A water-sand hysteresis mode discrimination method, device, equipment, medium and product

Abstract

The present application relates to the technical field of hydrology prediction, and discloses a water-sediment lag mode discrimination method, device, equipment, medium and product, comprising: acquiring current rainfall data collected by each rainfall station in a basin, and geographical data and basin structure data of the basin; calculating rainfall total amount, rainfall duration and maximum rainfall intensity, and initial rainfall gravity center and rainfall moving direction according to the current rainfall data and the geographical data; calculating the structural connectivity coefficient of the basin according to the basin structure data; inputting the rainfall total amount, rainfall duration, maximum rainfall intensity, initial rainfall gravity center, rainfall moving direction and structural connectivity coefficient into a lag mode judgment model to obtain a water-sediment lag mode judgment result; and bringing the rainfall data and geographical elevation data in the predicted basin into the pre-trained lag mode judgment model for mode judgment, thereby achieving the purpose of pre-judging the possible lag mode and water-sediment risk.

Description

Technical Field

[0001] This invention relates to the field of hydrological prediction technology, specifically to a method, apparatus, equipment, medium, and product for identifying water and sediment lag patterns. Background Technology

[0002] Accurately identifying the water and sediment lag pattern in a watershed is of vital practical significance for flood control safety, ecological protection, and water conservancy project management. Traditional understanding often focuses on the instantaneous impact of flood peak flow. However, the water and sediment lag effect reveals that the greatest risks of sediment transport and deposition often lurk during the receding stage after the flood peak. At this time, the drop in water level can easily lead to a lapse in vigilance, but the high-sediment-laden flow continues to erode the foundations of engineering projects and silt up river channels, creating a passive situation of "small floods causing major disasters."

[0003] Traditional discrimination methods are based on the "hysteresis loop method" of hydrological and sediment process lines during a single flood event at the watershed outlet. This method constructs hysteresis loops by plotting the asynchronous relationship between runoff and sediment concentration during a single flood event. The shape of the hysteresis loop is used as a basis for identifying water and sediment lag patterns. However, this method is essentially a post-event interpretation tool, only applicable after the flood event has occurred and outlet data has been collected. It cannot predict potential lag patterns and water and sediment risks based on the spatiotemporal characteristics of rainfall and the watershed's background structure during or before rainfall occurs. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, medium, and product for identifying water and sediment lag patterns, in order to solve the problem of being unable to predict potential lag patterns and water and sediment risks based on the spatiotemporal characteristics of rainfall and the baseline structure of the watershed when or before rainfall occurs.

[0005] In a first aspect, the present invention provides a method for determining water-sediment lag patterns, comprising:

[0006] Acquire current rainfall data collected by various rain gauge stations within the watershed, as well as geographical data and watershed structure data of the watershed;

[0007] The total rainfall, duration of rainfall, and maximum rainfall intensity, as well as the initial center of gravity of rainfall and direction of rainfall movement, are calculated based on current rainfall data and geographical data.

[0008] Calculate the structural connectivity coefficient of the watershed based on watershed structure data;

[0009] The total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient are input into the lag mode judgment model to obtain the water and sediment lag model judgment results.

[0010] By acquiring rainfall data measured by rain gauge stations in the watershed, along with geographical and structural data, we can obtain data related to water and sediment lag patterns. By inputting this data into a pre-trained lag pattern judgment model, we can determine the water and sediment lag patterns generated by the current rainfall. This solves the problem that traditional methods cannot predict water and sediment lag patterns. Furthermore, the method of acquiring this data is simpler and more convenient than the traditional method of acquiring water and sediment flow data at the watershed outlet.

[0011] In one optional implementation, the initial center of gravity of rainfall and the direction of rainfall movement for each rainfall period are calculated based on current rainfall data and geographic data, including:

[0012] The centroid coordinates of rainfall for each time period are calculated based on the total number of rain gauge stations in the basin, the coordinates of each rain gauge station, and the rainfall amount in the current rainfall data collected by each rain gauge station. The centroid coordinates of rainfall for the first time period are used as the initial centroid coordinates of rainfall.

[0013] The confluence path length for each time period is calculated based on the distance between the centroid coordinates of rainfall and the outlet coordinates of the watershed for each time period.

[0014] The direction of rainfall movement for each time period is calculated based on the difference in the confluence path length between adjacent time periods.

[0015] This paper presents a method to divide rainfall into different stages and calculate the centroid of rainfall and the direction of its movement for each stage based on measurements from rain gauge stations within the watershed. This method yields the total rainfall, rainfall duration, maximum rainfall intensity, initial centroid of rainfall, and the direction of rainfall movement for each stage. Dividing rainfall into different stages also refines the rainfall process, making the prediction results more accurate.

