the next moment are in line with expectations, which means that the predicted value of the three-dimensional displacement of each working base point at the next moment and the predicted value of the three-dimensional displacement of each observation point at the next moment are not greater than the preset three-dimensional displacement threshold.

[0132] S26 calculates the deviation and obtains the historical 3D displacement of the observation point and the real-time 3D displacement at the current moment. The historical 3D displacements are arranged in chronological order of acquisition time to obtain a historical 3D displacement sequence, and a historical 3D displacement curve is plotted based on the historical 3D displacement sequence.

[0133] A safety monitoring model is established to fit the historical three-dimensional displacement curve, and the theoretical three-dimensional displacement at the current moment is obtained based on the safety monitoring model. The theoretical three-dimensional displacement is the expected displacement value at the current moment calculated according to the safety monitoring model.

[0134] For example, the historical three-dimensional displacement sequence is: 0.8, 0.85, 0.89, 0.9, 0.95, 0.97. In this example, a quadratic polynomial is used for fitting, and the safety monitoring model is:

[0135] ;

[0136] After recursive fitting, we obtained A=-0.0029, B=0.0386, and C=0.7714.

[0137] That is, the calculation model of the safety monitoring model in this example is:

[0138] .

[0139] In other embodiments, a first-order polynomial or a higher-order polynomial may be used for recursive fitting.

[0140] Determine whether the difference between the real-time 3D displacement and the theoretical 3D displacement is greater than a preset difference threshold. If the difference is greater than the preset difference threshold, a deviation alarm signal is output, prompting relevant personnel to conduct further inspection and processing of the observation point and its surrounding area; if the difference is not greater than the preset difference threshold, no action is taken.

[0141] This embodiment first sets at least one benchmark point, multiple working benchmark points, and observation points on both sides of the riverbank in the target monitoring section area, constructs a three-dimensional coordinate system with the benchmark point as the origin, and determines the original three-dimensional coordinates of each working benchmark point and observation point. Then, a monitoring benchmark network is constructed. The coordinates of the working benchmark point are remeasured after a preset period of time. Based on this, the displacement changes in the horizontal and vertical directions are calculated, and the three-dimensional displacement is analyzed. If the displacement exceeds a preset threshold, an alarm signal for the working benchmark point is issued; otherwise, the coordinates of the observation point are remeasured. After obtaining the remeasured coordinates of the observation point, its coordinates are updated with the displacement change of the working benchmark point to obtain the absolute three-dimensional coordinates. The three-dimensional displacement of the observation point is calculated. If the threshold is exceeded, an alarm is issued; otherwise, a prediction is performed. A statistical analysis model is constructed, and the coordinates and displacement data of the working benchmark point and the observation point are input to obtain a predicted displacement change value. If it does not meet the expectations, an alarm is issued; if it meets the expectations, a deviation is calculated. The historical and real-time three-dimensional displacements of the observation point are obtained, a historical curve is plotted, and a safety monitoring model is established. Based on this, the theoretical three-dimensional displacement at the current moment is obtained. If the difference between the real-time value and the theoretical value exceeds the preset difference threshold, a deviation alarm signal is issued; otherwise, no action is taken.

[0142] Example 3: This example differs from Example 1 in that the method further comprises:

[0143] River channel data collection is performed using a drone equipped with a lidar system. LiDAR emits a laser beam into the river's surroundings. Upon hitting an object, the laser beam reflects back. A sensor receives the reflected signal and records the laser's flight time. Based on the speed of light and flight time, the distance between the object and the drone can be accurately calculated. Combining the drone's positioning information (such as GPS coordinates) and attitude data (such as pitch, roll, and yaw angles), the precise position of each measurement point in three-dimensional space can be determined, thereby generating three-dimensional point cloud data of the river channel.

[0144] Utilize professional 3D modeling software (such as CloudCompare and MeshLab) or algorithms (such as the Poisson reconstruction algorithm) to convert 3D point cloud data into a 3D point cloud model of the river. These software and algorithms automatically generate a triangular mesh model based on the distribution and characteristics of the point cloud data, restoring the surface morphology and topographic features of the river.

[0145] Clearly define inspection objectives, such as monitoring changes in river water quality, inspecting the condition of surrounding structures, and observing river ecology. Different inspection objectives may require specific focus areas and parameters. Based on the 3D point cloud model and the inspection objectives, a path planning algorithm (such as the A algorithm or Dijkstra algorithm) is used to develop the drone's initial inspection route. This algorithm considers factors such as the river's topography and obstacle distribution to ensure the drone can safely and efficiently cover all inspection areas. For example, the A algorithm estimates a cost function from the starting point to the end point to find the optimal flight path while avoiding obstacles.

