Flood entering judgment method for medium and large hydropower stations in Guangxi province based on dynamic rainfall threshold value

Abstract

The invention discloses a Guangxi medium and large hydropower station flood entering judgment method based on a dynamic rainfall threshold value, and relates to the technical field of hydrology and water resource and hydraulic dispatching decision-making auxiliary systems, and the method comprises the steps: obtaining and preprocessing rainfall and evaporation data in a drainage basin, and constructing a weight coefficient through a Thiessen polygon method; calculating basin surface rainfall and early-stage influence rainfall by adopting a weighting method; and calculating a dynamic flood entering rainfall threshold according to the index model, and determining a flood entering day in combination with a continuity discrimination method. According to the dynamic threshold model provided by the method, the early-stage hydrological state of the drainage basin is fully considered, dynamic adjustment of the flood entry threshold is realized by adopting an exponential function, and the physical rationality, the spatial representativeness and the aging precision of the model are greatly improved in combination with the surface rainfall and a continuous confirmation mechanism.

Description

Technical Field

[0001] This invention relates to the field of hydrological and water resources and hydraulic dispatching decision support system technology, specifically a method for judging the flood season of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds. Background Technology

[0002] In hydrological and meteorological research and water resource management, flood season entry assessment is a crucial component of flood warning and reservoir scheduling decisions, and has long been a focus of attention. Southern my country, especially Guangxi, with its complex terrain, significant spatial and temporal variations in rainfall, and frequent extreme weather events, demands higher precision in flood season identification. Traditional flood season entry assessment methods primarily rely on fixed dates or rainfall indicators, combined with monitoring data from regional hydrological stations. These methods played a fundamental role in early flood control systems. However, with the development of technologies such as meteorological remote sensing, Geographic Information Systems (GIS), and hydrological modeling, big data-based flood season assessment methods have emerged. These methods often incorporate parameters with more physical mechanisms, such as soil moisture and areal rainfall, aiming to improve the timeliness and accuracy of assessments.

[0003] However, existing methods still have significant limitations. First, most methods use static threshold settings, failing to dynamically adapt to the impact of previous rainfall and evaporation conditions on soil moisture, leading to delayed judgments or false alarms. Second, the traditional judgment mechanism of "cumulative rainfall exceeding a certain value over three consecutive days" or "station rainfall exceeding the warning line" fails to consider the spatial heterogeneity of rainfall, especially in the complex karst landforms of Guangxi, where the rainfall response characteristics of different sub-basins vary significantly, making it difficult for a unified indicator to accurately reflect the actual flood season status. Third, existing methods generally lack a systematic data preprocessing workflow, particularly lacking scientific handling of the uneven distribution of rain gauges, resulting in biases in areal rainfall calculations. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing flood season judgment methods lack dynamic response mechanisms, fail to consider the spatial distribution of rainfall, have non-standard data processing procedures, and are prone to misjudgment or omission. The problem is how to construct an intelligent flood season judgment mechanism that adapts to the complex terrain and climate characteristics of Guangxi.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds, comprising acquiring and preprocessing rainfall and evaporation data within the basin, constructing weighting coefficients using the Thiessen polygon method; calculating the basin surface rainfall and previous impact rainfall using a weighted method; calculating the dynamic flood season entry rainfall threshold based on an index model, and determining the flood season entry date by combining a continuity discrimination method.

[0007] As a preferred embodiment of the method for determining the flood season of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in this invention, the acquisition and preprocessing of rainfall and evaporation data within the basin includes: collecting daily rainfall data from rainfall monitoring stations within and around the control basin of the power station, and acquiring daily evaporation data from at least one representative meteorological station; simultaneously acquiring basic geographic information data, including DEM elevation raster, basin boundary vector data, and latitude and longitude coordinates of each station; performing quality control on the original data sequence, removing outliers, and imputing missing values ​​using a weighted average of adjacent stations; rainfall data must have spatial accessibility and temporal density, and data segments with missing lengths exceeding 3 days should be marked and not included in the areal rainfall calculation process; if evaporation data is missing, it is fitted and completed using historical monthly average evaporation; all data must have a unified time base and unit system.