[0016] In one optional implementation, the geographic data includes a two-dimensional geographic planar map of the watershed, the watershed structure data includes slope data, upstream catchment area data, confluence path length data, and weighting factors, and the structural connectivity coefficient includes average structural connectivity data and coefficient of variation. The calculation of the watershed structural connectivity coefficient based on the watershed structure data includes:

[0017] The watershed is divided into multiple calculation cells according to the geographical two-dimensional plan view of the watershed;

[0018] The structural connectivity data of each calculation cell is calculated based on the slope data, upstream catchment area data, confluence path length data, and weighting factors.

[0019] Based on the structural connectivity data of each calculated cell, calculate the average structural connectivity data and coefficient of variation of the watershed.

[0020] A method is presented to divide a two-dimensional planar map into multiple computational cells and calculate the structural connectivity data of each cell. Based on the structural connectivity data of each cell, the average structural connectivity data and coefficient of variation required for the lag mode judgment model are calculated. In this process, the average structural connectivity data quantifies the runoff and sediment transport efficiency within the watershed, and the coefficient of variation measures the degree of unevenness in the spatial distribution of connectivity, thereby more accurately judging the water and sediment lag mode.

[0021] In one optional implementation, the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient are input into the lag mode judgment model to obtain the water and sediment lag model judgment result, which further includes:

[0022] Assign importance values ​​to and display for each data point, including total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient.

[0023] A method for assigning significance values ​​to each input value is presented, which enhances the transparency and reliability of the lag pattern judgment model.

[0024] In one optional implementation, the hysteresis pattern judgment model training method includes:

[0025] Acquire historical rainfall data, historical geographical data, historical watershed structure data, and historical water and sediment lag models corresponding to each historical rainfall event from various rain gauge stations within the watershed.

[0026] Historical rainfall characteristics are calculated based on historical rainfall data and historical geographic data. These characteristics include total historical rainfall, historical rainfall duration, historical maximum rainfall intensity, as well as the historical initial rainfall center of gravity and historical rainfall movement direction.

[0027] Calculate the historical structural connectivity coefficient of the watershed based on historical watershed structural data;

[0028] Training data is constructed based on the historical rainfall characteristics, historical structural connectivity coefficients, and historical water and sediment lag models of each historical rainfall event.

[0029] The initial model was trained using training data to obtain a water and sediment lag pattern judgment model.

[0030] By training the lag pattern judgment model with the same type of data from historical rainfall data as in the actual forecasting process, the accuracy of the forecast model is guaranteed and its reliability is higher.

[0031] In one alternative implementation, the lag pattern judgment model uses a support vector machine model for judgment and classification.

[0032] This embodiment provides a method for identifying water and sediment lag patterns, which uses a support vector machine model for judgment. This method can reduce the impact of individual erroneous judgments, making the judgment results of the water and sediment lag model more scientific, reasonable and credible.

[0033] Secondly, the present invention provides a water-sediment lag mode discrimination device, comprising:

[0034] The data acquisition module is used to acquire current rainfall data collected by various rain gauge stations within the watershed, as well as geographical data and watershed structure data of the watershed;

[0035] The rainfall calculation module is used to calculate the total rainfall, rainfall duration, and maximum rainfall intensity, as well as the initial rainfall centroid and rainfall movement direction, based on the current rainfall data and geographic data.

[0036] The connectivity calculation module is used to calculate the structural connectivity coefficient of a watershed based on watershed structure data.

[0037] The pattern discrimination module is used to input the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient into the lag pattern judgment model to obtain the water and sediment lag model judgment result.

[0038] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform a water and sediment lag mode discrimination method as described in the first aspect or any corresponding embodiment.

[0039] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute a water-sediment lag pattern discrimination method as described in the first aspect or any corresponding embodiment thereof.

[0040] Fifthly, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute a water and sediment lag pattern discrimination method according to the first aspect above or any corresponding embodiment thereof. Attached Figure Description

[0041] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention;

[0043] Figure 2 This is a schematic diagram of the first process of a water-sediment lag mode discrimination method according to an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of the second process of a water-sediment lag mode discrimination method according to an embodiment of the present invention;

[0045] Figure 4 This is a schematic diagram of the third process of a water-sediment lag mode discrimination method according to an embodiment of the present invention;

[0046] Figure 5 This is a schematic diagram of the support vector machine classification ROC curve and confusion matrix constructed based on rainfall characteristics and structural connectivity according to an embodiment of the present invention.

[0047] Figure 6 This is a structural block diagram of a water-sediment lag mode discrimination device according to an embodiment of the present invention;

[0048] Figure 7 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0051] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0052] As an optional application scenario of this invention, such as Figure 1As shown, the water and sediment lag pattern discrimination system may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0053] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.

