Hydrological data intelligent analysis method and system

Abstract

The invention relates to the technical field of data analysis, in particular to an intelligent hydrological data analysis method and system, and the method comprises the following steps: collecting water level, flow velocity and precipitation data, calculating the change rate of adjacent moments, marking hydrological parameter abrupt change points, carrying out the statistics of time intervals, predicting the duration of extreme events, and obtaining the distribution characteristics of the hydrological abrupt change points. According to the method, by analyzing the time interval and the duration time of the data points, the distribution characteristics of hydrological abrupt change can be mastered in the early stage, response can be made in advance, the propagation sequence and the diffusion rate of the abrupt change points in different river reaches can be analyzed, and accurate division of the influence range and the risk level of hydrological events is enhanced; effective management of hydrological event space risks is achieved, understanding of hydrological key influence factors is deepened by screening key variables and evaluating the evolution relation of the variables along with time, accuracy of short-term and long-term hydrological risk prediction is enhanced, and a scientific basis is provided for dynamic early warning of hydrological abnormal events.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and particularly to an intelligent analysis method and system for hydrological data. Background Art

[0002] The technical field of data analysis is used to extract valuable information and insights from raw data. This field relies on statistics, algorithm development, and machine learning techniques to process and analyze data, aiming to support the decision-making process, optimize business operations, and predict future events. Data analysis technology is widely applied in multiple fields such as business, scientific research, medical care, and environmental monitoring. The core content covers aspects such as data cleaning, data integration, data mining, and data visualization. Through a systematic analysis process, data analysis technology can process large datasets and extract key indicators and patterns.

[0003] Among them, the intelligent analysis method and system for hydrological data refer to technical solutions specifically designed to process and analyze hydrological environment data. The technical matters targeted by this patent theme cover the collection, processing, and analysis of hydrological data, mainly achieved through the comprehensive analysis of data such as water level, flow rate, and precipitation. The solution includes data collection devices, data storage and analysis systems, and software application programs for interpreting and processing data. This system integrates various hydrological data sources, uses specific processing techniques to ensure the accuracy and practicality of data, and supports water resource management and prediction of hydrological events.

[0004] In actual operation of the prior art, although it can process and analyze a large amount of hydrological data, it lacks in-depth analysis of the characteristics of mutation points and a systematic risk assessment method, which limits its effectiveness in predicting hydrological events and managing risks. For example, it fails to make full use of key information such as the distribution characteristics and diffusion rate of mutation points, resulting in a less timely or accurate response to hydrological events. The prior art fails to achieve differential management of different river sections and time periods during processing, resulting in the inability to effectively distinguish various hydrological risk levels, and insufficient implementation of early warnings and response measures for major hydrological events in practical applications, affecting the overall efficiency and safety of water resource management. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose an intelligent analysis method and system for hydrological data.

[0006] To achieve the above purpose, the present invention adopts the following technical solution. An intelligent analysis method for hydrological data includes the following steps: S1: Collect water level, flow velocity, and precipitation data, calculate the change rate at adjacent moments, mark the mutation points of hydrological parameters, count the time interval, predict the duration of extreme events, and obtain the distribution characteristics of hydrological mutation points; S2: Based on the distribution characteristics of the hydrological mutation points, analyze the distribution density of the mutation points in different hydrological basins, count the propagation order of the mutation points between the upstream, middle, and downstream river sections, analyze the diffusion rate of the mutation points, identify the influence range of the mutation points, divide the spatial risk levels of hydrological events, and obtain the spatial diffusion trend of the hydrological mutation points; S3: Based on the spatial diffusion trend of the hydrological mutation points, screen key variables, analyze the change trends of the key variables in different time periods, evaluate the evolution relationship of the key variables over time, identify the key influencing factors, and obtain the key hydrological influencing factors; S4: Use the key hydrological influencing factors to evaluate the contribution degree of multiple variables to hydrological risk prediction, monitor the short-term changes in short-term rainfall, river water level fluctuations, and flow velocity mutations, calculate the weight changes of multiple variables within a short time window, analyze the change probability of the hydrological state, and adjust the key variables according to the state transition probability to obtain the hydrological state evolution path.

[0007] As a further solution of the present invention, the distribution characteristics of the hydrological mutation points include the priority area of the change rate, the mutation time point, and the corresponding time interval. The spatial diffusion trend of the hydrological mutation points includes the mutation propagation sequence from upstream to downstream, the mutation density in multiple river sections, and the corresponding spatial risk level. The key hydrological influencing factors include the change trend of precipitation, the key flow velocity threshold, and the time relationship affecting the water level. The hydrological state evolution path includes the weight dynamics of key variables, the state change probability in different stages, and the corresponding risk adjustment record.

[0008] As a further solution of the present invention, the steps for obtaining the distribution characteristics of the hydrological mutation points are specifically as follows: S111: Collect water level, flow velocity, and precipitation data, calculate the values at adjacent moments to obtain the change rate at adjacent moments. According to the fluctuation of the change rate, identify the dispersion degree of the change rate and the data fluctuation interval, and generate an initial set of hydrological parameter change rate mutation points; S112: Based on the initial set of hydrological parameter change rate mutation points, count the time intervals between adjacent mutation points, screen the mutation points that conform to the time distribution law, and eliminate the mutation points with abnormal time intervals. Use the formula: ; Calculate the standardized time deviation degree of the mutation points, and screen the mutation points with reasonable time intervals based on the numerical values to obtain the time distribution set of hydrological mutation points; Among them, represents the standardized time deviation degree of the mutation point , represents the time interval between adjacent mutation points, represents the mean value of the mutation point time intervals, represents the number of mutation points; S113: Using the set of time distributions of the hydrological mutation points, combining the distribution density and continuity characteristics of the mutation points, predict the duration of extreme events, and conduct segmented statistics on the prediction results to analyze the duration distribution of extreme events in the differential interval, so as to obtain the distribution characteristics of the hydrological mutation points.

