A dam area deformation early warning method and system based on laser scanner

Abstract

The present invention relates to the field of dam area deformation early warning technology, and proposes a dam area deformation early warning method and system based on a laser scanner, which includes the following steps: selecting a drone, equipping the drone with a laser scanner and a Beidou positioning system; planning the monitoring flight path of the drone, continuously emitting lasers and receiving reflected signals through the laser scanner, and using the Beidou positioning system to record the scanning position information of the laser scanner in real time to obtain original three-dimensional point cloud data; integrating the three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system through a registration algorithm to obtain a three-dimensional model of the dam area; using a statistical method and a preset threshold to perform a data consistency check on the original deformation data set, screening out abnormal values ​​in the original deformation data set, and obtaining a deformation data set; identifying the deformation data set through a dam area deformation risk assessment model, obtaining a dam area deformation risk level, and taking corresponding early warning measures based on different dam area deformation risk levels. The present invention improves the accuracy of deformation data, realizes accurate judgment of deformation risk levels, and takes corresponding early warning measures according to different risk levels, thereby enhancing the safety of the dam area.

Description

Technical Field

[0001] The present invention relates to the technical field of dam area deformation early warning, and in particular to a dam area deformation early warning method and system based on a laser scanner. Background Art

[0002] Dam area safety is an important engineering issue, which requires real-time monitoring and early warning of dam deformation. Traditional dam area deformation monitoring methods mainly rely on manual inspections and the deployment of a small number of sensors, which have problems such as limited monitoring range and untimely data acquisition, making it difficult to timely detect and warn of dam area deformation risks.

[0003] The existing technology uses drones equipped with laser scanners to perform 3D modeling and deformation monitoring of the dam area, which can fully cover the entire dam area, obtain high-precision 3D point cloud data through laser scanning, and analyze the deformation characteristics of the dam area based on multi-phase point cloud data. However, the existing methods are less effective in data consistency verification and outlier removal, and are easily affected by noise and interference, which affects the accuracy of deformation data. Summary of the invention

[0004] In view of this, the present invention proposes a dam area deformation early warning method and system based on a laser scanner, which solves the problem that the prior art has poor effects in data consistency verification and outlier elimination and is easily affected by noise and interference.

[0005] The technical solution of the present invention is implemented as follows: In the first aspect, the present invention provides a dam area deformation early warning method based on a laser scanner, comprising the following steps:

[0006] S1, selecting a UAV suitable for the complex environment of the dam area, and equipping the UAV with a laser scanner and a Beidou positioning system;

[0007] S2, based on the geographical data of the dam area, planning the monitoring flight path of the UAV to cover the entire dam area, continuously emitting lasers through the laser scanner and receiving reflected signals, and using the Beidou positioning system to record the scanning position information of the laser scanner in real time to obtain original three-dimensional point cloud data;

[0008] S3, preprocessing the original three-dimensional point cloud data to obtain three-dimensional point cloud data, integrating the three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system through a registration algorithm, and obtaining a three-dimensional model of the dam area;

[0009] S4, extracting the original deformation data set of the dam area three-dimensional model at different time points, using a statistical method and a preset threshold to perform a data consistency check on the original deformation data set, filtering out abnormal values ​​in the original deformation data set, and obtaining a deformation data set, wherein the deformation data set includes a displacement, a deformation rate, and a deformation direction;

[0010] S5, constructing a dam area deformation risk assessment model, identifying the deformation data set through the dam area deformation risk assessment model, obtaining the dam area deformation risk level, and taking corresponding early warning measures based on different dam area deformation risk levels.

[0011] Based on the above technical solution, preferably, step S2 includes:

[0012] S21, based on the geographical data of the dam area, the path planning algorithm is used to generate the initial flight path of the UAV, the dam area is divided into several grid units, a topological map of the dam area is constructed based on the grid units, and the optimal path is searched on the topological map through the heuristic search algorithm to generate the initial flight path;

[0013] S22, optimize and adjust the initial flight path according to the terrain characteristics and obstacle distribution of the dam area, identify the potential danger areas and blind spot areas in the initial flight path, adjust the flight altitude and flight angle of the initial flight path according to the potential danger areas and the blind spot areas to avoid the potential danger areas and cover the blind spot areas, and adjust the flight speed of the UAV and the scanning frequency of the laser scanner based on the endurance of the UAV and the effective scanning range of the laser scanner.

[0014] On the basis of the above technical solution, preferably, step S3 includes:

[0015] S31, performing preliminary outlier removal on the original three-dimensional point cloud data by a statistical filtering method to obtain first three-dimensional point cloud data, removing noise points on the first three-dimensional point cloud data by a statistical outlier filtering algorithm to obtain second three-dimensional point cloud data, downsampling the second three-dimensional point cloud data based on a voxel grid filtering algorithm to obtain third three-dimensional point cloud data, removing outlier points of the third three-dimensional point cloud data by a radius outlier filtering algorithm to obtain fourth three-dimensional point cloud data, and smoothing the fourth three-dimensional point cloud data based on a Gaussian filter and a bilateral filtering algorithm to obtain preprocessed three-dimensional point cloud data;

[0016] S32, select fixed reference scanning data as the reference coordinate system, select corresponding point pairs in the source point cloud and the target point cloud, calculate the optimal rigid body transformation of the corresponding point pairs to minimize the square distance error between the source point cloud and the target point cloud, and apply the optimal rigid body transformation to the source point cloud; repeatedly select corresponding point pairs to calculate the rigid body transformation and apply it until the preset number of iterations is reached, and integrate all three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system to form a three-dimensional model of the dam area.