[0054] Traditional methods for identifying water and sediment lag patterns are based on the "hysteresis loop method" of hydrological and sediment process lines during a single flood event at the watershed outlet. This method constructs hysteresis loops by plotting the asynchronous relationship between runoff and sediment concentration during a single flood event. The shape of the hysteresis loop is used as a basis for identifying water and sediment lag patterns. However, this method can only be used after the flood event has occurred and outlet data has been collected. It cannot predict potential lag patterns and water and sediment risks during or before rainfall occurs. This invention provides a method for identifying water and sediment lag patterns. By establishing a predictive model and combining the spatiotemporal characteristics of rainfall with the watershed's background structure, it aims to predict potential lag patterns and water and sediment risks.

[0055] According to an embodiment of the present invention, a method for determining water and sediment lag patterns is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0056] This embodiment provides a method for determining water and sediment lag patterns, which can be used on mobile terminals such as mobile phones and tablets. Figure 2 This is a flowchart of a water-sediment lag pattern discrimination method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps:

[0057] Step S201: Obtain the current rainfall data collected by each rain gauge station within the watershed, as well as the geographical data and watershed structure data of the watershed.

[0058] The current rainfall data is used to present the specific conditions of the rainfall collected by rain gauge stations within the basin. Geographic data refers to geographic elevation data. Watershed structure data can be used to comprehensively characterize runoff and sediment transport capacity.

[0059] For example, current rainfall data can be obtained by measuring rainfall at various rain gauge stations within the watershed, while geographic data can be obtained from a digital elevation model.

[0060] Step S202: Calculate the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall center of gravity, and rainfall movement direction based on the current rainfall data and geographical data.

[0061] Total rainfall refers to the average of the rainfall recorded by all rainfall stations within a watershed from the start to the end of a single rainfall event. Total rainfall determines the potential scale of total runoff and total sediment transport. The larger the total rainfall, the larger the potential flood peak and sediment peak. Rainfall duration reflects the persistence of rainfall and is the total time from the start to the end of a flood event. Short-duration rainfall results in short transport paths for eroded soil, or soil that fails to reach the watershed rivers, meaning the sediment response may lag behind the water flow. Long-duration rainfall may have relatively uniform intensity, leading to fully saturated soil and a closer synchronization between erosion and sediment transport processes and the water flow process. Maximum rainfall intensity is the highest rainfall recorded per unit time by rainfall stations within the watershed during a single flood event. High rainfall intensity can rapidly splash soil and form surface runoff, which is the main driving force for sediment initiation. The initial rainfall centroid refers to the location of the rainfall centroid corresponding to the first time period when the current rainfall is divided into multiple time periods in the subsequent calculation process. The rainfall movement direction refers to the rainfall movement direction calculated based on the location of the rainfall centroid in different time periods. The initial rainfall centroid and the rainfall movement direction determine the arrival time sequence and superposition method of runoff and sediment production at the outlet section of different parts of the watershed.

[0062] Step S203: Calculate the structural connectivity coefficient of the watershed based on the watershed structure data.

[0063] Structural connectivity coefficient is used to characterize runoff and sediment transport capacity. The structural connectivity coefficient of a watershed directly affects the probability of eroded material being transported to the watershed outlet or water body. In areas with high connectivity, the surface structural barrier effect is weakened, and the hydraulic connection between the slope and the river channel is closer, resulting in a significant increase in sediment transport efficiency.

[0064] Step S204: Input the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient into the lag mode judgment model to obtain the water and sediment lag model judgment result.

[0065] The lag pattern judgment model refers to a judgment model trained before forecasting, used to judge the lag pattern of water and sediment under the current rainfall and geographical environment by combining the above calculated values. The water and sediment risk of the current lag pattern can be judged based on the results of the lag pattern judgment model.

[0066] Common water and sediment lag patterns can be divided into clockwise lag loops where the sediment peak lags significantly behind the flood peak, counterclockwise lag loops where the sediment peak leads or is basically synchronized with the flood peak, or figure-eight lag loops in complex cases.

[0067] Clockwise lag is a typical phenomenon where the peak sediment concentration occurs before the peak runoff. During flood recession, the rate of decrease in sediment concentration is faster than the rate of decrease in runoff, reflecting that the sediment mainly originates from rapid near-surface erosion or the scouring of sediment accumulated in the river channel in the early stages. Counterclockwise lag is the opposite of clockwise lag, characterized by the peak runoff occurring before the peak sediment concentration, or when the peak runoff and peak sediment concentration coincide, the sediment concentration along the decline path remains high, indicating that sediment requires a longer time to reach the watershed outlet, suggesting a continuous supply of sediment at the far end. A complex figure-eight loop is an alternating clockwise / counterclockwise loop in flood events with multiple peaks, indicating a mixed contribution of sediment from multiple sources. The figure-eight lag combines clockwise and counterclockwise lags, caused by multiple peaks in runoff and sediment concentration, and can be specifically divided into a combination of counterclockwise circulation at low flow and clockwise circulation at high flow, or a combination of clockwise circulation at low flow and counterclockwise circulation at high flow.