[0009] As a further solution of the present invention, the steps for obtaining the spatial diffusion trend of the hydrological mutation points are specifically as follows: S211: Based on the distribution characteristics of the hydrological mutation points, count the number of mutation points in multiple hydrological basins, and use the formula: ; Calculate the distribution density per unit area to obtain the distribution density of the hydrological mutation points; Among them, represents the distribution density of the hydrological mutation points in the hydrological basin , represents the number of mutation points in the hydrological basin , represents the area of the hydrological basin , represents the basin and the weight relationship between the adjacent basin , represents the number of mutation points in the adjacent hydrological basin , represents the area of the adjacent hydrological basin , represents the number of adjacent hydrological basins; S212: Using the distribution density of the hydrological mutation points, analyze the diffusion path of the mutation points among multiple river reaches according to the spatial distribution relationship of the upstream river reach, the middle river reach and the downstream river reach, and arrange them in time series to obtain the propagation order of the hydrological mutation points; S213: Using the propagation order of the hydrological mutation points, calculate the propagation rate of the mutation points among the differential river reaches, analyze the diffusion range, and combine the influence range of the mutation points and the characteristics of the hydrological basin to divide the spatial risk level of the hydrological events, so as to obtain the spatial diffusion trend of the hydrological mutation points.

[0010] As a further solution of the present invention, the steps for obtaining the key hydrological influencing factors are specifically as follows: S311: Based on the spatial diffusion trend of the hydrological mutation points, analyze the change amplitude of the variables in the differential time intervals, and screen the variables that change prominently in multiple time periods to obtain the key hydrological variables; S312: Using the key hydrological variables, analyze the change trend of the key hydrological variables in the time interval, and use the formula: ; Calculate the change rate of key variables, analyze the relationship between key variables over time, and obtain the time evolution relationship of key variables; Among them, represents the change rate of the key variable , and respectively represent the values of the key variable and within the time intervals , represents the time interval, represents the key variable and the adjacent variable between the influence weights, is the number of adjacent variables; S313: Based on the time evolution relationship of the key variables, evaluate the correlation degree between multiple variables, identify the influencing factors that affect the key variables, analyze the contribution degree of the influencing factors, and obtain the key hydrological influencing factors.

[0011] As a further solution of the present invention, the steps for obtaining the hydrological state evolution path are specifically as follows: S411: Use the key hydrological influencing factors to monitor the short-term changes in short-term rainfall, river water level fluctuations, and flow velocity mutations, calculate the weight changes of multiple variables within a short time window, and evaluate the influence degree of multiple variables on the hydrological state within the short time window based on the weight changes to obtain the variable influence coefficient; S412: Based on the variable influence coefficient, analyze the change probability of the hydrological state within different time windows, and use the formula: ; Calculate the hydrological state change probability; Among them, represents the hydrological state change probability, represents the th variable at time weight, represents time when the variable weight, represents time when the short-term fluctuation amplitude of the variable, represents the total number of variables; S413: Use the hydrological state change probability to adjust the criticality of multiple variables, analyze the evolution process of the hydrological state, and obtain the hydrological state evolution path.

[0012] As a further solution of the present invention, the method further includes step S5: S5: Analyze the evolution trends of the mutation points in terms of time and space through the hydrological state evolution path, identify the development directions of hydrological events, conduct dynamic early warning based on the trends of the mutation points, classify the risks of hydrological abnormal events, and obtain the early warning results of hydrological abnormal events; The early warning results of the hydrological abnormal events include the time evolution sequence of the mutation points, the spatial dynamic trends, and the risk level adjustment results.

[0013] As a further solution of the present invention, the method further includes step S5: S5: Analyze the evolution trends of the mutation points in terms of time and space through the hydrological state evolution path, identify the development directions of hydrological events, conduct dynamic early warning based on the trends of the mutation points, classify the risks of hydrological abnormal events, and obtain the early warning results of hydrological abnormal events; The early warning results of the hydrological abnormal events include the time evolution sequence of the mutation points, the spatial dynamic trends, and the risk level adjustment results.

[0014] As a further solution of the present invention, the specific steps for obtaining the early warning results of the hydrological abnormal events are as follows: S511: Based on the hydrological state evolution path, analyze the spatial and time distributions of the mutation points in different time windows, and use the formula: ; Calculate the density change of the mutation points in space to obtain the spatio-temporal change trends of the mutation points; Among them, represents the density of the mutation points within the time window ; represents the th mutation point's spatial position at time ; represents the spatial position at time ; represents the length of the time window, represents the number of mutation points; S512: Use the spatio-temporal change trends of the mutation points to judge the development directions of hydrological events, conduct linear fitting based on the change directions of the mutation points within different time windows, analyze the change trends, and combine with real-time data to identify the development trends of hydrological events, and obtain the development directions of hydrological events; S513: Use the development directions of the hydrological events to calculate the change rate of the trends of the mutation points, dynamically set the risk level thresholds, classify the risks of hydrological abnormal events, and obtain the early warning results of hydrological abnormal events.

[0015] The intelligent hydrological data analysis system is used to execute the above intelligent hydrological data analysis method, and the system includes: The hydrological mutation point identification module obtains river water level, water flow velocity and precipitation data, calculates the change rate at adjacent times, marks the mutation points, counts the time intervals between adjacent mutation points, counts the occurrence frequencies of mutation points in different time periods, identifies the fluctuation relationship of mutation points in seasonal hydrological changes, and obtains the distribution characteristics of hydrological mutation points; Based on the distribution characteristics of the hydrological mutation points, the hydrological mutation point spatial analysis module calculates the distribution density of mutation points in different hydrological basins, counts the propagation order of mutation points between the upper, middle and lower reaches of the river, analyzes the diffusion rate of mutation points, and divides the spatial risk level of hydrological events according to the distribution of mutation points, so as to obtain the spatial diffusion trend of hydrological mutation points; The hydrological key variable extraction module uses the spatial diffusion trend of the hydrological mutation points to screen the key variables that affect the change of mutation points during the development of hydrological events, analyzes the change trend of key variables in different time periods, and analyzes the dynamic evolution relationship of key variables in hydrological events, so as to obtain the key hydrological influencing factors; The hydrological state evolution analysis module uses the key hydrological influencing factors to evaluate the impact of short-term rainfall, river water level fluctuations and water flow velocity changes on the evolution of hydrological states, calculates the change rate of hydrological states within a short time window, calculates the state transition probability of different variables, and constructs a hydrological state evolution path according to the change of variables on the time axis; The hydrological abnormal event analysis module analyzes the evolution trend of mutation points in time and space through the hydrological state evolution path, identifies the development direction of hydrological events, classifies hydrological abnormal events, and conducts dynamic early warning according to the change trend of mutation points, so as to obtain the early warning results of hydrological abnormal events.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In the present invention, by collecting and calculating the change rates of water level, flow velocity and precipitation data, and by marking mutation points, extreme hydrological events can be effectively identified and predicted, and the time intervals and durations of data points are analyzed, so that the distribution characteristics of hydrological mutations can be grasped at an early stage and responses can be made in advance. The analysis of the propagation order and diffusion rate of mutation points in different river reaches enhances the accurate division of the influence range and risk level of hydrological events, and realizes the effective management of the spatial risk of hydrological events. By screening key variables and evaluating the evolution relationship of variables over time, the understanding of key hydrological influencing factors is deepened, the accuracy of short-term and long-term hydrological risk prediction is enhanced, the development direction and risk level of hydrological events are helped to be identified, and a scientific basis is provided for the dynamic early warning of hydrological abnormal events. Brief Description of the Drawings