[0146] The drone is controlled to patrol the river along the initial inspection route. During the inspection, the drone's onboard sensors (such as meteorological sensors and water quality sensors) collect real-time meteorological data (such as temperature, humidity, wind speed, and direction) and water quality data (such as pH, dissolved oxygen, and turbidity).

[0147] Collected meteorological and water quality data is uploaded to a cloud server via the drone's communication module (e.g., 4G / 5G transmission). The cloud server can store, analyze, and process the data, providing data support for subsequent decision-making. For example, by monitoring water quality trends over a long period of time, water pollution issues can be promptly identified and appropriate remediation measures implemented.

[0148] In other embodiments, the method further comprises:

[0149] Unmanned vessels are equipped with a variety of sensors, such as lidar, cameras, and sonar, to perceive the water surface and its surroundings. Lidar acquires three-dimensional information about the surface and the shore, cameras capture visual images of the water surface, and sonar detects underwater terrain and obstacles. The data collected by these sensors is integrated and processed to construct a river model. This model is updated in real time using algorithms such as simultaneous localization and mapping (SLAM).

[0150] Mark the water contours and the locations of obstacles in the river model. Marking the water contours helps determine the scope of the cleaning operation, while marking obstacles can prevent collisions between unmanned boats and the waterway. For example, the location and size of obstacles such as bridge piers, reefs, and abandoned ships can be marked. This embodiment uses a combination of manual and automatic marking methods. Manual marking involves the operator visually marking important feature points on the software interface; automatic marking uses algorithms such as image recognition and target detection to automatically identify and mark obstacles and water contours.

[0151] Based on the marked river model, the river is divided into multiple sub-areas. This division can be based on the river's geographical characteristics (such as length and width), obstacle distribution, or cleaning task priority. Within each sub-area, unmanned boats clean the waters according to pre-set cleaning strategies. Cleaning strategies can be tailored to the specific characteristics of the sub-areas, such as increasing the frequency and intensity of cleaning in heavily polluted areas. Unmanned boats can improve the water environment by salvaging floating debris and absorbing oil.

[0152] This embodiment can achieve comprehensive monitoring and efficient cleaning of the river through the collaborative operation of drones and unmanned boats, thereby improving the intelligent level of river management and the quality of the water environment.

[0153] Example 4: This example discloses a river safety monitoring system, which includes a processor and a memory.

[0154] The memory stores program code;

[0155] The processor executes the steps of the method when calling the program code in the memory.

[0156] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A river safety monitoring method, characterized in that: include: S11 Data Collection: Obtain riverbed elevation distribution data of the target monitoring section area, divide the target monitoring section area into multiple flow measurement zones, obtain section parameters of each flow measurement zone, each flow measurement zone is provided with a radar flow velocity sensor for collecting river flow velocity data, and a radar water level sensor for measuring first river water level data is provided in any flow measurement zone; S12 Modeling: Construct a hydraulic dynamics model, input the river flow velocity data of each flow measurement zone into the hydraulic dynamics model, and obtain the average flow velocity data of each flow measurement zone; S13 Flow calculation and alarm: Calculate the second river water level data of the remaining flow measurement sections based on the first river water level data and riverbed elevation distribution data; calculate the instantaneous flow based on the section parameters, the first river water level data, the second river water level data and the average flow velocity data; calculate the cumulative flow based on the instantaneous flow, and issue an alarm signal when the cumulative flow exceeds the preset flow threshold; Based on all river water velocity data, the surface velocity data of the target monitoring section area is calculated using the interpolation method. The vertical velocity data of each position in the target monitoring section area is calculated according to the preset velocity distribution calculation model. A vertical velocity data change curve is drawn based on the vertical velocity data. The horizontal axis of the vertical velocity data change curve is the shortest distance to any river bank, and the vertical axis is the vertical velocity data. Calculating a first-order derivative of the vertical flow velocity data change curve with respect to distance, performing discretization processing on the first-order derivative to obtain a discretization processing result, performing normalization processing on the discretization processing result to obtain first data, calculating a vertical average value based on the first data and the vertical flow velocity data, and updating the vertical average value to the river water flow velocity data: According to the division method of each flow measurement partition, the range of the first-order derivative value is divided into several intervals, and then each first-order derivative value is mapped to the corresponding interval, and the first-order derivative value is represented by the interval number or representative value. It is judged whether there is a first-order derivative value of 0 in each flow measurement partition. If so, the flow measurement partition is divided again to obtain multiple new flow measurement partitions, and S11 data collection is re-executed. If there are c first-order derivative values of 0 in the i-th flow measurement partition, it is necessary to divide the i-th flow measurement partition into c+1 new flow measurement partitions; if not, no processing is performed; Normalization processing maps the data range of the discretization processing result to [0, 1]. The discretization processing result after normalization is recorded as the first data. The first data is used as the weight to multiply the corresponding vertical flow velocity data. Then all the products are added and divided by the sum of the first data to obtain the vertical average value.