[0068] The initial rainfall centroid and rainfall movement direction indicate which areas the rainfall primarily affects. The total rainfall amount, duration, and maximum rainfall intensity determine the intensity and time course of runoff and sediment production. The structural connectivity coefficient describes the sediment transport efficiency determined by factors such as topography, soil, vegetation, and land use within the watershed. Therefore, based on the above data, lag patterns can be identified and predicted.

[0069] For example, in a clockwise lagging pattern, when the initial rainfall center of gravity is downstream and moves away from the outlet, runoff forms downstream first, sediment is transported rapidly over short distances, and sediment reaches its peak value first. As rainfall gradually moves upstream, upstream runoff continuously converges towards the downstream outlet, resulting in a runoff peak at the watershed outlet. Meanwhile, upstream sediment is deposited due to long-distance transport, leading to insufficient sediment supply and a decrease in sediment transport rate, forming a clockwise ring. When the maximum rainfall intensity is >30 mm / h and the watershed structure has high connectivity variation, the runoff has strong sediment-carrying capacity, resulting in severe erosion and sediment concentration reaching its peak earlier than the runoff. At high maximum rainfall intensity, high-intensity raindrop splashing will strip away a large amount of loose surface sediment before the runoff is fully formed. During the rising flow stage, sediment is rapidly carried away, and the sediment transport rate increases sharply, exhibiting a clockwise ring.

[0070] For example, when the initial rainfall center is shifted upstream and closer to the outlet, the flow rate reaches its peak first. During the flow rate decline phase, residual sediment from upstream continues to be transported, with the sediment transport rate decreasing more slowly than the flow rate, forming a counter-clockwise loop. During the flow rate decline phase, previously stripped sediment continues to be transported in the runoff, with the sediment transport rate decreasing more slowly, also exhibiting a counter-clockwise loop. When structural connectivity is low, sediment transport paths are obstructed. During the flow rate rise phase, vegetation or terraces intercept a large amount of sediment, with the sediment transport rate lagging behind the flow rate. During the flow rate decline phase, residual sediment in the gullies continues to be eroded, with the sediment transport rate decreasing more slowly than the flow rate, corresponding to a counter-clockwise loop.

[0071] For example, when the rainfall intensity is low at the beginning and high at the end and there are differences in structural connectivity, the low rainfall intensity in the early stage of rainfall results in insufficient sediment supply and slow growth of sediment transport rate. The curve rotates clockwise at first. In the later stage of rainfall, the runoff shear force increases suddenly, and deep sediment is eroded in large quantities. The growth rate of sediment transport rate exceeds the growth rate of flow rate, and the curve rotates counterclockwise, eventually forming a figure-eight lag.

[0072] This embodiment provides a method for identifying water and sediment lag patterns. By acquiring rainfall data measured by rain gauge stations in the watershed, as well as geographical and structural data of the watershed, calculations can be performed to obtain data related to water and sediment lag patterns. This data is then fed into a pre-trained lag pattern identification model to determine the water and sediment lag patterns generated by the current rainfall. This solves the problem that traditional methods cannot predict water and sediment lag patterns. Furthermore, the method of acquiring the above data is simpler and more convenient than the traditional method of acquiring water and sediment flow data at the watershed outlet.

[0073] This embodiment provides a method for determining water and sediment lag patterns. Figure 3 This is a flowchart of a water-sediment lag pattern discrimination method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps:

[0074] Step S301: Obtain current rainfall data collected by each rain gauge station within the watershed, as well as geographical data and watershed structure data. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0075] Step S302: Calculate the total rainfall, rainfall duration, and maximum rainfall intensity, as well as the initial rainfall center of gravity and rainfall movement direction, based on the current rainfall data and geographical data.

[0076] Specifically, step S302 includes:

[0077] Step S3021: Calculate the centroid coordinates of rainfall for each time period based on the total number of rain gauge stations in the basin, the coordinates of each rain gauge station, and the rainfall amount in the current rainfall data collected by each rain gauge station. Use the centroid coordinates of rainfall for the first time period as the initial centroid coordinates of rainfall.

[0078] Each time period refers to dividing rainfall into different time periods, which can be used to distinguish different stages of a rainfall event. The centroid of rainfall is an important indicator describing the concentration of rainfall spatial distribution. It is the weighted average location of the spatial distribution of rainfall within the watershed. Its calculation is based on the geographical coordinates of each rain gauge station and its rainfall. The centroid coordinates of rainfall refer to the coordinates of the centroid of rainfall in the current two-dimensional planar map of the watershed. The centroid coordinates of rainfall in each time period are calculated separately, and the centroid coordinates of rainfall in the first time period are used as the initial centroid coordinates of rainfall.