[0017] Figure 1 It is a schematic diagram of the working process of the present invention; Figure 2It is a flowchart for obtaining the distribution characteristics of hydrological mutation points in the present invention; Figure 3 It is a flowchart for obtaining the spatial diffusion trend of hydrological mutation points in the present invention; Figure 4 It is a flowchart for obtaining the key influencing factors of hydrology in the present invention; Figure 5 It is a flowchart for obtaining the evolution path of hydrological state in the present invention; Figure 6 It is a flowchart for obtaining the early warning result of hydrological abnormal events in the present invention. Specific implementation manners

[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0019] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more unless otherwise specifically defined.

[0020] Please refer to Figure 1 , the present invention provides a technical solution, a method for intelligent analysis of hydrological data, including the following steps: S1: Collect water level, flow velocity and precipitation data, calculate the change rate at adjacent moments, mark the hydrological parameter mutation points, count the time interval, predict the duration of extreme events, and obtain the distribution characteristics of hydrological mutation points; S2: Based on the distribution characteristics of hydrological mutation points, analyze the distribution density of mutation points in different hydrological basins, count the propagation order of mutation points between the upper, middle and lower reaches of the river section, analyze the diffusion rate of mutation points, identify the influence range of mutation points, and divide the spatial risk level of hydrological events to obtain the spatial diffusion trend of hydrological mutation points; S3: Based on the spatial diffusion trend of hydrological mutation points, screen key variables, analyze the change trend of key variables in different time periods, evaluate the evolution relationship of key variables over time, and identify key influencing factors to obtain key hydrological influencing factors; S4: Using the key hydrological influencing factors, evaluate the contribution degree of multiple variables to the hydrological risk prediction, monitor the short-term changes in short-term rainfall, river water level fluctuations, and sudden changes in flow velocity, calculate the weight changes of multiple variables within a short time window, analyze the change probability of the hydrological state in different time periods, and adjust the key variables according to the state transition probability to obtain the hydrological state evolution path; S5: Through the hydrological state evolution path, analyze the evolution trends of the mutation points in time and space, identify the development direction of hydrological events, conduct dynamic early warnings based on the mutation point trends, classify the risks of hydrological abnormal events, and obtain the early warning results of hydrological abnormal events; The distribution characteristics of hydrological mutation points include the priority area of the change rate, the mutation time point, and the corresponding time interval. The spatial diffusion trend of hydrological mutation points includes the mutation propagation sequence from upstream to downstream, the mutation density of multiple river reaches, and the corresponding spatial risk level. The key hydrological influencing factors include the change trend of precipitation, the key flow velocity threshold, and the time relationship affecting the water level. The hydrological state evolution path includes the dynamic weight of key variables, the state change probability in different stages, and the corresponding risk adjustment records. The early warning results of hydrological abnormal events include the time evolution sequence of mutation points, the spatial dynamic trend, and the risk level adjustment results.

[0021] Please refer to Figure 2 , and the specific steps for obtaining the distribution characteristics of hydrological mutation points are as follows: S111: Collect water level, flow velocity, and precipitation data, calculate the values at adjacent times to obtain the change rate at adjacent times, identify the dispersion degree of the change rate and the data fluctuation interval according to the fluctuation of the change rate, and generate the initial set of hydrological parameter change rate mutation points; Obtain water level, flow velocity, and precipitation data. The data acquisition equipment should be installed at hydrological stations in different regions. Use buoy-type water level gauges, ultrasonic flow meters, and automatic rain gauges for data acquisition. The water level, flow velocity, and precipitation data at each monitoring point are recorded with a sampling period of 1 hour and stored in the hydrological database. Call the data at adjacent times and calculate the data change rate between two adjacent times. The calculation method of the water level change rate is the difference between the current water level value and the water level value at the previous time. The flow velocity change rate is calculated in the same way, and the precipitation change rate is calculated as the difference between the cumulative precipitation in the current period and the precipitation in the previous period. Due to data volatility, small-amplitude change data need to be excluded, and only data points with obvious mutations are retained. The screening standard is based on the mean and standard deviation of the change amplitude of real-time data, and a threshold is set as the mutation determination standard. If the change rate is greater than the threshold , it is recorded as a mutation point. The threshold is determined according to the mean of the change amplitude of real-time hydrological data and the standard deviation ; Among them, Calculate the mean through long-term hydrological data, Calculate through the standard deviation formula; set the average water level change rate of a certain river section in the past year to 0.05 m / h and the standard deviation to 0.02 m / h, then the mutation determination threshold is: ; If the change rate of the data point is higher than 0.09, it is determined as a mutation point. Calculate and screen out the mutation point data of each hydrological parameter through the above method to generate the initial set of mutation points of the hydrological parameter change rate.