[0008] As a preferred embodiment of the method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in this invention, the construction of weight coefficients includes: using raster analysis tools in ArcGIS software to reclassify the DEM raster data of the power station's control basin to obtain a base map; generating an uncropped Thiessen polygon layer based on rainfall station coordinates using a cost allocation formula; converting the control basin DEM raster into a vector map as the basin boundary; using a masking tool to cover the entire control basin with the uncropped layer to obtain a masked Thiessen polygon layer for each rainfall station; and calculating the Thiessen polygon area for each rainfall station. and with total area Calculate the weighting coefficients based on the benchmark. , is represented as: in, .

[0009] As a preferred embodiment of the method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in this invention, the calculation of basin-wide rainfall and previous impact rainfall includes calculating basin-wide rainfall based on daily rainfall data and constructed weighting coefficients. , is represented as: in, For the first Rainfall at each rain gauge station on that day The weighting coefficients for each station are used to calculate the anterior rainfall impact using a hydrological recursive model. The simulated residual soil moisture state is represented as: in, The amount of rainfall that affected the day in advance, The amount of rainfall that affected the area yesterday. This represents yesterday's average rainfall. The daily decay coefficient is expressed as: in, This represents the actual evaporation rate for the day. This represents the multi-year average evaporation rate.

[0010] As a preferred embodiment of the method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in this invention, the calculation of the dynamic flood season entry rainfall threshold includes constructing a dynamic threshold function to model the correlation between previous influencing rainfall and the required flood season entry rainfall threshold. in, This indicates the threshold of rainfall required for the start of the flood season on that day. The amount of rainfall that affected the day's rainfall 、 、 For undetermined model parameters, This is the lower limit of the threshold. This represents the upper limit of the threshold. Water deficit, representing the total water shortage required to reach the critical runoff production condition in a drought-stricken watershed. Its value is dominated by historical events where a prolonged drought was followed by a sudden heavy rainfall that brought the flood season to begin. The runoff response coefficient characterizes the nonlinear response intensity of the watershed runoff generation mechanism to changes in previous wet conditions. Critical runoff capacity characterizes the minimum runoff resistance of the underlying surface system in a watershed under fully wet conditions. Its value is dominated by historical events in which relatively small rainfall amounts were sufficient to initiate the flood season during periods of high humidity.

[0011] As a preferred embodiment of the method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in this invention, the calculation of the dynamic flood season entry rainfall threshold further includes training using historical data samples of more than 30 years, including daily rainfall, evaporation, and official flood season entry date information; and using the annual flood season entry date... Data points at As a positive sample, before the flood season Using intraday data as negative samples, construct the objective function: in, This represents the number of flood season entries that the model did not detect in the positive samples. To determine the number of times the model misclassifies non-flood season days as flood season days, and with the optimization objective of minimizing the total error, the SCE-UA algorithm is used to determine the optimal parameter combination. 、 、 , make the function It fits historical judgment results and has generalization ability.

[0012] As a preferred embodiment of the method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in this invention, the method of determining the flood season entry date by combining the continuous discrimination method includes setting a flood season entry judgment window, performing judgments daily by the model, and calculating the daily rainfall. With threshold If satisfied Mark that day as the initial start of the flood season; if it continues to occur... If the first date that meets the conditions is the initial date of the flood season, then the first date that meets the conditions will be confirmed as the official date of the flood season for that year; if the date is interrupted, the count will be restarted and it will not be considered the start of the flood season; once the start of the flood season is confirmed, the model will immediately output the result and automatically send the information to the power plant dispatch system.

[0013] Another objective of this invention is to provide a flood season judgment system for medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds.

[0014] As a preferred embodiment of the flood season entry judgment system for medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in this invention, the system includes: a coefficient construction module, a rainfall analysis module, and a flood season entry judgment module; the coefficient construction module is used to acquire and preprocess rainfall and evaporation data within the watershed, and construct weighted coefficients using the Thiessen polygon method; the rainfall analysis module is used to calculate the watershed surface rainfall and the previous impact rainfall using a weighted method; the flood season entry judgment module is used to calculate the dynamic flood season entry rainfall threshold based on an index model, and determine the flood season entry date by combining a continuity discrimination method.

[0015] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of a method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds.

[0016] A computer-readable storage medium having a computer program stored thereon, wherein when executed by a processor, the computer program implements the steps of a method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds.