[0079] For example, the centroid coordinates of rainfall at each stage can be calculated using the following formula:

[0080]

[0081] in, For the first Rainfall at each rain gauge station; For the first Geographic coordinates of each rainfall station; The total number of rainfall stations within the watershed. The coordinates of the centroid of rainfall, Let x be the coordinate of the centroid of the rainfall on the x-axis. The coordinates of the centroid of the rainfall on the y-axis.

[0082] Step S3022: Calculate the confluence path length for each time period based on the distance between the centroid coordinates of rainfall and the outlet coordinates of the watershed for each time period.

[0083] Based on the coordinates of the centroid of rainfall and the outlet coordinates of the watershed for each time period, combined with the current watershed confluence path length map, the confluence path length for each time period can be obtained. The confluence path length refers to the path length of precipitation from the current centroid of rainfall to the outlet of the watershed.

[0084] For example, the confluence path length for each time period can be calculated using the following formula:

[0085]

[0086] in, This represents the length of the confluence path in segment j. This is a function for calculating the length of the confluence path. Let x be the coordinate of the centroid of the rainfall on the x-axis. Let be the coordinate of the centroid of the rainfall along the y-axis. The coordinates of the watershed outlet.

[0087] Step S3023: Calculate the direction of rainfall movement for each time period based on the difference in the confluence path length between adjacent time periods.

[0088] The direction of rainfall movement refers to the description of rainfall moving in a certain direction. It is used as input data for the lag pattern judgment model and is calculated based on the confluence path length of each adjacent time period.

[0089] For example, the direction of rainfall movement for each time period can be calculated using the following formula:

[0090]

[0091] in, Represents the length of the confluence path in segment j. Represents the direction of the movement of the center of gravity of rainfall. This indicates an increase in the length of the confluence path and that the center of gravity of rainfall is farther away from the watershed outlet; This indicates that the confluence path length is reduced, and the center of gravity of rainfall is closer to the watershed outlet. As an example, j can be 1, 2, 3, or 4.

[0092] Step S303: Calculate the structural connectivity coefficient of the watershed based on the watershed structure data. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0093] Step S304: Input the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient into the lag mode judgment model to obtain the water and sediment lag model judgment result. For details, please refer to... Figure 2 Step S204 of the illustrated embodiment will not be described again here.

[0094] This embodiment provides a method for identifying water and sediment lag patterns. It divides rainfall into different stages and calculates the center of gravity and direction of movement of the center of gravity of rainfall in each stage based on the measurement values ​​of rain gauge stations in the watershed. This method is used to obtain the total rainfall, rainfall duration, maximum rainfall intensity, initial center of gravity of rainfall, and direction of movement of rainfall in each stage. Dividing rainfall into different stages also refines the rainfall process, making the prediction results more accurate.

[0095] This embodiment provides a method for determining water and sediment lag patterns. Figure 4 This is a flowchart of a water-sediment lag pattern discrimination method according to an embodiment of the present invention, such as... Figure 4 As shown, the process includes the following steps:

[0096] Step S401: Obtain current rainfall data collected by each rain gauge station within the watershed, as well as the watershed's geographical data and watershed structure data. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0097] Step S402: Calculate the total rainfall, duration of rainfall, and maximum rainfall intensity, as well as the initial rainfall centroid and direction of rainfall movement, based on the current rainfall data and geographic data. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0098] Step S403: Calculate the structural connectivity coefficient of the watershed based on the watershed structure data.

[0099] Specifically, step S403 includes:

[0100] Step S4031: Divide the watershed into multiple calculation cells according to the geographical two-dimensional plan of the watershed.

[0101] A two-dimensional geographic map is a two-dimensional map used to describe the topographic features within a watershed. Dividing the two-dimensional map into multiple calculation cells facilitates the subsequent calculation of structural connectivity data in each cell. The more cells used, the more accurate the calculation results.

[0102] Step S4032: Calculate the structural connectivity data of each calculation cell based on the slope data, upstream catchment area data, confluence path length data, and weighting factors.

[0103] Weighting factors reflect the hindering effect of surface characteristics on runoff and sediment transport. They are determined based on factors such as vegetation cover, soil roughness, and land management. For example, high vegetation cover can block raindrops and slow down water flow; rougher soil increases resistance to water passage; and constructing steps or fences on the surface can intercept water flow, thus affecting the ease of water and sediment transport. Slope data in a cell refers to the slope data contained within the geographic elevation data of that cell, while upstream catchment area data refers to the total area of ​​the catchment area within the watershed.

[0104] For example, the structural connectivity data of each cell can be calculated using the following formula:

[0105]

[0106] in, The upslope component represents the probability of sediment transport from the upstream unit to the downstream unit. The downhill component represents the resistance that sediment encounters along its confluence path to reach the nearest "sink" (such as a river channel, reservoir, or outlet). It's the slope. It is the area of ​​the upstream catchment area. It is the length of the downstream confluence path of this unit. It is a weighting factor for a dimensionless quantity. This represents the structural connectivity data for the kth computational unit.