[0022] S112: Based on the initial set of mutation points of the hydrological parameter change rate, count the time intervals between adjacent mutation points, screen out the mutation points that conform to the time distribution law, and eliminate the mutation points with abnormal time intervals. Use the formula: ; Calculate the standardized time deviation of the mutation point, and screen out the mutation points with reasonable time intervals according to the value to obtain the time distribution set of hydrological mutation points; Among them, represents the standardized time deviation of the mutation point , represents the time interval between adjacent mutation points, represents the mean of the mutation point time intervals, represents the number of mutation points; Extract the time intervals between adjacent mutation points, calculate the mean and standard deviation of all adjacent mutation point time intervals to identify the time distribution that conforms to the hydrological mutation law. The time interval is calculated as the time difference between the mutation point and the previous mutation point , that is: ; Calculate all mean and standard deviation , set the time deviation screening threshold , calculate the mutation point time deviation as the screening criterion; Calculation example: If the mutation point interval times in a certain area are 3h, 5h, 8h, 6h, 7h in sequence, then: ; The standard deviation is: ; Given the mean , standard deviation , substitute into the formula to calculate the time deviation: ; ; ; Set a threshold , since , then eliminate the mutation points with large time deviations to obtain a set of mutation points that conform to the time distribution law. This result indicates that there are abnormal points in the time interval of the mutation points, and screening is required to ensure rationality. Combining this result, the selected mutation points are used for the prediction and analysis of subsequent extreme events.

[0023] S113: Use the time distribution set of hydrological mutation points, combine the distribution density and continuity characteristics of the mutation points, predict the duration of extreme events, and conduct segmented statistics on the prediction results to analyze the duration distribution of extreme events in the differential interval, so as to obtain the distribution characteristics of hydrological mutation points; Analyze the distribution density of the mutation points, count the number of mutation points in the differential time window, and the time window value refers to the typical duration of regional hydrological events. For example, flood events generally last for 12 - 48 hours, and drought events can last for weeks or months. Set the time window , calculate the mutation point density within the time window , the formula is: ; Among them, is the number of mutation points within the time window ; Set the set time window , if the number of mutation points in a certain area within 24 hours , then the mutation point density is: ; The higher the density, the greater the frequency of abnormal changes in the hydrological situation. Calculate the predicted duration of extreme events based on the mutation point density. Assume that the mutation points are normally distributed during the event duration, then the predicted duration can be calculated using the mean of the mutation point time intervals and the standard deviation , that is: ; Calculation example: If , then: , different mutation point density intervals can correspond to different types of hydrological events. Obtain the duration distribution of extreme events in each time interval to obtain the distribution characteristics of hydrological mutation points.

[0024] Please refer to Figure 3 , the specific steps for obtaining the spatial diffusion trend of hydrological mutation points are as follows: S211: Based on the distribution characteristics of hydrological mutation points, count the number of mutation points in multiple hydrological basins, and use the formula: ; Calculate the distribution density per unit area to obtain the hydrological mutation point distribution density; Among them, represents the hydrological mutation point distribution density within the hydrological basin , represents the number of mutation points within the hydrological basin , represents the area of the hydrological basin , represents the weight relationship between the basin and the adjacent basin , represents the number of mutation points within the hydrological adjacent basin , represents the area of the hydrological adjacent basin , represents the number of hydrological adjacent basins ; Obtain the number of mutation points within each hydrological basin. The specific method is to use hydrological indicators such as flow, precipitation, and evaporation in multi-year monitoring data, process the data using the mutation point identification method, and record the time and spatial location of the mutation. In a certain basin area, count the mutation points in the hydrological data in the past 10 years and obtain the total number of mutation points. Suppose 40 mutation points are detected in a certain basin, and the basin area is 800 square kilometers. Calculate the distribution density per unit area, that is, the number of mutation points per unit area. This calculation needs to consider the influence of adjacent basins and comprehensively evaluate the spatial influence of mutation points in each region; Substitute the data: Target basin: Number of mutation points , basin area square kilometers; Adjacent basins: The numbers of mutation points are respectively , ; Areas of adjacent basins are respectively , ; Influence weights are respectively , ; Calculate: ; ; ; ; ; The result shows that the distribution density of mutation points in this basin is low. According to the statistical data of hydrological events, this value indicates that the frequency of hydrological mutations in this basin is low, the distribution of mutation points is sparse, and subsequently, the hydrological risk trend in this area can be further evaluated by combining the propagation order of hydrological mutation points.

[0025] S212: Adopt the distribution density of hydrological mutation points, analyze the diffusion path of mutation points among multiple river reaches according to the spatial distribution relationship of the upstream river reach, middle river reach, and downstream river reach, and arrange them in time series to obtain the propagation order of hydrological mutation points; According to the spatial distribution relationship of the upstream river reach, middle river reach, and downstream river reach, count the diffusion path of mutation points among each river reach, analyze the change of their occurrence positions according to the time series of mutation points, form a time axis of the propagation of mutation points. In a certain basin, mutation points appeared in the upstream in 2015, spread to the middle reaches in 2017, and appeared in the downstream in 2020. By counting the positions of mutation points in each time period, calculating the number of mutation points stratified by river reach, and sorting them according to the time dimension to obtain the propagation order. The cumulative number of mutation points in the upstream of a certain basin is 50, 80 in the middle reaches, and 120 in the downstream. Since the mutation points gradually increase from the upstream to the downstream, the propagation order is "upstream-middle-lower", and the propagation order of hydrological mutation points is obtained.

[0026] S213: Use the propagation order of hydrological mutation points to calculate the propagation rate of mutation points among different differential river reaches, analyze the diffusion range, and combine the influence range of mutation points and the characteristics of the hydrological basin to divide the spatial risk level of hydrological events and obtain the spatial diffusion trend of hydrological mutation points; Calculate the propagation rate of mutation points among different river reaches and analyze the diffusion range. By calculating the time required for mutation points to spread between different river reaches and the propagation distance, the propagation rate per unit time is obtained. Mutation points in the upstream began to appear in 2015, mutations occurred in the middle reaches in 2017, and mutations occurred in the downstream in 2020. The basin distance from the upstream to the middle reaches is 80 kilometers, and the basin distance from the middle reaches to the downstream is 120 kilometers. Then the propagation rate of mutation points from the upstream to the middle reaches is: km / year; The propagation rate from the middle reaches to the downstream is: km / year; According to the hydrological characteristics and propagation rates of each basin, the influence range of the mutation points is divided. The influence range can be classified through the propagation rate threshold. When the propagation rate is lower than 20 km / year, it is a low-risk area; when it is between 20 and 50 km / year, it is a medium-risk area; when it is higher than 50 km / year, it is a high-risk area. Since the calculated propagation rate is 40 km / year, which belongs to the medium-risk area, it can be classified as a medium-risk level. Further, in combination with the basin characteristics, the spatial risk level of the area is divided to obtain the spatial diffusion trend of the hydrological mutation points.