[0017] The beneficial effects of this invention are as follows: The dynamic threshold model proposed in this invention for judging the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds fully considers the previous hydrological conditions of the basin, uses an exponential function to dynamically adjust the flood season entry threshold, and combines areal rainfall and continuous confirmation mechanisms to significantly improve the physical rationality, spatial representativeness, and timeliness accuracy of the model. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 The flowchart below shows the overall process of a method for determining the flood season of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds, as provided in Embodiment 1 of the present invention. Detailed Implementation

[0020] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0021] Example 1, referring to Figure 1 As an embodiment of the present invention, a method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds is provided, including: S1: Acquire and preprocess rainfall and evaporation data within the watershed, and construct weighting coefficients using the Thiessen polygon method.

[0022] Furthermore, acquiring and preprocessing rainfall and evaporation data within the basin includes collecting daily rainfall data from rainfall monitoring stations within and around the power station's controlled basin, and acquiring daily evaporation data from at least one representative meteorological station; simultaneously acquiring basic geographic information data, including DEM elevation raster, basin boundary vector data, and latitude and longitude coordinates of each station; performing quality control on the raw data sequence, removing outliers, and imputing missing values ​​using a weighted average of adjacent stations; rainfall data must have spatial accessibility and temporal density, and data segments with missing lengths exceeding 3 days should be marked and excluded from the areal rainfall calculation process; if evaporation data is missing, it should be fitted and completed using historical monthly average evaporation; all data must adhere to a unified time base and unit system.

[0023] It should be noted that the construction of weighting coefficients includes: using raster analysis tools in ArcGIS software to reclassify the DEM raster data of the power station's control basin to obtain a base map; generating uncropped Thiessen polygon layers based on rainfall station coordinates using a cost allocation formula; converting the control basin DEM raster into a vector map as the basin boundary; using a masking tool to cover the entire control basin with the uncropped layer to obtain masked Thiessen polygon layers for each rainfall station; and calculating the Thiessen polygon area for each rainfall station. and with total area Calculate the weighting coefficients based on the benchmark. , is represented as: in, .

[0024] It should also be noted that by comprehensively acquiring rainfall, evaporation, and geospatial information data within the power station's control basin, and standardizing the data (including outlier removal, missing value imputation, and unified format processing), the accuracy and continuity of the input data were ensured. Simultaneously, by constructing Thiessen polygons for rain gauge stations using a geographic information system, the influence area of ​​each rain gauge station on the entire basin was accurately modeled, and the spatial weight of each station was calculated. This solved the average error problem caused by the uneven distribution of rain gauge stations in traditional methods.

[0025] S2: The weighted method is used to calculate the watershed surface rainfall and the rainfall from previous influences.

[0026] Furthermore, the calculation of watershed rainfall and antecedent rainfall includes calculating watershed rainfall based on daily rainfall data and constructed weighting coefficients. , is represented as: in, For the first Rainfall at each rain gauge station on that day The weighting coefficients for each station are used to calculate the anterior rainfall impact using a hydrological recursive model. The simulated residual soil moisture state is represented as: in, The amount of rainfall that affected the day in advance, The amount of rainfall that affected the area yesterday. This represents yesterday's average rainfall. The daily decay coefficient is expressed as: in, This represents the actual evaporation rate for the day. This represents the multi-year average evaporation rate.

[0027] It should be noted that, based on the existing spatial weights, the average rainfall across the entire basin at any given time was further calculated, and a recursive soil moisture index was introduced, comprehensively considering the cumulative effect of previous rainfall and the continuous impact of evaporation on soil moisture. This approach not only preserves the influence of historical rainfall on the current hydrological state but also quantifies the current moisture level of the basin. A bridge is established between the basin's historical water dynamics and the current hydrological response, enabling flood season assessments to move beyond short-term heavy rainfall events and to possess a comprehensive understanding of continuous precipitation processes and previous meteorological conditions, thereby significantly enhancing the model's adaptability and stability under various climatic backgrounds.

[0028] S3: Calculate the dynamic flood season rainfall threshold based on the index model, and determine the flood season date by combining it with the continuity discrimination method.