2. The river safety monitoring method according to claim 1, characterized in that: The method further comprises: At least one reference point, multiple working reference points, and multiple observation points are set on the riverbanks on both sides of the target monitoring section area, a three-dimensional coordinate system is constructed with the location of the reference point as the coordinate origin, and the original three-dimensional coordinates of each working reference point and each observation point are determined; Select at least one reference point and at least one working reference point to construct a monitoring reference network, retest the three-dimensional coordinates of each working reference point after a preset time period, obtain the remeasured three-dimensional coordinates of each working reference point, and obtain the horizontal displacement change and vertical displacement change of each working reference point based on the original three-dimensional coordinates and the remeasured three-dimensional coordinates of each working reference point; The three-dimensional displacement of each working base point is calculated based on the displacement change in the horizontal direction and the displacement change in the vertical direction of each working base point, and the three-dimensional displacement of each working base point is judged in turn to see whether it is greater than the preset three-dimensional displacement threshold. If so, a working base point alarm signal is issued; if not, no processing is performed.

3. The river safety monitoring method according to claim 2, characterized in that: When the three-dimensional displacement of the working base point is not greater than a preset three-dimensional displacement threshold, the method further includes: Obtaining remeasured three-dimensional coordinates of the observation point, wherein the remeasured three-dimensional coordinates of the observation point and the remeasured three-dimensional coordinates of the working base point have the same timestamp; Based on the horizontal and vertical displacement changes of each working base point, the re-measured three-dimensional coordinates of the observation point are updated to obtain the absolute three-dimensional coordinates of the observation point. The three-dimensional displacement of each observation point is calculated according to the original three-dimensional coordinates and the absolute three-dimensional coordinates of the observation point. It is judged in turn whether the three-dimensional displacement of each observation point is greater than the preset three-dimensional displacement threshold. If so, an observation point alarm signal is issued; if not, no processing is performed.

4. The river safety monitoring method according to claim 3, characterized in that: When the three-dimensional displacement of the observation point is not greater than a preset three-dimensional displacement threshold, the method further includes: Construct a statistical analysis model, input the remeasured three-dimensional coordinates of each working base point and the three-dimensional displacement of each working base point into the statistical analysis model in sequence, and obtain the predicted value of the three-dimensional displacement of each working base point at the next moment; input the absolute three-dimensional coordinates of each observation point and the three-dimensional displacement of each observation point into the statistical analysis model in sequence, and obtain the predicted value of the three-dimensional displacement of each observation point at the next moment; Determine whether the predicted value of the three-dimensional displacement of each working base point at the next moment and the predicted value of the three-dimensional displacement of each observation point at the next moment are consistent with expectations. If so, no processing is performed; if not, an early warning signal is output.

5. The river safety monitoring method according to claim 4, characterized in that: The method further comprises: Obtain the historical three-dimensional displacement of the observation point and the real-time three-dimensional displacement at the current moment, arrange the historical three-dimensional displacements in chronological order of acquisition time to obtain a historical three-dimensional displacement sequence, and draw a historical three-dimensional displacement curve based on the historical three-dimensional displacement sequence; Establish a safety monitoring model for fitting the historical three-dimensional displacement curve, and obtain the theoretical three-dimensional displacement at the current moment based on the safety monitoring model; Determine whether the difference between the real-time three-dimensional displacement and the theoretical three-dimensional displacement is greater than a preset difference threshold. If so, output a deviation alarm signal; if not, do nothing.

6. The river safety monitoring method according to claim 1, characterized in that: The method also includes: using a laser radar carried by a drone to obtain three-dimensional point cloud data of the river channel, constructing a three-dimensional point cloud model of the river channel based on the three-dimensional point cloud data, formulating an initial inspection route of the drone according to the three-dimensional point cloud model and the inspection target and using a path planning algorithm, controlling the drone to inspect the river channel according to the initial inspection route, and collecting meteorological data and water quality data during the inspection process, and uploading the meteorological data and water quality data to the cloud.

7. The river safety monitoring method according to claim 6, characterized in that: The method further comprises: The sensors carried by the unmanned boat are used to perceive and model the water surface and boundary environment of the water area to obtain a river model. The water contour and the location of obstacles are marked in the river model to obtain a marked river model. Based on the marked river model, the river is divided into multiple sub-areas, and the water area is cleaned in each sub-area.

8. A river safety monitoring system, characterized in that: include: processor and memory, The memory stores program code; When the processor calls the program code in the memory, the steps of the method according to any one of claims 1 to 7 are executed.

Images