[0027] Please refer to Figure 4 , and the specific steps for obtaining the key hydrological influencing factors are as follows: S311: Based on the spatial diffusion trend of hydrological mutation points, analyze the change range of variables in different time intervals, screen out the variables that change significantly in multiple time periods to obtain the key hydrological variables; It is necessary to extract the mutation point information from the existing hydrological data, use the hydrological monitoring data of different time stages to calculate the change rates of key hydrological variables such as water level, flow rate, and precipitation, and screen out the points with sudden increases or decreases in a short time as mutation points. Statistically analyze the spatial distribution of mutation points in different hydrological basins and river reaches, calculate the number of mutation points per unit area in the upstream, middle, and downstream regions respectively, and compare the mutation point densities between different basins. Suppose the area of a certain basin is 100 square kilometers and the observed number of mutation points is 50, then the mutation point distribution density of this basin can be calculated as 0.5 per square kilometer. By comparing the distribution densities of different basins, select the basins with higher densities as the key research objects. In the selected key basins, compare the mutation point distribution in different time intervals. If the number of mutation points in a certain basin is 120 during 2010 - 2015 and rises to 180 during 2015 - 2020, then the change rate of the mutation points can be calculated as: ; Adopt the statistical regression method to analyze the change relationship between the mutation point distribution density and time to judge whether there is a stable growth trend or periodic fluctuation. Through the above analysis, variables that show large changes in multiple time periods can be screened out. If the water level fluctuation exceeds ±2 meters or the flow rate change exceeds ±20% is defined as the mutation point threshold, then further calculations can be performed on the mutation point data that meet this standard to obtain the key hydrological variables.

[0028] S312: Use the key hydrological variables to analyze the change trend of the key hydrological variables in the time interval, using the formula: ; Calculate the change rate of the key variables, analyze the relationship between the key variables and time change to obtain the time evolution relationship of the key variables; Among them, Represents the rate of change of the key variable , and respectively represent the time intervals and within which the values of the key variable are Represents the time interval Represents the influence weight between the key variable and the adjacent variable , is the number of adjacent variables; Detailed explanation of the formula and the derivation process of formula calculation: and respectively represent the time intervals and within which the values of the key variable are, and the parameters and are calculated from the hydrological monitoring data. Set the average water level of a certain river basin to be 2.4 meters between 2000 and 2005, and 3.2 meters between 2005 and 2010. Then: ; Represents the time interval, and the time interval is set to 5 years: ; Represents the influence weight between the key variable and the adjacent variable , represents the weighted sum of squares of changes between adjacent variables; Calculate the first part: ; The adjacent variable represents the variables that affect the water level change, such as flow rate and precipitation. During the calculation process, the monitoring data of the flow rate and precipitation are taken. The changes in the average flow rate in the two time intervals are: , ; The changes in the average precipitation in the two time intervals are: , ; Set the weight to measure the influence of adjacent variables on the key variable. The flow rate weight , and the precipitation weight ; Calculate the weighted sum of squares of changes: ; ; ; ; Square root calculation part: ; Calculate: ; The result shows that the change rate of the key variable is 109.45, indicating that the change rate of the hydrological variable is large within the time interval. Compared with the small value, this value represents a strong temporal evolution trend of the hydrological key variable. This value can be used to further analyze the temporal dynamic pattern of the key variable and predict the probability of hydrological mutation.

[0029] S313: Based on the temporal evolution relationship of the key variable, evaluate the correlation degree between multiple variables, identify the influencing factors affecting the key variable, analyze the contribution degree of the influencing factors, and obtain the key hydrological influencing factors; Iteratively analyze the correlation between variables. For the change of each variable in each time interval, calculate the correlation coefficient between variables, calculate the Pearson correlation coefficient between water level and flow rate, precipitation, and judge whether its correlation reaches a significant level. Set the correlation coefficient calculation between water level and flow rate to 0.85, while the correlation coefficient calculation between water level and precipitation is 0.40. Then it can be determined that the relationship between water level and flow rate is closer. On this basis, screen out the variables that have a greater impact on the key variable, calculate the contribution rate of its influencing factors, set the contribution rate of flow rate to the change of water level to 60%, the contribution rate of precipitation to 30%, and the remaining factors to 10%. Then it can be determined that the flow rate is the main influencing factor. Through the regression analysis method, further calculate the regression equation between variables, using the formula: ; Among them, represents the key variable , represents the flow rate, represents the precipitation, represents the remaining influencing factors; Through this calculation process, the main influencing factors of the key variable can be identified, and the key hydrological influencing factors can be obtained.

[0030] Please refer to Figure 5 , the specific steps for obtaining the hydrological state evolution path are as follows: S411: Using key hydrological influencing factors, monitor the short-term changes of short-term rainfall, river water level fluctuations and flow velocity mutations, calculate the weight changes of multiple variables in a short time window, evaluate the influence of multiple variables on the hydrological state in a short time window based on the weight changes, and obtain the variable influence coefficient; To obtain monitoring data on short-term rainfall, river water level fluctuations and flow rate mutations, it is necessary to collect hydrological data through real-time monitoring equipment. Rainfall can be obtained through rain gauges or remote sensing satellites, river water level fluctuations can be continuously monitored using river depth sounders, and flow rate mutations are detected using current meters or buoy velocity measuring equipment. All data are stored in the form of time series and recorded once a minute to ensure the capture of short-term changes. The monitoring data are preprocessed to remove outliers. The judgment standard for outliers is based on the three times standard deviation range of real-time data. If the rainfall data of a certain monitoring is 50mm, the average rainfall is 20mm, and the standard deviation is 5mm, then the data is an outlier and needs to be removed. The preprocessed data is used to calculate the weight changes of multiple variables in a short time window. The weight calculation adopts the normalization method, and the rainfall, water level fluctuation, and flow rate mutation are set to be expressed as , , , and its weight calculation formula is: ; in, For variables The weight of Representative The data value of the variable, is the total number of variables; In this way, the relative importance of each variable is calculated and the , then the weight of rainfall is: ; The weight of water level fluctuation is: ; The weight of the sudden change in flow velocity is: ; Get the variable influence coefficient.