[0107] Step S4033: Calculate the average structural connectivity data and coefficient of variation of the watershed based on the structural connectivity data of each calculation cell.

[0108] The average structural connectivity data and coefficient of variation of the watershed are calculated based on the structural connectivity data of each cell, and used as input data for the subsequent lag pattern judgment model. The coefficient of variation is used to describe the degree of dispersion of the structural connectivity data in each cell.

[0109] Step S404: Input the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient into the lag mode judgment model to obtain the water and sediment lag model judgment result. For details, please refer to... Figure 2 Step S204 of the illustrated embodiment will not be described again here.

[0110] This embodiment provides a method for identifying water and sediment lag patterns. It divides a two-dimensional planar map into multiple calculation cells and calculates the structural connectivity data of each cell. Based on the structural connectivity data of each cell, it calculates the average structural connectivity data and coefficient of variation required for the lag pattern identification model. In this process, the average structural connectivity data quantifies the runoff and sediment transport efficiency within the watershed, and the coefficient of variation measures the degree of unevenness in the spatial distribution of connectivity, thereby more accurately identifying water and sediment lag patterns.

[0111] In an optional embodiment, the water and sediment lag model judgment result obtained in step S404 above further includes: assigning and displaying importance values ​​for each data point, such as total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient.

[0112] Importance values ​​are used to reflect the degree of influence of each data point on the prediction results.

[0113] For example, Shapley additive interpretation can be used to conduct in-depth analysis of the model's prediction results in order to obtain the importance values ​​of each data point.

[0114] This embodiment provides a method for identifying water and sediment lag patterns, and gives a method for assigning importance values ​​to each input value, thereby enhancing the transparency and reliability of the lag pattern identification model.

[0115] In an alternative embodiment, Figure 5 This is a schematic diagram of the ROC curve and confusion matrix of the support vector machine classification model based on rainfall characteristics and structural connectivity, according to an embodiment of the present invention. Figure 5The diagram shows the accuracy of this water and sediment lag model for judging three lag models. In step S404, the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient are input into the lag model to obtain the water and sediment lag model judgment results. The training method for the lag model includes:

[0116] Step a1: Obtain historical rainfall data, historical geographical data, historical watershed structure data, and historical water and sediment lag models corresponding to each historical rainfall event collected by each rain gauge station within the watershed.

[0117] Historical rainfall data refers to historical rainfall data collected by various rain gauges within the current watershed. Historical geographic data refers to the topographic data of the watershed over historical periods. Historical watershed structure data is used to characterize runoff and sediment transport capacity over historical periods. Historical water and sediment lag models corresponding to each historical rainfall event refer to the patterns of water and sediment lag under historical rainfall. The above data are used for calculations in subsequent processes or as test and validation sets to train lag pattern judgment models.

[0118] Step a2: Calculate the historical rainfall characteristics based on historical rainfall data and historical geographical data. The historical rainfall characteristics include the total historical rainfall, the duration of historical rainfall, the maximum historical rainfall intensity, the center of gravity of the initial historical rainfall, and the direction of historical rainfall movement.

[0119] Historical rainfall characteristics are used as training data for the subsequent training of the lag pattern judgment model. The total historical rainfall, historical rainfall duration, and historical maximum rainfall intensity are calculated using historical data from rainfall stations.

[0120] For example, the calculation method for historical initial rainfall centroid and historical rainfall movement direction data is the same as the calculation method for initial rainfall centroid and rainfall movement direction.

[0121] Step a3: Calculate the historical structural connectivity coefficient of the watershed based on historical watershed structural data.

[0122] The historical structural connectivity coefficients are also derived from the training data of the training lag pattern judgment model. They are calculated using historical watershed structural data.

[0123] For example, the historical structural connectivity coefficient can be calculated in the same way as the structural connectivity coefficient.

[0124] Step a4: Construct training data based on the historical rainfall characteristics, historical structural connectivity coefficients, and historical water and sediment lag models of each historical rainfall event.

[0125] The training data is used to train the lag pattern judgment model.

[0126] For example, 70% of the data can be divided into a test set and 30% into a validation set. A support vector machine model can be used to perform multiple parameter optimizations to compare model performance and output the model classification accuracy and confusion matrix.

[0127] Step a5: Train the initial model using the training data to obtain the hysteresis pattern judgment model.

[0128] The initial model refers to the lag pattern judgment model that has not been trained on a training set. The initial lag pattern judgment model is trained using the method described above to obtain the trained lag pattern judgment model. A validation set can also be used to validate the trained lag pattern judgment model.