[0029] Furthermore, calculating the dynamic flood season rainfall threshold includes constructing a dynamic threshold function to model the correlation between previous influencing rainfall and the required flood season rainfall threshold: in, This indicates the threshold of rainfall required for the start of the flood season on that day. The amount of rainfall that affected the day's rainfall 、 、 For undetermined model parameters, This is the lower limit of the threshold. This represents the upper limit of the threshold. Water deficit, representing the total water shortage required to reach the critical runoff production condition in a drought-stricken watershed. Its value is dominated by historical events where a prolonged drought was followed by a sudden heavy rainfall that brought the flood season to begin. The runoff response coefficient characterizes the nonlinear response intensity of the watershed runoff generation mechanism to changes in previous wet conditions. Critical runoff capacity characterizes the minimum runoff resistance of the underlying surface system in a watershed under fully wet conditions. Its value is dominated by historical events in which relatively small rainfall amounts were sufficient to initiate the flood season during periods of high humidity.

[0030] It should be noted that calculating the dynamic flood season rainfall threshold also includes training with historical data samples from over 30 years, including daily rainfall, evaporation, and official flood season start dates; and including the annual flood season start date... Data points at As a positive sample, before the flood season Using intraday data as negative samples, construct the objective function: in, This represents the number of flood season entries that the model did not detect in the positive samples. To determine the number of times the model misclassifies non-flood season days as flood season days, and with the optimization objective of minimizing the total error, the SCE-UA algorithm is used to determine the optimal parameter combination. 、 、 , make the function It fits historical judgment results and has generalization ability.

[0031] It should also be noted that determining the start of the flood season using the continuity discrimination method includes setting a flood season judgment window, the model performing the judgment daily, and calculating the daily rainfall. With threshold If satisfied Mark that day as the initial start of the flood season; if it continues to occur... If the first date that meets the conditions is the initial date of the flood season, then the first date that meets the conditions will be confirmed as the official date of the flood season for that year; if the date is interrupted, the count will be restarted and it will not be considered the start of the flood season; once the start of the flood season is confirmed, the model will immediately output the result and automatically send the information to the power plant dispatch system.

[0032] It should also be noted that the rainfall threshold required to determine the start of the flood season is dynamically adjusted based on the current humidity state of the basin, automatically correcting itself as climatic conditions change. This avoids the applicability limitations of traditional fixed-threshold methods in different years and climatic contexts. Simultaneously, a continuous judgment mechanism is introduced; only when the preliminary flood season conditions are met for several consecutive days is the start of the flood season finally confirmed, effectively filtering out short-term disturbances caused by extreme weather. This enhances the sensitivity and robustness of the judgment mechanism, achieving a dynamic decision-making logic that better aligns with natural hydrological processes. It strengthens the model's resilience to extreme weather interference in actual operation, providing more timely, accurate, and intelligent flood season early warning data for power station flood control scheduling.

[0033] Example 2, an embodiment of the present invention, provides a flood season entry judgment system for medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds, including a coefficient construction module, a rainfall analysis module, and a flood season entry judgment module.

[0034] The coefficient construction module is used to acquire and preprocess rainfall and evaporation data within the watershed, and construct weighting coefficients using the Thiessen polygon method; the rainfall analysis module is used to calculate the watershed surface rainfall and previous impact rainfall using a weighted method; the flood season entry judgment module is used to calculate the dynamic flood season entry rainfall threshold based on the index model, and determine the flood season entry date by combining the continuity discrimination method.

Claims

1. A method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds, characterized in that, include: Acquire and preprocess rainfall and evaporation data within the watershed, and construct weighting coefficients using the Thiessen polygon method; The weighted method was used to calculate the watershed surface rainfall and the antecedent rainfall. The dynamic threshold for the onset of the flood season is calculated based on the index model, and the date of the onset of the flood season is determined by combining the continuity discrimination method.

2. The method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in claim 1, characterized in that: The acquisition and preprocessing of rainfall and evaporation data within the watershed includes, Daily rainfall data were collected from rain gauge stations within and around the power station's controlled watershed, and daily evaporation data were obtained from at least one representative meteorological station. Simultaneously, basic geographic information data is acquired, including DEM elevation raster, watershed boundary vector data, and latitude and longitude coordinates of each station. Quality control is performed on the original data sequence to remove outliers. Missing values ​​are imputed using the weighted average of adjacent stations. Rainfall data must have spatial accessibility and temporal density. Data segments with missing lengths exceeding 3 days should be marked and not included in the areal rainfall calculation process. If evaporation data is missing, it should be supplemented by fitting historical monthly average evaporation data; all data must be based on a consistent time base and unit system.