[0031] S412: Based on the variable influence coefficient, analyze the probability of change of hydrological state in the differentiated time window, using the formula: ; The probability of hydrological state change is calculated; in, represents the probability of hydrological state change, Representative variables in time The weight of The variable weight at representing time The short - term fluctuation amplitude of the variable at representing time represents the total number of variables; Detailed parameter description: The variable at time The normalized weight at this moment, which is obtained through normalization calculation. The normalization calculation method is as follows: ; Among them, The variable at time The original data value; This data value is collected in real - time by a hydrological monitoring device. For example, short - term rainfall is recorded by an automatic rain gauge, with the unit of mm; The river water level fluctuation is measured by a water level gauge, with the unit of m; The sudden change in flow velocity is measured by a current meter, with the unit of m / s; At a certain moment , the monitoring data is mm, m, m / s. Calculate the normalized weight: ; ; ; represents the weight of the previous time window, and the data source is the same as ; Set the monitoring data corresponding to the moment as , , then the calculation result is: ; ; ; The variable at time The short - term fluctuation amplitude at this time, which is quantified by calculating the standard deviation within the current time window. The calculation method is as follows: ; Among them, represents the time window length and is set minutes, represents the mean value of the variable within the window. In the past 5 minutes, the observed rainfall data is 22, 24, 26, 25, 23 mm, then the mean value is: ; Calculate the short-term fluctuation amplitude: ; ; Similarly, calculate the water level fluctuation and the flow velocity fluctuation , and set to obtain m, m / s; Derivation process of the formula calculation: Calculate the sum of the absolute values of the weight differences: ; ; Calculate the square root of the sum of the squares of the weights: ; ; Calculate the sum of the short-term fluctuation amplitudes: ; Calculate : ; This result indicates that the probability of the hydrological state change in the current time window is 0.0044. The small value means that the overall fluctuation of the hydrological state is relatively stable. Subsequently, this value can be compared with the preset change threshold to determine whether it is necessary to adjust the weights of the key variables. If exceeds the threshold, it indicates that a significant change in the hydrological state has occurred, and it is necessary to reallocate the variable weights to enhance the ability to identify hydrological state changes.

[0032] S413: Adopt the probability of hydrological state change, adjust the key importance of multiple variables, analyze the evolution process of the hydrological state, and obtain the hydrological state evolution path; Adjust the key importance of each variable and set the key importance adjustment threshold , if , it indicates that the hydrological state has changed and the weights of key variables need to be adjusted. In the previous time window, if the change amount of a certain variable exceeds 50% of the global change amount, the criticality of this variable is enhanced. If the change amount of the rainfall weight in a certain time window is 0.05, while the change amounts of water level fluctuation and flow velocity mutation are 0.02 and 0.01 respectively, then the proportion of the change amount of rainfall is: ; The result exceeds 50%. Therefore, the weight of rainfall needs to be increased by 10%. After adjustment, the weights are renormalized to obtain the hydrological state evolution path.

[0033] Please refer to Figure 6 , the specific steps for obtaining the warning result of hydrological anomaly events are as follows: S511: Based on the hydrological state evolution path, analyze the spatial and temporal distribution of mutation points under different time windows, and use the formula: ; Calculate the density change of mutation points in space to obtain the spatio-temporal change trend of mutation points; Among them, represents the density of mutation points within the time window , represents the th mutation point at time in space, represents the space position at time , represents the length of the time window, represents the number of mutation points; Parameter meaning and calculation process: represents the th mutation point at time in space, with the unit of meter, and this value is obtained in real time by monitoring equipment, setting the position of the water level mutation point recorded by the water level sensor or the coordinate information of the hydrological anomaly point obtained by the remote sensing satellite; represents the space position of the same mutation point at time , with the unit of meter, and has the same acquisition method as ; represents the number of mutation points, which is obtained by counting the monitoring data within a short time window. Set the monitored river section to be 10 kilometers. If 10 water level mutation points are found within a certain time window, then ; represents the length of the time window, in seconds, set to 300 seconds, i.e., 5 minutes, and this value is set according to the typical time scale of hydrological state changes. The size of the time window affects the calculation accuracy of mutation points. A small window can improve the spatio-temporal resolution, while a large window can more comprehensively reflect the evolution trend of hydrological events; Substitute the monitoring data for calculation, and set it within a certain time window seconds, the number of mutation points , and their spatial coordinates are as follows: Time : m, m, m, m, m; Time : m, m, m, m, m; Calculate the change in the position of the mutation point: ; ; Calculate the sum of squares of the positions of the mutation points: ; ; ; Calculate the square root: ; Calculate the mutation point density: ; The result shows that within this time window, the spatio-temporal distribution density of the mutation points is 0.000129, and the change in the number of mutation points per second per unit space is low, indicating that the evolution of hydrological events is relatively stable. If this value is much higher than the set reference value, it can be determined that the hydrological state has changed drastically.

[0034] S512: Use the spatio-temporal change trend of mutation points to judge the development direction of hydrological events. Perform linear fitting based on the change direction of mutation points within different time windows, analyze the change trend, and combine real-time data to identify the development trend of hydrological events and obtain the development direction of hydrological events; Judge the development direction of hydrological events, obtain the change in the position of mutation points within different time windows, calculate their spatial displacement vectors, and set the positions of the mutation points on a certain river section at time to be (10, 20), (15, 25), (20, 30), and at time When it becomes (12, 22), (18, 27), (24, 33), the displacement vectors of each mutation point are calculated , and (2, 2), (3, 2), (4, 3) are obtained respectively. Furthermore, the average direction vector of the mutation point group is calculated using the formula: ; The overall direction of the mutation point group is obtained. If the average directions in multiple time windows are consistent, it indicates that the hydrological event expands along this direction. It is necessary to calculate the spatial distribution variance of the mutation points to judge the concentration degree of the mutation points. A judgment criterion is set. If the variance of the mutation points is less than the set threshold, the mutation points change concentratedly and the direction of the hydrological event is stable. The threshold of the coordinate variance of the mutation points is set to 5. If the calculated distribution variance of the mutation points is 3, it indicates that the hydrological event develops along a stable direction, otherwise it is judged as diffusive development, and the development direction of the hydrological event is obtained.