[0129] As an example, such as Figure 5 As shown, the model evaluation results show that the area under the ROC curve (AUC) is 0.98 clockwise, 0.99 counterclockwise, and 0.93 for the combined figure-eight curve. The AUC value represents the model's accuracy in judging the three categories of water and sediment lag patterns. An AUC of 0.5 indicates that the model has no discriminatory power, and its predictive performance is equivalent to random guessing. An AUC of 1.0 indicates that the model is a perfect classifier, capable of separating positive and negative samples without error. This data indicates high discrimination between categories and a low probability of misclassification. A high AUC value indicates that the model has strong robustness and can cope with the problem of imbalanced sample sizes between different categories. Figure 6The confusion matrix shown and the performance evaluation results of the support vector machine model for watershed sediment lag pattern discrimination, as shown in the table below, verify the performance indicators of each performance index. The overall classification accuracy of the model reaches 85.96%, with the clockwise type showing the best identification efficiency. The sensitivity (1.00) and negative prediction rate (1.00) both reach the theoretical extreme values, indicating that the model can completely capture positive samples of this type of event without misclassification or missed detection. However, the positive prediction rate drops to 0.78 due to the influence of the category distribution, reflecting a deviation between the predicted distribution (0.47) and the actual category distribution (0.37), which may be related to the tendency of the dynamic characteristics of rapid response events from near-end sediment sources to be overgeneralized. The counterclockwise type has the highest balanced accuracy (0.93). The balance between sensitivity (0.92) and specificity (0.94) indicates that the model can effectively distinguish distant transport processes from other patterns. Its high positive prediction rate, detection rate (0.40), and matching degree of the predicted distribution (0.44) further verify the rationality of the selection of influencing factors. Due to training bias caused by insufficient sample size, the composite figure-eight pattern exhibits relatively low sensitivity (0.45) and balanced accuracy (0.73). However, its high specificity and negative prediction rate suggest that while the model struggles to actively identify such events, it can effectively exclude samples from non-composite figure-eight lag patterns, providing a reliable threshold for extreme event warnings. Model evaluation results confirm that the classification framework constructed based on the spatiotemporal heterogeneity characteristics of rainfall and structural connectivity indices can effectively analyze the causes of water and sediment lag patterns, making linearly inseparable problems linearly separable in high-dimensional space. The performance evaluation of the support vector machine model for watershed water and sediment lag pattern discrimination is shown below:

[0130]

[0131] This embodiment provides a method for identifying water and sediment lag patterns. The lag pattern identification model is trained using the same type of data from historical rainfall data and the actual prediction process, which ensures the accuracy of the prediction model and has higher reliability.

[0132] In an optional embodiment, the above-mentioned lag pattern judgment model uses a support vector machine model for judgment and classification.

[0133] In the process of making relevant judgments, three binary classification models based on support vector machines can be used to judge and classify the three water and sediment lag patterns. Multiple classification results can be integrated by combining the majority voting mechanism. The judgment result of the water and sediment lag model can be finally determined according to the direction tended by the majority judgment, so as to improve the accuracy and reliability of the final judgment.

[0134] This embodiment provides a method for identifying water and sediment lag patterns, which uses a support vector machine model for judgment. This method can reduce the impact of individual erroneous judgments, making the judgment results of the water and sediment lag model more scientific, reasonable and credible.

[0135] This embodiment also provides a water-sediment lag mode discrimination device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as already described. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0136] This embodiment provides a water-sediment lag pattern discrimination device, such as Figure 6 As shown, it includes:

[0137] The data acquisition module 501 is used to acquire current rainfall data collected by various rain gauge stations within the watershed, as well as geographical data and watershed structure data of the watershed.

[0138] The rainfall calculation module 502 is used to calculate the total rainfall, rainfall duration, and maximum rainfall intensity, as well as the initial rainfall center of gravity and rainfall movement direction, based on the current rainfall data and geographical data.

[0139] The connectivity calculation module 503 is used to calculate the structural connectivity coefficient of the watershed based on the watershed structure data.

[0140] The pattern discrimination module 504 is used to input the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction and structural connectivity coefficient into the lag pattern judgment model to obtain the water and sediment lag model judgment result.

[0141] The water and sediment lag mode device provided in this embodiment of the invention can execute a water and sediment lag mode method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method execution. Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments, and will not be repeated here.

[0142] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0143] The following is a detailed reference. Figure 7This diagram illustrates a suitable structural design for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 601, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 602 or a program loaded from memory 608 into random access memory (RAM) 603. RAM 603 also stores various programs and data required for the operation of the electronic device. The processor 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0144] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0145] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a memory 608, or installed from a ROM 602. When the computer program is executed by the processor 601, it performs the functions defined in the water-sediment lag mode method of an embodiment of the present invention.

[0146] Figure 7 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0147] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium after being downloaded via a network. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, it implements the water-sand lag mode method shown in the above embodiments.