[0017] Based on the above technical solution, preferably, step S4 includes:

[0018] S41, adaptively meshing the three-dimensional model of the dam area at each time point, calculating the three-dimensional displacement vectors of corresponding grid nodes between adjacent time points, smoothing the three-dimensional displacement vectors using a dynamic trend smoothing method, calculating the rate of change of the three-dimensional displacement vectors using a multi-scale deformation rate method to obtain the deformation rate, and determining the deformation direction based on a principal component analysis method;

[0019] S42, calculate the mean, standard deviation and skewness of the original deformation data set, dynamically set the outlier judgment threshold based on the data distribution characteristics, identify the potential outliers of the original deformation data set through an adaptive Z-score method, perform spatial clustering analysis on the potential outliers, identify the confirmed outliers of the original deformation data set, eliminate the confirmed outliers of the original deformation data set, and obtain the final deformation data set.

[0020] On the basis of the above technical solution, preferably, step S4 further includes:

[0021] The calculation formula of the dynamic trend smoothing method is:

[0022] S t =α·X t +(1-α)·S t-1 +γ·ΔX t ;

[0023] Among them, S t is the displacement after smoothing, X t is the original displacement, α is the smoothing coefficient, S t-1 is the displacement of the previous time step, γ is the trend correction coefficient, ΔX t is the displacement change;

[0024] The calculation formula of the multi-scale deformation rate method is:

[0025]

[0026] Where V is the deformation rate, D n and D n-1 are the displacements at the nth and n-1th time steps, respectively, t n and t n-1 D n and D n-1 The corresponding time step, w n is the weight coefficient of the nth time period, and N is the total number of time steps;

[0027] The calculation formula of the adaptive Z-score method is:

[0028]

[0029] Where Z is the adaptive Z-score value, X is the original data in the original deformed dataset, median(X) is the median of the original data in the original deformed dataset, MAD is the absolute median difference, and τ is the anti-zero constant;

[0030] The calculation formula of the spatial cluster analysis is:

[0031]

[0032] Among them, C is the spatial autocorrelation index, W i,j is the spatial weight matrix, M i and M j are the adaptive Z-score values ​​of point i and point j respectively, μM is the mean of M, σ M is the standard deviation of M, λ is the distance attenuation coefficient, d i,j is the spatial distance between point i and point j.

[0033] Based on the above technical solution, preferably, step S5 includes:

[0034] S51, integrating the basic warning indicators, auxiliary warning indicators and environmental impact indicators of the deformation data set to obtain a comprehensive warning index, wherein the basic warning indicators include displacement, deformation rate and deformation acceleration, the auxiliary warning indicators include the main direction of deformation and the spatial autocorrelation index, and the environmental impact indicators include rainfall and temperature change;

[0035] S52, obtaining historical deformation data and warning records, constructing a training set and a verification set according to the historical deformation data and the warning records, establishing an original dam area deformation risk assessment model based on the comprehensive warning index using an integrated learning method, training and testing the original dam area deformation risk assessment model through the training set and the verification set to obtain a dam area deformation risk assessment model, identifying the deformation data set through the dam area deformation risk assessment model to obtain a dam area deformation risk level, and taking corresponding warning measures based on different dam area deformation risk levels.

[0036] On the basis of the above technical solution, preferably, step S5 further includes:

[0037] The calculation formula of the comprehensive early warning index is:

[0038]

[0039] Among them, CWI is the comprehensive early warning index, w k is the weight of the kth early warning indicator, K is the total number of early warning indicators, and f k (x k ) is the membership function of the kth indicator, p is the norm parameter, ε is the environmental impact indicator magnification coefficient, vl is the weight of the lth environmental factor, e l is the standardized value of the lth environmental impact indicator, L is the total number of environmental impact indicators, and tanh(·) is the hyperbolic tangent activation function;

[0040] The calculation formula of the model evaluation function of the original dam area deformation risk assessment model is:

[0041]

[0042] Among them, Score is the model score of the original dam area deformation risk assessment model, β1, β2, and β3 are the weight coefficients of accuracy, recall, and precision, respectively, Accuracy, Recall, and Precision are accuracy, recall, and precision, respectively, Q is the number of training samples, δ is the time decay parameter, and exp(·) is the exponential function.