[0035] S513: Using the development direction of the hydrological event, calculate the change rate of the mutation point trend, dynamically set the risk level threshold, and classify the hydrological anomaly event to obtain the early warning result of the hydrological anomaly event; Calculate the change rate of the mutation point trend. The change rate of the mutation point trend is defined as the growth rate of the number of mutation points per unit time. The total number of mutation points between the statistical time and is changed, and the growth ratio is calculated using the formula: ; Among them, represents the mutation point growth rate, represents the total number of mutation points within the time , represents the total number of mutation points within the time ; If the risk level threshold is set, then when , it is determined as a high-risk area. Suppose the number of mutation points in a certain area increases from 5 to 8 within 10 minutes, then the growth rate is calculated and it is determined as a high-risk area. In addition, further analyze the spatial distribution trend of the mutation points, calculate its distribution evenness. If the mutation points are highly concentrated in a certain area, the risk level is relatively high. The Gini coefficient is used to calculate its evenness, and the hydrological anomaly event is classified according to the risk level to obtain the early warning result of the hydrological anomaly event.

[0036] The intelligent hydrological data analysis system is used to execute the above intelligent hydrological data analysis method. The system includes: The hydrological mutation point identification module obtains river water level, water flow velocity and precipitation data, calculates the change rate at adjacent times, marks the mutation points, counts the time intervals between adjacent mutation points, counts the occurrence frequencies of mutation points in different time periods, identifies the fluctuation relationship of mutation points in seasonal hydrological changes, and obtains the distribution characteristics of hydrological mutation points; Based on the distribution characteristics of hydrological mutation points, the hydrological mutation point spatial analysis module calculates the distribution density of mutation points in different hydrological basins, counts the propagation order of mutation points between the upper, middle and lower reaches of the river, analyzes the diffusion rate of mutation points, and divides the spatial risk level of hydrological events according to the distribution of mutation points, so as to obtain the spatial diffusion trend of hydrological mutation points; The hydrological key variable extraction module uses the spatial diffusion trend of hydrological mutation points to screen the key variables that affect the change of mutation points in the development process of hydrological events, analyzes the change trend of key variables in different time periods, analyzes the dynamic evolution relationship of key variables in hydrological events, and obtains the key hydrological influencing factors; The hydrological state evolution analysis module uses the key hydrological influencing factors to evaluate the impact of short-term rainfall, river water level fluctuations and water flow velocity changes on the evolution of hydrological states, calculates the change rate of hydrological states within a short time window, calculates the state transition probability of different variables, and constructs a hydrological state evolution path according to the change of variables on the time axis; The hydrological abnormal event analysis module analyzes the evolution trend of mutation points in time and space through the hydrological state evolution path, identifies the development direction of hydrological events, classifies hydrological abnormal events, and conducts dynamic early warning according to the change trend of mutation points, so as to obtain the early warning results of hydrological abnormal events.

[0037] The above is only the preferred embodiment of the present invention, and does not limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention still belong to the protection scope of the technical solution of the present invention.

Claims

1. An intelligent analysis method for hydrological data, characterized in that, it includes the following steps: S1: Collect water level, flow velocity and precipitation data, calculate the change rate at adjacent times, mark the mutation points of hydrological parameters, count the time intervals, predict the duration of extreme events, and obtain the distribution characteristics of hydrological mutation points; S2: Based on the distribution characteristics of the hydrological mutation points, analyze the distribution density of the mutation points in different hydrological basins, count the propagation order of the mutation points between the upper, middle and lower reaches of the river, analyze the diffusion rate of the mutation points, identify the influence range of the mutation points, and divide the spatial risk level of hydrological events to obtain the spatial diffusion trend of hydrological mutation points; S3: Based on the spatial diffusion trend of the hydrological mutation points, screen key variables, analyze the change trend of key variables in different time periods, evaluate the evolution relationship of key variables over time, and identify key influencing factors to obtain hydrological key influencing factors; S4: Use the hydrological key influencing factors to evaluate the contribution degree of multiple variables to hydrological risk prediction, monitor the short-term changes of short-term rainfall, river water level fluctuations and flow velocity mutations, calculate the weight changes of multiple variables within a short time window, analyze the change probability of hydrological states, and adjust the key variables according to the state transition probability to obtain the hydrological state evolution path.

2. The intelligent analysis method for hydrological data according to claim 1, characterized in that, the distribution characteristics of the hydrological mutation points include the change rate area at adjacent times, the mutation points of hydrological parameters, and the corresponding time intervals. The spatial diffusion trend of the hydrological mutation points includes the mutation propagation order from upstream to downstream, the mutation diffusion rate of multiple river reaches, and the corresponding spatial risk levels. The hydrological key influencing factors include the change trend of precipitation, the critical flow velocity threshold, and the time relationship affecting the water level. The hydrological state evolution path includes the dynamic weight of key variables, the change probability of hydrological states in different stages, and the corresponding risk adjustment records.

3. The intelligent analysis method for hydrological data according to claim 1, characterized in that, the specific steps for obtaining the distribution characteristics of the hydrological mutation points are as follows: S111: Collect water level, flow velocity and precipitation data, calculate the values at adjacent times to obtain the change rate at adjacent times. According to the fluctuation of the change rate, identify the dispersion degree of the change rate and the data fluctuation interval, and generate an initial set of mutation points of the hydrological parameter change rate; S112: Based on the initial set of mutation points of the hydrological parameter change rate, count the time intervals between adjacent mutation points, screen the mutation points that conform to the time distribution law, and eliminate the mutation points with abnormal time intervals. Use the formula: ； Calculate the standardized time deviation degree of the mutation points, and select the mutation points with reasonable time intervals based on the numerical values to obtain the time distribution set of hydrological mutation points; Among them, represents the standardized time deviation of the mutation point of, represents the time interval between adjacent mutation points, represents the mean of the time intervals of the mutation points, represents the number of mutation points; S113: Use the time distribution set of hydrological mutation points, combine the distribution density and continuity characteristics of the mutation points, predict the duration of extreme events, and conduct segmented statistics on the prediction results. Analyze the duration distribution of extreme events in different intervals to obtain the distribution characteristics of hydrological mutation points.