[0148] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0149] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for determining water-sediment lag patterns, characterized in that, The method includes: Acquire current rainfall data collected by each rain gauge station within the watershed, as well as the geographical data and watershed structure data of the watershed; The total rainfall, duration of rainfall, and maximum rainfall intensity, as well as the initial center of gravity of rainfall and direction of rainfall movement, are calculated based on the current rainfall data and geographical data. Calculate the structural connectivity coefficient of the watershed based on the watershed structure data; The total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient are input into the lag mode judgment model to obtain the water and sediment lag model judgment result. The geographic data includes a two-dimensional geographic planar map of the watershed; the watershed structure data includes slope data, upstream catchment area data, confluence path length data, and weighting factors; the structural connectivity coefficient includes average structural connectivity data and a coefficient of variation; and the calculation of the watershed structural connectivity coefficient based on the watershed structure data includes: The watershed is divided into multiple calculation cells according to the geographical two-dimensional plan view of the watershed; The structural connectivity data of each calculation cell is calculated based on the slope data, upstream catchment area data, confluence path length data, and weighting factors. Based on the structural connectivity data of each calculated cell, the average structural connectivity data and coefficient of variation of the watershed are calculated.

2. The method according to claim 1, characterized in that, Based on the current rainfall data and geographical data, the initial rainfall centroid and rainfall movement direction for each rainfall period are calculated, including: The centroid coordinates of rainfall for each time period are calculated based on the total number of rain gauge stations in the basin, the coordinates of each rain gauge station, and the rainfall amount in the current rainfall data collected by each rain gauge station. The centroid coordinates of rainfall for the first time period are used as the initial centroid coordinates of rainfall. The confluence path length for each time period is calculated based on the distance between the centroid coordinates of rainfall in each time period and the outlet coordinates of the watershed. The direction of rainfall movement for each time period is calculated based on the difference in the confluence path length between adjacent time periods.

3. The method according to claim 1, characterized in that, The total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient are input into the lag mode judgment model to obtain the water and sediment lag model judgment result, which also includes: The importance values ​​of the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction, and structural connectivity coefficient are assigned and displayed.

4. The method according to claim 1, characterized in that, Training methods for hysteresis pattern recognition models include: The historical rainfall data of each historical rainfall event collected by each rain gauge station in the basin, the historical geographical data of the basin, the historical watershed structure data, and the historical water and sediment lag model corresponding to each historical rainfall event are obtained. Historical rainfall characteristics are calculated based on the historical rainfall data and historical geographic data. These characteristics include the total historical rainfall, the duration of historical rainfall, the maximum historical rainfall intensity, the initial historical rainfall center of gravity, and the direction of historical rainfall movement. Calculate the historical structural connectivity coefficient of the watershed based on the historical watershed structural data; Training data is constructed based on the historical rainfall characteristics, historical structural connectivity coefficients, and historical water and sediment lag models of each historical rainfall event. The initial model is trained using the training data to obtain the lag pattern judgment model.

5. The method according to claim 4, characterized in that, The lag pattern judgment model uses a support vector machine model for judgment and classification.

6. A water-sediment lag mode discrimination device, characterized in that, The device includes: The data acquisition module is used to acquire current rainfall data collected by each rain gauge station within the watershed, as well as geographical data and watershed structure data of the watershed. The rainfall calculation module is used to calculate the total rainfall, rainfall duration, and maximum rainfall intensity, as well as the initial rainfall center of gravity and rainfall movement direction, based on the current rainfall data and geographical data. A connectivity calculation module is used to calculate the structural connectivity coefficient of the watershed based on the watershed structure data. The pattern discrimination module is used to input the total rainfall, rainfall duration, maximum rainfall intensity, initial rainfall centroid, rainfall movement direction and structural connectivity coefficient into the lag pattern judgment model to obtain the water and sediment lag model judgment result. The geographic data includes a two-dimensional geographic planar map of the watershed; the watershed structure data includes slope data, upstream catchment area data, confluence path length data, and weighting factors; the structural connectivity coefficient includes average structural connectivity data and a coefficient of variation; and the calculation of the watershed structural connectivity coefficient based on the watershed structure data includes: The watershed is divided into multiple calculation cells according to the geographical two-dimensional plan view of the watershed; The structural connectivity data of each calculation cell is calculated based on the slope data, upstream catchment area data, confluence path length data, and weighting factors. Based on the structural connectivity data of each calculated cell, the average structural connectivity data and coefficient of variation of the watershed are calculated.

7. An electronic device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform a water-sediment lag pattern discrimination method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute a water-sediment lag pattern discrimination method according to any one of claims 1 to 5.

9. A computer program product, characterized in that, Includes computer instructions, which are used to cause a computer to execute a water-sediment lag pattern discrimination method according to any one of claims 1 to 5.

Images