[0043] In a second aspect, the present invention further provides a dam area deformation early warning system based on a laser scanner, the system comprising:

[0044] A drone scanning module is used to select drones suitable for the complex environment of the dam area and equip the drones with laser scanners and Beidou positioning systems;

[0045] A path planning module is used to plan the monitoring flight path of the UAV based on the geographical data of the dam area to cover the entire dam area, continuously emit lasers through the laser scanner and receive reflected signals, and use the Beidou positioning system to record the scanning position information of the laser scanner in real time to obtain original three-dimensional point cloud data;

[0046] The dam area model module is used to pre-process the original three-dimensional point cloud data to obtain three-dimensional point cloud data, and integrate the three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system through a registration algorithm to obtain a three-dimensional model of the dam area;

[0047] An outlier elimination module is used to extract the original deformation data set of the dam area three-dimensional model at different time points, use statistical methods and preset thresholds to perform data consistency check on the original deformation data set, screen out the outliers in the original deformation data set, and obtain the deformation data set, wherein the deformation data set includes displacement, deformation rate, and deformation direction;

[0048] The deformation early warning module is used to construct a dam area deformation risk assessment model, identify the deformation data set through the dam area deformation risk assessment model, obtain the dam area deformation risk level, and take corresponding early warning measures based on different dam area deformation risk levels.

[0049] In a third aspect, the present invention further provides an electronic device, comprising: at least one processor, at least one memory, a communication interface and a bus;

[0050] Among them, the processor, memory, and communication interface communicate with each other through the bus, the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement steps such as a dam area deformation early warning method based on a laser scanner.

[0051] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions enable a computer to implement steps of a dam area deformation early warning method based on a laser scanner.

[0052] The dam area deformation early warning method and system based on laser scanner of the present invention has the following beneficial effects compared with the prior art:

[0053] (1) By using unmanned aerial vehicles equipped with laser scanners, the dam area is fully covered and high-precision three-dimensional point cloud data is obtained. The Beidou positioning system is used to record the scanning position information in real time. A unified three-dimensional model of the dam area is formed by preprocessing and integrating the original data. Statistical methods and preset thresholds are used to check the data consistency, effectively eliminate outliers, improve the accuracy of deformation data, and build a dam area deformation risk assessment model, so as to accurately judge the deformation risk level and take corresponding early warning measures according to different risk levels to enhance the safety of the dam area.

[0054] (2) The original deformation data are processed and outliers are eliminated by using the dynamic trend smoothing method, multi-scale deformation rate method and adaptive Z-score method to improve the quality and reliability of deformation data. Not only can key indicators such as displacement, deformation rate and deformation direction be accurately extracted, but abnormal data can also be adaptively identified and removed, which improves the data analysis capability and prediction accuracy of the entire early warning system, thereby enhancing the reliability of deformation risk assessment in the dam area.

[0055] (3) By constructing a comprehensive early warning index, the basic early warning indicators, auxiliary early warning indicators and environmental impact indicators are effectively integrated to comprehensively reflect the deformation risk status of the dam area. The hyperbolic tangent activation function is used for dimensionless processing, and the norm parameter is introduced to adjust the relative importance of each indicator. When constructing the dam area deformation risk assessment model, the integrated learning method is adopted and the time attenuation parameter is introduced. Targeted prevention and emergency measures are taken under different deformation risk conditions, thereby minimizing the potential losses of safety accidents in the dam area. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 This is a flow chart of a dam area deformation early warning method based on a laser scanner according to the present invention;

[0058] Figure 2 This is a structural diagram of a dam area deformation early warning system based on a laser scanner according to the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] See also Figure 1 The present invention provides a dam area deformation early warning method based on a laser scanner, comprising the following steps:

[0061] S1, selecting a UAV suitable for the complex environment of the dam area, and equipping the UAV with a laser scanner and a Beidou positioning system;

[0062] S2, based on the geographical data of the dam area, planning the monitoring flight path of the UAV to cover the entire dam area, continuously emitting lasers through the laser scanner and receiving reflected signals, and using the Beidou positioning system to record the scanning position information of the laser scanner in real time to obtain original three-dimensional point cloud data;

[0063] S3, preprocessing the original three-dimensional point cloud data to obtain three-dimensional point cloud data, integrating the three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system through a registration algorithm, and obtaining a three-dimensional model of the dam area;

[0064] S4, extracting the original deformation data set of the dam area three-dimensional model at different time points, using a statistical method and a preset threshold to perform a data consistency check on the original deformation data set, filtering out abnormal values ​​in the original deformation data set, and obtaining a deformation data set, wherein the deformation data set includes a displacement, a deformation rate, and a deformation direction;

[0065] S5, constructing a dam area deformation risk assessment model, identifying the deformation data set through the dam area deformation risk assessment model, obtaining the dam area deformation risk level, and taking corresponding early warning measures based on different dam area deformation risk levels.

[0066] Specifically, this embodiment uses a laser scanner mounted on an unmanned aerial vehicle to comprehensively cover the dam area and obtain high-precision three-dimensional point cloud data, uses the Beidou positioning system to record the scanning position information in real time, and forms a unified three-dimensional model of the dam area through preprocessing and integration of the original data. Statistical methods and preset thresholds are used to check data consistency, effectively eliminate outliers, improve the accuracy of deformation data, and construct a dam area deformation risk assessment model, so as to accurately judge the deformation risk level and take corresponding early warning measures according to different risk levels to enhance the safety of the dam area.

[0067] Specifically, step S1 of this embodiment selects a drone suitable for the complex environment of the dam area and equips it with a laser scanner and a Beidou positioning system, so as to fully cover the entire dam area and obtain high-precision three-dimensional point cloud data, thereby overcoming the limitations of deployment when the number of sensors is limited and ensuring comprehensive monitoring of the deformation of the dam area.