3. The method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in claim 2, characterized in that: The constructed weight coefficients include, The raster analysis tool in ArcGIS software was used to reclassify the DEM raster data of the power station's control basin to obtain a base map. Based on the coordinates of the rain gauge stations, an uncropped Thiessen polygon layer is generated using a cost allocation formula; Convert the control watershed DEM raster into a vector map to serve as the watershed boundary; By using the masking tool to fully control the uncropped layer area within the watershed, we obtain the masked Thiessen polygon layers for each rainfall station. Calculate the area of ​​the Thiessen polygon at each rainfall station. and with total area Calculate the weighting coefficients based on the benchmark. , represented as: in, .

4. The method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in claim 3, characterized in that: The calculation of watershed surface rainfall and antecedent rainfall includes... The watershed surface rainfall is calculated based on daily rainfall data and constructed weighting coefficients. , represented as: in, For the first Rainfall at each rain gauge station on that day The weighting coefficients for each station are used to calculate the anterior rainfall impact using a hydrological recursive model. The simulated residual soil moisture state is represented as: in, The amount of rainfall that affected the day in advance, The amount of rainfall that affected the area yesterday. This represents yesterday's average rainfall. The daily decay coefficient is expressed as: in, This represents the actual evaporation rate for the day. This represents the multi-year average evaporation rate.

5. The method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in claim 4, characterized in that: The calculation of the dynamic flood season rainfall threshold includes, Construct a dynamic threshold function to model the correlation between previous rainfall impacts and the required flood season rainfall threshold: in, This indicates the threshold of rainfall required for the start of the flood season on that day. The amount of rainfall that affected the day's rainfall 、 、 For undetermined model parameters, The lower limit of the threshold. This represents the upper limit of the threshold. Water deficit, representing the total water shortage required to reach the critical runoff production condition in a drought-stricken watershed. Its value is dominated by historical events where a prolonged drought was followed by a sudden heavy rainfall that brought the flood season to begin. The runoff response coefficient characterizes the nonlinear response intensity of the watershed runoff generation mechanism to changes in previous wet conditions. Critical runoff capacity characterizes the minimum runoff resistance of the underlying surface system in a watershed under fully wet conditions. Its value is dominated by historical events in which relatively small rainfall amounts were sufficient to initiate the flood season during periods of high humidity.

6. The method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in claim 5, characterized in that: The calculation of the dynamic flood season rainfall threshold also includes, The training was conducted using a historical data sample of over 30 years, including daily rainfall, evaporation, and official flood season date information. The start of the flood season each year Data points at As a positive sample, before the flood season Using intraday data as negative samples, construct the objective function: in, This represents the number of flood season entries that the model did not detect in the positive samples. To determine the number of times the model misclassifies non-flood season days as flood season days, and with the optimization objective of minimizing the total error, the SCE-UA algorithm is used to determine the optimal parameter combination. 、 、 , make the function It fits historical judgment results and has generalization ability.

7. The method for determining the flood season entry of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in claim 6, characterized in that: The method for determining the start of the flood season using the continuity discrimination method includes, A flood season entry threshold is set, and the model performs the assessment daily, calculating the daily rainfall. With threshold If satisfied This day is marked as the initial start of the flood season; If it occurs consecutively If the date of the initial start of the flood season is determined, the first date that meets the conditions will be confirmed as the official start of the flood season for that year. If the count is interrupted, it will restart and will not be considered the start of the flood season. Once the flood season is confirmed, the model immediately outputs the flood season result and automatically sends the information to the power plant dispatch system.

8. A flood season entry judgment system for medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds, employing the flood season entry judgment method for medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds as described in any one of claims 1 to 7, characterized in that: Includes a coefficient construction module, a rainfall analysis module, and a flood season entry judgment module; The coefficient construction module is used to acquire and preprocess rainfall and evaporation data within the watershed, and to construct weighting coefficients using the Thiessen polygon method. The rainfall analysis module is used to calculate the watershed surface rainfall and the previous impact rainfall using a weighted method. The flood season entry judgment module is used to calculate the dynamic flood season entry rainfall threshold based on the index model and determine the flood season entry date by combining the continuity discrimination method.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for determining the flood season of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds, as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for judging the flood season of medium and large-sized hydropower stations in Guangxi based on dynamic rainfall thresholds, as described in any one of claims 1 to 7.

Images