4. The intelligent analysis method for hydrological data according to claim 3, characterized in that, the specific steps for obtaining the spatial diffusion trend of the hydrological mutation points are as follows: S211: Based on the distribution characteristics of the hydrological mutation points, count the number of mutation points in multiple hydrological basins, and use the formula: ； Calculate the distribution density per unit area to obtain the hydrological mutation point distribution density; Among them, represents the distribution density of hydrological mutation points within the hydrological basin, represents the number of mutation points within the hydrological basin, represents the area of the hydrological basin represents the weight relationship between the basin and the adjacent basin, represents the number of mutation points within the adjacent hydrological basin, represents the area of the adjacent hydrological basin represents the number of adjacent hydrological basins ;​​ S212: Use the hydrological mutation point distribution density, analyze the diffusion paths of mutation points among multiple river sections according to the spatial distribution relationship of the upstream river section, the middle - reach river section, and the downstream river section, and arrange them in time series to obtain the hydrological mutation point propagation order; S213: Utilize the hydrological mutation point propagation order, calculate the propagation rate of mutation points between different river sections, analyze the diffusion range, and combine the influence range of mutation points and the characteristics of hydrological basins to divide the spatial risk level of hydrological events and obtain the spatial diffusion trend of hydrological mutation points.

5. The intelligent hydrological data analysis method according to claim 4, wherein, the specific steps for obtaining the key hydrological influencing factors are as follows: S311: Based on the spatial diffusion trend of the hydrological mutation points, analyze the change amplitude of variables in different time intervals, and screen the variables with prominent changes in multiple time periods to obtain the key hydrological variables; S312: Use the key hydrological variables, analyze the change trend of the key hydrological variables in the time interval, and use the formula: ； Calculate the change rate of the key variables, analyze the relationship between the key variables and time changes, and obtain the time evolution relationship of the key variables; Among them, represents the change rate of the key variable , and represent the values of the key variable and within the time intervals respectively, represents the time interval, represents the influence weight between the key variable and the adjacent variable is the number of adjacent variables; S313: Based on the time evolution relationship of the key variables, evaluate the correlation degree between multiple variables, identify the influencing factors affecting the key variables, and analyze the contribution degree of the influencing factors to obtain the key hydrological influencing factors.

6. The intelligent hydrological data analysis method according to claim 5, wherein, the specific steps for obtaining the hydrological state evolution path are as follows: S411: Use the key hydrological influencing factors to monitor the short - term changes in short - term rainfall, river - channel water - level fluctuations, and flow - velocity mutations, calculate the weight changes of multiple variables within a short - time window, and evaluate the influence degree of multiple variables on the hydrological state within the short - time window based on the weight changes to obtain the variable influence coefficient; S412: Based on the variable influence coefficient, analyze the change probability of the hydrological state in different time windows, and use the formula: ； Calculate to obtain the hydrological state change probability; Among them, represents the probability of hydrological state change, represents the -th variable's weight at time , represents the variable weight at time , represents the short-term fluctuation range of the variable at time , represents the total number of variables; S413: Use the hydrological state change probability to adjust the key nature of multiple variables, analyze the evolution process of the hydrological state, and obtain the hydrological state evolution path.

7. The intelligent hydrological data analysis method according to claim 1, wherein, the method further includes step S5: S5: Through the hydrological state evolution path, analyze the evolution trends of mutation points in time and space, identify the development direction of hydrological events, conduct dynamic early warning based on the mutation point trends, classify the risks of hydrological abnormal events, and obtain the early - warning results of hydrological abnormal events; The early - warning results of the hydrological abnormal events include the time evolution sequence of mutation points, the spatial dynamic trend, and the risk - level adjustment results.

8. The intelligent hydrological data analysis method according to claim 7, wherein, the specific steps for obtaining the early - warning results of the hydrological abnormal events are as follows: S511: Based on the hydrological state evolution path, analyze the spatial and temporal distribution of mutation points under different time windows, and use the formula: ； Calculate the density change of mutation points in space to obtain the spatio-temporal change trend of mutation points; Among them, represents the mutation point density within the time window, represents the th mutation point's spatial position at time , represents the spatial position at time , represents the time window length, represents the number of mutation points; S512: Use the spatio-temporal change trend of the mutation points to judge the development direction of hydrological events. Perform linear fitting according to the change direction of the mutation points within different time windows, analyze the change trend, and combine real-time data to identify the development trend of hydrological events to obtain the development direction of hydrological events; S513: Use the development direction of the hydrological events to calculate the change rate of the mutation point trend, dynamically set the risk level threshold, and classify the risk of hydrological anomaly events to obtain the early warning result of hydrological anomaly events.

9. A hydrological data intelligent analysis system Characterized in that According to the hydrological data intelligent analysis method described in any one of claims 1-8, the system includes: The hydrological mutation point identification module obtains river water level, water flow velocity and precipitation data, calculates the change rate between adjacent moments, marks the mutation points, counts the time interval between adjacent mutation points, counts the occurrence frequency of mutation points in different time periods, and identifies the fluctuation relationship of mutation points in seasonal hydrological changes to obtain the distribution characteristics of hydrological mutation points; The hydrological mutation point spatial analysis module, based on the distribution characteristics of the hydrological mutation points, calculates the distribution density of the mutation points in different hydrological basins, counts the propagation order of the mutation points between the upper, middle and lower reaches of the river, analyzes the diffusion rate of the mutation points, and divides the spatial risk level of hydrological events according to the distribution of the mutation points to obtain the spatial diffusion trend of hydrological mutation points; The hydrological key variable extraction module uses the spatial diffusion trend of the hydrological mutation points to screen the key variables that affect the change of the mutation points during the development of hydrological events, analyzes the change trend of the key variables in different time periods, and analyzes the dynamic evolution relationship of the key variables in hydrological events to obtain the key hydrological influencing factors; The hydrological state evolution analysis module uses the key hydrological influencing factors to evaluate the impact of short-term rainfall, river water level fluctuations and water flow velocity changes on the hydrological state evolution, calculates the hydrological state change rate within a short time window, calculates the state transition probability of different variables, and constructs a hydrological state evolution path based on the variable changes on the time axis; The hydrological anomaly event analysis module analyzes the evolution trend of the mutation points in time and space through the hydrological state evolution path, identifies the development direction of hydrological events, and classifies the hydrological anomaly events, and performs dynamic early warning according to the change trend of the mutation points to obtain the early warning result of hydrological anomaly events.

Images