[0068] In a specific embodiment, a UAV suitable for the complex environment of the dam area is first selected. The environment of the dam area is usually complex, with factors such as undulating terrain and climate change, so it is necessary to select a UAV platform with excellent performance and strong adaptability. Then, the selected UAV is equipped with a laser scanner and a Beidou positioning system. The laser scanner can emit lasers and receive reflected signals to obtain high-precision three-dimensional point cloud data. The Beidou positioning system can record the scanning position information of the laser scanner in real time to ensure the time and space information of data acquisition. By using a UAV equipped with a laser scanner and a Beidou positioning system, the entire dam area can be fully covered to obtain high-precision three-dimensional point cloud data.

[0069] Step S2 includes:

[0070] S21, based on the geographical data of the dam area, the path planning algorithm is used to generate the initial flight path of the UAV, the dam area is divided into several grid units, a topological map of the dam area is constructed based on the grid units, and the optimal path is searched on the topological map through the heuristic search algorithm to generate the initial flight path;

[0071] S22, optimize and adjust the initial flight path according to the terrain characteristics and obstacle distribution of the dam area, identify the potential danger areas and blind spot areas in the initial flight path, adjust the flight altitude and flight angle of the initial flight path according to the potential danger areas and the blind spot areas to avoid the potential danger areas and cover the blind spot areas, and adjust the flight speed of the UAV and the scanning frequency of the laser scanner based on the endurance of the UAV and the effective scanning range of the laser scanner.

[0072] Specifically, this embodiment generates the monitoring flight path of the UAV by using the path planning algorithm based on the geographic data of the dam area, ensuring that the UAV can cover the entire dam area. By dividing the dam area into several grid units, a topological map of the dam area is constructed, and the optimal path is searched on the topological map by the heuristic search algorithm, thereby achieving efficient flight path planning. In addition, the initial flight path is optimized and adjusted according to the terrain characteristics and obstacle distribution of the dam area, which can effectively identify potential dangerous areas and blind spots, and adjust the flight altitude and angle according to these areas to ensure that the UAV avoids dangerous areas and covers blind spots during flight. Not only does it improve the safety and effectiveness of monitoring, but it also ensures the effective scanning range of the laser scanner.

[0073] Step S3 includes:

[0074] S31, performing preliminary outlier removal on the original three-dimensional point cloud data by a statistical filtering method to obtain first three-dimensional point cloud data, removing noise points on the first three-dimensional point cloud data by a statistical outlier filtering algorithm to obtain second three-dimensional point cloud data, downsampling the second three-dimensional point cloud data based on a voxel grid filtering algorithm to obtain third three-dimensional point cloud data, removing outlier points of the third three-dimensional point cloud data by a radius outlier filtering algorithm to obtain fourth three-dimensional point cloud data, and smoothing the fourth three-dimensional point cloud data based on a Gaussian filter and a bilateral filtering algorithm to obtain preprocessed three-dimensional point cloud data;

[0075] S32, select fixed reference scanning data as the reference coordinate system, select corresponding point pairs in the source point cloud and the target point cloud, calculate the optimal rigid body transformation of the corresponding point pairs to minimize the square distance error between the source point cloud and the target point cloud, and apply the optimal rigid body transformation to the source point cloud; repeatedly select corresponding point pairs to calculate the rigid body transformation and apply it until the preset number of iterations is reached, and integrate all three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system to form a three-dimensional model of the dam area.

[0076] Specifically, this embodiment preprocesses the original three-dimensional point cloud data and uses a statistical filtering method to perform preliminary outlier removal, thereby removing noise points caused by environmental interference or equipment errors, thereby obtaining more reliable first three-dimensional point cloud data. A statistical outlier filtering algorithm and a voxel grid filtering algorithm are used to further remove noise and reduce the amount of data.

[0077] The removal of outliers through the radius outlier filtering algorithm and the smoothing processing of the Gaussian filter and the bilateral filtering algorithm can significantly improve the smoothness and consistency of the point cloud data and reduce random errors in the data.

[0078] Step S4 includes:

[0079] S41, adaptively meshing the three-dimensional model of the dam area at each time point, calculating the three-dimensional displacement vectors of corresponding grid nodes between adjacent time points, smoothing the three-dimensional displacement vectors using a dynamic trend smoothing method, calculating the rate of change of the three-dimensional displacement vectors using a multi-scale deformation rate method to obtain the deformation rate, and determining the deformation direction based on a principal component analysis method;

[0080] S42, calculate the mean, standard deviation and skewness of the original deformation data set, dynamically set the outlier judgment threshold based on the data distribution characteristics, identify the potential outliers of the original deformation data set through an adaptive Z-score method, perform spatial clustering analysis on the potential outliers, identify the confirmed outliers of the original deformation data set, eliminate the confirmed outliers of the original deformation data set, and obtain the final deformation data set.

[0081] Specifically, this embodiment performs adaptive meshing on the three-dimensional model of the dam area and calculates the three-dimensional displacement vectors of the corresponding grid nodes between adjacent time points. The displacement vectors are smoothed using the dynamic trend smoothing method to effectively eliminate measurement noise and random errors. The rate of change of the three-dimensional displacement vector is calculated using the multi-scale deformation rate method, which can reflect the deformation characteristics of the dam area at different time scales.

[0082] Step S4 also includes:

[0083] The calculation formula of the three-dimensional displacement vector is:

[0084] D=((x2-x1)·w x ,(y2-y1)·w y ,(z2-z1)·w z );

[0085] Where D is the three-dimensional displacement vector, (x1, y1, z1) and (x2, y2, z2) are the coordinates of adjacent time points, and w x 、w y 、w z The weight coefficients are dynamically adjusted in the x, y, and z directions according to the point cloud density and measurement accuracy;

[0086] The calculation formula of the dynamic trend smoothing method is:

[0087] S t =α·X t +(1-α)·S t-1 +γ·ΔX t ;

[0088] Among them, S t is the displacement after smoothing, X t is the original displacement, α is the smoothing coefficient, S t-1is the displacement of the previous time step, γ is the trend correction coefficient, ΔX t is the displacement change;

[0089] The calculation formula of the multi-scale deformation rate method is:

[0090]

[0091] Where V is the deformation rate, D n and D n-1 are the displacements at the nth and n-1th time steps, respectively, t n and t n-1 D n and D n-1 The corresponding time step, w n is the weight coefficient of the nth time period, and N is the total number of time steps;

[0092] The calculation formula of the adaptive Z-score method is:

[0093]

[0094] Where Z is the adaptive Z-score value, X is the original data in the original deformed dataset, median(X) is the median of the original data in the original deformed dataset, MAD is the absolute median difference, and τ is the anti-zero constant;

[0095] The calculation formula of the spatial cluster analysis is:

[0096]

[0097] Among them, C is the spatial autocorrelation index, W i,j is the spatial weight matrix, M i and M j are the adaptive Z-score values ​​of point i and point j respectively, μM is the mean of M, σ M is the standard deviation of M, λ is the distance attenuation coefficient, d i,j is the spatial distance between point i and point j.

[0098] Specifically, this embodiment can more accurately reflect the deformation of the dam area at different time points by considering the coordinate differences between adjacent time points and combining the dynamically adjusted weight coefficient (according to the point cloud density and measurement accuracy).

[0099] This embodiment combines the original displacement, smoothing coefficient, displacement of the previous time step and trend correction coefficient to suppress the influence of noise and extract a more realistic deformation trend. By introducing the displacement of different time steps and the weight coefficient of the time step, the deformation rate is analyzed at different time scales to ensure that the short-term and long-term deformation characteristics are captured.

[0100] By calculating the median and absolute median difference of the original data, the outlier judgment threshold is dynamically set, making the identification of outliers more flexible and accurate. The adaptive Z-score method can be adjusted according to the actual distribution of the data, improving the accuracy of outlier detection. By calculating the spatial autocorrelation index, outliers with similar characteristics in space can be identified. Spatial clustering analysis can effectively integrate spatial information and improve the accuracy of data processing.

[0101] The dynamically adjusted weight coefficients and adaptive outlier detection methods significantly enhance the reliability of data processing under different environmental conditions and can better adapt to the complex monitoring environment of the dam area. The application of the multi-scale deformation rate method makes the analysis of dam area deformation more comprehensive and can capture both short-term and long-term deformation dynamics.

[0102] By eliminating outliers, high-quality input data can be provided for the dam area deformation risk assessment model, thereby improving the response speed of the early warning system and ensuring that corresponding early warning measures are taken in a timely manner.

[0103] Step S5 includes:

[0104] S51, integrating the basic warning indicators, auxiliary warning indicators and environmental impact indicators of the deformation data set to obtain a comprehensive warning index, wherein the basic warning indicators include displacement, deformation rate and deformation acceleration, the auxiliary warning indicators include the main direction of deformation and the spatial autocorrelation index, and the environmental impact indicators include rainfall and temperature change;

[0105] S52, obtaining historical deformation data and warning records, constructing a training set and a verification set according to the historical deformation data and the warning records, establishing an original dam area deformation risk assessment model based on the comprehensive warning index using an integrated learning method, training and testing the original dam area deformation risk assessment model through the training set and the verification set to obtain a dam area deformation risk assessment model, identifying the deformation data set through the dam area deformation risk assessment model to obtain a dam area deformation risk level, and taking corresponding warning measures based on different dam area deformation risk levels.

[0106] Specifically, this embodiment integrates basic early warning indicators, auxiliary early warning indicators and environmental impact indicators to obtain a comprehensive early warning index that can comprehensively reflect the risk status of dam area deformation.

[0107] Based on basic early warning indicators such as displacement, deformation rate, deformation acceleration, as well as auxiliary indicators such as the main direction of deformation and spatial autocorrelation index, the deformation of the dam area can be evaluated from multiple dimensions.

[0108] The introduction of environmental impact indicators such as rainfall and temperature changes enables the risk assessment model to take into account the impact of the external environment on the deformation of the dam area, making the assessment results closer to the actual situation and improving the applicability of the model. By taking corresponding early warning measures for different risk levels, a dynamic response to the deformation of the dam area can be achieved.

[0109] Step S5 also includes:

[0110] The calculation formula of the comprehensive early warning index is:

[0111]

[0112] Among them, CWI is the comprehensive early warning index, w k is the weight of the kth early warning indicator, K is the total number of early warning indicators, and f k (x k ) is the membership function of the kth indicator, p is the norm parameter, ε is the environmental impact indicator magnification coefficient, v l is the weight of the lth environmental factor, e l is the standardized value of the lth environmental impact indicator, L is the total number of environmental impact indicators, and tanh(·) is the hyperbolic tangent activation function;

[0113] The calculation formula of the model evaluation function of the original dam area deformation risk assessment model is:

[0114]

[0115] Among them, Score is the model score of the original dam area deformation risk assessment model, β1, β2, and β3 are the weight coefficients of accuracy, recall, and precision, respectively, Accuracy, Recall, and Precision are accuracy, recall, and precision, respectively, Q is the number of training samples, δ is the time decay parameter, and exp(·) is the exponential function.

[0116] Specifically, this embodiment introduces multiple early warning indicators, including basic early warning indicators such as displacement, deformation rate, deformation acceleration, and auxiliary early warning indicators such as main deformation direction and spatial autocorrelation index, and combines environmental impact indicators, including rainfall and temperature changes, to achieve a comprehensive assessment of the deformation risk of the dam area. By dynamically adjusting the application weights and membership functions, the relative importance of different indicators in the comprehensive assessment is reflected.

[0117] By using the hyperbolic tangent activation function, the early warning indicators of different dimensions are processed to make them comparable in the comprehensive early warning index. In the original dam area deformation risk assessment model, a weighted combination of accuracy, recall and precision is adopted, and a time decay parameter and exponential function are introduced, so that the model can dynamically adjust its dependence on historical data.

[0118] By integrating multiple indicators, the comprehensive early warning index can more comprehensively reflect the risk status of dam area deformation. The introduction of time attenuation parameters and dynamic weight adjustment enables the risk assessment model to better adapt to the dynamic changes of dam area deformation. By taking corresponding early warning measures for different risk levels, more flexible early warning decisions can be made.

[0119] In a specific embodiment, the deformation risk of the dam area is divided into four levels: low risk, medium risk, high risk and extremely high risk.

[0120] Each level corresponds to a different deformation index threshold, including:

[0121] Low risk: displacement less than 5mm, deformation rate less than 1mm / month.

[0122] Medium risk: displacement is between 5mm and 10mm, and deformation rate is between 1mm and 3mm / month.

[0123] High risk: displacement between 10mm and 20mm, deformation rate between 3mm and 5mm / month.

[0124] Very high risk: displacement greater than 20 mm, deformation rate greater than 5 mm / month.

[0125] For low risks, regularly monitor and record deformation data, maintain observation of the dam area, ensure continuous updating of data, issue regular safety tips, and remind relevant personnel to pay attention to the normal operation of the dam area.

[0126] For medium risks, the monitoring frequency will be increased, detailed deformation data will be collected once a week, on-site inspections will be carried out, the structural safety of the dam will be assessed, local reinforcement will be carried out when necessary, risks will be reported to relevant management departments, and emergency plans will be formulated.

[0127] For high risks, the emergency response mechanism will be immediately activated, an expert team will be organized to conduct on-site assessments, a comprehensive structural inspection and assessment of the dam area will be carried out, reinforcement or repair will be carried out when necessary, entry of people around the dam area will be restricted to ensure safety, and warning information will be issued to the public to inform them of possible risks and safety measures.

[0128] For extremely high risks, immediately evacuate personnel from the dam area and surrounding areas to ensure personal safety, initiate comprehensive emergency response procedures, carry out emergency rescue and reinforcement work, set up a warning area, prohibit any personnel from entering, ensure safety, report the situation to the management department and relevant agencies, and coordinate resources for emergency response.

[0129] After implementing the early warning measures, continuously monitor the deformation of the dam area, evaluate the effectiveness of the early warning measures, and adjust the early warning level and corresponding measures in a timely manner based on the monitoring results to ensure safe management of the dam area.

[0130] See also Figure 2 The present invention also provides a dam area deformation early warning system based on a laser scanner, the system comprising:

[0131] A drone scanning module is used to select drones suitable for the complex environment of the dam area and equip the drones with laser scanners and Beidou positioning systems;

[0132] A path planning module is used to plan the monitoring flight path of the UAV based on the geographical data of the dam area to cover the entire dam area, continuously emit lasers through the laser scanner and receive reflected signals, and use the Beidou positioning system to record the scanning position information of the laser scanner in real time to obtain original three-dimensional point cloud data;

[0133] The dam area model module is used to pre-process the original three-dimensional point cloud data to obtain three-dimensional point cloud data, and integrate the three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system through a registration algorithm to obtain a three-dimensional model of the dam area;

[0134] An outlier elimination module is used to extract the original deformation data set of the dam area three-dimensional model at different time points, use statistical methods and preset thresholds to perform data consistency check on the original deformation data set, screen out the outliers in the original deformation data set, and obtain the deformation data set, wherein the deformation data set includes displacement, deformation rate, and deformation direction;

[0135] The deformation early warning module is used to construct a dam area deformation risk assessment model, identify the deformation data set through the dam area deformation risk assessment model, obtain the dam area deformation risk level, and take corresponding early warning measures based on different dam area deformation risk levels.

[0136] Specifically, the system of this embodiment realizes comprehensive monitoring of the complex dam area environment through the drone scanning module, overcomes the limitations of ground monitoring, and improves the monitoring coverage and efficiency. The dam area model module can pre-process and align the original three-dimensional point cloud data to build a high-precision three-dimensional model of the dam area. The outlier removal module effectively identifies and removes noise data through statistical analysis methods to ensure data quality and improve the accuracy of the analysis results. The deformation warning module constructs a risk assessment model based on historical data and environmental factors, which can identify the current deformation data and judge the risk level, take corresponding warning measures for different risk levels, and improve the safety of the system.

[0137] The present invention also discloses an electronic device, comprising: at least one processor, at least one memory communication interface and a bus: wherein the processor, memory and communication interface communicate with each other through the bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement a dam area deformation early warning method based on a laser scanner.

[0138] The present invention also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, wherein the computer instructions enable the computer to implement all or part of the steps of a dam area deformation early warning method based on a laser scanner according to an embodiment of the present invention. The storage medium includes: a U disk, a mobile hard disk, a read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk, and other media that can store program codes.

[0139] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A dam area deformation early warning method based on laser scanner, characterized in that: The following steps are involved: S1, selecting a UAV suitable for the complex environment of the dam area, and equipping the UAV with a laser scanner and a Beidou positioning system; S2, based on the geographical data of the dam area, planning the monitoring flight path of the UAV to cover the entire dam area, continuously emitting lasers through the laser scanner and receiving reflected signals, and using the Beidou positioning system to record the scanning position information of the laser scanner in real time to obtain original three-dimensional point cloud data; S3, preprocessing the original three-dimensional point cloud data to obtain three-dimensional point cloud data, integrating the three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system through a registration algorithm, and obtaining a three-dimensional model of the dam area; S4, extracting the original deformation data set of the dam area three-dimensional model at different time points, using a statistical method and a preset threshold to perform a data consistency check on the original deformation data set, filtering out abnormal values ​​in the original deformation data set, and obtaining a deformation data set, wherein the deformation data set includes a displacement, a deformation rate, and a deformation direction; S5, constructing a dam area deformation risk assessment model, identifying the deformation data set through the dam area deformation risk assessment model, obtaining the dam area deformation risk level, and taking corresponding early warning measures based on different dam area deformation risk levels.

2. A dam area deformation early warning method based on a laser scanner as claimed in claim 1, characterized in that: Step S2 includes: S21, based on the geographical data of the dam area, the path planning algorithm is used to generate the initial flight path of the UAV, the dam area is divided into several grid units, a topological map of the dam area is constructed based on the grid units, and the optimal path is searched on the topological map through the heuristic search algorithm to generate the initial flight path; S22, optimize and adjust the initial flight path according to the terrain characteristics and obstacle distribution of the dam area, identify the potential danger areas and blind spot areas in the initial flight path, adjust the flight altitude and flight angle of the initial flight path according to the potential danger areas and the blind spot areas to avoid the potential danger areas and cover the blind spot areas, and adjust the flight speed of the UAV and the scanning frequency of the laser scanner based on the endurance of the UAV and the effective scanning range of the laser scanner.

3. The dam area deformation early warning method based on laser scanner according to claim 1, characterized in that: Step S3 includes: S31, performing preliminary outlier removal on the original three-dimensional point cloud data by a statistical filtering method to obtain first three-dimensional point cloud data, removing noise points on the first three-dimensional point cloud data by a statistical outlier filtering algorithm to obtain second three-dimensional point cloud data, downsampling the second three-dimensional point cloud data based on a voxel grid filtering algorithm to obtain third three-dimensional point cloud data, removing outlier points of the third three-dimensional point cloud data by a radius outlier filtering algorithm to obtain fourth three-dimensional point cloud data, and smoothing the fourth three-dimensional point cloud data based on a Gaussian filter and a bilateral filtering algorithm to obtain preprocessed three-dimensional point cloud data; S32, select fixed reference scanning data as the reference coordinate system, select corresponding point pairs in the source point cloud and the target point cloud, calculate the optimal rigid body transformation of the corresponding point pairs to minimize the square distance error between the source point cloud and the target point cloud, and apply the optimal rigid body transformation to the source point cloud; repeatedly select corresponding point pairs to calculate the rigid body transformation and apply it until the preset number of iterations is reached, and integrate all three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system to form a three-dimensional model of the dam area.

4. The dam area deformation early warning method based on laser scanner according to claim 1, characterized in that: Step S4 includes: S41, adaptively meshing the three-dimensional model of the dam area at each time point, calculating the three-dimensional displacement vectors of corresponding grid nodes between adjacent time points, smoothing the three-dimensional displacement vectors using a dynamic trend smoothing method, calculating the rate of change of the three-dimensional displacement vectors using a multi-scale deformation rate method to obtain the deformation rate, and determining the deformation direction based on a principal component analysis method; S42, calculate the mean, standard deviation and skewness of the original deformation data set, dynamically set the outlier judgment threshold based on the data distribution characteristics, identify the potential outliers of the original deformation data set through an adaptive Z-score method, perform spatial clustering analysis on the potential outliers, identify the confirmed outliers of the original deformation data set, eliminate the confirmed outliers of the original deformation data set, and obtain the final deformation data set.

5. The dam area deformation early warning method based on laser scanner according to claim 4, characterized in that: Step S4 also includes: The calculation formula of the dynamic trend smoothing method is: S t =α·X t +(1-α)·S t-1 +γ·ΔX t ; Among them, S t is the displacement after smoothing, X t is the original displacement, α is the smoothing coefficient, S t-1 is the displacement of the previous time step, γ is the trend correction coefficient, ΔX t is the displacement change; The calculation formula of the multi-scale deformation rate method is: Where V is the deformation rate, D n and D n-1 are the displacements at the nth and n-1th time steps, respectively, t n and t n-1 D n and D n-1 The corresponding time step, w n is the weight coefficient of the nth time period, and N is the total number of time steps; The calculation formula of the adaptive Z-score method is: Where Z is the adaptive Z-score value, X is the original data in the original deformed dataset, median(X) is the median of the original data in the original deformed dataset, MAD is the absolute median difference, and τ is the anti-zero constant; The calculation formula of the spatial cluster analysis is: Among them, C is the spatial autocorrelation index, W i,j is the spatial weight matrix, M i and M j are the adaptive Z-score values ​​of point i and point j respectively, μM is the mean of M, σ M is the standard deviation of M, λ is the distance attenuation coefficient, d i,j is the spatial distance between point i and point j.

6. The dam area deformation early warning method based on laser scanner according to claim 1, characterized in that: Step S5 includes: S51, integrating the basic warning indicators, auxiliary warning indicators and environmental impact indicators of the deformation data set to obtain a comprehensive warning index, wherein the basic warning indicators include displacement, deformation rate and deformation acceleration, the auxiliary warning indicators include the main direction of deformation and the spatial autocorrelation index, and the environmental impact indicators include rainfall and temperature change; S52, obtaining historical deformation data and warning records, constructing a training set and a verification set according to the historical deformation data and the warning records, establishing an original dam area deformation risk assessment model based on the comprehensive warning index using an integrated learning method, training and testing the original dam area deformation risk assessment model through the training set and the verification set to obtain a dam area deformation risk assessment model, identifying the deformation data set through the dam area deformation risk assessment model to obtain a dam area deformation risk level, and taking corresponding warning measures based on different dam area deformation risk levels.

7. The dam area deformation early warning method based on laser scanner according to claim 6, characterized in that: Step S5 also includes: The calculation formula of the comprehensive early warning index is: Among them, CWI is the comprehensive early warning index, w k is the weight of the kth early warning indicator, K is the total number of early warning indicators, and f k (x k ) is the membership function of the kth indicator, p is the norm parameter, ε is the environmental impact indicator magnification coefficient, v l is the weight of the lth environmental factor, e l is the standardized value of the lth environmental impact indicator, L is the total number of environmental impact indicators, and tanh(·) is the hyperbolic tangent activation function; The calculation formula of the model evaluation function of the original dam area deformation risk assessment model is: Among them, Score is the model score of the original dam area deformation risk assessment model, β1, β2, and β3 are the weight coefficients of accuracy, recall, and precision, respectively, Accuracy, Recall, and Precision are accuracy, recall, and precision, respectively, Q is the number of training samples, δ is the time decay parameter, and exp(·) is the exponential function.

8. A dam area deformation early warning system based on laser scanner, characterized in that: The system comprises: A drone scanning module is used to select drones suitable for the complex environment of the dam area and equip the drones with laser scanners and Beidou positioning systems; A path planning module is used to plan the monitoring flight path of the UAV based on the geographical data of the dam area to cover the entire dam area, continuously emit lasers through the laser scanner and receive reflected signals, and use the Beidou positioning system to record the scanning position information of the laser scanner in real time to obtain original three-dimensional point cloud data; The dam area model module is used to pre-process the original three-dimensional point cloud data to obtain three-dimensional point cloud data, and integrate the three-dimensional point cloud data obtained by multiple scans of the laser scanner into a unified coordinate system through a registration algorithm to obtain a three-dimensional model of the dam area; An outlier elimination module is used to extract the original deformation data set of the dam area three-dimensional model at different time points, use statistical methods and preset thresholds to perform data consistency check on the original deformation data set, screen out the outliers in the original deformation data set, and obtain the deformation data set, wherein the deformation data set includes displacement, deformation rate, and deformation direction; The deformation early warning module is used to construct a dam area deformation risk assessment model, identify the deformation data set through the dam area deformation risk assessment model, obtain the dam area deformation risk level, and take corresponding early warning measures based on different dam area deformation risk levels.

9. An electronic device, characterized in that: include: at least one processor, at least one memory, a communication interface, and a bus; The processor, memory, and communication interface communicate with each other via the bus, the memory stores program instructions executable by the processor, and the processor calls the program instructions to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions enable a computer to implement the method according to any one of claims 1 to 7.

Images